Add Markdown Link Converter Script#8650
Add Markdown Link Converter Script#8650arman-boyakhchyan wants to merge 1 commit intoDevExpress:26_1from
Conversation
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
| } | ||
|
|
||
| const indent = indentMatch[0]; | ||
| const commentRegex = new RegExp(`^${escapeForRegex(indent)}(?:<!--\s*tab:\s*index\.html\s*-->|<!--\s*HTML\s*-->)\s*$`, 'i'); |
| } | ||
|
|
||
| const indent = indentMatch[0]; | ||
| const commentRegex = new RegExp(`^${escapeForRegex(indent)}(?:<!--\s*tab:\s*index\.html\s*-->|<!--\s*HTML\s*-->)\s*$`, 'i'); |
| } | ||
|
|
||
| const indent = indentMatch[0]; | ||
| const commentRegex = new RegExp(`^${escapeForRegex(indent)}(?:<!--\s*tab:\s*index\.html\s*-->|<!--\s*HTML\s*-->)\s*$`, 'i'); |
| } | ||
|
|
||
| const indent = indentMatch[0]; | ||
| const commentRegex = new RegExp(`^${escapeForRegex(indent)}(?:<!--\s*tab:\s*index\.html\s*-->|<!--\s*HTML\s*-->)\s*$`, 'i'); |
| } | ||
|
|
||
| const indent = indentMatch[0]; | ||
| const commentRegex = new RegExp(`^${escapeForRegex(indent)}(?:<!--\s*tab:\s*index\.html\s*-->|<!--\s*HTML\s*-->)\s*$`, 'i'); |
| } | ||
|
|
||
| const indent = indentMatch[0]; | ||
| const commentRegex = new RegExp(`^${escapeForRegex(indent)}(?:<!--\s*tab:\s*index\.html\s*-->|<!--\s*HTML\s*-->)\s*$`, 'i'); |
| } | ||
|
|
||
| const indent = indentMatch[0]; | ||
| const commentRegex = new RegExp(`^${escapeForRegex(indent)}(?:<!--\s*tab:\s*index\.html\s*-->|<!--\s*HTML\s*-->)\s*$`, 'i'); |
There was a problem hiding this comment.
Pull request overview
Adds a repository-local CLI script intended to convert HTML <a target="_blank" href="...">...</a> links into Markdown links and wires it into package.json as an npm script.
Changes:
- Introduces
tools/convert-links.jsto rewrite qualifying HTML anchor tags into Markdown link syntax. - Adds a
convert-linksnpm script entry inpackage.jsonto run the converter.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 4 comments.
| File | Description |
|---|---|
| tools/convert-links.js | New converter script to rewrite certain HTML links into Markdown format. |
| package.json | Adds an npm script entry to run the new converter. |
| "rename-topics": "ts-node ./tools/updateFolder.ts", | ||
| "rename-uid": "ts-node ./tools/rename-uid.ts ./api-reference uid-map.json" | ||
| "rename-uid": "ts-node ./tools/rename-uid.ts ./api-reference uid-map.json", | ||
| "convert-links": "ts-node ./tools/convert-links.js" |
There was a problem hiding this comment.
The new npm script runs a .js file via ts-node. By default, ts-node does not execute plain JavaScript unless it is configured with allowJs, so this script can fail in a clean environment. Either run it with node or rename the file to .ts and keep ts-node consistent with the other tools/*.ts scripts.
| "convert-links": "ts-node ./tools/convert-links.js" | |
| "convert-links": "node ./tools/convert-links.js" |
| function getFilePaths(targetPath) { | ||
| const stats = fs.statSync(targetPath); | ||
|
|
||
| if (stats.isFile()) { | ||
| return [targetPath]; | ||
| } | ||
|
|
||
| if (!stats.isDirectory()) { | ||
| return []; | ||
| } | ||
|
|
||
| return fs.readdirSync(targetPath, { withFileTypes: true }).flatMap((entry) => { | ||
| const entryPath = path.join(targetPath, entry.name); | ||
|
|
||
| if (entry.isDirectory()) { | ||
| return getFilePaths(entryPath); | ||
| } | ||
|
|
||
| return entry.isFile() ? [entryPath] : []; | ||
| }); |
There was a problem hiding this comment.
This script recursively enumerates all files under the provided path and reads them as UTF-8. If the user passes the repo root, it will traverse directories like images/, applications/, and even .git/, which is very slow and can also corrupt non-text files if a match happens to occur. Consider skipping well-known non-doc directories and/or filtering to documentation extensions (for example, only .md/.mdx).
| function convertFile(filePath) { | ||
| const content = fs.readFileSync(filePath, 'utf-8'); | ||
| const updatedContent = content.replace(linkRegex, (match, href, innerHtml, offset) => { | ||
| if (isInsideTableCell(content, offset) || isInsideCodeSnippet(content, offset)) { | ||
| return match; | ||
| } | ||
|
|
||
| return `[${innerHtml}](${href})`; | ||
| }); |
There was a problem hiding this comment.
Because there is no extension filtering, the converter will also rewrite HTML sample files under applications/** that contain <a ... target="_blank">...</a> (often with nested tags like <b>), producing invalid Markdown inside .html files. Limit conversion to Markdown files (or make the allowed extensions explicit via CLI args) to avoid breaking sample applications.
| let fs = require('fs'); | ||
| let path = require('path'); | ||
|
|
||
| const specifiedPath = process.argv[2]; | ||
| const linkRegex = /<a\b(?=[^>]*\btarget=["']_blank["'])[^>]*\bhref=["']([^"']*)["'][^>]*>([\s\S]*?)<\/a>/gi; | ||
|
|
||
| function isInsideTableCell(content, offset) { | ||
| const before = content.slice(0, offset).toLowerCase(); |
There was a problem hiding this comment.
This tool uses CommonJS require and untyped function signatures, while the existing scripts in tools/ are TypeScript (tools/updateFolder.ts, tools/rename-uid.ts) and are executed via ts-node. For consistency and easier maintenance, consider converting this script to TypeScript (ES imports + basic types) or, if it stays JavaScript, run it with node and keep tools/ scripts consistently authored/executed.
No description provided.