Edit in GitHubLog an issue

Adobe Express Document API Concepts

The structures that make up the Adobe Express Document Object Model, their features, and how they're reflected in the Reference documentation.

Overview

In this article, you'll dive deep into the architecture and key elements of the Adobe Express Document Object Model (DOM) and why the Reference Documentation is the primary tool to explore it. Understanding the structures at play, hierarchy, and inheritance system will help you develop add-ons that exploit their full potential.

Getting started with the DOM

The notion of Document Object Model (DOM) is key to any scripting environment; developers refer to scripting when their programs use tools primarily available through the application's User Interface (UI) but programmatically, with code. For example, the Grids tutorial add-on creates a grid system while operating with built-in elements like shapes and editing their dimensions, positions, fills, and blending modes. If you were to implement an unsupported object type, you'd need to go beyond the combination of existing tools—i.e., outside the scripting realm.1

It is essential to hierarchically organize the features that are surfaced2 to the scripting layer. For example, a shape is not just contained within a document: it may be included in a particular Group in a certain Artboard, which belongs to a specific Page of a Document. Additionally, Ellipses and Rectangles, as shapes, share some properties, such as the ability to be filled or stroked with a solid color; in that respect, though, they differ from a bitmap image, which can be stroked but not filled.

More formally stated, any scripting environment must expose information about the Containment Structures (parent/child relations, like groups nesting shapes) and Inheritance Hierarchies (shared attributes like the opacity property available to both) of any element it deals with. Such a collection of organized information is called the Document Object Model, or DOM.

Developers with a front-end background may instinctively associate the notion of DOM with HTML and the Browser. Although it has been the most common case for decades, all applications supporting scripting rely on their own Document Object Model—as those working with desktop versions of the Creative Cloud know very well. The Adobe Express add-on system is technically different from CEP extensions in Adobe After Effects or UXP plugins in Adobe Photoshop; still, the concept of the DOM is equally valid.

Coding along

You are encouraged to try the code snippets discussed here; please use this boilerplate to follow along. It contains a demo add-on in two states: the complete version (in the express-dimensions-end folder, whose code is also found in the Final Project section at the bottom of this page) and a starting point (express-dimensions-start) that you can also use to test your assumptions and inspect the Console. The Document API code belongs to the src/documentSandbox/code.js file.

Copied to your clipboard
// ...
import { editor, colorUtils } from "express-document-sdk";
function start() {
runtime.exposeApi({
log: () => {
const selectedNode = editor.context.selection[0];
console.log("Currently selected node", selectedNode);
},
// ...
});
}

Run the add-on, select any object in the UI, and click the "Log selected node" button: it'll be logged in the Console.

refs addon log

The Adobe Express DOM

An Adobe Express document is internally represented as a hierarchical tree of nodes. Some of them may branch: in other words, contain other nodes, host children—like an Artboard or a Group with some shapes or media in them. Other ones are so-called leaf nodes: the tree's endpoints, for example, a text or a line.

refs addon scenegraph

Such a structure is referred to as the Scenegraph, and it allows for efficient rendering, manipulation, and traversal of all scene elements.

Object-Oriented programming concepts

Understanding the hierarchical tree of nodes in the Adobe Express DOM requires a firm grasp of a handful of key concepts in object-oriented programming (OOP).

  • Classes: the blueprint for creating objects. They define a set of properties and methods the created objects (known as instances of that class) will have. For example, when users add a page to an Adobe Express document, a new PageNode class is instantiated. Such a Page instance comes by default with a peculiar set of accessors (properties) like width, height, and artboards and methods like removeFromParent().
  • Inheritance: classes can inherit properties and methods from an existing class. The LineNode inherits from the StrokableNode, which in turn inherits from the Node class, establishing a parent/child relation.
  • Class Extension: this concept is closely related to Inheritance. A child class inherits the methods and properties from its parent and extends them with its additional ones. TextNode extends Node, adding, among others, the text property and the setPositionInParent() method.
  • Interfaces: define a contract for classes without implementing any behavior themselves. They specify a set of methods and properties a class must have but do not define how they work internally. For example, the ContainerNode interface mandates the implementation of allChildren, children, parent, and type properties. Classes implementing this interface, such as ArtboardNode and GroupNode, are responsible for providing these properties' actual implementation. By doing so, these classes conform to the structure and requirements defined by the ContainerNode interface. Interfaces can be extended, too: the ColorFill inherits the type property from the Fill interface and extends it with an additional color.

