Reimplement MD026/no-trailing-punctuation using micromark tokens, ignore trailing GitHub emoji codes, improve tests (fixes #457).

demo/markdownlint-browser.js

@@ -81,6 +81,16 @@ module.exports.blockquotePrefixRe = blockquotePrefixRe;
 var linkReferenceDefinitionRe = /^ {0,3}\[([^\]]*[^\\])\]:/;
 module.exports.linkReferenceDefinitionRe = linkReferenceDefinitionRe;
 
+// Regular expression for identifying an HTML entity at the end of a line
+module.exports.endOfLineHtmlEntityRe =
+// eslint-disable-next-line max-len
+/&(?:#\d+|#[xX][\da-fA-F]+|[a-zA-Z]{2,31}|blk\d{2}|emsp1[34]|frac\d{2}|sup\d|there4);$/;
+
+// Regular expression for identifying a GitHub emoji code at the end of a line
+module.exports.endOfLineGemojiCodeRe =
+// eslint-disable-next-line max-len
+/:(?:[abmovx]|[-+]1|100|1234|(?:1st|2nd|3rd)_place_medal|8ball|clock\d{1,4}|e-mail|non-potable_water|o2|t-rex|u5272|u5408|u55b6|u6307|u6708|u6709|u6e80|u7121|u7533|u7981|u7a7a|[a-z]{2,15}2?|[a-z]{1,14}(?:_[a-z\d]{1,16})+):$/;
+
 // All punctuation characters (normal and full-width)
 var allPunctuation = ".,;:!?。，；：！？";
 module.exports.allPunctuation = allPunctuation;
@@ -4434,12 +4444,17 @@ module.exports = {
 
 
 
+function _createForOfIteratorHelper(o, allowArrayLike) { var it = typeof Symbol !== "undefined" && o[Symbol.iterator] || o["@@iterator"]; if (!it) { if (Array.isArray(o) || (it = _unsupportedIterableToArray(o)) || allowArrayLike && o && typeof o.length === "number") { if (it) o = it; var i = 0; var F = function F() {}; return { s: F, n: function n() { if (i >= o.length) return { done: true }; return { done: false, value: o[i++] }; }, e: function e(_e) { throw _e; }, f: F }; } throw new TypeError("Invalid attempt to iterate non-iterable instance.\nIn order to be iterable, non-array objects must have a [Symbol.iterator]() method."); } var normalCompletion = true, didErr = false, err; return { s: function s() { it = it.call(o); }, n: function n() { var step = it.next(); normalCompletion = step.done; return step; }, e: function e(_e2) { didErr = true; err = _e2; }, f: function f() { try { if (!normalCompletion && it["return"] != null) it["return"](); } finally { if (didErr) throw err; } } }; }
+function _unsupportedIterableToArray(o, minLen) { if (!o) return; if (typeof o === "string") return _arrayLikeToArray(o, minLen); var n = Object.prototype.toString.call(o).slice(8, -1); if (n === "Object" && o.constructor) n = o.constructor.name; if (n === "Map" || n === "Set") return Array.from(o); if (n === "Arguments" || /^(?:Ui|I)nt(?:8|16|32)(?:Clamped)?Array$/.test(n)) return _arrayLikeToArray(o, minLen); }
+function _arrayLikeToArray(arr, len) { if (len == null || len > arr.length) len = arr.length; for (var i = 0, arr2 = new Array(len); i < len; i++) arr2[i] = arr[i]; return arr2; }
 var _require = __webpack_require__(/*! ../helpers */ "../helpers/helpers.js"),
   addError = _require.addError,
   allPunctuationNoQuestion = _require.allPunctuationNoQuestion,
-  escapeForRegExp = _require.escapeForRegExp,
+  endOfLineGemojiCodeRe = _require.endOfLineGemojiCodeRe,
-  forEachHeading = _require.forEachHeading;
+  endOfLineHtmlEntityRe = _require.endOfLineHtmlEntityRe,
-var endOfLineHtmlEntityRe = /&#?[\da-zA-Z]+;$/;
+  escapeForRegExp = _require.escapeForRegExp;
+var _require2 = __webpack_require__(/*! ../helpers/micromark.cjs */ "../helpers/micromark.cjs"),
+  filterByTypes = _require2.filterByTypes;
 module.exports = {
   "names": ["MD026", "no-trailing-punctuation"],
   "description": "Trailing punctuation in heading",
@@ -4448,21 +4463,31 @@ module.exports = {
     var punctuation = params.config.punctuation;
     punctuation = String(punctuation === undefined ? allPunctuationNoQuestion : punctuation);
     var trailingPunctuationRe = new RegExp("\\s*[" + escapeForRegExp(punctuation) + "]+$");
-    forEachHeading(params, function (heading) {
+    var headings = filterByTypes(params.parsers.micromark.tokens, ["atxHeadingText", "setextHeadingText"]);
-      var line = heading.line,
+    var _iterator = _createForOfIteratorHelper(headings),
-        lineNumber = heading.lineNumber;
+      _step;
-      var trimmedLine = line.replace(/([^\s#])[\s#]+$/, "$1");
+    try {
-      var match = trailingPunctuationRe.exec(trimmedLine);
+      for (_iterator.s(); !(_step = _iterator.n()).done;) {
-      if (match && !endOfLineHtmlEntityRe.test(trimmedLine)) {
+        var heading = _step.value;
+        var endLine = heading.endLine,
+          startColumn = heading.startColumn,
+          text = heading.text;
+        var match = trailingPunctuationRe.exec(text);
+        if (match && !endOfLineHtmlEntityRe.test(text) && !endOfLineGemojiCodeRe.test(text)) {
           var fullMatch = match[0];
-        var column = match.index + 1;
+          var column = startColumn + match.index;
           var length = fullMatch.length;
-        addError(onError, lineNumber, "Punctuation: '".concat(fullMatch, "'"), null, [column, length], {
+          addError(onError, endLine, "Punctuation: '".concat(fullMatch, "'"), undefined, [column, length], {
             "editColumn": column,
             "deleteCount": length
           });
         }
-    });
+      }
+    } catch (err) {
+      _iterator.e(err);
+    } finally {
+      _iterator.f();
+    }
   }
 };
 

