element and a element. The

element shows
the value of the name state variable, and the

element, when clicked, updates the value
of the name state variable to Svelte. Due to the binding, when the button is clicked, it will also
update the username props in the component.
Here’s the updated version of the component:","

Name in Profile: {username}

username = \"World\"}>Update from Profile button>"," In this code snippet, we are looking at the component. This component receives a prop
called username, whose value is displayed inside a

element. We’ve also added a button, and
when clicked, it will update the value of the username prop to \"World\".
Because of the binding we established in the parent component, any change to the username prop
in this component will also update the name state variable in the parent component.
Indeed, if you click on the button in the parent component, both the name state variable in the parent
component and the username prop in the component will update to \"Svelte\". On
the other hand, if you click on the button in the component, both the name state variable
in the parent component and the username prop in the component will change to
\"World\". This is the power of two-way binding in Svelte, allowing you to easily synchronize data
between parent and child components.
\f One-way versus two-way data flow 41","When we talk about binding in Svelte, we often refer to two-way data binding. But what exactly does
two-way mean? And is there such a thing as one-way data binding? Let’s delve deeper into the concept
of data flow to clarify these terms.","One-way versus two-way data flow","When you pass data from one component to another component either through props or binding, data","flows from one component to another component. The term data flow refers to how data is passed or","transmitted between components or elements within a web application.","Understanding data flow is important when designing an application’s architecture, as it helps to","establish clear lines of communication between components and determine how information is shared","and updated throughout the system.","Data flow can be unidirectional (one-way) or bidirectional (two-way), depending on how data is","transferred between components.","In one-way data flow, data moves in a single direction, from a parent to a child component, or from","a component to a DOM element. This unidirectional flow is achieved through component props or","DOM element attributes.","For example, consider the following code snippets featuring two components, component A and","component B.","In component B, we define a prop named value:"," ","   export let value;","In component A, we import component B and pass the value of a variable named data to the prop
of component B:"," ","   import B from './B.svelte';","   let data;"," ","Here, the value of data in component A is passed to the value prop in component B, illustrating
the data flow from component A to component B.
\f42 Managing Props and State"," If you draw up the data flow in a diagram, this is what a one-way data flow looks like:"," Figure 3.1: One-way data flow"," The boxes show the data state and value prop, and the arrow shows how changing one value will
affect another.
In one-way data flow, changes to data in the parent component automatically propagate to the child
component, but the reverse is not true.
Building on the example with component A and component B, if the value of the data variable in
component A changes, this change would automatically update the value props in component B.
However, any changes made directly to the value props in component B would not affect the data
variable in component A. Data changes only flow from component A to component B, but not in the
reverse direction.
Having a one-way data flow makes the application easier to reason about and debug, as data changes
follow a predictable path. For example, when the data variable in component A changes unexpectedly,
isolating the issue becomes more straightforward. Because of the one-way data flow, we know that
any changes to the data variable in component A will only originate from component A, not from
component B.
On the other hand, two-way data flow allows data to flow in both directions, enabling automatic
updates between the parent and child components, or between a component and DOM elements.
Two-way data flow in Svelte is achieved through component or DOM element bindings.
For example, consider the following code snippets featuring two components, component C and
component D.
In component D, we define a prop named value:"," ","   export let value;","\f One-way versus two-way data flow 43","In component C, we import component D and bind the value of a variable named data to the prop
of component D:"," ","   import D from './D.svelte';"," ","Here, the value of data in component C is bound to the value prop in component D, illustrating
the two-way data flow between component C to component D.
Here is a diagram showing how the data flows in a two-way data flow:"," Figure 3.2: Two-way data flow","In a two-way data flow, when you change the data in the parent component, the child component is
automatically updated; conversely, changing the data in the child component automatically updates
the parent.
Building on the example with component C and component D, if the value of the data variable in
component C changes, this change would automatically update the value props in component D.
Conversely, unlike one-way data flow, any changes made directly to the value props in component D
would update the data variable in component C. Data changes flow from component C to component
D as well as from component D to component C.
While this bidirectional flow of data can make it easier to keep data in different components synchronized,
it can also introduce complexity and make the application harder to debug, since data changes can
originate from multiple sources.
For example, when the data variable in component C changes unexpectedly, isolating the issue becomes
more complex. Because of two-way data flow, any changes to the data variable in component C could
originate from component C, as well as from component D. This complexity can escalate further if the
value prop of component D is also bound to the prop of another component.
\f44 Managing Props and State"," In summary, one-way data flow offers predictability and easier debugging, while two-way data flow
makes it much easier to synchronize data across multiple components but at the cost of added complexity.
Now that we’ve delved into both one-way and two-way data flows via props, let’s explore how to create
state variables that derive their values from props."," Deriving states from props with a reactive declaration"," It’s common in Svelte to create new state variables based on the values of props."," For instance, a component might accept a date value as a prop and display a"," formatted date inside a element. To use the component, you might write"," the following:"," "," To display the date as formatted text, you could first define a variable named label, deriving its
value from the date prop:"," ","   export let date;","   // Deriving the 'label' variable from the 'date' prop","   let label = date.toLocaleDateString();"," {label}"," In this code snippet, we defined a variable called label and derived its value from the date prop
using the toLocaleDateString() method. This variable is then used inside a element
to display the formatted date.
In the preceding code snippet, the label variable is initialized when the component
is first created. However, if the date prop changes after the component has been initialized, the label
variable won’t update to reflect the new value. This is not the intended behavior, as the
component should always display the most up-to-date formatted date based on the latest date prop.
To solve this issue, you can use Svelte’s reactive declarations to keep the label variable updated
whenever the date prop changes.
Svelte’s reactive declarations utilize a special $: syntax to mark a statement as reactive. This means
that whenever the variables used in that statement change, the statement itself will be re-run.
Let’s modify our component code to use Svelte’s reactive declaration:","\f Deriving states from props with a reactive declaration 45","   export let date;","   // Deriving the 'label' variable from the 'date' prop","   $: label = date.toLocaleDateString();","In this code snippet, by modifying the declaration of label to a reactive declaration, the component
will automatically re-compute the label variable whenever the date prop is modified, ensuring
that the component is always displaying the most current formatted date.
So, how does Svelte know when a reactive declaration statement should be re-run?
Svelte re-runs a reactive declaration statement whenever any of its dependent variables change. The
Svelte compiler identifies these dependencies by analyzing the statement.
For example, in the reactive declaration $: label = date.toLocaleDateString();,
Svelte recognizes that the dependency for this statement is the date variable. Therefore, whenever
the date changes, the statement will re-run and update the value of the label variable.
A good rule of thumb for identifying dependencies in a reactive declaration is to look for any
variable on the right side of the equal sign (=). These variables are considered dependencies of the
reactive declaration.
You can include multiple dependencies within a single reactive declaration. For instance, let’s say
we want to add a new locale prop to the component. To use this new prop, you
might write something like this:"," ","In this code snippet, we pass in a new locale prop with the value de-DE to format the date
in German. To accommodate this new locale prop, we’ll need to modify our
component as follows:","
{label}","In this updated code, the reactive declaration $: label = date.toLocaleDateString(locale);","now has two dependencies: date and locale. Svelte will automatically re-run this statement","whenever either of these variables changes, ensuring that the label value stays up to date with the","latest date and locale props.","\f46 Managing Props and State"," Now that we’ve covered the basics of props, bindings, states, and derived states, it’s crucial to note that
as components become more complex, managing these elements can quickly become overwhelming.
In the next section, we’ll explore some tips for effectively managing complex derived states to keep
them manageable."," Managing complex derived states"," As your Svelte application grows more complex, it will likely involve a greater number of interconnected"," components with multiple props and derived states. When dealing with this complexity, tracking"," updates and changes can become a complex task. Each prop or state change can affect other parts of"," your component, making it challenging to manage and predict how your component will behave."," To make this easier, here are some guidelines to consider:"," • Maintain one-way data flow for derived states"," While it’s possible to derive state from props and other states, it’s crucial to maintain a one-way"," data flow to simplify both debugging and understanding. Consider the following Svelte example:"," "," This code snippet won’t compile in Svelte due to a cyclical dependency detected by the compiler.
Though workarounds may exist to circumvent this compile error, the code is purposely flawed
to highlight the issue.
Upon observing the data flow, we can see that valueC depends on valueA, valueD
depends on both valueB and valueC, and then valueC in turn depends on valueD.
Consequently, it’s unclear how valueC is actually calculated, it could be either valueA +
5 or the minimum of valueC and valueD / 2. Such complexity makes the code hard to
understand and increases the likelihood of bugs.
• Group similar logic together
Consider this tip a stylistic suggestion. When dealing with multiple reactive declarations, it’s
beneficial to group those that are related together. Utilize blank lines to create a visual separation
between unrelated reactive declarations. This not only improves readability but also aids in
code maintenance.
It is worth noting that the Svelte compiler takes care of execution order based on dependencies
regardless of how you arrange your declarations. For example, the following two code snippets
will behave identically due to Svelte’s handling of dependencies:
\f Managing complex derived states 47"," "," In the preceding snippet, b is declared reactively before c. However, it behaves identically to"," the following snippet, where c is declared before b. Let’s look at the next one:"," "," The Svelte compiler analyzes the dependencies of each declaration and executes them in the
correct order. In this case, it evaluates b = a * 2 before c = b * 2 since the latter depends
on the value of b, established by the former declaration.
• Avoid reassigning props directly
It might be tempting to modify the value of a prop directly, especially when you want to transform
its value or provide a default fallback. For example, you might consider writing something like
this to set a default value for an undefined prop:
"," While this code snippet may seem to work as intended, directly reassigning prop values within
a component can introduce confusion and make data flow unpredictable. This practice can
muddy the waters when you’re trying to trace the origin of data changes, making it unclear
whether the alteration occurred within the component itself or came from the parent component.
Instead, it’s better to declare a new state variable to manage this behavior:
"," In this improved example, we introduce a new variable, dataWithDefault, which takes on"," either the value of the data prop or a default value of 100 if data is undefined. This approach"," makes the component’s behavior more transparent and easier to debug.","\f48 Managing Props and State"," • Be cautious when updating derived states"," Modifying a derived state directly can introduce inconsistencies, particularly when that state"," is based on props or other state variables."," Consider the following example:"," "," "," "," In this example, we have a single prop named value and two derived states, double and
triple, which are two and three times the value of the value prop respectively. The component
features two input boxes, each bound to double and triple using two-way binding.
Here, triple can be modified in two ways: either by updating the value prop, which will
keep triple at three times the value of value, or by directly changing the value in the input
box, thereby directly altering triple.
If you type into the input box bound to triple, you’ll find that its value diverges from being
strictly three times the value prop. This inconsistency arises because triple is now subject
to changes from multiple sources, causing it to go out of sync with the original value.
If you were to map out a diagram illustrating the data flow between value double and
triple, you’d get the following diagram:"," Figure 3.3: Data flows between props, states, and the input elements
\f Updating props using derived states 49"," Notice that both double and triple have two incoming arrows, suggesting multiple
pathways for changing their values. Conversely, value has no incoming arrows, indicating
that modifying triple alone would cause value and double to go out of sync.
Therefore, it is recommended to refrain from manually updating derived states, as this complicates
debugging and makes the component’s behavior harder to reason about.
If you wish to modify the input while maintaining the synchrony between value, double,
and triple, a solution will be discussed in the following section.
By keeping these tips in mind, you can better manage complex derived states and make your components
more maintainable and easier to understand.
In our last tip, we highlighted that updating the derived states can lead to inconsistencies between
states and props and noted that there’s a solution to modify the input while keeping everything in
sync. That solution is what we will explore next.","Updating props using derived states","In an attempt to synchronize the value prop with changes to the input bound to triple, one might","be tempted to add another reactive declaration. This declaration would update the value prop to be","one-third of triple whenever triple changes. Here is the proposed modification:","

","As we discussed earlier, it’s best practice to maintain a one-way data flow for derived states to simplify
debugging and data management. Indeed, the Svelte compiler flags the preceding code snippet
for cyclical dependencies. This is because double is derived from value, and value is in turn
dependent on double.
However, Svelte’s compiler determines dependency relationships based solely on the reactive declarations.
By refactoring the code to make these relationships less obvious, you can bypass the compiler’s cyclical
dependency check. Here’s a refactored version of the code that does just that:","

"," In the provided code snippet, we’ve shifted the equations value = double / 2 and value = triple
/ 3 into separate functions named updateValueFromDouble and updateValueFromTriple.
This change lets us evade Svelte’s compiler warning about cyclical dependencies.
However, there’s a catch. If you try altering the triple input, it updates value but doesn’t refresh
double. This happens because Svelte avoids infinite update loops. Changing triple would set off
a chain reaction—updating value, then double, then back to value, and so on.
This is how the data flow looks right now:"," Figure 3.4: Data flows between props, states, and the input elements
\f Updating props using derived states 51","As you can see in the diagram, we update the value of value, double, and triple through reactive
declarations, creating a loop in the data flow, indicated by the bold arrows.
Therefore, using derived states to update their original properties via reactive declarations isn’t advisable.
A better approach to keep value, double, and triple in sync is to establish value as the single
source of truth. Since both double and triple are derived from value, any changes to the input
should first update value. This, in turn, triggers the reactive declarations to automatically recalculate
double and triple.
Here’s the updated code:","

updateValueFromDouble(e.target.value)} />

updateValueFromTriple(e.target.value)} />","In the preceding code snippet, we’ve shifted our approach to focus on value as the sole source of
truth. Instead of binding the input elements directly to double and triple, we’ve added event
listeners that update value based on user input. This change automatically updates double and
triple through reactive declarations, which then refresh the displayed values in the input fields.
\f52 Managing Props and State"," With the updated code, the data flow is now streamlined, as shown in the following diagram:"," Figure 3.5: Data flows between props, states, and the input elements"," The values for double and triple are directly derived from value, which then populates the
input fields. When you modify the input, it alters value directly, which in turn automatically updates
both double and triple, as well as the input fields themselves.
So, this is how you maintain synchronized states for value, double, and triple, by focusing on
the data flow and keeping a single source of truth."," Summary
Understanding how to handle props and state effectively is crucial for creating robust Svelte apps.
This chapter has shown you how Svelte uses props, bindings, and reactive declarations to facilitate
data passing and state changes across components.
It’s crucial to keep an eye on data flow within a component. Having a unified and organized data flow
makes the code easier to follow and debug. Good data management paves the way for building more
intricate and dynamic apps with ease.
In the next chapter, we’ll explore how to compose Svelte components to construct more
complex applications.
\f 4
Composing Components
As your application grows, cramming all the logic into a single component becomes impractical.
You’ll need to split your app into smaller, modular components and assemble them to form a more
complex application.
In this chapter, we’ll explore various techniques to combine components effectively. We’ll start by
examining how to inject custom content into a component using slots. Then, we’ll discuss how to
conditionally render different HTML elements within a component. We’ll also delve into recursive
components, useful for displaying nested or hierarchical data.
We’ll guide you through each topic with hands-on examples, ensuring the techniques you learn are
both practical and applicable in real-world scenarios. By the end of this chapter, you’ll have a richer
set of strategies to compose components in your Svelte applications.
In this chapter, you will learn the following:"," • Manipulating how a child component looks from the perspective of its parent
• Passing dynamic content through slots
• Rendering different HTML element and component types
• Creating recursive components for recursive data
• The Container/Presentational pattern","Let’s kick things off by exploring the various ways we can manipulate and control the content of a
child component.","Technical requirements
All the code in this chapter can be found at https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter04.
\f54 Composing Components"," Manipulating how a child component looks"," When you’re combining multiple components, you’ll need to manage how each child component"," appears and behaves. Even though a child component handles its own display and logic, it still offers"," controls to tweak its appearance and behavior. From the perspective of the parent component, you’ll"," want to coordinate these child components to achieve the desired overall functionality."," In this section, we’ll explore various ways to control the look of child components, ranging from the"," most to the least commonly used methods. Understanding these options will equip you with the tools"," to make your components both versatile and effective."," The list of options to control how the child component looks includes the following:"," • Controlling through props: This is perhaps the most straightforward way to influence the
behavior and appearance of a child component. By passing props from a parent component to a
child, you can make your child component highly dynamic and responsive to external changes.
To demonstrate how a parent component can control content using props in Svelte, consider
the following code example:

"," In the code snippet, we imported the Child component and passed the message prop to it.
The child component, implemented in the following code snippet, then uses this message
prop to display text within a

element:

{message}

"," As you can see, the parent component has the ability to dictate the text shown in the Child
component by modifying the message prop’s value. Controlling props is a simple yet effective
way to manipulate a child component’s content. If you’re interested in learning more about
props, we covered the topic extensively in Chapter 3.
• Controlling through context: Svelte’s context API allows components to share state without
explicitly passing props through every layer of the component tree. This is particularly useful
when you have deeply nested components.
Here’s an example of a parent component:
"," "," In the previous code, we established a context named message with the 'Hello from
parent' value, using setContext. Then, we imported and used the Child component
without sending any props to it.
The following is the code for the Child component:

{message}

"," Here, getContext is used to read the message context value, which is then displayed
within a

element. As demonstrated, the parent component can influence the text in the
child component’s

element by changing the context value.
For a deeper dive into Svelte’s context feature, you can refer to Chapter 8, where we will explore
the topic in greater detail.
• Controlling the styles: Manipulating a child component’s appearance isn’t solely about controlling
the data passed to it. It also involves tweaking or modifying its styles.
You can modify a component style through CSS custom properties. This approach allows for
greater design flexibility, and it ensures that child components can adapt to various contexts
or themes within the parent component or the broader application.
For an in-depth discussion on how to alter a component’s style using CSS custom properties,
feel free to refer to Chapter 2.
• Passing dynamic content through slot: Svelte’s slot feature allows you to insert custom content
into specific areas of a child component. This provides a flexible way to have greater control
over a component’s content without modifying its internal behavior.
We will talk about slots and how to use them in the following section.
As you can see, there are various ways to shape a child component’s appearance and behavior. When
we compose different components within a parent component, the goal is to make them work together
in a coordinated way. You can use a combination of the approaches discussed to achieve this.
Most of these methods will be covered in separate chapters, and in the following section, we’ll focus
on how to dynamically alter a child component’s look by passing content through slots.
\f56 Composing Components"," Passing dynamic content through slots"," When building complex applications, one size doesn’t always fit all. Balancing between a component’s"," modularity and a component’s flexibility for customization is crucial."," Take a generic Card component, for example. You might sometimes want to include special"," headlines, unique lists, or custom footers for specific use cases. It’s nearly impossible to anticipate"," every requirement, so it’s essential to design components that are both modular and maintainable,"," yet still open to customization."," This is where Svelte’s slot feature shines. Slots enable you to inject dynamic content into your components,"," making them incredibly versatile. Instead of designing a Card component that tries to include every"," possible feature, aim for a simple, clean base that can be enhanced through composition. This approach"," allows you to piece together more complex, customized components as your needs evolve."," In a Svelte component, a element is a placeholder within your component where you"," can inject any type of content from a parent component. Here’s how you can define a slot inside a"," Svelte component:"," ","

","   ","

"," In the preceding code snippet, we defined a element inside a

element. Any dynamic
content from the parent component will be inserted into the

element.
Now, to pass dynamic content from a parent component into the slot of a child component, you need
to place that content between the opening and closing tags of the child component:","

  

Special headline

"," In this code snippet, we passed in a

element between the opening and closing tags of the
component. This effectively replaces the element in the Card component with
the

element we provide.
There may be instances where you don’t need to pass dynamic content from a parent component to a
child component’s slot. For these cases, Svelte allows you to define default content within the
element, providing a fallback when no custom content is supplied by the parent.
\f Passing dynamic content through slots 57","Providing default slot content","To provide default content for your element, you can place them inside the element,","like so:","

  
    

this is the default content

  

","In the preceding code snippet, we defined

, containing the text This is the default
content within the element. This serves as the default content for the slot.
When you use this Card component in a parent component and don’t provide content for the slot,
as shown in the following code snippet, the default text will automatically appear:","
","By utilizing the default content in slots, you can create more flexible Svelte components that offer
sensible fallbacks when no custom content is provided.
In the examples we’ve seen so far, we’ve been limited to inserting dynamic content from the parent
component into just one spot. But what if we wanted multiple insertion points for dynamic content?
Let’s explore how to achieve that.","Having multiple slots with named slots","Slots are amazing, and you are not limited to a single slot either. You can define multiple slots in a","component and name them to target specific areas for dynamic content.","You can use the name attribute on the element to give the slot a name, as shown in the","following code:"," ","

","   

","     ","   

","   ","\f58 Composing Components"," In the previous code snippet, we defined two named slots, one named \"header\" and another
named \"footer\".
To target the dynamic content into these named slots, you will need to use the
element in the parent component:","

  
    

Special headline

  
  
    Custom footer
  
"," Here, we pass two elements into the Card component. The first has a slot
attribute set to \"header\", and the second slot attribute is set to \"footer\".
The name specified in the slot attribute will correspond with the named slots within the Card
component. The content within replaces the name=\"header\"> in the Card component. Similarly, the content within slot=\"footer\"> takes the place of in the component.
Sometimes, you will want the dynamic content from the parent component to be aware of the data
within the child component. To handle this scenario, Svelte offers a feature known as slot props, which
allows you to pass data from the child component to the content in the slot. Let’s explore how this works."," Passing data through slot props"," You can pass data from the child component back to the dynamic content in the parent component"," using slot props, like so:","   

","     ","   

"," In the code example, we passed in two additional attributes to the element besides the name
attribute, which is used to name the slot. These additional attributes, width and height, serve as
slot props; their values can be accessed in the parent component when you create dynamic content.
\f Rendering different HTML element or component types 59","Here is an example of how you can use these slot prop values when using :","

  
    

Dimension: {width} x {height}

  
","In the preceding code snippet, you can see that we use let:width and let:height within the
element. These are called let: bindings.
The let: bindings allow us to access the width and height slot props provided by the child Card
component. The values for the width and height are then used within an

element to display
the dimensions. This way, we can dynamically adjust content in the parent component, based on data
originating from the child component.
Now that we have covered how to pass dynamic content through slots, how to define multiple-named
slots, and how to pass data from a child component to the dynamic slot content in a parent component,
you are well equipped to create more versatile and reusable Svelte components."," Renderless component
One of many common patterns to compose components using slots is the Renderless component
pattern. This technique allows you to build components that are purely functional, leaving the
presentation details to the parent component. We have dedicated Chapter 11 to delving into
this fascinating topic.","Now that we’ve explored how to pass dynamic content through slots, let’s dive into another exciting
feature of Svelte that lets you dynamically render different element or component types.","Rendering different HTML element or component types","In any dynamic application, there comes a time when you need even more flexibility than what static","components or elements offer. What if you don’t know the type of element or component you’ll need","to render until runtime?","Let’s imagine that you’re building a form generator, and the type of form field – whether it’s ,",", or – is determined by dynamic data. How could you switch between these","components seamlessly, especially when they share the same set of props?","\f60 Composing Components"," One straightforward approach is to use Svelte’s {#if} blocks to conditionally render the component
you need. Here is an example code snippet:","
{#if type === \"input\"}
  
{:else if type === \"checkbox\"}
  
{:else}
  
{/if}"," In the code snippet, we use the {#if} blocks to choose among three different component types,
passing the same prop values to each. If you find yourself needing to manage more component types,
this could lead to a lengthy and hard-to-maintain series of conditional blocks.
Is there a more efficient way to handle this?
Svelte provides two specialized elements, and ,
precisely for this. The element lets you create different HTML element
types dynamically, based on variable data, whereas the element lets you
dynamically render different Svelte component types.
Here’s how you can rewrite the previous example using :","
\f Rendering different HTML element or component types 61","   this={DynamicComponent}
  value={value}
  onChange={onChange}
/>","In the preceding code snippet, the DynamicComponent variable is used to hold the type of
component that will be rendered. This component type is then passed to the this attribute within
the element. The element also accepts other
props such as value and onChange.
With the preceding code, what happens is that renders the designated
component stored in the this attribute, simultaneously forwarding any props passed to
to this dynamically rendered component. For example, if the value of
DynamicComponent is the Select component, then the preceding code is effectively the same
as this code:"," ","By using , we simplify the code and make it more maintainable. It’s also
easier to extend; adding another form element type would only require an additional condition
and assignment.
Now that we’ve explored the use of , let’s look at ,
which follows a similar pattern. The following is a sample code snippet that demonstrates the usage
of :","

  Click Me
","In the preceding code snippet, the type variable holds the type of HTML element we want to render
– in this example, it’s button. The tag dynamically creates an HTML element
of the type specified by type. So, if type is 'button', this will render as a

element,
containing the text \"Click Me\".
This approach is particularly useful when you want to switch the type of an HTML element, based
on some condition, without having to rewrite the entire block of code. All you need to do is change
the value of type.
To summarize, and offer a more efficient and
maintainable way to handle dynamic rendering needs. They provide a robust alternative to multiple
{#if} blocks, making your Svelte applications more flexible and easier to manage.
\f62 Composing Components"," Sometimes, when designing Svelte components for data visualization, we encounter recursive data
structures. In such cases, we need to build components that can recursively render themselves. Let’s
delve into how we can accomplish this next."," Creating recursive components for recursive data"," A recursive data structure is a data structure that is defined in terms of itself, meaning that it can be"," composed of smaller instances of the same type of structure. Recursive data is everywhere – think of"," a comment section where replies can have their own sub-replies, or a filesystem with nested folders."," Creating a component to display them in a frontend application can be challenging."," Imagine we have a variable called folder, which is an array containing either files or additional"," folders. In this example, the folder variable could look like this:"," const folder = [
  { type: 'file', name: 'a.js' },
  { type: 'file', name: 'b.js' },
  { type: 'folder', name: 'c', children: [
    { type: 'file', name: 'd.js' },
  ]},
];"," Currently, our folder variable is two levels deep. To represent this structure in a Svelte component,
you might think to use nested {#each} blocks:","

  {#each folder as item}
    
• 
        {#if item.type === 'file'}
          
  File: {item.name}
  
        {:else}
          
  Folder: {item.name}
  
          
  
            {#each item.children as child}
              
  • 
                  {#if child.type === 'file'}
                    
    File: {child.name}
    
                  {:else}
                    ...
                  {/if}
                
  
  \f Creating recursive components for recursive data 63","           {/each}
          
  
        {/if}
      

  {/each}

","In the preceding code snippet, we used an {#each} block to iterate over the items in the folder
variable, rendering either a file or a folder based on item.type. If item.type is a folder, we use
another nested {#each} block to go through its contents.
However, here’s the catch – the inner folder could also contain files or additional folders, making it
recursive. As a result, we end up with repeated code for the inner and outer {#each} blocks. This
works fine for a folder structure that’s only two levels deep, but what if it’s more complex? How can
we avoid duplicating the same code for each level of nesting?
Svelte offers an elegant solution to handle such recursive structures with the
element. The element allows a component to embed itself within its own markup,
thus making it easier to handle recursive data.
Here’s the updated code using the element:","

  {#each folder as item}
    
• 
        {#if item.type === 'file'}
          
  File: {item.name}
  
        {:else}
          
  Folder: {item.name}
  
          
        {/if}
      

  {/each}

","In the updated code snippet, we replaced the nested {#each} block with folder={item.children} />. This effectively re-renders the current component, passing
item.children as the folder prop.
The beauty of this is that it eliminates the need to initiate a new

element and duplicate {#each}
blocks for each nesting level. Instead, the component simply reuses itself, making it capable of handling
nested structures of any depth.
\f64 Composing Components"," When comparing this with the previous code snippet, you can immediately see the advantage – it’s
much cleaner and avoids repetitive code, making it easier to read and maintain, and it scales naturally
to handle any level of nested folders.
Now that we’ve discussed how to tackle recursive data with , let’s look at a practical
example in the following section, where we will create a Svelte-based JSON tree viewer."," Example – a JSON tree viewer"," In this section, we will walk you through building a JSON tree viewer component in Svelte. The JSON"," tree viewer component helps you visualize JSON data in a tree-like format. Along the way, we’ll make"," use of some of the advanced Svelte features we’ve covered in this chapter, including ,"," , and slots."," Before we start, let’s think about what our JSON Tree Viewer should look like and how it should"," behave. Essentially, a JSON object is made up of key-value pairs, where the values can be either"," primitives, arrays, or other nested objects. Our goal is to display these key-value pairs in a way that"," clearly represents the hierarchical structure of the JSON object."," So, let’s create a JsonTree component for our JSON tree viewer:"," ","   export let data;","
","   {#each Object.entries(data) as [key, value]}","     
• {key}: {value}
• ","   {/each}","
"," In the preceding code snippet, we defined a JsonTree component that accepts data as a prop. Inside
this component, we utilize Svelte’s {#each} block to generate a list of
• elements. Each element
  displays a key-value pair from the data object.
  However, these values can vary. They could be primitive types or nested objects. Take the following example:"," data = {
    name: 'John Doe',
    address: {
      street: '123 Main St'
    },
  }
  \f Example – a JSON tree viewer 65","In the preceding data object, the name key has a primitive string value, while the value for the
  address key is an object. For nested objects like this, we will need to use our JsonTree component
  recursively to render that nested object. This is where comes into play.
  Here’s the updated code:"," ","   export let data;","
  ","   {#each Object.entries(data) as [key, value]}","     
  • ","       {key}:","       {#if typeof value === 'object'}","         ","       {:else}","         {value}","       {/if}","   {/each}","
  ","In this updated code snippet, we introduced an {#if} block that renders differently based on
  whether the value is a primitive type or a nested object. For nested objects, we recursively render
  another JsonTree component using . This allows us to elegantly handle JSON
  objects of any depth.
  Now, in our current implementation, we haven’t made distinctions between different types of primitive
  values. Suppose you’d like to render various primitive types using specialized Svelte components. In that
  case, you could dynamically switch between different component types using .
  Let’s assume you have three separate Svelte components – StringValue for strings, NumberValue
  for numbers, and JsonValue for others (such as Booleans, null, and undefined). Here is how
  you can update the code to use :","
  
  
    {#each Object.entries(data) as [key, value]}
      
  • 
          {key}:
          {#if typeof value === 'object'}
            
          {:else}
             value={value} />
          {/if}
        
  
    {/each}
  
  "," In the updated code snippet, we created a utility function named getComponent to determine which
  component to render, based on the type of the primitive value. We then used this={getComponent(typeof value)} /> to dynamically load the appropriate component
  for each primitive type, allowing our JsonTree component to switch to different components for
  different types of data.
  Lastly, to make our JSON tree viewer more versatile, we can introduce named slots to customize the
  appearance of keys and values. By doing so, users can easily tailor the look of these elements according
  to their needs, while keeping our current design as the default fallback. Let’s update the code to add
  the two named slots:","
  
    {#each Object.entries(data) as [key, value]}
      
  • 
          {key}:
          {#if typeof value === 'object'}
            
          {:else}
            
               value={value} />
            
          {/if}
        
  
    {/each}
  
  
  \f The Container/Presentational pattern 67","In the updated code, we added two named slots – one to customize keys, called \"obj-key\", and
  another to customize values, called \"obj-value\". These slots receive the current key and value
  as slot props, enabling you to tailor their appearance in the parent component.
  For instance, if you wish to change how keys and values are displayed, here’s how you can write in
  your parent component:"," ","   ","     {key}","   ","   ","     {value}"," ","Here, we use Svelte’s special element to target the named slots, \"obj-key\"
  and \"obj-value\", in our JsonTree component. The let:key and let:value syntax allows
  us to access the current key and value being rendered, respectively.
  In the element, we wrapped the key in an tag and the value in a
  tag, offering custom styling for these elements. This customization overrides the default rendering
  provided by the JsonTree component. If you don’t specify any custom content, the component will
  fall back to its default rendering, offering both robustness and flexibility.
  By combining for recursion, for dynamic behavior,
  and slots for customization, we have created a flexible and powerful JSON tree viewer. This not only
  demonstrates Svelte’s capability to handle complex UI patterns but also serves as a practical example
  of component composition in Svelte.
  You can find the complete code for this exercise here: https://github.com/PacktPublishing/
  Real-World-Svelte/tree/main/Chapter04/06-exercise-svelte-json-tree.
  Before we wrap up this chapter, let us take a moment to discuss a helpful common pattern to organize
  our components, the Container/Presentational pattern.","The Container/Presentational pattern","As your application scales in complexity, you may find it beneficial to adopt specific design patterns or","guidelines to structure components. One such approach is the Container/Presentational pattern, which","divides a component into two categories, the Container component and the Presentational component:"," • Container components focus on functionality. They handle data fetching, state management, and
  user interactions. While they usually don’t render Document Object Model (DOM) elements
  themselves, they wrap around Presentational components and supply them with data and actions.
  \f68 Composing Components"," • Presentational components are all about the user interface. They get their data and event-handling
  functions exclusively through props, making them highly reusable and straightforward to test."," A common scenario where you’ll see this pattern in action is when using a UI component library.
  In this case, the library’s components serve as the presentational elements, focusing solely on how
  things look. Meanwhile, the components you create that utilize these library elements act as Container
  components, managing data and handling interactions such as clicks, typing, and dragging.
  The Component/Presentational pattern is useful in a few aspects:"," • Simplicity: Separating behavior from appearance makes your codebase easier to understand
  and maintain
  • Reusability: Since Presentational components are agnostic about data sources or user actions,
  you can reuse them in different parts of your application
  • Collaboration: With this division, designers can work on the Presentational components while
  developers focus on the Container components, streamlining development"," As you come to appreciate the benefits of the Component/Presentational pattern, there are specific
  situations where I think you should consider using it, such as the following:"," • When your application starts to grow: Managing everything in a single component can become
  confusing as complexity increases
  • When you find yourself repeating the same UI patterns: Creating reusable Presentational
  components can save time in the long run
  • When your team scales: As your development team grows, having a standardized way of
  building components can reduce learning curves and prevent code conflicts"," While the Container/Presentational pattern offers a structured approach to organizing your components,
  it’s not always the best fit – especially for smaller applications, where it might be overkill. Hopefully,
  with the insights provided, you’ll be better equipped to make informed decisions."," Summary
  In this chapter, we delved into various strategies for component composition in Svelte, each offering
  its own set of advantages and applicable scenarios. Mastering these techniques will equip you to build
  Svelte applications that are not only more dynamic but also easier to maintain and scale.
  We kicked off by discussing multiple ways to influence a child component. These ranged from
  controlling props or using Svelte’s context to customizing styles via CSS custom properties, and even
  dynamically passing content through slots.
  \f Summary 69","Then, we turned our attention to some of Svelte’s special elements. We explored
  and to dynamically render various HTML elements and Svelte components.
  We also learned about , which allows a component to reference itself, thereby
  facilitating the creation of recursive UI structures. We then applied these newfound skills to build a
  JSON tree viewer as an illustrative example.
  Finally, we touched upon a popular design pattern – the Container/Presentational pattern. We examined
  its advantages and considered scenarios where it would be beneficial to employ this approach.
  Armed with these advanced techniques, you are now better prepared to tackle complex challenges
  in your Svelte projects. As we conclude our four-chapter exploration of writing Svelte components,
  we’ll shift our focus in the upcoming chapter to another fundamental feature of Svelte – Svelte actions.
  \f\f Part 2: Actions","In this part, we embark on a journey to learn about Svelte actions. Across three chapters, we will
  delve into three different use cases of Svelte actions. Starting with the creation of custom events using
  actions, we’ll progress to integrating third-party JavaScript libraries with Svelte via actions. Finally, we’ll
  unveil techniques to progressively enhance our applications, harnessing the power of Svelte actions.
  This part has the following chapters:"," • Chapter 5, Custom Events with Actions"," • Chapter 6, Integrating Libraries with Actions"," • Chapter 7, Progressive Enhancements with Svelte Actions","\f\f 5"," Custom Events with Actions","Actions are one of the most powerful features of Svelte.","They are a lightweight alternative to a component that encapsulates logic and data into a reusable unit.","They help us reuse the same logic on different elements.","While components have life cycle methods such as onMount and onDestroy that run when all","the elements within the component are added to or removed from the DOM, actions are designed","to handle the logic for individual elements, running only when that specific element is added to or","removed from the DOM.","While components can receive and react to prop changes, you can pass data to actions from a parent","component to a child component. The actions will react when the data is changed and you can specify","how the action should react when that data changes.","Actions are simple yet amazingly versatile. You can use them for various things. In this and the","following chapters, we are going to explore some of the use cases of actions.","One of the many use cases of actions is to manage elements’ event listeners. Event listeners are very","common in web applications as they allow us to implement specific behaviors in response to user","actions. This makes our application more interactive and dynamic. So, it will be interesting to see how","Svelte actions can be used to help us manage event listeners.","In this chapter, we will start by looking at how we can use actions to help manage event listeners for","elements. We will follow that up with examples and exercises to reinforce the idea.","This chapter includes sections on the following:"," • Managing event listeners with actions"," • Creating custom events with actions","\f74 Custom Events with Actions"," Technical requirements
  You can find the projects of the chapter here: https://github.com/PacktPublishing/
  Real-World-Svelte/tree/main/Chapter05"," Defining actions
  Before we start to talk about using Svelte actions to create custom events, let’s quickly recap how to
  define an action in Svelte.
  In Svelte, an action is nothing but a function that follows an action contract. This means if a function
  follows a specific function signature, it is considered an action. Here is the function signature of an action:"," function action(node) {
    return {
      destroy() {}
    };
  }"," It is a function that optionally returns an object that has a destroy method."," In this case, since the action function follows the action contract, it is a Svelte action."," To use the Svelte action on an element, you can use the use: directive:","
  "," Here, we used the Svelte action named action on the div element.
  So, what will happen to the div element with a Svelte action?
  When the

  element is mounted to the DOM, Svelte will call the action function with the
  reference to the

  element:"," const action_obj = action(div);"," When the element is removed from the DOM, Svelte will call the destroy method from the object
  returned from calling the action function:"," if (action_obj && action_obj.destroy) action_obj.destroy();"," We can customize the behavior of a Svelte action by using parameters. We can pass in an extra
  parameter to the action function:"," function action(node, parameter) {","    update(parameter) {},","\f Reusing DOM event logic with custom events 75","    destroy() {},
    };
  }","Also, you can add another method to the returning object, update, which will be called when the
  value of the parameter changes:"," action_obj.update(new_value);","To pass in an additional parameter, you can specify it in a similar syntax as the value attribute:","

  ","Now that we know how to use and define an action in Svelte, let’s take a look at our first use case
  of actions.","Reusing DOM event logic with custom events","Before we start talking directly about actions, let’s look at an example of using the mousedown and","mouseup events to create a long-press behavior. We shall see how this simple example will lead us","on to Svelte actions:","
    on:mousedown={handleMousedown}
    on:mouseup={handleMouseup}
  />","In the preceding example, we tried to implement a long-press behavior in a button. The idea is to
  press and hold the button for more than two seconds and then perform some action. As we detect it’s
  a long press, we print 'long press!' into the console.
  To implement the long-press behavior, I attached two event listeners: mousedown and mouseup.
  The two event listeners work together. mousedown starts counting down using setTimeout for
  two seconds, and mouseup clears the countdown using clearTimeout. If the user did not hold
  \f76 Custom Events with Actions"," onto the button for long enough, the timeout would not be triggered and it would not be considered
  a long press. Note that to coordinate the timers between the two event listeners, the timer variable
  is shared across the event listeners.
  As you can see, to implement a long-press behavior, you will need two event listeners and one
  shared variable.
  Now, what if you need to have this long-press behavior on another button?
  You can’t really share the same event listener and the variable, as you might want a different duration
  to be considered as a timeout or a different behavior when a long press happens.
  So, you would have to redeclare them again and remember to pair the right event with the right
  event listener."," Encapsulating logic into a component"," A way to recreate a different long-press button is to encapsulate it as a component, putting all the"," long-press button logic into a component and reusing the component as a means to reuse the logic."," When you define a component, you define the logic as well as the elements in the component. This"," means that if we put both the long-press logic and the button element into the component, we have"," to use the long-press logic together with the button element and no other elements."," If you want to customize the element, maybe by using a different element, having a different style,"," showing a different text content, or adding more event listeners, you would have to define the styles,"," text contents, or event listeners as props of the component and pass them into the button element"," in the component:"," "," ","   ","

"," In the preceding code, we passed in extra attributes from the props into the button element along
with forwarding two events, click and dblclick, from the button element out to the component.
\f Reusing DOM event logic with custom events 77","The point I am trying to make here is that if you wish to reuse event listener logic via components, you
will find yourself having to take care of other attributes that go along with the element in the component.
We can do much more with a component, but if we are trying to reuse just the long-press behavior, then
reusing it by defining it in a component is a bit overkill, and it can grow unmanageable rather quickly.
So, what other options do we have?","Encapsulating logic into an action","A better option is to use an action to encapsulate the long-press behavior.","Let’s just do that and then I’ll explain why using an action is a better approach:"," function longPress(node) {
  let timer;
  function handleMousedown() {
    timer = setTimeout(() => {
      console.log('long press!');
    }, 2000);
  }
  function handleMouseup() {
    clearTimeout(timer);
  }
  node.addEventListener('mousedown', handleMousedown);
  node.addEventListener('mouseup', handleMouseup);
  return {
    destroy() {
     node.removeEventListener('mousedown', handleMousedown);
     node.removeEventListener('mouseup', handleMouseup);
    }
  }
}","With the action defined (as shown in the preceding code), we can use the action over multiple elements:"," Button one"," Button two","You can apply the action to a different type of element:"," Hold on to me","You can also use it alongside other attributes or event listeners:"," ","\f78 Custom Events with Actions"," I hope you can see that the longPress action encapsulates only the long-press behavior. Unlike
the LongPressButton component, the longPress action can easily be reused in any element.
So, as a rule of thumb, when abstracting logic from an element, if you are abstracting it together with
the elements, it is okay to use a component. But if you only need to abstract the logic behavior from
an element, use actions.
Actions are a great tool for abstracting out element-level logic. But there is still one missing piece of the
puzzle: how should we add a different long-press handler for a different element? We’ll see how next."," Passing parameters to an action"," So, how do we customize the behavior of our longPress action?"," You could have probably guessed the answer since you’ve seen it in the previous section. We can"," customize the behavior of an action by passing parameters into the action."," For example, if we want to have a different function to handle a long-press action on different button"," elements, we can pass a different function into the action through action parameters:"," "," "," We will then receive this function within the second argument of the action function:"," function longPress(node, fn) {","   // ..."," We call the function passed in when it is considered a long press:"," fn();"," We can similarly pass in other parameters, such as the duration to be considered as a long press.
In Svelte actions, you can only pass in one parameter. To pass in multiple parameters, you will have
to turn them into an object and pass them in as one parameter.
Over in our longPress action, we want to pass the function to be called when the longPress
action is detected, as well as the duration to be considered a long press, into the longPress action.
To pass both the function and the duration in, we create an object that has them as object values and
pass the object as the action parameter:","   use:longPress={{
    onLongPress: doSomething1,
\f Reusing DOM event logic with custom events 79","     duration: 5000
  }} />","Some of the parameters in the object could be optional, so we might need to provide a default value
when reading them within the action:"," function longPress(node, { onLongPress, duration = 1000 }) {","   // if not specified, the default duration is 1s","At this stage, you might wonder, can the onLongPress be optional too?
It does not make much sense in our case since the main goal of our action is to detect a long press
and call the onLongPress function.
However, this is a good question.
In some other actions, you may have function handlers that are optional.
For example, what if we have a gesture action that can detect different gestures performed on
the element? In that case, each gesture callback function can be optional since you may only be
interested in one of the gestures:"," function gesture(node, { onSwipe, onDrag, onPinch }) { }","The onSwipe, onDrag, and onPinch function handlers in this gesture action are optional
and could be undefined.
Instead of creating an empty function as a fallback, we should instead check whether the function is
defined before calling it:"," if (typeof onSwipe === 'function') onSwipe();","This way, we don’t have to create a function unnecessarily.","However, this gets messier when you have multiple callback functions and you need to check whether","each of them has been defined before calling them.","Is there a better way to handle this?","Yes, there is.","In fact, a more idiomatic Svelte way of having an action to notify or call a function when something","happens is to dispatch a custom event.","For example, to know whether the user has long-pressed the element, it would be natural","to listen to the 'longpress' custom event on the element:"," ","\f80 Custom Events with Actions"," However, there’s no native event called 'longpress'."," But no worries, we can create a custom 'longpress' event."," Luckily, we have the longPress action to determine when the user long-presses a button. What"," we need to create a custom 'longpress' event is to trigger the 'longpress' event after"," determining the user has long-pressed a button in our longPress action. And so, in the line of"," code where we have determined that the user is long-pressing the button, we can dispatch a custom"," event from the button:"," node.dispatchEvent(new CustomEvent('longpress'));"," Let’s listen to the longpress custom event and create the custom 'longpress' event in our
action. Here is the final code:","
{/* do something */}} />"," Putting it in words, we add a longPress action to the button element, which adds logic to determine
whether the button is being long-pressed. When the button is long-pressed, the longPress action
dispatches a custom event called 'longpress' on the element. To react to and trigger specific
behaviors when the custom 'longpress' event is dispatched on the element, we can listen to the
event by using on:longpress with an event handler.
\f Example – validating form inputs with custom events 81","It may feel like a roundabout way to call a function from an action by dispatching an event, but doing
it this way has a few pros:"," • Whether we listen to the 'longpress' event on the button or not, the action could still
dispatch the 'longpress' custom event. So, with this approach, we don’t need to check
whether the handler is defined or not.
• Listening to the 'longpress' event using on: instead of passing the function directly into
the action would mean that you could use other Svelte features that come with Svelte’s on:
directive. For example, to only listen to the 'longpress' event once, you can use the |once
event modifier, for example, on:longpress|once.","Another way of describing what we have done with the longPress action is that the longPress
action enhances the button element and provides a new event, 'longpress', that can be listened
to on the element.
Now that we’ve learned how we can define Svelte actions, and how we can use actions to create new
events that we can listen to on an element, let’s look at a few more examples that use this technique.","Example – validating form inputs with custom events","The example that we are going to explore is using actions to validate form inputs.","When you add an input element to your form, you can add attributes such as required, minlength,","and min to indicate that the input value has to pass the constraint validation or else would be","considered invalid.","However, by default, such a validation check is only done during form submission. There’s no real-","time feedback on whether your input is valid as you type.","To make the input element validate as you type, we need to add an 'input' event listener (which will","be called on every keystroke as we type in the input element) and call input.checkValidity()","to validate the input. Now, let’s do just that:"," event.target.checkValidity()} />","As you call the checkValidity() method, if the input is indeed invalid, then it will trigger the
'invalid' event:","   on:input={(event) => event.target.checkValidity()}
  on:invalid={(event) => console.log(event.target.validity)}
/>","Unfortunately, there’s no 'valid' event. So, there’s no way to tell whether the input has passed
the validation.
\f82 Custom Events with Actions"," It would be great if there were an event called 'validate' in which within the event details, we
can tell whether the input is valid or not. If it isn’t, we could get an error message about why the input
is invalid.
Here’s an example of how we could use the 'validate' event:"," {","   if (event.detail.isValid) {","     errorMessage = '';","   } else {","     errorMessage = event.detail.errorMessage;","   }"," }} />"," There isn’t an event called 'validate', but we can create this custom event ourselves. So, why not
create an action to create this custom event for us?
This logic is well suited to be written as an action for the following reasons:"," • It can be reused in other input elements."," • This logic itself is not involved in creating or updating elements. If it were, it would probably"," be better suited to be a component."," Let us write this action:"," 1. Firstly, this action involves listening to the 'input' event listener. So, in the code, as shown,
we are going to add an event listener at the start of the action and remove the 'input' event
listener in the destroy method. This means that whenever an element that uses this action
is added to the DOM, it will listen to the 'input' event, and when it is removed from the
DOM, the event listener will be automatically removed:
function validateOnType(input) {
  input.addEventListener('input', onInput);
  return {
    destroy() {
      input.removeEventListener('input', onInput);
    }
  };
}"," 2. Next, within the input handler, we are going to call checkValidity() to check whether
the input is valid. If the input is invalid, then we will read input.validity and determine
the error message:
function validateOnType(input) {
  function onInput() {
\f Example – validating form inputs with custom events 83","     const isValid = input.checkValidity();","     const errorMessage = isValid ? '' : getErrorMessage(input."," validity);","   }","   // ..."," }","3. Finally, we will dispatch the custom 'validate' event with isValid and errorMessage
as event details:
function validateOnType(input) {
  function onInput() {
// ...
input.dispatchEvent(
  new CustomEvent(
    'validate',
    { detail: { isValid, errorMessage } }
  )
);
  }
  // ...
}","4. Now, with this action completed, we can enhance the element by adding a new
'validate' event, which will be called as-you-type, letting you know whether the input
is currently valid or invalid. It will also show the corresponding error message:
  use:validateOnType
  on:validate={(event) => console.log(event.detail)}
/>"," To make the validation results more useful to the user, you can use the result from the
'validate' event to modify element styles, such as setting the input border color to red
when the validation result is invalid, as shown in the following snippet:

  class:invalid={!isValid}
  use:validateOnType
  on:validate={(event) => isValid = event.detail.isValid}
/>
"," Are you getting the hang of writing actions that create custom events?
Let’s try the next one as an exercise. This time, we’ll tackle one of the most common user interactions,
drag and drop."," Exercise – creating a drag-and-drop event"," A drag-and-drop behavior means clicking on an element, moving the mouse while holding down the"," click to drag the element across the screen to the desired location, and then releasing the mouse click"," to drop the element in the new location."," A drag-and-drop behavior thus involves coordination between multiple events, namely, 'mousedown',"," 'mousemove', and 'mouseup'."," As we perform the drag-and-drop motion, what we are interested in knowing is when the dragging"," starts, how far the element is dragged, and when the dragging ends."," These three events can be translated into three custom events: 'dragStart', 'dragMove',"," and 'dragEnd'."," Let us try to implement the drag-and-drop behavior as an action that will create these three custom events:","   use:dragAndDrop
  on:dragStart={...}
  on:dragMove={...}
  on:dragEnd={...}
/>"," You can check the answer to this exercise here: https://github.com/PacktPublishing/"," Real-World-Svelte/tree/main/Chapter05/03-drag-and-drop."," Summary
In this chapter, we saw how to define an action. We talked about one of the common patterns of
actions, which is to create custom events. This allows us to encapsulate DOM event logic into custom
events and reuse them across elements.
In the next chapter, we will look at the next common pattern of actions, which is integrating
third-party libraries.
\f 6
Integrating Libraries
with Actions
There are a lot of JavaScript UI libraries out there on the internet. However, at the time of writing this
book, Svelte is relatively new. Not all the UI libraries out there are written using Svelte and written
specifically for Svelte. But that does not mean that we can’t use them in our Svelte component.
There are many ways to integrate third-party JavaScript UI libraries into Svelte. In this chapter, we are
going to explore how we can do it using Svelte actions.
We will start by integrating an imaginary UI library, slowly building up our case for why Svelte actions
are suitable for the job. Along the way, I will explain how to use Svelte actions for different scenarios
and show you where Svelte actions fall short. I’ll discuss my reasonings and personal opinions on
when to choose Svelte actions and when to choose otherwise.
Following that, I will show you some real-world UI library examples. After that, we will explore
integrating UI libraries written in other frameworks, such as React and Vue, with a few more examples.
By the end of this chapter, you will see that you are not limited to using only UI libraries written in
Svelte in your Svelte application – you can reuse any UI library that is available on the internet.
This chapter covers the following topics:"," • Integrating JavaScript UI libraries into Svelte"," • Why we should use actions to integrate UI libraries and other alternatives"," • Integrating UI libraries written in other frameworks into Svelte","Technical requirements
You can find the examples and code for this chapter here: https://github.com/
PacktPublishing/Real-World-Svelte/tree/main/Chapter06.
\f86 Integrating Libraries with Actions"," Integrating vanilla JavaScript UI libraries into Svelte"," First, we will explore UI libraries that are written in vanilla JavaScript. When we use the phrase vanilla"," JavaScript, we’re referring to plain JavaScript, or JavaScript in the absence of frameworks or libraries."," There are many reasons a UI library is written in vanilla JavaScript:"," • Performance reasons – it would be much easier to optimize without the abstractions from the
web framework
• The library author’s personal preference to be framework-agnostic
• The library was created predating any modern web frameworks"," For us, vanilla JavaScript UI libraries are great because they do not depend on any specific framework
runtime, which is an extra overhead on top of the UI library itself.
For example, if we use a calendar component library that is implemented in React, then besides
installing the calendar component library, we would need to install React’s framework as well.
This additional dependency leads to an increased bundle size and potential conflicts with Svelte.
Therefore, when using a component library in Svelte, it is generally preferable to choose a library that
does not depend on any specific framework.
Now that we have learned about why vanilla JavaScript UI libraries are great, let’s discuss how we can
integrate them into Svelte. In this chapter, we will explore integrating libraries in Svelte using Svelte
actions, which begs the question, why do we choose to use Svelte actions to integrate a UI library?"," Why use Svelte actions to integrate a UI library?"," In the previous chapter, we explored how Svelte actions are useful for adding custom event handlers."," At the same time, Svelte actions function as element-level life cycle functions, making them very useful"," for interfacing with third-party libraries. Now, let’s explore why that’s the case."," Let’s take a calendar component library as an example. For simplicity’s sake and to not get bogged"," down by implementation details, let’s imagine the library as an imaginary library instead of using any"," real-life calendar component library. This allows us to focus on the general problem itself rather than"," the specific library’s implementation details."," We will look at some real-life UI libraries afterward."," To decide where the calendar component will be added to the DOM, component libraries usually"," require us to specify a container element to house the library component."," For example, here, ImaginaryCalendar requires us to pass the container element as part of the"," argument for the constructor:"," new ImaginaryCalendar({ container: containerElement })","\f Integrating vanilla JavaScript UI libraries into Svelte 87","To get the reference to an element in Svelte, we can use bind:this:","

","The containerElement variable is only updated with the reference of the element after the element
is mounted, so it can only be referred to in onMount:"," ","Note that we are keeping the reference of the calendar instance because we can use it to call calendar
methods to get or set values:"," calendar.setDate(date);","Also, we call calendar.cleanup() when the component unmounts for cleanup purposes."," Note
The calendar library is just an imaginary example. However, most UI libraries will provide
similar APIs or methods to retrieve or modify the component instance’s internal state and
clean up when they are no longer in use.","We need to be extra careful when we are using the calendar instance here. We want to avoid
referencing the calendar instance before it is initialized to prevent encountering a reference error
as the calendar instance is only declared and initialized after onMount:
To be safe, we should check whether the calendar instance is defined before calling any of its
methods. In the code example below, we verify if the calendar instance is defined, before calling the
calendar.setDate() method."," if (calendar) calendar.setDate(date).
\f88 Integrating Libraries with Actions"," The need for this extra level of cautiousness is more obvious when the calendar is conditionally created:","
{#if someCondition}
  

{/if}"," In the preceding code, you can see that

is conditionally created based on someCondition.
That is why in onMount, we need to check whether containerElement is available before we create
ImaginaryCalendar using containerElement as a container. And the calendar instance
is only available if ImaginaryCalendar is created, and thus only available if someCondition
is true.
The preceding code illustrates one of the many possibilities where the calendar instance could
be undefined.
One thing to note about this code is that it is not behaviorally correct as it does not attempt to create
ImaginaryCalendar whenever the value of someCondition turns from false to true and
does cleanup whenever it turns back to false.
This is where Svelte actions shine."," Using Svelte actions"," By altering the preceding code so that it uses actions, you will see that we do not need the extra check"," to ensure that containerElement is available before instantiating ImaginaryCalendar."," The following code shows how such an action could be implemented. Here, the name of our Svelte"," action is calendar:","
{#if someCondition}
  

{/if}","This is because, when using Svelte actions, the action function will only be called with the reference
of the element whenever the element is created and mounted onto the DOM.
When the condition changes to such that the

element is removed from the DOM, the destroy
method of the action will be called to clean things up.
Using Svelte actions, we can now create as many ImaginaryCalendar instances as we want within
one component by adding the action to different HTML elements:
To prove my point, in the code snippet below, in addition to the original

element you’ve seen
in the previous examples, I am adding another

element and three more

elements
using the {#each} block. I am then applying the calendar actions to all four

elements to
create four more calendars, and we do not run into any errors for having multiple calendars at once."," {#if someCondition}
  

{/if}

{#each [1, 2, 3] as item}
  

{/each}","If we were to use bind:this and onMount, we would have to repeat ourselves multiple
times by declaring multiple containerElement variables multiple times and instantiate
ImaginaryCalendar with each of the containerElement variables.
Now, with the calendar instance encapsulated within the action, how should we call the calendar
instance method to update the calendar state from the outside?
That’s what the action data is for!
\f90 Integrating Libraries with Actions"," Adding data to Svelte actions"," In the previous section, we created a calendar action and instantiated an ImaginaryCalendar"," instance inside that action. If we want to call an instance method of ImaginaryCalendar outside"," the calendar action, such as calling calendar.setDate(date) to set the date of the calendar,"," what should we do?"," There’s no way to call calendar.setDate(date) outside of the calendar action since the"," calendar instance is defined within the calendar action. A workaround is to pass date through"," action data – that is, we can provide date as action data and call calendar.setDate(date)"," with the date passed in."," For example, in the following code snippet, we’re passing date into the calendar action:","

"," In the calendar action, we call calendar.setDate(date) with the date passed in. In addition
to that, we define an update method in the action, such that whenever the date that’s passed to the
calendar action changes, Svelte will call calendar.setDate(date):","

"," Here, we can pass different dates to the different calendar instances:"," {#each dates as date}","   

"," {/each}"," That’s great!
\f Integrating vanilla JavaScript UI libraries into Svelte 91","Now, what if you want to call a different calendar instance method, such as calendar.setMode(),
whenever the mode changes?
You could pass both date and mode into the action:","

","In that case, the calendar action would need to handle both date and mode:"," function calendar(node, { date, mode }) {","   const calendar = new ImaginaryCalendar({ container: containerElement"," });","   calendar.setDate(date);","   calendar.setMode(mode);","   return {","     update({ date: newDate, mode: newMode }) {","       calendar.setDate(newDate);","       calendar.setMode(newMode);","     },","     destroy() { ... }","The update method of the calendar action will be called whenever either date or mode
changes. This means that in the preceding code, we are calling both calendar.setDate() and
calendar.setMode() whenever either date or mode changes. This may not have any apparent
consequences, but we may be doing an unnecessary job.
A workaround for this is to keep track and always check whether date or mode have changed in the
update method. This is how we can do this:","       if (date !== newDate) {","         calendar.setDate(newDate);","         date = newDate;","       }","       if (mode !== newMode) {","         calendar.setMode(newMode);","         mode = newMode;","\f92 Integrating Libraries with Actions","     },
    destroy() { ... }
  };
}"," In the preceding code, we’re checking if newDate is different from the current date, and if it is
different, then we call the calendar.setDate() method and update our current reference of
date. We do a similar thing for mode.
This works. However, as you can see, it is more code and more complex than what we set up when we
first started creating the calendar action.
And what if you want to call a calendar instance method that is not tied to any data, such
as calendar.refreshDates()?
This is where using actions falls short."," An alternative to Svelte actions"," Remember the previous example, where we used bind:this and onMount to initialize Imag-"," inaryCalendar?"," We said that the approach is not versatile enough and falls short if we need to do the following:"," • Conditionally render the container and create ImaginaryCalendar"," • Have multiple calendars within the same component"," These drawbacks are all true, but there’s a use case in which it is perfectly fine to use bind:this
and onMount to initialize ImaginaryCalendar. This is when the conditions we mentioned
previously are never true:"," • We do not need to conditionally render the container"," • We do not need to have multiple calendar instances within the same component (this is not"," exactly true, but we will come back to this)"," I’m not sure whether you are thinking the same thing right now but allow me to cut the suspense.
This is when we want to use ImaginaryCalendar as a Svelte component.
Within the ImaginaryCalendar Svelte component itself, we will have only one container element,
and it will always be available:"," ","   import { onMount } from 'svelte';","   let containerElement;","\f Integrating vanilla JavaScript UI libraries into Svelte 93","   let calendar;
  onMount(() => { ... });