Compared to other Adobe desktop applications, the Adobe Express DOM features a remarkably clean class hierarchy, where every element has a well-defined lineage that traces back to a minimal set of root classes.

refs addon hierarchy

How to read the Reference

With this knowledge, you can use the Document APIs reference as the primary tool to study and learn about the Adobe Express DOM.

refs addon doc

There is a comprehensive list of Classes (blueprints), Interfaces (contracts), and Constants in the left navigation bar. Familiarize yourself with this content and learn how to read it.

Using the EllipseNode as an example, you can find:

  • The class name, always PascalCase.
  • A short description of what it represents and its role in the scenegraph.
  • The inheritance tree: in this case, EllipseNode is a subclass of FillableNode; following the chain, you'll find that, in turn, this inherits from StrokableNode, which eventually comes from the Node root class.
  • A list of accessors (properties), like stroke, opacity, etc.
  • A list of methods, like removeFromParent() and setPositionInParent().

refs addon ref

Some accessors are read-only, for instance, parent or rotation; some have getters and setters, like locked or fill. Properties can support a range of value kinds, from primitive values to objects, class instances, or collections. Let's break down the translation property as an example.

refs addon accessor

It's split into two parts: get (the getter, when you read the property) and set (the setter, when you write it). From the description, you see that it's a way to find out about the node's coordinates relative to its parent. The return type for the getter (wrapped with < > angle brackets) is { x: number, y: number}, i.e. an object with numeric x and y properties. You also read that this property is inherited from the FillableNode class that EllipseNode extends. The setter expects a value, which the Parameters table describes as of type Object, with the same numeric x and y properties; it returns void (i.e., nothing). Given all this, it's possible to confidently write something along these lines.

Copied to your clipboard
const ellipse = editor.createEllipse();
editor.context.insertionParent.children.append(ellipse); // attach to the scenegraph
console.log(ellipse.translation); // { x: 0, y: 0 }
ellipse.translation = { x: 100, y: 50 };

Some properties rely on interfaces to define their type. The ellipse's stroke happens to be of type Stroke, an interface whose contract mandates the implementation of five different properties: color, width, position, dashPattern, and dashOffset.

refs addon stroke

The importance of Constants

Let's use the stroke to demonstrate the use of Constants in the DOM. They are a named import—mind the lowercase spelling.

Copied to your clipboard
import { editor, colorUtils, constants } from "express-document-sdk";

They represent a safe, user-friendly way to refer to internal values (subject to change) that developers should not directly manipulate. For example, the stroke's position property is of type StrokePosition, which happens to be enumerable—a fixed set of pre-defined values.

refs addon strokeposition

Internally, the center, inside, and outside positions are represented with the integers 0, 1, and 2. You should instead use the StrokePosition constant and its available members:

  • StrokePosition.center
  • StrokePosition.inside
  • StrokePosition.outside

Using constants will ensure expected outcomes, in case Adobe Express engineers remap values in the future. This is the code needed to conclude the stroke example.

Copied to your clipboard
const ellipse = editor.createEllipse();
editor.context.insertionParent.children.append(ellipse);
ellipse.stroke = {
color: { red: 0.7, green: 0.6, blue: 1, alpha: 1 },
dashOffset: 0,
dashPattern: [4, 2],
// position: 1 ❌
position: constants.StrokePosition.inside, // ✅
width: 3,
};

The Document Stats tutorial features an add-on that goes through all elements in the scenegraph and groups them by type, providing a count of each: ComplexShape, Group, etc.

