remove-markdown 0.0.6 → 0.2.2

package/README.md

@@ -1,5 +1,7 @@
+[![CircleCI](https://circleci.com/gh/stiang/remove-markdown.svg?style=svg&circle-token=cac2feef7dc90e6b8578aec361be369412be1c6a)](https://circleci.com/gh/stiang/remove-markdown)
+
 ## What is it?
-**remove-markdown** is a node.js/io.js module that will remove (strip) Markdown formatting from a text. "Markdown formatting" means pretty much anything that doesn’t look like regular text, like square brackets, asterisks etc.
+**remove-markdown** is a node.js module that will remove (strip) Markdown formatting from a text. "Markdown formatting" means pretty much anything that doesn’t look like regular text, like square brackets, asterisks etc.
 
 ## When do I need it?
 The typical use case is to display an excerpt of a Markdown text, without the actual Markdown (or rendered HTML, for that matter), for example in a list of posts.
@@ -12,18 +14,28 @@ npm install remove-markdown
 
 ## Usage
 ```js
-var removeMd = require('remove-markdown');
-var markdown = '# This is a heading\n\nThis is a paragraph with [a link](http://www.disney.com/) in it.';
-var plainText = removeMd(markdown); // plainText is now 'This is a heading\n\nThis is a paragraph with a link in it.'
+const removeMd = require('remove-markdown');
+const markdown = '# This is a heading\n\nThis is a paragraph with [a link](http://www.disney.com/) in it.';
+const plainText = removeMd(markdown); // plainText is now 'This is a heading\n\nThis is a paragraph with a link in it.'
 ```
 
-You can also supply an options object to the function. Currently, the only option you can set is whether to strip list leaders:
+You can also supply an options object to the function. Currently, the following options are supported:
 
 ```js
-var plainText = removeMd(markdown, {stripListLeaders: false}); // The default is true
+var plainText = removeMd(markdown, {
+  stripListLeaders: true , // strip list leaders (default: true)
+  listUnicodeChar: '',     // char to insert instead of stripped list leaders (default: '')
+  gfm: true                // support GitHub-Flavored Markdown (default: true)
+});
 ```
 
-This will retain any list characters (`*, -, +, (digit).`).
+Setting `stripListLeaders` to false will retain any list characters (`*, -, +, (digit).`).
+
+## TODO
+PRs are very much welcome.
+* Allow the RegEx expressions to be customized per rule
+* Make the rules more robust, support more edge cases
+* Add more (comprehensive) tests
 
 ## Credits
 The code is based on [Markdown Service Tools - Strip Markdown](http://brettterpstra.com/2013/10/18/a-markdown-service-to-strip-markdown/) by Brett Terpstra.

package/index.js

@@ -1,15 +1,33 @@
 module.exports = function(md, options) {
   options = options || {};
+  options.listUnicodeChar = options.hasOwnProperty('listUnicodeChar') ? options.listUnicodeChar : false;
   options.stripListLeaders = options.hasOwnProperty('stripListLeaders') ? options.stripListLeaders : true;
+  options.gfm = options.hasOwnProperty('gfm') ? options.gfm : true;
+
+  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, '');
 
-  var output = md;
   try {
     if (options.stripListLeaders) {
-      output = output.replace(/^([\s\t]*)([\*\-\+]|\d\.)\s+/gm, '$1');
+      if (options.listUnicodeChar)
+        output = output.replace(/^([\s\t]*)([\*\-\+]|\d+\.)\s+/gm, options.listUnicodeChar + ' $1');
+      else
+        output = output.replace(/^([\s\t]*)([\*\-\+]|\d+\.)\s+/gm, '$1');
+    }
+    if (options.gfm) {
+      output = output
+        // Header
+        .replace(/\n={2,}/g, '\n')
+        // Strikethrough
+        .replace(/~~/g, '')
+        // Fenced codeblocks
+        .replace(/`{3}.*\n/g, '');
     }
     output = output
       // Remove HTML tags
-      .replace(/<(.*?)>/g, '$1')
+      .replace(/<[^>]*>/g, '')
       // Remove setext-style headers
       .replace(/^[=\-]{2,}\s*$/g, '')
       // Remove footnotes?
@@ -19,18 +37,24 @@ module.exports = function(md, options) {
       .replace(/\!\[.*?\][\[\(].*?[\]\)]/g, '')
       // Remove inline links
       .replace(/\[(.*?)\][\[\(].*?[\]\)]/g, '$1')
+      // Remove blockquotes
+      .replace(/^\s{0,3}>\s?/g, '')
       // Remove reference-style links?
       .replace(/^\s{1,2}\[(.*?)\]: (\S+)( ".*?")?\s*$/g, '')
       // Remove atx-style headers
-      .replace(/^\#{1,6}\s*([^#]*)\s*(\#{1,6})?/gm, '$1')
-      .replace(/([\*_]{1,2})(\S.*?\S)\1/g, '$2')
+      .replace(/^(\n)?\s{0,}#{1,6}\s+| {0,}(\n)?\s{0,}#{0,} {0,}(\n)?\s{0,}$/gm, '$1$2$3')
+      // Remove emphasis (repeat the line to remove double emphasis)
+      .replace(/([\*_]{1,3})(\S.*?\S{0,1})\1/g, '$2')
+      .replace(/([\*_]{1,3})(\S.*?\S{0,1})\1/g, '$2')
+      // Remove code blocks
       .replace(/(`{3,})(.*?)\1/gm, '$2')
-      .replace(/^-{3,}\s*$/g, '')
+      // Remove inline code
       .replace(/`(.+?)`/g, '$1')
+      // Replace two or more newlines with exactly two? Not entirely sure this belongs here...
       .replace(/\n{2,}/g, '\n\n');
   } catch(e) {
     console.error(e);
-    return md;    
+    return md;
   }
   return output;
-}
+};