","You can then use this component conditionally:","
{#if someCondition}
  
{/if}","Alternatively, you can use it however many times as you like:"," {#if someCondition}
  
{/if}


{#each array as item}
  
{/each}","Here, we’ve swapped out elements using actions,

, with Svelte components,
.
This is perfectly normal.
In the previous chapter, we contemplated between abstracting logic through components or abstracting
through actions.
In this scenario, we are looking at abstracting the logic of instantiating a UI library using an element
as a container, and we could abstract it into a Svelte action or a Svelte component.
Both options are equally fine.
Both options are designed for this.
So, which options should you choose? Let’s find out.","Choosing between Svelte actions and Svelte components","Here are my personal preferences when faced with a choice between either option.","\f94 Integrating Libraries with Actions"," Choose a Svelte action to integrate with a UI library when you are looking for an option that provides
the following aspects:"," • Is more lightweight. A Svelte component has slightly more overhead compared to a Svelte action.
• Only passes zero to one data into the UI library component instance. If you were to pass in
two or more pieces of data into the action, then the update method of the action will be called
whenever any part of the data changes."," You should choose a Svelte component if you are looking for an option that provides the following:"," • Allows more room for optimization and finer control"," • Allows you to call UI library component instance methods directly"," • Allows you to pass in child content into the UI component library"," We did not discuss this much, but integrating a UI library as a component opens up the possibility
of passing additional content into the UI component library through slots:","
  
  
    {date}
  
"," If you are interested in learning more about slots and how to compose components in Svelte, read
Chapter 4, where we explored this topic extensively. Now that we’ve covered how we could integrate
a UI library using Svelte actions, why we should use Svelte actions, as well as the alternatives and
considerations, let’s take a look at a real-world example, Tippy.js."," Example – integrating Tippy.js"," Tippy.js is a tooltip, popover, dropdown, and menu library."," I do not have any affiliation with the Tippy.js library, and the reason I chose Tippy.js as an example"," is purely by chance. Nonetheless, Tippy.js has a nice and simple API, making it a good candidate for"," an example."," First, let’s look at the Tippy.js documentation: https://atomiks.github.io/tippyjs/."," After installing the tippy.js library using a package manager of our choice, we can then import"," Tippy.js into our code:"," import tippy from 'tippy.js';"," import 'tippy.js/dist/tippy.css';","\f Example – integrating Tippy.js 95","Now, we can initialize tippy with the following constructor function:"," tippy('#id');
tippy(document.getElementById('my-element'));","Here, we pass in the element where Tippy.js should provide a tooltip.","You can specify any customizations of the tooltip's content through the data attributes of the element,","which Tippy.js will pick up as it initializes:","

","Alternatively, you can pass this in the constructor:"," tippy(element, { content: 'hello' });","To update the content after initialization, call the Tippy.js setContent method:"," tooltipInstance.setContent(\"bye\");","To permanently destroy and clean up the tooltip instance, Tippy.js provides the destroy method:"," tooltipInstance.destroy();","Here, we have all we need to create a tippy action. We have methods to do the following:"," • Create the tippy tooltip – tippy(…)"," • Clean up the tippy tooltip – tooltipInstance.destroy()"," • Update the tippy tooltip – tooltipInstance.setContent(…)","Let’s take a look at what the tippy action should look like.
Here is how I want it to look:","

","In the preceding code snippet, we applied our tippy action to a

element. The content within
the tooltip created by Tippy.js should be passed into the tippy action as action data, represented by
the tooltipContent variable. Whenever tooltipContent changes, the action should react
to it and update the tooltip.
So, let’s write our tippy Svelte action. Here’s the scaffolding for an action:"," function tippy(element, content) {","   // TODO #1: initialize the library","     update(newContent) {","\f96 Integrating Libraries with Actions","       // TODO #2: do something when action data changes","     destroy() {","       // TODO #3: clean up","     }"," As you can see, we created the tippy action based on the Svelte action contract: a function that
returns an object with destroy and update methods.
I left three TODOs in the code, each marking different stages of Svelte actions. Let’s go through each
one and fill them up. The first TODO is where the action will be called after the element is created and
mounted onto the DOM. Here, we are given the element the action is applied to and the action data,
and we should use it to initialize the Tippy.js tooltip:"," // TODO #1: initialize the library"," const tooltipInstance = tippy(element, { content });"," The second TODO is inside the update method. This method will be called every time the action data
changes. Here, we need to call the Tippy.js tooltip instance to reflect the dataset in the Svelte component:"," // TODO #2: do something when action data changes"," tooltipInstance.setContent(newContent);"," The third TODO is inside the destroy method. This method will be called after the element is
removed from the DOM. Here, we need to do the cleanup on the Tippy.js tooltip instance that we
created in the action:"," // TODO #3: clean up"," tooltipInstance.destroy();"," And that’s it – we now have a working tippy action that integrates the Tippy.js tooltip and will show
a tooltip with customizable content whenever we hover over the element.
You can find the complete code on GitHub: https://github.com/PacktPublishing/
Practical-Svelte/tree/main/Chapter06/01-tippy.
Let’s look at one more example, through which I want to show you one more thing that you can do
with actions when integrating with UI libraries.
The UI library we are going to look at next is CodeMirror.
\f Example – integrating CodeMirror 97","Example – integrating CodeMirror","CodeMirror is a code editor component that has many great features for editing, such as syntax","highlighting, code folding, and more.","You can find the CodeMirror documentation at https://codemirror.net/.","At the time of writing, CodeMirror is currently at version 5.65.9.","After installing the codemirror library using the package manager of our choice, we can import","codemirror into our code:"," import CodeMirror from 'codemirror';"," import 'codemirror/lib/codemirror.css';","Now, we can initialize CodeMirror with the following constructor function:"," const myCodeMirror = CodeMirror(document.body);","Here, we pass in the element where we want the CodeMirror code editor to be.
Before I continue, at this point, note that we are looking for the same set of things from CodeMirror:"," • Methods to initialize CodeMirror"," • Any method needed to clean up a CodeMirror instance"," • Any method to update a CodeMirror instance","I am going to leave it to you to complete the checklist and figure it out.","However, allow me to draw your attention to one particular API from the CodeMirror instance:"," myCodeMirror.on('change', () => { ... });","The on method from CodeMirror allows the CodeMirror instance to listen to events and react to them.
So, if we want to add event listeners to the CodeMirror instance from outside of the action, how
should we do it?
In the previous chapter, we saw that we could create custom events on the element using actions.
This means that we can allow users to listen to the 'change' event from the element that uses the
codemirror action:","

","\f98 Integrating Libraries with Actions"," To make this happen, you can dispatch an event from within the action:"," function codemirror(element) {
  const editor = CodeMirror(element);
  editor.on('change', () => {
    // trigger 'change' event on the element
    // whenever the editor changes
    element.dispatchEvent(new CustomEvent('change'));
  });
}"," Remember to check whether you need to clean up or unlisten to any event in the destroy method
so that you don’t cause any unwanted behavior.
And that’s it!
The rest of the action is left to you as an exercise.
You can find the complete code on GitHub: https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter06/02-codemirror.
In this section, we learned how to integrate vanilla UI libraries into Svelte. However, not all UI libraries
are implemented independently without any framework. Sometimes, the library you are looking for
might be implemented in a different framework, such as React or Vue. In such cases, how can you
integrate them into a Svelte application? That is what we will explore next."," Using UI libraries written in other frameworks"," It is not impossible to use components from other frameworks in Svelte."," However, doing so will introduce the framework’s runtime and other overheads that come along with"," the framework. The runtime usually includes code to handle reactivity and normalize browser APIs"," and events. Each framework usually ships its own code for this logic and does not share it with other"," frameworks. The runtime for React version 18.2.0 weighs 6.4 kB when minified, which is additional"," code you need to include when you want to use a React component within Svelte."," So, this is not recommended unless it is necessary."," The reason this section has been included in this book is more for educational purposes and to"," demonstrate that this is possible, as well as what needs to be done to make it happen."," Creating components in various frameworks"," Each framework usually provides an API that takes in a container element and the framework"," component as the root of the application.","\f Using UI libraries written in other frameworks 99","In this section, we’re going to take a look at React and Vue, the two most popular JavaScript frameworks
at the time of writing.
For example, in React 18, we use the createRoot API:"," import ReactDOM from 'react-dom';"," const root = ReactDOM.createRoot(container);"," root.render(

Hello, world

);","The preceding code uses the JSX syntax, which is not part of the standard JavaScript language syntax.
It is syntactic sugar for jsx:"," import { jsx } from 'react/jsx-runtime';"," const root = ReactDOM.createRoot(container)"," root.render(jsx('h1', { children: 'Hello, world' }));","You will have to write the preceding code if you do not configure any transpiling process in your code
to turn JSX syntax into valid JavaScript.
On the other hand, In Vue 3, there is the createApp API:"," import { createApp } from 'vue'"," import App from './App.vue'"," const app = createApp(App);"," app.mount(container);","The Vue framework uses the word application in the documentation, mentioning that the createApp
method is used to create a new application instance. The word application is aptly used as integrating
component libraries written in other frameworks is very much like starting a new sub-application
within our Svelte application.
You may have also started to notice the similarity between the APIs from these frameworks and other
UI libraries that we’ve seen so far – all of them take in a container element so that they know where
to render the content or apply changes.
Similar to integrating UI libraries using actions, after figuring out what APIs we can use to render a
component within a container element, the next thing we must check is whether there is any API to
clean up when it is no longer needed.","Cleaning up the components in various frameworks","Depending on the underlying framework of the component library, different APIs are provided to","clean up whenever the component is no longer needed.","In React, there is a method for this called unmount:"," root.unmount();
\f100 Integrating Libraries with Actions"," In Vue, it is also called unmount:"," app.unmount();"," The next thing we need to check is whether there is any API to pass data into our component and
API so that they can be updated later."," Updating the component in various frameworks"," Similar to different frameworks having different APIs for cleaning up, frameworks provide different"," APIs to update the component with new data."," If you are familiar with React, you can pass data into a React component through props:"," "," This is akin to the props in a Svelte component."," The preceding code desugars into the following:"," jsx(Component, { prop_name: value });"," To update the props of the component, React allows us to call root.render again with the same
component but with a different prop value:"," root.render(jsx(App, { prop_name: 123 }));"," // some time later"," root.render(jsx(App, { prop_name: 456 }));"," React will reconcile this internally and figure out how to update the DOM.
On the other hand, in Vue, you can pass data through props in the createApp API:"," const app = createApp(Component, { prop_name: value });"," However, to the best of my knowledge, there’s no straightforward way to update the props’ value from
the outside.
However, you could use a Vue Composition API, such as ref(), to create a reactive and mutable
ref. With that, you could modify the ref instead of updating the props directly:"," const value = ref(123);"," const app = createApp(Component, { prop_name: ref });"," value.value = 456;"," It is okay if you are not familiar with how React and Vue work. This book is for Svelte, not for React or Vue.
\f Integrating react-calendar into Svelte 101","The most important thing to take home from this is that when integrating a UI library, whether it is
in vanilla JavaScript, React, or Vue, we look for three things:"," • A method to create the component with a container element"," • A method to clean up the component instance"," • A method to pass in data and update the data","If you are familiar with a framework, you will be able to figure out a way to do all these things.
With that out of the way, let’s take a look at a real-world example, where we will integrate a React
calendar library, react-calendar.","Integrating react-calendar into Svelte","The react-calendar library is a calendar component library written in React.","You can read more about it here: https://projects.wojtekmaj.pl/react-calendar/.","The react-calendar library takes in various props for customization purposes. But for demonstration","purposes, we are only going to focus on two props, value and onChange, which allow us to control","the selected date of the library.","We pass the selected date through a prop named value. The onChange prop, on the other hand, is","used to pass in an event handler that will be called when the value changes from within the calendar","component. We saw how we could handle event handlers in a UI library in the previous section when","we discussed CodeMirror.","So, here is what I think using the calendar action would look like:","   use:calendar={selectedDate}
  on:change={(event) => selectedDate = event.detail}
/>","Here, event.detail is the data attached to the custom 'change' event, which would be the date
value that’s sent from the react-calendar component through the onChange props.
Now that we know what our calendar action would look like, let’s write the action out.
Again, here’s the scaffolding of an action:"," function calendar(element, date) {","   // TODO #1: render the react-calendar into the element","   // TODO #2: the onChange handler to dispatch a new custom event","     update(newDate) {","\f102 Integrating Libraries with Actions","       // TODO #3: re-render the calendar again when there's a new date
value
    },
    destroy() {
      // TODO #4: clean up
    }
  };
}"," Here, I’ve created a basic code structure of a Svelte action and left a few TODOs within the code. The
first two TODOs are to set up the calendar instance with the element the action is applied to. The
third TODO is to handle when a new date is passed into the action, and the last TODO is to clean up
when the element is removed from the DOM.
So, let’s fill up the TODOs.
For the first TODO, let’s create a React root and render our react-calendar component:"," import { jsx } from 'react/jsx-runtime';"," import ReactDOM from 'react-dom';"," function calendar(element, date) {","   # TODO #1: render the react-calendar into the element","   const app = ReactDOM.createRoot(element);","   app.render(jsx(Calendar, { value: date, onChange }));"," Here, we passed in onChange, which we have not defined yet."," Let’s do this now:"," # TODO #2: the onChange handler to dispatch a new custom event"," function onChange(value) {","   element.dispatchEvent(","   new CustomEvent('change', { detail: value })","   );"," In the preceding code snippet, whenever onChange is called, we will dispatch a new custom event,
with value passed in as the detail for the custom event.
The third TODO is the content for the update method. Whenever a new date value is passed in from
the action, we will re-render the Calendar component again:"," // TODO #3: re-render the calendar again when there's a new date value
app.render(jsx(Calendar, { value: newDate, onChange }));
\f Summary 103","In the last TODO, in the destroy method, we unmount our Calendar component:"," // TODO #4: clean up"," app.unmount();","And that’s it.","You can find the complete code on GitHub: https://github.com/PacktPublishing/","Real-World-Svelte/tree/main/Chapter06/03-react-calendar.","With that, you’ve written a Svelte action that integrates a component library from a different framework,","React, into Svelte, and you set up and updated the component’s value in a controlled manner.","Summary
In this chapter, we learned how to use actions to integrate UI libraries, either written in vanilla JavaScript
or any other frameworks into Svelte. We went through two real-world examples – integrating Tippy.js
and react-calendar into Svelte using Svelte actions. In both examples, we went through a step-
by-step process of writing out a Svelte action. We started by creating the structure of a Svelte action
and then filled up the steps within the action for when the Svelte action is initialized as the element is
created, when the data changes, and when the element is removed from the DOM. We also discussed
why we choose to use Svelte actions, as well as the other alternatives and considerations when it comes
to integrating UI libraries.
In the next chapter, we will look at the next common pattern of actions, which is to progressively
enhance your elements.
\f\f 7
Progressive Enhancement
with Actions
Progressive enhancement is a design philosophy in web development that emphasizes providing content
and core functionality to everyone while delivering an enhanced experience to users who can afford it.
In this chapter, we will start with a more in-depth discussion of what progressive enhancement is.
There are many ways to achieve progressive enhancement in your application; we will explore one of
them by using Svelte actions. I will explain my reasoning as to why I think Svelte actions are designed
for this use case.
Toward the end of the chapter, we will go through a few examples of using Svelte actions to progressively
enhance our application.
By the end of this chapter, you will be able to build an application that follows the principles of
progressive enhancement and supports as many user devices as possible.
This chapter covers the following topics:"," • What is progressive enhancement?"," • Why use Svelte actions for progressive enhancement?"," • Examples of progressive enhancements using Svelte actions","Technical requirements
You can find the examples and code for this chapter here: https://github.com/
PacktPublishing/Real-World-Svelte/tree/main/Chapter07.
\f106 Progressive Enhancement with Actions"," What is progressive enhancement?"," The most important thing in progressive enhancement is missing from the phrase itself. Something"," that is implied here is where are we progressively enhancing from."," The main idea of progressive enhancement is to provide a great baseline of essential content and"," core functionality to everyone, regardless of the browser software, device hardware, or the quality"," of the internet connection. Older browser software may not support newer JavaScript syntaxes and"," CSS features; older device hardware may take up more time to process and render your web page; a"," slower internet connection may take longer to load the resources needed to display your web page."," How do we ensure our web page stays usable for as many users as possible? Think about this for a"," while—I will come back to it later."," For users who can afford better browsers, more powerful hardware, and higher internet bandwidth,"," we progressively provide an enhanced experience to them. We leverage the power of JavaScript and"," CSS to surprise and delight our users."," How do we differentiate between the users and decide when to deliver an enhanced experience?"," There is another term that is often used to compare with progressive enhancement, and that is"," graceful degradation. Graceful degradation starts from a feature-rich baseline and gracefully handles"," the situation when the user browser can no longer support the feature by replacing it with a simpler"," alternate experience. A lot of the time, these features start with a more complex assumption, and so"," execution-wise, it is much harder to gracefully degrade to all kinds of users."," Progressive enhancement, on the other hand, starts from a baseline that works for most users and"," slowly works its way up by adding more features. So, we can be sure that when a new feature is not"," loaded or does not work, users will still have a basic working web page."," So, let’s get back to our question: How do we make sure that our web page is usable to all users? We’ll"," uncover that in the following section."," Progressively enhancing the web experience"," One of the ways is to make sure that we follow the standards. HTML, CSS, and JavaScript are the"," main languages of the web. We make sure we only use language features that are part of the standard"," specifications. Features that have been part of the specifications for longer have a higher probability"," of being implemented by all browsers. The latest and hottest features are less likely to be available in"," all browsers."," So, build your web pages with standard HTML, CSS, and JavaScript."," This then leads to the next question: How do we deliver a differentiating experience to users based on"," their browser, device, and network capabilities?","\f What is progressive enhancement? 107","There are many ways to do this.","One key idea for most approaches is to build your application layer by layer. Start with the first","layer of core functionality and make sure everything works. Then, add subsequent layers to enhance","the experience.","Building web pages in layers","One example of building web pages in layers is to build the base content and functionality in HTML","as the base layer, then add styles, transitions, and animations using CSS as the next layer. Finally,","complex interactivity is added using JavaScript as the final layer.","This aligns with how the browser loads your website.","Whenever a user visits your website, the first thing the browser downloads from your website is HTML.","HTML describes your content. HTML tags such as

,

, and describe how your","content should be laid out on the screen. HTML tags such as , , and describe","how users can submit data and interact with your content.","With HTML, your website should already provide basic content and functionality to the user.","And if you have within the HTML, with reference to external","CSS files, the browser will then make separate requests to download the CSS resource and parse and","apply the CSS styles to your document. This will, in turn, enhance the look and feel of the default","browser styles. HTML can offer basic layouts, but with CSS, you can have advanced layouts, such as","flex layouts, grid layouts, and so on.","On the other hand, if you have
Hello","Regardless of how the preview action is implemented when rendering the preceding code on the
server side, Svelte will generate the following HTML:"," Hello","This is because Svelte actions never run on the server side.","\f110 Progressive Enhancement with Actions"," As soon as your user receives the HTML response, they can start clicking on the link and navigating
to new locations. You now have a workable application with just the HTML.
Depending on the user’s network condition, your JavaScript code, compiled from your Svelte
component, may take a longer time to arrive. But that does not stop the user from using the hyperlinks
to navigate away.
Only when the JavaScript is loaded and executed does Svelte run the preview action with the
element that is on the DOM and enhances the element’s behavior.
The point here is that, as much as possible, we make the core functionality of our application workable
with just the HTML, and we add a layer of enhancement with JavaScript, which may come at a much
later time depending on the user’s network condition.
Enough with the progressive enhancement philosophy. Let’s look at how we can implement this
preview action.
We want the preview action to show a floating popup containing the content of the link when we
move our mouse cursor over the link and hide it as we move our mouse cursor away.
We can achieve that with 'mouseover' and 'mouseout' events. Here’s how:"," function preview(element) {
  element.addEventListener('mouseover', onMouseOver);
  element.addEventListener('mouseout', onMouseOut);
  return {
    destroy() {
          element.removeEventListener('mouseover', onMouseOver);
          element.removeEventListener('mouseout', onMouseOut);
        }
  };
}"," In the preceding code snippet, we add 'mouseover' and 'mouseout' event listeners at the
beginning of the preview action. Additionally, we ensure proper cleanup by removing both the
'mouseover' and 'mouseout' event listeners in the destroy method. Before we figure out
how to implement onMouseOver and onMouseOut, we need to first decide what the floating
popup would look like and how would we lay it out in the DOM.
To show the content of the link target, we are going to use the

Open navigation menu

Close suggestionsSucheSuche

enChange Language

Hochladen

Laden...

Laden...

User Settings

close menu

Welcome to Scribd!

• Hochladen
• Read for free
• FAQ and support
• Language (EN)
• Anmelden

Skip carousel

• What is Scribd?
• Academic
• Professional
• Culture
• Hobbies & Crafts
• Personal Growth
• All Documents

• Academic Documents

  • Foreign Language Studies
    • Chinese
    • ESL
  • Science & Mathematics
    • Astronomy & Space Sciences
    • Biology
  • Study Aids & Test Prep
    • Book Notes
    • College Entrance Exams

  • Teaching Methods & Materials
    • Early Childhood Education
    • Education Philosophy & Theory

  All categories
• Professional Documents

  • Business
    • Business Analytics
    • Human Resources & Personnel Management
  • Career & Growth
    • Karriere
    • Job Hunting
  • Computers
    • Applications & Software
    • CAD-CAM

  • Finance & Money Management
    • Accounting & Bookkeeping
    • Auditing
  • Law
    • Business & Financial
    • Contracts & Agreements
  • Politics
    • American Government
    • International Relations
  • Technology & Engineering
    • Automotive
    • Aviation & Aeronautics

  All categories
• Culture Documents

  • Kunst
    • Antiques & Collectibles
    • Architecture
  • Biography & Memoir
    • Artists and Musicians
    • Entertainers and the Rich & Famous
  • Comics & Graphic Novels
  • History
    • Ancient
    • Modern
  • Philosophy

  • Language Arts & Discipline
    • Composition & Creative Writing
    • Linguistics
  • Literary Criticism
  • Social Science
    • Anthropology
    • Archaeology
  • True Crime

  All categories
• Hobbies & Crafts Documents

  • Cooking, Food & Wine
    • Beverages
    • Courses & Dishes
  • Games & Activities
    • Card Games
    • Fantasy Sports
  • Home & Garden
    • Crafts & Hobbies
    • Gardening

  • Sports & Recreation
    • Baseball
    • Basketball

  All categories
• Personal Growth Documents

  • Lifestyle
    • Beauty & Grooming
    • Fashion
  • Religion & Spirituality
    • Buddhism
    • Christianity
  • Self-Improvement
    • Addiction
    • Mental Health

  • Wellness
    • Body, Mind, & Spirit
    • Diet & Nutrition

  All categories

Tan Li Hau - Real-World Svelte - Supercharge Your Apps With Svelte 4 by Mastering Advanced Web Development Concepts-Packt (2023)

Uploaded by

AntonioMolina

0 ratings0% found this document useful (0 votes)

4K views

282 pages

AI-enhanced title and description

Document Information

click to expand document information

The document is about a book titled "Real-World Svelte" that teaches advanced web development concepts for supercharging apps with Svelte 4, written by Tan Li Hau. It provides information on the book's author, publisher, copyright details and is dedicated to the author's loved ones for their support. The foreword promotes the author as an excellent maintainer and educator within the Svelte community.

Original Description:

Real-World Svelte

Original Title

Tan Li Hau - Real-World Svelte_ Supercharge your apps with Svelte 4 by mastering advanced web development concepts-Packt (2023)

Copyright

© © All Rights Reserved

Available Formats

PDF, TXT or read online from Scribd

Share this document

Share or Embed Document

Sharing Options

• Share on Facebook, opens a new window
  Facebook
• Share on Twitter, opens a new window
  Twitter
• Share on LinkedIn, opens a new window
  LinkedIn
• Share with Email, opens mail client
  E-Mail
• Copy link
  Copy link

Did you find this document useful?

0%0% found this document useful, Mark this document as useful

0%0% found this document not useful, Mark this document as not useful

Is this content inappropriate?

Report this Document

The document is about a book titled "Real-World Svelte" that teaches advanced web development concepts for supercharging apps with Svelte 4, written by Tan Li Hau. It provides information on the book's author, publisher, copyright details and is dedicated to the author's loved ones for their support. The foreword promotes the author as an excellent maintainer and educator within the Svelte community.

Copyright:

© All Rights Reserved

Available Formats

Download as PDF, TXT or read online from Scribd

Flag for inappropriate content

Download now

Download as pdf or txt

SaveSave Tan Li Hau - Real-World Svelte_ Supercharge your a... For Later

0 ratings0% found this document useful (0 votes)

4K views282 pages

Tan Li Hau - Real-World Svelte - Supercharge Your Apps With Svelte 4 by Mastering Advanced Web Development Concepts-Packt (2023)

Uploaded by

AntonioMolina

AI-enhanced title and description

The document is about a book titled "Real-World Svelte" that teaches advanced web development concepts for supercharging apps with Svelte 4, written by Tan Li Hau. It provides information on the book's author, publisher, copyright details and is dedicated to the author's loved ones for their support. The foreword promotes the author as an excellent maintainer and educator within the Svelte community.

Copyright:

© All Rights Reserved

Available Formats

Download as PDF, TXT or read online from Scribd

Flag for inappropriate content

SaveSave Tan Li Hau - Real-World Svelte_ Supercharge your a... For Later

0%0% found this document useful, Mark this document as useful

0%0% found this document not useful, Mark this document as not useful

Embed

Aktie

Print

Download now

Download as pdf or txt

Jump to Page

You are on page 1of 282

Search inside document

Real-World Svelte

Supercharge your apps with Svelte 4 by mastering advanced

web development concepts

Tan Li Hau

BIRMINGHAM—MUMBAI
Real-World Svelte
Copyright © 2023 Packt Publishing
All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted
in any form or by any means, without the prior written permission of the publisher, except in the case
of brief quotations embedded in critical articles or reviews.
Every effort has been made in the preparation of this book to ensure the accuracy of the information
presented. However, the information contained in this book is sold without warranty, either express
or implied. Neither the author, nor Packt Publishing or its dealers and distributors, will be held liable
for any damages caused or alleged to have been caused directly or indirectly by this book.
Packt Publishing has endeavored to provide trademark information about all of the companies and
products mentioned in this book by the appropriate use of capitals. However, Packt Publishing cannot
guarantee the accuracy of this information.

Group Product Manager: Rohit Rajkumar

Publishing Product Manager: Bhavya Rao
Book Project Manager: Aishwarya Mohan
Content Development Editor: Debolina Acharyya
Technical Editor: K Bimala Singha
Copy Editor: Safis Editing
Proofreader: Safis Editing
Indexer: Hemangini Bari
Production Designer: Ponraj Dhandapani
DevRel Marketing Coordinators: Nivedita Pandey, Namita Velgekar, and Anamika Singh

First published: November 2023

Production reference: 1011123

Published by Packt Publishing Ltd.

Grosvenor House
11 St Paul’s Square
Birmingham
B3 1RB, UK

ISBN 978-1-80461-603-1

www.packtpub.com
To my beloved wife, Chai Ying, whose love and belief in me have been my anchor, and to my parents,
Eng Swee and Sook Hwa, who have always been my guiding stars – this book is dedicated to you, my
pillars of strength and inspiration.
– Tan Li Hau
 Foreword
One of the hardest parts of maintaining an open source project is finding people to maintain it with.
It’s not enough for someone to be a good programmer, though that is certainly an essential ingredient.
They must also be gifted at communication (working in public means constantly explaining yourself
to an unseen audience), empathetic (understanding users’ frustrations and needs is a large part of a
maintainer’s responsibility), patient (on a large-enough project, seemingly simple changes can take
months to land), farsighted (anticipating how apparently unrelated things might interact with each
other in future is very important), and aligned with the values and goals of the wider community.
Above all, they must be fun to work with, otherwise what’s the point?
Such people are exceedingly rare.
Tan Li Hau epitomizes these rare qualities. His contributions to the Svelte project are wide-ranging
and significant — only two people have made more code commits to the core repository — and he has
developed elegant solutions for some of the thorniest problems. Everyone who uses the framework
has benefitted from his work.
And yet that’s not what Li Hau is best known for, because in addition to being an excellent steward of
the code base, he is a widely followed educator, whose “hello, hello” catchphrase introduces countless
informative videos on every aspect of Svelte and other related tools.
As such, I can’t think of anyone better qualified to write the book that you’re now reading. Within
these pages, you’ll learn everything you need to build Svelte apps of all different shapes and sizes and
gain an expert-level understanding of the underlying concepts. You’ve made a great choice, and we’re
very excited to have you as part of the Svelte community. Welcome!
Rich Harris, creator of Svelte
October 2023
 Contributors

About the author

Tan Li Hau is a frontend developer at Shopee and a core maintainer of Svelte. He has delivered multiple
conference talks and workshops on Svelte. Passionate about sharing his knowledge, Li Hau regularly
contributes to the community through blog posts, YouTube videos, and books. He aspires to inspire
others to explore and learn about Svelte and other modern web technologies.

I want to thank the people who have been close to me and supported me, especially my wife, Chai
Ying, my parents, Eng Swee and Sook Hwa, and my siblings, Li Khai and Li Huey.
About the reviewer
Shareej V K has over 14 years of experience in frontend, Node.js, and full stack development and
leadership. He currently works as an educator as well as a consultant, focusing on full stack development
training and SvelteKit consulting.
He is a frontend and full stack framework specialist, experienced in many of the frontend frameworks,
including Angular, React, Vue, and Svelte, and the full stack meta-frameworks SvelteKit and Next.js.
He has a master’s degree in psychology and believes tech educators can do more if they understand
the psychological aspects of learning. According to him, learning is a process that demands time and
perseverance. He loves exploring and implementing effective methods for technical training.
 Table of Contents
Prefacexiii

Part 1: Writing Svelte Components

1
Lifecycles in Svelte 3
Technical requirements 3 Which component to register? 8
Understanding the Svelte lifecycle Reusing lifecycle functions in Svelte
functions4 components9
Initializing the component 5 Exercise 1 – Update counter 12
Mounting the component 5
Updating the component 6
Composing lifecycle functions into
Destroying the component 6
reusable hooks 13
Coordinating lifecycle functions
The one rule for calling lifecycle across components 15
functions7
Exercise 2 – Scroll blocker 16
Refactoring lifecycle functions 7
Summary16

2
Implementing Styling and Theming 17
Technical requirements 17 Adding the class attribute 21
Styling Svelte components in six Simplifying the class attribute with the class:
different ways 18 directive21
Applying styles from external CSS files 22
Styling with the style attribute 18
Choosing which method to style Svelte
Using style: directives 18
components24
Adding the <style> block 19
viii Table of Contents

Styling Svelte with Tailwind CSS 25 Example – implementing a dark/light theme

Setting up Tailwind CSS 25 mode30
Allowing users to change the styles of a
Theming Svelte components with component31
CSS custom properties 27 Aligning the fallback value 33
Defining CSS custom properties 28
Summary34

3
Managing Props and State 35
Technical requirements 36 One-way versus two-way data flow 41
Defining props and state 36 Deriving states from props with a
Defining props 36 reactive declaration 44
Defining state 37 Managing complex derived states 46
Props versus state 37 Updating props using derived states 49
Understanding bindings 39 Summary52

4
Composing Components 53
Technical requirements 53 Rendering different HTML element
Manipulating how a child component or component types 59
looks54 Creating recursive components for
Passing dynamic content through slots 56 recursive data 62
Providing default slot content 57 Example – a JSON tree viewer 64
Having multiple slots with named slots 57 The Container/Presentational pattern 67
Passing data through slot props 58 Summary68
 Table of Contents ix

Part 2: Actions 71
5
Custom Events with Actions 73
Technical requirements 74 Passing parameters to an action 78
Defining actions 74 Example – validating form inputs
Reusing DOM event logic with with custom events 81
custom events 75 Exercise – creating a drag-and-drop
Encapsulating logic into a component 76 event84
Encapsulating logic into an action 77 Summary84

6
Integrating Libraries with Actions 85
Technical requirements 85 Example – integrating CodeMirror 97
Integrating vanilla JavaScript UI Using UI libraries written in other
libraries into Svelte 86 frameworks98
Why use Svelte actions to integrate a UI library? 86 Creating components in various frameworks 98
Using Svelte actions 88 Cleaning up the components in various
Adding data to Svelte actions 90 frameworks99
An alternative to Svelte actions 92 Updating the component in various
Choosing between Svelte actions and Svelte frameworks100
components93
Integrating react-calendar into Svelte 101
Example – integrating Tippy.js 94 Summary103

7
Progressive Enhancement with Actions 105
Technical requirements 105 Svelte actions for progressive enhancements 108
What is progressive enhancement? 106 Example – previewing a link with a
Progressively enhancing the web experience 106 <a> element 109
Building web pages in layers 107
x Table of Contents

Example – progressively enhancing a Implementing the enhance action 115

form114
Summary117
The default form behavior 114

Part 3: Context and Stores

8
Context versus Stores 121
Defining Svelte context 121 Creating Svelte stores with built-in functions 130
Using an object as a context key 124 Auto-subscribing to a Svelte store 131
Changing the context value 124 Choosing between a Svelte context
Defining the Svelte store 126 and a Svelte store 133
Using the observer pattern 126 Passing dynamic context using a Svelte store 136
Defining the subscribe function 127 Summary137
Defining a Svelte store 128

9
Implementing Custom Stores 139
Technical requirements 140 Debouncing store value changes 149
Creating a Svelte store from user Subscribing and unsubscribing original store
events140 on demand 152
Deriving a new Svelte store with the derived()
Ensuring the event listener is only added once 141
function153
Exercise144
Using the derived method 155
Creating an undo/redo store 145 Exercise156
Implementing the undo/redo logic 147
Summary157
Exercise148

Creating a debounced higher-order

Svelte store 149
 Table of Contents xi

10
State Management with Svelte Stores 159
Technical requirements 159 Tip 3 – derive a smaller state object from a
bigger state object 165
Managing states with Svelte stores 159
Tip 1 – simplify complex state updates with Using state management libraries
unidirectional data flow 163 with Svelte 166
Tip 2 – prefer smaller self-contained state Example – using Valtio as a Svelte store 167
objects over one big state object 163 Exercise – transforming an XState state
machine into a Svelte store 173

Summary173

11
Renderless Components 175
Technical requirements 175 Exercise 1 – a renderless autocomplete
component181
What are renderless components? 176
Exploring reusable renderless Turning a declarative description
components177 into imperative instructions 182
Example – building a renderless carousel Writing declarative Canvas components 185
component178 Exercise 2 – expanding shape components 189
Writing a renderless carousel component 180
Summary190

12
Stores and Animations 191
Technical requirements 191 Exercise – creating an animating line graph 197
Introducing the tweened and spring Creating custom tweened
stores191 interpolators198
Using the tweened and spring stores 193 Examples – creating an animated
Using tweened and spring stores with arrays image preview 200
and objects 194
Summary204
Examples – creating an animated
graph with the tweened and spring
stores195
xii Table of Contents

Part 4: Transitions
13
Using Transitions 207
Technical requirements 208 The global modifier 215
Adding transitions to elements 208 How Svelte transition works under
Svelte’s built-in transitions 209 the hood 217
Customizing a transition 209 Creating animations with CSS 217
When are the transitions played? 211 Creating animations with JavaScript 218
Animating transitions in Svelte 219
Handling mixed transition and static elements 213
Other Svelte logical blocks for transitions 213 Summary221

14
Exploring Custom Transitions 223
Technical requirements 223 Writing a custom CSS transition
The transition contract 223 using the css function 229
The css function 225 Writing a custom JavaScript
The tick function 228 transition using the tick function 237
Summary243

15
Accessibility with Transitions 245
Technical requirements 245 Reducing motion for Svelte transition248
What is web accessibility? 246 Having alternative transitions for
Understanding user preference with inaccessible users 250
prefers-reduced-motion246 Summary252

Index253

Other Books You May Enjoy 260

 Preface
In today’s digital age, web development is ever-evolving, with new tools and frameworks emerging
almost daily. Among them, Svelte stands out, being voted in the top place as the most beloved framework
in a recent developer survey (https://survey.stackoverflow.co/2022/#section-
most-loved-dreaded-and-wanted-web-frameworks-and-technologies). Svelte
brings a fresh perspective by optimizing performance and providing an intuitive design, as well as
being fully featured.
While other resources out there cover Svelte’s vast features, this book offers a unique lens and approach.
We’ll delve deep into key Svelte features, demystifying core concepts. Through hands-on, real-world
examples, we’ll not only teach the “how” but also the “why” behind each approach. By understanding
the underlying principles and thought processes, you’ll be equipped to seamlessly integrate what you
learn into your Svelte projects.

Who this book is for

This book is tailored for web developers and software engineers who possess a foundational understanding
of JavaScript, CSS, and general web development practices. Whether you're new to Svelte and are eager
to dive deep or have dabbled with its basics but seek to elevate your expertise through real-world
examples and patterns, this guide is for you. Beyond mere instruction, the content delves into the 'why'
and 'how' behind each concept, ensuring readers not only grasp the material but can also effectively
apply it in professional contexts and diverse projects. If you're ready to not just learn about Svelte but
to master its intricacies and practical applications, then this book is your next essential read.

What this book covers

Chapter 1, Lifecycles in Svelte, provides an overview of Svelte’s lifecycles, their respective functions, the
rules for invoking them, and strategies to reuse and compose these lifecycle functions.
Chapter 2, Implementing Styling and Theming, dives into six unique methods to style Svelte components.
You will also learn the essentials of theming Svelte components, from defining themes to enabling
user customization.
Chapter 3, Managing Props and State, deepens your understanding of props and state within Svelte.
This chapter demystifies props, state, and bindings, and discusses the distinctions between one-way
and two-way data bindings. It also showcases deriving state from props.
xiv Preface

Chapter 4, Composing Components, provides techniques to control content within child components
from their parent components. You will explore the <slot> element and various Svelte special
elements, such as <svelte:self> and <svelte:fragment>.
Chapter 5, Custom Events with Actions, kickstarts your exploration of Svelte actions over three chapters.
This chapter starts by exploring the idea of creating custom events using Svelte actions.
Chapter 6, Integrating Libraries with Actions, provides a hands-on guide to integrating a third-party
JavaScript library into Svelte using actions.
Chapter 7, Progressive Enhancements with Svelte Actions, unpacks the concept of progressive enhancements
and helps you understand how Svelte actions can be leveraged to progressively enhance your
Svelte application.
Chapter 8, Context versus Stores, delves into Svelte context and stores. You’ll learn how and when to
use Svelte context and stores.
Chapter 9, Implementing Custom Stores, teaches you how to implement custom Svelte stores through
a practical step-by-step guide, going through multiple real-world examples along the way.
Chapter 10, State Management with Svelte Stores, arms you with practical tips on managing application
state in Svelte applications. You will also learn how to use third-party state management libraries in Svelte.
Chapter 11, Renderless Components, explores the concept of the renderless component, a type of
reusable component that does not render any HTML elements of its own. We will systematically go
through implementing such a component.
Chapter 12, Stores and Animations, explores the built-in tweened and spring stores. You’ll learn
how to apply them in your Svelte application, and how to customize the interpolation for these
animating stores.
Chapter 13, Using Transitions, provides a comprehensive understanding of transitions in Svelte. You’ll
learn how to use transitions in Svelte, when and how transitions are played, and how they work under
the hood.
Chapter 14, Exploring Custom Transitions, explores the idea of writing a custom transition in Svelte.
You’ll learn about the Svelte transition contract, and with practical examples, you’ll be guided step by
step through creating a custom transition in Svelte.
Chapter 15, Accessibility with Transitions, sheds light on accessibility considerations in transitions,
particularly for users with vestibular disorders. You’ll gain insights into crafting responsible transitions
that respect user preferences and cater to all.
 Preface xv

To get the most out of this book

You will need to have basic knowledge of web development and a basic understanding of JavaScript,
CSS, and HTML.

Software/hardware covered in the book Operating system requirements

Svelte 4 Windows, macOS, or Linux
JavaScript

If you are using the digital version of this book, we advise you to type the code yourself or access
the code from the book’s GitHub repository (a link is available in the next section). Doing so will
help you avoid any potential errors related to the copying and pasting of code.

Download the example code files

You can download the example code files for this book from GitHub at https://github.com/
PacktPublishing/Real-World-Svelte. If there’s an update to the code, it will be updated
in the GitHub repository.
We also have other code bundles from our rich catalog of books and videos available at https://
github.com/PacktPublishing/. Check them out!

Conventions used
There are a number of text conventions used throughout this book.
Code in text: Indicates code words in text, database table names, folder names, filenames, file
extensions, pathnames, dummy URLs, user input, and Twitter handles. Here is an example: “Mount
the downloaded WebStorm-10*.dmg disk image file as another disk in your system.”
A block of code is set as follows:

const folder = [
  { type: 'file', name: 'a.js' },
  { type: 'file', name: 'b.js' },
  { type: 'folder', name: 'c', children: [
    { type: 'file', name: 'd.js' },
  ]},
];
xvi Preface

When we wish to draw your attention to a particular part of a code block, the relevant lines or items
are set in bold:

<!-- filename: JsonTree.svelte -->

<script>
  export let data;
</script>
<ul>
  {#each Object.entries(data) as [key, value]}
    <li>
      {key}:
      {#if typeof value === 'object'}
        <svelte:self data={value} />
      {:else}
        {value}
      {/if}
    <li>
  {/each}
</ul>

Bold: Indicates a new term, an important word, or words that you see onscreen. For instance, words
in menus or dialog boxes appear in bold. Here is an example: “When you have a <form> element,
by default when you hit the Submit button, it will navigate to the location indicated by the action
attribute, carrying along with it the value filled in the <input> elements within the <form> element.”

Tips or important notes

Appear like this.

Get in touch
Feedback from our readers is always welcome.
General feedback: If you have questions about any aspect of this book, email us at customercare@
packtpub.com and mention the book title in the subject of your message.
Errata: Although we have taken every care to ensure the accuracy of our content, mistakes do happen.
If you have found a mistake in this book, we would be grateful if you would report this to us. Please
visit www.packtpub.com/support/errata and fill in the form.
Piracy: If you come across any illegal copies of our works in any form on the internet, we would
be grateful if you would provide us with the location address or website name. Please contact us at
[email protected] with a link to the material.
 xvii

If you are interested in becoming an author: If there is a topic that you have expertise in and you
are interested in either writing or contributing to a book, please visit authors.packtpub.com.

Share Your Thoughts

Once you’ve read Real-World Svelte, we’d love to hear your thoughts! Please click here to go straight
to the Amazon review page for this book and share your feedback.
Your review is important to us and the tech community and will help us make sure we’re delivering
excellent quality content.
xviii

Download a free PDF copy of this book

Thanks for purchasing this book!
Do you like to read on the go but are unable to carry your print books everywhere?
Is your eBook purchase not compatible with the device of your choice?
Don’t worry, now with every Packt book you get a DRM-free PDF version of that book at no cost.
Read anywhere, any place, on any device. Search, copy, and paste code from your favorite technical
books directly into your application.
The perks don’t stop there, you can get exclusive access to discounts, newsletters, and great free content
in your inbox daily
Follow these simple steps to get the benefits:

1. Scan the QR code or visit the link below

https://packt.link/free-ebook/9781804616031

2. Submit your proof of purchase

3. That’s it! We’ll send your free PDF and other benefits to your email directly
 Part 1: Writing Svelte
Components

In this section, we will lay the foundation for writing Svelte components. We will kick things off by
delving into the lifecycles of Svelte components. Then, we will learn how to style and theme our Svelte
components. After that, we will explore the intricacies of data passing between components and finally
wrap up with techniques to compose components into a cohesive Svelte application.
This part has the following chapters:

• Chapter 1, Lifecycles in Svelte

• Chapter 2, Implementing Styling and Theming
• Chapter 3, Managing Props and State
• Chapter 4, Composing Components
 1
Lifecycles in Svelte
Svelte is a frontend framework. You can use Svelte to build websites and web applications. A Svelte
application is made up of components. You write a Svelte component within a file with .svelte
extension. Each .svelte file is one Svelte component.
When you create and use a Svelte component, the component goes through various stages of the
component lifecycle. Svelte provides lifecycle functions, allowing you to hook into the different stages
of the component.
In this chapter, we will start by talking about the various lifecycles and the lifecycle functions in
Svelte. With a clear idea of lifecycles in mind, you will then learn the basic rule of using the lifecycle
functions. This is essential, as you will see that this understanding will allow us to use the lifecycle
functions in a lot of creative ways.
This chapter contains sections on the following topics:

• What are Svelte lifecycle functions?

• The rule of calling lifecycle functions
• How to reuse and compose lifecycle functions

Technical requirements
Writing Svelte applications is very easy and does not require any paid tools. Despite the added value
of most paid tools, we decided to use only free tools to make the content of this book available to you
without any limitations.
4 Lifecycles in Svelte

You will require the following:

• Visual Studio Code as the integrated development environment (https://code.

visualstudio.com/)
• A decent web browser (Chrome, Firefox, or Edge, for instance)
• Node.js as the JavaScript runtime environment (https://nodejs.org/)

All the code examples for this chapter can be found on GitHub at: https://github.com/
PacktPublishing/Real-World-Svelte/tree/main/Chapter01
Code for all chapters can be found at https://github.com/PacktPublishing/Real-
World-Svelte.

Understanding the Svelte lifecycle functions

When using a Svelte component, it goes through different stages throughout its lifetime: mounting,
updating, and destroying. This is similar to a human being. We go through various stages in our lifetime,
such as birth, growth, old age, and death, throughout our lifetime. We call the different stages lifecycles.
Before we talk about lifecycles in Svelte, let’s look at a Svelte component.

<script>
  import { onMount, beforeUpdate, afterUpdate, onDestroy } from
'svelte';
  let count = 0;
  onMount(() => { console.log('onMount!'); });
  beforeUpdate(() => { console.log('beforeUpdate!'); });
  afterUpdate(() => { console.log('afterUpdate!'); });
  onDestroy(() => { console.log('onDestroy!'); });
</script>
<button on:click={() => { count ++; }}>
  Counter: {count}
</button>

Can you tell me when each part of the code is executed?

Not every part of the code is executed at once; different parts of the code are executed at different
stages of the component lifecycle.
A Svelte component has four different lifecycle stages: initializing, mounting, updating, and destroying.
 Understanding the Svelte lifecycle functions 5

Initializing the component

When you create a component, the component first goes through the initialization phase. You can
think of this as the setup phase, where the component sets up its internal state.
This is where lines 2–7 are being executed.
The count variable is declared and initialized. The onMount, beforeUpdate, afterUpdate,
and onDestroy lifecycle functions are called, with callback functions passed in, to register them at
the specific stages of the component lifecycles.
After the component is initialized, Svelte starts to create elements in the template, in this case, a
<button> element and text elements for "Counter: " and {count}.

Mounting the component

After all the elements are created, Svelte will insert them in order into the Document Object Model
(DOM). This is called the mounting phase, where elements are mounted onto the DOM.
If you add Svelte actions to an element, then the actions are called with the element:

<script>
  function action(node) {}
</script>
<div use:action>

We will explore Svelte actions in more depth in Chapter 5 to 7.

If and when you add event listeners to the element, this is when Svelte will attach the event listeners
to the element.
In the case of the preceding example, Svelte attaches the click event listener onto the button after
it is inserted into the DOM.
When we add bindings to an element, the bound variable gets updated with values from the element:

<script>
  let element;
</script>
<div bind:this={element} />

This is when the element variable gets updated with the reference to the <div> element created
by Svelte.
If and when you add transitions to an element, this is when the transitions are initialized and start playing.
6 Lifecycles in Svelte

The following snippet is an example of adding a transition to an element. You can add a transition to
an element using the transition:, in:, and out: directives. We will explore more about Svelte
transitions in Chapter 13 to 15:

<div in:fade />

After all the directives, use: (actions), on: (event listeners), bind: bindings, in:, transition:
(transitions), are processed, the mounting phase comes to an end by calling all the functions registered
in the onMount lifecycle functions.
This is when the function on line 4 is executed, and you will see "onMount!" printed in the logs.

Updating the component

When you click on the button, the click event listener is called. The function on line 9 is executed.
The count variable is incremented.
Right before Svelte modifies the DOM based on the latest value of the count variable, the functions
registered in the beforeUpdate lifecycle function are called.
The function on line 5 is executed, and you will see the text "beforeUpdate!" printed in the logs.
At this point, if you attempt to retrieve the text content within the button, it would still be "Counter:
0".
Svelte then proceeds to modify the DOM, updating the text content of the button to "Counter: 1".
After updating all the elements within the component, Svelte calls all the functions registered in the
afterUpdate lifecycle function.
The function on line 6 is executed, and you will see the text "afterUpdate!" printed in the logs.
If you click on the button again, Svelte will go through another cycle of beforeUpdate, and then
update the DOM elements, and then afterUpdate.

Destroying the component

A component that is conditionally shown to a user will remain while the condition holds; when the
condition no longer holds, Svelte will proceed to destroy the component.
Let’s say the component in our example now enters the destroy stage.
Svelte calls all the functions registered in the onDestroy lifecycle function. The function on line 7
is executed, and you will see the text "onDestroy!" printed in the logs.
After that, Svelte removes the elements from the DOM.
 The one rule for calling lifecycle functions 7

Svelte then cleans up the directives if necessary, such as removing the event listeners and calling the
destroy method from the action.
And that’s it! If you try to recreate the component again, a new cycle starts again.
The Svelte component lifecycle starts with initializing, mounting, updating, and destroying. Svelte
provides lifecycle methods, allowing you to run functions at different stages of the component.
Since the component lifecycle functions are just functions exported from 'svelte', can you import
and use them anywhere? Are there any rules or constraints when importing and using them?
Let’s find out.

The one rule for calling lifecycle functions

The only rule for calling component lifecycle functions is that you should call them during component
initialization. If no component is being initialized, Svelte will complain by throwing an error.
Let’s look at the following example:

<script>
  import { onMount } from 'svelte';
  function buttonClicked() {
    onMount(() => console.log('onMount!'));
  }
</script>
<button on:click={buttonClicked} />

When you click on the button, it will call buttonClicked, which will call onMount. As no
component is being initialized when onMount is being called, (the component above has initialized
and mounted by the time you click on the button), Svelte throws an error:

Error: Function called outside component initialization

Yes, Svelte does not allow lifecycle functions to be called outside of the component initialization phase.
This rule dictates when you can call the lifecycle functions. What it does not dictate is where or how
you call the lifecycle functions. This allows us to refactor lifecycle functions and call them in other ways.

Refactoring lifecycle functions

If you look carefully at the rule for calling lifecycle functions, you will notice that it is about when you
call them, and not where you call them.
It is not necessary to call lifecycle functions at the top level within the <script> tag.
8 Lifecycles in Svelte

In the following example, the setup function is called during component initialization, and in turn
calls the onMount function:

<script>
  import { onMount } from 'svelte';
  setup();
  function setup() {
    onMount(() => console.log('onMount!'));
  }
</script>

Since the component is still initializing, this is perfectly fine.

It is also not necessary to import the onMount function within the component. As you see in the
following example, you can import it in another file; as long as the onMount function is called during
component initialization, it is perfectly fine:

// file-a.js
import { onMount } from 'svelte';
export function setup() {
  onMount(() => console.log('onMount!'));
}

In the preceding code snippet, we’ve moved the setup function we defined previously to a new
module called file-a.js. Then, in the original Svelte component, rather than defining the setup
function, we import it from file-a.js, shown in the following code snippet:

<script>
  import { setup } from './file-a.js';
  setup();
</script>

Since the setup function calls the onMount function, the same rule applies to the setup function
too! You can no longer call the setup function outside component initialization.

Which component to register?

Looking at just the setup function, you may be wondering, when you call the onMount function,
how does Svelte know which component’s lifecycle you are referring to?
Internally, Svelte keeps track of which component is initializing. When you call the lifecycle functions,
it will register your function to the lifecycle of the component that is being initialized.
 Reusing lifecycle functions in Svelte components 9

So, the same setup function can be called within different components and registers the onMount
function for different components.
This unlocks the first pattern in this chapter: reusing lifecycle functions.

Reusing lifecycle functions in Svelte components

In the previous section, we learned that we can extract the calling of lifecycle functions into a function
and reuse the function in other components.
Let’s look at an example. In this example, after the component is added to the screen for 5 seconds, it will
call the showPopup function. I want to reuse this logic of calling showPopup in other components:

<script>
  import { onMount } from 'svelte';
  import { showPopup } from './popup';
  onMount(() => {
    const timeoutId = setTimeout(() => {
      showPopup();
    }, 5000);
    return () => clearTimeout(timeoutId);
  });
</script>

Here, I can extract the logic into a function, showPopupOnMount:

// popup-on-mount.js
import { onMount } from 'svelte';
import { showPopup } from './popup';
export function showPopupOnMount() {
  onMount(() => {
    const timeoutId = setTimeout(() => {
      showPopup();
    }, 5000);
    return () => clearTimeout(timeoutId);
  });
}

And now, I can import this function and reuse it in any component:

<script>
  import { showPopupOnMount } from './popup-on-mount';
  showPopupOnMount();
</script>
10 Lifecycles in Svelte

You may be wondering, why not only extract the callback function and reuse that instead?

// popup-on-mount.js
import { showPopup } from './popup';
export function showPopupOnMount() {
  const timeoutId = setTimeout(() => {
    showPopup();
  }, 5000);
  return () => clearTimeout(timeoutId);
}

Over here, we extract only setTimeout and clearTimeout logic into showPopupOnMount,
and pass the function into onMount:

<script>
  import { onMount } from 'svelte';
  import { showPopupOnMount } from './popup-on-mount';
  onMount(showPopupOnMount);
</script>

In my opinion, the second approach of refactoring and reusing is not as good as the first approach.
There are a few pros in extracting the entire calling of the lifecycle functions into a function, as it
allows you to do much more than you can otherwise:

• You can pass in different input parameters to your lifecycle functions.

Let’s say you wish to allow different components to customize the duration before showing the
popup. It is much easier to pass that in this way:
<script>
  import { showPopupOnMount } from './popup-on-mount';
  showPopupOnMount(2000); // change it to 2s
</script>

• You can return values from the function.

Let’s say you want to return the timeoutId used in the onMount function so that you can
cancel it if the user clicks on any button within the component.
It is near impossible to do so if you just reuse the callback function, as the value returned from
the callback function will be used to register for the onDestroy lifecycle function:
<script>
  import { showPopupOnMount } from './popup-on-mount';
  const timeoutId = showPopupOnMount(2000);
</script>
<button on:click={() => clearTimeout(timeoutId)} />
 Reusing lifecycle functions in Svelte components 11

See how easy it is to implement it to return anything if we write it this way:

// popup-on-mount.js
export function showPopupOnMount(duration) {
  let timeoutId;
  onMount(() => {
    timeoutId = setTimeout(() => {
      showPopup();
    }, duration ?? 5000);
    return () => clearTimeout(timeoutId);
  });
  return timeoutId;
}

• You can encapsulate more logic along with the lifecycle functions.
Sometimes, the code in your lifecycle functions callback function does not work in a silo; it
interacts with and modifies other variables. To reuse lifecycle functions like this, you must
encapsulate those variables and logic into a reusable function.
To illustrate this, let’s look at a new example.
Here, I have a counter that starts counting when a component is added to the screen:
<script>
  import { onMount } from 'svelte';
  let counter = 0;
  onMount(() => {
    const intervalId = setInterval(() => counter++, 1000);
    return () => clearInterval(intervalId);
  });
</script>
<span>{counter}</span>

The counter variable is coupled with the onMount lifecycle functions; to reuse this logic,
the counter variable and the onMount function should be extracted together into a
reusable function:
import { writable } from 'svelte/store';
import { onMount } from 'svelte';
export function startCounterOnMount() {
  const counter = writable(0);
  onMount(() => {
    const intervalId = setInterval(() => counter.update($counter
=> $counter + 1), 1000);
    return () => clearInterval(intervalId);
  });
  return counter;
}
12 Lifecycles in Svelte

In this example, we use a writable Svelte store to make the counter variable reactive. We
will delve more into Svelte stores in Part 3 of this book.
For now, all you need to understand is that a Svelte store allows Svelte to track changes in a
variable across modules, and you can subscribe to and retrieve the value of the store by prefixing
a $ in front of a Svelte store variable. For example, if you have a Svelte store named counter,
then to get the value of the Svelte store, you would need to use the $counter variable.
Now, we can use the startCounterOnMount function in any Svelte component:
<script>
  import { startCounterOnMount } from './counter';
  const counter = startCounterOnMount();
</script>
<span>{$counter}</span>

I hope I’ve convinced you about the pros of extracting the calling of lifecycle functions into a function.
Let’s try it out in an example.

Exercise 1 – Update counter

In the following example code, I want to know how many times the component has gone through
the update cycle.
Using the fact that every time the component goes through the update cycle, the afterUpdate callback
function will be called, I created a counter that will be incremented every time the afterUpdate
callback function is called.
To help us measure only the update count of a certain user operation, we have functions to start
measuring and stop measuring, so the update counter is only incremented when we are measuring:

<script>
  import { afterUpdate } from 'svelte';

  let updateCount = 0;
  let measuring = false;
  afterUpdate(() => {
    if (measuring) {
      updateCount ++;
    }
  });
  function startMeasuring() {
    updateCount = 0;
    measuring = true;
  }
  function stopMeasuring() {
 Composing lifecycle functions into reusable hooks 13

    measuring = false;
  }
</script>
<button on:click={startMeasuring}>Measure</button>
<button on:click={stopMeasuring}>Stop</button>
<span>Updated {updateCount} times</span>

To reuse all the logic of the counter: – the counting of update cycles and the starting and stopping
of the measurement – we should move all of it into a function, which ends up looking like this:

<script>
  import { createUpdateCounter } from './update-counter';
  const { updateCount, startMeasuring, stopMeasuring } =
createUpdateCounter();
</script>
<button on:click={startMeasuring}>Measure</button>
<button on:click={stopMeasuring}>Stop</button>
<span>Updated {$updateCount} times</span>

The update counter returns an object that contains the updateCount variable and the
startMeasuring and stopMeasuring functions.
The implementation of the createUpdateCounter function is left as an exercise to you, and
you can check the answer at https://github.com/PacktPublishing/Real-World-
Svelte/tree/main/Chapter01/01-update-counter.
We’ve learned how to extract a lifecycle function and reuse it, so let’s take it up a notch and reuse
multiple lifecycle functions in the next pattern: composing lifecycle functions.

