remove-markdown 0.5.5 → 0.6.1

package/README.md

@@ -1,3 +1,7 @@
+[![npm version](https://img.shields.io/npm/v/remove-markdown.svg)](https://www.npmjs.com/package/remove-markdown)
+[![npm downloads](https://img.shields.io/npm/dm/remove-markdown.svg)](https://npm-compare.com/remove-markdown/#timeRange=THREE_YEARS)
+[![GitHub Actions Build Status](https://github.com/zuchka/remove-markdown/actions/workflows/default.yaml/badge.svg)](https://github.com/zuchka/remove-markdown/actions/workflows/default.yaml)
+
 ## What is it?
 **remove-markdown** is a node.js module that will remove (strip) Markdown formatting from text.
 *Markdown formatting* means pretty much anything that doesn’t look like regular text, like square brackets, asterisks etc.

package/index.js

@@ -12,7 +12,7 @@ module.exports = function(md, options) {
   var output = md || '';
 
   // Remove horizontal rules (stripListHeaders conflict with this rule, which is why it has been moved to the top)
-  output = output.replace(/^(-\s*?|\*\s*?|_\s*?){3,}\s*/gm, '');
+  output = output.replace(/^ {0,3}((?:-[\t ]*){3,}|(?:_[ \t]*){3,}|(?:\*[ \t]*){3,})(?:\n+|$)/gm, '');
 
   try {
     if (options.stripListLeaders) {
@@ -29,29 +29,22 @@ module.exports = function(md, options) {
         .replace(/~{3}.*\n/g, '')
         // Strikethrough
         .replace(/~~/g, '')
-        // Fenced codeblocks
-        .replace(/`{3}.*\n/g, '');
+        // Fenced codeblocks with backticks
+        .replace(/```(?:.*)\n([\s\S]*?)```/g, (_, code) => code.trim());
     }
     if (options.abbr) {
       // Remove abbreviations
       output = output.replace(/\*\[.*\]:.*\n/, '');
     }
-    output = output
-    // Remove HTML tags
-      .replace(/<[^>]*>/g, '')
-
-    var htmlReplaceRegex = new RegExp('<[^>]*>', 'g');
-    if (options.htmlTagsToSkip.length > 0) {
-      // Using negative lookahead. Eg. (?!sup|sub) will not match 'sup' and 'sub' tags.
-      var joinedHtmlTagsToSkip = '(?!' + options.htmlTagsToSkip.join("|") + ')';
-
-      // Adding the lookahead literal with the default regex for html. Eg./<(?!sup|sub)[^>]*>/ig
+    
+    let htmlReplaceRegex = /<[^>]*>/g
+    if (options.htmlTagsToSkip && options.htmlTagsToSkip.length > 0) {
+      // Create a regex that matches tags not in htmlTagsToSkip
+      const joinedHtmlTagsToSkip = options.htmlTagsToSkip.join('|')
       htmlReplaceRegex = new RegExp(
-          '<' +
-          joinedHtmlTagsToSkip +
-          '[^>]*>', 
-          'ig'
-      );
+        `<(?!\/?(${joinedHtmlTagsToSkip})(?=>|\s[^>]*>))[^>]*>`,
+        'g',
+      )
     }
 
     output = output
@@ -79,7 +72,7 @@ module.exports = function(md, options) {
       //   1. Either there is a whitespace character before opening _ and after closing _.
       //   2. Or _ is at the start/end of the string.
       .replace(/(^|\W)([_]+)(\S)(.*?\S)??\2($|\W)/g, '$1$3$4$5')
-      // Remove code blocks
+      // Remove single-line code blocks (already handled multiline above in gfm section)
       .replace(/(`{3,})(.*?)\1/gm, '$2')
       // Remove inline code
       .replace(/`(.+?)`/g, '$1')

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "remove-markdown",
-  "version": "0.5.5",
+  "version": "0.6.1",
   "description": "Remove Markdown formatting from text",
   "main": "index.js",
   "scripts": {

package/test/remove-markdown.js

@@ -57,6 +57,24 @@ describe('remove Markdown', function () {
       expect(removeMd(string)).to.equal(expected);
     });
 
+    it('should strip simple multiline code tags', function () {
+      const string = '```\ncode\n```';
+      const expected = 'code';
+      expect(removeMd(string)).to.equal(expected);
+    });
+    
+    it('should strip complex multiline code blocks with language specified', function () {
+      const string = '```javascript\nconst x = 1;\nconst y = 2;\nconsole.log(x + y);\n```';
+      const expected = 'const x = 1;\nconst y = 2;\nconsole.log(x + y);';
+      expect(removeMd(string)).to.equal(expected);
+    });
+    
+    it('should strip multiline code blocks with multiple paragraphs', function () {
+      const string = 'Text before\n\n```\ncode line 1\n\ncode line 2\n```\n\nText after';
+      const expected = 'Text before\n\ncode line 1\n\ncode line 2\n\nText after';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
     it('should leave hashtags in headings', function () {
       const string = '## This #heading contains #hashtags';
       const expected = 'This #heading contains #hashtags';
@@ -87,6 +105,12 @@ describe('remove Markdown', function () {
       expect(removeMd(string)).to.equal(expected);
     });
 
+    it('should not mistake a horizontal rule when symbols are mixed ', function () {
+      const string = 'Some text on a line\n\n--*\n\nA line below';
+      const expected = 'Some text on a line\n\n--*\n\nA line below';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
     it('should remove horizontal rules', function () {
       const string = 'Some text on a line\n\n---\n\nA line below';
       const expected = 'Some text on a line\n\nA line below';
@@ -161,6 +185,12 @@ describe('remove Markdown', function () {
       expect(removeMd(string)).to.equal(expected);
     });
 
+    it('should strip list items with bold word in the beginning', function () {
+      const string = 'Some text on a line\n\n- **A** list Item\n- **Another** list item';
+      const expected = 'Some text on a line\n\nA list Item\nAnother list item';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
     it('should handle paragraphs with markdown', function () {
       const paragraph = '\n## This is a heading ##\n\nThis is a paragraph with [a link](http://www.disney.com/).\n\n### This is another heading\n\nIn `Getting Started` we set up `something` foo.\n\n  * Some list\n  * With items\n    * Even indented';
       const expected = '\nThis is a heading\n\nThis is a paragraph with a link.\n\nThis is another heading\n\nIn Getting Started we set up something foo.\n\n  Some list\n  With items\n    Even indented';
@@ -195,5 +225,16 @@ describe('remove Markdown', function () {
       const expected = 'Heading in italic';
       expect(removeMd(string)).to.equal(expected);
     });
+
+    it('should skip specified HTML tags when htmlTagsToSkip option is provided', () => {
+      const markdown =
+        '<div>HTML content <sub>Superscript</sub> <span>span text</span></div>'
+      const result = removeMd(markdown, {htmlTagsToSkip: ['sub']})
+      expect(result).to.equal('HTML content <sub>Superscript</sub> span text')
+      const result2 = removeMd(markdown, {htmlTagsToSkip: ['sub', 'span']})
+      expect(result2).to.equal(
+        'HTML content <sub>Superscript</sub> <span>span text</span>',
+      )
+    })
   });
 });