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "remove-markdown",
-  "version": "0.0.6",
+  "version": "0.2.2",
   "description": "Remove Markdown formatting from text",
   "main": "index.js",
   "scripts": {
@@ -20,6 +20,7 @@
   },
   "homepage": "https://github.com/stiang/remove-markdown",
   "devDependencies": {
+    "chai": "^4.0.2",
     "mocha": "^2.1.0",
     "should": "^5.0.0"
   }

package/test/remove-markdown.js

@@ -1,14 +1,139 @@
 'use strict';
+const expect = require('chai').expect;
+const removeMd = require('../');
 
-var fs = require('fs'),
-    path = require('path'),
-    should = require('should'),
-    removeMd = require('../'),
-    markdown = fs.readFileSync(path.resolve(__dirname, 'markdown.md')).toString(),
-    result = fs.readFileSync(path.resolve(__dirname, 'result.txt')).toString();
+describe('remove Markdown', function () {
+  describe('removeMd', function () {
+    it('should leave a string alone without markdown', function () {
+      const string = 'Javascript Developers are the best.';
+      expect(removeMd(string)).to.equal(string);
+    });
 
-describe('remove-markdown', function () {
-  it('should remove markdown', function () {
-    removeMd(markdown).should.eql(result);
+    it('should strip out remaining markdown', function () {
+      const string = '*Javascript* developers are the _best_.';
+      const expected = 'Javascript developers are the best.';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
+    it('should leave non-matching markdown markdown', function () {
+      const string = '*Javascript* developers* are the _best_.';
+      const expected = 'Javascript developers* are the best.';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
+    it('should leave non-matching markdown, but strip empty anchors', function () {
+      const string = '*Javascript* [developers]()* are the _best_.';
+      const expected = 'Javascript developers* are the best.';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
+    it('should strip HTML', function () {
+      const string = '<p>Hello World</p>';
+      const expected = 'Hello World';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
+    it('should strip anchors', function () {
+      const string = '*Javascript* [developers](https://engineering.condenast.io/)* are the _best_.';
+      const expected = 'Javascript developers* are the best.';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
+    it('should strip img tags', function () {
+      const string = '![bear](https://placebear.com/640/480)*Javascript* developers are the _best_.';
+      const expected = 'Javascript developers are the best.';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
+    it('should strip code tags', function () {
+      const string = 'In `Getting Started` we set up `something` foo.';
+      const expected = 'In Getting Started we set up something foo.';
+      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';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
+    it('should remove emphasis', function () {
+      const string = 'I italicized an *I* and it _made_ me *sad*.';
+      const expected = 'I italicized an I and it made me sad.';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
+    it('should remove double emphasis', function () {
+      const string = '**this sentence has __double styling__**';
+      const expected = 'this sentence has double styling';
+      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';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
+    it('should remove horizontal rules with space-separated asterisks', 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';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
+    it('should remove blockquotes', function () {
+      const string = '>I am a blockquote';
+      const expected = 'I am a blockquote';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
+    it('should remove blockquotes with spaces', function () {
+      const string = '> I am a blockquote';
+      const expected = 'I am a blockquote';
+      expect(removeMd(string)).to.equal(expected);
+    });
+
+    it('should remove indented blockquotes', function () {
+        var tests = [
+            { string: ' > I am a blockquote', expected: 'I am a blockquote' },
+            { string: '  > I am a blockquote', expected: 'I am a blockquote' },
+            { string: '   > I am a blockquote', expected: 'I am a blockquote' },
+        ];
+        tests.forEach(function (test) {
+            expect(removeMd(test.string)).to.equal(test.expected);
+        });
+    });
+
+    it('should not remove greater than signs', function () {
+      var tests = [
+          { string: '100 > 0', expected: '100 > 0' },
+          { string: '100 >= 0', expected: '100 >= 0' },
+          { string: '100>0', expected: '100>0' },
+          { string: '> 100 > 0', expected: '100 > 0' },
+          { string: '1 < 100', expected: '1 < 100' },
+          { string: '1 <= 100', expected: '1 <= 100' },
+      ];
+      tests.forEach(function (test) {
+          expect(removeMd(test.string)).to.equal(test.expected);
+      });
+    });
+
+    it('should strip unordered list leaders', 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 strip ordered list leaders', function () {
+      const string = 'Some text on a line\n\n9. A list Item\n10. 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';
+      expect(removeMd(paragraph)).to.equal(expected);
+    });
   });
-});
+});

package/test/markdown.md

@@ -1,11 +0,0 @@
-## This is a heading ##
-
-This is a paragraph with [a link](http://www.disney.com/).
-
-### This is another heading
-
-In `Getting Started` we set up `something` foo.
-
-* Some list
-* With items
-  * Even indented

package/test/result.txt

@@ -1,11 +0,0 @@
-This is a heading 
-
-This is a paragraph with a link.
-
-This is another heading
-
-In Getting Started we set up something foo.
-
-Some list
-With items
-  Even indented