Lexical

Adds the lexical WYSIWYG editor as a field type.

strapi-plugin-lexical

Integrates the Lexical WYSIWYG editor as a custom field in Strapi. Basically a port of Lexical playground into strapi environment with some nice extras.

screenshot-lexical

Alpha Software
This plugin is in active development. Contributions in the form of bug reports, feature suggestions, and pull requests are highly encouraged!

Table of contents

strapi-plugin-lexical

Installation

Install the plugin:
```
npm install strapi-plugin-lexical
```

Enable the plugin:

1// ./config/plugins.js
2{
3 lexical: {
4   enabled: true,
5 },
6
7};

Include the required CSS and Prism.js in your Strapi admin:

1// ./src/admin/app.js
2import "strapi-plugin-lexical/dist/style.css";
3import "prismjs";

Add Vite support for Prism.js:

Install the plugin:

npm install --save-dev vite-plugin-prismjs

Update your Vite configuration:

1// ./src/admin/vite.config.js
2import { mergeConfig } from "vite";
3import prismjs from "vite-plugin-prismjs";
4
5export default (config) =>
6  mergeConfig(config, {
7    plugins: [
8      prismjs({
9        languages: "all", // Load all languages or customize as needed
10      }),
11    ],
12  });

Note: Prism.js is required even if you don't plan to support code blocks. If you find a workaround to avoid this, please share it with us via a pull request or issue. We happily skip this installation step if we can!

Usage

A new Lexical custom field type will be available in the Strapi content-type builder.
Currently, it supports features migrated from the Lexical playground.
For rendering content on your frontend, consider using libraries like payload-lexical-react-renderer or similar tools.

Handling Media and Internal Links

This plugin ensures reliable rendering of images and internal links by maintaining relationships between rich text content and referenced entities. By using a regular media field and automatically generated or your own link components we can ensure that all referenced media and internal links are readily available for your frontend, always reflecting the latest data.

Important: This is readme section is WIP. We will update this soon and give better examples on how to query and render images and links

Opt-in Mechanism

To enable this feature, you have to create secondary fields:

With the suffix Media (e.G. YourFieldNameMedia): This field must be a multiple media field with editing disabled.
With the suffix Links (e.G. YourFieldNameLinks): This must be a component, either use our pregenerated Links component or build your own. Important: It should only contain relation fields and the field name must match the linked collection name.

Integration in Lexical documents

Media is stored as a custom Lexical nodes, while store relations to strapi content with a custom URL format for links. These are automatically parsed and extracted into the fields you created above.

Media References

Images are stored as strapi-image node in Lexical.
Other file types are planned but not yet supported.
The structure is rather simple, as you can see:

strapi-image Lexical Node Data Structure:

1{ "documentId": "id_of_media_asset" }

Internal Links

Internal links are stored using the regular link node in Lexical.
The URL follows the format:
strapi://collectionName/documentId
This ensures that even if a page’s slug changes, links remain valid.

Rendering Strapi media and links

There are two options:

Fetch while rendering

@todo

Benefit: Less code, multiple API calls while rendering

adjust the rendering functions of each lexical node
actually.. can it even be asnyc? double check... this is the not recommended way anyways...

Example: Fetching the Latest API Data for a link

1const [collectionName, documentId] = linkNode.url.replace("strapi://", "").split("/");
2const articles = client.collection(collectionName);
3const singleArticle = await articles.findOne(documentId);
4// your link generation logic here ...
5return `/${singleArticle.locale}/blog/${singleArticle.slug}`

Prefetch and inject for rendering

@todo

To render media and links, we have to query the data from our media field and fields within the link component, then inject the data into our rendering process.

Benefit: only one API call, more control

fetch fields
iterate through the lexical document
inject the document from the strapi api response into the lexical node for later rendering
the data is now available when rendering the lexical node in your renderer

Example Renderer with NextJS:

