remark-abbr-start

example.js

import fs from "node:fs/promises";
import { unified } from "unified";
import remarkParse from "remark-parse";
import remarkRehype from "remark-rehype";
import rehypeStringify from "rehype-stringify";
import myCustomRemarkAbbrPlugin from "./index.js";

const file = await unified()
  .use(remarkParse)
  .use(myCustomRemarkAbbrPlugin)
  .use(remarkRehype)
  .use(rehypeStringify)
  .process(await fs.readFile("example.md"));

console.log(file);

example.md

This is an abbreviation: HTML.

*[HTML]: HyperText Markup Language

index.js

import { visit } from "unist-util-visit";
import { findAndReplace } from "mdast-util-find-and-replace";
import escapeStringRegexp from "escape-string-regexp";

// This plugin does two things:
// a) First, it tells micromark and friends how to parse abbreviation definitions.
// b) Much later, it replaces abbreviations use, in the AST, with custom nodes.
/** @type {import('unified').Plugin<[], import('mdast').Root>} */
export default function myCustomRemarkAbbr() {
  const data = this.data();

  // Register things that deal with parsing.
  // Micromark extensions define how to parse syntax.
  // This is a micromark extension.
  // It would look like <https://github.com/micromark/micromark/blob/main/packages/micromark-core-commonmark/dev/lib/definition.js>.
  add("micromarkExtensions", myCustomMicromarkAbbrExtension);
  // This is a `mdast-util-from-markdown` extension.
  // `mdast-util-from-markdown` extensions define how to handle the resulting tokens/events from micromark extensions and build nodes from them.
  // It would look like <https://github.com/syntax-tree/mdast-util-math/blob/4670ae469c47b53d9661e7a116a869493f4b1800/index.js#L24>.
  // But instead of math, it would handle the tokens that are added by your custom micromark extension.
  // The rest of the codes expects the resulting nodes from your extension to look as follows:
  // `{type: 'my-abbr-definition-node-type', id: string, title: string}`.
  add("fromMarkdownExtensions", myCustomMdastUtilFromMarkdownAbbrExtension);
  // To do: if you need to compile back to markdown again, you need to define how to turn nodes back into a string of markdown.
  // If you need this, it would look like <https://github.com/syntax-tree/mdast-util-math/blob/4670ae469c47b53d9661e7a116a869493f4b1800/index.js#L125>.
  // But instead of math, it would handle the nodes that are created by your `mdast-util-from-markdown` extension.
  // add("toMarkdownExtensions", myCustomMicromarkAbbrToMarkdownExtension);

  // This a) finds all defined abbr definitions, b) replaces all abbr uses with what is defined.
  return function transform(tree) {
    /** @type {Record<string, string>} */
    const definitionToTitle = Object.create(null);

    visit(tree, "my-abbr-definition-node-type", (node) => {
      // To do: prefer the first of duplicates?
      definitionToTitle[node.id] = node.title;
    });

    // Create a regex to look for ID use.
    let search = new RegExp(
      Object.keys(definitionToTitle).map(escapeStringRegexp).join("|"),
      "g"
    );

    // Replace ID use with custom abbr use nodes.
    findAndReplace(tree, search, (/** @type {string} */ $0) => {
      let title = definitionToTitle;

      return {
        type: "my-abbr-use-node-type",
        value: $0,
        // We are in markdown, but we want to define how our custom nodes get turned into HTML.
        // For more on this, see: <https://github.com/remarkjs/remark-rehype#what-is-this>.
        // For more on these fields, see: <https://github.com/syntax-tree/mdast-util-to-hast#fields-on-nodes>.
        data: {
          hTagName: "abbr",
          hProperties: {
            title,
          },
        },
      };
    });
  };

  /**
   * @param {string} field
   * @param {unknown} value
   */
  function add(field, value) {
    const list = /** @type {unknown[]} */ (
      // Other extensions
      data[field] ? data[field] : (data[field] = [])
    );

    list.push(value);
  }
}

package.json

{
  "name": "example",
  "version": "1.0.0",
  "description": "",
  "main": "index.js",
  "scripts": {
    "test": "echo \"Error: no test specified\" && exit 1"
  },
  "keywords": [],
  "author": "",
  "license": "ISC",
  "devDependencies": {
    "rehype-stringify": "^9.0.3",
    "remark-parse": "^10.0.1",
    "remark-rehype": "^10.1.0",
    "unified": "^10.1.2"
  },
  "dependencies": {
    "@types/mdast": "^3.0.10",
    "escape-string-regexp": "^5.0.0",
    "mdast-util-find-and-replace": "^2.2.1",
    "unist-util-visit": "^4.1.0"
  }
}

Hi, @wooorm,

How can I find the myCustomMicromarkAbbrExtension and myCustomMdastUtilFromMarkdownAbbrExtension

I couldn't find any remark abbreviation plugin which works, so far. I need help.

You are commenting on a pseudocode example that explains in the comments how to do that?

Re extending markdown? You most likely shouldn't: https://github.com/micromark/micromark#extending-markdown

I’m not sure how you found this gist, it was I think privately helping someone with a ton of legacy markdown.
I recommend going with directives if you can!

Thank you very much for quick reply. I will check it out.