Replace helpers.urlRe with helpers.urlFe to address "innefficient regular expression" CodeQL alert, introduce function expressions as an alternative, test more inputs.

demo/markdownlint-browser.js

@@ -47,9 +47,6 @@ module.exports.inlineCommentStartRe = inlineCommentStartRe;
 const htmlElementRe = /<(([A-Za-z][A-Za-z0-9-]*)(?:\s[^`>]*)?)\/?>/g;
 module.exports.htmlElementRe = htmlElementRe;
 // Regular expressions for range matching
-module.exports.urlRe =
-    // eslint-disable-next-line max-len
-    /(?:http|ftp)s?:\/\/(?:[^\s()<>\]"'`]|\([^\s<>\]"'`]*\))*\b(?:[-#/]|\([^\s<>\]"'`]*\))*/ig;
 module.exports.listItemMarkerRe = /^([\s>]*)(?:[*+-]|\d+[.)])\s+/;
 module.exports.orderedListItemMarkerRe = /^[\s>]*0*(\d+)[.)]/;
 // Regular expression for all instances of emphasis markers
@@ -1137,6 +1134,109 @@ function expandTildePath(file, os) {
     return homedir ? file.replace(/^~($|\/|\\)/, `${homedir}$1`) : file;
 }
 module.exports.expandTildePath = expandTildePath;
+/**
+ * RegExp.exec-style implementation of function expressions.
+ *
+ * @param {Function} funcExp Function that takes string and returns
+ * [index, length] or null.
+ * @param {string} input String to search.
+ * @returns {string[] | null} RegExp.exec-style [match] with an index property.
+ */
+function funcExpExec(funcExp, input) {
+    // Start or resume match
+    // @ts-ignore
+    const lastIndex = funcExp.lastIndex || 0;
+    const result = funcExp(input.slice(lastIndex));
+    if (result) {
+        // Update lastIndex and return match
+        const [subIndex, length] = result;
+        const index = lastIndex + subIndex;
+        // @ts-ignore
+        funcExp.lastIndex = index + length;
+        const match = [input.slice(index, index + length)];
+        // @ts-ignore
+        match.index = index;
+        return match;
+    }
+    // Reset lastIndex and return no match
+    // @ts-ignore
+    funcExp.lastIndex = 0;
+    return null;
+}
+module.exports.funcExpExec = funcExpExec;
+const urlFeProtocolRe = /(?:http|ftp)s?:\/\//i;
+const urlFeAutolinkTerminalsRe = / |$/;
+const urlFeBareTerminalsRe = /[ ,!`'"\]]|$/;
+const urlFeNonTerminalsRe = "-#/";
+const urlFePunctuationRe = /\p{Punctuation}/u;
+const urlFePrefixToPostfix = new Map([
+    [" ", " "],
+    ["`", "`"],
+    ["'", "'"],
+    ["\"", "\""],
+    ["‘", "’"],
+    ["“", "”"],
+    ["«", "»"],
+    ["*", "*"],
+    ["_", "_"],
+    ["(", ")"],
+    ["[", "]"],
+    ["{", "}"],
+    ["<", ">"],
+    [">", "<"]
+]);
+/**
+ * Function expression that matches URLs.
+ *
+ * @param {string} input Substring to search for a URL.
+ * @returns {Array | null} [index, length] of URL or null.
+ */
+function urlFe(input) {
+    // Find start of URL by searching for protocol
+    const match = input.match(urlFeProtocolRe);
+    if (match) {
+        // Look for matching pre/postfix characters (ex: <...>)
+        const start = match.index || 0;
+        const length = match[0].length;
+        const prefix = input[start - 1] || " ";
+        const postfix = urlFePrefixToPostfix.get(prefix);
+        // @ts-ignore
+        let endPostfix = input.indexOf(postfix, start + length);
+        if (endPostfix === -1) {
+            endPostfix = input.length;
+        }
+        // Look for characters that terminate a URL
+        const terminalsRe = (prefix === "<") ? urlFeAutolinkTerminalsRe : urlFeBareTerminalsRe;
+        const endTerminal = start + input.slice(start).search(terminalsRe);
+        // Determine tentative end of URL
+        let end = Math.min(endPostfix, endTerminal);
+        if (prefix === " ") {
+            // If the URL used " " as pre/postfix characters, trim the end
+            if (input[end - 1] === ")") {
+                // Trim any ")" beyond the last "(...)" pair
+                const lastOpenParen = input.lastIndexOf("(", end - 2);
+                if (lastOpenParen <= start) {
+                    end--;
+                }
+                else {
+                    const nextCloseParen = input.indexOf(")", lastOpenParen + 1);
+                    end = nextCloseParen + 1;
+                }
+            }
+            else {
+                // Trim unwanted punctuation
+                while (!urlFeNonTerminalsRe.includes(input[end - 1]) &&
+                    urlFePunctuationRe.test(input[end - 1])) {
+                    end--;
+                }
+            }
+        }
+        return [start, end - start];
+    }
+    // No match
+    return null;
+}
+module.exports.urlFe = urlFe;
 
