Merge branch 'fixes'

lib/marked.js

@@ -16,27 +16,31 @@ var block = {
   hr: /^( *[\-*_]){3,} *\n/,
   heading: /^ *(#{1,6}) *([^\0]+?) *#* *\n+/,
   lheading: /^([^\n]+)\n *(=|-){3,}/,
-  blockquote: /^ *>[^\n]*(?:\n *>[^\n]*)*/,
-  list: /^( *)([*+-]|\d+\.) [^\0]+?(?:\n{2,}(?! )|\s*$)(?!\1\2|\1\d+\.)/,
+  blockquote: /^( *>[^\n]+(\n[^\n]+)*)+/,
+  list: /^( *)([*+-]|\d+\.) [^\0]+?(?:\n{2,}(?! )|\s*$)(?!\1bullet)/,
   html: /^ *(?:comment|closed|closing) *(?:\n{2,}|\s*$)/,
   paragraph: /^([^\n]+\n?(?!body))+\n*/,
   text: /^[^\n]+/
 };
 
+block.list = (function() {
+  var list = block.list.source;
+
+  list = list
+    .replace('bullet', /(?:[*+-](?!( *[-*]){2,})|\d+\.)/.source)
+
+  return new RegExp(list);
+})();
+
 block.html = (function() {
   var html = block.html.source;
 
   html = html
     .replace('comment', /<!--[^\0]*?-->/.source)
     .replace('closed', /<(\w+)[^\0]+?<\/\1>/.source)
-    .replace('closing', /<\w+[^>]*>/.source);
+    .replace('closing', /<\w+(?!:\/|@)\b(?:"[^"]*"|'[^']*'|[^>])*>/.source);
 
-    // Better regexes:
-    // .replace('closed', /<(\w+)\b[^\0]*?<\/\1>/.source)
-    // .replace('closing', /<\w+(?!:\/|@)\b[^>]*>/.source);
-
-  return new
-    RegExp(html);
+  return new RegExp(html);
 })();
 
 block.paragraph = (function() {
@@ -76,9 +80,9 @@ block.lexer = function(str) {
            .replace(/\t/g, '    ');
 
   str = str.replace(
-    /^ {0,3}\[([^\]]+)\]: *([^ ]+)(?: +"([^\n]+)")? *$/gm,
+    /^ {0,3}\[([^\]]+)\]: *([^ ]+)(?: +["(]([^\n]+)[")])? *$/gm,
     function(__, id, href, title) {
-      links[id] = {
+      links[id.toLowerCase()] = {
         href: href,
         title: title
       };
@@ -175,9 +179,6 @@ block.token = function(str, tokens, top) {
 
       cap = cap[0].replace(/^ *>/gm, '');
 
-      // Better conformance:
-      // cap = cap[0].replace(/^ *> ?/gm, '');
-
       // Pass `top` to keep the current
       // "toplevel" state. This is exactly
       // how markdown.pl works.
@@ -202,7 +203,7 @@ block.token = function(str, tokens, top) {
 
       // Get each top-level item.
       cap = cap[0].match(
-        /^( *)([*+-]|\d+\.)[^\n]*(?:\n(?!\1\2|\1\d+\.)[^\n]*)*/gm
+        /^( *)([*+-]|\d+\.)[^\n]*(?:\n(?!\1(?:[*+-]|\d+\.))[^\n]*)*/gm
       );
 