helpers/helpers.js

@@ -39,6 +39,16 @@ module.exports.blockquotePrefixRe = blockquotePrefixRe;
 const linkReferenceDefinitionRe = /^ {0,3}\[([^\]]*[^\\])\]:/;
 module.exports.linkReferenceDefinitionRe = linkReferenceDefinitionRe;
 
+// Regular expression for identifying an HTML entity at the end of a line
+module.exports.endOfLineHtmlEntityRe =
+  // eslint-disable-next-line max-len
+  /&(?:#\d+|#[xX][\da-fA-F]+|[a-zA-Z]{2,31}|blk\d{2}|emsp1[34]|frac\d{2}|sup\d|there4);$/;
+
+// Regular expression for identifying a GitHub emoji code at the end of a line
+module.exports.endOfLineGemojiCodeRe =
+  // eslint-disable-next-line max-len
+  /:(?:[abmovx]|[-+]1|100|1234|(?:1st|2nd|3rd)_place_medal|8ball|clock\d{1,4}|e-mail|non-potable_water|o2|t-rex|u5272|u5408|u55b6|u6307|u6708|u6709|u6e80|u7121|u7533|u7981|u7a7a|[a-z]{2,15}2?|[a-z]{1,14}(?:_[a-z\d]{1,16})+):$/;
+
 // All punctuation characters (normal and full-width)
 const allPunctuation = ".,;:!?。，；：！？";
 module.exports.allPunctuation = allPunctuation;

lib/md026.js

@@ -2,10 +2,10 @@
 
 "use strict";
 
-const { addError, allPunctuationNoQuestion, escapeForRegExp, forEachHeading } =
+const { addError, allPunctuationNoQuestion, endOfLineGemojiCodeRe,
-  require("../helpers");
+  endOfLineHtmlEntityRe, escapeForRegExp } = require("../helpers");
+const { filterByTypes } = require("../helpers/micromark.cjs");
 
-const endOfLineHtmlEntityRe = /&#?[\da-zA-Z]+;$/;
 
 module.exports = {
   "names": [ "MD026", "no-trailing-punctuation" ],
@@ -18,19 +18,26 @@ module.exports = {
     );
     const trailingPunctuationRe =
       new RegExp("\\s*[" + escapeForRegExp(punctuation) + "]+$");
-    forEachHeading(params, (heading) => {
+    const headings = filterByTypes(
-      const { line, lineNumber } = heading;
+      params.parsers.micromark.tokens,
-      const trimmedLine = line.replace(/([^\s#])[\s#]+$/, "$1");
+      [ "atxHeadingText", "setextHeadingText" ]
-      const match = trailingPunctuationRe.exec(trimmedLine);
+    );
-      if (match && !endOfLineHtmlEntityRe.test(trimmedLine)) {
+    for (const heading of headings) {
+      const { endLine, startColumn, text } = heading;
+      const match = trailingPunctuationRe.exec(text);
+      if (
+        match &&
+        !endOfLineHtmlEntityRe.test(text) &&
+        !endOfLineGemojiCodeRe.test(text)
+      ) {
         const fullMatch = match[0];
-        const column = match.index + 1;
+        const column = startColumn + match.index;
         const length = fullMatch.length;
         addError(
           onError,
-          lineNumber,
+          endLine,
           `Punctuation: '${fullMatch}'`,
-          null,
+          undefined,
           [ column, length ],
           {
             "editColumn": column,
@@ -38,6 +45,6 @@ module.exports = {
           }
         );
       }
-    });
+    }
   }
 };

