feat: validate relative links fragments

Similar to https://github.com/DavidAnson/markdownlint/blob/main/doc/md051.md

Fixes #2

BREAKING CHANGE: Validate links fragments in relative links

package-lock.json

@@ -9,6 +9,9 @@
       "version": "0.0.0-development",
       "hasInstallScript": true,
       "license": "MIT",
+      "dependencies": {
+        "markdown-it": "13.0.1"
+      },
       "devDependencies": {
         "@commitlint/cli": "17.6.5",
         "@commitlint/config-conventional": "17.6.5",
@@ -1619,8 +1622,7 @@
     "node_modules/argparse": {
       "version": "2.0.1",
       "resolved": "https://registry.npmjs.org/argparse/-/argparse-2.0.1.tgz",
-      "integrity": "sha512-8+9WqebbFzpX9OR+Wa6O29asIogeRMzcGtAINdpMHHyAg10f05aSFVBbcEqGf/PXw1EjAZ+q2/bEBg3DvurK3Q==",
+      "integrity": "sha512-8+9WqebbFzpX9OR+Wa6O29asIogeRMzcGtAINdpMHHyAg10f05aSFVBbcEqGf/PXw1EjAZ+q2/bEBg3DvurK3Q=="
-      "dev": true
     },
     "node_modules/argv-formatter": {
       "version": "1.0.0",
@@ -2447,7 +2449,6 @@
       "version": "3.0.1",
       "resolved": "https://registry.npmjs.org/entities/-/entities-3.0.1.tgz",
       "integrity": "sha512-WiyBqoomrwMdFG1e0kqvASYfnlb0lp8M5o5Fw2OFq1hNZxxcNk8Ik0Xm7LxzBhuidnZB/UtBqVCgUz3kBOP51Q==",
-      "dev": true,
       "engines": {
         "node": ">=0.12"
       },
@@ -4372,7 +4373,6 @@
       "version": "4.0.1",
       "resolved": "https://registry.npmjs.org/linkify-it/-/linkify-it-4.0.1.tgz",
       "integrity": "sha512-C7bfi1UZmoj8+PQx22XyeXCuBlokoyWQL5pWSP+EI6nzRylyThouddufc2c1NDIcP9k5agmN9fLpA7VNJfIiqw==",
-      "dev": true,
       "dependencies": {
         "uc.micro": "^1.0.1"
       }
@@ -4872,7 +4872,6 @@
       "version": "13.0.1",
       "resolved": "https://registry.npmjs.org/markdown-it/-/markdown-it-13.0.1.tgz",
       "integrity": "sha512-lTlxriVoy2criHP0JKRhO2VDG9c2ypWCsT237eDiLqi09rmbKoUetyGHq2uOIRoRS//kfoJckS0eUzzkDR+k2Q==",
-      "dev": true,
       "dependencies": {
         "argparse": "^2.0.1",
         "entities": "~3.0.1",
@@ -5023,8 +5022,7 @@
     "node_modules/mdurl": {
       "version": "1.0.1",
       "resolved": "https://registry.npmjs.org/mdurl/-/mdurl-1.0.1.tgz",
-      "integrity": "sha512-/sKlQJCBYVY9Ers9hqzKou4H6V5UWc/M59TH2dvkt+84itfnq7uFOMLpOiOS4ujvHP4etln18fmIxA5R5fll0g==",
+      "integrity": "sha512-/sKlQJCBYVY9Ers9hqzKou4H6V5UWc/M59TH2dvkt+84itfnq7uFOMLpOiOS4ujvHP4etln18fmIxA5R5fll0g=="
-      "dev": true
     },
     "node_modules/meow": {
       "version": "8.1.2",
@@ -10471,8 +10469,7 @@
     "node_modules/uc.micro": {
       "version": "1.0.6",
       "resolved": "https://registry.npmjs.org/uc.micro/-/uc.micro-1.0.6.tgz",
-      "integrity": "sha512-8Y75pvTYkLJW2hWQHXxoqRgV7qb9B+9vFEtidML+7koHUFapnVJAZ6cKs+Qjz5Aw3aZWHMC6u0wJE3At+nSGwA==",
+      "integrity": "sha512-8Y75pvTYkLJW2hWQHXxoqRgV7qb9B+9vFEtidML+7koHUFapnVJAZ6cKs+Qjz5Aw3aZWHMC6u0wJE3At+nSGwA=="
-      "dev": true
     },
     "node_modules/uglify-js": {
       "version": "3.17.4",

package.json

@@ -42,6 +42,9 @@
     "prepublishOnly": "pinst --disable",
     "postpublish": "pinst --enable"
   },
+  "dependencies": {
+    "markdown-it": "13.0.1"
+  },
   "devDependencies": {
     "@commitlint/cli": "17.6.5",
     "@commitlint/config-conventional": "17.6.5",

src/index.js

@@ -3,42 +3,12 @@
 const { pathToFileURL } = require('node:url')
 const fs = require('node:fs')
 
