Ignore the content of (valid) HTML comments when linting (fixes #64).

.eslintrc.json

@@ -57,7 +57,7 @@
         "max-lines": "off",
         "max-nested-callbacks": "error",
         "max-params": ["error", 7],
-        "max-statements": ["error", 30],
+        "max-statements": ["error", 32],
         "max-statements-per-line": "error",
         "multiline-ternary": "off",
         "new-cap": "error",

README.md

@@ -115,6 +115,12 @@ See [Rules.md](doc/Rules.md) for more details.
 
 ## Configuration
 
+Text passed to `markdownlint` is parsed as Markdown, analyzed, and any issues reported.
+Two kinds of text are ignored:
+
+* [HTML comments](https://www.w3.org/TR/html5/syntax.html#comments)
+* [Front matter](https://jekyllrb.com/docs/frontmatter/) (see `options.frontMatter` below)
+
 Rules can be enabled, disabled, and configured via `options.config` (described
 below) to define the expected behavior for a set of inputs. To enable or disable
 rules within a file, add one of these markers to the appropriate place (HTML

lib/markdownlint.js

@@ -104,6 +104,8 @@ function lintContent(
       }
     }
   }
+  // Ignore the content of HTML comments
+  content = shared.clearHtmlCommentText(content);
   // Parse content into tokens and lines
   var tokens = md.parse(content, {});
   var lines = content.split(shared.newLineRe);

lib/shared.js

@@ -7,8 +7,9 @@ module.exports.newLineRe = /\r\n|\r|\n/;
 module.exports.frontMatterRe = /^(---|\+\+\+)$[^]*?^\1$(\r\n|\r|\n)/m;
 
 // Regular expression for matching inline disable/enable comments
-module.exports.inlineCommentRe =
+var inlineCommentRe =
   /<!--\s*markdownlint-(dis|en)able((?:\s+[a-z0-9_-]+)*)\s*-->/ig;
+module.exports.inlineCommentRe = inlineCommentRe;
 
 // readFile options for reading with the UTF-8 encoding
 module.exports.utf8Encoding = { "encoding": "utf8" };
@@ -26,3 +27,36 @@ module.exports.assign = assign;
 module.exports.clone = function clone(obj) {
   return assign({}, obj);
 };
+
+// Replaces the text of all properly-formatted HTML comments with whitespace
+// This preserves the line/column information for the rest of the document
+// Trailing whitespace is avoided with a '\' character in the last column
+// See https://www.w3.org/TR/html5/syntax.html#comments for details
+var htmlCommentBegin = "<!--";
+var htmlCommentEnd = "-->";
+function clearHtmlCommentText(text) {
+  var i = 0;
+  while ((i = text.indexOf(htmlCommentBegin, i)) !== -1) {
+    var j = text.indexOf(htmlCommentEnd, i);
+    if (j === -1) {
+      j = text.length;
+      text += "\\";
+    }
+    var comment = text.slice(i + htmlCommentBegin.length, j);
+    if ((comment.length > 0) &&
+        (comment[0] !== ">") &&
+        (comment[comment.length - 1] !== "-") &&
+        (comment.indexOf("--") === -1) &&
+        (text.slice(i, j + htmlCommentEnd.length)
+          .search(inlineCommentRe) === -1)) {
+      var blanks = comment
+        .replace(/[^\r\n]/g, " ")
+        .replace(/ ([\r\n])/g, "\\$1");
+      text = text.slice(0, i + htmlCommentBegin.length) +
+        blanks + text.slice(j);
+    }
+    i = j + htmlCommentEnd.length;
+  }
+  return text;
+}
+module.exports.clearHtmlCommentText = clearHtmlCommentText;

test/ignore-comments.md

@@ -0,0 +1,44 @@
+# ignore-comments.md
+
+Hard	tab {MD010}
+
+<!-- Hard	tab -->
+
+<!--Hard	tab-->
+
+<!--
+Hard	tab
+-->
+
+<!--
+Hard	tab
+
+Hard	tab
+-->
+
+<!--
+Hard	tab {MD010}
+Invalid--comment
+Hard	tab {MD010}
+-->
+
+Te<!-- Hard	tab -->xt
+
+Te<!-- Hard	tab -->xt {MD009} 
+
+T<!-- Hard	tab -->ex<!-- Hard	tab -->t
+
+Te<!--
+Hard	tab
+-->xt
+
+Te<!--
+Hard	tab
+-->xt {MD009} 
+
+Te<!-- Trailing space 
+-->xt
+
+<!-- markdownlint-disable MD010 -->
+
+Hard	tab

test/markdownlint-test.js

@@ -1142,6 +1142,158 @@ module.exports.validateConfigSchema = function validateConfigSchema(test) {
   test.done();
 };
 