 
 /***/ }),
@@ -3756,7 +3856,7 @@ module.exports = {
 "use strict";
 // @ts-check
 
-const { addErrorContext, filterTokens, urlRe, withinAnyRange } = __webpack_require__(/*! ../helpers */ "../helpers/helpers.js");
+const { addErrorContext, filterTokens, funcExpExec, urlFe, withinAnyRange } = __webpack_require__(/*! ../helpers */ "../helpers/helpers.js");
 const { codeBlockAndSpanRanges, htmlElementRanges, referenceLinkImageData } = __webpack_require__(/*! ./cache */ "../lib/cache.js");
 const htmlLinkRe = /<a(?:|\s[^>]+)>[^<>]*<\/a\s*>/ig;
 module.exports = {
@@ -3785,8 +3885,9 @@ module.exports = {
                 while ((match = htmlLinkRe.exec(line)) !== null) {
                     lineExclusions.push([lineIndex, match.index, match[0].length]);
                 }
-                while ((match = urlRe.exec(line)) !== null) {
+                while ((match = funcExpExec(urlFe, line)) !== null) {
                     const [bareUrl] = match;
+                    // @ts-ignore
                     const matchIndex = match.index;
                     const bareUrlLength = bareUrl.length;
                     const prefix = line.slice(0, matchIndex);
@@ -4444,7 +4545,7 @@ module.exports = {
 "use strict";
 // @ts-check
 
-const { addErrorDetailIf, escapeForRegExp, forEachLine, forEachLink, linkReferenceDefinitionRe, urlRe, withinAnyRange } = __webpack_require__(/*! ../helpers */ "../helpers/helpers.js");
+const { addErrorDetailIf, escapeForRegExp, forEachLine, forEachLink, funcExpExec, linkReferenceDefinitionRe, urlFe, withinAnyRange } = __webpack_require__(/*! ../helpers */ "../helpers/helpers.js");
 const { codeBlockAndSpanRanges, htmlElementRanges, lineMetadata } = __webpack_require__(/*! ./cache */ "../lib/cache.js");
 module.exports = {
     "names": ["MD044", "proper-names"],
@@ -4465,7 +4566,8 @@ module.exports = {
             }
             else {
                 let match = null;
-                while ((match = urlRe.exec(line)) !== null) {
+                while ((match = funcExpExec(urlFe, line)) !== null) {
+                    // @ts-ignore
                     exclusions.push([lineIndex, match.index, match[0].length]);
                 }
                 forEachLink(line, (index, _, text, destination) => {

helpers/helpers.js

@@ -23,9 +23,6 @@ const htmlElementRe = /<(([A-Za-z][A-Za-z0-9-]*)(?:\s[^`>]*)?)\/?>/g;
 module.exports.htmlElementRe = htmlElementRe;
 
 // Regular expressions for range matching
-module.exports.urlRe =
-  // eslint-disable-next-line max-len
-  /(?:http|ftp)s?:\/\/(?:[^\s()<>\]"'`]|\([^\s<>\]"'`]*\))*\b(?:[-#/]|\([^\s<>\]"'`]*\))*/ig;
 module.exports.listItemMarkerRe = /^([\s>]*)(?:[*+-]|\d+[.)])\s+/;
 module.exports.orderedListItemMarkerRe = /^[\s>]*0*(\d+)[.)]/;
 