-/**
+const {
- * Calls the provided function for each matching token.
+  filterTokens,
- *
+  addError,
- * @param {Object} params RuleParams instance.
+  convertHeadingToHTMLFragment,
- * @param {string} type Token type identifier.
+  getMarkdownHeadings
- * @param {Function} handler Callback function.
+} = require('./utils.js')
- * @returns {void}
- */
-const filterTokens = (params, type, handler) => {
-  for (const token of params.tokens) {
-    if (token.type === type) {
-      handler(token)
-    }
-  }
-}
-
-/**
- * Adds a generic error object via the onError callback.
- *
- * @param {Object} onError RuleOnError instance.
- * @param {number} lineNumber Line number.
- * @param {string} [detail] Error details.
- * @param {string} [context] Error context.
- * @param {number[]} [range] Column and length of error.
- * @param {Object} [fixInfo] RuleOnErrorFixInfo instance.
- * @returns {void}
- */
-const addError = (onError, lineNumber, detail, context, range, fixInfo) => {
-  onError({
-    lineNumber,
-    detail,
-    context,
-    range,
-    fixInfo
-  })
-}
 
 const customRule = {
   names: ['relative-links'],
@@ -70,12 +40,37 @@ const customRule = {
 
         if (hrefSrc != null) {
           const url = new URL(hrefSrc, pathToFileURL(params.name))
-          url.hash = ''
           const isRelative =
             url.protocol === 'file:' && !hrefSrc.startsWith('/')
-          if (isRelative && !fs.existsSync(url)) {
+          if (isRelative) {
-            const detail = `Link "${hrefSrc}" is dead`
+            const detail = `Link "${hrefSrc}" is not valid`
-            addError(onError, lineNumber, detail)
+
+            if (!fs.existsSync(url)) {
+              addError(onError, lineNumber, detail)
+              return
+            }
+
+            if (type === 'link_open' && url.hash !== '') {
+              const fileContent = fs.readFileSync(url, { encoding: 'utf8' })
+              const headings = getMarkdownHeadings(fileContent)
+
+              /** @type {Map<string, number>} */
+              const fragments = new Map()
+
+              const headingsHTMLFragments = headings.map((heading) => {
+                const fragment = convertHeadingToHTMLFragment(heading)
+                const count = fragments.get(fragment) ?? 0
+                fragments.set(fragment, count + 1)
+                if (count !== 0) {
+                  return `${fragment}-${count}`
+                }
+                return fragment
+              })
+
+              if (!headingsHTMLFragments.includes(url.hash)) {
+                addError(onError, lineNumber, detail)
+              }
+            }
           }
         }
       })

src/utils.js

@@ -0,0 +1,120 @@
+const MarkdownIt = require('markdown-it')
+
+/**
+ * Calls the provided function for each matching token.
+ *
+ * @param {Object} params RuleParams instance.
+ * @param {string} type Token type identifier.
+ * @param {Function} handler Callback function.
+ * @returns {void}
+ */
+const filterTokens = (params, type, handler) => {
+  for (const token of params.tokens) {
+    if (token.type === type) {
+      handler(token)
+    }
+  }
+}
+
+/**
+ * Adds a generic error object via the onError callback.
+ *
+ * @param {Object} onError RuleOnError instance.
+ * @param {number} lineNumber Line number.
+ * @param {string} [detail] Error details.
+ * @param {string} [context] Error context.
+ * @param {number[]} [range] Column and length of error.
+ * @param {Object} [fixInfo] RuleOnErrorFixInfo instance.
+ * @returns {void}
+ */
+const addError = (onError, lineNumber, detail, context, range, fixInfo) => {
+  onError({
+    lineNumber,
+    detail,
+    context,
+    range,
+    fixInfo
+  })
+}
+
+/**
+ * Converts a Markdown heading into an HTML fragment according to the rules
+ * used by GitHub.
+ *
+ * Taken from <https://github.com/DavidAnson/markdownlint/blob/d01180ec5a014083ee9d574b693a8d7fbc1e566d/lib/md051.js#L19>
+ *
+ * @param {string} inlineText Inline token for heading.
+ * @returns {string} Fragment string for heading.
+ */
+const convertHeadingToHTMLFragment = (inlineText) => {
+  return (
+    '#' +
+    encodeURIComponent(
+      inlineText
+        .toLowerCase()
+        // RegExp source with Ruby's \p{Word} expanded into its General Categories
+        // eslint-disable-next-line max-len
+        // https://github.com/gjtorikian/html-pipeline/blob/main/lib/html/pipeline/toc_filter.rb
+        // https://ruby-doc.org/core-3.0.2/Regexp.html
+        .replace(
+          /[^\p{Letter}\p{Mark}\p{Number}\p{Connector_Punctuation}\- ]/gu,
+          ''
+        )
+        .replace(/ /gu, '-')
+    )
+  )
+}
+
+const headingTags = new Set(['h1', 'h2', 'h3', 'h4', 'h5', 'h6'])
+const ignoredTokens = new Set(['heading_open', 'heading_close'])
+
+/**
+ * Gets the headings from a Markdown string.
+ * @param {string} content
+ * @returns {string[]}
+ */
+const getMarkdownHeadings = (content) => {
+  const markdownIt = new MarkdownIt({ html: true })
+  const tokens = markdownIt.parse(content, {})
+
+  /** @type {string[]} */
+  const headings = []
+
+  /** @type {string | null} */
+  let headingToken = null
+
+  for (const token of tokens) {
+    if (headingTags.has(token.tag)) {
+      if (token.type === 'heading_open') {
+        headingToken = token.markup
+      } else if (token.type === 'heading_close') {
+        headingToken = null
+      }
+    }
+
+    if (ignoredTokens.has(token.type)) {
+      continue
+    }
+
+    if (headingToken === null) {
+      continue
+    }
+
+    headings.push(
+      `${token.children
+        .map((token) => {
+          return token.content
+        })
+        .join('')}`
+    )
+  }
+
+  return headings
+}
+
+module.exports = {
+  filterTokens,
+  addError,
+  convertHeadingToHTMLFragment,
+  getMarkdownHeadings
+}