       i = 0;
@@ -282,16 +283,16 @@ block.token = function(str, tokens, top) {
  */
 
 var inline = {
-  escape: /^\\([\\`*{}\[\]()#+\-.!_])/,
+  escape: /^\\([\\`*{}\[\]()#+\-.!_>])/,
   autolink: /^<([^ >]+(@|:\/)[^ >]+)>/,
   gfm_autolink: /^(\w+:\/\/[^\s]+[^.,:;"')\]\s])/,
-  tag: /^<!--[^\0]*?-->|^<\/?\w+[^>]*>/,
-  link: /^!?\[((?:\[[^\]]*\]|[^\[\]])*)\]\(([^\)]*)\)/,
-  reflink: /^!?\[((?:\[[^\]]*\]|[^\[\]])*)\]\s*\[([^\]]*)\]/,
+  tag: /^<!--[^\0]*?-->|^<\/?\w+(?:"[^"]*"|'[^']*'|[^>])*>/,
+  link: /^!?\[((?:\[[^\]]*\]|[^\[\]]|\[|\](?=[^[\]]*\]))*)\]\(([^\)]*)\)/,
+  reflink: /^!?\[((?:\[[^\]]*\]|[^\[\]]|\[|\](?=[^[\]]*\]))*)\]\s*\[([^\]]*)\]/,
   nolink: /^!?\[((?:\[[^\]]*\]|[^\[\]])*)\]/,
   strong: /^__(?=\S)([^\0]*?\S)__(?!_)|^\*\*(?=\S)([^\0]*?\S)\*\*(?!\*)/,
-  em: /^\b_(?=\S)([^\0]*?\S)_\b|^\*(?=\S)([^\0]*?\S)\*/,
-  code: /^`([^`]+)`|^``([^\0]+?)``/,
+  em: /^\b_(?=\S)([^\0]*?\S)_\b|^\*(?=\S)((?:\*\*|[^\0])*?)\*(?!\*)/,
+  code: /^(`+)([^\0]*?[^`])\1(?!`)/,
   br: /^ {2,}\n(?!\s*$)/,
   text: /^[^\0]+?(?=body|$)/
 };
@@ -382,6 +383,11 @@ inline.lexer = function(str) {
     if (cap = inline.link.exec(str)) {
       str = str.substring(cap[0].length);
       text = /^\s*<?([^\s]*?)>?(?:\s+"([^\n]+)")?\s*$/.exec(cap[2]);
+      if (!text) {
+        out += cap[0][0];
+        str = cap[0].substring(1) + str;
+        continue;
+      }
       link = {
         href: text[1],
         title: text[2]
@@ -395,7 +401,7 @@ inline.lexer = function(str) {
         || (cap = inline.nolink.exec(str))) {
       str = str.substring(cap[0].length);
       link = (cap[2] || cap[1]).replace(/\s+/g, ' ');
-      link = links[link];
+      link = links[link.toLowerCase()];
       if (!link) {
         out += cap[0][0];
         str = cap[0].substring(1) + str;

test/fix.js

@@ -60,14 +60,6 @@ fs.readdirSync(dir).forEach(function(file) {
   fs.writeFileSync(file, html);
 })();
 
-// fix bad grammar
-(function() {
-  var file = dir + '/ordered_and_unordered_lists.text';
-  var text = fs.readFileSync(file, 'utf8');
-  text = text.replace(/(\n\*\sasterisk\s3\n\n)(\*\s\*\s\*)/, '$1\n$2');
-  fs.writeFileSync(file, text);
-})();
-
 // fix strange markup that isnt likely
 // to exist in the reality
 (function _(ext) {

test/index.js

@@ -101,10 +101,11 @@ main.bench = function(name, func) {
     load();
     // remove these as they affect the
     // comparison to older benchmark times.
-    Object.keys(files).forEach(function(name) {
-      if (name.indexOf('gfm') === 0) delete files[name];
+    fs.readdirSync(__dirname + '/new').forEach(function(name) {
+      if (name.split('.').pop() === 'html') return;
+      if (name === 'main.text') return;
+      delete files[name];
     });
-    delete files['toplevel_paragraphs.text'];
   }
 
   var start = Date.now()