To log the type property is acceptable in this specific case, although the proper way to check against node types involves constants; the type itself is an internal string value mapped to the SceneNodeType enumerable.

Copied to your clipboard
if (someNode.type === "ab:Artboard") { /* ... */ }
if (someNode.type === constants.SceneNodeType.artboard) { /* ... */ }

Helper functions

Let's return to strokes and fills: they are expressed as plain objects; in fact, all the properties the Stroke interface requires, like dashPattern, were previously provided. The editor.makeStroke() utility will fill in the default values when you don't provide them.

Copied to your clipboard
const ellipse = editor.createEllipse();
editor.context.insertionParent.children.append(ellipse);
ellipse.stroke = editor.makeStroke({
color: { red: 0.7, green: 0.6, blue: 1, alpha: 1 },
});

Here, you get a 1-pixel, solid stroke of the provided color, the only property you've explicitly supplied. There's a similar helper function for the ColorFill interface.

Copied to your clipboard
ellipse.fill = editor.makeColorFill({
red: 0.6,
green: 0.9,
blue: 0.6,
alpha: 1,
});

If you wonder where all the help is since you've to type the entire color object manually anyway, let me remind you that the fill property is of type ColorFill: such an interface expects a color property (the one provided above) as well as a type expressed by the FillType.color constants. Without the makeColorFill() method, you must write the following.

Copied to your clipboard
ellipse.fill = {
color: { red: 0.6, green: 0.9, blue: 0.6, alpha: 1 },
type: constants.FillType.color, // 👈
};

Speaking of colors, the ColorUtils module can output a proper Color from either a partial RGB object (where the alpha property is optional, defaulting to 1) or a Hex string with or without alpha, providing the reverse function as well.

Copied to your clipboard
const gecko = colorUtils.fromHex("#9de19a"); // alpha is optional
// or
const gecko = colorUtils.fromRGB(157, 225, 154); // alpha is optional
ellipse.fill = editor.makeColorFill(gecko);
const geckoHex = colorUtils.toHex(gecko); // "#9de19aff" 👈 alpha is included

Types matter

CLI versions from "1.1.1" onwards now scaffold add-ons with type definitions for JavaScript and TypeScript projects. If you're unfamiliar with TypeScript, these additional .d.ts and tsconfig.json files offer significant benefits. The .d.ts files, or TypeScript declarations, contain type information about the APIs. They describe the shape of JavaScript objects, providing TypeScript type-checking and autocompletion capabilities (in the form of IntelliSense in VSCode). The tsconfig.json files, on the other hand, are configuration files necessary to provide type information for global types in the sandbox runtime. For instance, the sandbox' Console only exposes a subset of the available methods: the tsconfig.json file (via typeRoots) informs the IntelliSense engine only to show the methods that are actually available.

The bottom line is that .d.ts and tsconfig.json files in your JavaScript (and TypeScript) projects give code editors knowledge about the Adobe Express document sandbox APIs: it's used to provide code completion and type checking, which can help you avoid errors and write code faster.

refs addon intellisense

While IntelliSense is undoubtedly handy, type safety can prevent errors that are otherwise difficult to foresee. Let me show you a revealing example.

Several classes in the Adobe Express DOM have a children property, like GroupNode or ArtboardNode—both implementing the ContainerNode interface. Intuitively, their children are the contained elements: a group can hold bitmap images, shapes, and whatnot. ContainerNode also mandates the implementation of an allChildren property, which, apparently, serves the same purpose (we'll get to the difference in a moment).

The type for children and allChildren is defined as Iterable<Node>, where the Iterable is basically a collection that supports a for...of loop.

Copied to your clipboard
for (const node of someIterable) {
console.log(node.type);
}

In previous versions of Adobe Express, the Console would log allChildren as an actual Array of Nodes.

refs addon allchildren

If allChildren is an Array, the following code will work just fine, won't it?