test/basic.test.js

@@ -5,16 +5,17 @@ const { markdownlint } = require('markdownlint').promises
 
 const relativeLinks = require('../src/index.js')
 
-test('ensure we validate correctly', async () => {
+test('ensure the rule validate correctly', async () => {
   const lintResults = await markdownlint({
     files: ['test/fixtures/Valid.md', 'test/fixtures/Invalid.md'],
     config: {
+      default: false,
       'relative-links': true
     },
     customRules: [relativeLinks]
   })
   assert.equal(lintResults['test/fixtures/Valid.md'].length, 0)
-  assert.equal(lintResults['test/fixtures/Invalid.md'].length, 2)
+  assert.equal(lintResults['test/fixtures/Invalid.md'].length, 3)
 
   assert.equal(
     lintResults['test/fixtures/Invalid.md'][0]?.ruleDescription,
@@ -22,7 +23,7 @@ test('ensure we validate correctly', async () => {
   )
   assert.equal(
     lintResults['test/fixtures/Invalid.md'][0]?.errorDetail,
-    'Link "./basic.test.js" is dead'
+    'Link "./basic.test.js" is not valid'
   )
 
   assert.equal(
@@ -31,6 +32,15 @@ test('ensure we validate correctly', async () => {
   )
   assert.equal(
     lintResults['test/fixtures/Invalid.md'][1]?.errorDetail,
-    'Link "../image.png" is dead'
+    'Link "../image.png" is not valid'
+  )
+
+  assert.equal(
+    lintResults['test/fixtures/Invalid.md'][2]?.ruleDescription,
+    'Relative links should be valid'
+  )
+  assert.equal(
+    lintResults['test/fixtures/Invalid.md'][2]?.errorDetail,
+    'Link "./Valid.md#not-existing-heading" is not valid'
   )
 })

test/fixtures/Invalid.md

@@ -3,3 +3,17 @@
 [basic.js](./basic.test.js)
 
 ![Image](../image.png)
+
+[Link fragment](./Valid.md#not-existing-heading)
+
+## Existing Heading
+
+### Repeated Heading
+
+Text
+
+### Repeated Heading
+
+Text
+
+### Repeated Heading

test/fixtures/Valid.md

@@ -11,3 +11,11 @@
 [External https link 2](https:./external.https)
 
 [External ftp link](ftp:./external.ftp)
+
+[Link fragment](./Invalid.md#existing-heading)
+
+[Link fragment Repeated 0](./Invalid.md#repeated-heading)
+
+[Link fragment Repeated 1](./Invalid.md#repeated-heading-1)
+
+[Link fragment Repeated 2](./Invalid.md#repeated-heading-2)

test/utils.test.js

@@ -0,0 +1,37 @@
+const test = require('node:test')
+const assert = require('node:assert/strict')
+
+const {
+  convertHeadingToHTMLFragment,
+  getMarkdownHeadings
+} = require('../src/utils.js')
+
+test('utils', async (t) => {
+  await t.test('convertHeadingToHTMLFragment', async () => {
+    assert.strictEqual(
+      convertHeadingToHTMLFragment('Valid Fragments'),
+      '#valid-fragments'
+    )
+    assert.strictEqual(
+      convertHeadingToHTMLFragment('Valid Heading With Underscores _'),
+      '#valid-heading-with-underscores-_'
+    )
+    assert.strictEqual(
+      convertHeadingToHTMLFragment(
+        `Valid Heading With Quotes ' And Double Quotes "`
+      ),
+      '#valid-heading-with-quotes--and-double-quotes-'
+    )
+    assert.strictEqual(
+      convertHeadingToHTMLFragment('🚀 Valid Heading With Emoji'),
+      '#-valid-heading-with-emoji'
+    )
+  })
+
+  await t.test('getMarkdownHeadings', async () => {
+    assert.deepStrictEqual(
+      getMarkdownHeadings('# Hello\n\n## World\n\n## Hello, world!\n'),
+      ['Hello', 'World', 'Hello, world!']
+    )
+  })
+})