Composing lifecycle functions into reusable hooks

So far, we’ve mainly talked about reusing one lifecycle function. However, there’s nothing stopping us
from grouping multiple lifecycle functions to perform a function.
Here’s an excerpt from the example at https://svelte.dev/examples/update. The example
shows a list of messages. When new messages are added to the list, the container will automatically
scroll to the bottom to show the new message. In the code snippet, we see that this automatic scrolling
behavior is achieved by using a combination of beforeUpdate and afterUpdate:

<script>
  import { beforeUpdate, afterUpdate } from 'svelte';
  let div;
  let autoscroll;
  beforeUpdate(() => {
    autoscroll = div && (div.offsetHeight + div.scrollTop) > (div.
14 Lifecycles in Svelte

scrollHeight - 20);
  });
  afterUpdate(() => {
    if (autoscroll) div.scrollTo(0, div.scrollHeight);
  });
</script>
<div bind:this={div} />

To reuse this autoscroll logic in other components, we can extract the beforeUpdate and
afterUpdate logic together into a new function:

export function setupAutoscroll() {

  let div;
  let autoscroll;
  beforeUpdate(() => {
    autoscroll = div && (div.offsetHeight + div.scrollTop) > (div.
scrollHeight - 20);
  });
  afterUpdate(() => {
    if (autoscroll) div.scrollTo(0, div.scrollHeight);
  });
  return {
  setDiv(_div) {
  div = _div;
    },
  };
}

We can then use the extracted function, setupAutoScroll, in any component:

<script>
  import { setupAutoscroll } from './autoscroll';
  const { setDiv } = setupAutoscroll();
  let div;
  $: setDiv(div);
</script>
<div bind:this={div} />

In the refactored setupAutoscroll function, we return a setDiv function to allow us to update

the reference of the div used within the setupAutoscroll function.
As you’ve seen, by adhering to the one rule of calling lifecycle functions during component initialization,
you can compose multiple lifecycle functions into reusable hooks. What you’ve learned so far is sufficient
for composing lifecycle functions, but there are more alternatives on the horizon. In the upcoming
chapters, you’ll explore Svelte actions in Chapter 5 and the Svelte store in Chapter 8, expanding your
options further. Here’s a sneak peek at some of these alternatives.
 Coordinating lifecycle functions across components 15

An alternative implementation could be to make div a writable store and return it from the
setupAutoscroll function. This way, we could bind to the div writable store directly instead
of having to call setDiv manually.
Alternatively, we could return a function that follows the Svelte action contract and use the action
on the div:

export function setupAutoscroll() {

  let div;
  // ...
  return function (node) {
    div = node;
    return {
      destroy() {
        div = undefined;
      },
    };
  };
}

setupAutoscroll now returns an action, and we use the action on our div container:

<script>
  import { setupAutoscroll } from './autoscroll';
  const autoscroll = setupAutoscroll();
</script>
<div use:autoscroll />

We will discuss the Svelte action contract in more detail later in the book.
We’ve seen how we can extract lifecycle functions into a separate file and reuse it in multiple Svelte
components. Currently, the components call the lifecycle functions independently and function as
standalone units. Is it possible to synchronize or coordinate actions across components that uses the
same lifecycle functions? Let’s find out.

Coordinating lifecycle functions across components

As we reuse the same function across components, we can keep track globally of the components that
use the same lifecycle function.
Let me show you an example. Here, I would like to keep track of how many components on the screen
are using our lifecycle function.
16 Lifecycles in Svelte

To count the number of components, we can define a module-level variable and update it within our
lifecycle function:

import { onMount, onDestroy } from 'svelte';

import { writable } from 'svelte/store';
let counter = writable(0);
export function setupGlobalCounter() {
  onMount(() => counter.update($counter => $counter + 1));
  onDestroy(() => counter.update($counter => $counter - 1));
  return counter;
}

As the counter variable is declared outside the setupGlobalCounter function, the same
counter variable instance is used and shared across all the components.
When any component is mounted, it will increment the counter, and any component that is referring
to the counter will get updated with the latest counter value.
This pattern is extremely useful when you want to set up a shared communication channel between
components and tear it down in onDestroy when the component is being destroyed.
Let’s try to use this technique in our next exercise.

Exercise 2 – Scroll blocker

Usually, when you add a pop-up component onto the screen, you want the document to not be
scrollable so that the user focuses on the popup and only scrolls within the popup.
This can be done by setting the overflow CSS property of the body to "hidden".
Write a reusable function used by pop-up components that disables scrolling when the pop-up component
is mounted. Restore the initial overflow property value when the pop-up component is destroyed.
Do note that it is possible to have more than one pop-up component mounted on the screen at once,
so you should only restore the overflow property value when all the popups are destroyed.
You can check the answer at https://github.com/PacktPublishing/Real-World-
Svelte/tree/main/Chapter01/02-scroll-blocker.

Summary
In this chapter, we went through the lifecycles of a Svelte component. We saw the different stages of a
component lifecycle and learned when the lifecycle function callbacks will be called.
We also covered the rule of calling lifecycle functions. This helps us to realize the different patterns
of reusing and composing lifecycle functions.
In the next chapter, we will start to look at the different patterns for styling and theming a Svelte component.
 2
Implementing
Styling and Theming
Without styling, an h1 element within a component will look the same as another h1 element from
another component. Svelte allows you to use Cascading Style Sheets (CSS), a language used for styling
and formatting web content, to style your elements, giving them a different look and feel.
In this chapter, we will start by talking about different ways to style a Svelte component. We will then
see some examples, including integrating a popular CSS framework, Tailwind CSS, into Svelte.
Following that, we will talk about themes. When you have a set of styles consistently applied throughout
Svelte components, you’ll see an overall styling theme in your components. We will talk about how to
synchronize the styles across components, as well as how to let users of the components customize them.
By the end of the chapter, you will have learned various methods of styling and will be comfortable in
choosing the right approach and applying the right methods, depending on the scenario.
This chapter includes sections on the following:

• Ways to style a Svelte component

• Ways to style a Svelte component with Tailwind CSS
• Applying themes to Svelte components

Technical requirements
You can find the code used in this chapter on GitHub: https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter02.
18 Implementing Styling and Theming

Styling Svelte components in six different ways

In a Svelte component, you have elements that define the structure and content. With styling, you can
change the look and feel of the elements beyond the browser default.
Svelte components can be styled in six different ways. Let’s explore the different ways to apply a style
to elements within a Svelte component.

Styling with the style attribute

Firstly, you can add inline styles to an element with the style attribute:

<div style="color: blue;" />

The preceding snippet will turn the color of the text within div to blue.
Similar to the style attribute in HTML elements, you can add multiple CSS styling declarations:

<div style="color: blue; font-size: 2rem;" />

The syntax of adding multiple CSS styling declarations in Svelte is the same as you would do in HTML.
In the preceding snippet, we change the text within div to be blue in color and 2 rem in size.
The value of the style attribute is a string. You can form the style attribute with dynamic expressions:

<div style="color: {color};" />

Sometimes it gets messy when you have multiple CSS styling declarations within the style attribute:

<div style="color: {color}; font-size: {fontSize}; background:

{background}; border-top: {borderTop};" />

Using style: directives

Svelte provides style: directives, which allow you to split the style attribute into several attributes,
which is hopefully more readable after adding line breaks and indentations:

<div
  style:color={color}
  style:font-size={fontSize}
  style:background={background}
  style:border-top={borderTop}
/>

The style: directive follows the following syntax:

style:css-property-name={value}
 Styling Svelte components in six different ways 19

The CSS property name can be any CSS property, including CSS custom properties:

<div style:--main-color={color} />

And if the name of the style matches the name of the value it depends on, then you can use the
shorthand form of the style: directive:

<div
  style:color
  style:font-size={fontSize}
  style:background
  style:border-top={borderTop}
/>

A style declared in the style: directive has higher priority than the style attribute. In the following
example, the h1 text color is red instead of blue:

<div style:color="red" style="color: blue;" />

Besides adding inline styles to style an element one by one, the next approach allows us to write CSS
selectors to target multiple elements and style them together.

Adding the <style> block

In each Svelte component, you can have one top-level <style> block.
Within the <style> block, you can have CSS rules, targeting elements within the component:

<div>First div</div>
<div>Second div</div>
<style>
  div {
    color: blue;
  }
</style>

This approach is useful when you want to apply the same style across multiple elements within
the component.
In the preceding code, would the CSS rule turn all the div elements in the document to blue?
No. The CSS rules within the <style> block are scoped to the component, meaning the CSS rules
will only be applied to the elements within the component, and not div elsewhere in your application.
So you do not have to worry about the CSS rule inside the <style> block changing elements outside
the component.
20 Implementing Styling and Theming

But, how does this work? How does Svelte make sure that the CSS rules only apply to the element
within the same component?
Let’s explore that.

How Svelte scopes CSS rules within a component

Let’s digress for a bit to understand how Svelte makes sure that the CSS rules within a component
are scoped.
When Svelte compiles a component, the Svelte compiler goes through each CSS rule and attempts to
match each element with the selector of the CSS rule:

<div>row</div>
<style>
  div { color: red; }
</style>

Whenever an element matches the selector, the Svelte compiler will generate a CSS class name that
is unique to the component and apply it to the element. At the same time, Svelte limits the scope of
the elements being selected by the CSS rule by including a class selector of the generated class name
in the selector.
The transformation of the element and the CSS rule looks like this:

<div class="svelte-q5jdbb">row</div>
<style>
  div.svelte-q5jdbb { color: red; }
</style>

Here, "svelte-g5jdbb" is the unique CSS class name that is generated by calculating the hash
value of the CSS content. The hash value will be different when the CSS content changes. Since the
Svelte compiler only applies the CSS class name to the elements within the component, it is unlikely
that the style will be applied to other elements outside the component.
This transformation happens during compilation by default. There’s nothing additional that you need
to do. The example here is more for illustration purposes only.
Knowing that the CSS rules within the <style> block are scoped, when you want to apply styles to
all of the elements of the same node name, using the CSS type selector for your CSS rule is sufficient
most of the time. However, if you want to style only some of the elements, say only the second div
element in the preceding example, you can add a class attribute to the second div element to
differentiate between other div elements.
And yes, adding a class attribute is our next way of styling our Svelte component.
 Styling Svelte components in six different ways 21

Adding the class attribute

The CSS class selector in CSS works the same way in Svelte.
When you add a class attribute to an element, you can target it using the CSS class selector.
In the following example, we added the highlight class to the second div, and thus only the
second div has a yellow background:

<div>First div</div>
<div class="highlight">Second div</div>
<style>
  .highlight {
    background-color: yellow;
  }
</style>

The value of the class attribute can be a string or a dynamic expression.

You can conditionally apply classes to the element:

<div class="{toHighlight ? "highlight" : ""} {toBold ? "bold" : ""}"

/>

In the preceding example, when the value of both toHighlight and toBold is true, the class
attribute value evaluates to "highlight bold". Thus, two classes, highlight and bold, are
applied to the div element.
This pattern of conditionally applying classes to an element based on a variable is so common that
Svelte provides the class: directive to simplify it.

Simplifying the class attribute with the class: directive

In the previous example, we conditionally applied the highlight class when the toHighlight
variable was truthy, and bold when the toBold variable was truthy.
This can be simplified with the class: directive, where you have the following:

class:class-name={condition}

To simplify the previous example with the class: directive, we have the following:

<div class:highlight={toHighlight} class:bold={toBold} />

Just like the style: attribute, you can further simplify to a class: directive shorthand if the name
of the class is the same as the name of the variable for the condition.
22 Implementing Styling and Theming

If the condition for adding the highlight class is instead a variable named highlight, then the
preceding example can be rewritten as follows:

<div class:highlight />

Putting all of them together, in the following code example, the div element has a yellow background
when the highlight variable is true, and a transparent background otherwise:

<script>
  export let highlight = true;
</script>
<div class:highlight />
<style>
  .highlight {
    background-color: yellow;
  }
</style>

All the approaches of applying styles to an element that we’ve explored so far have the CSS declarations
written within the Svelte component. However, it is possible to define styles outside of the Svelte component.

Applying styles from external CSS files

Let’s say you add a style element to the HTML of your application, like so:

<html>
  <head>
    <style>.title { color: blue }</style>
  </head>
</html>

Alternatively, you could include external CSS files in the HTML of your application, like so:

<html>
  <head>
    <link rel="stylesheet" href="external.css">
  <head>
</html>

In both cases, the CSS rules written in them are applied globally to all elements in the application,
including elements within your Svelte component.
If you are using build tools such as webpack, Rollup, or Vite to bundle your application, it is common
to configure your build tools to import CSS files using the import statement, just like importing
 Styling Svelte components in six different ways 23

any JS files (some tools, such as Vite, even have been configured to allow importing CSS files such as
any JS files by default!):

import './style.css';

Importing CSS Modules

In Vite, when you name your CSS files ending with .module.css, the CSS file is considered a
CSS Modules file. A CSS Module is a CSS file where all the class names defined within the file are
locally scoped:

/* filename: style.module.css */
.highlight {
  background-color: yellow;
}

This means that the CSS class names specified within a CSS Module will not conflict with any class
names specified elsewhere, even with class names that have the same name.
This is because the build tool will transform the class names within CSS Modules to something unique
that is unlikely to have conflicts with any other names.
The following is an example of how the CSS rules in the preceding CSS Modules would turn out after
the build:

/* the class name 'highlight' transformed into 'q3tu41d' */

.q3tu41d {
  background-color: yellow;
}

When importing a CSS Module from a JavaScript module, the CSS Module exports an object, containing
the mapping of the original class name to the transformed class name:

import styles from './style.module.css';

styles.highlight; // 'q3tu41d'

In the preceding snippet, the imported styles module is an object, and we can get the transformed
class name, 'q3tu41d', through styles.highlight.
This allows you to use the transformed class name in your Svelte component:

<script>
  import styles from './style.module.css';
</script>
<div class="{styles.highlight}" />
24 Implementing Styling and Theming

We’ve seen six different ways of styling a Svelte component, but how do you choose when to use
which one?

Choosing which method to style Svelte components

Each method that we’ve seen so far has its pros and cons. Most of the time, choosing which method
to style your Svelte component is up to personal preference and convenience.
Here are some of my personal preferences when choosing the method to style my Svelte component:

• <style> block over inline styles:

Most of the time, I find that the style I am writing has less to do with the logic of showing the
elements, and more with how the elements look. So, I find it clutters my flow of reading the
component when having the styles in line along with the elements. I prefer to have all my styles
in one place in the <style> block.
• Controlling styles using the style: directive and the class: directive:
When the style property of an element is dependent on a variable, I would use the style:
directive or the class: directive instead of the style attribute or the class attribute. I
find this cleaner to read, as well as finding it a strong signal telling me that the style of the
element is dynamic.
When I am changing only one style property based on one variable, I would use the style:
directive. However, when changing more than one style property with the same variable, I prefer
declaring a CSS class to group all the styles together and controlling it via the class: directive.
• CSS Modules for reusing CSS declarations in multiple Svelte components:
At the time of writing this book, there’s no built-in method in Svelte to share the same CSS
declarations in multiple Svelte components. So, you might want to share the CSS declarations
through CSS Modules.
However, more often than not, when you need the same CSS declarations for elements across
multiple Svelte components, you have a less-than-perfect component structure. It may be
possible to abstract the elements out into a Svelte component.
We’ve seen how we can define our own styles within and outside of a Svelte component; however,
sometimes it’s much easier to adopt styles written by others, rather than designing our own styles.
In the next section, we are going to explore using a popular CSS framework, Tailwind CSS, in Svelte.
 Styling Svelte with Tailwind CSS 25

Styling Svelte with Tailwind CSS

Tailwind CSS is a utility-first CSS framework. It comes with predefined classes, such as flex, pt-4,
and text-center, which you can use directly in your markup:

<div class="flex pt-4 text-center" />

We are going to use Vite’s Svelte template as a base to set up Tailwind CSS. If you are not familiar with
setting up Vite’s Svelte template, here are the quick steps to set it up:

1. Run the Vite setup tool:

npm create vite@latest my-project-name -- --template svelte

This will generate a new folder named my-project-name containing the basic files necessary
for a Svelte project.
2. Step into the my-project-name folder and install the dependencies:
cd my-project-name
npm install

3. Once the dependencies are installed, you can start the development server:
npm run dev

With the Svelte project up and running, let’s look at what we need to do to set up Tailwind CSS.

Setting up Tailwind CSS

Tailwind CSS has come up with a tailwindcss CLI tool that has made the setup so much easier.
Follow these steps:

1. To set up Tailwind CSS in a Svelte + Vite project, we first install the dependencies that are needed:
npm install -D tailwindcss postcss autoprefixer

PostCSS is a tool that transforms CSS files. PostCSS supports plugins, which extend the
functionality of PostCSS to perform different transformations. Tailwind CSS is a PostCSS plugin
that will transform the @tailwind directive – a Tailwind CSS directive, which you’ll see later.
2. After tailwindcss is installed, run the command to generate tailwind.config.js
and postcss.config.js:
npx tailwindcss init -p
26 Implementing Styling and Theming

The tailwind.config.js file keeps the configuration for Tailwind CSS. Tailwind CSS
works by scanning template files for class names and generating the corresponding styles into
a CSS file.
3. Specify the location of where our Svelte components are in tailwind.config.js so that
Tailwind CSS knows where to scan Svelte components for class names:
module.exports = {
  content: [
    // all files ends with .svelte in the src folder
    "./src/**/*.svelte"
  ],
};

The postcss.config.js file keeps the configuration for PostCSS. The default configuration
generated is good for now.
4. Create a CSS file in ./src/main.css to add the Tailwind directives:
@tailwind base;
@tailwind components;
@tailwind utilities;

5. Import the newly created ./src/main.css in the ./src/main.js file:

import './src/main.css';

This is similar to importing external CSS files, which we saw earlier on.
6. Start the Vite dev server:
npm run dev

And you can start using Tailwind CSS!

Try adding the text-center and text-sky-400 Tailwind CSS classes onto the elements
in your Svelte component.
For example, head over to ./src/App.svelte and add the following: <h1 class="text-
center text-sky-400">Hello World</h1>
You just created an <h1> element with two Tailwind CSS classes applied: text-center and text-
sky-400. This will make the text in the <h1> element center-aligned and in a sky-blue color. As
you use Tailwind CSS utility class names in your Svelte component, the style declarations for those
classes are generated and replace the @tailwind utilities directive.
Tailwind CSS is able to extract class names from both the class attribute and the class: directive,
allowing you to statically or conditionally apply the Tailwind CSS styles to the elements:

<script>
  export let condition = false;
 Theming Svelte components with CSS custom properties 27

</script>

<h1 class="text-center">Center aligned title</h1>

<h1 class:text-center={condition}>
  Conditionally center aligned
</h1>

Tailwind CSS comes with a lot of utility classes: you can learn more about Tailwind CSS at https://
tailwindcss.com/.
Since Tailwind CSS classes are globally available, you can apply the same CSS classes to maintain the
same look and feel across Svelte components within the same application. However, if you are doing
so with your own style declarations, you may have to re-specify the same colors or dimensions in the
style declarations across components to maintain the same look. Changing the value in the future
would also be a problem, as we would have to search for usages by the value and update them.
CSS custom properties are a solution for this. They allow us to define a value in one place and
then reference it in multiple other places. Let’s look at how we can use CSS custom properties in
Svelte components.

Theming Svelte components with CSS custom properties

Let’s take a quick knowledge check on CSS custom properties:

• You define CSS custom properties like any other CSS properties, except that the name of the
CSS custom property starts with two dashes:
--text-color: blue;

• To reference the CSS custom property, use the var() function:

color: var(--text-color);

• CSS custom properties cascade like any other CSS properties:

.foo {
  --text-color: blue;
}
div {
  --text-color: red;
}

The value of the --text-color CSS custom property for <div> elements is red, except
for the <div> elements with a class named foo.
28 Implementing Styling and Theming

• The value of CSS custom properties is inherited from their parent:

<div>
  <span />
</div>

If the value of --text-color for <div> in the preceding example is red, then without
other CSS rules applied to <span>, the value of --text-color for <span> is also red.

Defining CSS custom properties

To specify a set of dimensions and colors for the Svelte components, we can define them as CSS custom
properties at the root component of our application:

<div>
<ChildComponent />
</div>
<style>
div {
  --text-color: #eee;
  --background-color: #333;
  --text-size: 14px;
}
</style>

In the preceding example, we define the CSS custom property at the root element of the component,
the <div> element. As a CSS custom property value inherits from a parent, the child elements and
elements within the child components inherit the value from the <div> element.
As we are defining the CSS custom property at the <div> element of the component, elements that
are not the descendant of the <div> element will not be able to access the value.
If, instead, you would like to define the variable for all elements, even elements that are not the
descendant of the root element of our root component, you can define the CSS custom property at
the root of the document using the :root selector:

<style>
  :root {
    --text-color: #eee;
    --background-color: #333;
    --text-size: 14px;
  }
</style>

You do not need to use the :global() pseudo-selector for :root, as it will always refer to the root
of the document, and never be scoped to the component.
 Theming Svelte components with CSS custom properties 29

The :global() pseudo-selector is used in CSS Modules to define styles that apply globally, outside
the local module scope. In Svelte, when used within a component’s <style> block, it allows you
to define CSS rules that won’t be scoped to the component, making them available and applicable to
elements across the entire Svelte application.
As an alternative to defining the CSS custom properties in the <style> block, you can define them
directly on the element itself through the style attribute:

<div style="--text-color: #eee; --background-color: #333; --text-size:

14px;">
  <ChildComponent />
</div>

As you may recall, doing so with the style: directive makes the code look tidier:

<div
  style:--text-color="#eee"
  style:--background-color="#333"
  style:--text-size="14px">
  <ChildComponent />
</div>

To reference the value in the child component, you use the var() function:

<style>
  div {
   color: var(--text-color);
  }
</style>

The great thing about using the CSS custom property is that we could dynamically change the value
of the CSS custom property, and the element’s style referencing the CSS custom property will be
updated automatically.
For example, I can specify the value of --text-color of <div> to be #222 or #eee, based on
a condition:

<script>
  export let condition = false;
</script>
<div style:--text-color={condition ? '#222' : '#eee'} >
  <ChildComponent />
</div>

When the condition is true, the value of var(--text-color) is #222, and the value changes
to #eee when the condition changes to false.
30 Implementing Styling and Theming

As you can see, CSS custom properties make it much easier to synchronize the style of elements.
Now, let’s look at a real-world example of using CSS custom properties: creating a dark/light theme mode.

Example – implementing a dark/light theme mode

Dark mode is a color scheme where you have light-colored text on a dark-colored background. The
idea behind dark mode is that it reduces the light coming from the screen while maintaining color
contrast ratios so that the content is still readable. Less light coming from the device makes it more
comfortable to read, especially in a low-light environment.
Most operating systems allow users to set their preference on whether to use a dark or light theme,
and major browsers support a media query, prefers-color-scheme, to indicate a user’s
system preference:

@media (prefers-color-scheme: dark) {

  /* Dark theme styles go here */
}
@media (prefers-color-scheme: light) {
  /* Light theme styles go here */
}

Before we start, let’s decide on the variables needed.

To simplify things, we only change the background color and text color, so that would be --background-
color and --text-color.
It is possible that your application has other colors, such as accent colors, shadow colors, and border
colors, which would need to have different colors for dark and light themes.
Since these colors are going to be applied everywhere, we are going to define that on the root element
with the :root pseudo-class:

<style>
  @media (prefers-color-scheme: dark) {
    :root {
      --background-color: black;
      --text-color: white;
    }
  }
  @media (prefers-color-scheme: light) {
    :root {
      --background-color: white;
      --text-color: black;
    }
 Theming Svelte components with CSS custom properties 31

  }
</style>

Now, in our Svelte components, we will need to start setting the color of the text to use var(--
text-color):

<style>
  * {
    color: var(--text-color);
  }
</style>

And that’s it; the color of the text will be white when the system preference is on the dark theme, and
black when it is on the light theme.
With the inheriting nature of CSS custom properties, the value of the CSS custom property will be
determined by the closest parent element that has set the value.
This opens the door to allowing component users to specify the style of a component without having
to override style declarations through CSS rules of higher specificity.

Allowing users to change the styles of a component

Let’s say you style your component with CSS in the <style> block, like so:

<p>Hello World</p>
<style>
  p {
   color: red;
  }
</style>

If you want to modify the color of the p element from outside of the component, you’ll need to know
the following:

• The CSS selector that has a higher specificity to override the styles:
Here, we attempt to use the div :global(p) selector to override the color; however, without
knowing the implementation detail of the component, we do not know for sure whether our
selector has a higher specificity:
<div>
  <Component />
</div>
<style>
32 Implementing Styling and Theming

  div :global(p) {
    color: blue;
  }
</style>

• The element structure of the component:

To know which element’s color to override, we would have to know the element structure of
the component and whether the element containing the text whose color we would like to
change is a paragraph element.
The CSS rules and element structure of a component should not be part of the component’s public
API. Overriding the style of a component via a higher specificity CSS rule is not recommended.
Small tweaks on the CSS rules or element structure will most likely break our CSS rule overrides.
A better approach is to expose a list of CSS custom properties that can be used to override the
styles of the component:
<p>Hello World</p>
<style>
  p {
   color: var(--text-color, red);
  }
</style>

The var() function accepts an optional second parameter, which is the fallback value if the
variable name in the first parameter does not exist.
If you use the component without defining --text-color, then the color of the paragraph
will fall back to red.
The color of the paragraph in the following code snippet is red:
<div>
  <Component />
</div>

However, if --text-color is set, then the value of --text-color is used instead:

<div style:--text-color="blue">
  <Component />
</div>

In the preceding code snippet, the color of the paragraph is blue instead of red.
Setting CSS custom properties outside of a component is so common that Svelte provides a
shorthand to pass CSS custom properties into a component directly, without needing to create
a div element to wrap the component yourself:
<Component --text-color="blue" />
 Theming Svelte components with CSS custom properties 33

This is equivalent to the following:

<div style="display:contents;--text-color:blue;">
  <Component />
</div>

Here, "display:contents" is to make sure that the extra div does not participate in the
layout of the contents of <Component />.
If we are specifying a fallback value whenever we are using the CSS custom properties, we may find
ourselves repeating the fallback value a few more times. It would be a hassle if we are going to change
the fallback value. Let’s see how we can align the fallback value.

Aligning the fallback value

If we are using var(--text-color, red) across elements, you may quickly realize that we
should also define a CSS custom property for the fallback value, lest we will be repeating the value
multiple times, and it will potentially be troublesome to find and replace all of them in the future.
To define another CSS custom property, you will have to define it at the root element of your component.
If the value is local to your component and its descendent components only, then you should not
define the CSS custom property at the document root element via :root:

<p>Hello World</p>
<style>
  p {
    --fallback-color: red;
    color: var(--text-color, var(--fallback-color));
  }
</style>

This approach, however, requires us to use var(--fallback-color) wherever we are using var(-
-text-color).
A slightly better approach is to define a new CSS custom property that will have the value of --text-
color if defined, or red as a fallback:

<style>
  p {
   --internal-text-color: var(--text-color, red);
   color: var(--internal-text-color);
  }
</style>

This way, the value of var(--internal-text-color) will always be defined, and it is more
convenient to use just one CSS custom property for elements thereafter.
34 Implementing Styling and Theming

Summary
In this chapter, we went through six different methods to style a Svelte component. So, do you know
which method you are going to use to style your Svelte component? You should now know to choose
the approach that is best suited for the scenario.
We then saw how to use Tailwind CSS in a Svelte project. It takes some initial setup to get Tailwind up
and running at the beginning, but CSS frameworks such as Tailwind CSS usually come with predefined
CSS classes, and most of the time, you use a class attribute or the class: directive to apply them.
Finally, we covered how we can use the CSS custom property to theme Svelte components and how
to allow component users to customize the style of a component. You can now create and share Svelte
components while allowing others to have different styling than the default styles that you’ve created.
In the next chapter, we will look at how to manage the props and states of a Svelte component.
 3
Managing Props and State
In the world of web development, managing data effectively is crucial. Whether it’s the information
that flows between components or the internal state of a component, proper data management is the
backbone of a functional and responsive web application.
In this chapter, we will delve into the core concepts of managing props and state within a Svelte application.
First, we’ll clarify what props and states in Svelte are, laying the groundwork for understanding more
advanced topics. We then explore the concept of bindings, a feature in Svelte for keeping state and
element values or component props in sync.
We’ll then explore data flow within components, highlighting the differences between one-way data
flow and two-way data flow and why they matter. Moving on, we’ll discuss how to derive state from
props using Svelte’s reactive declarations. To conclude, we’ll offer tips for managing complex derived
states and explain how to update props based on those derived states.
By the end of the chapter, you’ll have a solid understanding of how to manage data within a Svelte
component, being equipped with practical tips and strategies to tackle common challenges effectively.
In this chapter, you will learn the following:

• Defining props and state

• Understanding bindings
• One-way versus two-way data flow
• Deriving states from props with a reactive declaration
• Managing complex derived states
• Updating props using derived states

Before we start to talk about props and state, let’s first define what props and state are in Svelte.
36 Managing Props and State

Technical requirements
You can find all the code samples used in this chapter on GitHub at https://github.com/
PacktPublishing/Real-World-Svelte/tree/main/Chapter03

Defining props and state

In Svelte, both props and state are used to manage data within components. Props are a way to pass
data from a parent component to a child component. This makes the child component flexible and
reusable, as it can get different data from the parent as needed.
On the other hand, state is data that is initialized and managed internally within a component, unlike
props, which are received from an external source. State allows a component to be self-contained
and modular.

Defining props
Let’s start with props. Props in Svelte are defined using the export keyword. When you export a
variable in a Svelte component, it becomes a prop that you can pass data to from a parent component.
Here is a simple example:

<!-- file: Child.svelte -->

<script>
  export let message;
</script>
<h1>{message}</h1>

In the preceding code snippet, we defined a Svelte component in a file named Child.svelte. In
the Svelte component, message is a prop. You can pass data to message from a parent component
like so:

<!-- file: Parent.svelte -->

<script>
  import Child from './Child.svelte';
</script>
<Child message="Hello, World!" />

In the preceding code snippet, we define another Svelte component in a file named Parent.
svelte. In the component, we import and use the Child component from Child.svelte. As
the Parent component includes the Child component, the Parent component is considered the
parent component of the imported Child component.
 Defining props and state 37

In the parent component, you can set the message props of the child component to the value
"Hello, World!" by passing "Hello, World!" through the message attribute of the
<Child /> component, as shown in the preceding code snippet.
In summary, props are defined using the export keyword, and their values are passed from the
parent component to the child component.

Defining state
Next, let’s look at the state. State is any data that is used and managed within a component. It is not
passed in from a parent component like props. Instead, it is defined within the component itself.
Here’s an example that illustrates state:

<!-- file: Counter.svelte -->

<script>
  let count = 0;
  function increment() {
    count += 1;
  }
</script>

<button on:click={increment}>Click me</button>

<p>{count}</p>

In this example, count is a state variable. It’s not passed in as a prop but is defined and managed
within the Counter component. When you click the button, the increment function is called,
which modifies the count state.
In summary, props are variables that are passed into a component from a parent component, whereas
the state is data that is managed within a component itself.

Props versus state

If you look closely, both props and state represent data. The difference between them depends on the
context of the component you are considering.
For example, let’s consider two components, component A and component B.
Let’s begin with component A:

<!-- A.svelte -->

<script>
  export let height;
</script>
38 Managing Props and State

In component A, we define a props called height.

Now take a look at component B:

<!-- B.svelte -->

<script>
  import A from './A.svelte';
  let height = 0;
</script>
<A height={height} />

In component B, we define a state called height and pass its value as a prop to component A.
From the perspective of component B, height is considered a state, but from the viewpoint of
component A, it’s considered a prop. Whether a variable is a prop or a state depends on the context
of the component in which it is being viewed. At their core, they are essentially the same thing.
In this example, due to the way JavaScript passes primitive variables by value, it may not be immediately
obvious that the height variable in both component A and component B refers to the same thing.
However, if we define an object as state and pass it to another component through props, it becomes
clear that both the state in one component and the props in the other component refer to the same
object reference.
Let’s modify our example to illustrate this point:

<!-- A.svelte -->

<script>
  export let height;
  setInterval(() => console.log('A:', height), 1000);
</script>
<!-- B.svelte -->
<script>
  import A from './A.svelte';
  let height = { value: 100 };
  setInterval(() => console.log('B:', height), 1000);
</script>
<A height={height} />

In this code snippet, I’ve added a setInterval function to print out the value of the height
variable every second in both components. In component B, I’ve modified the state height to be
an object. Because objects in JavaScript are passed by reference, the state height in component B
that is passed as a prop to component A is passed by reference as well. This means the height prop
in component A is referencing the same object as the state height in component B.
 Understanding bindings 39

If we add a <button> element in component B to mutate the height object, as shown here, you’ll be
able to see that both component A and component B print out the same updated value of the height
variable in the console. This is because they are printing out the value of the same object reference:

<button on:click={() => { height.value += 10; }} />

Clicking on the button from the preceding code snippet will result in the console printing out A: {
value: 110 } and B: { value: 110 }. This demonstrates that the height variable in both
components A and B refer to the same object reference. When height.value is changed to 110
in component B, the change is also reflected in the height variable in component A.
Now that we understand what props and state are in Svelte, let’s talk about bindings next.

Understanding bindings
Bindings in Svelte allow you to keep the value of a component’s state in sync with the value of an
<input /> element. If the state changes, the input updates; conversely, if the input changes, the
state updates as well.
The following code snippet is an example of creating a binding in Svelte:

<script>
  let name = "John";
</script>
<input bind:value={name} />

Bindings are created through the bind: directive. In the preceding code snippet, the input element’s
value is bound to the name variable. When you type in the input, the name variable will update
automatically. Conversely, when you change the value of the name variable, the input element’s value
will also automatically update.
As demonstrated, bindings create a two-way data flow, enabling data changes to propagate from the
element to the component state, and from the component state into the element.
The previous example demonstrates binding on elements, but bindings can also work on components.
You can use the bind: directive to link the props of a component with your component’s state, as
shown in the following code:

<script>
  import Profile from './Profile.svelte';
  let name = "John";
</script>
<Profile bind:username={name} />
40 Managing Props and State

In the preceding code snippet, we bind the username props of the <Profile> component to the
name state variable. When you update the name state variable, the value of the username prop
will automatically reflect the new value; conversely, if you update the value of the username prop
from within the <Profile> component, the value of the name state variable will automatically
update to match.
To further demonstrate this behavior, let’s make a slight modification to the code. Here’s the updated
version of the component:

<script>
  import Profile from './Profile.svelte';
  let name = "John";
</script>
<p>Name from App: {name}</p>
<Profile bind:username={name} />
<button on:click={() => name = "Svelte"}>Update from App</button>

In this code snippet, we’ve added a <p> element and a <button> element. The <p> element shows
the value of the name state variable, and the <button> element, when clicked, updates the value
of the name state variable to Svelte. Due to the binding, when the button is clicked, it will also
update the username props in the <Profile> component.
Here’s the updated version of the <Profile> component:

<script>
  export let username;
</script>
<p>Name in Profile: {username}</p>
<button on:click={() => username = "World"}>Update from Profile</
button>

In this code snippet, we are looking at the <Profile> component. This component receives a prop
called username, whose value is displayed inside a <p> element. We’ve also added a button, and
when clicked, it will update the value of the username prop to "World".
Because of the binding we established in the parent component, any change to the username prop
in this <Profile> component will also update the name state variable in the parent component.
Indeed, if you click on the button in the parent component, both the name state variable in the parent
component and the username prop in the <Profile> component will update to "Svelte". On
the other hand, if you click on the button in the <Profile> component, both the name state variable
in the parent component and the username prop in the <Profile> component will change to
"World". This is the power of two-way binding in Svelte, allowing you to easily synchronize data
between parent and child components.
 One-way versus two-way data flow 41

When we talk about binding in Svelte, we often refer to two-way data binding. But what exactly does
two-way mean? And is there such a thing as one-way data binding? Let’s delve deeper into the concept
of data flow to clarify these terms.

One-way versus two-way data flow

When you pass data from one component to another component either through props or binding, data
flows from one component to another component. The term data flow refers to how data is passed or
transmitted between components or elements within a web application.
Understanding data flow is important when designing an application’s architecture, as it helps to
establish clear lines of communication between components and determine how information is shared
and updated throughout the system.
Data flow can be unidirectional (one-way) or bidirectional (two-way), depending on how data is
transferred between components.
In one-way data flow, data moves in a single direction, from a parent to a child component, or from
a component to a DOM element. This unidirectional flow is achieved through component props or
DOM element attributes.
For example, consider the following code snippets featuring two components, component A and
component B.
In component B, we define a prop named value:

<!-- filename: B.svelte -->

<script>
  export let value;
</script>

In component A, we import component B and pass the value of a variable named data to the prop
of component B:

<!-- filename: A.svelte -->

<script>
  import B from './B.svelte';
  let data;
</script>
<B value={data} />

Here, the value of data in component A is passed to the value prop in component B, illustrating
the data flow from component A to component B.
42 Managing Props and State

If you draw up the data flow in a diagram, this is what a one-way data flow looks like:

Figure 3.1: One-way data flow

The boxes show the data state and value prop, and the arrow shows how changing one value will
affect another.
In one-way data flow, changes to data in the parent component automatically propagate to the child
component, but the reverse is not true.
Building on the example with component A and component B, if the value of the data variable in
component A changes, this change would automatically update the value props in component B.
However, any changes made directly to the value props in component B would not affect the data
variable in component A. Data changes only flow from component A to component B, but not in the
reverse direction.
Having a one-way data flow makes the application easier to reason about and debug, as data changes
follow a predictable path. For example, when the data variable in component A changes unexpectedly,
isolating the issue becomes more straightforward. Because of the one-way data flow, we know that
any changes to the data variable in component A will only originate from component A, not from
component B.
On the other hand, two-way data flow allows data to flow in both directions, enabling automatic
updates between the parent and child components, or between a component and DOM elements.
Two-way data flow in Svelte is achieved through component or DOM element bindings.
For example, consider the following code snippets featuring two components, component C and
component D.
In component D, we define a prop named value:

<!-- filename: D.svelte -->

<script>
  export let value;
</script>
 One-way versus two-way data flow 43

In component C, we import component D and bind the value of a variable named data to the prop
of component D:

<!-- filename: C.svelte -->

<script>
  import D from './D.svelte';
  let data;
</script>
<D bind:value={data} />

Here, the value of data in component C is bound to the value prop in component D, illustrating
the two-way data flow between component C to component D.
Here is a diagram showing how the data flows in a two-way data flow:

Figure 3.2: Two-way data flow

In a two-way data flow, when you change the data in the parent component, the child component is
automatically updated; conversely, changing the data in the child component automatically updates
the parent.
Building on the example with component C and component D, if the value of the data variable in
component C changes, this change would automatically update the value props in component D.
Conversely, unlike one-way data flow, any changes made directly to the value props in component D
would update the data variable in component C. Data changes flow from component C to component
D as well as from component D to component C.
While this bidirectional flow of data can make it easier to keep data in different components synchronized,
it can also introduce complexity and make the application harder to debug, since data changes can
originate from multiple sources.
For example, when the data variable in component C changes unexpectedly, isolating the issue becomes
more complex. Because of two-way data flow, any changes to the data variable in component C could
originate from component C, as well as from component D. This complexity can escalate further if the
value prop of component D is also bound to the prop of another component.
44 Managing Props and State

In summary, one-way data flow offers predictability and easier debugging, while two-way data flow
makes it much easier to synchronize data across multiple components but at the cost of added complexity.
Now that we’ve delved into both one-way and two-way data flows via props, let’s explore how to create
state variables that derive their values from props.

Deriving states from props with a reactive declaration

It’s common in Svelte to create new state variables based on the values of props.
For instance, a <DateLabel /> component might accept a date value as a prop and display a
formatted date inside a <label> element. To use the <DateLabel> component, you might write
the following:

<DateLabel date={new Date(2023,5,5)} />

To display the date as formatted text, you could first define a variable named label, deriving its
value from the date prop:

<!-- filename: DateLabel.svelte -->

<script>
  export let date;
  // Deriving the 'label' variable from the 'date' prop
  let label = date.toLocaleDateString();
</script>
<label>{label}</label>

In this code snippet, we defined a variable called label and derived its value from the date prop
using the toLocaleDateString() method. This variable is then used inside a <label> element
to display the formatted date.
In the preceding code snippet, the label variable is initialized when the <DateLabel> component
is first created. However, if the date prop changes after the component has been initialized, the label
variable won’t update to reflect the new value. This is not the intended behavior, as the <DateLabel>
component should always display the most up-to-date formatted date based on the latest date prop.
To solve this issue, you can use Svelte’s reactive declarations to keep the label variable updated
whenever the date prop changes.
Svelte’s reactive declarations utilize a special $: syntax to mark a statement as reactive. This means
that whenever the variables used in that statement change, the statement itself will be re-run.
Let’s modify our component code to use Svelte’s reactive declaration:

<!-- filename: DateLabel.svelte -->

<script>
 Deriving states from props with a reactive declaration 45

  export let date;

  // Deriving the 'label' variable from the 'date' prop
  $: label = date.toLocaleDateString();
</script>

In this code snippet, by modifying the declaration of label to a reactive declaration, the component
will automatically re-compute the label variable whenever the date prop is modified, ensuring
that the <DateLabel> component is always displaying the most current formatted date.
So, how does Svelte know when a reactive declaration statement should be re-run?
Svelte re-runs a reactive declaration statement whenever any of its dependent variables change. The
Svelte compiler identifies these dependencies by analyzing the statement.
For example, in the reactive declaration $: label = date.toLocaleDateString();,
Svelte recognizes that the dependency for this statement is the date variable. Therefore, whenever
the date changes, the statement will re-run and update the value of the label variable.
A good rule of thumb for identifying dependencies in a reactive declaration is to look for any
variable on the right side of the equal sign (=). These variables are considered dependencies of the
reactive declaration.
You can include multiple dependencies within a single reactive declaration. For instance, let’s say
we want to add a new locale prop to the <DateLabel> component. To use this new prop, you
might write something like this:

<DateLabel date={new Date(2023, 5, 5)} locale="de-DE" />

In this code snippet, we pass in a new locale prop with the value de-DE to format the date
in German. To accommodate this new locale prop, we’ll need to modify our <DateLabel>
component as follows:

<script>
  export let date;
  export let locale;
  // Reactive declaration with multiple dependencies
  $: label = date.toLocaleDateString(locale);
</script>
<label>{label}</label>

In this updated code, the reactive declaration $: label = date.toLocaleDateString(locale);

now has two dependencies: date and locale. Svelte will automatically re-run this statement
whenever either of these variables changes, ensuring that the label value stays up to date with the
latest date and locale props.
46 Managing Props and State

Now that we’ve covered the basics of props, bindings, states, and derived states, it’s crucial to note that
as components become more complex, managing these elements can quickly become overwhelming.
In the next section, we’ll explore some tips for effectively managing complex derived states to keep
them manageable.

Managing complex derived states

As your Svelte application grows more complex, it will likely involve a greater number of interconnected
components with multiple props and derived states. When dealing with this complexity, tracking
updates and changes can become a complex task. Each prop or state change can affect other parts of
your component, making it challenging to manage and predict how your component will behave.
To make this easier, here are some guidelines to consider:

• Maintain one-way data flow for derived states

While it’s possible to derive state from props and other states, it’s crucial to maintain a one-way
data flow to simplify both debugging and understanding. Consider the following Svelte example:
<script>
  export let valueA;
  export let valueB;
  $: valueC = valueA + 5;
  $: valueD = valueB + valueC;
  $: valueC = Math.min(valueC, valueD / 2);
</script>

This code snippet won’t compile in Svelte due to a cyclical dependency detected by the compiler.
Though workarounds may exist to circumvent this compile error, the code is purposely flawed
to highlight the issue.
Upon observing the data flow, we can see that valueC depends on valueA, valueD
depends on both valueB and valueC, and then valueC in turn depends on valueD.
Consequently, it’s unclear how valueC is actually calculated, it could be either valueA +
5 or the minimum of valueC and valueD / 2. Such complexity makes the code hard to
understand and increases the likelihood of bugs.
• Group similar logic together
Consider this tip a stylistic suggestion. When dealing with multiple reactive declarations, it’s
beneficial to group those that are related together. Utilize blank lines to create a visual separation
between unrelated reactive declarations. This not only improves readability but also aids in
code maintenance.
It is worth noting that the Svelte compiler takes care of execution order based on dependencies
regardless of how you arrange your declarations. For example, the following two code snippets
will behave identically due to Svelte’s handling of dependencies:
 Managing complex derived states 47

<script>
  export let a;
  $: b = a * 2;
  $: c = b * 2;
</script>

In the preceding snippet, b is declared reactively before c. However, it behaves identically to

the following snippet, where c is declared before b. Let’s look at the next one:
<script>
  export let a;
  $: c = b * 2;
  $: b = a * 2;
</script>

The Svelte compiler analyzes the dependencies of each declaration and executes them in the
correct order. In this case, it evaluates b = a * 2 before c = b * 2 since the latter depends
on the value of b, established by the former declaration.
• Avoid reassigning props directly
It might be tempting to modify the value of a prop directly, especially when you want to transform
its value or provide a default fallback. For example, you might consider writing something like
this to set a default value for an undefined prop:
<script>
  export let data;
  $: data = data ?? 100;
</script>

While this code snippet may seem to work as intended, directly reassigning prop values within
a component can introduce confusion and make data flow unpredictable. This practice can
muddy the waters when you’re trying to trace the origin of data changes, making it unclear
whether the alteration occurred within the component itself or came from the parent component.
Instead, it’s better to declare a new state variable to manage this behavior:
<script>
  export let data;
  $: dataWithDefault = data ?? 100;
</script>

In this improved example, we introduce a new variable, dataWithDefault, which takes on

either the value of the data prop or a default value of 100 if data is undefined. This approach
makes the component’s behavior more transparent and easier to debug.
48 Managing Props and State

• Be cautious when updating derived states

Modifying a derived state directly can introduce inconsistencies, particularly when that state
is based on props or other state variables.
Consider the following example:
<script>
  export let value;
  $: double = value * 2;
  $: triple = value * 3;
</script>
<input bind:value={double} type="number" />
<input bind:value={triple} type="number" />

In this example, we have a single prop named value and two derived states, double and
triple, which are two and three times the value of the value prop respectively. The component
features two input boxes, each bound to double and triple using two-way binding.
Here, triple can be modified in two ways: either by updating the value prop, which will
keep triple at three times the value of value, or by directly changing the value in the input
box, thereby directly altering triple.
If you type into the input box bound to triple, you’ll find that its value diverges from being
strictly three times the value prop. This inconsistency arises because triple is now subject
to changes from multiple sources, causing it to go out of sync with the original value.
If you were to map out a diagram illustrating the data flow between value double and
triple, you’d get the following diagram:

Figure 3.3: Data flows between props, states, and the input elements
 Updating props using derived states 49

Notice that both double and triple have two incoming arrows, suggesting multiple
pathways for changing their values. Conversely, value has no incoming arrows, indicating
that modifying triple alone would cause value and double to go out of sync.
Therefore, it is recommended to refrain from manually updating derived states, as this complicates
debugging and makes the component’s behavior harder to reason about.
If you wish to modify the input while maintaining the synchrony between value, double,
and triple, a solution will be discussed in the following section.
By keeping these tips in mind, you can better manage complex derived states and make your components
more maintainable and easier to understand.
In our last tip, we highlighted that updating the derived states can lead to inconsistencies between
states and props and noted that there’s a solution to modify the input while keeping everything in
sync. That solution is what we will explore next.

Updating props using derived states

In an attempt to synchronize the value prop with changes to the input bound to triple, one might
be tempted to add another reactive declaration. This declaration would update the value prop to be
one-third of triple whenever triple changes. Here is the proposed modification:

<script>
  export let value;
  $: double = value * 2;
  $: triple = value * 3;
  $: value = double / 2;
  $: value = triple / 3;
</script>
<input bind:value={double} type="number" />
<input bind:value={triple} type="number" />

As we discussed earlier, it’s best practice to maintain a one-way data flow for derived states to simplify
debugging and data management. Indeed, the Svelte compiler flags the preceding code snippet
for cyclical dependencies. This is because double is derived from value, and value is in turn
dependent on double.
However, Svelte’s compiler determines dependency relationships based solely on the reactive declarations.
By refactoring the code to make these relationships less obvious, you can bypass the compiler’s cyclical
dependency check. Here’s a refactored version of the code that does just that:

<script>
  export let value;
  $: double = value * 2;
  $: triple = value * 3;
50 Managing Props and State

  $: updateValueFromDouble(double);
  $: updateValueFromTriple(triple);
  function updateValueFromDouble(double) {
    value = double / 2;
  }
  function updateValueFromTriple(triple) {
    value = triple / 3;
  }
</script>
<input bind:value={double} type="number" />
<input bind:value={triple} type="number" />

In the provided code snippet, we’ve shifted the equations value = double / 2 and value = triple
/ 3 into separate functions named updateValueFromDouble and updateValueFromTriple.
This change lets us evade Svelte’s compiler warning about cyclical dependencies.
However, there’s a catch. If you try altering the triple input, it updates value but doesn’t refresh
double. This happens because Svelte avoids infinite update loops. Changing triple would set off
a chain reaction—updating value, then double, then back to value, and so on.
This is how the data flow looks right now:

Figure 3.4: Data flows between props, states, and the input elements
 Updating props using derived states 51

As you can see in the diagram, we update the value of value, double, and triple through reactive
declarations, creating a loop in the data flow, indicated by the bold arrows.
Therefore, using derived states to update their original properties via reactive declarations isn’t advisable.
A better approach to keep value, double, and triple in sync is to establish value as the single
source of truth. Since both double and triple are derived from value, any changes to the input
should first update value. This, in turn, triggers the reactive declarations to automatically recalculate
double and triple.
Here’s the updated code:

<script>
  export let value;
  $: double = value * 2;
  $: triple = value * 3;
  function updateValueFromDouble(double) {
    value = double / 2;
  }
  function updateValueFromTriple(triple) {
    value = triple / 3;
  }
</script>
<input value={double} type="number" on:change={e =>
updateValueFromDouble(e.target.value)} />
<input value={triple} type="number" on:change={e =>
updateValueFromTriple(e.target.value)} />

In the preceding code snippet, we’ve shifted our approach to focus on value as the sole source of
truth. Instead of binding the input elements directly to double and triple, we’ve added event
listeners that update value based on user input. This change automatically updates double and
triple through reactive declarations, which then refresh the displayed values in the input fields.
52 Managing Props and State

With the updated code, the data flow is now streamlined, as shown in the following diagram:

Figure 3.5: Data flows between props, states, and the input elements

The values for double and triple are directly derived from value, which then populates the
input fields. When you modify the input, it alters value directly, which in turn automatically updates
both double and triple, as well as the input fields themselves.
So, this is how you maintain synchronized states for value, double, and triple, by focusing on
the data flow and keeping a single source of truth.

Summary
Understanding how to handle props and state effectively is crucial for creating robust Svelte apps.
This chapter has shown you how Svelte uses props, bindings, and reactive declarations to facilitate
data passing and state changes across components.
It’s crucial to keep an eye on data flow within a component. Having a unified and organized data flow
makes the code easier to follow and debug. Good data management paves the way for building more
intricate and dynamic apps with ease.
In the next chapter, we’ll explore how to compose Svelte components to construct more
complex applications.
 4
Composing Components
As your application grows, cramming all the logic into a single component becomes impractical.
You’ll need to split your app into smaller, modular components and assemble them to form a more
complex application.
In this chapter, we’ll explore various techniques to combine components effectively. We’ll start by
examining how to inject custom content into a component using slots. Then, we’ll discuss how to
conditionally render different HTML elements within a component. We’ll also delve into recursive
components, useful for displaying nested or hierarchical data.
We’ll guide you through each topic with hands-on examples, ensuring the techniques you learn are
both practical and applicable in real-world scenarios. By the end of this chapter, you’ll have a richer
set of strategies to compose components in your Svelte applications.
In this chapter, you will learn the following:

• Manipulating how a child component looks from the perspective of its parent
• Passing dynamic content through slots
• Rendering different HTML element and component types
• Creating recursive components for recursive data
• The Container/Presentational pattern

Let’s kick things off by exploring the various ways we can manipulate and control the content of a
child component.

Technical requirements
All the code in this chapter can be found at https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter04.
54 Composing Components

Manipulating how a child component looks

When you’re combining multiple components, you’ll need to manage how each child component
appears and behaves. Even though a child component handles its own display and logic, it still offers
controls to tweak its appearance and behavior. From the perspective of the parent component, you’ll
want to coordinate these child components to achieve the desired overall functionality.
In this section, we’ll explore various ways to control the look of child components, ranging from the
most to the least commonly used methods. Understanding these options will equip you with the tools
to make your components both versatile and effective.
The list of options to control how the child component looks includes the following:

• Controlling through props: This is perhaps the most straightforward way to influence the
behavior and appearance of a child component. By passing props from a parent component to a
child, you can make your child component highly dynamic and responsive to external changes.
To demonstrate how a parent component can control content using props in Svelte, consider
the following code example:
<script>
  import Child from './Child.svelte';
  let message = 'Hello from Parent';
</script>
<Child message={message} />

In the code snippet, we imported the Child component and passed the message prop to it.
The child component, implemented in the following code snippet, then uses this message
prop to display text within a <p> element:
<script>
  export let message;
</script>
<p>{message}</p>

As you can see, the parent component has the ability to dictate the text shown in the Child
component by modifying the message prop’s value. Controlling props is a simple yet effective
way to manipulate a child component’s content. If you’re interested in learning more about
props, we covered the topic extensively in Chapter 3.
• Controlling through context: Svelte’s context API allows components to share state without
explicitly passing props through every layer of the component tree. This is particularly useful
when you have deeply nested components.
Here’s an example of a parent component:
<script>
  import Child from './Child.svelte';
 Manipulating how a child component looks 55

  import { setContext } from 'svelte';

  setContext('message', 'Hello from parent');
</script>
<Child />

In the previous code, we established a context named message with the 'Hello from
parent' value, using setContext. Then, we imported and used the Child component
without sending any props to it.
The following is the code for the Child component:
<script>
  import { getContext } from 'svelte';
  const message = getContext('message');
</script>
<p>{message}</p>

Here, getContext is used to read the message context value, which is then displayed
within a <p> element. As demonstrated, the parent component can influence the text in the
child component’s <p> element by changing the context value.
For a deeper dive into Svelte’s context feature, you can refer to Chapter 8, where we will explore
the topic in greater detail.
• Controlling the styles: Manipulating a child component’s appearance isn’t solely about controlling
the data passed to it. It also involves tweaking or modifying its styles.
You can modify a component style through CSS custom properties. This approach allows for
greater design flexibility, and it ensures that child components can adapt to various contexts
or themes within the parent component or the broader application.
For an in-depth discussion on how to alter a component’s style using CSS custom properties,
feel free to refer to Chapter 2.
• Passing dynamic content through slot: Svelte’s slot feature allows you to insert custom content
into specific areas of a child component. This provides a flexible way to have greater control
over a component’s content without modifying its internal behavior.
We will talk about slots and how to use them in the following section.
As you can see, there are various ways to shape a child component’s appearance and behavior. When
we compose different components within a parent component, the goal is to make them work together
in a coordinated way. You can use a combination of the approaches discussed to achieve this.
Most of these methods will be covered in separate chapters, and in the following section, we’ll focus
on how to dynamically alter a child component’s look by passing content through slots.
56 Composing Components

Passing dynamic content through slots

When building complex applications, one size doesn’t always fit all. Balancing between a component’s
modularity and a component’s flexibility for customization is crucial.
Take a generic Card component, for example. You might sometimes want to include special
headlines, unique lists, or custom footers for specific use cases. It’s nearly impossible to anticipate
every requirement, so it’s essential to design components that are both modular and maintainable,
yet still open to customization.
This is where Svelte’s slot feature shines. Slots enable you to inject dynamic content into your components,
making them incredibly versatile. Instead of designing a Card component that tries to include every
possible feature, aim for a simple, clean base that can be enhanced through composition. This approach
allows you to piece together more complex, customized components as your needs evolve.
In a Svelte component, a <slot> element is a placeholder within your component where you
can inject any type of content from a parent component. Here’s how you can define a slot inside a
Svelte component:

<!-- filename: Card.svelte -->

<div class="card">
  <slot></slot>
</div>

In the preceding code snippet, we defined a <slot> element inside a <div> element. Any dynamic
content from the parent component will be inserted into the <div> element.
Now, to pass dynamic content from a parent component into the slot of a child component, you need
to place that content between the opening and closing tags of the child component:

<script>
  import Card from './Card.svelte';
</script>
<Card>
  <h1>Special headline</h1>
</Card>

In this code snippet, we passed in a <h1> element between the opening and closing tags of the
<Card> component. This effectively replaces the <slot> element in the Card component with
the <h1> element we provide.
There may be instances where you don’t need to pass dynamic content from a parent component to a
child component’s slot. For these cases, Svelte allows you to define default content within the <slot>
element, providing a fallback when no custom content is supplied by the parent.
 Passing dynamic content through slots 57

Providing default slot content

To provide default content for your <slot> element, you can place them inside the <slot> element,
like so:

<div class="card">
  <slot>
    <div>this is the default content</div>
  </slot>
</div>

In the preceding code snippet, we defined <div>, containing the text This is the default
content within the <slot> element. This serves as the default content for the slot.
When you use this Card component in a parent component and don’t provide content for the slot,
as shown in the following code snippet, the default text will automatically appear:

<script>
  import Card from './Card.svelte';
</script>
<Card />

By utilizing the default content in slots, you can create more flexible Svelte components that offer
sensible fallbacks when no custom content is provided.
In the examples we’ve seen so far, we’ve been limited to inserting dynamic content from the parent
component into just one spot. But what if we wanted multiple insertion points for dynamic content?
Let’s explore how to achieve that.

Having multiple slots with named slots

Slots are amazing, and you are not limited to a single slot either. You can define multiple slots in a
component and name them to target specific areas for dynamic content.
You can use the name attribute on the <slot> element to give the slot a name, as shown in the
following code:

<!-- filename: Card.svelte -->

<div class="card">
  <header>
    <slot name="header"></slot>
  </header>
  <footer>
    <slot name="footer"></slot>
  </footer>
</div>
58 Composing Components

In the previous code snippet, we defined two named slots, one named "header" and another
named "footer".
To target the dynamic content into these named slots, you will need to use the <svelte:fragment>
element in the parent component:

<script>
  import Card from './Card.svelte';
</script>
<Card>
  <svelte:fragment slot="header">
    <h1>Special headline</h1>
  </svelte:fragment>
  <svelte:fragment slot="footer">
    Custom footer
  </svelte:fragment>
</Card>

Here, we pass two <svelte:fragment> elements into the Card component. The first has a slot
attribute set to "header", and the second slot attribute is set to "footer".
The name specified in the slot attribute will correspond with the named slots within the Card
component. The content within <svelte:fragment slot="header"> replaces the <slot
name="header"> in the Card component. Similarly, the content within <svelte:fragment
slot="footer"> takes the place of <slot name="footer"> in the component.
Sometimes, you will want the dynamic content from the parent component to be aware of the data
within the child component. To handle this scenario, Svelte offers a feature known as slot props, which
allows you to pass data from the child component to the content in the slot. Let’s explore how this works.