Copied to your clipboard
const selectedNode = editor.context.selection[0];
// If we have a group
if (selectedNode.type === constants.SceneNodeType.group) {
// Filtering the ellipse shapes
const ellipses = selectedNode.allChildren.filter(
// 👈 danger! ❌
(node) => node.type === constants.SceneNodeType.ellipse
);
console.log(`Number of ellipses in this group: ${ellipses.length}`);
}

In fact, it used to work, but it doesn't anymore; however, it's not Adobe Express' fault. Why, then?

The allChildren property is defined in its contract to be of type Iterable. The key point here is the nature of the contract itself rather than the specific implementation. For instance, when allChildren was internally implemented as an Array, this was in line with the contract since Arrays are indeed iterable and support the for...of loop. However, Arrays offer additional functionalities beyond the requirements of the allChildren contract. Developers should focus on using features that the contract explicitly supports rather than relying on capabilities that might be available in the current implementation of the property but are outside the contract's scope. This approach ensures compatibility and reliability, irrespective of the property's underlying implementation changes. If you must use Array methods, convert the iterable with the static method Array.from() first.

Copied to your clipboard
const ellipses = Array.from(selectedNode.allChildren) // 👈
.filter((node) => node.type === constants.SceneNodeType.ellipse);

To finally unravel the allChildren purpose mystery, let's see what the documentation says about it.

[...] nodes with a more specific structure can hold children in various discrete "slots"; this allChildren list includes all such children and reflects their overall display z-order.

If you inspect a MediaContainerNode class, which is instantiated every time you place an image, it has two peculiar properties: maskShape and mediaRectangle. They hold the shape that masks the bitmap (in UI terms, the Crop—by default, a rectangle with the same image dimensions) and the ImageRectangleNode itself. They are the "structures" the documentation refers to; therefore, you'll find them in its allChildren property. Other notable examples are maskShape in Groups and artboards in Pages.

refs addon mediacontainer

Classes and Interfaces

The Reference documentation does a good job listing all the classes and interfaces that make up the Adobe Express DOM, but there are classification differences worth knowing.

Live Object classes

Concrete classes like RectangleNode and GroupNode represent the DOM's object elements with live setters. Some of these classes can be instantiated and used directly through factory methods.

Copied to your clipboard
import { editor } from "express-document-sdk";
const rect = editor.createRectangle();
const group = editor.createGroup();
group.children.append(rect);
editor.context.insertionParent.children.append(group);

Currently, only some classes of this kind have their equivalent create* method (for example, PathNode lacks it), but it's fair to expect that most of them will, eventually. Live Object classes are the ones whose instances make up the Adobe Express scenegraph: PageNode, ArtboardNode, StrokeShapeNode, LineNode, and so on. A particular case is represented by the UnknownNode, which acts as a safety placeholder when dealing with actual DOM objects that don't yet have a corresponding class.

Collection classes

In Adobe Express, most collections (children is a notable example) are expressed as Lists: groups of homogeneous elements with properties such as first, last, and length, and dedicated methods like append(), clear(), etc. They are iterable with for...of loops.

Some of these collections are Generic, like ItemList<T>, ReadOnlyItemList<T>, and RestrictedItemList<T>, where the <T> means they are designed to be type-agnostic and can work with various data types. The <T> acts as a placeholder for the type of objects that the collection will hold or manage.

An implementation of such generic classes is the ArtboardList (which subclasses RestrictedItemList<Artboard>) and PageList (subclassing RestrictedItemList<Page>), found respectively in the PageNode.artboards and ExpressRootNode.pages properties.

Abstract classes

These are classes that serve as a base for other classes. Used by Adobe to rationalize the DOM structure, they provide a common set of properties and methods, but typically cannot be instantiated themselves nor manipulated by add-on developers. Instead, they are meant to be inherited and extended by other subclasses. These subclasses can then implement the abstract class's methods. In the Reference, you can identify abstract classes such as FillableNode, Node, and BaseNode.