@@ -1187,3 +1184,110 @@ function expandTildePath(file, os) {
   return homedir ? file.replace(/^~($|\/|\\)/, `${homedir}$1`) : file;
 }
 module.exports.expandTildePath = expandTildePath;
+
+/**
+ * RegExp.exec-style implementation of function expressions.
+ *
+ * @param {Function} funcExp Function that takes string and returns
+ * [index, length] or null.
+ * @param {string} input String to search.
+ * @returns {string[] | null} RegExp.exec-style [match] with an index property.
+ */
+function funcExpExec(funcExp, input) {
+  // Start or resume match
+  // @ts-ignore
+  const lastIndex = funcExp.lastIndex || 0;
+  const result = funcExp(input.slice(lastIndex));
+  if (result) {
+    // Update lastIndex and return match
+    const [ subIndex, length ] = result;
+    const index = lastIndex + subIndex;
+    // @ts-ignore
+    funcExp.lastIndex = index + length;
+    const match = [ input.slice(index, index + length) ];
+    // @ts-ignore
+    match.index = index;
+    return match;
+  }
+  // Reset lastIndex and return no match
+  // @ts-ignore
+  funcExp.lastIndex = 0;
+  return null;
+}
+module.exports.funcExpExec = funcExpExec;
+
+const urlFeProtocolRe = /(?:http|ftp)s?:\/\//i;
+const urlFeAutolinkTerminalsRe = / |$/;
+const urlFeBareTerminalsRe = /[ ,!`'"\]]|$/;
+const urlFeNonTerminalsRe = "-#/";
+const urlFePunctuationRe = /\p{Punctuation}/u;
+const urlFePrefixToPostfix = new Map([
+  [ " ", " " ],
+  [ "`", "`" ],
+  [ "'", "'" ],
+  [ "\"", "\"" ],
+  [ "‘", "’" ],
+  [ "“", "”" ],
+  [ "«", "»" ],
+  [ "*", "*" ],
+  [ "_", "_" ],
+  [ "(", ")" ],
+  [ "[", "]" ],
+  [ "{", "}" ],
+  [ "<", ">" ],
+  [ ">", "<" ]
+]);
+
+/**
+ * Function expression that matches URLs.
+ *
+ * @param {string} input Substring to search for a URL.
+ * @returns {Array | null} [index, length] of URL or null.
+ */
+function urlFe(input) {
+  // Find start of URL by searching for protocol
+  const match = input.match(urlFeProtocolRe);
+  if (match) {
+    // Look for matching pre/postfix characters (ex: <...>)
+    const start = match.index || 0;
+    const length = match[0].length;
+    const prefix = input[start - 1] || " ";
+    const postfix = urlFePrefixToPostfix.get(prefix);
+    // @ts-ignore
+    let endPostfix = input.indexOf(postfix, start + length);
+    if (endPostfix === -1) {
+      endPostfix = input.length;
+    }
+    // Look for characters that terminate a URL
+    const terminalsRe =
+      (prefix === "<") ? urlFeAutolinkTerminalsRe : urlFeBareTerminalsRe;
+    const endTerminal = start + input.slice(start).search(terminalsRe);
+    // Determine tentative end of URL
+    let end = Math.min(endPostfix, endTerminal);
+    if (prefix === " ") {
+      // If the URL used " " as pre/postfix characters, trim the end
+      if (input[end - 1] === ")") {
+        // Trim any ")" beyond the last "(...)" pair
+        const lastOpenParen = input.lastIndexOf("(", end - 2);
+        if (lastOpenParen <= start) {
+          end--;
+        } else {
+          const nextCloseParen = input.indexOf(")", lastOpenParen + 1);
+          end = nextCloseParen + 1;
+        }
+      } else {
+        // Trim unwanted punctuation
+        while (
+          !urlFeNonTerminalsRe.includes(input[end - 1]) &&
+          urlFePunctuationRe.test(input[end - 1])
+        ) {
+          end--;
+        }
+      }
+    }
+    return [ start, end - start ];
+  }
+  // No match
+  return null;
+}
+module.exports.urlFe = urlFe;

lib/md034.js

@@ -2,7 +2,7 @@
 
 "use strict";
 
-const { addErrorContext, filterTokens, urlRe, withinAnyRange } =
+const { addErrorContext, filterTokens, funcExpExec, urlFe, withinAnyRange } =
   require("../helpers");
 const { codeBlockAndSpanRanges, htmlElementRanges, referenceLinkImageData } =
   require("./cache");
@@ -34,8 +34,9 @@ module.exports = {
         while ((match = htmlLinkRe.exec(line)) !== null) {
           lineExclusions.push([ lineIndex, match.index, match[0].length ]);
         }
-        while ((match = urlRe.exec(line)) !== null) {
+        while ((match = funcExpExec(urlFe, line)) !== null) {
           const [ bareUrl ] = match;
+          // @ts-ignore
           const matchIndex = match.index;
           const bareUrlLength = bareUrl.length;
           const prefix = line.slice(0, matchIndex);

lib/md044.js

@@ -3,7 +3,8 @@
 "use strict";
 
 const { addErrorDetailIf, escapeForRegExp, forEachLine, forEachLink,
-  linkReferenceDefinitionRe, urlRe, withinAnyRange } = require("../helpers");
+  funcExpExec, linkReferenceDefinitionRe, urlFe, withinAnyRange } =
+  require("../helpers");
 const { codeBlockAndSpanRanges, htmlElementRanges, lineMetadata } =
   require("./cache");
 