Passing data through slot props

You can pass data from the child component back to the dynamic content in the parent component
using slot props, like so:

<!-- filename: Card.svelte -->

<div class="card">
  <header>
    <slot name="header" width={30} height={50}></slot>
  </header>
</div>

In the code example, we passed in two additional attributes to the <slot> element besides the name
attribute, which is used to name the slot. These additional attributes, width and height, serve as
slot props; their values can be accessed in the parent component when you create dynamic content.
 Rendering different HTML element or component types 59

Here is an example of how you can use these slot prop values when using <svelte:fragment>:

<script>
  import Card from './Card.svelte';
</script>
<Card>
  <svelte:fragment slot="header" let:width let:height>
    <h1>Dimension: {width} x {height}</h1>
  </svelte:fragment>
</Card>

In the preceding code snippet, you can see that we use let:width and let:height within the
<svelte:fragment> element. These are called let: bindings.
The let: bindings allow us to access the width and height slot props provided by the child Card
component. The values for the width and height are then used within an <h1> element to display
the dimensions. This way, we can dynamically adjust content in the parent component, based on data
originating from the child component.
Now that we have covered how to pass dynamic content through slots, how to define multiple-named
slots, and how to pass data from a child component to the dynamic slot content in a parent component,
you are well equipped to create more versatile and reusable Svelte components.

Renderless component
One of many common patterns to compose components using slots is the Renderless component
pattern. This technique allows you to build components that are purely functional, leaving the
presentation details to the parent component. We have dedicated Chapter 11 to delving into
this fascinating topic.

Now that we’ve explored how to pass dynamic content through slots, let’s dive into another exciting
feature of Svelte that lets you dynamically render different element or component types.

Rendering different HTML element or component types

In any dynamic application, there comes a time when you need even more flexibility than what static
components or elements offer. What if you don’t know the type of element or component you’ll need
to render until runtime?
Let’s imagine that you’re building a form generator, and the type of form field – whether it’s <Input>,
<Checkbox>, or <Select> – is determined by dynamic data. How could you switch between these
components seamlessly, especially when they share the same set of props?
60 Composing Components

One straightforward approach is to use Svelte’s {#if} blocks to conditionally render the component
you need. Here is an example code snippet:

<script>
  import Input from './Input.svelte';
  import Checkbox from './Checkbox.svelte';
  import Select from './Select.svelte';
  let type = "input"; // Could be "checkbox" or "select"
</script>
{#if type === "input"}
  <Input value={value} onChange={onChange} />
{:else if type === "checkbox"}
  <Checkbox value={value} onChange={onChange} />
{:else}
  <Select value={value} onChange={onChange} />
{/if}

In the code snippet, we use the {#if} blocks to choose among three different component types,
passing the same prop values to each. If you find yourself needing to manage more component types,
this could lead to a lengthy and hard-to-maintain series of conditional blocks.
Is there a more efficient way to handle this?
Svelte provides two specialized elements, <svelte:element> and <svelte:component>,
precisely for this. The <svelte:element> element lets you create different HTML element
types dynamically, based on variable data, whereas the <svelte:component> element lets you
dynamically render different Svelte component types.
Here’s how you can rewrite the previous example using <svelte:component>:

<script>
  import Input from './Input.svelte';
  import Checkbox from './Checkbox.svelte';
  import Select from './Select.svelte';
  let type = "input"; // Could be "checkbox" or "select"
  let DynamicComponent;
  if (type === "input") {
    DynamicComponent = Input;
  } else if (type === "checkbox") {
    DynamicComponent = Checkbox;
  } else {
    DynamicComponent = Select;
  }
</script>
<svelte:component
 Rendering different HTML element or component types 61

  this={DynamicComponent}
  value={value}
  onChange={onChange}
/>

In the preceding code snippet, the DynamicComponent variable is used to hold the type of
component that will be rendered. This component type is then passed to the this attribute within
the <svelte:component> element. The <svelte:component> element also accepts other
props such as value and onChange.
With the preceding code, what happens is that <svelte:component> renders the designated
component stored in the this attribute, simultaneously forwarding any props passed to
<svelte:component> to this dynamically rendered component. For example, if the value of
DynamicComponent is the Select component, then the preceding code is effectively the same
as this code:

<Select value={value} onChange={onChange} />

By using <svelte:component>, we simplify the code and make it more maintainable. It’s also
easier to extend; adding another form element type would only require an additional condition
and assignment.
Now that we’ve explored the use of <svelte:component>, let’s look at <svelte:element>,
which follows a similar pattern. The following is a sample code snippet that demonstrates the usage
of <svelte:element>:

<script>
  let type = 'button'; // could be 'div', 'p', 'a', etc.
</script>
<svelte:element this={type}>
  Click Me
</svelte:element>

In the preceding code snippet, the type variable holds the type of HTML element we want to render
– in this example, it’s button. The <svelte:element> tag dynamically creates an HTML element
of the type specified by type. So, if type is 'button', this will render as a <button> element,
containing the text "Click Me".
This approach is particularly useful when you want to switch the type of an HTML element, based
on some condition, without having to rewrite the entire block of code. All you need to do is change
the value of type.
To summarize, <svelte:element> and <svelte:component> offer a more efficient and
maintainable way to handle dynamic rendering needs. They provide a robust alternative to multiple
{#if} blocks, making your Svelte applications more flexible and easier to manage.
62 Composing Components

Sometimes, when designing Svelte components for data visualization, we encounter recursive data
structures. In such cases, we need to build components that can recursively render themselves. Let’s
delve into how we can accomplish this next.

Creating recursive components for recursive data

A recursive data structure is a data structure that is defined in terms of itself, meaning that it can be
composed of smaller instances of the same type of structure. Recursive data is everywhere – think of
a comment section where replies can have their own sub-replies, or a filesystem with nested folders.
Creating a component to display them in a frontend application can be challenging.
Imagine we have a variable called folder, which is an array containing either files or additional
folders. In this example, the folder variable could look like this:

const folder = [
  { type: 'file', name: 'a.js' },
  { type: 'file', name: 'b.js' },
  { type: 'folder', name: 'c', children: [
    { type: 'file', name: 'd.js' },
  ]},
];

Currently, our folder variable is two levels deep. To represent this structure in a Svelte component,
you might think to use nested {#each} blocks:

<script>
  export let folder;
</script>
<ul>
  {#each folder as item}
    <li>
      {#if item.type === 'file'}
        <div>File: {item.name}</div>
      {:else}
        <div>Folder: {item.name}</div>
        <ul>
          {#each item.children as child}
            <li>
              {#if child.type === 'file'}
                <div>File: {child.name}</div>
              {:else}
                ...
              {/if}
            </li>
 Creating recursive components for recursive data 63

          {/each}
        </ul>
      {/if}
    </li>
  {/each}
</ul>

In the preceding code snippet, we used an {#each} block to iterate over the items in the folder
variable, rendering either a file or a folder based on item.type. If item.type is a folder, we use
another nested {#each} block to go through its contents.
However, here’s the catch – the inner folder could also contain files or additional folders, making it
recursive. As a result, we end up with repeated code for the inner and outer {#each} blocks. This
works fine for a folder structure that’s only two levels deep, but what if it’s more complex? How can
we avoid duplicating the same code for each level of nesting?
Svelte offers an elegant solution to handle such recursive structures with the <svelte:self>
element. The <svelte:self> element allows a component to embed itself within its own markup,
thus making it easier to handle recursive data.
Here’s the updated code using the <svelte:self> element:

<script>
  export let folder;
</script>
<ul>
  {#each folder as item}
    <li>
      {#if item.type === 'file'}
        <div>File: {item.name}</div>
      {:else}
        <div>Folder: {item.name}</div>
        <svelte:self folder={item.children} />
      {/if}
    </li>
  {/each}
</ul>

In the updated code snippet, we replaced the nested {#each} block with <svelte:self
folder={item.children} />. This effectively re-renders the current component, passing
item.children as the folder prop.
The beauty of this is that it eliminates the need to initiate a new <ul> element and duplicate {#each}
blocks for each nesting level. Instead, the component simply reuses itself, making it capable of handling
nested structures of any depth.
64 Composing Components

When comparing this with the previous code snippet, you can immediately see the advantage – it’s
much cleaner and avoids repetitive code, making it easier to read and maintain, and it scales naturally
to handle any level of nested folders.
Now that we’ve discussed how to tackle recursive data with <svelte:self>, let’s look at a practical
example in the following section, where we will create a Svelte-based JSON tree viewer.

Example – a JSON tree viewer

In this section, we will walk you through building a JSON tree viewer component in Svelte. The JSON
tree viewer component helps you visualize JSON data in a tree-like format. Along the way, we’ll make
use of some of the advanced Svelte features we’ve covered in this chapter, including <svelte:self>,
<svelte:component>, and slots.
Before we start, let’s think about what our JSON Tree Viewer should look like and how it should
behave. Essentially, a JSON object is made up of key-value pairs, where the values can be either
primitives, arrays, or other nested objects. Our goal is to display these key-value pairs in a way that
clearly represents the hierarchical structure of the JSON object.
So, let’s create a JsonTree component for our JSON tree viewer:

<!-- filename: JsonTree.svelte -->

<script>
  export let data;
</script>
<ul>
  {#each Object.entries(data) as [key, value]}
    <li>{key}: {value}<li>
  {/each}
</ul>

In the preceding code snippet, we defined a JsonTree component that accepts data as a prop. Inside
this component, we utilize Svelte’s {#each} block to generate a list of <li> elements. Each element
displays a key-value pair from the data object.
However, these values can vary. They could be primitive types or nested objects. Take the following example:

data = {
  name: 'John Doe',
  address: {
    street: '123 Main St'
  },
}
 Example – a JSON tree viewer 65

In the preceding data object, the name key has a primitive string value, while the value for the
address key is an object. For nested objects like this, we will need to use our JsonTree component
recursively to render that nested object. This is where <svelte:self> comes into play.
Here’s the updated code:

<!-- filename: JsonTree.svelte -->

<script>
  export let data;
</script>
<ul>
  {#each Object.entries(data) as [key, value]}
    <li>
      {key}:
      {#if typeof value === 'object'}
        <svelte:self data={value} />
      {:else}
        {value}
      {/if}
    <li>
  {/each}
</ul>

In this updated code snippet, we introduced an {#if} block that renders differently based on
whether the value is a primitive type or a nested object. For nested objects, we recursively render
another JsonTree component using <svelte:self>. This allows us to elegantly handle JSON
objects of any depth.
Now, in our current implementation, we haven’t made distinctions between different types of primitive
values. Suppose you’d like to render various primitive types using specialized Svelte components. In that
case, you could dynamically switch between different component types using <svelte:component>.
Let’s assume you have three separate Svelte components – StringValue for strings, NumberValue
for numbers, and JsonValue for others (such as Booleans, null, and undefined). Here is how
you can update the code to use <svelte:component>:

<script>
  import StringValue from './StringValue.svelte';
  import NumberValue from './NumberValue.svelte';
  import JsonValue from './JsonValue.svelte';

  export let data;

  function getComponent(type) {
    if (type === 'string') return StringValue;
    if (type === 'number') return NumberValue;
66 Composing Components

    return JsonValue;
  }
</script>
<ul>
  {#each Object.entries(data) as [key, value]}
    <li>
      {key}:
      {#if typeof value === 'object'}
        <svelte:self data={value} />
      {:else}
        <svelte:component this={getComponent(typeof value)}
value={value} />
      {/if}
    </li>
  {/each}
</ul>

In the updated code snippet, we created a utility function named getComponent to determine which
component to render, based on the type of the primitive value. We then used <svelte:component
this={getComponent(typeof value)} /> to dynamically load the appropriate component
for each primitive type, allowing our JsonTree component to switch to different components for
different types of data.
Lastly, to make our JSON tree viewer more versatile, we can introduce named slots to customize the
appearance of keys and values. By doing so, users can easily tailor the look of these elements according
to their needs, while keeping our current design as the default fallback. Let’s update the code to add
the two named slots:

<ul>
  {#each Object.entries(data) as [key, value]}
    <li>
      <slot name="obj-key" key={key}>{key}:<slot>
      {#if typeof value === 'object'}
        <svelte:self data={value} />
      {:else}
        <slot name="obj-value" value={value}>
          <svelte:component this={getComponent(typeof value)}
value={value} />
        </slot>
      {/if}
    </li>
  {/each}
</ul>
 The Container/Presentational pattern 67

In the updated code, we added two named slots – one to customize keys, called "obj-key", and
another to customize values, called "obj-value". These slots receive the current key and value
as slot props, enabling you to tailor their appearance in the parent component.
For instance, if you wish to change how keys and values are displayed, here’s how you can write in
your parent component:

<JsonTree data={{a: 1, b: 2}}>

  <svelte:fragment slot="obj-key" let:key>
    <em>{key}</em>
  </svelte:fragment>
  <svelte:fragment slot="obj-value" let:value>
    <u>{value}</u>
  </svelte:fragment>
</JsonTree>

Here, we use Svelte’s <svelte:fragment> special element to target the named slots, "obj-key"
and "obj-value", in our JsonTree component. The let:key and let:value syntax allows
us to access the current key and value being rendered, respectively.
In the <svelte:fragment> element, we wrapped the key in an <em> tag and the value in a <u>
tag, offering custom styling for these elements. This customization overrides the default rendering
provided by the JsonTree component. If you don’t specify any custom content, the component will
fall back to its default rendering, offering both robustness and flexibility.
By combining <svelte:self> for recursion, <svelte:component> for dynamic behavior,
and slots for customization, we have created a flexible and powerful JSON tree viewer. This not only
demonstrates Svelte’s capability to handle complex UI patterns but also serves as a practical example
of component composition in Svelte.
You can find the complete code for this exercise here: https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter04/06-exercise-svelte-json-tree.
Before we wrap up this chapter, let us take a moment to discuss a helpful common pattern to organize
our components, the Container/Presentational pattern.

The Container/Presentational pattern

As your application scales in complexity, you may find it beneficial to adopt specific design patterns or
guidelines to structure components. One such approach is the Container/Presentational pattern, which
divides a component into two categories, the Container component and the Presentational component:

• Container components focus on functionality. They handle data fetching, state management, and
user interactions. While they usually don’t render Document Object Model (DOM) elements
themselves, they wrap around Presentational components and supply them with data and actions.
68 Composing Components

• Presentational components are all about the user interface. They get their data and event-handling
functions exclusively through props, making them highly reusable and straightforward to test.

A common scenario where you’ll see this pattern in action is when using a UI component library.
In this case, the library’s components serve as the presentational elements, focusing solely on how
things look. Meanwhile, the components you create that utilize these library elements act as Container
components, managing data and handling interactions such as clicks, typing, and dragging.
The Component/Presentational pattern is useful in a few aspects:

• Simplicity: Separating behavior from appearance makes your codebase easier to understand
and maintain
• Reusability: Since Presentational components are agnostic about data sources or user actions,
you can reuse them in different parts of your application
• Collaboration: With this division, designers can work on the Presentational components while
developers focus on the Container components, streamlining development

As you come to appreciate the benefits of the Component/Presentational pattern, there are specific
situations where I think you should consider using it, such as the following:

• When your application starts to grow: Managing everything in a single component can become
confusing as complexity increases
• When you find yourself repeating the same UI patterns: Creating reusable Presentational
components can save time in the long run
• When your team scales: As your development team grows, having a standardized way of
building components can reduce learning curves and prevent code conflicts

While the Container/Presentational pattern offers a structured approach to organizing your components,
it’s not always the best fit – especially for smaller applications, where it might be overkill. Hopefully,
with the insights provided, you’ll be better equipped to make informed decisions.

Summary
In this chapter, we delved into various strategies for component composition in Svelte, each offering
its own set of advantages and applicable scenarios. Mastering these techniques will equip you to build
Svelte applications that are not only more dynamic but also easier to maintain and scale.
We kicked off by discussing multiple ways to influence a child component. These ranged from
controlling props or using Svelte’s context to customizing styles via CSS custom properties, and even
dynamically passing content through slots.
 Summary 69

Then, we turned our attention to some of Svelte’s special elements. We explored <svelte:element>
and <svelte:component> to dynamically render various HTML elements and Svelte components.
We also learned about <svelte:self>, which allows a component to reference itself, thereby
facilitating the creation of recursive UI structures. We then applied these newfound skills to build a
JSON tree viewer as an illustrative example.
Finally, we touched upon a popular design pattern – the Container/Presentational pattern. We examined
its advantages and considered scenarios where it would be beneficial to employ this approach.
Armed with these advanced techniques, you are now better prepared to tackle complex challenges
in your Svelte projects. As we conclude our four-chapter exploration of writing Svelte components,
we’ll shift our focus in the upcoming chapter to another fundamental feature of Svelte – Svelte actions.
 Part 2: Actions

In this part, we embark on a journey to learn about Svelte actions. Across three chapters, we will
delve into three different use cases of Svelte actions. Starting with the creation of custom events using
actions, we’ll progress to integrating third-party JavaScript libraries with Svelte via actions. Finally, we’ll
unveil techniques to progressively enhance our applications, harnessing the power of Svelte actions.
This part has the following chapters:

• Chapter 5, Custom Events with Actions

• Chapter 6, Integrating Libraries with Actions
• Chapter 7, Progressive Enhancements with Svelte Actions
 5
Custom Events with Actions
Actions are one of the most powerful features of Svelte.
They are a lightweight alternative to a component that encapsulates logic and data into a reusable unit.
They help us reuse the same logic on different elements.
While components have life cycle methods such as onMount and onDestroy that run when all
the elements within the component are added to or removed from the DOM, actions are designed
to handle the logic for individual elements, running only when that specific element is added to or
removed from the DOM.
While components can receive and react to prop changes, you can pass data to actions from a parent
component to a child component. The actions will react when the data is changed and you can specify
how the action should react when that data changes.
Actions are simple yet amazingly versatile. You can use them for various things. In this and the
following chapters, we are going to explore some of the use cases of actions.
One of the many use cases of actions is to manage elements’ event listeners. Event listeners are very
common in web applications as they allow us to implement specific behaviors in response to user
actions. This makes our application more interactive and dynamic. So, it will be interesting to see how
Svelte actions can be used to help us manage event listeners.
In this chapter, we will start by looking at how we can use actions to help manage event listeners for
elements. We will follow that up with examples and exercises to reinforce the idea.
This chapter includes sections on the following:

• Managing event listeners with actions

• Creating custom events with actions
74 Custom Events with Actions

Technical requirements
You can find the projects of the chapter here: https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter05

Defining actions
Before we start to talk about using Svelte actions to create custom events, let’s quickly recap how to
define an action in Svelte.
In Svelte, an action is nothing but a function that follows an action contract. This means if a function
follows a specific function signature, it is considered an action. Here is the function signature of an action:

function action(node) {
  return {
    destroy() {}
  };
}

It is a function that optionally returns an object that has a destroy method.

In this case, since the action function follows the action contract, it is a Svelte action.
To use the Svelte action on an element, you can use the use: directive:

<div use:action />

Here, we used the Svelte action named action on the div element.
So, what will happen to the div element with a Svelte action?
When the <div> element is mounted to the DOM, Svelte will call the action function with the
reference to the <div> element:

const action_obj = action(div);

When the element is removed from the DOM, Svelte will call the destroy method from the object
returned from calling the action function:

if (action_obj && action_obj.destroy) action_obj.destroy();

We can customize the behavior of a Svelte action by using parameters. We can pass in an extra
parameter to the action function:

function action(node, parameter) {

  return {
   update(parameter) {},
 Reusing DOM event logic with custom events 75

   destroy() {},
  };
}

Also, you can add another method to the returning object, update, which will be called when the
value of the parameter changes:

action_obj.update(new_value);

To pass in an additional parameter, you can specify it in a similar syntax as the value attribute:

<div use:action={value} />

Now that we know how to use and define an action in Svelte, let’s take a look at our first use case
of actions.

Reusing DOM event logic with custom events

Before we start talking directly about actions, let’s look at an example of using the mousedown and
mouseup events to create a long-press behavior. We shall see how this simple example will lead us
on to Svelte actions:

<script>
  let timer;
  function handleMousedown() {
    timer = setTimeout(() => {
      console.log('long press!');
    }, 2000);
  }
  function handleMouseup() {
    clearTimeout(timer);
  }
</script>
<button
  on:mousedown={handleMousedown}
  on:mouseup={handleMouseup}
/>

In the preceding example, we tried to implement a long-press behavior in a button. The idea is to
press and hold the button for more than two seconds and then perform some action. As we detect it’s
a long press, we print 'long press!' into the console.
To implement the long-press behavior, I attached two event listeners: mousedown and mouseup.
The two event listeners work together. mousedown starts counting down using setTimeout for
two seconds, and mouseup clears the countdown using clearTimeout. If the user did not hold
76 Custom Events with Actions

onto the button for long enough, the timeout would not be triggered and it would not be considered
a long press. Note that to coordinate the timers between the two event listeners, the timer variable
is shared across the event listeners.
As you can see, to implement a long-press behavior, you will need two event listeners and one
shared variable.
Now, what if you need to have this long-press behavior on another button?
You can’t really share the same event listener and the variable, as you might want a different duration
to be considered as a timeout or a different behavior when a long press happens.
So, you would have to redeclare them again and remember to pair the right event with the right
event listener.

Encapsulating logic into a component

A way to recreate a different long-press button is to encapsulate it as a component, putting all the
long-press button logic into a component and reusing the component as a means to reuse the logic.
When you define a component, you define the logic as well as the elements in the component. This
means that if we put both the long-press logic and the button element into the component, we have
to use the long-press logic together with the button element and no other elements.
If you want to customize the element, maybe by using a different element, having a different style,
showing a different text content, or adding more event listeners, you would have to define the styles,
text contents, or event listeners as props of the component and pass them into the button element
in the component:

<!-- LongPressButton.svelte -->

<button
  // besides the longpress behavior
  on:mousedown={handleMousedown}
  on:mouseup={handleMouseup}
  // you need to pass down props as attributes
  {...$$props}
  // and also forward events up
  on:click
  on:dblclick
>
  <slot />
</button>

In the preceding code, we passed in extra attributes from the props into the button element along
with forwarding two events, click and dblclick, from the button element out to the component.
 Reusing DOM event logic with custom events 77

The point I am trying to make here is that if you wish to reuse event listener logic via components, you
will find yourself having to take care of other attributes that go along with the element in the component.
We can do much more with a component, but if we are trying to reuse just the long-press behavior, then
reusing it by defining it in a component is a bit overkill, and it can grow unmanageable rather quickly.
So, what other options do we have?

Encapsulating logic into an action

A better option is to use an action to encapsulate the long-press behavior.
Let’s just do that and then I’ll explain why using an action is a better approach:

function longPress(node) {
  let timer;
  function handleMousedown() {
    timer = setTimeout(() => {
      console.log('long press!');
    }, 2000);
  }
  function handleMouseup() {
    clearTimeout(timer);
  }
  node.addEventListener('mousedown', handleMousedown);
  node.addEventListener('mouseup', handleMouseup);
  return {
    destroy() {
     node.removeEventListener('mousedown', handleMousedown);
     node.removeEventListener('mouseup', handleMouseup);
    }
  }
}

With the action defined (as shown in the preceding code), we can use the action over multiple elements:

<button use:longPress>Button one</button>

<button use:longPress>Button two</button>

You can apply the action to a different type of element:

<span use:longPress>Hold on to me</span>

You can also use it alongside other attributes or event listeners:

<button use:longPress class="..." on:click={...} />

78 Custom Events with Actions

I hope you can see that the longPress action encapsulates only the long-press behavior. Unlike
the LongPressButton component, the longPress action can easily be reused in any element.
So, as a rule of thumb, when abstracting logic from an element, if you are abstracting it together with
the elements, it is okay to use a component. But if you only need to abstract the logic behavior from
an element, use actions.
Actions are a great tool for abstracting out element-level logic. But there is still one missing piece of the
puzzle: how should we add a different long-press handler for a different element? We’ll see how next.

Passing parameters to an action

So, how do we customize the behavior of our longPress action?
You could have probably guessed the answer since you’ve seen it in the previous section. We can
customize the behavior of an action by passing parameters into the action.
For example, if we want to have a different function to handle a long-press action on different button
elements, we can pass a different function into the action through action parameters:

<button use:longPress={doSomething1} />

<button use:longPress={doSomething2} />

We will then receive this function within the second argument of the action function:

function longPress(node, fn) {

  // ...
}

We call the function passed in when it is considered a long press:

fn();

We can similarly pass in other parameters, such as the duration to be considered as a long press.
In Svelte actions, you can only pass in one parameter. To pass in multiple parameters, you will have
to turn them into an object and pass them in as one parameter.
Over in our longPress action, we want to pass the function to be called when the longPress
action is detected, as well as the duration to be considered a long press, into the longPress action.
To pass both the function and the duration in, we create an object that has them as object values and
pass the object as the action parameter:

<button
  use:longPress={{
    onLongPress: doSomething1,
 Reusing DOM event logic with custom events 79

    duration: 5000
  }} />

Some of the parameters in the object could be optional, so we might need to provide a default value
when reading them within the action:

function longPress(node, { onLongPress, duration = 1000 }) {

  // if not specified, the default duration is 1s
}

At this stage, you might wonder, can the onLongPress be optional too?
It does not make much sense in our case since the main goal of our action is to detect a long press
and call the onLongPress function.
However, this is a good question.
In some other actions, you may have function handlers that are optional.
For example, what if we have a gesture action that can detect different gestures performed on
the element? In that case, each gesture callback function can be optional since you may only be
interested in one of the gestures:

function gesture(node, { onSwipe, onDrag, onPinch }) { }

The onSwipe, onDrag, and onPinch function handlers in this gesture action are optional
and could be undefined.
Instead of creating an empty function as a fallback, we should instead check whether the function is
defined before calling it:

if (typeof onSwipe === 'function') onSwipe();

This way, we don’t have to create a function unnecessarily.

However, this gets messier when you have multiple callback functions and you need to check whether
each of them has been defined before calling them.
Is there a better way to handle this?
Yes, there is.
In fact, a more idiomatic Svelte way of having an action to notify or call a function when something
happens is to dispatch a custom event.
For example, to know whether the user has long-pressed the <button> element, it would be natural
to listen to the 'longpress' custom event on the <button> element:

<button on:longpress={doSomething1} />

80 Custom Events with Actions

However, there’s no native event called 'longpress'.

But no worries, we can create a custom 'longpress' event.
Luckily, we have the longPress action to determine when the user long-presses a button. What
we need to create a custom 'longpress' event is to trigger the 'longpress' event after
determining the user has long-pressed a button in our longPress action. And so, in the line of
code where we have determined that the user is long-pressing the button, we can dispatch a custom
event from the button:

node.dispatchEvent(new CustomEvent('longpress'));

Let’s listen to the longpress custom event and create the custom 'longpress' event in our
action. Here is the final code:

<script>
  function longPress(node, { duration = 1000 } = {}) {
    let timer;
    function handleMousedown() {
      timer = setTimeout(() => {
        node.dispatchEvent(new CustomEvent('longpress'));
      }, duration);
    }
    function handleMouseup() {
      clearTimeout(timer);
    }
    node.addEventListener('mousedown', handleMousedown);
    node.addEventListener('mouseup', handleMouseup);
    return {
      destroy() {
        node.removeEventListener('mousedown', handleMousedown);
        node.removeEventListener('mouseup', handleMouseup);
     }
    }
  }
</script>
<button use:longPress on:longpress={() => {/* do something */}} />

Putting it in words, we add a longPress action to the button element, which adds logic to determine
whether the button is being long-pressed. When the button is long-pressed, the longPress action
dispatches a custom event called 'longpress' on the element. To react to and trigger specific
behaviors when the custom 'longpress' event is dispatched on the element, we can listen to the
event by using on:longpress with an event handler.
 Example – validating form inputs with custom events 81

It may feel like a roundabout way to call a function from an action by dispatching an event, but doing
it this way has a few pros:

• Whether we listen to the 'longpress' event on the button or not, the action could still
dispatch the 'longpress' custom event. So, with this approach, we don’t need to check
whether the handler is defined or not.
• Listening to the 'longpress' event using on: instead of passing the function directly into
the action would mean that you could use other Svelte features that come with Svelte’s on:
directive. For example, to only listen to the 'longpress' event once, you can use the |once
event modifier, for example, on:longpress|once.

Another way of describing what we have done with the longPress action is that the longPress
action enhances the button element and provides a new event, 'longpress', that can be listened
to on the element.
Now that we’ve learned how we can define Svelte actions, and how we can use actions to create new
events that we can listen to on an element, let’s look at a few more examples that use this technique.

Example – validating form inputs with custom events

The example that we are going to explore is using actions to validate form inputs.
When you add an input element to your form, you can add attributes such as required, minlength,
and min to indicate that the input value has to pass the constraint validation or else would be
considered invalid.
However, by default, such a validation check is only done during form submission. There’s no real-
time feedback on whether your input is valid as you type.
To make the input element validate as you type, we need to add an 'input' event listener (which will
be called on every keystroke as we type in the input element) and call input.checkValidity()
to validate the input. Now, let’s do just that:

<input on:input={(event) => event.target.checkValidity()} />

As you call the checkValidity() method, if the input is indeed invalid, then it will trigger the
'invalid' event:

<input
  on:input={(event) => event.target.checkValidity()}
  on:invalid={(event) => console.log(event.target.validity)}
/>

Unfortunately, there’s no 'valid' event. So, there’s no way to tell whether the input has passed
the validation.
82 Custom Events with Actions

It would be great if there were an event called 'validate' in which within the event details, we
can tell whether the input is valid or not. If it isn’t, we could get an error message about why the input
is invalid.
Here’s an example of how we could use the 'validate' event:

<input on:validate={(event) => {

  if (event.detail.isValid) {
    errorMessage = '';
  } else {
    errorMessage = event.detail.errorMessage;
  }
}} />

There isn’t an event called 'validate', but we can create this custom event ourselves. So, why not
create an action to create this custom event for us?
This logic is well suited to be written as an action for the following reasons:

• It can be reused in other input elements.

• This logic itself is not involved in creating or updating elements. If it were, it would probably
be better suited to be a component.

Let us write this action:

1. Firstly, this action involves listening to the 'input' event listener. So, in the code, as shown,
we are going to add an event listener at the start of the action and remove the 'input' event
listener in the destroy method. This means that whenever an element that uses this action
is added to the DOM, it will listen to the 'input' event, and when it is removed from the
DOM, the event listener will be automatically removed:
function validateOnType(input) {
  input.addEventListener('input', onInput);
  return {
    destroy() {
      input.removeEventListener('input', onInput);
    }
  };
}

2. Next, within the input handler, we are going to call checkValidity() to check whether
the input is valid. If the input is invalid, then we will read input.validity and determine
the error message:
function validateOnType(input) {
  function onInput() {
 Example – validating form inputs with custom events 83

    const isValid = input.checkValidity();

    const errorMessage = isValid ? '' : getErrorMessage(input.
validity);
  }
  // ...
}

3. Finally, we will dispatch the custom 'validate' event with isValid and errorMessage
as event details:
function validateOnType(input) {
  function onInput() {
// ...
input.dispatchEvent(
  new CustomEvent(
    'validate',
    { detail: { isValid, errorMessage } }
  )
);
  }
  // ...
}

4. Now, with this action completed, we can enhance the <input> element by adding a new
'validate' event, which will be called as-you-type, letting you know whether the input
is currently valid or invalid. It will also show the corresponding error message:
<input
  use:validateOnType
  on:validate={(event) => console.log(event.detail)}
/>

To make the validation results more useful to the user, you can use the result from the
'validate' event to modify element styles, such as setting the input border color to red
when the validation result is invalid, as shown in the following snippet:
<script>
  let isValid = false;
</script>
<input
  class:invalid={!isValid}
  use:validateOnType
  on:validate={(event) => isValid = event.detail.isValid}
/>
<style>
84 Custom Events with Actions

  .invalid { border: red 1px solid; }

</style>

Are you getting the hang of writing actions that create custom events?
Let’s try the next one as an exercise. This time, we’ll tackle one of the most common user interactions,
drag and drop.

Exercise – creating a drag-and-drop event

A drag-and-drop behavior means clicking on an element, moving the mouse while holding down the
click to drag the element across the screen to the desired location, and then releasing the mouse click
to drop the element in the new location.
A drag-and-drop behavior thus involves coordination between multiple events, namely, 'mousedown',
'mousemove', and 'mouseup'.
As we perform the drag-and-drop motion, what we are interested in knowing is when the dragging
starts, how far the element is dragged, and when the dragging ends.
These three events can be translated into three custom events: 'dragStart', 'dragMove',
and 'dragEnd'.
Let us try to implement the drag-and-drop behavior as an action that will create these three custom events:

<div
  use:dragAndDrop
  on:dragStart={...}
  on:dragMove={...}
  on:dragEnd={...}
/>

You can check the answer to this exercise here: https://github.com/PacktPublishing/

Real-World-Svelte/tree/main/Chapter05/03-drag-and-drop.

Summary
In this chapter, we saw how to define an action. We talked about one of the common patterns of
actions, which is to create custom events. This allows us to encapsulate DOM event logic into custom
events and reuse them across elements.
In the next chapter, we will look at the next common pattern of actions, which is integrating
third-party libraries.
 6
Integrating Libraries
with Actions
There are a lot of JavaScript UI libraries out there on the internet. However, at the time of writing this
book, Svelte is relatively new. Not all the UI libraries out there are written using Svelte and written
specifically for Svelte. But that does not mean that we can’t use them in our Svelte component.
There are many ways to integrate third-party JavaScript UI libraries into Svelte. In this chapter, we are
going to explore how we can do it using Svelte actions.
We will start by integrating an imaginary UI library, slowly building up our case for why Svelte actions
are suitable for the job. Along the way, I will explain how to use Svelte actions for different scenarios
and show you where Svelte actions fall short. I’ll discuss my reasonings and personal opinions on
when to choose Svelte actions and when to choose otherwise.
Following that, I will show you some real-world UI library examples. After that, we will explore
integrating UI libraries written in other frameworks, such as React and Vue, with a few more examples.
By the end of this chapter, you will see that you are not limited to using only UI libraries written in
Svelte in your Svelte application – you can reuse any UI library that is available on the internet.
This chapter covers the following topics:

• Integrating JavaScript UI libraries into Svelte

• Why we should use actions to integrate UI libraries and other alternatives
• Integrating UI libraries written in other frameworks into Svelte

Technical requirements
You can find the examples and code for this chapter here: https://github.com/
PacktPublishing/Real-World-Svelte/tree/main/Chapter06.
86 Integrating Libraries with Actions

Integrating vanilla JavaScript UI libraries into Svelte

First, we will explore UI libraries that are written in vanilla JavaScript. When we use the phrase vanilla
JavaScript, we’re referring to plain JavaScript, or JavaScript in the absence of frameworks or libraries.
There are many reasons a UI library is written in vanilla JavaScript:

• Performance reasons – it would be much easier to optimize without the abstractions from the
web framework
• The library author’s personal preference to be framework-agnostic
• The library was created predating any modern web frameworks

For us, vanilla JavaScript UI libraries are great because they do not depend on any specific framework
runtime, which is an extra overhead on top of the UI library itself.
For example, if we use a calendar component library that is implemented in React, then besides
installing the calendar component library, we would need to install React’s framework as well.
This additional dependency leads to an increased bundle size and potential conflicts with Svelte.
Therefore, when using a component library in Svelte, it is generally preferable to choose a library that
does not depend on any specific framework.
Now that we have learned about why vanilla JavaScript UI libraries are great, let’s discuss how we can
integrate them into Svelte. In this chapter, we will explore integrating libraries in Svelte using Svelte
actions, which begs the question, why do we choose to use Svelte actions to integrate a UI library?

Why use Svelte actions to integrate a UI library?

In the previous chapter, we explored how Svelte actions are useful for adding custom event handlers.
At the same time, Svelte actions function as element-level life cycle functions, making them very useful
for interfacing with third-party libraries. Now, let’s explore why that’s the case.
Let’s take a calendar component library as an example. For simplicity’s sake and to not get bogged
down by implementation details, let’s imagine the library as an imaginary library instead of using any
real-life calendar component library. This allows us to focus on the general problem itself rather than
the specific library’s implementation details.
We will look at some real-life UI libraries afterward.
To decide where the calendar component will be added to the DOM, component libraries usually
require us to specify a container element to house the library component.
For example, here, ImaginaryCalendar requires us to pass the container element as part of the
argument for the constructor:

new ImaginaryCalendar({ container: containerElement })

 Integrating vanilla JavaScript UI libraries into Svelte 87

To get the reference to an element in Svelte, we can use bind:this:

<script>
  let containerElement;
</script>
<div bind:this={containerElement} />

The containerElement variable is only updated with the reference of the element after the element
is mounted, so it can only be referred to in onMount:

<script>
  import { onMount } from 'svelte';
  let containerElement;
  let calendar;
  onMount(() => {
    calendar = new ImaginaryCalendar({ container: containerElement });
    return () => calendar.cleanup();
  });
</script>

Note that we are keeping the reference of the calendar instance because we can use it to call calendar
methods to get or set values:

calendar.setDate(date);

Also, we call calendar.cleanup() when the component unmounts for cleanup purposes.

Note
The calendar library is just an imaginary example. However, most UI libraries will provide
similar APIs or methods to retrieve or modify the component instance’s internal state and
clean up when they are no longer in use.

We need to be extra careful when we are using the calendar instance here. We want to avoid
referencing the calendar instance before it is initialized to prevent encountering a reference error
as the calendar instance is only declared and initialized after onMount:
To be safe, we should check whether the calendar instance is defined before calling any of its
methods. In the code example below, we verify if the calendar instance is defined, before calling the
calendar.setDate() method.

if (calendar) calendar.setDate(date).
88 Integrating Libraries with Actions

The need for this extra level of cautiousness is more obvious when the calendar is conditionally created:

<script>
  import { onMount } from 'svelte';
  let containerElement;
  let calendar;
  onMount(() => {
    if (containerElement) {
      calendar = new ImaginaryCalendar({ container: containerElement
});
      return () => calendar.cleanup();
    }
  });
</script>
{#if someCondition}
  <div bind:this={containerElement} />
{/if}

In the preceding code, you can see that <div> is conditionally created based on someCondition.
That is why in onMount, we need to check whether containerElement is available before we create
ImaginaryCalendar using containerElement as a container. And the calendar instance
is only available if ImaginaryCalendar is created, and thus only available if someCondition
is true.
The preceding code illustrates one of the many possibilities where the calendar instance could
be undefined.
One thing to note about this code is that it is not behaviorally correct as it does not attempt to create
ImaginaryCalendar whenever the value of someCondition turns from false to true and
does cleanup whenever it turns back to false.
This is where Svelte actions shine.

Using Svelte actions

By altering the preceding code so that it uses actions, you will see that we do not need the extra check
to ensure that containerElement is available before instantiating ImaginaryCalendar.
The following code shows how such an action could be implemented. Here, the name of our Svelte
action is calendar:

<script>
  function calendar(containerElement) {
    const calendar = new ImaginaryCalendar({ container:
containerElement });
    return {
 Integrating vanilla JavaScript UI libraries into Svelte 89

      destroy() {
        calendar.cleanup();
      }
    };
  }
</script>
{#if someCondition}
  <div use:calendar />
{/if}

This is because, when using Svelte actions, the action function will only be called with the reference
of the element whenever the element is created and mounted onto the DOM.
When the condition changes to such that the <div> element is removed from the DOM, the destroy
method of the action will be called to clean things up.
Using Svelte actions, we can now create as many ImaginaryCalendar instances as we want within
one component by adding the action to different HTML elements:
To prove my point, in the code snippet below, in addition to the original <div> element you’ve seen
in the previous examples, I am adding another <div> element and three more <div> elements
using the {#each} block. I am then applying the calendar actions to all four <div> elements to
create four more calendars, and we do not run into any errors for having multiple calendars at once.

{#if someCondition}
  <div use:calendar />
{/if}
<!-- Look we can have as many calendars as we want -->
<div use:calendar />
{#each [1, 2, 3] as item}
  <div use:calendar />
{/each}

If we were to use bind:this and onMount, we would have to repeat ourselves multiple
times by declaring multiple containerElement variables multiple times and instantiate
ImaginaryCalendar with each of the containerElement variables.
Now, with the calendar instance encapsulated within the action, how should we call the calendar
instance method to update the calendar state from the outside?
That’s what the action data is for!
90 Integrating Libraries with Actions

Adding data to Svelte actions

In the previous section, we created a calendar action and instantiated an ImaginaryCalendar
instance inside that action. If we want to call an instance method of ImaginaryCalendar outside
the calendar action, such as calling calendar.setDate(date) to set the date of the calendar,
what should we do?
There’s no way to call calendar.setDate(date) outside of the calendar action since the
calendar instance is defined within the calendar action. A workaround is to pass date through
action data – that is, we can provide date as action data and call calendar.setDate(date)
with the date passed in.
For example, in the following code snippet, we’re passing date into the calendar action:

<div use:calendar={date} />

In the calendar action, we call calendar.setDate(date) with the date passed in. In addition
to that, we define an update method in the action, such that whenever the date that’s passed to the
calendar action changes, Svelte will call calendar.setDate(date):

<script>
  function calendar(containerElement, date) {
    const calendar = new ImaginaryCalendar({ container:
containerElement });
    calendar.setDate(date);
    return {
      update(newDate) {
        calendar.setDate(newDate);
      },
      destroy() {
        calendar.cleanup();
      }
    };
  }
</script>
<div use:calendar={new Date(2022, 10, 5)} />

Here, we can pass different dates to the different calendar instances:

{#each dates as date}

  <div use:calendar={date} />
{/each}

That’s great!
 Integrating vanilla JavaScript UI libraries into Svelte 91

Now, what if you want to call a different calendar instance method, such as calendar.setMode(),
whenever the mode changes?
You could pass both date and mode into the action:

<div use:calendar={{ date, mode }} />

In that case, the calendar action would need to handle both date and mode:

function calendar(node, { date, mode }) {

  const calendar = new ImaginaryCalendar({ container: containerElement
});
  calendar.setDate(date);
  calendar.setMode(mode);
  return {
    update({ date: newDate, mode: newMode }) {
      calendar.setDate(newDate);
      calendar.setMode(newMode);
    },
    destroy() { ... }
  };
}

The update method of the calendar action will be called whenever either date or mode
changes. This means that in the preceding code, we are calling both calendar.setDate() and
calendar.setMode() whenever either date or mode changes. This may not have any apparent
consequences, but we may be doing an unnecessary job.
A workaround for this is to keep track and always check whether date or mode have changed in the
update method. This is how we can do this:

function calendar(node, { date, mode }) {

  const calendar = new ImaginaryCalendar({ container: containerElement
});
  calendar.setDate(date);
  calendar.setMode(mode);
  return {
    update({ date: newDate, mode: newMode }) {
      if (date !== newDate) {
        calendar.setDate(newDate);
        date = newDate;
      }
      if (mode !== newMode) {
        calendar.setMode(newMode);
        mode = newMode;
      }
92 Integrating Libraries with Actions

    },
    destroy() { ... }
  };
}

In the preceding code, we’re checking if newDate is different from the current date, and if it is
different, then we call the calendar.setDate() method and update our current reference of
date. We do a similar thing for mode.
This works. However, as you can see, it is more code and more complex than what we set up when we
first started creating the calendar action.
And what if you want to call a calendar instance method that is not tied to any data, such
as calendar.refreshDates()?
This is where using actions falls short.

An alternative to Svelte actions

Remember the previous example, where we used bind:this and onMount to initialize Imag-
inaryCalendar?
We said that the approach is not versatile enough and falls short if we need to do the following:

• Conditionally render the container and create ImaginaryCalendar

• Have multiple calendars within the same component

These drawbacks are all true, but there’s a use case in which it is perfectly fine to use bind:this
and onMount to initialize ImaginaryCalendar. This is when the conditions we mentioned
previously are never true:

• We do not need to conditionally render the container

• We do not need to have multiple calendar instances within the same component (this is not
exactly true, but we will come back to this)

I’m not sure whether you are thinking the same thing right now but allow me to cut the suspense.
This is when we want to use ImaginaryCalendar as a Svelte component.
Within the ImaginaryCalendar Svelte component itself, we will have only one container element,
and it will always be available:

<!-- ImaginaryCalendarComponent.svelte -->

<script>
  import { onMount } from 'svelte';
  let containerElement;
 Integrating vanilla JavaScript UI libraries into Svelte 93

  let calendar;
  onMount(() => { ... });
</script>
<div bind:this={containerElement} />

You can then use this component conditionally:

<script>
  import ImaginaryCalendarComponent from './
ImaginaryCalendarComponent.svelte';
</script>
{#if someCondition}
  <ImaginaryCalendarComponent />
{/if}

Alternatively, you can use it however many times as you like:

{#if someCondition}
  <ImaginaryCalendarComponent />
{/if}
<!-- Look we can have as many calendars as we want -->
<ImaginaryCalendarComponent />
{#each array as item}
  <ImaginaryCalendarComponent />
{/each}

Here, we’ve swapped out elements using actions, <div use:calendar />, with Svelte components,
<ImaginaryCalendarComponent />.
This is perfectly normal.
In the previous chapter, we contemplated between abstracting logic through components or abstracting
through actions.
In this scenario, we are looking at abstracting the logic of instantiating a UI library using an element
as a container, and we could abstract it into a Svelte action or a Svelte component.
Both options are equally fine.
Both options are designed for this.
So, which options should you choose? Let’s find out.

Choosing between Svelte actions and Svelte components

Here are my personal preferences when faced with a choice between either option.
94 Integrating Libraries with Actions

Choose a Svelte action to integrate with a UI library when you are looking for an option that provides
the following aspects:

• Is more lightweight. A Svelte component has slightly more overhead compared to a Svelte action.
• Only passes zero to one data into the UI library component instance. If you were to pass in
two or more pieces of data into the action, then the update method of the action will be called
whenever any part of the data changes.

You should choose a Svelte component if you are looking for an option that provides the following:

• Allows more room for optimization and finer control

• Allows you to call UI library component instance methods directly
• Allows you to pass in child content into the UI component library

We did not discuss this much, but integrating a UI library as a component opens up the possibility
of passing additional content into the UI component library through slots:

<ImaginaryCalendarComponent>
  <!—Customize how each cell of the calendar looks ––>
  <svelte:fragment sl"t="c"ll" let:date>
    {date}
  </svelte:fragment>
</ImaginaryCalendarComponent>

If you are interested in learning more about slots and how to compose components in Svelte, read
Chapter 4, where we explored this topic extensively. Now that we’ve covered how we could integrate
a UI library using Svelte actions, why we should use Svelte actions, as well as the alternatives and
considerations, let’s take a look at a real-world example, Tippy.js.

Example – integrating Tippy.js

Tippy.js is a tooltip, popover, dropdown, and menu library.
I do not have any affiliation with the Tippy.js library, and the reason I chose Tippy.js as an example
is purely by chance. Nonetheless, Tippy.js has a nice and simple API, making it a good candidate for
an example.
First, let’s look at the Tippy.js documentation: https://atomiks.github.io/tippyjs/.
After installing the tippy.js library using a package manager of our choice, we can then import
Tippy.js into our code:

import tippy from 'tippy.js';

import 'tippy.js/dist/tippy.css';
 Example – integrating Tippy.js 95

Now, we can initialize tippy with the following constructor function:

tippy('#id');
tippy(document.getElementById('my-element'));

Here, we pass in the element where Tippy.js should provide a tooltip.

You can specify any customizations of the tooltip's content through the data attributes of the element,
which Tippy.js will pick up as it initializes:

<button data-tippy-content="hello" />

Alternatively, you can pass this in the constructor:

tippy(element, { content: 'hello' });

To update the content after initialization, call the Tippy.js setContent method:

tooltipInstance.setContent("bye");

To permanently destroy and clean up the tooltip instance, Tippy.js provides the destroy method:

tooltipInstance.destroy();

Here, we have all we need to create a tippy action. We have methods to do the following:

• Create the tippy tooltip – tippy(…)

• Clean up the tippy tooltip – tooltipInstance.destroy()
• Update the tippy tooltip – tooltipInstance.setContent(…)

Let’s take a look at what the tippy action should look like.
Here is how I want it to look:

<div use:tippy={tooltipContent} />

In the preceding code snippet, we applied our tippy action to a <div> element. The content within
the tooltip created by Tippy.js should be passed into the tippy action as action data, represented by
the tooltipContent variable. Whenever tooltipContent changes, the action should react
to it and update the tooltip.
So, let’s write our tippy Svelte action. Here’s the scaffolding for an action:

function tippy(element, content) {

  // TODO #1: initialize the library
  return {
    update(newContent) {
96 Integrating Libraries with Actions

      // TODO #2: do something when action data changes

    },
    destroy() {
      // TODO #3: clean up
    }
  };
}

As you can see, we created the tippy action based on the Svelte action contract: a function that
returns an object with destroy and update methods.
I left three TODOs in the code, each marking different stages of Svelte actions. Let’s go through each
one and fill them up. The first TODO is where the action will be called after the element is created and
mounted onto the DOM. Here, we are given the element the action is applied to and the action data,
and we should use it to initialize the Tippy.js tooltip:

// TODO #1: initialize the library

const tooltipInstance = tippy(element, { content });

The second TODO is inside the update method. This method will be called every time the action data
changes. Here, we need to call the Tippy.js tooltip instance to reflect the dataset in the Svelte component:

// TODO #2: do something when action data changes

tooltipInstance.setContent(newContent);

The third TODO is inside the destroy method. This method will be called after the element is
removed from the DOM. Here, we need to do the cleanup on the Tippy.js tooltip instance that we
created in the action:

// TODO #3: clean up

tooltipInstance.destroy();

And that’s it – we now have a working tippy action that integrates the Tippy.js tooltip and will show
a tooltip with customizable content whenever we hover over the element.
You can find the complete code on GitHub: https://github.com/PacktPublishing/
Practical-Svelte/tree/main/Chapter06/01-tippy.
Let’s look at one more example, through which I want to show you one more thing that you can do
with actions when integrating with UI libraries.
The UI library we are going to look at next is CodeMirror.
 Example – integrating CodeMirror 97

Example – integrating CodeMirror

CodeMirror is a code editor component that has many great features for editing, such as syntax
highlighting, code folding, and more.
You can find the CodeMirror documentation at https://codemirror.net/.
At the time of writing, CodeMirror is currently at version 5.65.9.
After installing the codemirror library using the package manager of our choice, we can import
codemirror into our code:

import CodeMirror from 'codemirror';

import 'codemirror/lib/codemirror.css';

Now, we can initialize CodeMirror with the following constructor function:

const myCodeMirror = CodeMirror(document.body);

Here, we pass in the element where we want the CodeMirror code editor to be.
Before I continue, at this point, note that we are looking for the same set of things from CodeMirror:

• Methods to initialize CodeMirror

• Any method needed to clean up a CodeMirror instance
• Any method to update a CodeMirror instance

I am going to leave it to you to complete the checklist and figure it out.

However, allow me to draw your attention to one particular API from the CodeMirror instance:

myCodeMirror.on('change', () => { ... });

The on method from CodeMirror allows the CodeMirror instance to listen to events and react to them.
So, if we want to add event listeners to the CodeMirror instance from outside of the action, how
should we do it?
In the previous chapter, we saw that we could create custom events on the element using actions.
This means that we can allow users to listen to the 'change' event from the element that uses the
codemirror action:

<div use:codemirror on:change={onChangeHandler} />

98 Integrating Libraries with Actions

To make this happen, you can dispatch an event from within the action:

function codemirror(element) {
  const editor = CodeMirror(element);
  editor.on('change', () => {
    // trigger 'change' event on the element
    // whenever the editor changes
    element.dispatchEvent(new CustomEvent('change'));
  });
}

Remember to check whether you need to clean up or unlisten to any event in the destroy method
so that you don’t cause any unwanted behavior.
And that’s it!
The rest of the action is left to you as an exercise.
You can find the complete code on GitHub: https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter06/02-codemirror.
In this section, we learned how to integrate vanilla UI libraries into Svelte. However, not all UI libraries
are implemented independently without any framework. Sometimes, the library you are looking for
might be implemented in a different framework, such as React or Vue. In such cases, how can you
integrate them into a Svelte application? That is what we will explore next.

Using UI libraries written in other frameworks

It is not impossible to use components from other frameworks in Svelte.
However, doing so will introduce the framework’s runtime and other overheads that come along with
the framework. The runtime usually includes code to handle reactivity and normalize browser APIs
and events. Each framework usually ships its own code for this logic and does not share it with other
frameworks. The runtime for React version 18.2.0 weighs 6.4 kB when minified, which is additional
code you need to include when you want to use a React component within Svelte.
So, this is not recommended unless it is necessary.
The reason this section has been included in this book is more for educational purposes and to
demonstrate that this is possible, as well as what needs to be done to make it happen.

Creating components in various frameworks

Each framework usually provides an API that takes in a container element and the framework
component as the root of the application.
 Using UI libraries written in other frameworks 99

In this section, we’re going to take a look at React and Vue, the two most popular JavaScript frameworks
at the time of writing.
For example, in React 18, we use the createRoot API:

import ReactDOM from 'react-dom';

const root = ReactDOM.createRoot(container);
root.render(<h1>Hello, world</h1>);

The preceding code uses the JSX syntax, which is not part of the standard JavaScript language syntax.
It is syntactic sugar for jsx:

import { jsx } from 'react/jsx-runtime';

const root = ReactDOM.createRoot(container)
root.render(jsx('h1', { children: 'Hello, world' }));

You will have to write the preceding code if you do not configure any transpiling process in your code
to turn JSX syntax into valid JavaScript.
On the other hand, In Vue 3, there is the createApp API:

import { createApp } from 'vue'

import App from './App.vue'
const app = createApp(App);
app.mount(container);

The Vue framework uses the word application in the documentation, mentioning that the createApp
method is used to create a new application instance. The word application is aptly used as integrating
component libraries written in other frameworks is very much like starting a new sub-application
within our Svelte application.
You may have also started to notice the similarity between the APIs from these frameworks and other
UI libraries that we’ve seen so far – all of them take in a container element so that they know where
to render the content or apply changes.
Similar to integrating UI libraries using actions, after figuring out what APIs we can use to render a
component within a container element, the next thing we must check is whether there is any API to
clean up when it is no longer needed.

Cleaning up the components in various frameworks

Depending on the underlying framework of the component library, different APIs are provided to
clean up whenever the component is no longer needed.
In React, there is a method for this called unmount:

root.unmount();
100 Integrating Libraries with Actions

In Vue, it is also called unmount:

app.unmount();

The next thing we need to check is whether there is any API to pass data into our component and
API so that they can be updated later.

Updating the component in various frameworks

Similar to different frameworks having different APIs for cleaning up, frameworks provide different
APIs to update the component with new data.
If you are familiar with React, you can pass data into a React component through props:

<Component prop_name={value} />

This is akin to the props in a Svelte component.

The preceding code desugars into the following:

jsx(Component, { prop_name: value });

To update the props of the component, React allows us to call root.render again with the same
component but with a different prop value:

root.render(jsx(App, { prop_name: 123 }));

// some time later
root.render(jsx(App, { prop_name: 456 }));

React will reconcile this internally and figure out how to update the DOM.
On the other hand, in Vue, you can pass data through props in the createApp API:

const app = createApp(Component, { prop_name: value });

However, to the best of my knowledge, there’s no straightforward way to update the props’ value from
the outside.
However, you could use a Vue Composition API, such as ref(), to create a reactive and mutable
ref. With that, you could modify the ref instead of updating the props directly:

const value = ref(123);

const app = createApp(Component, { prop_name: ref });
// some time later
value.value = 456;

It is okay if you are not familiar with how React and Vue work. This book is for Svelte, not for React or Vue.
 Integrating react-calendar into Svelte 101

The most important thing to take home from this is that when integrating a UI library, whether it is
in vanilla JavaScript, React, or Vue, we look for three things:

• A method to create the component with a container element

• A method to clean up the component instance
• A method to pass in data and update the data

If you are familiar with a framework, you will be able to figure out a way to do all these things.
With that out of the way, let’s take a look at a real-world example, where we will integrate a React
calendar library, react-calendar.

Integrating react-calendar into Svelte

The react-calendar library is a calendar component library written in React.
You can read more about it here: https://projects.wojtekmaj.pl/react-calendar/.
The react-calendar library takes in various props for customization purposes. But for demonstration
purposes, we are only going to focus on two props, value and onChange, which allow us to control
the selected date of the library.
We pass the selected date through a prop named value. The onChange prop, on the other hand, is
used to pass in an event handler that will be called when the value changes from within the calendar
component. We saw how we could handle event handlers in a UI library in the previous section when
we discussed CodeMirror.
So, here is what I think using the calendar action would look like:

<div
  use:calendar={selectedDate}
  on:change={(event) => selectedDate = event.detail}
/>

Here, event.detail is the data attached to the custom 'change' event, which would be the date
value that’s sent from the react-calendar component through the onChange props.
Now that we know what our calendar action would look like, let’s write the action out.
Again, here’s the scaffolding of an action:

function calendar(element, date) {

  // TODO #1: render the react-calendar into the element
  // TODO #2: the onChange handler to dispatch a new custom event
  return {
    update(newDate) {
102 Integrating Libraries with Actions

      // TODO #3: re-render the calendar again when there's a new date
value
    },
    destroy() {
      // TODO #4: clean up
    }
  };
}

Here, I’ve created a basic code structure of a Svelte action and left a few TODOs within the code. The
first two TODOs are to set up the calendar instance with the element the action is applied to. The
third TODO is to handle when a new date is passed into the action, and the last TODO is to clean up
when the element is removed from the DOM.
So, let’s fill up the TODOs.
For the first TODO, let’s create a React root and render our react-calendar component:

import { jsx } from 'react/jsx-runtime';

import ReactDOM from 'react-dom';
function calendar(element, date) {
  # TODO #1: render the react-calendar into the element
  const app = ReactDOM.createRoot(element);
  app.render(jsx(Calendar, { value: date, onChange }));
  // ...
}

Here, we passed in onChange, which we have not defined yet.

Let’s do this now:

# TODO #2: the onChange handler to dispatch a new custom event

function onChange(value) {
  element.dispatchEvent(
  new CustomEvent('change', { detail: value })
  );
}

In the preceding code snippet, whenever onChange is called, we will dispatch a new custom event,
with value passed in as the detail for the custom event.
The third TODO is the content for the update method. Whenever a new date value is passed in from
the action, we will re-render the Calendar component again:

// TODO #3: re-render the calendar again when there's a new date value
app.render(jsx(Calendar, { value: newDate, onChange }));
 Summary 103

In the last TODO, in the destroy method, we unmount our Calendar component:

// TODO #4: clean up

app.unmount();

And that’s it.