Static classes

Editor and ColorUtils are special Singleton classes that aren't meant to be instantiated or extended; the former acts as the entry point for APIs that read or modify the document's content, whereas the latter provides static utility methods for color manipulation. They are brought into the scope as lowercase named imports from the "express-document-sdk".

Copied to your clipboard
import { editor, colorUtils } from "express-document-sdk";

Object (POJO) Interfaces

Such interfaces define the properties of actual JavaScript objects that must be created and used, for example, to set a shape's fill and stroke (respectively, the ColorFill or Stroke interfaces) or fed to utility functions like colorUtils.fromRGB() that expect a parameter that implements the Color interface. They are the contracts that establish the shape of actual JavaScript objects that developers use in their code. POJO is an acronym that stands for "Plain Old Java Object", which in this context refers to a plain JavaScript object.

Implementable Interfaces

Implementable interfaces like IFillableNode are only meant to be implemented by classes: they define a contract of properties and methods to which a class must adhere.

In summary, the distinction between all the listed categories lies in their purpose and usage: "concrete" classes and object interfaces are used to create actual objects (either JavaScript objects, node instances, or collections), while abstract classes and implementable interfaces provide structure and behaviors that other classes can inherit or implement.

From theory to practice

Experimenting with newly acquired knowledge is one of the most effective methods to test it. Let's say you have an idea for an add-on that traces the dimensions of the selected object in the style of technical drawings.

refs addon draw

The production of such an add-on would require a number of stages, starting from the MVP (Minimum Viable Product) feature set to the UI. Here, you'll focus exclusively on the DOM prototyping; that is to say, you'll try to figure out the code building blocks by navigating the documentation reference alone—it will be an excellent exercise to get accustomed to it. Every step will be carefully described here; for simplicity, the add-on will be restricted to drawing dimensions on MediaContainer objects, assuming no crop has been applied.

To follow along, add one button to the index.html file and attach a drawDimensions() function to it in code.js.

Copied to your clipboard
<body>
<sp-theme scale="medium" color="light" theme="express">
<sp-button id="logNode">Log selected node</sp-button>
<sp-button id="drawDimensions">Draw Dimensions</sp-button>
</sp-theme>
</body>

Getting the context

The first part is to get and validate the selected node. There's a promising Context class in the reference documentation that "contains the user's current selection state, indicating the content they are focused on". Excellent! But how can you access it, though? The Editor class is the Document API entry point; it may be worth paying a visit to its page. Lo and behold, it exposes a context property, which returns a context instance: you can use it to check if there's a selection and whether it's of the right type—hasSelection and selection will do the job.

Copied to your clipboard
import { editor, constants, colorUtils } from "express-document-sdk";
runtime.exposeApi({
drawDimensions: () => {
if (
editor.context.hasSelection &&
editor.context.selection[0].type === constants.SceneNodeType.mediaContainer
) {
// ...
}
},
});

Please note the use of the SceneNodeType constant, as discussed earlier, instead of the "MediaContainer" string literal.

Getting the dimensions

Now that you've made sure only the right kind of node is handled, let's get its dimensions; since you'll need to draw a line, it may be helpful to have its coordinates in space (relative to its parent container) as well. According to the reference, the MediaContainerNode class provides a translation property that returns an object with x and y properties, which you need. There are no width and height, though; where to look? MediaContainerNode also features a mediaRectangle property, which, upon inspection, is of type ImageRectangleNode: this holds the actual media and offers both width and height properties. In future versions, Adobe Express will make available a proper bounds object for this purpose.