@@ -27,7 +28,8 @@ module.exports = {
         exclusions.push([ lineIndex, 0, line.length ]);
       } else {
         let match = null;
-        while ((match = urlRe.exec(line)) !== null) {
+        while ((match = funcExpExec(urlFe, line)) !== null) {
+          // @ts-ignore
           exclusions.push([ lineIndex, match.index, match[0].length ]);
         }
         forEachLink(line, (index, _, text, destination) => {

test/markdownlint-test-helpers.js

@@ -1309,7 +1309,7 @@ test("expandTildePath", (t) => {
   t.is(helpers.expandTildePath("~/dir/file", null), "~/dir/file");
 });
 
-test("urlRe", (t) => {
+test("urlFe", (t) => {
   t.plan(1);
   const input = `
 Text ftp://example.com text
@@ -1329,6 +1329,11 @@ Text https://example.com/path() text
 Text https://example.com/path(path) text
 Text https://example.com/path(path)path text
 Text https://example.com/path-(path) text
+Text https://example.com/path(() text
+Text https://example.com/path()) text
+Text https://example.com/path(()) text
+Text https://example.com/path((())) text
+Text https://example.com/path()() text
 Text (https://example.com/path) text
 Text <https://example.com/path> text
 Text >https://example.com/path< text
@@ -1350,24 +1355,35 @@ Text *https://example.com* text
 Text **https://example.com** text
 Text _https://example.com_ text
 Text __https://example.com__ text
-Text https://example.com. Text
+Text https://example.com. text
 Text https://example.com, text
 Text https://example.com; text
 Text https://example.com: text
-Text https://example.com? Text
+Text https://example.com? text
-Text https://example.com! Text
+Text https://example.com! text
-Text https://example.com。 Text
+Text https://example.com。 text
-Text https://example.com， Text
+Text https://example.com， text
-Text https://example.com； Text
+Text https://example.com； text
-Text https://example.com： Text
+Text https://example.com： text
-Text https://example.com！ Text
+Text https://example.com！ text
 Text https://example.com,text
 Text https://example.com.path text
 Text https://example.com?path text
 Text https://example.com!text
+Text https://example.com.. text
+Text https://example.com... text
+Text https://example.com.co text
+Text <https://example.com/path text> text
+Text <https://example.com/path.path> text
+Text <https://example.com/path,path> text
+Text <https://example.com/path;path> text
+Text <https://example.com/path:path> text
+Text <https://example.com/path?path> text
+Text <https://example.com/path!path> text
 [https://example.com/path](https://example.com/path)
 [ https://example.com/path](https://example.com/path)
 [https://example.com/path ](https://example.com/path)
+https://example.com/ text https://example.com/path text https://example.com/
 https://example.com
  https://example.com
 https://example.com 
@@ -1390,6 +1406,11 @@ Text  text
 Text  text
 Text  text
 Text  text
+Text  text
+Text ) text
+Text ) text
+Text )) text
+Text  text
 Text () text
 Text <> text
 Text >< text
@@ -1409,26 +1430,37 @@ Text <a href="">link</a> text
 Text <a href=""></a> text
 Text ** text
 Text **** text
-Text _ text
 Text __ text
-Text . Text
+Text ____ text
+Text . text
 Text , text
 Text ; text
 Text : text
-Text ? Text
+Text ? text
-Text ! Text
+Text ! text
-Text 。 Text
+Text 。 text
-Text ， Text
+Text ， text
-Text ； Text
+Text ； text
-Text ： Text
+Text ： text
-Text ！ Text
+Text ！ text
-Text 
+Text ,text
 Text  text
 Text  text
-Text 
+Text !text
+Text .. text
+Text ... text
+Text  text
+Text < text> text
+Text <> text
+Text <> text
+Text <> text
+Text <> text
+Text <> text
+Text <> text
 []()
 [ ]()
 [ ]()
+ text  text 
 
  
  
@@ -1437,7 +1469,8 @@ Text
   for (let line of input) {
     const urlRanges = [];
     let match = null;
-    while ((match = helpers.urlRe.exec(line)) !== null) {
+    while ((match = helpers.funcExpExec(helpers.urlFe, line)) !== null) {
+      // @ts-ignore
       urlRanges.push([ match.index, match[0].length ]);
     }
     urlRanges.reverse();