You can find the complete code on GitHub: https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter06/03-react-calendar.
With that, you’ve written a Svelte action that integrates a component library from a different framework,
React, into Svelte, and you set up and updated the component’s value in a controlled manner.

Summary
In this chapter, we learned how to use actions to integrate UI libraries, either written in vanilla JavaScript
or any other frameworks into Svelte. We went through two real-world examples – integrating Tippy.js
and react-calendar into Svelte using Svelte actions. In both examples, we went through a step-
by-step process of writing out a Svelte action. We started by creating the structure of a Svelte action
and then filled up the steps within the action for when the Svelte action is initialized as the element is
created, when the data changes, and when the element is removed from the DOM. We also discussed
why we choose to use Svelte actions, as well as the other alternatives and considerations when it comes
to integrating UI libraries.
In the next chapter, we will look at the next common pattern of actions, which is to progressively
enhance your elements.
 7
Progressive Enhancement
with Actions
Progressive enhancement is a design philosophy in web development that emphasizes providing content
and core functionality to everyone while delivering an enhanced experience to users who can afford it.
In this chapter, we will start with a more in-depth discussion of what progressive enhancement is.
There are many ways to achieve progressive enhancement in your application; we will explore one of
them by using Svelte actions. I will explain my reasoning as to why I think Svelte actions are designed
for this use case.
Toward the end of the chapter, we will go through a few examples of using Svelte actions to progressively
enhance our application.
By the end of this chapter, you will be able to build an application that follows the principles of
progressive enhancement and supports as many user devices as possible.
This chapter covers the following topics:

• What is progressive enhancement?

• Why use Svelte actions for progressive enhancement?
• Examples of progressive enhancements using Svelte actions

Technical requirements
You can find the examples and code for this chapter here: https://github.com/
PacktPublishing/Real-World-Svelte/tree/main/Chapter07.
106 Progressive Enhancement with Actions

What is progressive enhancement?

The most important thing in progressive enhancement is missing from the phrase itself. Something
that is implied here is where are we progressively enhancing from.
The main idea of progressive enhancement is to provide a great baseline of essential content and
core functionality to everyone, regardless of the browser software, device hardware, or the quality
of the internet connection. Older browser software may not support newer JavaScript syntaxes and
CSS features; older device hardware may take up more time to process and render your web page; a
slower internet connection may take longer to load the resources needed to display your web page.
How do we ensure our web page stays usable for as many users as possible? Think about this for a
while—I will come back to it later.
For users who can afford better browsers, more powerful hardware, and higher internet bandwidth,
we progressively provide an enhanced experience to them. We leverage the power of JavaScript and
CSS to surprise and delight our users.
How do we differentiate between the users and decide when to deliver an enhanced experience?
There is another term that is often used to compare with progressive enhancement, and that is
graceful degradation. Graceful degradation starts from a feature-rich baseline and gracefully handles
the situation when the user browser can no longer support the feature by replacing it with a simpler
alternate experience. A lot of the time, these features start with a more complex assumption, and so
execution-wise, it is much harder to gracefully degrade to all kinds of users.
Progressive enhancement, on the other hand, starts from a baseline that works for most users and
slowly works its way up by adding more features. So, we can be sure that when a new feature is not
loaded or does not work, users will still have a basic working web page.
So, let’s get back to our question: How do we make sure that our web page is usable to all users? We’ll
uncover that in the following section.

Progressively enhancing the web experience

One of the ways is to make sure that we follow the standards. HTML, CSS, and JavaScript are the
main languages of the web. We make sure we only use language features that are part of the standard
specifications. Features that have been part of the specifications for longer have a higher probability
of being implemented by all browsers. The latest and hottest features are less likely to be available in
all browsers.
So, build your web pages with standard HTML, CSS, and JavaScript.
This then leads to the next question: How do we deliver a differentiating experience to users based on
their browser, device, and network capabilities?
 What is progressive enhancement? 107

There are many ways to do this.

One key idea for most approaches is to build your application layer by layer. Start with the first
layer of core functionality and make sure everything works. Then, add subsequent layers to enhance
the experience.

Building web pages in layers

One example of building web pages in layers is to build the base content and functionality in HTML
as the base layer, then add styles, transitions, and animations using CSS as the next layer. Finally,
complex interactivity is added using JavaScript as the final layer.
This aligns with how the browser loads your website.
Whenever a user visits your website, the first thing the browser downloads from your website is HTML.
HTML describes your content. HTML tags such as <p>, <div>, and <table> describe how your
content should be laid out on the screen. HTML tags such as <form>, <input>, and <a> describe
how users can submit data and interact with your content.
With HTML, your website should already provide basic content and functionality to the user.
And if you have <link rel="stylesheet"> within the HTML, with reference to external
CSS files, the browser will then make separate requests to download the CSS resource and parse and
apply the CSS styles to your document. This will, in turn, enhance the look and feel of the default
browser styles. HTML can offer basic layouts, but with CSS, you can have advanced layouts, such as
flex layouts, grid layouts, and so on.
On the other hand, if you have <script> tags in your HTML, the browser will seek them out and
load the referenced JavaScript files, and as soon as the JavaScript files are downloaded, the browser
will parse them and execute them. JavaScript can be used to dynamically make changes to the DOM,
perform computations, and add interactivity to the website.
Without JavaScript, HTML forms alone allow users to submit data; however, upon submission, the browser
will navigate to a new location based on the form action. With JavaScript, you can make an asynchronous
HTTP request to send data to the server while the user can continue browsing on the same page.
So, as you can see, serving HTML as a base experience layer and adding CSS and JavaScript on top
for an enhanced experience is progressive enhancement.
Users with older browsers, slower hardware, and lower internet bandwidth can still view and interact
with your website with just HTML while waiting for CSS and JavaScript to be downloaded and executed
for a more enhanced experience.
Hopefully, you are now convinced about the first HTML approach. But what if your website content
is dynamic? How do you generate dynamic HTML for the user? Do you need to write separate code
to generate dynamic HTML?
108 Progressive Enhancement with Actions

No, you don’t.

Svelte supports server-side rendering (SSR). What that means is that the same Svelte component can
be used to render content on the browser as well as generate HTML on the server side.
You can set it up yourself (however, that’s beyond the scope of this book), or you can use meta
frameworks such as SvelteKit, which comprehensively sets out how everything should work.
One thing to take away from here is that no matter your setup, it is possible to write your Svelte
component as is and have the same Svelte component code that you write used in both generating
HTML on the server side and rendering content on the browser side.
This begs the question: Does all the code work the same way on both the server side and the browser
side? Is there code that only runs on the server but not on the browser, or the other way around?
Well, not all code runs both on the server side and on the browser. Svelte actions, along with bind:
directives and on: event listeners, are Svelte features that do not run on both the server side and the
browser side. Svelte actions only run on the browser side and not on the server side. That’s because
Svelte actions run after an element is added to the DOM. Since there’s no DOM when generating
HTML strings on the server side, Svelte actions do not run on the server side.
This makes Svelte actions the perfect candidate for progressive enhancements.

Svelte actions for progressive enhancements

In the previous section, we learned about progressive enhancements and the concept of building web
pages in layers to achieve progressive enhancements.
In this section, we will delve deeper into the role of Svelte actions, which enable us to add an extra layer of
interactivity to existing HTML elements, making them a natural fit for creating progressive enhancements.
Let’s begin by examining a code example in Svelte to understand how Svelte actions fit into this approach.
Now, imagine the following code:

<button use:enhance />

When rendered from the server side, you get HTML that looks like this:

<button></button>

The HTML button element alone should be able to do what a button element is supposed to do: be
clickable and able to submit forms.
But as JavaScript is loaded and the Svelte component code is executed, the enhance action is run
with the button element, allowing the action code to enhance the button element.
 Example – previewing a link with a <a> element 109

Here are some examples of what the action could do: show a helpful tooltip upon hovering, provide
a loading indicator when pressed, and so on.
Users with older browsers that are having trouble running the Svelte component code on the client
side might still be able to use and interact with just the default HTML button element and experience
an unenhanced version of the web page.
As we have gathered from this section, Svelte actions allow us to add another layer of interactivity to
existing HTML elements. They are a naturally good candidate for designing progressive enhancements.
With that, let’s look at a few examples of using actions to progressively enhance HTML elements.
The first example we are going to look at is progressively enhancing the <a> element.

Example – previewing a link with a <a> element

In our first example, we will explore how to progressively enhance a <a> element to display a preview
when hovered upon.
Here, the browser receives HTML that contains a <a> tag with a href attribute, like this:

<a href="..." />

It then creates a hyperlink. When you click on the hyperlink, the browser will navigate to the destination
specified in the href attribute.
This is the default behavior of the <a> element.
As the user loads the JavaScript, we want to make the <a> element do more than just navigate to a
new location.
We are going to enhance the <a> element by having it show the destination location when hovering
over it.
To do that, we are going to create a preview action and use it on the <a> element:

<script>
  function preview(element) {
  }
</script>
<a href="..." use:preview>Hello</a>

Regardless of how the preview action is implemented when rendering the preceding code on the
server side, Svelte will generate the following HTML:

<a href="..." />Hello</a>

This is because Svelte actions never run on the server side.

110 Progressive Enhancement with Actions

As soon as your user receives the HTML response, they can start clicking on the link and navigating
to new locations. You now have a workable application with just the HTML.
Depending on the user’s network condition, your JavaScript code, compiled from your Svelte
component, may take a longer time to arrive. But that does not stop the user from using the hyperlinks
to navigate away.
Only when the JavaScript is loaded and executed does Svelte run the preview action with the <a>
element that is on the DOM and enhances the <a> element’s behavior.
The point here is that, as much as possible, we make the core functionality of our application workable
with just the HTML, and we add a layer of enhancement with JavaScript, which may come at a much
later time depending on the user’s network condition.
Enough with the progressive enhancement philosophy. Let’s look at how we can implement this
preview action.
We want the preview action to show a floating popup containing the content of the link when we
move our mouse cursor over the link and hide it as we move our mouse cursor away.
We can achieve that with 'mouseover' and 'mouseout' events. Here’s how:

function preview(element) {
  element.addEventListener('mouseover', onMouseOver);
  element.addEventListener('mouseout', onMouseOut);
  return {
    destroy() {
          element.removeEventListener('mouseover', onMouseOver);
          element.removeEventListener('mouseout', onMouseOut);
        }
  };
}

In the preceding code snippet, we add 'mouseover' and 'mouseout' event listeners at the
beginning of the preview action. Additionally, we ensure proper cleanup by removing both the
'mouseover' and 'mouseout' event listeners in the destroy method. Before we figure out
how to implement onMouseOver and onMouseOut, we need to first decide what the floating
popup would look like and how would we lay it out in the DOM.
To show the content of the link target, we are going to use the <iframe> element, which allows us
to embed another HTML page into the current one:

<iframe src="..." />

To make the <iframe> element float above other content instead of being part of the document
flow, we will need to modify the CSS position property of the <iframe> element by using either
position: fixed or position: absolute.
 Example – previewing a link with a <a> element 111

If we used position: fixed on the <iframe> element, then the <iframe> element would be
positioned relative to the viewport. To place the <iframe> element right next to the <a> element,
we will have to figure out the position of the <a> element relative to the viewport and calculate the
top and left values to place our <iframe> element.
On the other hand, if we used position: absolute, then the <iframe> element would be
positioned relative to the nearest positioned parent element. We could place the <iframe> element
inside the <a> element and make the <a> element a positioned parent element by specifying
position: relative on the <a> element (the position: relative CSS property is to
be positioned relative to its current position). The <iframe> element would then be positioned
relative to its parent, the <a> element.
Either approach has its pros and cons. I am going to use the second approach here, which is to use
position: absolute. I will have to modify the content and the position CSS property of
the <a> element, but I could get away from needing to perform calculations if I use position:
absolute instead of position: fixed.
Here’s what the DOM will look like after we place the <iframe> element inside the <a> element:

<a href="..." style="position: relative">Hello<iframe src="..."

style="position: absolute"/></a>

Our task now is to programmatically create and insert the <iframe> element in the onMouseOver
function. The following code snippet illustrates how to do this:

function preview(element) {
  // make the <a> element position relative
  element.style.position = 'relative';
  let iframe;
  function onMouseOver() {
    iframe = document.createElement('iframe');
    iframe.src = element.getAttribute('href');
    iframe.style.position = 'absolute';
    iframe.style.left = 0;
    iframe.style.top = '100%';
    element.appendChild(iframe);
  }
  // ...
}

In the preceding code snippet, we set the CSS position property of the <a> element to 'relative'.
In the onMouseOver function, which will be called when the mouse is hovered over the <a> element,
we programmatically create a <iframe> element, style it, and insert it into the <a> element.
112 Progressive Enhancement with Actions

In the previous code, we are using the DOM API such as document.createElement() and
element.appendChild(). Since the <iframe> element is programmatically created, we are
also programmatically modifying its style attribute.
Luckily, in this example, we are only creating one element, but you can imagine how this could easily
grow messy if we were to create more elements.
Since we are learning Svelte here and Svelte is designed to abstract out these imperative DOM
instructions into declarative Svelte syntax, why not leverage Svelte in our action?
We can replace the previous imperative code with a Svelte component, like so:

<!-- IframePopup.svelte -->

<script>
  export let src;
</script>
<iframe {src} />
<style>
  iframe {
    position: absolute;
    left: 0;
    top: 100%;
  }
</style>

The preceding code snippet shows a Svelte component that contains a <iframe> element. This
<iframe> element is equivalent to the one we created programmatically in the previous code snippet.
It has the same CSS styles applied.
The Svelte component exposes a prop called src, and the value of the src prop will be used to set
the value of the src attribute of the <iframe> element. Now, instead of calling the DOM APIs to
create the <iframe> element, we can instantiate our Svelte component and pass the desired src
value as the src prop to the component. In Svelte, you can instantiate a component by using the
new keyword along with the component’s constructor, passing in any required props as part of the
constructor arguments. The Svelte component will then render the <iframe> element with the
specified src attribute value based on the passed prop. This simplifies the process of creating and
managing a <iframe> element within our Svelte application:

import IframePopup from './IframePopup.svelte';

function preview(element) {
  // ...
  function onMouseOver() {
    iframe = new IframePopup({
      // target specifies where we want the component
      // to be inserted into
 Example – previewing a link with a <a> element 113

      target: element,
      // we are passing the href value into the
      // component through props
      props: { src: element.getAttribute('href') },
    });
  }
  // ...
}

In the preceding code snippet, we’ve replaced the DOM operations in the onMouseOver function
to instantiate the IframePopup Svelte component.
We need to remember to remove the popup when we move our mouse away from the link. Here’s
how we can do that:

function onMouseOut() {
  iframe.$destroy();
}

Although we are only inserting one HTML element into the DOM through Svelte actions, we’ve seen
that it’s much more manageable to encapsulate that into a Svelte component and instantiate the Svelte
component instead of manually creating HTML elements.
We can then leverage Svelte to create scoped styles, as well as add other interactive logic to the elements
we are going to create.
We could also add transitions to elements; instead of appearing abruptly upon hovering, we can make
the popup fade by using a fade transition, like so:

<!-- IframePopup.svelte -->

<script>
  import { fade } from 'svelte/transition';
  // ...
</script>
<iframe {src} transition:fade />

By default, transitions are not played when the component is first created. So, we need to pass in the
intro: true option to play the transition:

iframe = new IframePopup({

  target: element,
  props: { src: element.getAttribute('href') },
  // play transition when created
  intro: true,
});
114 Progressive Enhancement with Actions

In the preceding code snippet, we pass in intro: true to the IframePopup constructor.
.

We now have a link that shows a preview in a popup that fades in upon hovering. Try simulating
loading the page with a slow network. Most browsers provide developer tools to simulate network
speed. For example, if you are using Google Chrome, then you can open the Developer Tools, find
the Network conditions tab, look for the Network throttling section, and choose the Slow 3G preset.
Try reloading your page, and you’ll find that once you see the link (albeit the JavaScript file is still
loading), the link is immediately working; clicking it will navigate you to the destination. As the
JavaScript is finally loaded into the browser, your link is now enhanced, and you are now able to hover
over the link and see a preview.
With our link preview action done, let’s look at another common component in a web app, forms,
and see how we can progressively enhance a form element.

Example – progressively enhancing a form

A <form> element is a section of a document that can contain inputs that will be used to
submit information.
By default, when you submit a form, the browser will navigate to a URL to process the form submission.
This means that the user will lose the state they are in as they navigate away from the current page
when they submit the form.
However, with the ability to make asynchronous requests through the browser fetch API, we can
now submit data through API requests without leaving the current page, and stay where we are.
This means that if the site is playing music, video, or animation, they will still be playing while we
make asynchronous API calls.
Our task now is to create an action to enhance the form element so that the enhanced form will not
navigate to a new location, but rather submit the form data asynchronously.
For lack of a better name, I am going to call this enhancing action enhance.
Before we proceed to implement the enhance action, let’s recap on the default form behavior.

The default form behavior

When you have a <form> element, by default when you hit the Submit button, it will navigate to the
location indicated by the action attribute, carrying along with it the value filled in the <input>
elements within the <form> element.
For example, imagine you have the following form:

<form action="/foo">
  <input name="name" />
 Example – progressively enhancing a form 115

  <input name="address" />

  <button>Submit</button>
</form>

When you hit the Submit button, the browser will navigate to /foo?name=xxx&address=yyy,
carrying the form data via query parameters.
A <form> element can define the HTTP method to use to submit the form, by specifying the
method attribute.
For example, the following form will navigate to /foo via a POST request:

<form action="/foo" method="post">...</form>

Form data will be sent as a request body for the POST request.
Depending on the server’s implementation for the /foo endpoint, the server can choose how to
process the data and what to show on the /foo page. Sometimes, the server may decide to redirect
back to the current page after processing the data. In that case, having an action that can replace the
default form action and submit the form data asynchronously instead would be very useful, since we
will eventually come back to the same page.
Now we know the default form behavior, let’s figure out what we need to implement the enhance action.

Implementing the enhance action

Let’s break the problem down.
First, we need to figure out how to know when a user submits a form. Then, we need to prevent the
default form behavior and then make an asynchronous API call to submit the form asynchronously,
and finally reset the form to the initial state, similar to what you would see after the server redirects
back to the same page.
To figure out when the user submits a form, we could listen to the 'submit' event on the
<form> element:

form.addEventListener('submit', handleSubmit);

To prevent the default form behavior, we call event.preventDefault() on the 'submit'

event listener to prevent the default submit behavior:

function handleSubmit(event) {
  event.preventDefault();
}
116 Progressive Enhancement with Actions

To make the API call to submit the form asynchronously, we need to first find out where we are submitting
the form. We can get this information from reading the action property of the form instance:

form.action; // "https://domain/foo"

We can also determine the preferred HTTP request method from the form instance:

form.method // "post"

To get the form data submitted, we can use the FormData interface:

const data = new FormData(form);

With the URL, request method, and data, we can use the fetch API to submit the form:

fetch(form.action, {
  method: form.method,
  body: new FormData(form),
});

Finally, to reset the form, we can use the reset() method:

form.reset()

Putting everything together, we have the following:

<script>
  function enhance(form) {
    async function handleSubmit(event) {
      event.preventDefault();
      const response = await fetch(form.action, {
        method: form.method,
        body: new FormData(form),
      });
      form.reset();
    }
    form.addEventListener('submit', handleSubmit);
    return {
      destroy() {
        form.removeEventListener('submit', handleSubmit);
      }
    };
  }
</script>
<form action="/foo" method="post" use:enhance>...</form>
 Summary 117

Now, try to submit the form after the JavaScript is loaded. You’ll notice that a network request is made
to submit the form, while you remain on the same page without navigating away.
Try to disable the JavaScript or simulate a slow network speed. You’ll notice that you can still submit
the form, while the JavaScript is still being loaded. However, this time round, you submit through the
default browser behavior, which will navigate you away from the page.
Here we have it—a workable form by default, but progressively enhanced to submit form data without
leaving the page if the JavaScript is being loaded.
There’s a lot that can be improved on the enhance action. I’ll leave that as an exercise:

• Modify the enhance action to allow passing in a callback function that will be called after
the form submission is successfully made.
• What happens if the form submission fails? How should you handle such a case?
• Right now, the enhance action submits the form data through the request body; however,
when the form method is "get", form data should be passed through query parameters.
Modify the enhance action to handle the "get" form method.

Summary
In this chapter, we explained what progressive enhancement is and why it is important. Following up
on that, we learned how we can use Svelte actions to progressively enhance our elements.
We went through two different examples of progressive enhancements—enhancing a link to make it
show a preview popup, and enhancing form elements to submit a form asynchronously.
In the past three chapters, we’ve seen three different patterns and use cases of Svelte actions, creating
custom events, integrating UI libraries, and progressive enhancements. What you can do with Svelte
actions is not limited to the three different use cases that we’ve discussed so far, but hopefully, these
patterns have opened your imagination and made you see what is possible with Svelte actions.
With that, we are moving on to the next part of the book. We will be exploring Svelte context and stores
from Chapter 8 to Chapter 12, exploring their various use cases, such as in state management, creating
renderless components, and using them for animations. We will start by defining and comparing Svelte
contexts and Svelte stores in the next chapter.
 Part 3: Context and Stores

In this part, we will dive deep into Svelte’s two core features, Svelte context and Svelte stores. Over the
five chapters, we will explore different scenarios for using Svelte context and stores. We’ll start our
exploration by defining Svelte context and stores. This will be followed by a deep dive into implementing
custom stores and strategies for managing application state using Svelte stores. Following that, we will
learn how to use Svelte context to create a renderless component. Finally, we will learn how to create
animations using Svelte stores.
This part has the following chapters:

• Chapter 8, Context versus Stores

• Chapter 9, Implementing Custom Stores
• Chapter 10, State Management with Svelte Stores
• Chapter 11, Renderless Components
• Chapter 12, Stores and Animations
 8
Context versus Stores
A Svelte application can be composed of one or many Svelte components. A Svelte component can be
seen as a standalone unit, encapsulating its own reactive data and logic. In the previous chapter, we
learned how two Svelte components – in particular, components in a parent and child relationship
– communicate and pass data between each other. In this chapter, however, we are going to explore
communication and passing data between components beyond the parent and child relationship.
Svelte provides two primitives to pass data across Svelte components – Svelte context and Svelte
stores. Svelte context allows you to pass data from an ancestor to all children, while Svelte stores use
the observer pattern to allow you to access reactive data across multiple unrelated Svelte components.
In the coming five chapters, we are going to explore the different use cases of Svelte context and Svelte
stores. In this chapter, we will cover what Svelte context and Svelte stores are.
We will talk about when to use Svelte context and/or Svelte stores and the considerations for choosing
them. We will then proceed with an example of a combination of both Svelte context and Svelte stores
– a Svelte context store.
By the end of this chapter, you’ll be proficient in using Svelte stores and Svelte context in your Svelte
application. You will also understand when to use them effectively.
In this chapter, we will cover the following topics:

• Defining Svelte context and Svelte stores

• When to use Svelte context and Svelte stores
• Creating dynamic context using Svelte stores

Defining Svelte context

When you need to pass data from a parent component to a child component, the first thing you should
think of is using props:

<Component props={value} />

122 Context versus Stores

What if you need to pass data from a parent component to a grandchild component? You could pass
data as props from the parent component to the child component, and then from the child component
to the grandchild component:

<!-- Parent.svelte -->

<Child props={value} />

<!-- Child.svelte -->

<script>
  export let props;
</script>
<GrandChild props={props} />

What if you need to pass data from a parent component to a great-grandchild component?
You could follow a process similar to what we did in the preceding code, passing the data through
layers of components to reach the great-grandchild component.
This approach is called prop drilling. It is akin to drilling a hole through layers of components via
props. This is frowned upon in most cases due to the following reasons:

• It is hard to trace where the data comes from.

Whenever you want to trace where the data comes from in the child component, you may
endlessly trace up through layers of the parent component, jumping through different Svelte
component files.
This slows you down and makes it harder to reason with the data flow.
• It is hard to trace where the data leads to.
The data that’s passed down through props into the child component is not to be used by the
child component directly, but to be passed through it to its child component. You would have
to step through the layer of components to find out where the data is finally being used.
You may lose sight of where the data goes and have less confidence in making changes to the
data that’s passed down.
• It is hard to restructure the component hierarchy.
When you add a new component in between the layers, you need to make sure to still pass the
props through the new component from its parent to its children.
When you move the components around, you need to make sure the child component still gets
the props it needs by checking the chain of parent components.
 Defining Svelte context 123

Note
Keeping this in mind, when the component tree is small and simple, even with its drawback, passing
props around may still be the simplest way to pass data from a parent to its child components.

So, what is the alternative to prop drilling? Svelte context.

Svelte context is a method that provides data to all child components, no matter how many levels
down the component tree they are.
A component tree is like a family tree for components. You have a parent component at the top,
and one level down its child components, and one more level down is the child components of the
child components:

Figure 8.1: Component tree diagram

In the preceding component tree diagram, the top-left node represents where setContext is called,
and all shaded nodes under the node can access the context value using getContext. To set a
context value in a component, you can use setContext():

<script>
  import { setContext } from 'svelte';
  setContext("key", value);
</script>

All its child components and child components’ child components will be able to read the context
value through getContext():

<script>
  import { getContext } from 'svelte';
  const value = getContext("key");
</script>
124 Context versus Stores

If you paid attention to the preceding snippets, you may have noticed that we have been using strings
as the context key when setting and reading context. However, you might wonder whether it is possible
to use other data types as the context key. Let’s look into this.

Using an object as a context key

What is the underlying mechanism of Svelte context?
Svelte context is implemented using a JavaScript Map, which means you can set values and read values
out of Svelte context using a key, the same way you would with a JavaScript map.
This also means that you can set more than one key to Svelte context:

<script>
  setContext("item", item);
  setContext("order", order);
</script>

You just need to make sure you read them out using the same key you set the context with.
As I mentioned earlier, Svelte context is implemented using a JavaScript Map, and the keys of a Map
can be of any type, including functions, objects, or any primitives; you are not constrained to using
just a String key value:

<script>
  const object = {};
  setContext(object, value);
</script>

And just as how JavaScript Map works, if you set a context using an object, then you would need to
use the same object instance to read the value out of the Svelte context.

Changing the context value

One thing to take note of when using setContext and getContext is that these functions need to
be called during component initialization. ReadChapter 1 to recap on what component initialization is.
If we can call setContext after component initialization, then this leads us to the next question –
how would we change the context value?
Look at the following snippet:

<script>
  let itemId = 123;
  setContext("itemid", itemId);
 Defining Svelte context 125

  itemId = 456;
</script>

When calling setContext in line 3, we are passing the value of the itemId variable to the
setContext function. Reassigning the itemId variable in line 5 would not make the context
value change.
This is how JavaScript works. If you call a function with a primitive-type variable, then the value of
the variable is being passed in, and reassigning the variable outside of the function would not change
the value of the variable that’s read from inside the function.
How about passing an object as a context value? Let’s see how that works:

<script>
  let item = { id: 123 };
  setContext("item", item);

  item.id = 456;
</script>

In JavaScript, objects are passed by reference. This means that the Svelte context and the item variable
outside of the setContext function are referring to the same object. Mutating the object modifies
the same referenced object and thus the changes can be seen when reading the Svelte context:

<script>
  const item = getContext("item");
</script>
{item.id}

However, you may have noticed that after you render {item.id} onto the DOM, the value shown
in the DOM does not change when you mutate it in the parent component.
This does not mean item.id has not changed. If you try to print out item.id on an interval, you
will notice that item.id has changed, but the value in the DOM remains the same:

const item = getContext("item");

setInterval(() => {
  console.log(item.id);
}, 1000);

Why would this happen?

Svelte tracks variable mutations and reassignments within a Svelte component and instruments
operations to update the DOM to reflect the changes. However, what this means is that changes that
happen outside of the component are not tracked, and therefore the DOM does not reflect such changes.
126 Context versus Stores

So, what should we do to make Svelte aware of changes to a variable outside of the component?
This is where the Svelte store comes in.

Defining the Svelte store

To understand why Svelte reactivity is limited within a Svelte component, we must first understand
how Svelte’s reactivity works.
Unlike some other frameworks, Svelte reactivity works during build time. As Svelte compiles a Svelte
component into JavaScript, Svelte looks at each variable and tracks the variable to see when the
variable changes.
Instead of tracking all the variables throughout the application, Svelte limits itself to only analyzing and
compiling one file at a time. This allows Svelte to compile multiple Svelte component files in parallel but
also means that a Svelte component would not be aware of variable changes that happen in other files.
A common situation where a variable change is not tracked is when the variable is defined in a separate
file and imported into the current component.
In the following code snippet, the quantity variable is imported from a separate file. Svelte will not
track any changes to the quantity variable that may have occurred in that file:

<script>
  import { quantity } from './item';
</script>
<p>Quantity: {quantity}</p>

If you attempt to modify the variable outside of the Svelte component, then Svelte is not able to track
that. So, Svelte has no idea when you modify the variable and therefore will not be able to update the
DOM when that happens.
To make Svelte aware of the changes outside of the component and update the DOM accordingly,
we will need to design a mechanism at runtime that will notify Svelte whenever a variable changes.
For this, we can draw inspiration from the observer pattern. The observer pattern is a design pattern
that lets you define a subscription mechanism to notify multiple objects when an event happens.

Using the observer pattern

Here, instead of just importing the quantity variable, I am also importing a subscribe function,
which we will define later:

import { quantity, subscribe } from './item';

The idea of the subscribe function is such that we can subscribe to know when quantity changes.
 Defining the Svelte store 127

Here, we assume that subscribe takes in a callback function, which will be called whenever quantity
changes. The callback function takes in a parameter that gives us the latest value of quantity:

import { quantity, subscribe } from './item';

subscribe((newQuantity) => { ... });

So, now, although Svelte still cannot track changes to the quantity variable outside of the component,
we can use the subscribe function to tell Svelte when that happens.
As an example of how to tell Svelte this, we can define another variable called _quantity that
initializes to be the same value as quantity.
Whenever quantity changes, the callback function that’s passed into the subscribe function should
be called with the new quantity value. We will use this as an opportunity to update _quantity
to the new quantity value:

<script>
  import { quantity, subscribe } from './item';
  let _quantity = quantity;
  subscribe((newQuantity) => { _quantity = newQuantity; });
</script>
<p>Quantity: {_quantity}</p>

Since the _quantity variable is defined within the component, and we update the value of the
variable within the component (in the _quantity = newQuantity statement), Svelte can track
the update of _quantity. And since the _quantity variable tracks the changes of the quantity
variable itself, you can see that the DOM updates whenever quantity changes.
However, all of this depends on the subscribe function, which would call the callback function
whenever the value of quantity changes.
So, let’s see how we can define the subscribe function.

Defining the subscribe function

There are multiple ways of defining the subscribe function.
Here, we are going to use an array and name it subscribers so that we can keep track of all the
functions being called with subscribe. Then, when we attempt to update the value of quantity,
we will iterate through the subscribers array to get each of the subscriber functions and call
them one by one:

let subscribers = [];

function subscribe(fn) {
  subscribers.push(fn);
}
128 Context versus Stores

function notifySubscribers(newQuantity) {
  subscribers.forEach(fn => {
    fn(newQuantity);
  });
}

As an example, here, we want to update quantity to 20. To make sure that the subscribers are
notified of the changes, we call notifySubscribers with the updated value at the same time so
that each subscribers is notified with the latest value for quantity:

quantity = 20;
notifySubscribers(quantity);

Take your time when connecting the implementation of the subscribe and notifySubscribers
functions in the preceding code back to the Svelte component code in the previous section. You
will see that whenever we call notifySubscribers, the callback function that’s passed into the
subscribe function will be called. _quantity will be updated, and the value in the DOM will
be updated.
So, it doesn’t matter if you are modifying quantity outside the Svelte component, so long as you
call the notifySubscribers function. With the new value of quantity, Svelte will update the
DOM elements to reflect the latest value of quantity.
With the observer pattern, we are now able to define and update variables across Svelte components.
You will see this pattern a lot in Svelte. Svelte encapsulates the idea of subscribe and
notifySubscribers into a concept called a Svelte store. So, let’s explore what it means to be a
Svelte store and explore what first-class support Svelte is providing for Svelte stores.

Defining a Svelte store

A Svelte store is any object that follows the Svelte store contract.
This means that any object that follows the Svelte store contract can be referred to as a Svelte store. As
a Svelte store, there are a few syntactic sugars and built-in functions that come with it.
Before we get ahead of ourselves, let’s take a look at what a Svelte store contract is.
The Svelte store contract requires an object to have a subscribe method and an optional set method:

const store = {
  subscribe() {},
  set() {},
};
 Defining the Svelte store 129

I have not told you about the specific requirements for the subscribe and set methods, but I
hope you can see the similarity in the subscribe and set methods in the Svelte store contract to
the subscribe and notifySubscribers functions illustrated in the previous section.
But something is missing here. Where do we place the store value, or the corresponding quantity
variable from the previous section?
Well, the store value is not part of the Svelte store contract, and we will soon explain why.
Let’s continue with our requirements for the subscribe and set methods:

• The subscribe method has to return a function to unsubscribe from the store.
This allows the subscriber to stop receiving updates of the latest store value.
For example, if we use an array to keep track of the subscriber functions being called with the
subscribe function, then we can use the returned function from subscribe to remove
the subscriber function from the array since the subscriber function is no longer needed
to receive any new updates from the store:
const subscribers = [];
const store = {
  subscribe(fn) {
    // add the fn to the list of subscribers
    subscribers.push(fn);
    // return a function to remove the fn from the list of
subscribers
    return () => {
      subscribers.splice(subscribers.indexOf(fn), 1);
    };
  }
};

• When the subscribe method is being called with a function, the function must be called
immediately and synchronously with the store value.
If the function is not called immediately, the store value is assumed to be undefined:
let storeValue = 10;
const store = {
  subscribe(fn) {
    // immediately call the function with the store value
    fn(storeValue);
    // ...
  },
};
130 Context versus Stores

This requirement means that to read the store value out from the Svelte store, you will need to
use the subscribe method:
let storeValue;
store.subscribe((value) => {
  storeValue = value;
});
console.log(storeValue);

In the preceding snippet, you can see that if the callback function that’s passed into the
subscribe method is not being called immediately and synchronously, then immediately
in the next statement where we console out the value of storeValue, you will see that the
value of storeValue remains undefined.
• The set method of a Svelte store takes in a new store value and returns nothing:
store.set(newValue);

The set method is supposed to update the value of the store. Naturally, we would implement
the set method such that it will notify all the store subscribers of the latest store value:
const store = {
  // ...
  set(newValue) {
    // notify subscribers with new store value
    for(const subscriber of subscribers) {
      subscriber(newValue);
    }
  },
};

With that, we’ve gone through the requirements of a Svelte store contract. Along the way, we’ve also
seen snippets of code on implementing each requirement of a Svelte store. By putting them together,
we will have a Svelte store.
Creating a Svelte store is such a common use case that Svelte has provided a few built-in functions
to help us with creating one.

Creating Svelte stores with built-in functions

Svelte provides a sub-package that exports a few built-in functions for Svelte stores. You can import
them from the 'svelte/store' package.
Here is the list of built-in Svelte store functions:

• readable() helps create a readable Svelte store. Since the set method in a Svelte contract
is optional, a readable store is a store that does not implement the set method.
 Defining the Svelte store 131

To update the store value, the readable() function takes in a callback function that will
be called when the store is being subscribed, and the callback function is called with a set
function that can be used to update the store value:
const store = readable(initialValue, (set) => {
  // update store value
  set(newValue);
});

The set function in the callback function can be called numerous times. In the following example,
we are calling the set function every second to update the store value to the latest timestamp:
const store = readable(Date.now(), (set) => {
  setInterval(() => {
    // update store value to the current timestamp
    set(Date.now());
  }, 1000);
});

• writable() helps create a writable Svelte store. This is similar to the readable store, except
it implements the set method:
const store = writable(initialValue);
store.set(newValue);

• derived() creates a new Svelte store, deriving from existing stores.

We will explore derived() in more detail in the next chapter when we talk about creating
custom stores.
With readable(), writable(), and derived(), you can easily create a new Svelte store
without having to implement the Svelte store contract yourself.
So, we have built-in methods to create a Svelte store, but do we have any built-in methods for using
Svelte stores? Let’s find out.

Auto-subscribing to a Svelte store

Since all Svelte stores follow the Svelte store contract, all Svelte stores have the subscribe method
and, optionally, the set method. We can use the store.subscribe() method to subscribe to
the latest store value, and store.set() to update the store value:

<script>
  import { onMount } from 'svelte';
  let storeValue;
  onMount(() => {
    // use `subscribe` to subscribe to latest store value
132 Context versus Stores

    const unsubscribe = store.subscribe(newStoreValue => {

      storeValue = newStoreValue;
    });
    return () => unsubscribe();
  });
  function update(newValue) {
    // use `set` to update store value
    store.set(newValue);
  }
</script>
<p>{storeValue}</p>

When using a Svelte store in a Svelte component, we only subscribe to the Svelte store when needed
(usually as we mount the Svelte component). In the preceding snippet, we subscribe to the store
exactly after the Svelte component is mounted by calling the store.subscribe method inside
an onMount callback.
It is important to unsubscribe from new store value changes when it is no longer necessary. This is
usually when we unmount and destroy the Svelte component. In the preceding snippet, we return a
function in an onMount callback, which will be called when the component is unmounted. In the
function, we call the unsubscribe function return from the store.subscribe method.
This is in accordance with the Svelte store contract, where the store.subscribe method has to
return a function to unsubscribe from the store.
In a Svelte component, we need to remember to call store.subscribe during onMount, and
remember to clean up by calling unsubscribe during onDestroy.
This can become verbose, so Svelte provides a way to auto-subscribe to a Svelte store in a Svelte component.
When you have a variable that references a store in a Svelte component, you can auto-subscribe to
the store and access the value of the store through the $ prefixed variable name of the store variable.
For example, let’s say you have a Svelte store variable named count, as shown here:

<script>
  import { writable } from 'svelte/store';
  const count = writable();
</script>
 Choosing between a Svelte context and a Svelte store 133

In this case, you can auto-subscribe to the count Svelte store and access the store value through $count:

<script>
  import { writable } from 'svelte/store';
  const count = writable();
  console.log($count);
</script>

This is equivalent to subscribing to the store and assigning the latest store value to the $count variable,
However, when doing it this way, you no longer need to explicitly call the count.subscribe
method to subscribe to the store and call the unsubscribe function to unsubscribe from it.
If you paid attention to the code, you may have noticed that we did not declare the $count variable
at all. However, it is magically available, automatically declared by Svelte as Svelte builds the Svelte
component code.
This also assumes that whenever you are using the variable that starts with $, the variable without the
$ prefix is assumed to be a Svelte store.
Also, because of how Svelte auto declares the $ prefixed variable, it disallows any declaration of any
variable with a variable name starting with the $ sign.
What happens if I assign a new value to the $ prefixed variable? Doing so is equivalent to calling the
set method of the store:

$count = 123;
// is equivalent to
count.set(123);

So, now that we’ve learned about the Svelte context and Svelte store, let’s discuss when we should use
a Svelte context and/or a Svelte store.

Choosing between a Svelte context and a Svelte store

The Svelte context and Svelte store are designed for very different use cases.
Here’s a recap: the Svelte context helps pass data from a parent component to all descendent components,
while a Svelte store helps make data reactive across multiple Svelte components.
Although both the Svelte context and Svelte store are meant to pass data across Svelte components,
they are designed for different use cases. So, choosing when to use a Svelte context and Svelte store
is never an either-or situation.
134 Context versus Stores

You can use either a Svelte context, a Svelte store, or both to pass the same data across Svelte components.
To decide which one to use, I’ve come up with a 2x2 decision matrix:

Figure 8.2: A decision matrix for choosing a Svelte store, a Svelte context, or both

In this 2x2 decision matrix, there are two dimensions: local-global and static-reactive.
Depending on the kind of data you are passing, the data should fall into one of the four quadrants.
Then, we can decide on the best way of passing the data across components.
So, let’s take a closer look at what each dimension means.
In the local-global dimension, we determine whether the data should have the same value globally across
the entire application or have separate local versions among components that are close to each other.
For example, language preference data falls into being global instead of local. There tends to be only
one piece of language preference data throughout the entire application so that the language preference
is consistent in the application.
On the other hand, chart settings in a dashboard of charts could be local data. Multiple Svelte
components, such as chart axis, chart data, and chart grid, within the same chart share the same data,
yet different charts could have different chart settings. There’s no one single piece of data throughout
the entire application. So, in this case, it is more toward local in the local-global dimension.
If the data is going to be global throughout the entire application, the data can be declared in a JavaScript
module and imported from anywhere within the application:

// language-preference.js
export const languagePreference = ...;
 Choosing between a Svelte context and a Svelte store 135

On the other hand, if the data is going to be local, the data can be declared as a Svelte context. This
allows the child component to get different values, depending on where the component lives in the
component tree hierarchy:

<!-- Chart.svelte -->

<script>
  import { setContext } from 'svelte';
  setContext('chart', ...);
</script>

<!-- ChartAxis.svelte -->

<script>
  import { getContext } from 'svelte';
  // chartSettings depending on which chart it falls under
  const chartSettings = getContext('chart');
</script>

In the static-reactive dimension, we determine whether the data should be static, meaning it wouldn’t
change through the life cycle of the application, or should be dynamic, where the value of the data
will change as the user interacts with the application.
An example of static data would be the theme of an application. The value of the data could be determined
based on dynamic conditions, but once the value is determined at the start of the application, the
value does not change throughout the application life cycle. The application theme is a good example
of such a scenario. Usually, the theme of an application is determined at the start as the application
loads, and the theme stays throughout the application.
On the other hand, an example of dynamic data would be chart data. Chart data is dynamic and can
be changed throughout the application’s life cycle.
If the data is going to be static throughout the application’s life cycle, then the data can be declared
using a normal JavaScript variable:

let theme = 'dark';

However, if the data is going to be dynamic, and needs to be reactive across multiple components, the
data should be declared as a Svelte store:

import { writable } from 'svelte/store';

let chartData = writable();
136 Context versus Stores

If we combine the two dimensions, we get the following:

• Static global: The data is declared as a normal JavaScript variable in a JavaScript module and
exported for Svelte components to import
• Dynamic global: The data is declared as a Svelte store in a JavaScript module and exported
for Svelte components to import
• Static local: The data is declared as a Svelte context with a normal JavaScript variable as the
Svelte context value
• Dynamic local: The data is declared as a Svelte context with the Svelte store as the Svelte
context value

With that, we’ve seen how we can use a Svelte context and a Svelte store individually, but when we pass
dynamic local data across Svelte components, we use both a Svelte context and Svelte store together.
So, how do you combine both a Svelte context and a Svelte store? Let’s find out.

Passing dynamic context using a Svelte store

To make Svelte context data dynamic and reactive across components, we need to pass a Svelte store
as the Svelte context data instead of a normal JavaScript variable.
This is very similar to importing a Svelte store from a JavaScript module, except we are not importing
the Svelte store; instead, we are sending the Svelte store through the Svelte context.
To begin, we are creating a Svelte store, and passing the Svelte store into the context:

<script>
  import { writable } from 'svelte/store';
  import { setContext } from 'svelte';

  // declare a Svelte store

  let data = writable(0);
  setContext('data', data);
</script>

Note that we are passing the store into the Svelte context directly, instead of passing the store value.
In the child component, we can read the value out from the context through getContext():

<script>
  import { getContext } from 'svelte';
  const data = getContext('data');
</script>
 Summary 137

Since data is a Svelte store, we can reference the Svelte store value using the $ prefixed variable:

<script>
  import { getContext } from 'svelte';
  const data = getContext('data');
</script>
<p>{$data}</p>

To test whether the reactivity works, we can set a new value to the data store in the parent component:

<script>
  let data = writable(0);
  setContext('data', data);
  function update() {
    $data = 123;
  }
</script>

This goes both ways. If you attempt to update the store value from the child component, the store value
in the parent component will be updated too. Since you are getting the same store value through the
context in all the children components, any component using the same store will be updated as well.

Summary
In this chapter, we learned what Svelte context and Svelte store are.
Although both Svelte context and Svelte store are meant for sharing data across multiple Svelte
components, they are designed and used for different reasons.
Svelte context is meant for sharing the same data across all descendant components in the component
tree, while Svelte store is meant for sharing reactivity across Svelte components.
Then, we explored the decision matrix on when to use a Svelte context, when to use a Svelte store,
and when to use both.
This chapter served as an introduction to the Svelte context and the Svelte store. By now, you should
have a good understanding of what they are and how they work, and feel confident in knowing when
to use them. As we move forward, we will explore practical use cases that involve Svelte context and
Svelte store, allowing you to apply these powerful concepts effectively in real-world scenarios.
In the next chapter, we will dive deeper into the topic of the Svelte store and look at how to create a
custom one.
 9
Implementing Custom Stores
In the last chapter, we learned that a Svelte store is any object that follows a Svelte store contract, and
encapsulating data within a Svelte store allows the data to be shared and used across multiple Svelte
components reactively. The Svelte component keeps the DOM up to date with the data, even though
the data is modified outside of the Svelte component.
We learned about two of Svelte’s built-in methods for creating a Svelte store—namely readable()
and writable(), which create a readable and writable store. The two methods follow the Svelte
contract and create a very basic Svelte store. However, besides using a Svelte store to encapsulate data,
we can also encapsulate logic with the data, making the Svelte store modular and highly reusable.
In this chapter, we will be creating custom Svelte stores—Svelte stores that encapsulate custom data
logic. We will go through three distinct examples; each example will serve as a guide, brimming with
tips and tricks for creating your own custom Svelte stores.
First up, we’ll examine how to turn user events into store values, specifically turning click counts into
a Svelte store.
Second, we’ll explore a custom store that goes beyond the basic set method to modify its value.
We’ll look at an undo/redo store, which incorporates additional methods for reverting or redoing
changes to its value.
Lastly, we’ll turn our attention to higher-order stores. While not a custom store in itself, a higher-
order store is a function that accepts a Svelte store as input and returns an enhanced version of it. This
chapter includes sections on the following topics:

• Creating a Svelte store from user events

• Creating an undo/redo store
• Creating a debounced higher-order Svelte store

So, without further ado, let’s dive into creating our first custom Svelte store.
140 Implementing Custom Stores

Technical requirements
All the code in this chapter can be found on GitHub: https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter09

Creating a Svelte store from user events

A Svelte store stores data, but where does the data come from?
It could be from user interaction or user input, which calls an event handler that updates the store
value by calling store.set().
What if we can encapsulate the user events and event handler logic into the store so that we do not
need to call store.set() manually?
For example, we are going to have a Svelte store to calculate how many times the user clicks on the
screen. Instead of manually adding an event listener on the document, if there’s a way to create a Svelte
store and update it every time there’s a new click, how would that look? In short, how about having a
custom Svelte store that can do all of that for us?
It would be great if we could reuse this Svelte store the next time we have a similar need, instead of
having to manually set it up again.
So, let’s try to implement this click counter custom Svelte store.
Let's first scaffold the Svelte store:

const subscribers = [];

let clickCount = 0;
const store = {
  subscribe: (fn) => {
    fn(clickCount);
    subscribers.push(fn);
    return () => {
      subscribers.splice(subscribers.indexOf(fn), 1);
    }
  },
}

In the preceding code snippet, we created a basic Svelte store based on the Svelte store contract with
just a subscribe() method. The value of the store is stored in a variable called clickCount, and
in the subscribe() method, we keep track of all the subscribers using the subscribers array.
Notice that we need to synchronously call fn in the subscribe() method with the value of the
store; this lets the subscriber know the current value of the store.
 Creating a Svelte store from user events 141

If you use this store in a Svelte component right now (as shown in the following code snippet), you
will see 0 on the screen. That is the current value of the store at this point:

<script>
  import { store } from './store.js';
</script>
{$store}

In the preceding code snippet, we import store into a Svelte component and display the store
value using $store. Now, let’s listen to the click events to update the count value of the store.
Instead of listening to the click events at the start of our program, we should start subscribing only
when there’s a subscriber.
Here’s the updated code:

const store = {
  subscribe: (fn) => {
    // ...
    document.addEventListener('click', () => {
      clickCount++;
      // notify subscribers
      subscribers.forEach(subscriber => subscriber(clickCount));
    });
  },
};

In the preceding code snippet, we add the click event listener inside the subscribe method,
and whenever the document is clicked, we increment the clickCount value by 1 and notify all
the subscribers of the latest value of clickCount.
If you try to use this Svelte store in multiple Svelte components right now, you will realize that every
time you click, the store value increments by more than one.

Ensuring the event listener is only added once

Why is the store value incremented more than once when it is only clicked once?
If you look at the current implementation closely, you will realize that we call document.
addEventListener on every subscribe method call. When you use the Svelte store in multiple
Svelte components, each Svelte component subscribes to the store changes individually. If there are
five components subscribed to the store, then five click event listeners will be added to document.
As a result, a single click on document will trigger five event listeners, causing the clickCount
value to increase by five each time. This means the store value will go up by more than one for every
click. To fix this behavior, we can still call addEventListener within the subscribe method,
but we need to only call addEventListener once, even though the subscribe method could
be called more than once.
142 Implementing Custom Stores

We can use a flag to indicate if we have already called addEventListener, and make sure not to
call addEventListener again when the flag is set to true, as shown here:

let called = false;

const store = {
  subscribe: (fn) => {
    // ...
    if (!called) {
      called = true;
      document.addEventListener('click', () => {... });
    }
  },
};

In the preceding code snippet, we add a variable named called and use it to prevent document
from adding click event listeners more than once.
This works, but there’s a better way to implement this.
Instead of having a new flag to indicate whether we have called addEventListener, we can use
any existing variables to determine if we should call addEventListener. We know that we should
call addEventListener once the subscribe method is being called, and it should not be called
subsequently when we are adding more subscribers; we can use the length of the subscribers
array to determine if we should call the addEventListener method.
If there are no subscribers currently, then we know this is the first time the subscribe method is
being called. In this case, we should call addEventListener. On the other hand, if there are existing
subscribers—that is, the length of subscribers is more than zero, we know that the subscribe
method has been called before, and therefore we should not call addEventListener again.
So, here’s the updated code to use the length of the subscribers array instead of the variable called
to determine if the document should add click event listeners:

const store = {
  subscribe: (fn) => {
    // ...
    if (subscribers.length === 0) {
      document.addEventListener('click', () => {... });
    }
  },
};

In the preceding code snippet , we replaced the !called condition with subscribers.length
=== 0.
 Creating a Svelte store from user events 143

Now that we have added a click event listener when a subscriber is subscribing to our Svelte store, we
need to clean it up when all subscribers have unsubscribed from the Svelte store.
To clean it up, we are going to call removeEventListener to remove the click event listeners
from document. The unsubscribe function can be called multiple times, but we should only
call removeEventListener when there are no more subscribers in the subscribers array:

const store = {
  subscribe: (fn) => {
    return () => {
      subscribers.splice(subscribers.indexOf(fn), 1);
      if (subscribers.length === 0) {
        document.removeEventListener('click', () => {...});
      }
    };
  },
};

In the preceding code snippet, the return function of the subscribe method is used for
unsubscribing the store. In the function, we added a check to see if the number of subscribers
has dropped to zero; if so, we will remove the click event listeners from document.
It is common that when you are creating a Svelte store, you need to keep track of the subscribers
list, ensure that you only set up the event listeners once, and clean it up only after there are no
more subscribers.
As you’ve seen in this and the previous section, we've been through a lot of steps to create a Svelte
store from user events, manage a subscribers array, and decide when to add or remove click
event listeners from document. In the next section, we’ll explore a simpler way to achieve the same
goal using Svelte’s built-in methods and using Svelte’s built-in readable() function.
Svelte provides built-in methods such as readable() to make our lives easier when creating a
Svelte store.
Since the store value is only updated from click events, and not from elsewhere, readable(),
out of the two most important methods—readable() and writable()—is good enough for
our use case.
We will use the readable() function to create our click event store. There are two arguments for the
readable() function. The first argument is for the initial value of the store. The second argument
of the readable() function takes in a function that will be called when there’s a first subscriber and
not be called for the subsequent subscribers. If the function returns another function, the returned
function will be called when the last subscriber unsubscribes from the store. This is perfect for us to
add and remove click event listeners on document.
144 Implementing Custom Stores

Let’s take a look at the updated code rewritten using readable:

let clickCount = 0;
const store = readable(clickCount, (set) => {
  const onClick = () => set(++clickCount);
  document.addEventListener('click', onClick);
  return () => {
    document.removeEventListener('click', onClick);
  };
});

In the preceding code snippet, we utilize readable() to handle both the creation and the cleanup
of the click event listener, making our code cleaner and more efficient.
Comparing it with our implementation in the last section, where we managed the subscribers
list ourselves, you can see how using readable() allows us to declutter away from maintaining
the subscribers array manually and focus on implementing the logic.
It is good to know the Svelte contract and how to implement a Svelte store from scratch. But in most
real-world scenarios, it is easier to create a Svelte store out of readable() or writable() and
leave the nitty-gritty detail to the built-in function.
Now that you have learned how to create a Svelte store out of click events, let's practice with an exercise.

Exercise
Let’s begin our exercise, where we will implement a Svelte store in which the value of the store comes
from the scroll position of the document:

<script>
  const scrollPosition = createStore();
  function createStore() {
    // Your code here
  }
</script>
Scroll position {$scrollPosition}

In the code snippet, a scrollPosition store is set up using a function called createStore().
Your task is to implement the createStore() function to actually create a scrollPosition store.
You can find the solution at the following link: https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter09/02-exercise-scroll-position
 Creating an undo/redo store 145

Now that we have seen how to create a Svelte store where the store value comes from an event, let’s
take a look at a different kind of Svelte store.

Creating an undo/redo store

Typically, we change a store’s value using the set method. However, the next custom Svelte store we’ll
explore provides additional custom methods to update its store value. The next custom Svelte store that
we are going to look at is an undo/redo store. It is similar to a writable store where you can subscribe
to and set a new store value. But an undo/redo store also comes with two more methods, undo and
redo, which revert the store value backward or forward, based on the history of the store value.
Here’s a snippet of how you would use an undo/redo store:

<script>
  let value = createUndoRedoStore();
  $value = 123;
  $value = 456;
  $value = 789;
  value.undo(); // $value now goes back to 456
  value.undo(); // $value now goes back to 123
  value.redo(); // $value now turns to 456
</script>
Value: {$value}

In the provided code snippet, the createUndoRedoStore() function generates an undo/redo

store. Initially, we set the store’s value to 123, followed by updates to 456 and then 789. When we
call the store’s undo method, the value reverts to 456 and then to 123. Subsequently, using the redo
method takes the store’s value back to 456. Now that we understand how undo/redo store functions,
how do we go about creating one in Svelte?
Firstly, the undo/redo store is going to have four methods: subscribe, set, undo, and redo. The
subscribe and set methods are based on the Svelte store contract and are the reason an undo/
redo store is considered a Svelte store. The undo and redo methods, on the other hand, are two
additional methods that we define.
A JavaScript object can contain different methods and properties, but as long as it has the subscribe
and set methods, with the method signature following the Svelte store contract, we consider the object
a Svelte store. You can use a $-prefixed variable to auto-subscribe to the Svelte store and reference
the Svelte store value.
146 Implementing Custom Stores

Now, to implement this undo/redo store, we know that without the undo/redo feature, the store behaves
just like a writable store. So, we are going to implement the undo/redo store based on a writable store,
as shown in the following code snippet:

function createUndoRedoStore() {
  const store = writable();
  return store;
}

In the preceding code snippet, we set the stage for our createUndoRedoStore() function. We
start by using Svelte’s writable() function to create a writable store, which will be the base for
our undo/redo store.
But as we are setting a new value into the undo/redo store, we need to keep track of the history of
store values so that we can undo or redo them.
To do that, we are going to intercept the set method of the writable store, like so:

function createUndoRedoStore() {
  const store = writable();
  function set(value) {
    store.set(value);
  }
  return {
    subscribe: store.subscribe,
    set: set,
  };
}

In the preceding code, I return a new object. While the subscribe method is the same as the
original writable store’s subscribe method, the set method is now a new function. We still call
the writable store’s set method in the set function, so the behavior has not changed much.
But now, when we call the set method of the undo/redo store, we are calling the set function first,
before relaying this to the set method of the underlying writable store. This allows us to add additional
logic into the set function that will be run whenever we are setting a new value to the undo/redo store.
Before we get ahead of ourselves, let’s not forget that we also need to add two more methods, undo
and redo, into the undo/redo store. Here’s how we do this:

function createUndoRedoStore() {
  const store = writable();
  function set(value) { store.set(value); }
  function undo() {}
  function redo() {}
  return {
 Creating an undo/redo store 147

    subscribe: store.subscribe,
    set: set,
    undo: undo,
    redo: redo,
  };
}

In the preceding code snippet, we add two additional methods, undo and redo, to the object returned
by createUndoRedoStore(). We will go ahead and implement these methods in the next steps.
And now, we have the basic structure of our undo/redo store.
You can treat the preceding code as a template for creating custom Svelte stores. We use a writable
store as a base and return a new object. The returned object is considered a Svelte store because, by
having the subscribe and set methods, it adheres to the Svelte store contract. If we want to
add logic to the subscribe or set method, we can build a new function based on the original
subscribe and set methods from the writable store. In addition to that, we can also add more
methods to the returned object.

Implementing the undo/redo logic

Now, to implement the undo/redo logic, we are creating two arrays, undoHistory and redoHistory,
to record the history of values that we can replay when we call undo() or redo().
Whenever the set function is called, we will add the value as a new entry to undoHistory so that
we can replay it later on when undo() is called. When undo() is called, we will push the latest entry
in undoHistory into redoHistory so that we can redo what we’ve just undone.
Let’s go ahead and implement the logic we’ve just described:

function createUndoRedoStore() {
  const store = writable();
  const undoHistory = [];
  const redoHistory = [];
  function set(value) {
    undoHistory.push(value);
    redoHistory.length = 0; // resets redoHistory
    store.set(value);
  }
  function undo() {
    if (undoHistory.length <= 1) return;
    redoHistory.push(undoHistory.pop());
    store.set(undoHistory[undoHistory.length – 1]);
  }
  function redo() {
    if (redoHistory.length === 0) return;
148 Implementing Custom Stores

    const value = redoHistory.pop();

    undoHistory.push(value);
    store.set(value);
  }
  // ...
}

In the preceding code snippet, we’ve implemented the undo and redo function using two arrays:
undoHistory and redoHistory. We’ve also added checks to see if there are any values in these
arrays before performing an undo or redo action. This ensures that we don’t attempt to undo or redo
when there’s no history to revert to or advance from. So, you have now learned how to create a custom
Svelte store that extends from a writable store and adds new behavior to the original set method, as
well as adding new methods to the store.
It’s time to have an exercise.