Copied to your clipboard
if ( /* ... */ ) {
// grabbing the selected node
const selectedNode = editor.context.selection[0];
// getting its coordinates
const nodeTranslation = selectedNode.translation;
// getting its dimensions from the mediaRectangle, destucturing and renaming them on the fly
const { width: nodeWidth, height: nodeHeight } = selectedNode.mediaRectangle;
}

Drawing a line

Time to draw the line. The Editor class includes a createLine() factory method, which returns a LineNode instance. The LineNode class, you learn from the reference, has startX, startY, endX, and endY properties: they only implement the getter, though—hence, are read-only. Scrolling through the methods, you find setEndPoints(), which expects the same parameters and is used as a setter for them.

Copied to your clipboard
// ...
const hLine = editor.createLine();
hLine.setEndPoints(nodeTranslation.x, nodeTranslation.y - 20, nodeTranslation.x + nodeWidth, nodeTranslation.y - 20);

Here, an arbitrary 20px margin is added between the horizontal line and the selected node; this could very well be a parameter chosen by the user from a UI control you can add later. The line is extended to a nodeWidth length, i.e., the mediaRectangle's. You can finally add it to the scenegraph.

Copied to your clipboard
editor.context.insertionParent.children.append(hLine);

The line is there, but it lacks the proper endpoints. The reference helps us out again: the LineNode has a startArrowHeadType and endArrowHeadType properties, whose value is an enumerable provided by the ArrowHeadType constant.3 There are several options available: let's pick triangularFilled.

Copied to your clipboard
// ...
hLine.startArrowHeadType = hLine.endArrowHeadType = constants.ArrowHeadType.triangularFilled;

As live objects, setting all the properties before or after appending the line to the scenegraph doesn't really matter; the result is the same.

Adding the text

Next up, you need to add the text. The Editor class provides a createText() method, which returns a TextNode instance. The TextNode class has a text property, which expects a string. Mind you, the setter implements parameter validation; if you were to assign a number, it would throw an error.

Copied to your clipboard
// ...
const hText = editor.createText();
hText.text = `${Math.trunc(nodeWidth).toString()}px`;
editor.context.insertionParent.children.append(hText);

The text appears on the top-left corner of the document, which is the default insertion point. setPositionInParent() would be perfect to move it to the right spot, just above the line. At the time of this writing, a TextNode doesn't feature width and height properties, which would be helpful to set an appropriate localRegistrationPoint; hence, you'll have to do with translation.

Copied to your clipboard
// ...
hText.translation = {
x: nodeTranslation.x + nodeWidth / 2,
y: nodeTranslation.y - 30,
};

An extra 10 pixels padding from the line has been added, but this could also be a parameter. Let's group the line and the text together to move them around as a single entity. The Editor class provides a createGroup() method, which returns a GroupNode instance. The GroupNode class provides a children property: you can append() both the line and the text to it.

Copied to your clipboard
// ...
const hGroup = editor.createGroup();
editor.context.insertionParent.children.append(hGroup);
hGroup.children.append(hLine, hText);

Repeating the process

From this point, creating the vertical line and the text is a matter of copying/pasting and changing a few parameters. The only new element may be text rotation: the TextNode class has a setRotationInParent() method. It expects a number in degrees (-90, in our case) and a localRotationPoint object (implementing the Point interface), which is the point to rotate around in the node's local coordinates. In this case, {x: 0, y: 0} will do.

Copied to your clipboard
const vText = editor.createText();
vText.text = `${Math.trunc(nodeHeight).toString()}px`;
editor.context.insertionParent.children.append(vText);
vText.translation = {
x: nodeTranslation.x - 30,
y: nodeTranslation.y + nodeHeight / 2,
};
vText.setRotationInParent(-90, { x: 0, y: 0 }); // 👈

Refactoring and optimizing the code

In the final add-on code, there are three buttons: one logs the selected node, one draws the dimensions as you've just seen, and the last one is a refactored version that also draws dashed lines (red and thinner) connecting the dimensions to the object's corners.

refs addon refactor

For brevity's sake, only a few relevant additions to the code will be mentioned below—please refer to the full sample for the complete picture.

