feat(husky): check deletions and broken fragments in URLs

.github/workflows/pr-check_url-issues.yml

@@ -0,0 +1,30 @@
+name: Check URL issues
+
+on:
+  pull_request:
+    branches:
+      - main
+    paths:
+      - "files/**/*.md"
+
+jobs:
+  check_url_issues:
+    #if: github.repository == 'mdn/content'
+    runs-on: ubuntu-latest
+
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Setup Node.js environment
+        uses: actions/setup-node@v4
+        with:
+          node-version-file: ".nvmrc"
+          cache: yarn
+
+      - name: Check URL deletions and broken fragments
+        run: |
+          echo "::add-matcher::.github/workflows/url-issues-problem-matcher.json"
+          git fetch origin main
+          node scripts/log-url-issues.js --workflow

.github/workflows/url-issues-problem-matcher.json

@@ -0,0 +1,18 @@
+{
+  "problemMatcher": [
+    {
+      "owner": "log-url-issues",
+      "severity": "error",
+      "pattern": [
+        {
+          "regexp": "^(ERROR|WARN|INFO):(.+):(\\d+):(\\d+):(.+)$",
+          "severity": 1,
+          "file": 2,
+          "line": 3,
+          "column": 4,
+          "message": 5
+        }
+      ]
+    }
+  ]
+}

.lintstagedrc.json

@@ -6,5 +6,6 @@
     "prettier --write"
   ],
   "tests/**/*.*": "yarn test:front-matter-linter",
-  "*.{svg,png,jpeg,jpg,gif}": "yarn filecheck"
+  "*.{svg,png,jpeg,jpg,gif}": "yarn filecheck",
+  "*": "node scripts/log-url-issues.js"
 }

files/en-us/mdn/writing_guidelines/page_structures/syntax_sections/index.md

@@ -102,7 +102,7 @@ new Date(year, monthIndex, day, hours, minutes, seconds, milliseconds)
 
 ##### Formal syntax
 
-Formal syntax notation (using [BNF](https://en.wikipedia.org/wiki/Backus%E2%80%93Naur_form)) should not be used in the Syntax section — instead use the expanded multiple-line format [described above](multiple_linesoptional_parameters).
+Formal syntax notation (using [BNF](https://en.wikipedia.org/wiki/Backus%E2%80%93Naur_form)) should not be used in the Syntax section — instead use the expanded multiple-line format [described above](#multiple_linesoptional_parameters).
 
 While the formal notation provides a concise mechanism for describing complex syntax, it is not familiar to many developers, and can _conflict_ with valid syntax for particular programming languages. For example, "`[ ]`" indicates both an "optional parameter" and a JavaScript {{jsxref("Array")}}. You can see this in the formal syntax for {{jsxref("Array.prototype.slice()")}} below:
 

files/en-us/mozilla/add-ons/webextensions/api/runtime/setuninstallurl/index.md

@@ -7,7 +7,7 @@ browser-compat: webextensions.api.runtime.setUninstallURL
 
 {{AddonSidebar()}}
 
-Sets the URL to be visited when the extension is uninstalled. This can be used to clean up server-side data, do analytics, or implement surveys. The URL can be up to 1023 characters. This limit used to be 255, see [Browser compatibility](browser_compatibility) for more details.
+Sets the URL to be visited when the extension is uninstalled. This can be used to clean up server-side data, do analytics, or implement surveys. The URL can be up to 1023 characters. This limit used to be 255, see [Browser compatibility](#browser_compatibility) for more details.
 
 This is an asynchronous function that returns a [`Promise`](/en-US/docs/Web/JavaScript/Reference/Global_Objects/Promise).
 