Exercise
Now that we’ve learned how to create a custom store that provides custom methods to manipulate the
underlying store values, let’s have an exercise to build another such custom store, a tweened store. A
tweened store is a Svelte store that can contain only numeric values. When you set the value of a tweened
Svelte store, the tweened Svelte store takes a fixed duration to update its store value to the set value.
For example, let’s suppose the tweened Svelte store is set to 0 initially:

const store = createTweenedStore(0); // $store = 0

Then, you set the value of the store to 10:

$store = 10;

The store value is not set to 10 directly but increases from 0 to 10 over a configurable fixed duration—
say, 1 second.
Now you’ve learned the behavior of a tweened store, let’s implement a tweened Svelte store. You can
find the code for the tweened store on GitHub at https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter09/04-exercise-tweened-store.
Now that we’ve learned how to create a custom store, let’s turn our attention to another concept, a
higher-order store—a function that takes in a Svelte store as an input and returns a more specialized,
custom store based on that input.
 Creating a debounced higher-order Svelte store 149

Creating a debounced higher-order Svelte store

The two preceding sections that we’ve seen so far each created a new Svelte store. In this section, we
are going to look at how we can create a higher-order store.
The concept of a higher-order store is inspired by a higher-order function, where functions are treated
just like any other data. This means that you can pass functions as arguments to other functions or
return them as values.
In a similar concept, we are going to create a function that treats stores just like any data, taking a
Svelte store as an argument and then returning a Svelte store.
The idea of a higher-order Svelte store is to create a function that enhances an existing Svelte store.
A higher-order Svelte store is a function that takes in a Svelte store and returns a new Svelte store, an
enhanced version of the input Svelte store.
The example that we are going to use to illustrate this idea will create a debounce higher-order
Svelte store.
The debounce function that I am going to create will take in a Svelte store and return a new Svelte
store that has its store value debounced based on the input Svelte store value:

<script>
  function debounce(store) { ... }
  const store = writable();
  const debouncedStore = debounce(store);
</script>
Store value: {$store}
Debounced store value: {$debouncedStore}

In the preceding code snippet, in the fourth line, I demonstrated how to use the debounce function
that we’re going to implement in this section. This debounce function accepts a Svelte store as an
argument and returns an enhanced version of it, which we’ll call debouncedStore. To showcase
the debounce functionality, the values of both the original store and debouncedStore arguments
are displayed side by side.
Before we start implementing the debounce function, let’s talk about what debounce is.

Debouncing store value changes

In engineering, debounce is the process of removing unwanted input noises from user input. In web
development, debounce is used when there are too many user events, and we want to trigger an event
handler or process the event only after the user event settles.
150 Implementing Custom Stores

Here’s an example of debounce. When implementing an autocomplete search, we wouldn’t want to

trigger the search for an autocomplete result on every character the user types; rather, we’d only start
searching after the user stops typing. This saves resources as the autocomplete result may no longer
be usable as soon as the user types the next character.
Applying the concept of debounce to a Svelte store, we are going to create a new debounced Svelte
store based on an input Svelte store. When the input Svelte store value updates, the debounced Svelte
store only updates after the Svelte store updates settle.
If we are going to create a debounced Svelte store from scratch, we can build it up based on a writable
store. I showed you how to create a custom Svelte store from a writable store in the last section, so
I hope you know how to do that. Do try it out yourself; when you are done, compare it with my
implementation at https://github.com/PacktPublishing/Real-World-Svelte/
tree/main/Chapter09/05-debounce-store.
But in this section, we are going to create a higher-order store. We already have a Svelte store. It could
be a writable store or an undo/redo store. Our debounce function is going to take in the Svelte store
and return a debounced version of the Svelte store.
Let’s start with the basic structure of a debounce higher-order store.
The debounce function is going to return a new Svelte store. It is still much easier to build a new
Svelte store based on a writable store than to implement it from scratch, create a subscribe function,
and maintain an array of subscribers.
Here’s the basic outline of the debounce function:

function debounce(store) {
  const debounced = writable();
  return {
    subscribe: debounced.subscribe,
    set: store.set,
  };
}

In the preceding code snippet, the returned Svelte store is going to be based on the writable store, so
the subscribe method is going to be the writable store’s subscribe method.
The set method is going to be the original store’s set method, instead of a new set function. We
are not going to create a separate set function where it sets the original store and tries to set out a
debounced store with the debounce logic.
We are not going to do like it’s done in this snippet:

function debounce(store) {
  const debounced = writable();
  function set(value) {
 Creating a debounced higher-order Svelte store 151

    store.set(value);
    // some debounce logic and call debounce.set(value);
  }
  return {
    subscribe: debounced.subscribe,
    set: set,
  };
}

In the preceding code snippet, we intercepted the set method and relayed it to both the original store
and the debounced store. We are not going to do that because we want to preserve the original Svelte
store logic. When a value is passed to the set method, it may undergo transformations, especially if
the store is a custom Svelte store. The value argument in the set method may not be the same as
the final store value. So instead, we let the original store handle these potential transformations and
subscribe to the original Svelte store to get the final store value. We use the final store value and update
the debounced store as the original store value changes, as illustrated in the following code snippet:

function debounce(store) {
  const debounced = writable();
  store.subscribe(value => {
    // some debounce logic and call debounce.set(value);
  });
  // ...
}

In the preceding code snippet, I show how to subscribe to the original Svelte store rather than
intercepting its set method. This approach allows us to keep the original store’s logic intact while
still benefiting from its features. It also takes care of situations when the original Svelte store value
could change by other methods. An example of this is the undo() and redo() methods of the
undo/redo store. If we only intercepted the value from the set function, as in the preceding code
snippet, then the debounced Svelte store would not change when the original undo/redo store was
being undone or redone.
To implement the debounce logic, we are going to use a timeout to update the debounced Svelte store.
If there are new changes to the original Svelte store within the timeout, then we are going to cancel
the previous timeout and set a new timeout. If not, then we assume the changes have settled, and we
update the debounced store.
Here’s the updated code snippet with the debounce logic:

function debounce(store) {
  const debounced = writable();
  let timeoutId = null;
  store.subscribe(value => {
    if (timeoutId !== null) clearTimeout(timeoutId);
152 Implementing Custom Stores

    timeoutId = setTimeout(() => {

      timeoutId = null;
      debounced.set(value);
    }, 200);
  });
  // ...
}

In the preceding code snippet, we use the setTimeout function within the subscribe callback
to set a timeout for 200 ms. During this time, if the original store’s value changes again, the existing
timeout will be cleared, and a new one will be set. But if there’s a new value within the 200 ms time
period, we will update the debounced store with the latest value from the original store.
Try out the debounced store and see that the store value is now debounced from the changes.
One thing you might notice is that we subscribed to the original Svelte store but haven’t unsubscribed
from it. Let’s address that next.

Subscribing and unsubscribing original store on demand

Before we wrap up, there’s one small issue with the debounced store that we’ve implemented so
far. We subscribed to the original Svelte store at the creation of the debounced Svelte store and
we never unsubscribed from it, even though the debounced Svelte store may no longer be in use.
We should only start subscribing to the original Svelte store when there’s already a subscriber to the
debounced Svelte store and unsubscribe when there is no longer any subscriber.
We could intercept the subscribe method and attempt to unsubscribe from the original Svelte
store when the debounced Svelte store is being unsubscribed, like so:

function debounce(store) {
  const debounced = writable();
  function subscribe(fn) {
    const debouncedUnsubscribe = debounced.subscribe(fn);
    const unsubscribe = store.subscribe(...);
    return () => {
      debouncedUnsubscribe();
      unsubscribe();
    };
  }
  // ...
  return {
    subscribe: subscribe,
    set: store.set,
  };
}
 Creating a debounced higher-order Svelte store 153

In the updated code snippet, the subscribe method now subscribes to both the debounced
store and the original Svelte store. In the return function, which is used to unsubscribe from the
debounced store, we also make sure to unsubscribe from the original Svelte store.
However, this is not quite right. Doing it this way means that we subscribe to the original store every
time we subscribe to the debounced store and unsubscribe from the original store every time we
unsubscribe from the debounced store.
We should aim to subscribe to the original store just once, regardless of how many subscribers there
are. Similarly, we should unsubscribe from it only when there are no subscribers left. Does this
sound familiar?
We ran into a similar conundrum in the previous section when we were trying to implement a click-
counter Svelte store. We tried to maintain a list of subscribers before we found out that Svelte provides
a built-in writable() function that neatly takes care of that for us.
So, is there a built-in function that allows us to only subscribe to another store once a store is being
subscribed, and only unsubscribe from that store once there are no more subscribers in a store?
Yes—this is where derived() comes into play.

Deriving a new Svelte store with the derived() function

derived() is a built-in function from Svelte that takes in one or multiple Svelte stores and returns
a new Svelte store whose store value is based on the input Svelte stores.
For example, you can use derived() to define a Svelte store whose store value is always double that
of another Svelte store, as seen in the following code snippet:

import { writable, derived } from 'svelte/store';

const store = writable(1);
const double = derived(store, value => value * 2);

Or, this could be a Svelte store whose store value is the sum of two Svelte stores, as shown here:

const store1 = writable(1);

const store2 = writable(2);
const sum = derived(
  [store1, store2],
  ([$store1, $store2]) => $store1 + $store2
);
154 Implementing Custom Stores

Note
You may notice that we are using a $-prefixed variable in the callback function of the derived()
function; for example, $store1 and $store2. However, it does not work the same way as
the $-prefixed variable at the top level of the script, where it auto-subscribes to the store and
is referenced as the store value.
In this case, it is merely a convention that some people use to denote that the variable is used
to reference the value of a store. It is the same as any other variable name and nothing more.

When using derived(), Svelte only subscribes to the input Svelte stores—for example, store1
and store2—only when the returned Svelte store—for example, sum—is being subscribed to, and
will unsubscribe from the input Svelte stores when there are no more subscribers to the returned
Svelte store.
Before we rewrite our debounce higher-order store using derived(), let’s dive a little bit more
into the derived() function first.
The derived() function provides two approaches to determine the store value of the returned
Svelte store: synchronous or asynchronous.
The synchronous approach means that as soon as any of the input Svelte store’s values change, the store
value of the returned Svelte store is determined synchronously. By the same token, the asynchronous
approach means it is determined asynchronously, which means the store value can be set later after
the input Svelte store’s value has changed.
The two examples that I showed at the beginning of this section use the synchronous approach, where
the store value of the returned Svelte store is synchronously calculated and set right after the input
Svelte store has changed.
The asynchronous approach can be useful when the store value of the returned Svelte store is determined
through some asynchronous operations.
To determine whether the derived store is using the synchronous or asynchronous approach, the
derived() method looks at the function signature of the callback function. If it takes in one
parameter, then it is considered a synchronous approach. If it takes in two, then it is an asynchronous
approach. To get a better idea, check out the following code snippet:

derived(store, ($store) => ...) // synchronous

derived(store, ($store, set) => ...) // asynchronous

In both cases, the callback function will be called every time any of the input Svelte store values change.
If the derived store uses the synchronous approach, the returned value of the callback function is used
as the new store value of the returned Svelte store.
 Creating a debounced higher-order Svelte store 155

If the derived store uses the asynchronous approach, then the second parameter of the callback function
is a set function. The set function can be used to set the value of the returned store at any time.
The returned value of the callback function, however, will be treated as a cleanup function that will
be called right before the callback function is called again.
Now that we are more familiar with the derived() function, let's rewrite our debounce higher-
order store using derived().

Using the derived method

Here is the updated code for our debounced store rewritten using the derived() method:

function debounce(store) {
  const debounced = derived(store, (value, set) => {
    let timeoutId = setTimeout(() => {
      timeoutId = null;
      set(value);
    }, 200);
    return () => {
      if (timeoutId !== null) clearTimeout(timeoutId);
    };
  });
  return {
    subscribe: debounced.subscribe,
    set: store.set,
  };
}

Instead of creating a new debounced store using writable(), we derived it from the original
store using the derived() method. We shifted the setTimeout function into the derived()
method callback. When there’s a new value from the original Svelte store, we will clear the timeout
and set a new timeout.
The derived() function will subscribe to the original Svelte store once only when the debounced
store is being subscribed to and unsubscribe from the original Svelte store when there are no more
subscribers for the debounced store.
Since we are only going to update the debounced store asynchronously after a timeout, we pass
a callback function with two parameters into the derived() function. We call the set function
to set the value of the debounced store after the timeout. If the value of the original Svelte store
changes before the timeout, then the returned function of the callback function will be called first
before the callback function is called again with the updated store value of the original Svelte store.
In the returned function, we clear the timeout since we no longer need it anymore.
156 Implementing Custom Stores

One last thing before we conclude—sometimes, the original Svelte store may contain additional methods,
such as the undo() and redo() methods in the case of an undo/redo store. These methods should
also be defined in the debounced store returned by our higher-order function. This ensures that the
enhanced store maintains all the same methods and behaviors as the input store while adding the
debouncing feature. You can see an illustration of this in the following code snippet:

const value = createUndoRedoStore();

const debouncedValue = debounce(value);
$debouncedValue = 123;
$debouncedValue = 456;
debouncedValue.undo(); // $debouncedValue reverts to 123

To return all the methods from the original Svelte store, we use the spread operator (...) to spread
all the methods from the original Svelte store:

function debounce(store) {
  const debounced = derived(...);
  return {
    ...store,
    subscribe: debounced.subscribe,
  };
}

In this case, we do not even need to have set: store.set since the set method will be spread
into the returned Svelte store too!
That’s it! In this section, you’ve learned another trick—creating a higher-order store, a function that
takes in an existing Svelte store and returns a new Svelte store enhanced with new behavior. Instead
of building a custom Svelte store with all the logic in it, we can now create smaller well-encapsulated
higher-order stores and put them into a custom Svelte store that you need.

Exercise
Now, it’s time for an exercise.
The undo/redo store in the previous section is created as a custom Svelte store. Can you create an
undo/redo higher-order store that turns any Svelte store into an undo/redo store with two additional
methods, undo() and redo()?
Here’s an example of how the undo/redo higher-order store would be used:

import { writable } from 'svelte/store';

const originalStore = writable(100);
const undoRedoStore = undoRedo(originalStore);
$undoRedoStore = 42;
 Summary 157

undoRedoStore.undo(); // store value now goes back to 5

The preceding code snippet shows how we use the undo/redo higher-order store, undoRedo. The
undoRedo function takes in a store, named originalStore, and returns a new store based on
originalStore that has the undo/redo capability. For example, if we set a new value and then call
the undo method, the store value will revert to its original state, which is 5 in this case.
You can find the solution for this exercise on GitHub here:
https://github.com/PacktPublishing/Real-World-Svelte/tree/main/
Chapter09/07-exercise-undo-redo-higher-order-store
In this section, we’ve explored the process of creating a debounce higher-order Svelte store using
the derived() method. I look forward to you applying this knowledge in real-world scenarios.

Summary
In this chapter, we looked at three different examples and learned three different techniques with a
Svelte store.
We explored how to turn any user events into a data source for a Svelte store, learned how to create a
Svelte store from scratch, and also learned how to use the built-in readable() function to make
the process much simpler.
After that, we explored making a custom Svelte store with additional methods, building a new Svelte
store based on the built-in writable store.
Lastly, we learned to create a higher-order store, a function that takes in a Svelte store and returns an
enhanced version of the input store. In the example, we see how we can turn any Svelte store into a
debounced version of itself.
By understanding these techniques, you’re now equipped to manage state in Svelte more effectively
to craft more scalable and maintainable applications.
In the next chapter, we will look at state management in Svelte—namely, how to do state management
with Svelte stores.
 10
State Management with
Svelte Stores
Every user interface control has a state. A checkbox has a checked-unchecked state. A textbox’s state
is its current input value. A table’s state is the data displayed and the columns currently being sorted.
Sometimes when you have multiple user interface controls present across the screen at the same time,
you need to synchronize the state across them—this is where state management comes in.
In this chapter, we are going to discuss managing states in Svelte using Svelte stores. We will start with
why we should use Svelte stores, and then discuss tips that will help you when using Svelte stores for
state management.
Next, we will go into the topic of using state management libraries. We will talk about why and how
to use them. With that, we will be going through a few examples of integrating third-party state
management libraries into Svelte through Svelte stores.
This chapter includes sections on the following topics:

• Managing states with Svelte stores

• Using state management libraries with Svelte

Technical requirements
The code in this chapter can be found on GitHub: https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter10.

Managing states with Svelte stores

When building an interactive user interface, the first thing we consider is determining the necessary
states to represent the various components and interactions.
160 State Management with Svelte Stores

For example, in the following snippet, I have a login form that has a few components, including two
inputs, one checkbox, and one button:

<Input name="username" />

<Input name="password" />
<Checkbox label="Show Password" />
<Button>Submit</Button>

Each of the Svelte components has multiple states, as outlined here:

• The <Input /> component has an input value state and an error state set during validation
• The <Checkbox /> component has a checked/unchecked state, checked to reveal the
password in the input
• The password <Input /> component has an additional state to reveal/hide the password
• The <Button /> component has an enabled/disabled state, disabled when the form is incomplete

Each of these states can be represented by a variable. The variables themselves can be interrelated.
Let us examine how these variables can be related to each other. For example, when the checkbox
is checked, the password input needs to reveal the password. When the name and password inputs
are filled, the button needs to turn into an enabled state. When the button is pressed, validation is
performed, and the inputs’ error state is updated accordingly.
The art of managing multiple states across components to form a cohesive experience is called
state management.
It would be clearer and easier to group all these states together and manage them as one, rather than
separate individual states. In the following snippet, I group all of the aforementioned variables into
one variable object:

const state = {
  nameValue: '',
  nameError: null,
  passwordValue: '',
  passwordError: null,
  revealPassword: false,
  submitDisabled: true,
}

Here, you can see the related states in one place, making it easier to see and tell if something is wrong.
 Managing states with Svelte stores 161

To illustrate this, let’s say the nameValue and submitDisabled variables are kept as two separate
variables in two separate components; in the Input component, we have the nameValue variable:

<!-- Input.svelte -->

<script>
  export let nameValue = '';
</script>
<input bind:value={nameValue} />

And in the Button component, we have the submitDisabled variable:

<!-- Button.svelte -->

<script>
  export let submitDisabled = false;
</script>
<button disabled={submitDisabled}>Submit</button>

From the code, you can see that it is not straightforward to tell whether the nameValue and
submitDisabled variables are related. If one of the states is not updated as expected, you don’t
have one convenient place to inspect both states at once; you will have to inspect them in each
separate component.
For example, if the submitDisabled state in the <Button /> component did not change to
false to enable the button after typing in the name input, you would need to find and inspect the
nameValue variable in a separate component (the <Input /> component).
If, instead, the states are grouped into a state object, then you can inspect the state object and
check the nameValue and the submitDisabled properties and see if they are set correctly. Have
a look at the following example:

const state = {
  nameValue: '',
  submitDisabled: true,
};

Now that we’ve established that we should group multiple related states into one state object, the next
question is this: If we could group states using a simple JavaScript object, why would it be necessary
to group states using a Svelte store?
Well, the truth is, it is not always necessary. In some cases, using a simple JavaScript object to group
states can work just as effectively. However, a Svelte store offers additional benefits and functionality
that can enhance your state management experience.
162 State Management with Svelte Stores

As we explored in Chapter 8, a Svelte store is useful as it can be passed around components and have
its updates propagate across components. Then, in Chapter 9, we saw how useful a Svelte store is,
being able to encapsulate logic within the Svelte store, as well as being able to define data logic right
next to the data itself.
So, for the sake of the chapter, we are going to proceed with grouping multiple related states into a
state object using a Svelte store. Here’s a short snippet of code of how it would look:

<script>
  import { writable } from 'svelte/store';
  const state = writable({
    nameValue: '',
    nameError: null,
    passwordValue: '',
    passwordError: null,
    revealPassword: false,
    submitDisabled: true,
  });
</script>

Here, we use a writable() store for creating the state object because we will be modifying the
state object.
The state object can be defined either inside or outside a Svelte component. Since our state object is
a Svelte store, we can import the state object into any component, and any updates to the state object
will be propagated across all Svelte components that use the state object—so, it doesn’t matter in which
file we define the state object in.
If we define our state object in a Svelte component, then we can pass our state object to the component’s
child component through props. Here’s an example of doing so:

<script>
  const state = writable({ name: 'Svelte' });
</script>
<Input {state} />

On the other hand, if we define our state object in a JavaScript module, then we can import our state
object into any Svelte components that will be using the state object. The following is a snippet of
importing a Svelte store state object into a Svelte component:

<script>
  // import state object defined in './state.js'
  import { state } from './state.js';
</script>
 Managing states with Svelte stores 163

In Chapter 8, we learned how to read, subscribe to, and update a Svelte store. Whether the data stored
inside a Svelte store is a state object or anything else, there is no difference in terms of the operations
to read, subscribe, and update the Svelte store state object like any other Svelte store.
But as the state is used more frequently in a growing project, you will have more state values inside
a state store object. As the state store gets larger and more complex, your application may get
slower, and you may feel the need to optimize/improve it. Therefore, I am dedicating the following
subsections to sharing my tips and opinions on state management with a Svelte store.
Of course, you shouldn’t apply all these methods blindly. As said by Donald Knuth, the famous
computer scientist, “Premature optimization is the root of all evil”—you should measure and evaluate
whether any of the following tips apply. As I go through the tips, I will explain what they are and why
and when are they useful.

Tip 1 – simplify complex state updates with unidirectional data

flow
When your application becomes more complex, with multiple components and multiple places to
update your state, managing state updates becomes even more challenging.
How would you know when managing state updates becomes unmanageable? You may run into
situations such as this:

• Your application state changes, and you find it hard to track down why the state changed
• You change a certain state value, but unknowingly, this causes other seemingly unrelated state
values to update as well, and you have a hard time working out why

These are signs that the state updates have grown complex and become difficult to manage. In such
situations, having unidirectional data flow can help simplify state management.
With unidirectional data flow, state updates are managed in a single direction, and the flow of data
can be easily traced and debugged.
There are state management libraries such as Redux, MobX, and XState that help to enforce a single
data flow and enable you to reason with the state changes. You may be curious about how to use these
state management libraries with a Svelte store; I will be covering them later in this chapter and will
use one of the libraries as an example.

Tip 2 – prefer smaller self-contained state objects over one big

state object
In Svelte, when a state object changes, all components that use that state object are updated, even if the
specific state value that a component uses did not change. This means that if a state object becomes
larger and more complex, with many unrelated state values, updating that state object can trigger
unnecessary updates to many components in the application.
164 State Management with Svelte Stores

Here, I am going to use a product details page of an e-commerce web application as an example. In a
typical e-commerce web app, you have a page that shows the details of a product. You will see other
information, such as the shopping cart and product reviews, on the same page as well.
If I use one state object for all the information on the page, it may look like this:

<script>
  let state = {…}
</script>
<ShoppingCart cart={$state.cart} />
<ProductDetails product={$state.product} />
<ProductRatings ratings={$state.ratings} />

In the preceding code, we use the same state object, $state, across three components: ShoppingCart,
ProductDetails, and ProductRatings. When any part of the $state state object changes,
such as changing $state.cart, all three components will be triggered to update.
This is undesirable—multiple components updating unnecessarily could lead to slower performance
and a less responsive user interface.
So, it is recommended to split big state objects into smaller state objects. Using the same example, that
would mean splitting the $state state object into three smaller state objects, like so:

<script>
  let cartState = {…};
  let productState = {…};
  let ratingState = {…};
</script>
<ShoppingCart cart={$cartState} />
<ProductDetails product={$productState} />
<ProductRatings ratings={$ratingState} />

This way, changing the shopping cart’s state, $cartState, would not trigger an update on the
<ProductDetails /> and <ProductRatings /> components.
So, if you have big state objects and you find components update unnecessarily and the performance
of your application is impacted by it, then consider breaking these big state objects down into smaller
state objects.
But what if the state object is still big, yet the different values in the state objects are closely related
and you are unable to break it apart into smaller state objects? Well, there is still hope, which leads
us to our third tip.
 Managing states with Svelte stores 165

Tip 3 – derive a smaller state object from a bigger state object

If state values are related to each other and so you can’t break a big state object into smaller state
objects, we can create smaller state objects that derive from the big state object. Let me show you
some code to explain this clearly.
Let’s say that you have a state object, userInfo, that has two closely related state values, $userInfo.
personalDetails and $userInfo.socials, as shown here:

import { writable } from 'svelte/store';

const userInfo = writable({
  personalDetails: {...},
  socials: {...},
});

You may realize that one part of the userInfo state object doesn’t change as often as the other. But
whenever any part of userInfo changes, all the components that use either the $userInfo.
personalDetails or $userInfo.socials state values will be updated.
To ensure that components using only $userInfo.socials are updated exclusively when
$userInfo.socials changes, one way would be to break the state object into smaller, more
focused state objects, like so:

import { writable } from 'svelte/store';

const userPersonalDetails = writable({...});
const userSocials = writable({...});

As you can see, you now have two separate state objects, userPersonalDetails and userSocials.
But this would mean that places where you previously updated the userInfo state object would
have to change since userInfo is now split into two separate state objects.
Here is how you would change the code:

// previously
$userInfo = newUserInfo;
// now
$userPersonalDetails = newUserInfo.personalDetails;
$userSocials = newUserInfo.socials;

So, the question now is this: Is there an alternative to not having to change this, yet being able to update
the components that use $userInfo.socials only when $userInfo.socials changes?
166 State Management with Svelte Stores

I believe I’ve leaked the answer already. The alternative is to derive a new state object. In the following
code snippet, I am deriving a new userSocials state object from the userInfo state object:

import { derived } from 'svelte/store';

const userSocials = derived(userInfo, $userInfo => $userInfo.socials);

The component that uses $userSocials will only update whenever the userSocial state
changes. The userSocial state changes only when the $userInfo.socials changes. So, when
the component uses $userSocials instead of $userInfo.socials, it will not update when
any other part of the userInfo state object changes.
I believe that seeing is believing, and it will be much clearer to see and interact with an example
to get this idea forward. So, I’ve prepared some demo examples at https://github.com/
PacktPublishing/Real-World-Svelte/tree/main/Chapter10/01-user-social,
and you can try them out and see what I mean.
Let’s quickly recap the three tips:

• Whenever the state update logic is complex and convoluted, introduce some state management
libraries to enforce simpler and unidirectional data flows
• If the state object gets too big, and state changes update more components than needed, break
the state object into smaller state objects
• If you can’t split a Svelte store state object, consider deriving it into a smaller state object

So, we’ve gone through my three general tips for managing complex stores in a Svelte application; let’s
now elaborate more on my first tip on how to use state management libraries with Svelte.

Using state management libraries with Svelte

If you google State management library for frontend development, at the time of writing, you will get
list after list of libraries, such as Redux, XState, MobX, Valtio, Zustand, and many more.
These libraries have their own take on how states should be managed, each with different design
considerations and design constraints. For the longevity of the content of this book, we are not going
to compare and analyze each of them since these libraries will change and evolve over time and
potentially be replaced by newer alternatives.
It is worth noting that some of the state management libraries are written for a specific web framework.
For example, at the time of writing, the Jōtai library (https://jotai.org/) is written specifically
for React, which means you can only use Jōtai if you write your web application in React.
On the other hand, there are framework-agnostic state management libraries. An example is XState
(https://xstate.js.org/), which can be used by any web framework as the XState team has
created packages such as @xstate/svelte to work with Svelte, @xstate/react for the React
framework, @xstate/vue for Vue, and many more.
 Using state management libraries with Svelte 167

While the @xstate/svelte package is tailored for seamless integration of XState in Svelte, not all
state management libraries offer this level of compatibility. Nevertheless, there are several other state
management libraries that you can use in Svelte, and integrating them is straightforward. In fact, I will
provide some examples to showcase how simple it is to integrate these libraries in Svelte and utilize
Svelte’s first-class capabilities for working with stores.
One such state management library that works seamlessly with Svelte is Valtio (https://github.
com/pmndrs/valtio), a minimalist library that turns objects into self-aware proxy states. We
are going to explore how we can use Valtio in Svelte, by turning Valtio’s proxy state into a Svelte store
and using the Svelte store’s $-prefixed syntax to subscribe to the proxy state changes and access the
proxy state value.

Example – using Valtio as a Svelte store

Before we start talking about how to use Valtio in Svelte, let’s look at how to use Valtio on its own.

Using Valtio

Valtio turns the object you pass to it into a self-aware proxy state. In the following code snippet, we
create a Valtio state named state through the proxy() method:

import { proxy } from 'valtio/vanilla';

const state = proxy({ count: 0, text: 'hello' });

To update the proxy state, you can make changes to it the same way you would to a normal JavaScript
object. For example, we can increment the value of state.count by mutating the value directly,
as follows:

setInterval(() => {
  state.count++;
}, 1000);

To be notified of modifications in the proxy state, we can use Valtio’s subscribe method, as
illustrated here:

import { subscribe } from 'valtio';

const unsubscribe = subscribe(state, () => {
  console.log('state object changed');
});

In this case, every time we modify the state, the callback function passed into the subscribe method
will be called, and we will see a new log in the console, printing 'state object changed'.
168 State Management with Svelte Stores

Valtio also allows you to subscribe only to a portion of a state. For example, in the following snippet,
we subscribe only to the changes made to state.count:

const unsubscribe = subscribe(state.count, () => {

  console.log('state count changed');
});

Since in this case, we are subscribing only to the changes made to state.count, then modifying
state.text would not see a 'state count changed' log added to the console as the change
to state.text is not subscribed.
The Valtio proxy state is meant for tracking update changes. To read the latest value of the proxy state,
we should use the snapshot() method to create a snapshot object, as follows:

import { subscribe, snapshot } from 'valtio';

const unsubscribe = subscribe(state, () => {
  const obj = snapshot(state);
});

Using Valtio as a Svelte store

Now that we’ve learned about basic Valtio operations and concepts, let’s create a simple counter application
to see how we can use Valtio in Svelte. Firstly, we create a proxy state for our counter application:

// filename: data.js
import { proxy } from 'valtio';
export const state = proxy({ count1: 0, count2: 0 });

Here, I am creating two counters, count1 and count2, which will later allow us to experiment with
subscribing to a specific portion of the proxy state. This way, we can observe whether our application
updates only when one of the counters changes.
Also, we are creating the proxy state in a separate file, data.js, rather than declaring it inside a
Svelte component; this way, we can import the state separately in each Svelte component later on.
In addition, let’s create two functions to increment the counters:

export function increment1() {

  state.count1 ++;
}
export function increment2() {
  state.count2 ++;
}
 Using state management libraries with Svelte 169

Now, let’s import the proxy state into our Svelte component; I have created two buttons to increment
each counter separately:

<script>
  import { state, increment1, increment2 } from './data.js';
</script>
Count #1: {state.count1}
Count #2: {state.count2}
<button on:click={increment1}>Increment 1</button>
<button on:click={increment2}>Increment 2</button>

If you try clicking on the buttons, you’ll realize that they are not working as expected— the counters
are not incrementing.
However, if you add the following code to data.js and click on the button, the console will print
out the current value of the counters, indicating that the counters are incrementing successfully:

import { subscribe } from 'valtio';

subscribe(state, () => {
  console.log(state);
});

As you can see, the counters are updating as expected. The issue, therefore, does not lie in the inability
to increment the counters, but rather in the failure to display the changes on the screen. It is possible
that Svelte is not recognizing the changes to the counters, and therefore, it is not updating the elements
to show the latest value of the counters.
So, what can we do to make Svelte aware of the changes to Valtio’s proxy state?
One approach would be to use a Svelte store, as it provides a mechanism for Svelte components to
react to changes in data. We can turn Valtio’s proxy state into a Svelte store. Then, by subscribing to
the store, Svelte will be aware of the changes to the state and will update the elements accordingly.
This approach of converting states from other state management libraries into a Svelte store in order
to take advantage of Svelte’s built-in update mechanism is very common. It allows developers to use
their preferred state management solution while still taking advantage of Svelte’s reactive capabilities.
So, let’s see how we can turn Valtio’s proxy state into a Svelte store. To start off, I am creating a function
called valtioStateToSvelteStore:

function valtioStateToSvelteStore(proxyState) {
  return {};
}
170 State Management with Svelte Stores

Before creating the Svelte store from the Valtio proxy state, let’s have a quick recap on what the Svelte
store contract is. The Svelte store contract dictates that a Svelte store should have a subscribe
method and an optional set method. The subscribe method takes a callback function as its only
parameter, which will be called whenever the store’s value changes, and returns an unsubscribe
function to stop further notifications.
So, let’s define the subscribe method in the returned object of the valtioStateToSvelteStore
function, like so:

function valtioStateToSvelteStore(proxyState) {
  return {
    subscribe(fn) {
    },
  };
}

The initial value of the Svelte store can be defined by calling the callback function synchronously with
the value of the proxy state:

function valtioStateToSvelteStore(proxyState) {
  return {
    subscribe(fn) {
      fn(snapshot(proxyState));
    },
  };
}

Our next step is to subscribe to changes in the Valtio proxy state:

function valtioStateToSvelteStore(proxyState) {
  return {
    subscribe(fn) {
      fn(snapshot(proxyState));
      return subscribe(proxyState, () => {
        fn(snapshot(proxyState));
      });
    },
  };
}

Based on the Svelte store contract, within the subscribe method, we need to return a function to
unsubscribe the callback function from the Svelte store. In our code, the reason we return the return
value of the subscribe function from Valtio in our subscribe method is that the return value
of Valtio’s subscribe function is a function to unsubscribe from the proxy state.
 Using state management libraries with Svelte 171

A function to unsubscribe from changes is just what we need. Isn’t this convenient?
It’s no coincidence that most state management libraries provide methods for subscribing to state
changes and unsubscribing from them, just like what we need to define a Svelte store. This is because
both Svelte stores and state management libraries are designed based on the Observer pattern we
discussed in Chapter 8. In summary, to turn a state management library’s state into a Svelte store, we
need to understand how the library works and how its APIs translate into the Svelte store contract.
Now that we have a function to transform a Valtio’s proxy state into a Svelte store, let’s try to use it by
running the following code:

<script>
  import { state, increment1, increment2, valtioStateToSvelteStore }
from './data.js';
  const store = valtioStateToSvelteStore(state);
</script>
Count #1: {$store.count1}
Count #2: {$store.count2}

Clicking on any button in the component, you will see that the counter works perfectly fine now.
So, this is how we turn a Valtio proxy state into a Svelte store.
The next thing I would like to explore is creating a Svelte store that subscribes only to partial updates
of the Valtio proxy state. By selecting only a specific portion of the state to monitor, we can ensure
that the Svelte store updates only when a particular part of the state changes.
Before we do that, let’s add a few more lines to show you what I mean:

<script>
  // ...
  $: console.log('count1: ', $store.count1);
  $: console.log('count2: ', $store.count2);
</script>

As you click on either of the increment buttons, you will notice that both reactive statements are called,
indicating that $store is updated whenever either count1 or count2 is updated.
As discussed in the third tip earlier in the chapter, if state changes cause unnecessary code to run, we
can derive a smaller state from the original state to only subscribe to partial updates. So, let’s do that:

<script>
  import { state, increment1, increment2, valtioStateToSvelteStore }
from './data.js';
  const count1 = valtioStateToSvelteStore(state.count1);
  const count2 = valtioStateToSvelteStore(state.count2);
  $: console.log('count1: ', $count1);
172 State Management with Svelte Stores

  $: console.log('count2: ', $count2);

</script>

Here, instead of turning state into a Svelte store, we are turning state.count1 into a Svelte
store. This allows us to create separate Svelte stores that only subscribe to a portion of the proxy state.
This should work, but unfortunately, it doesn’t. The reason for this has nothing to do with our code
but with the data structure of our state. state.count1 is a primitive value, which Valtio is unable
to subscribe to.
To work around this, I’m going to change the data type of state.count1 from a primitive value
to an object:

export const state = proxy({

  count1: { value: 0 },
  count2: { value: 0 },
});
export function increment1() {
  state.count1.value ++;
}

In the preceding code snippet, we changed state.count1 to an object with a property called
value. We still keep the code of deriving the count1 Svelte store from the state.count1 proxy
state. So, now, the derived Svelte store of count1 would be an object, and to get the value of the
count, we will be referring to $count1.value instead of $count1:

$: console.log('count1: ', $count1.value);

Now, when you click on the button labeled Increment 1, you will only see logs printed for count1.
Conversely, when you click on the other button, labeled Increment 2, you will see new logs printed
only for count2. The reactive statement will only run when the respective counter is incremented.
That’s because our count1 and count2 stores only update when a specific part of the Valtio proxy
state changes.
So, let’s conclude what we’ve done so far.
We explored the integration of Valtio in Svelte by converting the Valtio proxy state into a Svelte store.
Step by step, we implemented the Svelte store contract by leveraging Valtio’s built-in methods for
subscribing and unsubscribing. Toward the end, we explored how to subscribe to partial updates of
the proxy state, minimizing unnecessary reactivity.
I believe you are eager to try it out yourself by now, so let’s use the next state management library as
an exercise.
 Summary 173

Exercise – transforming an XState state machine into a Svelte

store
XState (https://xstate.js.org/) is a JavaScript library for managing states using state
machines. In this exercise, you are going to transform an XState state machine into a Svelte store.
XState provides a package named @xstate/svelte that includes utilities specifically designed for
integrating XState in Svelte. Although this package can be a source of inspiration for your task, the
use of @xstate/svelte is, however, not permitted for this exercise. The goal here is to challenge
you to implement the function to transform an XState state machine into a Svelte store without the
aid of pre-existing integration utilities.
To get started with this exercise, follow these steps:

1. Understand XState: If you’re not already familiar with XState, spend some time reading its
documentation and experimenting with its features. Understand how state machines work and
how XState implements them.
2. Create a state machine: Using XState, create a state machine that you’d like to convert into a
Svelte store. This could be a simple machine with a few states and transitions, or something
more complex if you’re comfortable with it.
3. Transform the state machine into a Svelte store: Now comes the key part of the exercise.
You’ll need to write a function that converts your XState state machine into a Svelte store. This
involves subscribing to state changes in the machine and forwarding them to the Svelte store,
as well as mapping actions from the store back to transitions in the state machine.
4. Test your implementation: After you’ve completed your conversion, ensure you thoroughly
test it. Change states in your store and observe whether the same changes are reflected in your
state machine and vice versa.

If you’re stuck, take a step back and refer to the documentation of XState and this chapter, or the
source code of @xstate/svelte.

Summary
In this chapter, we learned about how we can manage our application states with Svelte.
At the beginning of the chapter, we discussed some of the tips for managing states in Svelte through a
Svelte store. These tips will be useful to you as your Svelte application grows larger and more complex.
One of the tips that we discussed is using state management libraries to manage data changes and data
flows. That’s why we spent the second half of the chapter exploring how we can use state management
libraries in Svelte, by transforming the state management libraries’ state into Svelte stores.
In the next chapter, we are going to explore how we can use a combination of both a Svelte context and
a Svelte store to create renderless components—logical components that do not render any content.
 11
Renderless Components
A renderless component is an advanced concept in Svelte that allows developers to create reusable
components without rendering any HTML elements within the component itself.
This technique is particularly useful when leveraging Svelte to render on a canvas or in a 3D context,
where the rendering of an HTML template by Svelte is not required. Instead, the canvas and Web
Graphics Library (WebGL) offer an imperative API to produce graphics on the canvas. With the
renderless component technique, it becomes possible to design components that enable users to
describe the canvas declaratively, allowing the component to translate it into imperative instructions.
Another use case for a renderless component is to create components that only manage states and
behaviors, leaving the parent component control over what should actually be rendered. This will come
in handy when developing a component library and you want to make it easy for users to customize
how your component looks and lets you have control over the state and behaviors.
In this chapter, we will utilize Svelte context to create our renderless component. If you are new to
Svelte context, please refer to Chapter 8, where we explain its meaning.
Initially, we will explore the concept of renderless components and subsequently build techniques to create
them. As we develop our renderless component, we will share a few examples of renderless components.
At the end of the chapter, you should possess the ability to employ the renderless component technique
to convert imperative APIs into declarative components.
The topics discussed in this chapter are as follows:

• What are renderless components?

• Building renderless components with Svelte context

Technical requirements
You can find the code used in this chapter on GitHub: https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter11.
176 Renderless Components

What are renderless components?

A renderless component, as its name implies, is a type of component that does not render any HTML
elements of its own.
You might wonder, what’s the purpose of a component that doesn’t render anything?
Well, despite not rendering HTML, there are still several useful things that a component can do,
including the following:

• Accepting props, processing their values, and triggering side effects as their values change:
Even though the prop values are not used directly in the template, they are still reactive. You
can write reactive statements with props in the component and have them run whenever the
prop values change. You can see this in the following example code snippet:
<script>
  export let title;
  export let description;
  $: document.title = `${title} - ${description}`;
</script>

Even though the title and description props are not used in the template, both title
and description are used in a reactive statement. Whenever the title or description
props change, the reactive statement in line 4 will rerun and update the title of the document.
Setting the document title, updating the cookie value, and modifying context are good examples
of side effects.
• Exhibiting component lifecycles, and performing actions as the component mounts and
gets destroyed: The component lifecycles, such as onMount and onDestroy, still run even
if no elements are mounted and destroyed in the component.
• Providing data to child elements via context setup or by passing data through the <slot>
element: Even though a renderless component does not render any HTML elements, it can
still render a <slot> element that allows the user of the component to pass in child elements
or components.
For example, in the following Parent component, we render a default slot:
<!-- Parent.svelte -->
<slot />

Then, the user of the Parent component can pass in child elements or components under
the Parent component:
<script>
  import Parent from './Parent.svelte';
</script>
<Parent>
 Exploring reusable renderless components 177

  <Child /> <!-- example of child components -->

  <div /> <!-- example of child elements -->
</Parent>

The Parent component can pass data to the child components or elements through two
avenues – setting context data and setting slot props.
You can read more about setting slot props in Chapter 4.
• Communicating with the parent component using context: Similar to any other component,
a renderless component can use getContext() to retrieve the context value set by its
parent component. Depending on the type of context value provided by the parent, we can
use the context value to communicate with the parent or inform it about the existence of the
child component.
In the Writing declarative canvas components section later in the chapter, you will see an example
of such a technique.
Throughout the upcoming sections, we will challenge ourselves by creating components that perform
only the operations listed in the preceding list, without rendering any HTML elements.
In this chapter, we will explore two use cases for renderless components, which are turning reusable
renderless components and declarative descriptions into imperative instructions. Let’s take an in-depth
look at them.

Exploring reusable renderless components

The first use case for renderless components involves creating components that solely focus on the
logic of the component. These components are not your typical ones, such as buttons or text inputs.
Instead, think of components with slightly complex logic, such as carousels, tabs, or drop-down menus.
Although the logic of a carousel component is relatively standard, its appearance can vary significantly
based on where and how it is used.
So, how can we create a reusable carousel component that can look different based on where it is used?
One solution is to create a carousel component that only contains the carousel logic, without any
specific styling or HTML structure. Then, the consumer of the component can decide how the carousel
should look by passing in their own styling and HTML structure. This allows for greater flexibility and
customization, making the component more versatile and reusable in different contexts.
For example, the carousel component could accept props such as items, which would determine the
list of items in the carousel. The carousel component could render a slot element that takes in slot
props such as currentIndex and setIndex, which would be the index of the currently active
item and the function to set the index. This allows the carousel component to manage the carousel
item cycling logic, while leaving the consumer of the carousel component to determine the actual
carousel styling and structure.
178 Renderless Components

By separating the carousel logic from the specific styling and structure, we can create a more modular
and reusable component that can be used in various contexts, without having to rewrite the same
logic over and over again. This is the power of renderless components – they allow us to create
components that focus solely on their core functionality, without being tied to any specific rendering
or styling requirements.
Of course, before we proceed, it’s essential to clarify when not to use a renderless component to create
a carousel. If you require a fully functional carousel component, including the design and style, a
renderless component may not be suitable for your needs, as its main purpose is to handle the logic
and behavior of the carousel without dictating its appearance.
Ultimately, the decision to create a renderless component for a carousel depends on your specific
needs and goals. Consider your project requirements and design preferences before deciding on the
best approach.
In the next section, I will show you step by step how we can build a renderless carousel component.

Example – building a renderless carousel component

A carousel component is a UI component that displays a set of items in a loop. It allows users to view
a collection of items in a slideshow-like format. A carousel component is usually found on websites
such as e-commerce platforms, news portals, and social media platforms.

Figure 11.1: An example of a carousel component

We will create a renderless carousel component that accepts a list of items through props named
items. The carousel component will render a slot element to allow customization of its appearance.
The slot element will take in a few slot props:

• currentIndex: This represents the index of the currently displayed item in the carousel.
• currentItem: This represents the currently displayed item in the carousel.
 Exploring reusable renderless components 179

• setCurrentIndex: This is a function that can be used to update the current index of the
carousel. It can be used to implement custom navigation controls.
• next and prev: These slot props are functions that can be used to navigate to the next or
previous item in the carousel. They can be used to implement custom navigation controls or
respond to user input, such as clicks or swipes.

These slot props allow the consumer to decide how to use them to build their own carousel UI.
To determine the appropriate slot props for the carousel component, we consider the essential
states and functions required for users to build their own carousel UI. In this case, the key states are
currentIndex and currentItem, while the necessary functions to interact with the carousel
UI include setCurrentIndex, next, and prev, which are helpful to implement custom
navigation controls.
An example of a carousel UI is one that displays the current item and has buttons to navigate forward
and backward through the carousel:

<Carousel {items} let:currentItem let:next let:prev>

  <button on:click={prev}>{'<'}</button>
  <img src={currentItem} />
  <button on:click={next}>{'>'}</button>
</Carousel>

In the preceding code snippet, we use the currentItem, next, and prev slot props to build a
simple carousel UI. We decide how to use the slot props and control the structure and styling of
the HTML elements.
Another example of a carousel UI is displaying a list of item numbers at the bottom of it, enabling the
user to quickly jump to a selected item by clicking on its number:

<Carousel {items} let:currentItem let:setCurrentIndex>

  <img src={currentItem} />
  {#each items as _, index}
    <button on:click={() => setCurrentIndex(index)}>
      {index}
    </button>
  {/each}
</Carousel>

In the preceding code, I used Svelte’s {#each} block to create a list of buttons, labeled with a different
index number. Each of the buttons has a click event listener that sets the current index of the carousel
to the index number of the button. When you click on the button, the carousel will jump to the item
at the index that is specified by the button.
180 Renderless Components

The second carousel UI is different from the first carousel UI, and you can create and style a totally
different carousel UI. You can see that the appearance of the carousel UI is entirely up to the user.
The renderless carousel component focuses on the carousel logic and allows the user to decide on
the UI structure and styling.
Without further ado, let’s explore how to write the carousel component.

Writing a renderless carousel component

In the previous section, we decided on the props of our Carousel component, items, and decided
that it should have a default <slot>.
Based on this information, let’s create the structure of our Carousel component:

<script>
  export let items;
</script>
<slot>

In the preceding code snippet, the <slot> element will be the only element we will have in our
Carousel component. This <slot> element is needed; otherwise, the child elements inside the
<Carousel> component, indicated in the following code snippet, will be ignored and discarded if
there’s no <slot> element in the Carousel component:

<Carousel>
  <!-- content over here will be ignored if the Carousel don't have a
<slot> element -->
</Carousel>

In the <slot> element, we define the following slot props – currentIndex, currentItem,
setCurrentIndex, prev, and next:

<slot {currentIndex} {currentItem} {setCurrentIndex} {prev} {next} />

However, these slot props are not defined in our Carousel component yet, so let’s define them.
We initialize currentIndex to 0 at the start of the component:

let currentIndex = 0;

setCurrentIndex is used to update currentIndex to the value being passed in:

const setCurrentIndex = (value) => { currentIndex = value; };

 Exploring reusable renderless components 181

currentItem will be the item in the items array at the index position of currentIndex. Here,
I’ll use the reactive statements so that we will have a new currentItem slot prop whenever the
items array or currentIndex changes:

$: currentItem = items[currentIndex];

Finally, the prev and next functions will be used to set currentIndex, based on its current value:

const prev = () => setCurrentIndex((currentIndex - 1 + items.length) %

items.length);
const next = () => setCurrentIndex((currentIndex + 1) % items.length);

In the preceding snippet, I used % items.length so that the index is always within the bounds of
the length of the items array. This way, we ensure that the carousel can loop back to the beginning
of the items array after reaching the end, creating a seamless looping effect for the user.
And that’s it. If you add all the preceding code snippets to the Carousel component, you’ll have a
working renderless component. The full code is available on GitHub athttps://github.com/
PacktPublishing/Real-World-Svelte/tree/main/Chapter11/01-carousel
as well.
Creating a renderless component is not hard; it doesn’t take us much time to create a renderless
Carousel component. The key here is to figure out the props for our renderless component, then
figure out the slot props needed, and finally, create a default <slot> element with the slot props.
Before we move on, let’s have an exercise to create a reusable renderless component yourself.

Exercise 1 – a renderless autocomplete component

An autocomplete component is an input with an auto-suggest dropdown that shows a list of matching
options as a user types in the input. It helps users quickly and easily find and select an option from a
large set of choices. autocomplete components are commonly used in search bars, forms, and other
areas where users need to enter data quickly and accurately.
In this exercise, we’ll create a renderless autocomplete component.
The autocomplete component will take in only one prop, a search function, which takes in a search
value and returns Promise, which resolves to an array of string results.
What about the slot props that our renderless Autocomplete component would provide?
We need three states for our autocomplete component:

• value: To represent the current value in the input box

• searching: A Boolean value to represent whether the autocomplete currently searches
for results
• suggestions: An array of autocomplete results returned from the search function
182 Renderless Components

For a user to interact with the Autocomplete component, we need two functions:

• setValue: To update the value of the input box

• selectSuggestion: To select the suggestion and apply it to the input box

So, the preceding three states and two functions will be the slot props for our
Autocomplete component.
Here is an example of how the Autocomplete component could be used:

<Autocomplete {search} let:value let:setValue let:searching

let:suggestions let:selectSuggestion>
  <input {value} on:input={event => setValue(event.currentTarget.
value)}>
  {#if searching}Searching...{/if}
  {#if suggestions}
    <ul>
      {#each suggestions as suggestion}
        <li on:click={() =>
selectSuggestion(suggestion)}>{suggestion}</li>
      {/each}
    </ul>
  {/if}
</Autocomplete>

In the preceding code snippet, we use the Autocomplete component to render an autocomplete
textbox, with an <input> element and an <ul> element in the default slot.
The <input> element uses the value and setValue slot props to access and modify value,
held by the Autocomplete component.
The <ul> element uses the suggestions slot prop to showcase the list of suggestions provided by
the Autocomplete component and the selectSuggestion slot props are used in the click
event handler of the <li> element to select the chosen suggestion and apply it to the textbox.
A sample solution can be found at https://github.com/PacktPublishing/Real-World-
Svelte/tree/main/Chapter11/02-autocomplete.
Now, let us look at the second use case for a renderless component.

Turning a declarative description into imperative

instructions
The second use case for a renderless component involves allowing users to describe their needs
declaratively and then translating them into imperative instructions.
 Turning a declarative description into imperative instructions 183

A good example of this use case is when working with a canvas or WebGL.
For example, in a canvas, to create a red rectangle with a green border, you would need to use imperative
APIs to create and style the rectangle:

ctx.fillStyle = 'red';
ctx.strokeStyle = 'green';
ctx.rect(10, 10, 100, 100);
ctx.stroke();
ctx.fill();

Step by step, we instruct the canvas context to set fillStyle and strokeStyle and then draw
a rectangle, based on the fill color and stroke color set.
When interacting with the canvas in an imperative manner, the code focuses on how to do things rather
than what to do. This can result in code that is difficult to read and maintain, with a lot of low-level
details that can make it hard to see the bigger picture.
Conversely, if you write code declaratively, you describe what you want to happen, rather than how
it should happen. This makes the code more expressive and easier to read, as well as more flexible
and reusable.
Continuing with the example of the red rectangle, we can create a Svelte component that handles drawing
the rectangle on the canvas. Instead of manually coding the instructions to draw the rectangle, we
can simply describe how we want it to appear on the canvas through the component. The component
then takes care of rendering the rectangle on the canvas for us.
Here’s an example of code that describes the same rectangle through a Svelte component:

<script>
  let x = 10, y = 10, height = 100, width = 100;
</script>
<Canvas>
  <Rectangle
    fill="red" stroke="green"
    {x} {y} {height} {width}
  />
</Canvas>

In the preceding code snippet, we see a Rectangle component nested within a Canvas component.
The Rectangle component has its fill, stroke, x position, y position, width, and height specified.
The following diagram illustrates how the Rectangle component from the preceding code renders
a red square with a green border.
184 Renderless Components

Figure 11.2: A red square with a green border

Even though explicit instructions to draw on the canvas are nowhere to be seen in the code, we can
visualize a rectangle being drawn on the canvas based on the Rectangle component. This rectangle,
with a width and height of 100 px each, is positioned 10 px from the left and 10 px from the top, filled
with a red color and a green border.
By creating a renderless component that handles the low-level canvas instructions, we can separate the
logic of drawing the rectangle from the specifics of how it should be rendered. This allows for more
flexibility in how the rectangle is displayed on the canvas, as well as easier maintenance of the code.
Furthermore, by allowing a user to describe their needs declaratively, we create a more intuitive and
user-friendly interface to work with the canvas and other low-level technologies. This can lead to
faster development times and a more enjoyable development experience.
Let me show you an example of how declaratively describing a canvas is much faster than instructing it
imperatively. If I want to animate the size of the rectangle, rather than coding the animation manually
using imperative APIs, I can simply animate the height and width variables' value, just like the
following code snippet:

<script>
  let x = 10, y = 10, height = 100, width = 100;
  setInterval(() => {
    height += 10;
    width += 10;
  }, 100);
</script>
<Canvas>
  <Rectangle
    fill="red" stroke="green"
    {x} {y} {height} {width}
  />
</Canvas>
 Turning a declarative description into imperative instructions 185

Although the use of setInterval may not be the best to create an animation, the preceding code
snippet attempts to demonstrate how easy it is to change the height and width of the rectangle in
the canvas.
In the preceding code snippet, we update height and width on every interval. We pass height
and width into the Rectangle component. As you can see, the height and width of the rectangle
on the canvas increases by 10 px every 100 ms.
Conversely, if we were to animate it imperatively, we would have to clear the canvas and redraw
a rectangle with a new height and width on each interval. These implementation details are now
abstracted away in the declarative component, making it easier to reason about the animation and
modify it in the future.
Overall, using renderless components to handle low-level imperative tasks can greatly improve
the readability, maintainability, and flexibility of our code, while also making it more accessible
and user-friendly.
So, how do we implement the Canvas and the Rectangle component shown in the preceding
code snippet? Let’s find out.

Writing declarative Canvas components

We’ll start with the Canvas component.
Similar to the component structure of the Carousel component in the previous section, the Canvas
component will render a <slot> element to insert all the child components.
However, what’s different is that it will also render a <canvas> element, which is what we will
interact with and draw on:

<script>
  import { onMount } from 'svelte';
  let canvas, ctx;
  onMount(() => {
    ctx = canvas.getContext('2d');
  });
</script>
<canvas bind:this={canvas} />
<slot />

In the preceding code snippet, we bind the reference of the <canvas> element to the variable named
canvas. After the <canvas> element is mounted onto the DOM, we get the drawing context of
the <canvas> element and assign it to the variable named ctx.
186 Renderless Components

If you recall the previous code example, we placed the <Rectangle> component inside the
<Canvas> component to draw a rectangle onto the canvas. The <Rectangle> component does
not receive any data or slot props from the <Canvas> component. So, how does the <Rectangle>
component inform the <Canvas> component that it is a child of the <Canvas> component? How
does the <Canvas> component communicate with the <Rectangle> component to know what
to draw and how to draw on its <canvas> element?
If you recall Chapter 8, we introduced Svelte context as a mechanism to set data and make it available
to child components. Svelte context allows parent components to share data and functions with the
child components without explicitly passing them as props.
We can use Svelte context for communication between the <Canvas> and <Rectangle> components.
The <Canvas> component can set the context with the drawing context, ctx:

<script>
  import { setContext } from 'svelte';
  // setting the context with the drawing context
  setContext('canvas', () => ctx);
</script>

When the <Rectangle> component is rendered as a child of the <Canvas> component, it can
access the context set by the <Canvas> component and retrieve the ctx variable. In onMount,
<Rectangle> retrieves the drawing context of the <canvas> element and draws a rectangle on
the canvas:

<script>
  import { getContext, onMount } from 'svelte';
  const getCtx = getContext('canvas');
  onMount(() => {
    const ctx = getCtxt();
    // draws a rectangle onto the canvas
    ctx.fillRect(...);
  });
</script>

The reason we pass a function that returns ctx instead of directly passing ctx in the context is that
the ctx value is only available in onMount after the <canvas> element is mounted onto the DOM,
yet setContext has to be called during component initialization, which is before onMount. In
the <Rectangle> component, you should only call getCtx()within onMount to retrieve the
latest value of ctx.
By leveraging Svelte context, the <Canvas> and <Rectangle> components can maintain a clear
and efficient communication channel. The <Canvas> component creates the canvas and provides
the drawing context, while the <Rectangle> component accesses the canvas and performs the
drawing tasks.
 Turning a declarative description into imperative instructions 187

To complete the <Rectangle> component, we need to redraw the rectangle whenever the x or y
positions or the width and height dimensions change. To do that, we will use a reactive statement:

<script>
  // ...
  export let x, y, width, height;
  $: draw(x, y, width, height);
  function draw(x, y, width, height) {
    const ctx = getCtx();
    ctx.fillRect(x, y, width, height);
  }
</script>

In the preceding reactive statement, the draw function is rerun whenever the x, y, width, or height
values change. By using this approach, the <Rectangle> component can efficiently update its
appearance in response to changes in position or dimensions, ensuring that the rendered rectangle
always reflects the latest state.
However, you might notice that it doesn’t work as expected. You’ll see that the new rectangle is drawn
on top of the old rectangle. This is because we need to clear the canvas before drawing another rectangle.
We cannot do that with our draw function, as it would lead to undesirable results if we have more
than one <Rectangle> component. Each component would clear the canvas before drawing its
own rectangle, resulting in only the last <Rectangle> component being visible.
To fix this issue, we need the <Canvas> component to clear the canvas before redrawing all the
child rectangles. This function can be called whenever any of the child rectangle components request
an update.
Let us define a redrawCanvas function in the <Canvas> component to redraw the canvas. In
the redrawCanvas function, we first clear the canvas through ctx.clearRect(). Here is the
update code for the <Canvas> component:

<script>
  // ...
  function redrawCanvas() {
    ctx.clearRect(0, 0, ctx.canvas.width, ctx.canvas.height);
    // TODO: here we need to redraw all the rectangles
  }

  // Provide both the drawing context and the clearCanvas function to
child components
  setContext('canvas', { getCtx: () => ctx, redrawCanvas });
</script>
188 Renderless Components

In the redrawCanvas function, we want to redraw all the rectangles after we clear the canvas. But
how do we do that? One idea is that instead of providing the drawing context, ctx, to all the child
components and letting the components decide when and how to draw on the canvas, we could
provide a function for the components to register their draw functions. This way, the <Canvas>
component can call the draw functions when it needs to redraw the canvas.
In the <Canvas> component, we change the getCtx function into registerDrawFunction:

<script>
  const drawFunctions = new Set();
  function registerDrawFunction(drawFn) {
    drawFunctions.add(drawFn);
    return () => {
      drawFunctions.delete(drawFn);
    };
  }

  function redrawCanvas() {
    ctx.clearRect(0, 0, ctx.canvas.width, ctx.canvas.height);
    // Redraw all the rectangles
    for (const drawFn of drawFunctions) {
      drawFn(ctx);
    }
  }

  setContext('canvas', { registerDrawFunction, redrawCanvas });

</script>

In the preceding redrawCanvas function, we loop through the drawFunctions registered from
child components and call them with the drawing context, ctx. This way, we don’t need to provide
ctx through Svelte context, yet the child components can get the latest ctx in their draw functions.
Finally, let us register our draw function in the <Rectangle> component:

<script>
  // ...
  const { registerDrawFunction, redrawCanvas } = getContext('canvas');
  function draw(ctx) {
    ctx.fillRect(x, y, width, height);
  }
  onMount(() => {
    // register the draw function
    const unregister = registerDrawFunction(draw);
    return () => {
      unregister();
 Turning a declarative description into imperative instructions 189

      redrawCanvas();
    };
  });
  // call redrawCanvas when x, y, height, width changes
  $: x, y, height, width, redrawCanvas();
</script>

In the onMount callback, we register the draw function of the <Rectangle> component through
Svelte context. When the component is destroyed and removed from the DOM, the <Rectangle>
component unregisters itself and calls the redrawCanvas function. This ensures that the <Canvas>
component is updated, and the canvas is cleared of the removed rectangle.
Moreover, by calling the redrawCanvas function in a reactive statement whenever x, y, height,
or width changes, the <Rectangle> component ensures that its position and dimensions are
accurately reflected on the canvas. This way, the <Canvas> component always maintains an up-to-
date visual representation of its child components.
Now, the <Canvas> component has full control over redrawing the entire canvas, and the <Rectangle>
components can register their draw functions with the <Canvas> component. This approach ensures
that the canvas is always cleared before redrawing and allows multiple <Rectangle> components
to coexist without interfering with each other’s drawings.
We now have functional <Canvas> and <Rectangle> components. Throughout the process of
creating these components, we have transformed imperative canvas operations into more manageable,
declarative components.
To facilitate communication between parent and child components, we utilized Svelte context as the
communication channel. As demonstrated in the preceding code snippets, the parent <Canvas>
component maintains a list of draw functions from its child components, enabling it to invoke
them as needed. This general pattern can be applied to parent components that need to track and call
methods from their child components.
While the code we’ve written is functional, it may still require some refinements. To access the
full working code, including any additional features or optimizations necessary for a polished and
comprehensive implementation, please visit the GitHub repository: https://github.com/
PacktPublishing/Real-World-Svelte/tree/main/Chapter11/03-canvas.