The code for drawing the dimensions is moved to a separate dimensions.js file. This update includes exporting drawDimensions() and drawDimensionsRefactored() alongside the implementation of a private createDimensionLine(). This function abstracts the creation of a line and its accompanying text by accepting width, height, translation, orientation, and margin within an options object.

Copied to your clipboard
// ...
const createDimensionLine = ({ width, height, translation, orientation, margin }) => {
/* ... */
};
// ...
const drawDimensions = () => {
/* ... */
};
const drawDimensionsRefactored = () => {
// ...
createDimensionLine({
/* ... */
}); // 👈
createDimensionLine({
/* ... */
}); // 👈
// ...
};
export { drawDimensions, drawDimensionsRefactored };

A rather unorthodox warning system is implemented to alert the user when selecting an unsupported node type. Through the Communication API outlined in this tutorial, a flashWrongElement() function, defined in the iframe UI, is available to the Document Sandbox.

Copied to your clipboard
addOnUISdk.ready.then(async () => {
// ...
runtime.exposeApi({
flashWrongElement(id) {
const button = document.getElementById(id);
const oldText = button.textContent;
const oldColor = button.style.backgroundColor;
button.textContent = "WRONG NODE";
button.style.backgroundColor = "#ff0000";
setTimeout(() => {
button.textContent = oldText;
button.style.backgroundColor = oldColor;
}, 1000);
},
});
});

When the Document Sandbox code detects an unsupported node type, it reaches out to the iframe UI flashWrongElement() method (exposed via proxy), sending the button id as a parameter. As a result, the button blinks red for a second, as its CSS and textContent property are temporarily changed.

refs addon unsupported

Next Steps

As an exercise, you use the code sample found below and expand it to build a more complete Dimensions add-on; for example, you can:

  • support more node types;
  • add UI elements:
    • a slider to control the dimensions' distance from the object;
    • dropdowns to choose the line's style (solid, dashed, dotted), and the arrowhead type;
    • a checkbox to toggle the extra dashed lines;
    • a color picker to change the line's color—see the Grids add-on for an example.

Lessons Learned

Let's review the key concepts discussed in this article.

  • DOM Structure: the DOM is a hierarchical tree of nodes with parent/child relationships and well-defined inheritance.
  • OOP principles: classes, inheritance, class extension, and interfaces are crucial Object Oriented programming notions.
  • Constants: represent a safe way to refer to internal values and check against enumerable properties.
  • Reference documentation: the Adobe Express Document API reference is a key resource: it provides comprehensive information on available classes, interfaces, and their properties and methods.
  • Classes and Interfaces: you can differentiate between live object classes, collection classes, abstract classes, static classes, object interfaces, and implementable interfaces.
  • Types and IntelliSense: using .d.ts and tsconfig.json files in your projects can enhance code completion and type checking, leading to faster and error-free coding.
  • You've also learned how to prototype an add-on by navigating the Reference documentation alone.

Final Project

The code for this project can be found below, as well as in the express-add-on-samples repository.

Copied to your clipboard
<body>
<sp-theme scale="medium" color="light" theme="express">
<sp-button id="logNode">Log selected node</sp-button>
<sp-button id="drawDimensions">Draw Dimensions</sp-button>
<sp-button id="drawDimensionsRefactored">Draw Dimensions (refactored)</sp-button>
</sp-theme>
</body>


  1. Creating entirely novel features for desktop applications typically involves using SDKs and low-level languages such as C++. Adobe Express doesn't allow for this kind of customization.
  2. Typically, not every feature available in an application is by default surfaced to Scripting—it's not uncommon for it to be a smaller subset of the UI, features-wise.
  3. At the time of this writing, these properties are not yet available for selection in the Adobe Express user interface. It's one of those rare cases where scripting is mightier than the UI!
  • Privacy
  • Terms of Use
  • Do not sell or share my personal information
  • AdChoices
Copyright © 2024 Adobe. All rights reserved.