package.json

@@ -73,12 +73,14 @@
     "ava": "5.3.1",
     "babel-loader": "9.1.2",
     "c8": "8.0.0",
+    "character-entities": "2.0.2",
     "eslint": "8.43.0",
     "eslint-plugin-es": "4.1.0",
     "eslint-plugin-jsdoc": "46.2.6",
     "eslint-plugin-n": "16.0.0",
     "eslint-plugin-regexp": "1.15.0",
     "eslint-plugin-unicorn": "47.0.0",
+    "gemoji": "8.1.0",
     "globby": "13.2.0",
     "js-yaml": "4.1.0",
     "markdown-it-for-inline": "0.1.1",

test/headings-with-emoji.md

@@ -0,0 +1,25 @@
+# headings-with-emoji
+
+## Known Issues :bug:
+
+## Love :heartpulse:
+
+## :tada:
+
+## :checkered_flag:
+
+## :clock930:
+
+## :t-rex:
+
+## Boba:bubble_tea:
+
+<!-- markdownlint-disable heading-style -->
+
+## Fix the :bug: ##
+
+Another :heartpulse:
+--------------------
+
+:eyes:
+------

test/headings-with-html-entities.md

@@ -12,6 +12,8 @@
 
 ## Copyright 2004 ©
 
+## Copyright 2004 ©
+
 ## Copyright 2005 ©
 
 ## Copyright 2006 © ##

test/markdownlint-test-helpers.js

@@ -1283,3 +1283,21 @@ Empty bracket pairs: [text3[]][]
   };
   return markdownlint(options).then(() => null);
 });
+
+test("endOfLineHtmlEntityRe", async(t) => {
+  const { characterEntities } = await import("character-entities");
+  const entities = Object.keys(characterEntities);
+  t.plan(entities.length);
+  for (const entity of entities) {
+    t.true(helpers.endOfLineHtmlEntityRe.test(`-&${entity};`), entity);
+  }
+});
+
+test("endOfLineGemojiCodeRe", async(t) => {
+  const { gemoji } = await import("gemoji");
+  const emojis = gemoji.flatMap((i) => i.names);
+  t.plan(emojis.length);
+  for (const emoji of emojis) {
+    t.true(helpers.endOfLineGemojiCodeRe.test(`-:${emoji}:`), emoji);
+  }
+});

test/snapshots/markdownlint-test-scenarios.js.md

@@ -14772,6 +14772,40 @@ Generated by [AVA](https://avajs.dev).
       `,
     }
 
+## headings-with-emoji.md
+
+> Snapshot 1
+
+    {
+      errors: [],
+      fixed: `# headings-with-emoji␊
+      ␊
+      ## Known Issues :bug:␊
+      ␊
+      ## Love :heartpulse:␊
+      ␊
+      ## :tada:␊
+      ␊
+      ## :checkered_flag:␊
+      ␊
+      ## :clock930:␊
+      ␊
+      ## :t-rex:␊
+      ␊
+      ## Boba:bubble_tea:␊
+      ␊
+      <!-- markdownlint-disable heading-style -->␊
+      ␊
+      ## Fix the :bug: ##␊
+      ␊
+      Another :heartpulse:␊
+      --------------------␊
+      ␊
+      :eyes:␊
+      ------␊
+      `,
+    }
+
 ## headings-with-html-entities.md
 
 > Snapshot 1
@@ -14789,7 +14823,7 @@ Generated by [AVA](https://avajs.dev).
             deleteCount: 1,
             editColumn: 31,
           },
-          lineNumber: 22,
+          lineNumber: 24,
           ruleDescription: 'Trailing punctuation in heading',
           ruleInformation: 'https://github.com/DavidAnson/markdownlint/blob/v0.0.0/doc/md026.md',
           ruleNames: [
@@ -14808,7 +14842,7 @@ Generated by [AVA](https://avajs.dev).
             deleteCount: 1,
             editColumn: 34,
           },
-          lineNumber: 24,
+          lineNumber: 26,
           ruleDescription: 'Trailing punctuation in heading',
           ruleInformation: 'https://github.com/DavidAnson/markdownlint/blob/v0.0.0/doc/md026.md',
           ruleNames: [
@@ -14827,7 +14861,7 @@ Generated by [AVA](https://avajs.dev).
             deleteCount: 1,
             editColumn: 31,
           },
-          lineNumber: 26,
+          lineNumber: 28,
           ruleDescription: 'Trailing punctuation in heading',
           ruleInformation: 'https://github.com/DavidAnson/markdownlint/blob/v0.0.0/doc/md026.md',
           ruleNames: [
@@ -14850,6 +14884,8 @@ Generated by [AVA](https://avajs.dev).
       ␊
       ## Copyright 2004 &#x000A9;␊
       ␊
+      ## Copyright 2004 &#X000A9;␊
+      ␊
       ## Copyright 2005 &#169;␊
       ␊
       ## Copyright 2006 &#169; ##␊