test/new/case_insensitive_refs.html

@@ -0,0 +1 @@
+<p><a href="/url">hi</a></p>

test/new/case_insensitive_refs.text

@@ -0,0 +1,3 @@
+[hi]
+
+[HI]: /url

test/new/escape_angles.html

@@ -0,0 +1 @@
+<div style=">"/>

test/new/escape_angles.text

@@ -0,0 +1 @@
+<div style=">"/>

test/new/lazy_blockquotes.html

@@ -0,0 +1,4 @@
+<blockquote>
+  <p>hi there
+bud</p>
+</blockquote>

test/new/lazy_blockquotes.text

@@ -0,0 +1,2 @@
+> hi there
+bud

test/new/naked_angles.html

@@ -0,0 +1 @@
+<p>></p>

test/new/naked_angles.text

@@ -0,0 +1 @@
+\>

test/new/nested_code.html

@@ -0,0 +1 @@
+<p><code>hi ther `` ok ```</code></p>

test/new/nested_code.text

@@ -0,0 +1 @@
+````` hi ther `` ok ``` `````

test/new/nested_em.html

@@ -0,0 +1 @@
+<p><em>test <strong>test</strong></em></p>

test/new/nested_em.text

@@ -0,0 +1 @@
+*test **test***

test/new/nested_square_link.html

@@ -0,0 +1 @@
+<p><a href="/url">the <code>]</code> character</a></p>

test/new/nested_square_link.text

@@ -0,0 +1 @@
+[the `]` character](/url)

test/new/not_a_link.html

@@ -0,0 +1 @@
+<p>\[test](not a link)</p>

test/new/not_a_link.text

@@ -0,0 +1 @@
+\\[test](not a link)

test/new/ref_paren.html

@@ -0,0 +1 @@
+<p><a href="/url" title="there">hi</a></p>

test/new/ref_paren.text

@@ -0,0 +1,3 @@
+[hi]
+
+[hi]: /url (there)

test/new/sublist.html

@@ -0,0 +1,5 @@
+<ul>
+<li>test</li>
+<li>test</li>
+<li>test</li>
+</ul>

test/new/sublist.text

@@ -0,0 +1,3 @@
+* test
++ test
+- test

test/tests/case_insensitive_refs.html

@@ -0,0 +1 @@
+<p><a href="/url">hi</a></p>

test/tests/case_insensitive_refs.text

@@ -0,0 +1,3 @@
+[hi]
+
+[HI]: /url

test/tests/escape_angles.html

@@ -0,0 +1 @@
+<div style=">"/>

test/tests/escape_angles.text

@@ -0,0 +1 @@
+<div style=">"/>

test/tests/lazy_blockquotes.html

@@ -0,0 +1,4 @@
+<blockquote>
+  <p>hi there
+bud</p>
+</blockquote>

test/tests/lazy_blockquotes.text

@@ -0,0 +1,2 @@
+> hi there
+bud

test/tests/naked_angles.html

@@ -0,0 +1 @@
+<p>></p>

test/tests/naked_angles.text

@@ -0,0 +1 @@
+\>

test/tests/nested_code.html

@@ -0,0 +1 @@
+<p><code>hi ther `` ok ```</code></p>

test/tests/nested_code.text

@@ -0,0 +1 @@
+````` hi ther `` ok ``` `````

test/tests/nested_em.html

@@ -0,0 +1 @@
+<p><em>test <strong>test</strong></em></p>

test/tests/nested_em.text

@@ -0,0 +1 @@
+*test **test***

test/tests/nested_square_link.html

@@ -0,0 +1 @@
+<p><a href="/url">the <code>]</code> character</a></p>

test/tests/nested_square_link.text

@@ -0,0 +1 @@
+[the `]` character](/url)

test/tests/not_a_link.html

@@ -0,0 +1 @@
+<p>\[test](not a link)</p>

test/tests/not_a_link.text

@@ -0,0 +1 @@
+\\[test](not a link)

test/tests/ordered_and_unordered_lists.text

@@ -15,7 +15,6 @@ Asterisks loose:
 
 *	asterisk 3
 
-
 * * *
 
 Pluses tight:

test/tests/ref_paren.html

@@ -0,0 +1 @@
+<p><a href="/url" title="there">hi</a></p>

test/tests/ref_paren.text

@@ -0,0 +1,3 @@
+[hi]
+
+[hi]: /url (there)

test/tests/sublist.html

@@ -0,0 +1,5 @@
+<ul>
+<li>test</li>
+<li>test</li>
+<li>test</li>
+</ul>

test/tests/sublist.text

@@ -0,0 +1,3 @@
+* test
++ test
+- test