1// LexicalRenderer.tsx
2import Image from "next/image";
3import Link from "next/link";
4
5import type { Media_Plain } from "@strapi/common/schemas-to-ts/Media";
6import { Links_Plain } from "@strapi/components/links/interfaces/Links";
7
8import React from "react";
9
10import clsx from "clsx/lite";
11
12import { createPath } from "@/utils/paths";
13
14import type {
15  AbstractNode,
16  ElementRenderer,
17  ElementRenderers,
18  Node,
19  PayloadLexicalReactRendererContent,
20} from "lexical-renderer-atelier-disko";
21import {
22  defaultElementRenderers,
23  PayloadLexicalReactRenderer,
24} from "lexical-renderer-atelier-disko";
25
26type StrapiImageNode = {
27  documentId: string;
28  entity: Media_Plain;
29} & AbstractNode<"strapi-image">;
30
31type NodeAll = Node | StrapiImageNode;
32
33const elementRenderers: ElementRenderers & {
34  "strapi-image": ElementRenderer<StrapiImageNode>;
35} = {
36  ...defaultElementRenderers,
37  // Define your custom lexical nodes here
38  link: (element, children, parent, className) => (
39    <Link
40      href={element.url}
41      className={className}
42      target={element.newTab ? "_blank" : "_self"}
43    >
44      {children}
45    </Link>
46  ),
47  "strapi-image": (element, children, parent, className) => (
48    <Image
49      className={clsx("mx-auto", className)}
50      src={`${process.env.NEXT_PUBLIC_IMAGE_BASE_URL}${element.entity.url}`}
51      alt={element.entity.alternativeText}
52      width={Math.floor(element.entity.width / 2)}
53      height={Math.floor(element.entity.height / 2)}
54      sizes={`(max-width: 768px) 100vw, ${Math.floor(
55        (element.entity.width / 2) * 1.25
56      )}px`}
57      loading="lazy"
58    />
59  ),
60};
61
62export function LexicalRenderer({
63  children,
64  classNames,
65  media,
66  links,
67}: {
68  children: PayloadLexicalReactRendererContent;
69  classNames?: { [key: string]: string };
70  media?: Media_Plain[];
71  links?: Links_Plain;
72}) {
73  // Inject media and links into our lexical document
74  const injectedDocument = React.useMemo(() => {
75    if (!children) {
76      return null;
77    }
78
79    if (media || links) {
80      const injectStrapiEntities = (nodes: NodeAll[]) => {
81        for (const node of nodes) {
82          // Media (Images only for now)
83          if (node.type === "strapi-image" && media?.length) {
84            const foundMedia = media.find(
85              // @ts-expect-error documentId is there. the ts schema plugin is just outdated :(
86              ({ documentId }) => documentId === node.documentId
87            );
88            if (foundMedia) {
89              node.entity = foundMedia;
90            }
91          }
92
93          // Links
94          if (
95            node.type === "link" &&
96            links &&
97            node.url.indexOf("strapi://") === 0
98          ) {
99            // Extract info from strapi link
100            const [collectionName, linkDocumentId] = (node.url as string)
101              .replace("strapi://", "")
102              .split("/") as [keyof Links_Plain, string];
103            if (links[collectionName]) {
104              // Find linked document
105              const foundCollectionDocument = links[collectionName].find(
106                ({ documentId }) => documentId === linkDocumentId
107              );
108              if (foundCollectionDocument) {
109                // Generate page link with helper function
110                node.url = createPath(
111                  collectionName,
112                  foundCollectionDocument.locale,
113                  foundCollectionDocument.slug
114                );
115              }
116            }
117          }
118          if (node.type !== "strapi-image" && node.children) {
119            injectStrapiEntities(node.children);
120          }
121        }
122      };
123
124      injectStrapiEntities(children.root.children);
125    }
126
127    return children;
128  }, [children, media, links]);
129
130  if (!children || !injectedDocument) {
131    return null;
132  }
133
134  return (
135    <PayloadLexicalReactRenderer
136      content={injectedDocument}
137      classNames={classNames}
138      elementRenderers={elementRenderers}
139    />
140  );
141}
142
143export const lexicalToPlaintext = (json: { root: Node }) => {
144  const traverse = (node: Node): string => {
145    if (node.type === "text" && node.text) return node.text;
146    if (node.children) return node.children.map(traverse).join(" ");
147    return "";
148  };
149  return traverse(json.root);
150};

Roadmap

v0 - Alpha

Implement basic functionality.
Port features from the Lexical playground as the initial foundation.
Integrate Strapi Media Assets and enable linking to Strapi Collection Entries
Create field presets:
- Simple, Complex, and Full (selectable during field setup).
Gather community feedback.
Look for a potential co-maintainer.

v1 - Stable

Introduce plugin-based architecture:
- Allow users to extend functionality with their own plugins.
Enable configuration of presets via plugin settings.
Open to community ideas! Submit an issue.

Contributing

We welcome contributions! Here’s how you can help:

Report bugs or suggest features via the issue tracker.
Submit pull requests to improve functionality or documentation.
Share your feedback and ideas to shape the plugin’s future.

Resources

🛠️ Sponsored by hashbite.net | support & custom development available

We welcome everyone to post issues, fork the project, and contribute via pull requests. Together we can make this a better tool for all of us!

If the contribution process feels too slow or complex for your needs, hashbite.net can quickly implement features, fix bugs, or develop custom variations of this plugin on a paid basis. Just reach out through their website for direct support.