Exercise 2 – expanding shape components

In this exercise, we challenge you to create additional shape components to expand the capabilities
of your existing <Canvas> component.
190 Renderless Components

Examples of shapes you can create include the following:

• A <Circle> component that takes in x, y, radius, and color as props. The component
should draw a circle on the canvas at the given coordinates and with the specified radius and color.
• A <Line> component that takes in x1, y1, x2, y2, and color as props. The component
should draw a line on the canvas between the two sets of coordinates with the specified color.
• Other shape components, such as <Ellipse> and <Triangle>.

You may need to refer to the Canvas API documentation to learn how to draw different shapes.
You can find the code for the <Circle> and <Line> components at https://github.com/
PacktPublishing/Real-World-Svelte/tree/main/Chapter11/03-canvas.

Summary
Throughout this chapter, we have delved into the concept of renderless components in Svelte and
explored their various use cases. Understanding renderless components equips you with a new toolset
to create reusable components. A renderless component emphasizes reusability by focusing on the
core logic, state, and behavior, leaving the visual presentation flexible for customization.
By using slot props, we demonstrated how to build a renderless component that is reusable and gives
users control over its appearance, while maintaining the component logic and transforming imperative
operations into declarative Svelte components.
We also presented practical examples of transforming imperative operations into declarative Svelte
components. We demonstrated how to create <Canvas> and <Rectangle> components that
draw a rectangle on a canvas, which can change in size dynamically.
In the next chapter, we will explore how Svelte stores and animations can be combined to create fluid,
animated applications.
 12
Stores and Animations
In this chapter, we will delve into the world of Svelte animations, focusing on the power and versatility
of the tweened and spring stores. The tweened and spring stores are writable stores in which
their store value changes over time when the set or update method is invoked, enabling us to
develop more complex and visually appealing animations. By effectively harnessing these stores, you
can elevate the user experience and create applications that are both dynamic and captivating.
We begin this chapter by delving into the tweened and spring stores, learning how to create
animations using these stores. Following that, we explore interpolation and the use of custom
interpolations. Throughout the chapter, we examine various examples, such as animated graphs and
image lightboxes, to illustrate the concepts. By the end of this chapter, you will have acquired the skills
necessary to harness the tweened and spring stores effectively, enabling you to create intricate
and engaging animations in your Svelte projects.
This chapter covers the following topics:

• Introduction to the tweened and spring stores

• Custom interpolation and its usage
• Animating with the tweened and spring stores

Technical requirements
You can find the code used in this chapter on GitHub: https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter12

Introducing the tweened and spring stores

Let us begin our journey into the world of Svelte animations by understanding the concept of tweened
and spring stores.
The tweened and spring stores are writable stores that typically hold numeric values. To see the
features they offer, let us compare them with a regular numeric variable.
192 Stores and Animations

If you are not familiar with writable stores, you can check out Chapter 8, where we extensively explained
Svelte stores and creating writable Svelte stores using the built-in writable() function.
Usually, when you have a numeric variable and you update the variable, the value of the variable changes
instantly. In the following example, we have a numeric variable, height, whose initial value is 10.
When we assign a new value of 20 to the variable, the value of the variable changes to 20 immediately:

let height = 10;

height = 20;

If we use this numeric variable to represent the height of an element or the progress in a progress
bar, the height or progress jumps to the new value as soon as it is assigned. These abrupt changes can
be jarring.
So, how can we ensure a smooth transition when updating the target value?
Svelte provides two built-in stores, tweened and spring, specifically designed for storing numeric
values and allowing for smooth transitions to new values over a specified duration.
Let’s look at an example to get a clearer idea.
In the example, we create a tweened store with an initial value of 10:

import { tweened } from 'svelte/motion';

const height = tweened(10, {
  duration: 1000 /* 1 second */
});

Then, we assign a new value of 20 to the store:

$height = 20;

When that is done, the store value gradually increases from 10 to 20 over one second. If we use this
store value as the height of an element, the element’s height will smoothly grow or shrink toward the
target value as we assign a new value to the store, resulting in a visually appealing and fluid transition.
Let us try changing the height from a tweened store to a spring store. Do not worry about the
options in the spring function, as we will explain them in the next section:

import { spring } from 'svelte/motion';

const height = spring(10, {
  stiffness: 0.1,
  damping: 0.25
});

Now, as you assign a new value to the store, you will notice that, similar to when using the tweened
store, the store value changes over time to the new value but at a different rate.
 Introducing the tweened and spring stores 193

As you can see, the tweened and spring stores are powerful features in Svelte that enable you
to create smooth animations and transitions in your applications. These stores allow for eased value
changes. When used as component states, they allow state updates in a more natural and fluid manner.
The difference between a tweened store and a spring store is that a tweened store provides a
way to smoothly transition between two values over a specified duration using an easing function,
whereas a spring store is designed for physics-based animations, where elements behave as if they
were attached to a spring.
Let us dive in and look at how we can use the tweened and spring stores.

Using the tweened and spring stores

tweened is a store that smoothly transitions between numeric values over a specified duration using
a chosen easing function.
Here is how you can create a tweened store:

import { tweened } from 'svelte/motion';

import { cubicOut } from 'svelte/easing';
const progress = tweened(0, {
  duration: 1000,
  easing: cubicOut,
  delay: 500,
});

In the preceding snippet, we created a tweened store called progress, with 0 as the initial store
value. When you set a new value for the progress store, the store value of the progress store
stays the same for 0.5 seconds, then transitions to the new value in 1 second, using the cubicOut
easing function.
The function signature for the tweened store is as follows:

import { tweened } from 'svelte/motion';

const value = tweened(initialValue, options);

initialValue is the initial numeric value of the store.

options is an object containing the following properties:

• duration (default: 400): The duration of the transition in milliseconds.

• easing (default: linear): The easing function to use for the transition.
Svelte provides various easing functions in the svelte/easing module, such as linear,
quadIn, and expoOut. You can also create a custom easing function.
• delay (default: 0): The delay in milliseconds before the transition starts.
194 Stores and Animations

On the other hand, spring is a store that smoothly transitions between numeric values using a
spring-based physics simulation.
Here’s how you can create a spring store:

import { spring } from 'svelte/motion';

const position = spring(0, {
  stiffness: 0.2,
  damping: 0.5,
  precision: 0.001,
});

In the preceding snippet, we created a spring store called position, with 0 as the initial store
value. When you set a new value for the position store, this value will bounce toward the target
value and oscillate around the target value for a while until it settles at it. The amplitude and duration
of the oscillations depend on the stiffness, damping, and precision values configured.
The function signature for the spring store is as follows:

import { spring } from 'svelte/motion';

const value = spring(initialValue, options);

initialValue is the initial numeric value of the store.

options is an object containing the following properties:

• stiffness (default: 0.15): The stiffness of the spring. Higher values result in a stiffer spring,
which causes quicker and more forceful transitions.
• damping (default: 0.8): The damping coefficient of the spring. Higher values result in more
damping, which causes the spring to settle more quickly.
• precision (default: 0.01): The threshold at which the spring is considered to be at rest.
Smaller values result in more accurate simulations but may take longer to settle.

Using tweened and spring stores with arrays and objects

Both tweened and spring stores can handle not just single numeric values, but also arrays of
numbers and objects with numeric properties. This makes it easy to create complex animations
involving multiple values.
When you pass an array of numbers as the initial value, the stores will smoothly transition each element
of the array independently. Here’s an example using an array of two numbers:

import { tweened } from 'svelte/motion';

const coordinates = tweened([0, 0], { duration: 1000 });
// Updating the coordinates
$coordinates = [100, 200];
 Examples – creating an animated graph with the tweened and spring stores 195

Similarly, when you pass an object with numeric properties as the initial value, the stores will smoothly
transition each property independently. Here’s an example using an object with two numeric properties:

import { tweened } from 'svelte/motion';

const position = tweened({ x: 0, y: 0 }, { duration: 1000 });
// Updating the position
$position = { x: 100, y: 200 };

When using arrays or objects, you can access and use the individual values in your Svelte component
as follows:

<script>
  import { tweened } from 'svelte/motion';
  const position = tweened({ x: 0, y: 0 }, { duration: 1000 });
</script>
<div style="transform: translate({$position.x}px, {$position.y}px)"></
div>

This capability to handle arrays and objects makes the tweened and spring stores even more
versatile and powerful, enabling you to create intricate animations with ease.
Now that we know how to use the tweened and spring stores, let us use them to create an
animated graph.

Examples – creating an animated graph with the tweened

and spring stores
In this section, we will explore an example that demonstrates the power of the tweened and spring
stores. We will create an animated bar chart where the bars dynamically resize to reflect updated data
values. By adding animation to the bar chart, we can effectively highlight data changes and provide
insights into complex datasets.
Firstly, let us create a bar chart component:

<script>
  let data = generateData(10);
  function generateData(length) {
    const result = new Array(length);
    for (let i = 0; i < length; i ++) {
      result[i] = Math.random() * 300;
    }
    return result;
  }
</script>
<style>
196 Stores and Animations

  .bar {
    background-color: steelblue;
    height: 50px;
  }
</style>
<div>
  {#each data as value}
    <div class="bar" style="width: {value}px"></div>
  {/each}
</div>

In the code snippet provided, we initialize the data variable with an array of 10 randomly generated
items, created using the generateData function. The generateData function takes a length
parameter and creates an array of randomly generated data of the specified length.
With the data array, we use the {#each} block to create a <div> element for each item in the
array, setting the width of the <div> element to the value of the corresponding item in the array.
As a result, we have a horizontal bar chart displaying 10 bars with randomly generated widths.
To make things more engaging, we will update the values of the bar chart at fixed intervals:

import { onMount } from 'svelte';

onMount(() => {
  const intervalId = setInterval(() => {
    data = generateData(10);
  }, 1000);
  return () => clearInterval(intervalId);
});

We initiate an interval of 1 second using setInterval as soon as the component is mounted. On

each interval, we update the data by regenerating it with generateData(10).
With this new code addition, you’ll observe that the horizontal bars alter their width during each
interval. The width of the horizontal bars adjusts suddenly, creating a jarring visual effect.
Now, let’s utilize the tweened store to make the bars grow or shrink smoothly every time the
data changes:

1. First, let’s import the tweened store from svelte/motion. We’ll then wrap the data array
with the tweened store. As shown in the previous section, we can pass an array of numbers
to the tweened function to create a tweened store. This store will smoothly transition each
number in the array independently when updated:
import { tweened } from 'svelte/motion';
const data = tweened(generateData(10));
 Examples – creating an animated graph with the tweened and spring stores 197

2. Now, since data is a Svelte store, we need to update its value using the $data variable instead
when making changes:
$data = generateData(10);

3. Similarly, when we want to iterate through the store value of data in the {#each} block, we
need to use the $data variable as well:
{#each $data as value}

Putting all these changes together, you will now observe that the horizontal bars grow and shrink
smoothly, greatly improving the visual appeal and user experience of the bar chart.
As you can see, creating a smoothly animating chart using the tweened store is quite straightforward.
Before we move on to the next section, we encourage you to try this on your own: replace the tweened
function with spring and observe the changes in the animation.

Exercise – creating an animating line graph

Now that you’ve seen how to create an animating bar chart, it’s time for you to have a shot at creating
an animating line graph.
You can use the d3-shape library (https://github.com/d3/d3-shape), which offers a
convenient line method that generates an SVG path based on an array of items. Here is an example
of using the line method:

const pathGenerator = line().x((d, i) => i).y((d) => d);

const path = pathGenerator(data);

In the preceding code snippet, we utilize the line method to create a pathGenerator function,
which generates an SVG path by mapping the array’s values to the y coordinates. You can create a line
graph by using the returned SVG path with the <path> element in an SVG.
Once you’ve completed your implementation, feel free to compare your results with our example
at https://github.com/PacktPublishing/Real-World-Svelte/tree/main/
Chapter12/02-line-chart. Good luck and have fun experimenting with your animated
line graph!
After familiarizing ourselves with using the tweened and spring stores with numbers, arrays, and
objects, it is now time for us to incorporate non-numeric values. This will be achieved through the
creation of custom tweened interpolators, which we will explore in the next section.
198 Stores and Animations

Creating custom tweened interpolators

Sometimes, you may want to transition between non-numeric values, such as colors. Fortunately, this
doesn’t prevent us from using the tweened store. The tweened function offers an option to define
custom interpolation between two values.
In this section, we’ll explore how to create a custom interpolation to smoothly transition between
two colors.
But what is interpolation?
When transitioning between two values, interpolating means generating intermediate values between
the values, to create a smooth transition.
For example, consider a tweened store initialized at 0, and we set it to 100. The tweened store
generates intermediate values between 0 and 100, such as 20, 40, 60, and so on, while updating the
store value with these intermediate values. As a result, during the transition from 0 to 100, the store
value smoothly changes, providing a visually appealing progression from 0 to 100.
This process of generating intermediate values is known as interpolation.
The default tweened store is capable of interpolating between two numbers, two arrays of numbers,
and two objects with numeric property values. However, it doesn’t know how to interpolate between
two colors. In such cases, we can pass a custom interpolation function when creating a tweened store.
The function signature of the tweened store interpolation looks like this:

function interpolate(a, b) {
  return function (t) {
    // calculate the intermediate value between 'a' and 'b' based on
't'
  };
}

In this function, a and b represent the starting and ending values, while t is a value between 0 and
1 indicating the progress of the transition. The interpolate function should return another
function that calculates and returns the intermediate value based on t, the progress of the transition.
For example, an interpolate function that interpolates between two numbers linearly looks like this:

function interpolate(a, b) {
  return function (t) {
    return a + t * (b – a);
  };
}
 Creating custom tweened interpolators 199

The interpolate function returns another function that takes a progress value, t, and calculates
a linear interpolation between a and b based on the t value. When t is 0, the function returns a,
and when t is 1, it returns b. For values of t between 0 and 1, the result is proportionally between a
and b.To create an interpolation function for colors, we could break the colors down into individual
red, green, and blue (RGB) components and interpolate each of the components separately. After
interpolating each component, we could then recombine them to form the intermediate color.
Alternatively, we could use a library that has already implemented such interpolation. A good example
of such a library is d3-interpolate (https://github.com/d3/d3-interpolate). By
using a well-tested library such as d3-interpolate, we can save time and ensure that our color
interpolation is accurate and efficient.
Here’s an example using d3-interpolate:

import { interpolateRgb } from 'd3-interpolate';

function interpolate(a, b) {
  const interpolateColor = interpolateRgb(a, b);
  return function (t) {
    return interpolateColor(t);
  };
}

In the preceding code snippet, we import the interpolateRgb function from d3-interpolate,
which returns an interpolator function for the colors a and b. We then create our custom interpolate
function, interpolateColor, which returns a function that calculates the intermediate color
based on the progress, t.
To use our custom interpolate function when creating a tweened store, we can pass the
interpolate function in the second argument:

tweened(color, { interpolate: interpolate });

And that’s it; you can now create a tweened store for colors that can smoothly transition between them.
You can find a code example of using color interpolation on GitHub at https://github.com/
PacktPublishing/Real-World-Svelte/tree/main/Chapter12/04-interpolation.
By now, you’ve learned how to use a tweened store to create animated graphs and how to transition
between non-numeric values using a custom interpolate function.
Let’s explore more examples using the tweened and spring stores to create fluid user interfaces.
200 Stores and Animations

Examples – creating an animated image preview

In this example, we’ll create an image preview feature that allows users to view a larger, more detailed
version of a thumbnail image when they click on it, enhancing the user’s visual experience and allowing
them to inspect images more closely.
While building this feature, you’ll see how we can utilize the spring store to create a more fluid
and natural user experience, making the transitions between images and their larger previews feel
smooth and engaging.
To begin, let’s create a list of images that will be displayed on our page:

<script>
  const images = [
    "path/to/image1.jpg",
    "path/to/image2.jpg",
    // ...more image paths
  ];
  const imgElements = [];
</script>
<div class="image-container">
  {#each images as image, index}
    <div>
      <img src={image} bind:this={imgElements[index]} />
    </div>
  {/each}
</div>

In this example, we create an array of images containing the paths to our image files. We use the
{#each} block to loop through the images and create a <div> element containing an <img>
element for each image.
In the preceding snippet, the <style> section is omitted because it is not essential for understanding
how the code functions. If you would like to know what the styles look like, you can find them
at https://github.com/PacktPublishing/Real-World-Svelte/tree/main/
Chapter12/03-image-preview.
We keep the references to the <img> elements inside the imgElements variable. This will be
useful later.
To preview the images and close the preview, we need to implement two functions, openPreview
and closePreview, along with a variable, selectedImageIndex, to keep track of the currently
previewed image:

<script>
  let selectedImageIndex = -1;
 Examples – creating an animated image preview 201

  function openPreview(index) {
    selectedImageIndex = index;
  }
  function closePreview() {
    selectedImageIndex = -1;
  }
</script>
...
      <img
        src={image}
        bind:this={imgElements[index]}
        on:click={() => openPreview(index)}
      />

In the preceding code snippet, we initialize selectedImageIndex to -1, indicating that no image
is selected. The openPreview function sets selectedImageIndex, while closePreview
unsets it. Lastly, we add a click event listener to call openPreview for the clicked image.
To create a black backdrop for our image preview, we add a <div>element, which will have the
.backdrop class only when an image is selected. Clicking on the backdrop would close the preview:

<div
  class:backdrop={selectedImageIndex !== -1}
  on:click={closePreview}
/>

To display the image preview, our goal is to emphasize, enlarge, and center the image on the screen.
In order to accomplish this, we must determine the target width and height for the enlarged image
and calculate the x and y positions required to center it.
For simplicity, let’s assume that the image has a 1:1 aspect ratio. We’ll set the target width and height
to be 80% of the smaller value between the window height and window width. With the target height
and width determined, we can use these values to calculate the x and y positions required to center
the image on the screen. Let’s see how:

<script>
  let selectedImageIndex = -1;
  let width, height, left, top;
  function openPreview(index) {
    selectedImageIndex = index;
    width = Math.min(window.innerWidth, window.innerHeight) * 0.8;
    height = width; // same as width, assuming 1:1 ratio
    left = (window.innerWidth - width) / 2;
    top = (window.innerHeight - height) / 2;
  }
202 Stores and Animations

</script>
...
      <img
        src={image}
        bind:this={imgElements[index]}
        on:click={() => openPreview(index)}
        style={selectedImageIndex === index ? `
          position: fixed;
          left: ${left}px;
          top: ${top}px;
          width: ${width}px;
          height: ${height}px;
        ` : ''}
      />

In the code snippet, we set the style for the selected image. When an image is selected, we use
position: fixed to position it, allowing us to set the left and top positions of the image
relative to the viewport. At this point, we have a simple image preview component.
Now, let’s move on to an interesting question: how can we use the spring store to make the preview
more fluid, rather than abruptly placing the image at the center of the screen?
One idea is to use transform: translate instead of directly setting the left and top positions
to center the image. We can keep the left and top positions unchanged, and use transform:
translate to move the image to the center of the screen. The values for the translation offsets
would come from the spring store.
The reason for using transform: translate instead of updating the left and top positions is
that it allows for smoother and more efficient animations, as it doesn’t trigger layout recalculations
and repaints as frequently as updating positional properties such as left and top. Also, using
transform: translate makes it much easier to reset the image back to its original position
by simply resetting the translation offset back to 0.
Similarly, we can apply this idea to the width and height of the image. We can maintain the original
image dimensions and utilize transform: scale to resize the image.
With this idea ready, let’s get to the code. Here, I initialize the transformation as a spring store:

const transform = spring(

  { translate: { x: 0, y: 0 }, scale: { x: 1, y: 1 } },
  { stiffness: 0.1, damping: 0.25 }
);

The default value of the transform spring store is a 0 translation offset and 1 as the scale.
 Examples – creating an animated image preview 203

To keep the image in its original position after setting it to use position: fixed,
we need to get the <img> element’s current position and dimension, which can be obtained
through getBoundingClientRect:

function openPreview(index) {
  // ...
  const rect = imgElements[index].getBoundingClientRect();
  left = rect.left;
  top = rect.top;
  width = rect.width;
  height = rect.height;
}

The previous formula we calculated for the left, top, width, and height values to center and
enlarge the image will be our target left, top, width, and height. They will be used to calculate
the translation offset and scale:

  const targetWidth = Math.min(window.innerWidth, window.innerHeight)

* 0.8;
  const targetHeight = targetWidth;
  const targetLeft = (window.innerWidth - targetWidth) / 2;
  const targetTop = (window.innerHeight - targetHeight) / 2;
  $transform = {
    translate: {
      x: targetLeft - left,
      y: targetTop - top
    },
    scale: {
      x: targetWidth / width,
      y: targetHeight / height
    },
  };

The translation offset is calculated using the difference between the target position and the actual
position, while the scale is the ratio between the target width and height compared to the actual
width and height.
Incorporating the transform values into the <img> styles looks as follows:

<img
  style={selectedImageIndex === index ? `
  ...
  transform: translate(${$transform.translate.x}px, ${$transform.
translate.y}px) scale(${$transform.scale.x}, ${$transform.scale.y});
  ` : ''}
204 Stores and Animations

With these changes, the image will smoothly spring into the center when clicked, creating a more
fluid and natural user experience.
Now it’s your turn to try it for yourself. Experiment with creating an opacity value using the spring
store and use this value to adjust the dimness of the image preview backdrop. This will further enhance
the fluidity and visual appeal of the image preview component.

Summary
In this chapter, we looked at the tweened and spring stores from Svelte.
We explored how to use the tweened and spring stores to create smooth animations and transitions,
enhancing the visual appeal and user experience. By working with custom interpolation functions and
applying them to non-numeric values, such as colors, we’ve expanded the possibilities for creating
dynamic and engaging user interface elements.
Throughout the chapter, we’ve seen multiple examples of the tweened and spring stores in action,
seeing how easy it is to use the tweened and spring stores to create animations. Hopefully, you
are now more comfortable using the tweened and spring stores in your Svelte projects.
This is our last chapter discussing Svelte context and Svelte stores. In the next chapter, we will look
into transitions, namely, how to use transitions in our Svelte components.
 Part 4: Transitions

In this final section, we will delve into Svelte transitions. We will begin by understanding how to
incorporate built-in transitions in our Svelte components. Following that, we will guide you through
creating your own custom transitions. To end the section, we will wrap up by emphasizing accessibility
and how we can create an accessible application with transitions that caters to all users.
This part has the following chapters:

• Chapter 13, Using Transitions

• Chapter 14, Exploring Custom Transitions
• Chapter 15, Accessibility with Transition
 13
Using Transitions
Transitions are essential in creating smooth and engaging user experiences. By defining how elements
appear, disappear, or change within a user interface, transitions can turn ordinary interactions into
captivating experiences that leave lasting impressions on users.
Over the next three chapters, we will explore the topic of transitions in Svelte, beginning with a
comprehensive understanding of how to use transitions in Svelte.
In this chapter, we will start by learning how to add transitions to elements in Svelte. We will explore
the different transition directives and learn how to customize the transitions.
After that, we will discuss when and how the transitions are being played. We will look at different
scenarios, such as where there’s a mix of elements with and without transitions, or when the elements
are within nested logical blocks.
To truly master transitions, it’s important to understand the inner workings of the Svelte transition
system. We will conclude the chapter by examining the underlying mechanics and offering insights
that will help you optimize your use of transitions in your projects.
By the end of this chapter, you will have a solid foundation in Svelte transitions, allowing you to create
engaging and dynamic user interfaces with ease.
This chapter includes sections on the following:

• How to add transitions to elements

• When the transitions for the elements are played
• How transitions work under the hood
208 Using Transitions

Technical requirements
You can find the code used in this chapter on GitHub: https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter13.

Adding transitions to elements

Svelte provides a simple and powerful way to add transitions to your application elements. The
framework offers built-in transition functions that can be easily applied to elements, allowing for
smooth animations and seamless user experiences. You can also define your own custom transitions,
which we will learn about in the next chapter.
Transitions in Svelte are applied to elements when the elements are mounted or unmounted from the
DOM. This ensures that elements appear and disappear gracefully, rather than just abruptly popping
in and out of view.
To add a transition to an element in Svelte, you can use the transition: directive with the desired
transition function. Here’s an example of adding a transition to an element:

<script>
  import { fade } from 'svelte/transition';
</script>
<div transition:fade>some text here</div>

In the preceding code snippet, we imported fade from svelte/transition and applied it to
the <div> element.
What you will see with the preceding code is that when the <div> element is mounted onto the
DOM, the <div> element will smoothly fade in. When the <div> element is unmounted from the
DOM, the <div> element will smoothly fade out.
The transition: directive sets the transitions played when the element is both mounted onto the
DOM and unmounted from the DOM. If you want to have finer control over which transitions are
played as the element is mounted or unmounted, you can use the in: and out: directives instead:

<script>
  import { fade, blur } from 'svelte/transition';
</script>
<div in:fade out:blur>some text here</div>

In the preceding code snippet, we applied fade as the in transition and blur as the out transition.
When the <div> element is mounted onto the DOM, the <div> element will smoothly fade in.
When the <div> element is unmounted from the DOM, the <div> element will smoothly blur out.
 Adding transitions to elements 209

Thus, the transition: directive is essentially a shorthand for both in: and out: transitions. In
other words, the transitions applied to the two elements in the following snippet are functionally identical:

<div transition:blur>some text here</div>

So, the preceding code snippet is similar to the following code snippet:

<div in:blur out:blur>some text here</div>

From the preceding examples, we have seen two of Svelte’s built-in transitions, fade and blur – let
us look at more of them!

Svelte’s built-in transitions

Svelte’s built-in transitions are exported from the svelte/transition module.
The following list provides an overview of Svelte’s built-in transitions:

• fade: This transition smoothly fades an element in or out, adjusting its opacity over time
• blur: The blur transition gradually applies or removes a blur effect on an element
• slide: The slide transition makes an element slide smoothly in or out of view
• fly: The fly transition makes an element smoothly translate from a specified x and y offset
• scale: This transition causes an element to grow or shrink in size while appearing or disappearing
• draw: The draw transition creates a drawing or erasing effect on SVG paths

As you go through the list of built-in transitions, you may notice that some of these transitions rely on
user-specified values. For example, the fly transition depends on the specified x and y offset from
which the element should fly when transitioning in.

Customizing a transition
To make use of these transitions with their required values, you can pass a configuration object
containing the necessary properties to the transition directive:

<script>
  import { fly } from 'svelte/transition';
</script>
<div transition:fly={{ x: 200, y: 100 }}>Content goes here</div>

In the preceding code snippet, we apply the fly transition with the specified x and y offsets, indicating
that the element will fly in from 200 pixels to the right and 100 pixels down. By providing the appropriate
values, you can achieve a wide range of customized transition effects in your Svelte components.
210 Using Transitions

This approach can be particularly useful when you want an element to fly to a different location as it
transitions out, compared to the location from which it flies in during the transition.
Instead of using the transition: directive and having only one configuration for both in and out
transitions, you can separate it into the in: and out: directives and pass different configuration
objects to each directive.
An example of this can be seen in the following code snippet:

<script>
  import { fly } from 'svelte/transition';
</script>
<div in:fly={{ x: 200, y: 100 }} out:fly={{ x: -200, y: 50 }}>Content
goes here</div>

The <div> element flies in from 200 pixels to the right and 100 pixels down, while it flies out 200
pixels to the left and 50 pixels down. By separating the transition: directive into in: and out:
directives, you can control the in and out transition with different configuration objects, having more
intricate transition effects in your Svelte components.
In addition to custom configurations specific to each transition, all of Svelte’s built-in transitions accept
delay, duration, and easing as part of the transition configuration. These parameters allow you
to control the timing of your animations, providing greater flexibility in designing your user interface.
The delay parameter determines the waiting time before the transition begins while the duration
parameter specifies how long the transition lasts. By modifying these values, you can coordinate when a
transition begins and how long each transition takes, creating more complex and engaging animations.
Here is an example of adjusting the delay and duration values of a fade transition:

<script>
  import { fade } from 'svelte/transition';
</script>
<div transition:fade={{ delay: 500, duration: 1000 }}>Content goes
here</div>

In the preceding code snippet, the fade transition will start with a 500 ms delay after it is mounted
onto the DOM, and the transition will last for 1000 ms.
On the other hand, easing is a function that is responsible for controlling the pacing of the animation.
By adjusting the easing function, you can create animations that start slow and end fast, start fast
and end slow, or follow a custom pattern, giving you even more control over the look and feel of
your animations.
 When are the transitions played? 211

Svelte comes with a collection of built-in easing functions, which can be imported from svelte/
easing. These easing functions can then be applied to transitions, as seen in the following code:

<script>
  import { fade } from 'svelte/transition';
  import { quadInOut } from 'svelte/easing';
</script>
<div transition:fade={{ easing: quadInOut }}>Content goes here</div>

In the preceding code snippet, the fade transition uses the quadInOut easing function, which
causes the animation to start slowly, accelerate, and then end slowly. By incorporating different
easing functions into your transitions, you can create a variety of animations for your application.

Exercise – discovering Svelte’s built-in transitions

As a practice exercise, try to visit the official Svelte documentation and identify the list of configurable
properties for each built-in transition.
To get you started, here is a list of Svelte’s built-in transitions:

• fade
• blur
• fly
• slide
• scale
• draw

We know that transitions are played when elements are mounted or unmounted from the DOM, but
when and how exactly are the transitions played?
Let us explore the timing and manner in which the transitions are played in the next section.

When are the transitions played?

The transitions in Svelte are played when elements are added or removed from the DOM.
in: transitions are executed when an element is added to the DOM. This usually occurs when a
component is initialized or when a condition that controls the element’s rendering becomes true.
212 Using Transitions

For example, in an {#if} block, when the if condition turns from falsy to truthy, the elements
inside the {#if} block are added to the DOM. All the in: transitions applied to these elements will
be played simultaneously as soon as the elements are inserted into the DOM:

{#if condition}
  <div in:fade>some content</div>
  <div transition:blur>more content</div>
{/if}

In the preceding code snippet, as condition turns to true, both <div> elements will be inserted
into the DOM. As soon as both <div> elements are inserted, both the fade and blur transitions
will start playing simultaneously. Whether both the fade and blur transitions end at the same time
depends on the specified duration for each transition.
Conversely, out: transitions are executed when an element is removed from the DOM. This can
happen when a component is destroyed, or when a condition that controls the element’s rendering
becomes false.
The transition begins as soon as the element is scheduled for removal from the DOM. After the
transition is complete, the element is removed from the DOM.
Let us illustrate this with an example:

{#if condition}
  <div out:fade>some content</div>
  <div transition:blur>more content</div>
{/if}

In the preceding code snippet, when condition turns to false, both the <div> elements remain
in the DOM, even though the condition is no longer true. This is because the out: transition needs to
be played on both <div> elements before they are removed from the DOM. If the <div> elements
were removed from the DOM immediately, they would no longer be visible to users, rendering any
subsequent out: transitions ineffective and invisible.
Both fade and blur transitions will play simultaneously on both the <div> elements as out:
transitions. Similar to the in: transition, the duration of each transition depends on the specified
duration for each transition.
Once all the out: transitions have finished playing, both <div> elements will be removed from the
DOM together, making the DOM state consistent with the updated value of condition.
In the previous examples explaining when in: and out: transitions are played, all elements within the
{#if} block had transitions applied to them, resulting in the transitions being played simultaneously
in all the elements in the {#if} block. But, what happens if not all elements inside the {#if} block
have transitions applied? Let’s discuss that next.
 When are the transitions played? 213

Handling mixed transition and static elements

When some elements inside an {#if} block have transitions applied and others do not, Svelte handles
each element differently based on the specified transitions.
Let’s consider an example:

{#if condition}
  <div in:fade>Element with fade transition</div>
  <div>Static element without transition</div>
  <div transition:slide>Element with slide transition</div>
{/if}

In this example, when condition becomes true, the elements with transitions applied will animate
as they are inserted into the DOM, while the static element without a transition will simply appear
without any animation.
Based on the preceding code snippet, the second <div> element will be inserted and visible on the
DOM immediately, as the first and third <div> elements fade and slide in, respectively.
Similarly, when condition turns to false, the elements with out: transitions (in this case, only
the third <div> element, since the transition: directive implies both in: and out: transition)
will play their respective out transitions.
Based on the preceding code snippet, you will see that both the first and second <div> elements
remain unchanged, and the slide transition is played on the third <div> element. All elements within
the {#if} block will only be removed from the DOM together after all the out: transitions have
finished playing.
In summary, when you have a mix of elements with and without transitions inside the same logical
block, all the elements will be added into and removed from the DOM at the same time. Svelte animates
only the elements with transitions applied, while the static elements without transitions will be inserted
or removed without any animations.
So far, we have only seen examples using the {#if} block as a means of adding or removing elements,
but there are other logical blocks in Svelte that can also be used.
Let us look at what they are.

Other Svelte logical blocks for transitions

The {#if} block adds or removes elements based on the if condition. In addition to the {#if}
block, there are other logical blocks in Svelte that provide opportunities for applying transitions
when adding or removing elements, such as {#each}, {#await}, and {#key}. These blocks can
also have transitions applied to the elements they contain, providing a wide range of possibilities for
animating your user interface.
214 Using Transitions

For example, the {#each} block is used to iterate over a list of items and render elements for each
item. You can apply transitions to the elements within an {#each} block in a similar way as you
would with an {#if} block. Let’s look at an example of that:

{#each items as item (item.id)}

  <div in:fade out:slide>{item.name}</div>
{/each}

In this example, as new items are added or removed from the items array, the elements within the
{#each} block will have their respective in: and out: transitions played. When there is a new
item in the items array, the new <div> element will fade into the end of the list. When an element
is removed from the items array, the respective <div> element will slide out of the list. Using
transitions in a list allows you to create a dynamic and engaging user experience, providing clear
visual cues when items are added to or removed from the list.
Similarly, you can use transitions with {#await} and {#key} blocks to create visually appealing
animations while managing the addition and removal of elements in various scenarios.
The transition:, in:, and out: directives can be applied to any elements, and elements within
the same logical block will be added or removed at the same time. This also holds true for nested
logical blocks.
For example, let us consider the following code snippet:

{#if condition}
  <p transition:blur>paragraph 1</p>
  {#each items as item (item.id)}
    <div transition:fade>{item.name}</div>
  {/each}
  <p>paragraph 2</p>
{/if}

When condition changes from false to true, the following happens:

• The first <p> element with the blur transition will animate as it is inserted into the DOM
• Simultaneously, for each item in the items array, the <div> elements with the fade transition
will animate as they are inserted into the DOM
• The last <p> element, which does not have any transition, will simply appear in the DOM
without animation

Conversely, when condition changes from true to false, the following happens:

• The first <p> element with the blur transition will animate
• Simultaneously, for each item in the items array, the <div> elements with the fade
transition will animate
 When are the transitions played? 215

• The last <p> element, which does not have a transition, will remain unchanged
• Once all the transitions in the <p> element and all the <div> elements in the {#each} block
have finished, both the <p> and <div> elements will be removed from the DOM together

By using transitions in combination with nested logical blocks, you can create intricate animations
that enhance the user experience.
By default, transitions are only played when the nearest logical block causes an addition or removal
of the element. However, we can change this behavior with the global modifier.

The global modifier

Transitions being played only when the nearest logical block causes an addition or removal of the element
helps limit the number of simultaneous animations, making the user experience more focused and
less overwhelming. This is called the local mode; that is, transitions are only applied for local changes.
To change this behavior, we can apply the global modifier. The global modifier, when applied
to the transition:, in:, and out: directives, ensures that the animation is played whenever
the element is added or removed.
To apply the global modifier, simply suffix the directive with |global, like this:

{#if condition}
  <div in:fade|global>some text here</div>
{/if}

As per the preceding example, before applying the global modifier, the fade animation was only
being played when the nearest logical block, the {#if} block, triggered the insertion or removal of
the <div> element. This means that if another parent logical block caused the addition or removal
of the element, the animation would not be played. With the global modifier, the transition will be
played whenever the <div> element is added or removed, irrespective of which logical block causes it.
To elaborate further, let us look at the following nested {#if} block example:

{#if condition1}
  <div transition:fade>first div</div>
  {#if condition2}
    <div transition:fade>second div</div>
    <div transition:fade|global>third div</div>
  {/if}
{/if}

Let us start with condition1 as false and condition2 as true.

216 Using Transitions

When condition1 turns true, the three <div> elements will be inserted into the DOM together.
Since condition2 has always been true, at this point, the {#if} block that causes all the <div>
elements to be inserted is the one with condition1.
The first <div> element will fade in because its nearest logical block, {#if condition1}, is
responsible for the insertion of the <div> element.
The second <div> element will be immediately visible on the screen without playing the fade
transition. This is because, by default, the transition is in local mode, and its nearest logical block,
{#if condition2}, is not the cause of the <div> element being inserted at this point.
The third <div> element will fade in simultaneously with the first <div> element. Because the
<div> element has the |global modifier applied to its transition, it doesn’t matter which logical
block is responsible for its insertion. The transition will play regardless of the specific logical block
that causes the <div> element to be inserted.
Now what if condition1 turns from true to false?
The same logic applies; therefore, the second <div> element will remain unchanged, and only the
first and third <div> elements will fade out. Once the fade transition has finished, all three <div>
elements will be removed from the DOM.
In the scenarios we have walked through, the second <div> element’s fade transition has not been
played yet. So, when will the fade transition of the second <div> element be played?
To understand when the fade transition of the second <div> element will be played, let’s consider
the situation where condition1 remains true and condition2 changes from false to true.
When condition1 is true and condition2 changes from false to true, the second <div>
element will be inserted into the DOM. Since its nearest logical block, {#if condition2}, is now
responsible for the insertion, the transition:fade transition will be played.
As you can see, with the global modifier, we can change when transitions are played in response to
changes. Instead of playing the transition only when they are relevant to specific conditions affecting
the elements, we can change it to be played all the time.

Difference between Svelte 3 and Svelte 4

As we’ve explained earlier, Svelte transitions are in local mode by default. However, this is only
changed in Svelte 4, where in Svelte 3 it is the other way around. In Svelte 3, transitions are in
global mode by default, and you would need to apply the local modifier to the transition to
change it to local mode.

So far, we have covered how to add a transition to an element, using the transition:, in:, and
out: directives. We have learned when and how the transitions are played. Before we end the chapter,
let us dive deeper into the inner workings of transitions in Svelte to better understand their mechanics.
 How Svelte transition works under the hood 217

How Svelte transition works under the hood

Before we delve into the inner workings of Svelte transitions, let us first briefly discuss the general
methods for creating animations on the web. Understanding these fundamental concepts provides a
solid foundation for grasping how Svelte transitions work.
In general, you can create animations using either CSS or JavaScript.

Creating animations with CSS

To create animations using CSS, you can use the CSS animation property along with the
@keyframes rules. The @keyframes rule is used to define a sequence of styles, specifying the
CSS styles at each keyframe (from 0% to 100%) during the animation.
See this, for example:

@keyframes example {
  0% {
    opacity: 1;
    transform: scale(1);
  }
  100% {
    opacity: 0;
    transform: scale(1.75);
  }
}

In the preceding code snippet, we defined an animation keyframe named example that changes the
opacity from 100% to 0% and the scale from 1 to 1.75 simultaneously.
To apply the example animation to an element, we use the CSS animation property:

<div style="animation: example 4s 1s 1;">Animated element</div>

In the preceding code snippet, we set the animation to the example animation keyframe, with a
duration of four seconds, a delay of one second, and only play the animation exactly once.
The @keyframes rule is versatile. We have fine-grained control over the intermediate steps of
the animation sequence through @keyframes declarations. Combining it with the animation
property, we control how the animation would look, and when and how long it will play.
The advantages of using CSS to create animations are that it does not involve JavaScript, and the browser
can optimize for the CSS animation on its own. This saves JavaScript bandwidth, and as a result, the
animation can run smoothly even if you have intensive JavaScript tasks running concurrently. This
ensures a better performance and user experience, as the animations remain responsive and fluid
even under heavy processing loads.
218 Using Transitions

Creating animations with JavaScript

Creating animation using JavaScript involves manipulating the DOM elements’ styles and
properties dynamically.
As an example, let us write a fade-in animation using JavaScript.
To achieve this, we need to gradually change the element’s opacity from 0 to 1. To set a <div> element’s
opacity to 0 in JavaScript, we set it directly through the element’s style.opacity property:

div.style.opacity = 0;

In the preceding code snippet, we assumed that we have obtained a reference to the <div> element,
which we have stored in the variable named div. We then set the opacity of the <div> element to
0 through the div variable.
To animate the opacity of an element from one value to another, you would need to update the styles
at regular intervals over a specified duration of time.
Instead of setting a fixed interval through setInterval, the interval for updating styles is usually
achieved using the requestAnimationFrame method. requestAnimationFrame is a
browser method that optimizes the animation performance by calling the specified function before the
next repaint occurs. requestAnimationFrame helps ensure that your animations run smoothly
and efficiently by allowing the browser to determine the best time to update the styles, avoiding
unnecessary work or redundant repaints.
Here is an example of using requestAnimationFrame to create an animation:

let start;
const duration = 4000; // 4 seconds
function loop(timestamp) {
  if (!start) start = timestamp;
  // get the progress in percentage
  const progress = (timestamp – start) / duration;

  // Update the DOM element's styles based on progress

  if (progress > 1) {
    div.style.opacity = 0;
    div.style.transform = 'scale(1.75)';
  } else {
    div.style.opacity = 1 – progress;
    const scale = 1 + progress * 0.75;
    div.style.transform = `scale(${scale})`;
    // continue animating, schedule the next loop
    requestAnimationFrame(loop);
  }
 How Svelte transition works under the hood 219

}
// Start the animation
requestAnimationFrame(loop);

In the preceding code snippet, we schedule the loop function in the next animation frame until
the progress is finished. We calculate progress as the percentage of time passed over the total
duration of the animation. With the value of progress, we calculate the opacity and the scale of
the <div> element.
In this example, the end result of using JavaScript animation with requestAnimationFrame is the
same as the end result achieved with the examples of using CSS animations from the previous section.
The <div> element starts at opacity 1 and scale 1 at the beginning of the animation and ends up
with opacity 0 and scale 1.75 at the end of the animation.
Using JavaScript for animations offers more control over the animation logic, enabling you to create
complex and interactive animations that can respond to user input or other events.
However, one of the downsides of using JavaScript for animations is that it can be more resource-intensive
as the animations rely on the browser’s JavaScript engine to process and execute the animation logic.
Now that we have learned about the two different approaches to creating animations on the web,
which one does the Svelte transition use?
The answer is both.

Animating transitions in Svelte

Although all of Svelte’s built-in transitions use CSS for animating, Svelte is capable of animating
transitions using both CSS and JavaScript.
To animate the transition through CSS, Svelte generates a one-time-only @keyframes rule for each
element, based on the transition and the specified configuration object.
Let us look at a fly transition as an example:

<script>
  import { fly } from 'svelte/transition';
</script>
<div in:fly={{ x: 50, y: 30 }}>Some text here</div>

In the preceding code snippet, the fly transition is applied to a <div> element. In response to that,
Svelte generates a @keyframes rule that looks like this:

@keyframes fly-in-unique-id {
  0% {
    transform: translate(50px, 30px);
220 Using Transitions

    opacity: 0;
  }
  100% {
    transform: translate(0, 0);
    opacity: 1;
  }
}

This generated @keyframes rule will be applied to the element for the duration of the transition.
The unique-id part of the keyframe name ensures that each generated keyframe is unique and
doesn’t interfere with other elements.
Based on the specified duration and delay of the transition, Svelte will calculate the appropriate
timing for the animation and apply the generated @keyframes rule to the element, using the CSS
animation property. The element will then animate according to the specified transition,
duration, and delay.
For example, in the following code snippet, we have a fly transition applied to a <div> element
with a specified duration of 500 ms and a delay of 200 ms:

<script>
  import { fly } from 'svelte/transition';
</script>
<div in:fly={{ x: 50, y: 30, duration: 500, delay: 200 }}>Some text
here</div>

To animate the fly transition in the preceding code snippet, Svelte will generate the corresponding
keyframe animation, and apply the generated keyframe animation to the element with the specified
duration and delay:

div.style.animation = 'fly-in-unique-id 500ms 200ms 1';

It is also possible to animate a transition using JavaScript.

Svelte will schedule a loop through requestAnimationFrame to run the animation throughout
the specified duration.
We won’t delve further into the specifics of how the requestAnimationFrame loop works with
the animation at this point. In the next chapter, we will explore the creation of custom transitions using
JavaScript, which will provide a deeper understanding of how the requestAnimationFrame
loop interacts with animations and how to effectively utilize it for smooth, engaging transitions. Stay
tuned to learn more about crafting your own unique animations with Svelte.
 Summary 221

Summary
In this chapter, we learned how to add a transition onto an element. We explored the transition:,
in:, and out: directives, and how to customize them.
Following that, we looked at when and how the transitions are played. We discussed how the transitions
are played when we have a mix of elements with and without transitions, and also how the transitions
are played when used inside elements within nested logical blocks.
Last but not least, we dug deeper into how the transition animations are played by Svelte.
With this knowledge, you can now confidently apply transitions into elements when working with
Svelte. This will allow you to enhance the interactivity and visual appeal of your applications, thereby
providing a more engaging user experience.
In the next chapter, we will look beyond the built-in transitions and will explore the creation of
custom transitions.
 14
Exploring Custom Transitions
In this chapter, we will delve into the world of custom transitions in Svelte. So far, we have explored
Svelte’s built-in transitions and how they can be used to create engaging and dynamic user interfaces.
However, there may be situations where the built-in transitions don’t quite meet your requirements
and you want to create something more unique. This is where custom transitions come into play.
Custom transitions allow you to take full control over the animations and effects that you want to
achieve in your Svelte application. This chapter will guide you through the process of creating your
own custom transitions, whether they are CSS- or JavaScript-based. We will explore the transition
contract, which serves as the foundation for creating custom transitions, and provide practical examples
to help you get started.
By the end of this chapter, you will have a solid understanding of how to create custom transitions in
Svelte, and you will be equipped with the knowledge to implement them in your own projects, taking
your user interfaces to the next level.
This chapter includes sections on the following topics:

• The transition contract

• Writing a custom CSS transition using the css function
• Writing a custom JavaScript transition using the tick function

Technical requirements
This chapter will be somewhat code-heavy, but don’t worry—you can find all the code samples used
in this chapter on GitHub at https://github.com/PacktPublishing/Real-World-
Svelte/tree/main/Chapter14.

The transition contract

Before we dive into creating custom transitions, it’s essential to understand the foundation upon which
they are built: the transition contract.
224 Exploring Custom Transitions

If you’ve read Chapter 9, you’ll be familiar with the concept of a store contract. Just as a store is an
object that adheres to a specific store contract, a transition is a function that follows a transition
contract. By understanding and adhering to these contracts, you can create custom transitions that
integrate seamlessly with Svelte’s built-in transition system.
The transition contract consists of a single function that is responsible for the transition. This function
accepts two arguments:

• node: The target DOM element that the transition will be applied to
• params: An object containing configuration options

The function should return an object that describes how the transition should be executed. We will
delve into the details of this returned object later in the section.
Here is an example of a custom transition that adheres to the transition contract:

function customTransition(node, params) {

  const config = { ... };
  return config;
}

In the preceding code snippet, we have created a custom transition named customTransition.
We did this by declaring a customTransition function that takes in two parameters: node and
params. This function then returns an object—which we’ll refer to as config—that describes
the transition.
To relate the custom transition we’ve just created back to the usage of a transition in Svelte, here, we
see how the customTransition function is applied to a <div> element:

<div transition:customTransition={{ duration: 500 }}>some text</div>

When the <div> element is inserted or about to be removed from the DOM, Svelte will attempt to
play the transition. Svelte does this by calling the customTransition function with a reference
to the <div> element and the config object passed to the transition:

const config = customTransition(div, { duration: 500 });

This config object returned by customTransition will determine how the transition is played out.
Now, let’s focus on the requirements for the config object returned by the custom transition function.
The config object returned from a custom transition should include at least one of the following
properties or methods:

• delay: A number specified in milliseconds. This specifies how long to wait before the
transition begins.
 The transition contract 225

• duration: A number specified in milliseconds. The duration that the transition will play.
This determines how fast or slow the animation appears to the user.
• easing: A function for easing the transition. This function determines the rate of change in
the transition’s progress over time.
• css: A function that is called with two parameters: progress and remaining. Here,
progress is a value between 0 and 1 representing the progress of the transition, and the
value of the remaining parameter is equal to 1 - progress.
This function should return a string containing the CSS styles to be applied to the target
DOM element.
• tick: A function that is called repeatedly during the transition with two parameters: progress
and remaining. Here, progress is a value between 0 and 1 indicating the progress of the
transition, and the value of the remaining parameter is equal to 1 - progress.
This function can be used to update the DOM element’s styles based on the current progress.
Here is a more complete sample of a custom transition that follows the transition contract:

import { cubicInOut } from 'svelte/easing';

function customTransition(node, params) {
  return {
    duration: 1000,
    delay: 500,
    easing: cubicInOut,
    css: (progress) => `opacity: ${progress}`,
  };
}

In the preceding code snippet, our custom transition, named customTransition, returns an
object that describes the transition’s duration, delay, easing, and css styles.
We saw delay, duration, and easing in the previous chapter when applying transitions to
an element. Since the behavior of these properties remains the same in this context, let us focus on
something new: the css and tick functions.

The css function

As you might remember from the last section of the previous chapter, How Svelte transitions work under the
hood, Svelte animates transitions using a combination of CSS and JavaScript. It leverages CSS @keyframe
rules along with the animation property for CSS animations and the requestAnimationFrame
function for JavaScript animations.
226 Exploring Custom Transitions

The css function is used to generate CSS @keyframe rules for custom transitions. If the css
function is defined in the returned object of a custom transition, Svelte will call this function when
an element is inserted into the DOM or about to be removed from the DOM. Svelte will call the css
function as many times as necessary, depending on the duration and easing of the transition, in order
to generate the appropriate @keyframe rules.
The first parameter of the css function relates to the progress of the transition. progress is a
number between 0 and 1, where 0 represents the element being out of view, and 1 represents the
element being in its final position on the screen.
For example, when transitioning an element in after it has been inserted into the DOM, the value of
progress starts at 0 and moves toward 1. Conversely, when transitioning an element out before it
is removed from the DOM, the value of progress starts at 1 and moves toward 0.
You can use progress to calculate the CSS styles necessary to create your custom transition.
For example, if we want to create a transition that fades an element from transparent to fully visible,
we can use progress to calculate the opacity value throughout the transition:

• When the element is out of view (value of progress is 0), we want the element to be transparent
(value of opacity should be 0)
• When the element is in view (value of progress is 1), we want the element to be fully visible
(value of opacity should be 1)

The relationship between progress and opacity can be represented by the diagram shown in
Figure 14.1:

igure 14.1: Relationship between progress and opacity

 The transition contract 227

We are going to derive the value of opacity from the value of progress in the css function, like so:

function customTransition(node, params) {

  return {
    css: (progress) => `opacity: ${progress}`,
  };
}

Applying the transition in the preceding code will give you an element that fades from transparent
to visible when inserted into the DOM and fades back to transparent when removed from the DOM.
Let us take another example. Let us create a transition that flies the element from the right to its final
position. Here, the translation of the element changes throughout the transition, and we can use
progress to calculate the translation:

• When the element is out of view (value of progress is 0), we want the element to be on the
right (value of translateX is 100px)
• When the element is in view (value of progress is 1), we want the element to be at its final
position (value of translateX is 0px)

Here’s a diagram depicting the relationship between progress and translateX:

igure 14.2: Relationship between progress and translateX

Unlike the previous example, the value of translateX is the inverse of progress: when progress
is 0, translateX has a nonzero value; when progress is 1, translateX becomes 0.
So, to calculate the translateX value, we use 1 – progress times a value, as you can see in
the following code snippet:

function customTransition(node, params) {

  return {
    css: (progress) => `transform: translateX(${(1 – progress) * 100}
px)`,
  };
}

When applying the customTransition function in the preceding code snippet to an element,
as the element is added to the DOM, the element will fly in from the right to its final position. And
because calculating the inverse of progress (1 – progress) is so commonly done, the value
is provided as a second argument of the css function.
So, here is our custom transition again, but using the second argument instead to calculate the translation:
228 Exploring Custom Transitions

function customTransition(node, params) {

  return {
    css: (progress, remaining) => `transform: translateX(${remaining *
100}px)`,
  };
}

The css function returns a CSS string that can consist of multiple CSS declarations. You separate
each CSS declaration with a semicolon, just as you would in a style attribute of an element. For
example, let’s create a transition that combines fading in and translating from the right simultaneously:

function customTransition(node, params) {

  return {
    css: (progress, remaining) => `opacity: ${progress}; transform:
translateX(${remaining * 100}px); `,
  };
}

In the preceding code snippet, we combine both fading in and translating from the right simultaneously.
The returned CSS string contains multiple CSS declarations separated by semicolons, one for opacity
and another for transform, that will be applied during the transition.
Now that we have covered the css function, let us look at the tick function.

The tick function

The tick function serves as an alternative to the css function for creating custom transitions. Unlike
the css function, which is used to generate CSS @keyframe rules for the animation, the tick
function allows you to animate a transition using JavaScript. This can provide more fine-grained
control over the transition, enabling the creation of more complex animations that may not be easily
achieved with CSS alone.
The tick function is called repeatedly during the transition through requestAnimationFrame.
Similar to the css function, the tick function accepts two parameters: progress and remaining.
The progress parameter is a value between 0 and 1, where 0 represents the element being out of
view, and 1 represents the element being in its final position on the screen, while the remaining
parameter is equal to 1 – progress. These parameters can be used to modify the DOM element
the transition is applied to, based on the current progress of the transition.
 Writing a custom CSS transition using the css function 229

For example, if we want to create a fade-in transition using the tick function, you can update the
element’s opacity based on the progress value, as illustrated here:

function customTransition(node, params) {

  return {
    tick: (progress) => {
      node.style.opacity = progress;
    },
  };
}

Based on the preceding code snippet, Svelte triggers the tick function during every animation frame
throughout the transition.
When the element starts to appear, the progress value is 0, and we use this progress value to
set the initial opacity value of the element to 0.
As the transition continues, the tick function is called with progress values between 0 and 1,
and we update the element’s opacity value based on the progress value.
At the end of the transition, the tick function is called one last time with a progress value of 1.
At this point, we set the element’s opacity value to its final value of 1.
The tick function in the preceding code snippet operates similarly to the custom fade-in transition
we created using the css function. Both approaches modify the element’s opacity value throughout
the transition. The key difference lies in how they are executed.
Svelte calls the css function multiple times with different progress values at the beginning of the
transition to construct the CSS @keyframe rules. Once this is done, the css function is not called
again during the transition. The newly created CSS @keyframe rule is then applied to the element
through the CSS animation property. The element’s opacity value is then updated through CSS.
On the other hand, the tick function is called multiple times by Svelte during each animation frame
throughout the transition. On every tick call, the element’s opacity value is modified by JavaScript.
Now that we have learned about the transition contract, let us use this knowledge to create a few
more custom transitions.