+module.exports.clearHtmlCommentTextValid =
+function clearHtmlCommentTextValid(test) {
+  test.expect(1);
+  var validComments = [
+    "<!-- text -->",
+    "<!--text-->",
+    "<!-- -->",
+    "<!---->",
+    "<!---text-->",
+    "<!--text-text-->",
+    "<!--- -->",
+    "<!--",
+    "-->",
+    "<!--",
+    "",
+    "-->",
+    "<!--",
+    "",
+    "",
+    "-->",
+    "<!--",
+    "",
+    " text ",
+    "",
+    "-->",
+    "<!--text",
+    "",
+    "text-->",
+    "text<!--text-->text",
+    "text<!--",
+    "-->text",
+    "text<!--",
+    "text",
+    "-->text",
+    "<!--text--><!--text-->",
+    "text<!--text-->text<!--text-->text",
+    "<!--",
+    "text"
+  ];
+  var validResult = [
+    "<!--      -->",
+    "<!--    -->",
+    "<!-- -->",
+    "<!---->",
+    "<!--     -->",
+    "<!--         -->",
+    "<!--  -->",
+    "<!--",
+    "-->",
+    "<!--",
+    "",
+    "-->",
+    "<!--",
+    "",
+    "",
+    "-->",
+    "<!--",
+    "",
+    "     \\",
+    "",
+    "-->",
+    "<!--   \\",
+    "",
+    "    -->",
+    "text<!--    -->text",
+    "text<!--",
+    "-->text",
+    "text<!--",
+    "   \\",
+    "-->text",
+    "<!--    --><!--    -->",
+    "text<!--    -->text<!--    -->text",
+    "<!--",
+    "    \\"
+  ];
+  var actual = shared.clearHtmlCommentText(validComments.join("\n"));
+  var expected = validResult.join("\n");
+  test.equal(actual, expected);
+  test.done();
+};
+
+module.exports.clearHtmlCommentTextInvalid =
+function clearHtmlCommentTextInvalid(test) {
+  test.expect(1);
+  var invalidComments = [
+    "<!>",
+    "<!->",
+    "<!-->",
+    "<!--->",
+    "<!-->-->",
+    "<!--->-->",
+    "<!----->",
+    "<!------>",
+    "<!-- -- -->",
+    "<!-->-->",
+    "<!--> -->",
+    "<!--->-->",
+    "<!-->text-->",
+    "<!--->text-->",
+    "<!--text--->",
+    "<!--te--xt-->"
+  ];
+  var actual = shared.clearHtmlCommentText(invalidComments.join("\n"));
+  var expected = invalidComments.join("\n");
+  test.equal(actual, expected);
+  test.done();
+};
+
+module.exports.clearHtmlCommentTextNonGreedy =
+function clearHtmlCommentTextNonGreedy(test) {
+  test.expect(1);
+  var nonGreedyComments = [
+    "<!-- text --> -->",
+    "<!---text --> -->",
+    "<!--t--> -->",
+    "<!----> -->"
+  ];
+  var nonGreedyResult = [
+    "<!--      --> -->",
+    "<!--      --> -->",
+    "<!-- --> -->",
+    "<!----> -->"
+  ];
+  var actual = shared.clearHtmlCommentText(nonGreedyComments.join("\n"));
+  var expected = nonGreedyResult.join("\n");
+  test.equal(actual, expected);
+  test.done();
+};
+
+module.exports.clearHtmlCommentTextEmbedded =
+function clearHtmlCommentTextEmbedded(test) {
+  test.expect(1);
+  var embeddedComments = [
+    "text<!--text-->text",
+    "<!-- markdownlint-disable MD010 -->",
+    "text<!--text-->text",
+    "text<!-- markdownlint-disable MD010 -->text",
+    "text<!--text-->text"
+  ];
+  var embeddedResult = [
+    "text<!--    -->text",
+    "<!-- markdownlint-disable MD010 -->",
+    "text<!--    -->text",
+    "text<!-- markdownlint-disable MD010 -->text",
+    "text<!--    -->text"
+  ];
+  var actual = shared.clearHtmlCommentText(embeddedComments.join("\n"));
+  var expected = embeddedResult.join("\n");
+  test.equal(actual, expected);
+  test.done();
+};
+
 module.exports.trimPolyfills = function trimPolyfills(test) {
   var inputs = [
     "text text",