scripts/log-url-issues.js

@@ -0,0 +1,133 @@
+/*
+ * The script logs locations of affected URLs due to following reasons:
+ * - file deletion
+ * - Markdown header updates
+ */
+
+import fs from "node:fs/promises";
+import path from "node:path";
+import {
+  execGit,
+  getRootDir,
+  walkSync,
+  isImagePath,
+  getLocations,
+  IMG_RX,
+  stringToFragment,
+} from "./utils.js";
+
+const rootDir = getRootDir();
+const argLength = process.argv.length;
+const deletedSlugs = [];
+const fragmentDetails = [];
+
+function getDeletedSlugs() {
+  // git status --short --porcelain
+  let result = execGit(["status", "--short", "--porcelain"], { cwd: "." });
+
+  if (result.trim()) {
+    deletedSlugs.push(
+      ...result
+        .split("\n")
+        .filter(
+          (line) =>
+            /^\s*D\s+/gi.test(line) &&
+            line.includes("files/en-us") &&
+            (IMG_RX.test(line) || line.includes("index.md")),
+        )
+        .map((line) => line.replaceAll(/^\s*|files\/en-us\/|\/index.md/gm, ""))
+        .map((line) => line.split(/\s+/)[1]),
+    );
+  }
+  console.log("deletedSlugs", deletedSlugs);
+}
+
+function getFragmentDetails(fromStaging = true) {
+  let result = "";
+
+  if (fromStaging) {
+    // get staged and unstaged changes
+    result = execGit(["diff", "HEAD"], { cwd: "." });
+  } else {
+    // get diff between branch base and HEAD
+    result = execGit(["diff", "origin/main...HEAD"], { cwd: "." });
+  }
+
+  if (result.trim()) {
+    const segments = [
+      ...result.split("diff --git a/").filter((segment) => segment !== ""),
+    ];
+    for (const segment of segments) {
+      const path = segment
+        .substring(0, segment.indexOf(" "))
+        .replaceAll(/files\/en-us\/|\/index.md/gm, "");
+
+      const headerRx = /^-#+ .*$/gm;
+      const fragments = [...segment.matchAll(headerRx)]
+        .map((match) => match[0].toLowerCase())
+        .map((header) => header.replace(/-#+ /g, ""))
+        .map((header) => stringToFragment(header));
+
+      for (const fragment of fragments) {
+        fragmentDetails.push(`${path}#${fragment}`);
+      }
+    }
+  }
+  console.log("fragmentDetails", fragmentDetails);
+}
+
+if (process.argv[2] !== "--workflow") {
+  getDeletedSlugs();
+  getFragmentDetails();
+} else {
+  getFragmentDetails(false);
+}
+
+if (deletedSlugs.length < 1 && fragmentDetails.length < 1) {
+  console.log("Nothing to check. 🎉");
+  process.exit(0);
+}
+
+for await (const filePath of walkSync(getRootDir())) {
+  if (filePath.endsWith("index.md")) {
+    try {
+      const content = await fs.readFile(filePath, "utf-8");
+      const relativePath = filePath.substring(filePath.indexOf("files/en-us"));
+
+      // check deleted links
+      for (const slug of deletedSlugs) {
+        const locations = getLocations(
+          content,
+          new RegExp(`/${slug}[)># \"']`, "mig"),
+        );
+        if (locations.length) {
+          for (const location of locations) {
+            console.error(
+              `ERROR:${relativePath}:${location.line}:${location.column}:Slug '${slug}' has been deleted`,
+            );
+          }
+        }
+      }
+
+      // check broken URL fragment
+      for (const fragment of fragmentDetails) {
+        const locations = getLocations(content, fragment);
+        // check fragments in the same file
+        const urlParts = fragment.split("#");
+        if (filePath.includes(urlParts[0])) {
+          locations.push(...getLocations(content, urlParts[1]));
+        }
+        if (locations.length) {
+          for (const location of locations) {
+            console.error(
+              `ERROR:${relativePath}:${location.line}:${location.column}:URL fragment '${fragment}' is broken`,
+            );
+          }
+        }
+      }
+    } catch (e) {
+      console.error(`Error processing ${filePath}: ${e.message}`);
+      throw e;
+    }
+  }
+}

scripts/utils.js

@@ -2,7 +2,7 @@ import fs from "node:fs/promises";
 import path from "node:path";
 import childProcess from "node:child_process";
 
-const IMG_RX = /(\.png|\.jpg|\.svg|\.gif)$/gim;
+export const IMG_RX = /(\.png|\.jpg|\.svg|\.gif)$/gim;
 
 export async function* walkSync(dir) {
   const files = await fs.readdir(dir, { withFileTypes: true });
@@ -49,3 +49,41 @@ export function getRootDir() {
 export function isImagePath(path) {
   return IMG_RX.test(path);
 }
+
+/*
+ * Returns locations (line and column numbers) of 'searchValue' in the given 'content'.
+ */
+export function getLocations(content, searchValue) {
+  const lineLengths = content.split("\n").map((line) => line.length);
+  const searchRx =
+    searchValue instanceof RegExp
+      ? searchValue
+      : new RegExp(searchValue, "mig");
+  const matches = [...content.matchAll(searchRx)].map((match) => match.index);
+  const positions = [];
+
+  let currentPosition = 0;
+  lineLengths.forEach((lineLength, index) => {
+    lineLength += 1; // add '\n'
+    for (const match of matches) {
+      if (currentPosition < match && currentPosition + lineLength > match) {
+        positions.push({
+          line: index + 1,
+          column: match - currentPosition + 1,
+        });
+      }
+    }
+    currentPosition += lineLength;
+  });
+  return positions;
+}
+
+/*
+ * Convert Markdown header into URL slug.
+ */
+export function stringToFragment(text) {
+  return text
+    .trim()
+    .replace(/["#$%&+,/:;=?@[\]^`{|}~')(\\]/g, "")
+    .replace(/\s+/g, "_");
+}