Writing a custom CSS transition using the css function

The first custom transition we will be attempting to write together is an effect often witnessed in
presentations, commonly referred to as the “color swipe.” This effect stands out due to its dynamic sweep
of color that flows across the screen, creating a sense of energy that captivates the viewer’s attention.
The color swipe transition, as its name suggests, involves a sweeping change in color that takes place
over an object.
230 Exploring Custom Transitions

Picture this: you’re looking at a static screen, possibly a section of a website. Suddenly, a new color
begins to surface from one edge of the screen. Like a wave, this color spreads across the screen,
enveloping it. As soon as the color completely covers the screen, it starts to recede from the edge of
origin, revealing new content. When the color entirely withdraws, the new content is fully unveiled:

igure 14.3: The color swipe transition

The swipe can move in from any direction—it can move horizontally from left to right, vertically from
top to bottom, or even diagonally.
We are going to modify the color swipe transition to apply to paragraph (<p>) elements. When a <p>
element is added to the DOM, a wave of color will sweep over it, unveiling the text within the <p>
element upon completion of the transition. When the <p> element is removed from the DOM, the
reverse of the transition is played, concealing the text upon completion of the transition.
A visualization of the transition can be seen here:

igure 14.4: The color swipe transition on paragraphs

In this section, we will walk through step by step how to create this captivating color swipe transition
using Svelte.
Since the transition that conceals the text when the <p> element is being removed from the DOM
is the same as the transition that reveals the text when the <p> element is being added but played
backward, we will focus on the transition played when a <p> element is added to the DOM. This
is because when a transition is applied to an element, Svelte will play the same transition when the
element is removed, but in reverse. Therefore, by focusing on the transition played when an element
is added to the DOM, we effectively cover both scenarios.
So, let us get started on creating a transition.
Firstly, let us create the structure of our custom transition. Recall the transition contract—a transition
is a function that returns an object describing the transition:

<script>
  function colourSwipe(node) {
    // TODO: implement the transition here
    const config = {};
    return config;
  }
</script>
 Writing a custom CSS transition using the css function 231

<p transition:colourSwipe>Some text here</p>

In the preceding code snippet, we create a colourSwipe transition and apply it to the <p> element.
Our current task is to implement the colourSwipe transition by populating the config object.
The first two fields that we are going to add to the config object are duration and delay.
As shown in the following code snippet, we set the duration of the transition to 1 second, and the
transition will have no delay to start:

function colourSwipe(node) {
  const config = {
    duration: 1000,
    delay: 0,
  };
  return config;
}

However, often when creating a custom transition, you may want to allow the user of the transition
to customize the duration and delay based on where the transition is being applied.
For example, a user of the transition may want to have a delay of 200 milliseconds and a duration of 2
seconds by specifying them in the transition: directive, as shown in the following code snippet:

<div transition:colourSwipe={{ delay: 200, duration: 2000 }} />

These custom delays and durations specified in the transition: directive will be passed into
the colourSwipe transition as the second argument, which will use them in the config object:

function colourSwipe(node, params) {

  const config = {
    duration: params?.duration ?? 1000,
    delay: params?.delay ?? 0,
  };
  return config;
}

In the preceding code snippet, we use the values of params.duration and params.delay in
the config object and provide a default value when these parameters aren’t explicitly stated.
Now that we have specified the delay and the duration fields of the transition, let’s turn our
attention to the next field—easing.
232 Exploring Custom Transitions

We are going to use a linear easing, making the transition move at a constant speed, without any
acceleration or deceleration. As we did with duration or delay, we are going to make easing
customizable by the user. So, in the following code snippet, we set the value of easing based on
the user-specified easing. If it is left unspecified, we fall back on our default easing—linear easing:

import { linear } from 'svelte/easing';

function colourSwipe(node, params) {
  const config = {
    duration: params?.duration ?? 1000,
    delay: params?.delay ?? 0,
    easing: params?.easing ?? linear,
  };
  return config;
}

Typically, in the process of creating a custom transition, the duration, delay, and easing
fields are the simplest to set up. More often than not, we determine default duration, delay, and
easing values, and then offer the user the flexibility to tweak these to their liking.
Having figured out the duration, delay, and easing values, we now delve into the crux of the
transition: coming up with the CSS for the transitioning element.
If you observe the transition carefully, you will notice that the transition can be divided into two
distinct stages: the initial half involves the color block expanding to envelop the entire <p> element,
and the latter half corresponds to the color block retracting to unveil the text within the <p> element:

igure 14.5: The color swipe transition split into two halves, separated by dotted lines

Let’s explore how we can create these CSS rules to effectively bring to life these stages of the transition.
Before we start, let us create a css function in our colourSwipe transition:

function colourSwipe(node, params) {

  const config = {
    duration: params?.duration ?? 1000,
    delay: params?.delay ?? 0,
    easing: params?.easing ?? linear,
    css: (progress) => {}
  };
  return config;
}

We are going to fill in the css function in the preceding code snippet.
 Writing a custom CSS transition using the css function 233

It’s key to note that the progress value begins at 0 at the start of the transition and goes to 1 at
the end of the transition. As we are splitting the transition into two stages, the first stage will see the
progress value moving from 0 to 0.5, while the second stage advances from 0.5 to 1.
Hence, in our css function, we will implement different CSS rules for different stages of the transition:

css: (progress) => {

  if (progress <= 0.5) {
  } else {
  }
}

In the preceding code snippet, you can see that we have added an if block in the css function,
which uses the progress value to determine which sets of CSS rules to apply. For the first half of the
transition (progress <= 0.5), the first set of CSS rules is implemented. For the latter half (progress
> 0.5), the second set of rules is used. This way, we can customize the element’s appearance in distinct
ways throughout the two stages of the transition.
In the first half of our transition, we need to create a growing color block. To create this, we will apply
a linear gradient on the background of the element. The gradient will transition from a solid color to
a transparent color. By aligning the color stop point of the solid color and the transparent color at the
same location, we can create a sharp hard line in the gradient transition.
For example, if we want a solid red color block that occupies the left 25% of an element, we could
apply the following CSS:

background: linear-gradient(to right, red 0, 25%, transparent 25%);

In the preceding snippet, we have a linear gradient moving from left to right, with the red and
transparent colors sharing the same color stops at the 25% point. This creates a solid red block on the
leftmost 25% of the gradient, while the remaining 75% is transparent.
Our choice to use a linear gradient to achieve this color block, as opposed to superimposing another
element, shows the simplicity of this approach. It eliminates the necessity of creating an additional element.
Another benefit of setting a linear gradient on the background, instead of resizing the element, is to
have a resizing color block effect without actually resizing an element, which would cause re-layout
and layout shifts in the DOM. This way, the element with the applied CSS remains static in its original
position and size throughout the transition.
234 Exploring Custom Transitions

So, now we’ve figured out the CSS to use, let’s incorporate it into our transition’s css function.
Before that, we need to do some math. We intend to use the value of progress to calculate the
percentage of the element to be covered by the solid color block.
In the first half of the transition, the value of progress goes from 0 to 0.5.
Within this phase of the transition, the percentage of the element that needs to be covered should
range from 0% to 100%.
Consequently, by performing arithmetic calculations, we can conclude that the percentage value is
200 times the value of progress:

const percentage = progress * 200;

Let’s now integrate this into our css function:

css: (progress) => {

  if (progress <= 0.5) {
    const percentage = progress * 200;
    return `background: linear-gradient(to right, red 0,
${percentage}%, transparent ${percentage}%);`;
  } else {
  }
}

In the preceding code snippet, we use the calculated percentage to control the size of the solid color
block, making it grow from the left from 0% to 100% through the first half of the transition.
Now, let’s turn our attention to the second half of the transition, where the solid color block contracts
from full width toward the right edge of the element.
An alternate way to envision this is by considering the expansion of the transparent portion from
the left edge, covering from 0% to 100% of the element. This mirrors the first half of the transition,
with the key difference being that it’s now the transparent color, not the solid one, that grows to fully
envelop the element.
The formula for calculating the percentage value remains the same, but since the progress value
now ranges from 0.5 to 1, we need to subtract 0.5 from the progress value before multiplying
it by 200. So, the equation becomes this:

const percentage = (progress – 0.5) * 200

With this modification, our css function now becomes this:

css: (progress) => {

  if (progress <= 0.5) {
    const percentage = progress * 200;
 Writing a custom CSS transition using the css function 235

    return `background: linear-gradient(to right, red 0,

${percentage}%, transparent ${percentage}%);`;
  } else {
    const percentage = (progress – 0.5) * 200;
    return `background: linear-gradient(to right, transparent 0,
${percentage}%, red ${percentage}%);`;
  }
}

In this updated function, the solid color block and transparent areas dynamically resize according to
the calculated percentage during the transition, effectively creating the visual illusion of a color swipe.
Applying the transition to an element now, you might notice that although we have a functional
solid color swipe transition effect, there are a few elements that could be refined for a smoother
visual experience.
One prominent aspect is that the text within the element remains visible throughout the entire
transition. Ideally, it should stay hidden in the first half of the transition, when the solid color block
is expanding, and only be revealed in the second half as the color block shrinks.
The following screenshot illustrates this issue:

igure 14.6: The text is not hidden in the first half of the transition

To address this, we can set the text color to transparent during the first half of the transition, as shown
in the following code snippet:

css: (progress) => {

  if (progress <= 0.5) {
    const percentage = progress * 200;
    return `background: linear-gradient(to right, red 0,
${percentage}%, transparent ${percentage}%); color: transparent;`;
  } else { /* ... */ }
}
236 Exploring Custom Transitions

Another issue is that the solid color block remains red, regardless of the text color. Because we are
using the CSS background property to create a swipe effect, the text remains at the forefront, while
the solid color block is in the background.
This affects the text-revealing effect from the color block since the full text content becomes fully
visible in the second half of the transition. If the solid color block shared the same color as the text,
the text would blend with the background. This would create a visual illusion, giving the appearance
that the text is being revealed as the color block contracts.
The following screenshot illustrates this issue:

igure 14.7: The text and block color do not match

To resolve this, we need to find a way to obtain the color of the text and incorporate it into our linear
gradient background.
The window.getComputedStyle() function allows us to fetch the styles applied to an element.
We can use this function to get the color of the text at the beginning of the transition and use that
color for our gradient background:

function colourSwipe(node, params) {

  const { color } = window.getComputedStyle(node);
  const config = {
    css: (progress) => {
      if (progress <= 0.5) {
        const percentage = progress * 200;
        return `background: linear-gradient(to right, ${color} 0,
${percentage}%, transparent ${percentage}%); color: transparent;`;
      } else {
        const percentage = (progress – 0.5) * 200;
        return `background: linear-gradient(to right, transparent 0,
${percentage}%, ${color} ${percentage}%);`;
      }
    }
  };
  return config;
}

In the preceding revised code snippet, we replace red with the text color we fetched from the node
element’s computed style.
 Writing a custom JavaScript transition using the tick function 237

And there you have it—a customized color swipe effect, implemented as a Svelte transition. The
complete code can be found at https://github.com/PacktPublishing/Real-World-
Svelte/tree/main/Chapter14/01-css-transition.
We went step by step through creating a custom Svelte transition using CSS. Throughout the process,
we learned how to implement user-customizable attributes into our transition, such as duration,
delay, and easing. In our color swipe transition, we learned how to craft a multi-staged transition
and how to segment the progress parameter into various stages, utilize its value to compute the
CSS rules, and apply it to the element.
Hopefully, you are now equipped to create your own custom Svelte transition using CSS.
At the start of this chapter, we learned that a transition contract can include not only a css function
but also a tick function. The tick function allows us to modify elements during the transition.
We’ve explored how to use the css function to create a color swipe transition; in the next section,
we’ll delve into creating another custom transition, this time using the tick function.

Writing a custom JavaScript transition using the tick

function
The custom transition we will attempt to write in this section is a flipboard transition. The transition
emulates the mechanics of vintage airport departure boards. During this transition, each letter of the
text flips, cycling through characters until it finally lands on the correct one. The transition ends when
all the letters have settled into the right character.
238 Exploring Custom Transitions

The following diagram illustrates how the flipboard transition works to reveal the phrase Hello Svelte,
with the vertical axis representing the flow of time from top to bottom:

igure 14.8: Flipboard transition visualized

At the onset of the transition, letters begin to appear from left to right, starting as a dash (-) and then
flipping through random characters before settling on the correct one. This flipping motion continues
from left to right until all letters have aligned with their corresponding characters, unveiling the
intended phrase.
The entire transition process involves modifying characters within the element, transitioning from
a blank state through jumbled characters, and finally to the correct text. Since there are no style or
layout changes needed, we are not using the css function to implement this transition. The tick
function is the perfect candidate to implement this transition in Svelte.
Now that we’ve defined how the flipboard transition would look, let us start implementing this transition.
Building on what we learned with the color swipe transition in the previous section, the flipboard
transition begins in a similar way. Here is the basic code structure for our flipboard transition:

<script>
  function flipboard(node, params) {
    const config = {
      duration: params?.duration ?? 1000,
      delay: params?.delay ?? 0,
      easing: params?.easing ?? linear,
      tick: (progress) => {
        // TODO: implement the transition here
      },
    };
    return config;
  }
</script>
<p transition:flipboard>Hello Svelte.</p>

In the preceding code snippet, we have defined a flipboard function that adheres to the transition
contract. It takes in two arguments, node and params, and returns a transition configuration—an
object that describes the transition. As such, we are able to apply the flipboard function as a
transition using the transition: directive on a <div> element.
Within the flipboard transition, we’ve set up the basic parameters, such as defining the duration,
delay, and easing values, while leaving a placeholder for the tick function, where we will
implement the transition.
 Writing a custom JavaScript transition using the tick function 239

To create a flipboard transition, we start by obtaining the text content of the element that will be
transitioning. Then, each time the tick function is called, we determine the text to display based
on the progress value and update the element accordingly.
We can retrieve the text of an element using the following API:

const text = node.textContent;

Similarly, to set the text content of an element, we assign it through the same property:

node.textContent = text;

Incorporating these into the flipboard transition, here’s what we get:

function flipboard(node, params) {

  const text = node.textContent;
  const config = {
    // ...
    tick: (progress) => {
      let newText;
      // TODO: compute the newText based on `text` and progress value
      node.textContent = newText;
    },
  };
  return config;
}

In the preceding code snippet, we retrieve the text content of the element at the beginning of the
flipboard function, right before the transition starts to play on the element. The tick function,
called repeatedly on every animation frame, computes the new text value for the element, based
on the original text value and current progress value.
240 Exploring Custom Transitions

The tick function’s task is to determine how each letter should be displayed, based on the progress
value. Some letters might appear as a dash, some as random characters, some as their original value,
and others might be hidden.
For each letter, its display depends on its position relative to the full text’s length and the current
progress value. For instance, if a letter is positioned at 30% from the left, and the current progress
value is 0.5 (or 50%), then that letter should be displayed as is.
How do we determine these rules? What leads us to the conclusions just drawn?
We want the element to reveal all its original characters at the end of the transition. This means that
when the progress value reaches 1 (or 100%), all letters should display their original character. At
the midpoint, with a progress value of 0.5 (or 50%), 50% of the letters from the left should show
their original character, while the remaining 50% on the right should show either a dash, a random
character, or nothing at all.
To generalize, if a letter’s position from the left is less than the progress value, it should display its
original character. Otherwise, it may display a dash, a random character, or nothing.
The following diagram illustrates how the progress value and the displayed text are related:

igure 14.9: The red box shows which letters are displaying the original character at each progress value

The preceding diagram illustrates how the original characters are revealed at different values of
progress. As the progress value increases, more letters in the word display their original characters.
To implement the flipboard effect just described, we will loop through each character, determine its
relative position, and then decide whether it should be displayed. For characters whose position is
beyond the current progress value, we will display a blank space.
Here is the updated code:

tick: (progress) => {

  let newText = '';
  for (let i = 0; i < text.length; i++) {
    const position = i / text.length;
    if (position < progress) {
      // display the original character
      newText += text[i];
    } else {
      // display a blank space instead
      newText += ' ';
    }
  }
 Writing a custom JavaScript transition using the tick function 241

  node.textContent = newText;
},

With this code, the flipboard transition now either shows the original characters or a blank space
based on the progress value. When playing the transition, you will see the characters appear one by
one from left to right.
Having figured out when to show up the original characters, let us continue to determine when the
text should display dashes or random characters.
Using a similar idea, we can determine whether a letter should display a random character, a dash,
or nothing at all. We will introduce a constant number to manage the timing of these changes. If the
letter’s position is beyond this constant multiplied by the progress value, it will show up as nothing.
I’ve chosen a value of 2 for this constant, based on trial and error. It’s a value greater than 1 but not
too large, to create the effect of characters appearing gradually from the left.
Similarly, we could also introduce another constant to manage the display of dashes or random
characters. If the letter’s relative position is greater than this new constant times the progress value, but
less than 2, the letter will appear as a dash. Otherwise, it will be a random character. For this purpose,
I have chosen 1.5, positioning it midway between 1 and 2.
The following diagram visually represents these two constants and their effect on the transition:

igure 14.10: Relationship between the new text and the original text
242 Exploring Custom Transitions

In the preceding diagram, you can observe how the characters change during the transition. When the
progress value is 0.4, for example, letters at the 40% position are showing original characters, letters
at 40% - 60% (progress * 1.5) are showing random characters, letters at 60% - 80% (progress
* 2) are showing dashes, and anything beyond is not shown.
Here’s how the updated code for our flipboard transition looks:

tick: (progress) => {

  let newText = '';
  for (let i = 0; i < text.length; i++) {
    const position = i / text.length;
    if (position < progress) {
      // display the original character
      newText += text[i];
    } else if (position < progress * 1.5) {
      // display random characters
      newText += randomCharacter()
    } else if (position < progress * 2) {
      // display dash
      newText += '-';
    } else {
      // display a blank space instead
      newText += ' ';
    }
  }
  node.textContent = newText;
},

In the preceding code snippet, I’ve added two additional conditions to determine when to display
dashes or random characters.
The randomCharacter() function returns a randomly selected character, implemented as follows:

const chars =
'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz1234567890'
function randomCharacter() {
  return chars[Math.floor(Math.random() * chars.length)];
}

With this code, you have a flipboard transition! Try it out with an element and observe the effect. The
characters slowly appear one by one from left to right, starting as a dash, flipping through characters,
and finally settling with the correct character.
 Summary 243

You may notice one small issue: not all characters are the same width, so the overall width of the text
grows and shrinks. Since each letter is not aligned with its previous position, the flipping effect may
not be immediately obvious.
To fix this, you can use a monospace font. Monospace fonts, also known as fixed-width fonts, ensure
that each letter occupies the same horizontal space. Applying a monospace font to an element enhances
the flipping effect, making it more visually distinct.
For example, you can set the font as follows:

font-family: monospace;

In this section, we’ve explored how to create a flipboard transition, emulating the appearance of
vintage airport departure boards. We’ve learned how to control the appearance of characters based
on the progress of the transition, using random characters, dashes, and original text. And through
modifying the text of an element, we created a transition that is visually engaging.
The complete code for this section can be found at here:https://github.com/PacktPublishing/
Real-World-Svelte/tree/main/Chapter13.

Summary
In this chapter, we have explored how to create custom transitions in Svelte. We went through two
detailed examples that utilize the css function and the tick function.
Hopefully, you now feel well equipped to write your own custom transition in Svelte, enabling you to
craft a more engaging and unique user experience in Svelte.
In our next and final chapter, we will delve into how transitions can impact the accessibility of your
Svelte application, guiding you in the creation of an engaging and inclusive experience for all users.
 15
Accessibility with Transitions
In the past two chapters, we learned how to use transitions in Svelte. Transitions, when used correctly,
can enhance the user experience, guiding a user’s attention, providing feedback, and adding a layer
of polish to the interface. However, for users with vestibular disorders, these animations can be
uncomfortable or even debilitating. Therefore, it is essential to strike a balance between creating
engaging animations and ensuring that they do not negatively impact users with specific needs.
In this chapter, we will dive into the techniques available to make web transition more accessible for
users with a vestibular disorder, exploring CSS and JavaScript approaches to respect a user’s preferences
regarding motion.
By the end of this chapter, you will have a better understanding of web accessibility and how to create
more inclusive web applications that cater to all users, regardless of their specific needs or preferences.
In this chapter, we will cover the following:

• What is web accessibility?

• Understanding user preference with prefers-reduced-motion
• Reducing motion for a Svelte transition
• Having an alternative transition for inaccessible users

Let’s start by exploring what web accessibility is.

Technical requirements
All the code in this chapter can be found at the following link: https://github.com/
PacktPublishing/Real-World-Svelte/tree/main/Chapter15/01-accessible-
transition
246 Accessibility with Transitions

What is web accessibility?

Accessibility is the design of products, devices, services, or environments to be usable by as many
people as possible, regardless of any physical, sensory, or cognitive disabilities they may have.
It is vital to ensure that websites are accessible to all users. There are many disabilities that can potentially
affect a user’s experience on a website. Ensuring a website’s accessibility allows everyone, regardless of
their abilities, to have equal access to the same services and information that are available to everyone.
One of the many disabilities that can hinder a website’s user experience is a vestibular disorder. In this
chapter, we will specifically focus on enhancing accessibility for individuals with vestibular disorders.
Vestibular disorders are conditions that affect the inner ear and brain, and they can cause difficulties
with balance, spatial orientation, and movement perception. Imagine that your body’s natural sense
of balance isn’t working right. It’s like being dizzy or feeling tipsy. The ground under your feet doesn’t
feel stable, and the things you see around you seem to move on their own, even if you’re standing still.
For individuals with vestibular disorders, certain visual stimuli, such as moving or flashing content
on a web page, can trigger symptoms such as dizziness, nausea, or migraines.
We learned how to add transitions in Chapter 13 and Chapter 14 to make our application more
engaging for users. However, for users with vestibular disorders, these transitions can inadvertently
provide a negative experience. Most operating systems provide accessibility settings, enabling users
with vestibular disorders to reduce or remove animation. These accessibility preferences can be used
by web applications to create an inclusive user experience.
So, let’s explore how a web application accesses a user’s accessibility preference.

Understanding user preference with prefers-reduced-

motion
Most operating systems offer accessibility settings that allow users to disable animation effects. For
instance, in Windows 11, you can navigate to Settings | Accessibility | Visual Effects | Animation
Effects and uncheck the Animation Effects option to turn off animations.
 Understanding user preference with prefers-reduced-motion 247

igure 15.1: The Animation effects option in Window 11

In web applications, you can use the prefers-reduced-motion CSS media query to determine
whether a user has activated a setting on their device to reduce or eliminate non-essential motion.
Here is an example of how to use the prefers-reduced-motion CSS media query:

@media (prefers-reduced-motion: reduce) {

  div {
    /* Removes animation */
    animation: none;
  }
}

In the preceding code snippet, if a user has indicated a preference for reduced motion, we set the CSS
animation property to none to remove animation from the <div> element.
Alternatively, besides using CSS, you can also use JavaScript to determine a user’s preference for
reduced motion. The window.matchMedia method lets you check whether the web page matches
a given media query string.
248 Accessibility with Transitions

In the following code snippet, we use window.matchMedia to test whether the prefers-
reduced-motion CSS media query is matched:

const mediaQuery = window.matchMedia('(prefers-reduced-motion:

reduce)');
const matches = mediaQuery.matches;

If the user prefers reduced motion, then the value of matches in the preceding code snippet will be
true; otherwise, the value of matches will be false.
A user might change their accessibility preferences while browsing a web page. To get notified whenever
the user alters their preference for reduced motion, we can listen to the change event of the media
query. Here’s how:

const mediaQuery = window.matchMedia('(prefers-reduced-motion:

reduce)');
mediaQuery.addEventListener('change', () => {
  let matches = mediaQuery.matches;
});

In the preceding code snippet, whenever a user changes their preference for reduced motion, the
listener function passed into the change event handler will be invoked. It will evaluate the updated
user preferences through mediaQuery.matches.
Now that we’ve learned how to determine user preferences for reduced motion through prefers-
reduced-motion, let us take a look at how we can use it to reduce Svelte’s transition for users
with vestibular disorders.

Reducing motion for Svelte transition

After learning how to obtain a user’s preference for reduced motion, let’s now respect that preference by
reducing unnecessary motions in our transitions, which could potentially trigger vestibular discomfort.
In the following code block, there is an example of our Svelte component, which has a fly transition
applied to the list items:

<script>
  import { fly } from 'svelte/transition';
  export let list = [];
</script>
<ul>
  {#each list as item}
    <li transition:fly={{ x: 40 }}>{item}</li>
  {/each}
</ul>
 Reducing motion for Svelte transition 249

In the preceding code, whenever a new item is added to the list, a new <li> element will fly in from the
right and be inserted into the list. This flying motion could be a trigger for users with vestibular disorders.
However, the flying transition is not essential because the application will still function correctly
without it. Therefore, if a user has indicated a preference for reduced motion in the system settings,
we should respect that preference by reducing or removing the flying transition.
One way to achieve this is by setting the duration of the fly transition to 0. This way, the transition
will take no time to play and complete and, in effect, will not be played at all.
Here’s the modified version of the previous Svelte component, which does not play the fly transition
if the user prefers reduced motion:

<script>
  import { fly } from 'svelte/transition';
  export let list = [];
  const mediaQuery = window.matchMedia('(prefers-reduced-motion:
reduce)');
  let prefersReducedMotion = mediaQuery.matches;
  mediaQuery.addEventListener('change', () => {
    prefersReducedMotion = mediaQuery.matches;
  });
</script>
<ul>
  {#each list as item}
    <li transition:fly={{
      x: 40,
      duration: prefersReducedMotion ? 0 : 400,
    }}>{item}</li>
  {/each}
</ul>

In the preceding code snippet, we determine whether a user prefers reduced motion by checking
whether the CSS media query, prefers-reduced-motion: reduce, is matched, and we store
this information in a variable named prefersReducedMotion. If prefersReducedMotion
is true (meaning the user prefers reduced motion), then we set the duration of the fly transition
to 0. The user will not see any flying motion when a new item is added to the list.
On the other hand, if the user does not have a vestibular disorder and has not expressed a preference
for reduced motion effects, then prefersReducedMotion will be false. In this case, the
duration for the fly transition will be set to 400 ms, and a flying transition will be displayed for
every new item added to the list.
250 Accessibility with Transitions

However, not all transitions act as triggers for vestibular motion disturbances. For instance, the fade
transition, being a subtler animation, isn’t as disruptive to users with vestibular disorders. Instead
of entirely eliminating transitions by setting their duration to 0, we can opt to replace more intense
transitions with milder ones. We’ll delve into this approach in the following section.

Having alternative transitions for inaccessible users

Users with vestibular disorders may feel discomfort when exposed to motion-based animations, such
as scaling or panning large objects. However, they are generally less affected by subtler animations,
such as fading.
Switching all transitions to fading for users with vestibular disorders is not a one-size-fits-all solution.
It is always better to seek feedback from the users themselves.
We will continue using the same example from the previous section and explore how we can switch
from the fly transition to a fade transition when a user prefers reduced motion.
One thing to note is that, in Svelte, you are not allowed to apply more than one transition to an element.
For example, the following code is invalid and will result in a build error:

<div transition:fade transition:fly />

This means we can’t apply two transitions to an element and then decide which one to use. We must
find a way to switch between different transitions while applying only one transition to an element.
As we learned in Chapter 14 about creating custom transitions, a transition in Svelte is a function
that follows a transition contract. The return value from the function determines how the transition
will play out.
So, one approach to having a transition that switches between two transitions, based on a condition, is
to create a custom transition that returns different transition configurations, based on the condition.
Our custom transition would look something like the following code, returning different transition
configurations based on whether a user prefers reduced motion:

<script>
  function accessibleFly(node, params) {
    const mediaQuery = window.matchMedia('(prefers-reduced-motion:
reduce)');
    const matches = mediaQuery.matches;
    if (matches) {
      // user prefers reduced motion
      // return a fade transition
    } else {
      // return a fly transition
 Having alternative transitions for inaccessible users 251

    }
  }
</script>
<ul>
  {#each list as item}
    <li transition:accessibleFly={{ x: 40 }}>{item}</li>
  {/each}
</ul>

In the preceding code snippet, we defined an accessibleFly transition, which is a more accessible
fly transition that will switch to a fade transition if a user prefers reduced motion.
Now, we need to determine what to return in each of the conditional cases within our custom
accessibleFly transition.
It’s important to recall that a transition in Svelte is a JavaScript function. Therefore, we can call both
the fly and fade transitions as functions, and the return value will be the transition configuration
for each respective transition. By doing this, we can return these values from our accessibleFly
transition, effectively allowing our transition to be either a fly or fade transition, based on a
user’s preference.
Here’s the updated accessibleFly transition:

<script>
  function accessibleFly(node, params) {
    const mediaQuery = window.matchMedia('(prefers-reduced-motion:
reduce)');
    const matches = mediaQuery.matches;
    if (matches) {
      // user prefers reduced motion
      return fade(node, params);
    } else {
      return fly(node, params);
    }
  }
</script>

In the preceding code snippet, we return the value either from the fade or fly transition, depending
on a user’s preference. We pass the node and params values that are given to our accessibleFly
transition into both the fade and fly transitions. The node and params values specify which
element the transition is applied to and provide user parameters, such as duration and delay.
These are useful for the fade and fly transitions to determine how the transitions should be executed.
252 Accessibility with Transitions

With the preceding code changes, we now have an accessible fly transition named accessibleFly,
which, by default, will cause an element to fly as part of the transition. However, the accessibleFly
transition will cause the element to fade if a user has indicated a preference for reduced motion.
And there you have it–a transition that is engaging and appealing to most users, while still being
considerate of those who prefer reduced motion.
You can find the code for the accessibleFly transition at https://github.com/
PacktPublishing/Real-World-Svelte/tree/main/Chapter15/01-accessible-
transition.

Summary
In this chapter, we explored the importance of accessibility in web design and how to implement
transitions that consider the preferences of users with vestibular disorders. By understanding the
impact of motion-based animations on users with vestibular disorders, we can create more inclusive
and user-friendly web applications.
We learned about the prefers-reduced-motion media query, which allows us to detect whether
a user has indicated a preference for reduced motion in their system settings. Using this media query,
we can adjust our transitions to be less motion-heavy or remove them altogether for users who prefer
reduced motion.
We also discussed how to create custom transitions in Svelte for accessibility. We looked at an example
of a custom transition, named accessibleFly, that switches between a fly and a fade transition,
based on a user’s preferences for reduced motion. This custom transition is considerate of users with
vestibular disorders while still providing engaging and interesting transitions for other users.
In summary, accessibility is crucial in web design, and transitions are no exception. By considering
the preferences and needs of all users, including those with vestibular disorders, we can create more
inclusive and user-friendly web applications.
 Index
A
accessibility 246 built-in transitions
actions 73 blur 209
defining 74, 75 discovering 211
DOM event logic, reusing with draw 209
custom events 75 fade 209
drag-and-drop event, creating 84 fly 209
form inputs validation, with scale 209
custom events 81-84 slide 209
<a> element
enhancing, progressively 109-114
animated bar graph
C
creating, with tweened and Canvas components 185
spring stores 195-197 declarative Canvas components 185
animated image preview shape components 189
creating, with spring store 200-204 carousel component 178
animating line graph example 178, 179
creating, with tweened and spring stores 197 Cascading Style Sheets (CSS) 17
animations animations, creating with 217
creating, with CSS 217 child component
creating, with JavaScript 218, 219 controlling, through context 54, 55
Autocomplete component 181 controlling, through props 54
using 182 dynamic content, passing through slot 55
manipulating 54

B styles, controlling 55
CodeMirror 97
bindings 39, 40 integrating 97, 98
one-way, versus two-way data flow 41-43 reference link 97
254 Index

complex derived states debounced higher-order Svelte store

managing 46-49 creating 149
Component/Presentational pattern 68 derived method, using 155, 156
benefits 68 new store, deriving with derived()
component tree 123 function 153, 154
component types original store, subscribing 152, 153
rendering 59-61 original store, unsubscribing 152, 153
config object store value changes, debouncing 149-152
methods 224 declarative Canvas components
Container Components 67 writing 185-189
Container/Presentational pattern 67, 68 declarative description
CSS custom properties turning, into imperative
component styles modification, instructions 182-185
allowing 31, 32 derived states
dark/light theme mode props, updating with 49-52
implementation 30, 31 Document Object Model (DOM) 5
defining 28, 29 elements 67
fallback value, aligning 33 DOM event logic reuse, with
real-world example 30 custom events 75
Svelte components, theming with 27, 28 logic, encapsulating into action 77
css function 225-228 logic, encapsulating into component 76, 77
CSS Modules parameters, passing to action 78-81
importing 23 drag-and-drop event
custom CSS transition creating 84
writing, with css function 229-237 dynamic content
custom events passing, through slots 56
DOM event logic, reusing with 75 dynamic context
form inputs, validating with 81-84 passing, with Svelte store 136, 137
custom JavaScript transition dynamic global data 136
writing, with tick function 237-243 dynamic local data 136
custom transition
sample 225
custom tweened interpolation
F
creating, with tweened store 198, 199 <form> element 114
default form behavior 114, 115

D enhance action, implementing 115-117

enhancing, progressively 114
data flow form inputs
one-way, versus two-way data flow 41-43 validating, with custom events 81-84
 Index 255

G O
global modifier 215 one-way data flow 41, 42
applying, to transitions 215, 216
graceful degradation 106
P
H Presentational components 68
progressive enhancement 106
HTML element Svelte actions, using 108
rendering 59-61 web experience 106, 107
web pages, building in layers 107

J prop drilling approach 122

props
JavaScript defining 36, 37
animations, creating with 218, 219 state deriving from, reactive
Jotai library declaration used 44, 45
URL 166 updating, with derived states 49-52
JSON tree viewer component 64 versus state 37-39
building 64-67

R
L react-calendar library
lifecycle functions 4 integrating, in Svelte 101, 102
calling, rule 7 reference link 101
component, destroying 6 reactive declaration
component, initializing 5 used, for deriving state from props 44, 45
component, mounting 5 Rectangle component 183
components, registering 8 recursive component
component, updating 6 creating, for recursive data 62, 63
composing, into reusable hooks 13-15 recursive data structure 62
coordinating, across components 15, 16 Renderless component pattern 59
refactoring 7, 8 renderless components 176
reusing, in Svelte components 9-12 advantages 176, 177
Scroll blocker example 16 reusable renderless components 177, 178
update counter example 12, 13 renderless autocomplete
link component 181, 182
previewing, with <a> element 109-114 renderless carousel component,
building 178-180
256 Index

renderless carousel component, CSS Modules, importing 23

writing 180, 181 CSS rules, scoping 20
lifecycle functions, reusing 9-12

S method selection, for styling 24

style block, adding 19
server-side rendering (SSR) 108 styles, applying from external CSS files 22
shape components styling 18
expanding 189 styling, with style attribute 18
slots 56 styling, with style: directives 18, 19
data, passing through slot props 58, 59 theming, with CSS custom properties 27
default slot content, providing 57 Svelte context
dynamic content, passing through 56 component tree 123
multiple slots, with named slots 57, 58 defining 121, 122
spring store 193 object, using as context key 124
animated image preview, value, changing 124, 125
creating with 200-204 versus Svelte store 133-135
creating 194 Svelte lifecycle functions. See
function signature 194 lifecycle functions
state Svelte store
defining 37 auto-subscribing 131-133
deriving from props, reactive creating, from user events 140, 141
declaration used 44, 45 creating, with built-in function 130, 131
managing, with Svelte store 159-163 defining 126-130
managing tips 163-166 event listener, adding only once 141-144
versus props 37-39 implementing 144
state management libraries observer pattern, using 126, 127
using, with Svelte 166, 167 states, managing with 159-163
static global data 136 subscribe function, defining 127
static local data 136 used, for passing dynamic context 136, 137
Svelte Valtio, using as 167-172
react-calendar, integrating 101, 102 XState state machine, transforming into 173
state management libraries,
using with 166, 167
styling, with Tailwind CSS 25-27
T
Svelte components Tailwind CSS 17
class attribute, adding 21 setting up 25-27
class attribute, simplifying with Svelte, styling with 25
class: directive 21 URL 27
 Index 257

tick function 228, 229

used, for writing custom JavaScript
U
transition 237-243 UI libraries, written in other frameworks
Tippy.js 94 components, cleaning up 99
integrating 94-96 components, creating 98, 99
reference link 94 components, updating 100
transition contract 224, 225 using 98
arguments 224 undo/redo logic
css function 225-228 implementing 147, 148
tick function 228, 229 undo/redo store
transitions 207 creating 145-147
adding, to elements 208
animating, in Svelte 219, 220
built-in transitions 209
V
built-in transitions, discovering 211 Valtio
customizing 209-211 URL 167
global modifier, applying 215, 216 using 167, 168
inner workings 217 using, as Svelte store 167-172
logical blocks, using 213, 214 vanilla JavaScript UI libraries integration
mixed transition and static alternative, to Svelte actions 92, 93
elements, handling 213 data, adding to Svelte actions 90-92
play timeline 211, 212 into Svelte 86
tweened and spring stores 191, 192 Svelte actions, using 86-89
animated bar graph, creating with 195-197 Svelte actions, versus Svelte
animating line graph, creating with 197 components 93, 94
using 193, 194
using, with arrays and objects 194, 195
tweened store 193
W
creating 193 web accessibility 246
custom tweened interpolation, alternative transitions, for
creating with 198, 199 inaccessible users 250-252
function signature 193 reduced motion, for Svelte
two-way data flow 42, 43 transition 248, 249
user preference, accessing with prefers-
reduced-motion 246, 247
Web Graphics Library (WebGL) 175
258 Index

X
XState
URL 166, 173
XState state machine
transforming, into Svelte store 173
www.packtpub.com
Subscribe to our online digital library for full access to over 7,000 books and videos, as well as
industry leading tools to help you plan your personal development and advance your career. For more
information, please visit our website.

Why subscribe?
• Spend less time learning and more time coding with practical eBooks and Videos from over
4,000 industry professionals
• Improve your learning with Skill Plans built especially for you
• Get a free eBook or video every month
• Fully searchable for easy access to vital information
• Copy and paste, print, and bookmark content

Did you know that Packt offers eBook versions of every book published, with PDF and ePub files
available? You can upgrade to the eBook version at packtpub.com and as a print book customer, you
are entitled to a discount on the eBook copy. Get in touch with us at customercare@packtpub.
com for more details.
At www.packtpub.com, you can also read a collection of free technical articles, sign up for a range
of free newsletters, and receive exclusive discounts and offers on Packt books and eBooks.
 Other Books You May Enjoy
If you enjoyed this book, you may be interested in these other books by Packt:

Svelte with Test-Driven Development

Daniel Irvine
ISBN: 978-1-83763-833-8

• Create clear and concise Vitest unit tests helping the implementation of Svelte components
• Use Playwright and Cucumber.js to develop end-to-end tests that simulate user interactions
and test the functionality of your application
• Leverage component mocks to isolate and test individual components
• Write unit tests for a range of Svelte framework features
• Explore effective refactoring techniques to keep your Svelte application code and test suites clean
• Build high-quality Svelte applications that are well-tested, performant, and resilient to changes
 Other Books You May Enjoy 261

SvelteKit Up and Running

Dylan Hildenbrand
ISBN: 978-1-80461-548-5

• Gain a comprehensive understanding of the core technologies of SvelteKit

• Dive deep into SvelteKit’s file-based routing system and uncover basic and advanced concepts
• Master code organization in SvelteKit through effective data-loading techniques
• Elevate your application’s visibility by incorporating SEO and accessibility
• Streamline the deployment process of your SvelteKit apps by using adapters
• Optimize your SvelteKit app by exploring various configuration and tooling options
262

Packt is searching for authors like you

If you’re interested in becoming an author for Packt, please visit authors.packtpub.com and
apply today. We have worked with thousands of developers and tech professionals, just like you, to
help them share their insight with the global tech community. You can make a general application,
apply for a specific hot topic that we are recruiting an author for, or submit your own idea.
Hi!
I am Tan Li Hau, author of Real-World Svelte. I really hope you enjoyed reading this book and found
it useful for increasing your productivity and efficiency.
It would really help me (and other potential readers!) if you could leave a review on Amazon sharing
your thoughts on this book.
Go to the link below to leave your review:
https://packt.link/r/1804616036

Your review will help us to understand what’s worked well in this book, and what could be improved
upon for future editions, so it really is appreciated.
Best wishes,

Tan Li Hau
 263

Download a free PDF copy of this book

Thanks for purchasing this book!
Do you like to read on the go but are unable to carry your print books everywhere?
Is your eBook purchase not compatible with the device of your choice?
Don’t worry, now with every Packt book you get a DRM-free PDF version of that book at no cost.
Read anywhere, any place, on any device. Search, copy, and paste code from your favorite technical
books directly into your application.
The perks don’t stop there, you can get exclusive access to discounts, newsletters, and great free content
in your inbox daily
Follow these simple steps to get the benefits:

1. Scan the QR code or visit the link below

https://packt.link/free-ebook/9781804616031

2. Submit your proof of purchase

3. That’s it! We’ll send your free PDF and other benefits to your email directly

You might also like

• SOURABH SHARMA - Modern API Development With Spring 6 and Spring Boot 3 - Design Scalable, Viable, and Reactive APIs With REST, GRPC, and GraphQL-Packt Publishing (2023)

  

  Document494 pages

  SOURABH SHARMA - Modern API Development With Spring 6 and Spring Boot 3 - Design Scalable, Viable, and Reactive APIs With REST, GRPC, and GraphQL-Packt Publishing (2023)

  Javier Almary

  Noch keine Bewertungen
• Java Coding Problems (For Viis - Anghel Leonard

  

  Document762 pages

  Java Coding Problems (For Viis - Anghel Leonard

  Gerry Dela Cruz

  100% (1)
• Full Stack Development With Spring Boot 3 and React - Build Modern Web Applications Using The Power of Java, React, and TypeScript-Packt Publishing (2023)

  

  Document455 pages

  Full Stack Development With Spring Boot 3 and React - Build Modern Web Applications Using The Power of Java, React, and TypeScript-Packt Publishing (2023)

  21110917

  Noch keine Bewertungen
• • Magazines
  • Podcasts
  • Sheet music
• NET MAUI For C# Developers Jesse Liberty, Rodrigo Juarez, 2023 Packt

  

  Document296 pages

  NET MAUI For C# Developers Jesse Liberty, Rodrigo Juarez, 2023 Packt

  svwerwolf

  Noch keine Bewertungen
• Sinchan Banerjee - Scalable Data Architecture With Java - Build Efficient Enterprise-Grade Data Architecting Solutions Using Java-Packt Publishing (2022)

  

  Document382 pages

  Sinchan Banerjee - Scalable Data Architecture With Java - Build Efficient Enterprise-Grade Data Architecting Solutions Using Java-Packt Publishing (2022)

  jerry.voldo

  100% (1)
• Sandilands D. Puppet 8 For DevOps Engineers. Automate Your Infrastructure... 2023

  

  Document416 pages

  Sandilands D. Puppet 8 For DevOps Engineers. Automate Your Infrastructure... 2023

  pourlestele

  Noch keine Bewertungen
• Frameworkless Front End Development Do You Control Your Dependencies or Are They Controlling You 2Nd Edition Francesco Strazzullo Full Chapter

  

  Document68 pages

  Frameworkless Front End Development Do You Control Your Dependencies or Are They Controlling You 2Nd Edition Francesco Strazzullo Full Chapter

  greg.garrett128

  100% (13)
• 9781837634255-Android Ui Development With Jetpack Compose

  

  Document278 pages

  9781837634255-Android Ui Development With Jetpack Compose

  brigidmacedo

  Noch keine Bewertungen
• Microservices Design Patterns in .NET Making Sense of Microservices Design and Architecture Using .NET Core (Trevoir Williams) (Z-Library)

  

  Document298 pages

  Microservices Design Patterns in .NET Making Sense of Microservices Design and Architecture Using .NET Core (Trevoir Williams) (Z-Library)

  gdrg

  Noch keine Bewertungen
• Designing Hexagonal Architecture With Java

  

  Document460 pages

  Designing Hexagonal Architecture With Java

  Gustavo Marquez

  100% (2)
• Core and Vue - Js - 2021

  

  Document479 pages

  Core and Vue - Js - 2021

  Yousef Elsaeed

  Noch keine Bewertungen
• Go For DevOps Packops - Ir 2022

  

  Document878 pages

  Go For DevOps Packops - Ir 2022

  longhoangdazai

  100% (1)
• Packt Csharp Interview Guide

  

  Document362 pages

  Packt Csharp Interview Guide

  stroganovboris

  Noch keine Bewertungen
• Android UI Development With Jetpack Compose (Thomas Künneth) (Z-Library)

  

  Document349 pages

  Android UI Development With Jetpack Compose (Thomas Künneth) (Z-Library)

  Julio Cesar

  Noch keine Bewertungen
• Mastering Blazor WebAssembly (Ahmad Mozaffar, 2023) Packt

  

  Document370 pages

  Mastering Blazor WebAssembly (Ahmad Mozaffar, 2023) Packt

  svwerwolf

  Noch keine Bewertungen
• Hands-On Design Patterns With React Native Proven Techniques and Patterns For Efficient Native Mobile Development (Mateusz Grzesiukiewicz)

  

  Document297 pages

  Hands-On Design Patterns With React Native Proven Techniques and Patterns For Efficient Native Mobile Development (Mateusz Grzesiukiewicz)

  Martín Oña

  100% (1)
• LLVM Techniques, Tips, and Best Practices Clang and - Yih Hsu - 2021 - Anna's Archive

  

  Document370 pages

  LLVM Techniques, Tips, and Best Practices Clang and - Yih Hsu - 2021 - Anna's Archive

  belgacem

  100% (1)
• Tan Li Hau - Real-World Svelte - Supercharge Your Apps With Svelte 4 by Mastering Advanced Web Development Concepts-Packt (2023)

  

  Document282 pages

  Tan Li Hau - Real-World Svelte - Supercharge Your Apps With Svelte 4 by Mastering Advanced Web Development Concepts-Packt (2023)

  AntonioMolina

  Noch keine Bewertungen
• Mobile App Reverse Engineering Get Started With Discovering, Analyzing, and Exploring The Internals of Android and iOS Apps (Abhinav Mishra) (Z-Library)

  

  Document166 pages

  Mobile App Reverse Engineering Get Started With Discovering, Analyzing, and Exploring The Internals of Android and iOS Apps (Abhinav Mishra) (Z-Library)

  doolalos

  Noch keine Bewertungen
• Building Your Own JavaScript Framework

  

  Document247 pages

  Building Your Own JavaScript Framework

  Pranit Waydhane

  Noch keine Bewertungen
• Building Your Own JavaScript Framework

  

  Document247 pages

  Building Your Own JavaScript Framework

  Pranit Waydhane

  Noch keine Bewertungen
• Developing Web Components With Svelte Building A Library of Reusable UI Components

  

  Document350 pages

  Developing Web Components With Svelte Building A Library of Reusable UI Components

  Febrian Alif Hermawan

  Noch keine Bewertungen
• Simplifying State Management in React Native by Al

  

  Document203 pages

  Simplifying State Management in React Native by Al

  Tran quoc Huy

  100% (1)
• Learn Blockchain Programming With JavaScript (Eric Traub (Eric Traub) ) (Z-Library)

  

  Document669 pages

  Learn Blockchain Programming With JavaScript (Eric Traub (Eric Traub) ) (Z-Library)

  07 Kavya Sundara Moorthy

  Noch keine Bewertungen
• (3.1) William S Vincent - Django For APIs - Build Web APIs With Python and Django. 3.1 (2020)

  

  Document9 pages

  (3.1) William S Vincent - Django For APIs - Build Web APIs With Python and Django. 3.1 (2020)

  Rebeca Alvarez Cepeda

  Noch keine Bewertungen
• Infrastructure As Code For Beginners - Deploy and Manage Your Cloud-Based Services

  

  Document222 pages

  Infrastructure As Code For Beginners - Deploy and Manage Your Cloud-Based Services

  david flores

  100% (1)
• Practical Module Federation

  

  Document167 pages

  Practical Module Federation

  Atishay.Jain

  Noch keine Bewertungen
• Data Structures and The Java Collections Framework.9780470482674.56071

  

  Document760 pages

  Data Structures and The Java Collections Framework.9780470482674.56071

  nilsmart

  100% (2)
• Alex Boten - Cloud-Native Observability With OpenTelemetry-Packt Publishing PVT LTD (2022)

  

  Document386 pages

  Alex Boten - Cloud-Native Observability With OpenTelemetry-Packt Publishing PVT LTD (2022)

  alote1146

  Noch keine Bewertungen
• Real World Next - Js

  

  Document367 pages

  Real World Next - Js

  sate258

  100% (1)
• Enterprise React Development With UmiJS Douglas Alves Venancio 1

  

  Document198 pages

  Enterprise React Development With UmiJS Douglas Alves Venancio 1

  William Wesley

  Noch keine Bewertungen
• Real World Implementation of C# Design Patterns Bruce M Van Horn Packt

  

  Document443 pages

  Real World Implementation of C# Design Patterns Bruce M Van Horn Packt

  svwerwolf

  100% (1)
• 9781803243450-Template Metaprogramming With C

  

  Document480 pages

  9781803243450-Template Metaprogramming With C

  pixelfruitsstudio

  Noch keine Bewertungen
• Mern Stack Online

  

  Document22 pages

  Mern Stack Online

  devshreeenterprises23

  Noch keine Bewertungen
• Practical Design Patterns For Java Developers

  

  Document266 pages

  Practical Design Patterns For Java Developers

  todevg

  100% (1)
• Fundamentals of Trace and Log Analysis - A Pattern-Oriented Approach To Monitoring, Diagnostics, and Debugging - 1st Ed (2023)

  

  Document151 pages

  Fundamentals of Trace and Log Analysis - A Pattern-Oriented Approach To Monitoring, Diagnostics, and Debugging - 1st Ed (2023)

  ludickeys

  Noch keine Bewertungen
• Android Summit Learn Jetpack Compose by Example

  

  Document156 pages

  Android Summit Learn Jetpack Compose by Example

  mrnovoa

  Noch keine Bewertungen
• Svelte Custom Stores TypeScript Declaration - by David Dal Busco - Better Programming

  

  Document14 pages

  Svelte Custom Stores TypeScript Declaration - by David Dal Busco - Better Programming

  howesteve

  Noch keine Bewertungen
• Building Games With Flutter

  

  Document224 pages

  Building Games With Flutter

  Riyad

  Noch keine Bewertungen
• How To Build Android Apps With Kotlin 2-Packt Pul

  

  Document704 pages

  How To Build Android Apps With Kotlin 2-Packt Pul

  Rubén Vilchez

  Noch keine Bewertungen
• Testing Spring Boot Applications Demystified (Philip Riecks) (Z-Library)

  

  Document101 pages

  Testing Spring Boot Applications Demystified (Philip Riecks) (Z-Library)

  gdrg

  Noch keine Bewertungen
• Linux Kernel Programming Kaiwan N Billimoria

  

  Document769 pages

  Linux Kernel Programming Kaiwan N Billimoria

  akademicwisher

  Noch keine Bewertungen
• Alan Mellor - Test-Driven Development With Java - Create Higher-Quality Software by Writing Tests First With SOLID and Hexagonal Architecture (2023, Packt Publishing) - Libgen - Li

  

  Document348 pages

  Alan Mellor - Test-Driven Development With Java - Create Higher-Quality Software by Writing Tests First With SOLID and Hexagonal Architecture (2023, Packt Publishing) - Libgen - Li

  Gabriel Martinelli

  Noch keine Bewertungen
• Quarkus 1

  

  Document10 pages

  Quarkus 1

  Laszlo Kamler

  Noch keine Bewertungen
• Modern Android 13 Development

  

  Document628 pages

  Modern Android 13 Development

  kelvin kitonyo

  Noch keine Bewertungen
• Responsive Web Design With AngularJS Sample Chapter

  

  Document34 pages

  Responsive Web Design With AngularJS Sample Chapter

  Packt Publishing

  Noch keine Bewertungen
• jsp3 Toc

  

  Document11 pages

  jsp3 Toc

  niked12118

  Noch keine Bewertungen
• Test-Driven Development With C++ A Simple Guide To Writing Bug-Free Agile Code (Tanner, Abdul Wahid) (Z-Library)

  

  Document430 pages

  Test-Driven Development With C++ A Simple Guide To Writing Bug-Free Agile Code (Tanner, Abdul Wahid) (Z-Library)

  Jeff Gurguri

  Noch keine Bewertungen
• Chapter 5

  

  Document9 pages

  Chapter 5

  Sneha Nikam

  Noch keine Bewertungen
• Building Microservice Solutions

  

  Document35 pages

  Building Microservice Solutions

  WashynAceroM

  Noch keine Bewertungen
• 9781119711308

  

  Document352 pages

  9781119711308

  gogdodog

  Noch keine Bewertungen
• Michael Hartl - Learn Enough JavaScript To Be Dangerous - Write Programs, Publish Packages, and Develop Interactive Websites With JavaScript (2022, Addison-Wesley Professional) - Libgen - Li

  

  Document687 pages

  Michael Hartl - Learn Enough JavaScript To Be Dangerous - Write Programs, Publish Packages, and Develop Interactive Websites With JavaScript (2022, Addison-Wesley Professional) - Libgen - Li

  Maricruz Delgado

  Noch keine Bewertungen
• Hands-On High Performance With Go Boost and Optimize The Performance of Your Golang Applications at Scale With Resilience (Bob Strecansky)

  

  Document387 pages

  Hands-On High Performance With Go Boost and Optimize The Performance of Your Golang Applications at Scale With Resilience (Bob Strecansky)

  Ankit Kumar

  Noch keine Bewertungen
• Developing Graphics Frameworks Java Opengl

  

  Document309 pages

  Developing Graphics Frameworks Java Opengl

  Jose Capablanca

  Noch keine Bewertungen
• Getting Started With Angular (2023) Create and Deploy Angular App

  

  Document410 pages

  Getting Started With Angular (2023) Create and Deploy Angular App

  ap s

  Noch keine Bewertungen
• Eric Smith - Game Development With Rust and WebAssembly - Learn How To Run Rust On The Web While Building A Game-Packt Publishing (2022)

  

  Document476 pages

  Eric Smith - Game Development With Rust and WebAssembly - Learn How To Run Rust On The Web While Building A Game-Packt Publishing (2022)

  barath2k

  Noch keine Bewertungen
• Cracking Core Java Interviews - v3.5

  

  Document266 pages

  Cracking Core Java Interviews - v3.5

  aashish.kpn1

  100% (1)
• From Java To Kotlin PDF

  

  Document9 pages

  From Java To Kotlin PDF

  Mridupaban Dutta

  Noch keine Bewertungen
• Hands Software Architecture Golang

  

  Document700 pages

  Hands Software Architecture Golang

  Tarun Dugar

  Noch keine Bewertungen
• Analysis Node - Js Platform Web Application Security

  

  Document60 pages

  Analysis Node - Js Platform Web Application Security

  Sorin Gîtlan

  Noch keine Bewertungen
• Rapid Game Development Using Cocos2d JS An End To End Guide To 2D

  

  Document135 pages

  Rapid Game Development Using Cocos2d JS An End To End Guide To 2D

  lordinicus

  Noch keine Bewertungen
• The C Workshop Kickstart Your Career As A Software Developer With C (Jason Hales, Almantas Karpavicius Etc.) PDF

  

  Document780 pages

  The C Workshop Kickstart Your Career As A Software Developer With C (Jason Hales, Almantas Karpavicius Etc.) PDF

  Jose Luis Ferreira

  Noch keine Bewertungen
• Ultimate Salesforce LWC Developers' Handbook

  

  From Everand

  Ultimate Salesforce LWC Developers' Handbook

  Cihan Fethi Hizar

  Noch keine Bewertungen
• Building Modern Web Applications With JakartaEE, NoSQL Databases and Microservices: Create Web Applications Jakarta EE with Microservices, JNoSQL, ... and MicroProfile easily (English Edition)

  

  From Everand

  Building Modern Web Applications With JakartaEE, NoSQL Databases and Microservices: Create Web Applications Jakarta EE with Microservices, JNoSQL, ... and MicroProfile easily (English Edition)

  Geovanny Mendoza Gonzalez

  Noch keine Bewertungen
• Heroku Cloud Application Development

  

  From Everand

  Heroku Cloud Application Development

  Anubhav Hanjura

  Noch keine Bewertungen
• Frontend Roadmap

  

  Document1 page

  Frontend Roadmap

  anonim

  Noch keine Bewertungen
• Svelte Handbook

  

  Document40 pages

  Svelte Handbook

  bardi Gideon

  Noch keine Bewertungen
• 50+ Best Web Development Tools and Resources PDF

  

  Document20 pages

  50+ Best Web Development Tools and Resources PDF

  kibet caleb

  Noch keine Bewertungen
• Sveltekit Supabase Typescript Without Prisma

  

  Document9 pages

  Sveltekit Supabase Typescript Without Prisma

  sssakol1981

  Noch keine Bewertungen
• Beginning Shadow DOM API Get Up and Running - Alex Libby

  

  Document183 pages

  Beginning Shadow DOM API Get Up and Running - Alex Libby

  talwarapappu

  Noch keine Bewertungen
• 01 Svelte Components

  

  Document8 pages

  01 Svelte Components

  rahey63273

  Noch keine Bewertungen
• Cypress Component Testing 1

  

  Document2 pages

  Cypress Component Testing 1

  கார்த்திக் பெரியசாமி

  Noch keine Bewertungen
• Svelte Succinctly

  

  Document114 pages

  Svelte Succinctly

  Breno Santos

  Noch keine Bewertungen
• Remote-Frontend Jobs

  

  Document1 page

  Remote-Frontend Jobs

  Harshad Bhale

  Noch keine Bewertungen
• Large Scale Apps With Svelte and TypeScript by Dam

  

  Document273 pages

  Large Scale Apps With Svelte and TypeScript by Dam

  Tran quoc Huy

  100% (1)
• Svelte JS

  

  Document18 pages

  Svelte JS

  Chaithu Chaitanya

  Noch keine Bewertungen
• Zahidhossain Resume

  

  Document2 pages

  Zahidhossain Resume

  tamzidmustaine khan

  Noch keine Bewertungen
• Sveltekit en

  

  Document105 pages

  Sveltekit en

  Ivan Vilches

  Noch keine Bewertungen
• Svelte en

  

  Document108 pages

  Svelte en

  sam negro

  Noch keine Bewertungen
• 250+ Portfolio Ideas

  

  Document33 pages

  250+ Portfolio Ideas

  aditya12bone

  Noch keine Bewertungen
• Developing Web Components With Svelte Building A Library of Reusable Ui Components 2Nd Edition Alex Libby Full Chapter

  

  Document54 pages

  Developing Web Components With Svelte Building A Library of Reusable Ui Components 2Nd Edition Alex Libby Full Chapter

  george.wright130

  100% (18)
• 50+ Best Web Development Tools and Resources

  

  Document20 pages

  50+ Best Web Development Tools and Resources

  echidne

  Noch keine Bewertungen
• SvelteKit Up and Running Leverage The Power of A Next-Generation Web Framework To Build High-Performance Web Apps With Ease (Dylan Hildenbrand) (Z-Library)

  

  Document166 pages

  SvelteKit Up and Running Leverage The Power of A Next-Generation Web Framework To Build High-Performance Web Apps With Ease (Dylan Hildenbrand) (Z-Library)

  liang luyu

  100% (1)
• Svelte Custom Stores TypeScript Declaration - by David Dal Busco - Better Programming

  

  Document14 pages

  Svelte Custom Stores TypeScript Declaration - by David Dal Busco - Better Programming

  howesteve

  Noch keine Bewertungen
• Front End

  

  Document1 page

  Front End

  Divyanshu

  Noch keine Bewertungen
• Report File

  

  Document33 pages

  Report File

  Chaitanya Jagarwal

  Noch keine Bewertungen
• Frontend Dev Roadmap

  

  Document2 pages

  Frontend Dev Roadmap

  2217 Nishchay sharma

  Noch keine Bewertungen
• Frontend Developer Roadmap

  

  Document5 pages

  Frontend Developer Roadmap

  michael marco

  Noch keine Bewertungen
• Astro Getting Started

  

  Document3 pages

  Astro Getting Started

  Cristina Raluca Glăvan

  Noch keine Bewertungen