Assorted fixes & speed opts

lib/helpers.js

@@ -91,6 +91,23 @@ function escapeHtml(str) {
   return str;
 }
 
+function escapeHtmlKeepEntities(str) {
+  if (str.indexOf('&') >= 0) {
+    str = str.replace(/[&](?![#](x[a-f0-9]{1,8}|[0-9]{1,8});|[a-z][a-z0-9]{1,31};)/gi,'&');
+  }
+  if (str.indexOf('<') >= 0) { str = str.replace(/</g, '&lt;'); }
+  if (str.indexOf('>') >= 0) { str = str.replace(/>/g, '&gt;'); }
+  if (str.indexOf('"') >= 0) { str = str.replace(/"/g, '&quot;'); }
+  return str;
+}
+
+var UNESCAPE_MD_RE = /\\([\\!"#$%&'()*+,.\/:;<=>?@[\]^_`{|}~-])/g;
+
+function unescapeMd(str) {
+  if (str.indexOf('\\') < 0) { return str; }
+  return str.replace(UNESCAPE_MD_RE, '$1');
+}
+
 
 exports.isWhiteSpace = isWhiteSpace;
 exports.isEmpty = isEmpty;
@@ -100,3 +117,5 @@ exports.skipChars = skipChars;
 exports.getLines = getLines;
 exports.skipCharsBack = skipCharsBack;
 exports.escapeHtml = escapeHtml;
+exports.unescapeMd = unescapeMd;
+exports.escapeHtmlKeepEntities = escapeHtmlKeepEntities;

lib/lexer_block/fences.js

@@ -85,7 +85,7 @@ module.exports = function fences(state, startLine, endLine, silent) {
 
   state.tokens.push({
     type: 'fence',
-    params: params ? params.split(/\s+/g) : [],
+    params: params,
     content: getLines(state, startLine + 1, nextLine, len, true)
   });
 

lib/lexer_block/heading.js

@@ -55,7 +55,7 @@ module.exports = function heading(state, startLine, endLine, silent) {
   if (pos < max) {
     state.tokens.push({
       type: 'inline',
-      content: state.src.slice(pos, max)
+      content: state.src.slice(pos, max).trim()
     });
   }
   state.tokens.push({ type: 'heading_close', level: level });

lib/lexer_block/lheading.js

@@ -40,7 +40,7 @@ module.exports = function lheading(state, startLine, endLine, silent) {
   state.tokens.push({ type: 'heading_open', level: marker === 0x3D/* = */ ? 1 : 2 });
   state.tokens.push({
     type: 'inline',
-    content: state.src.slice(pos, max)
+    content: state.src.slice(pos, max).trim()
   });
   state.tokens.push({ type: 'heading_close', level: marker === 0x3D/* = */ ? 1 : 2 });
 

lib/lexer_block/paragraph.js

@@ -37,7 +37,7 @@ module.exports = function paragraph(state, startLine/*, endLine*/) {
   state.tokens.push({ type: 'paragraph_open' });
   state.tokens.push({
     type: 'inline',
-    content: getLines(state, startLine, nextLine, state.blkIndent, false)
+    content: getLines(state, startLine, nextLine, state.blkIndent, false).trim()
   });
   state.tokens.push({ type: 'paragraph_close' });
 

lib/lexer_inline.js

@@ -152,21 +152,10 @@ LexerInline.prototype.tokenize = function (state) {
 // Parse input string.
 //
 LexerInline.prototype.parse = function (str, options) {
-  var state = new StateInline(str, this, options),
+  var state = new StateInline(str, this, options);
-      tokens = state.tokens,
-      len, tok;
 
   this.tokenize(state);
 
-  // If last token is text - strip tailing spaces
-  len = tokens.length;
-  if (len) {
-    tok = tokens[len - 1];
-    if (tok.type === 'text' && tok.content.charCodeAt(tok.content.length - 1) === 0x20) {
-      tok.content = tok.content.replace(/ +$/, '');
-    }
-  }
-
   return state.tokens;
 };
 

lib/lexer_inline/autolink.js

@@ -1,7 +1,8 @@
 // Process autolinks '<protocol:...>'
 
 
-var escape = require('../helpers').escapeHtml;
+var escapeHtml = require('../helpers').escapeHtml;
+
 
 /*eslint max-len:0*/
 var EMAIL_RE    = /^<([a-zA-Z0-9.!#$%&'*+\/=?^_`{|}~-]+@[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?(?:\.[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*)>/;
@@ -24,11 +25,11 @@ module.exports = function autolink(state) {
 
     state.push({
       type: 'link_open',
-      href: escape(linkMatch[0].slice(1, -1))
+      href: linkMatch[0].slice(1, -1)
     });
     state.push({
       type: 'text',
-      content: escape(linkMatch[0].slice(1, -1))
+      content: escapeHtml(linkMatch[0].slice(1, -1))
     });
     state.push({ type: 'link_close' });
 
@@ -41,11 +42,11 @@ module.exports = function autolink(state) {
   if (emailMatch) {
     state.tokens.push({
       type: 'link_open',
-      href: 'mailto:' + escape(emailMatch[0].slice(1, -1))
+      href: 'mailto:' + emailMatch[0].slice(1, -1)
     });
     state.tokens.push({
       type: 'text',
-      content: escape(emailMatch[0].slice(1, -1))
+      content: escapeHtml(emailMatch[0].slice(1, -1))
     });
     state.tokens.push({ type: 'link_close' });
 

lib/lexer_inline/escape.js

@@ -1,6 +1,6 @@
 // Proceess escaped chars and hardbreaks
 
-var ESCAPED = '\\!"#$%&\'()*+,./:;<=>?@[]^_`{|}~-]'
+var ESCAPED = '\\!"#$%&\'()*+,./:;<=>?@[]^_`{|}~-'
                       .split('')
                       .map(function(ch) { return ch.charCodeAt(0); });
 

lib/lexer_inline/newline.js

@@ -26,7 +26,7 @@ module.exports = function escape(state) {
   }
 
   if (pmax > 0 && state.pending.charCodeAt(pmax) === 0x20) {
-    state.pending = state.pending.slice(0, -1);
+    state.pending = state.pending.replace(/ +$/, '');
   }
 
   state.pending += '\n';

lib/renderer.js

@@ -3,6 +3,8 @@
 
 var assign = require('object-assign');
 var escapeHtml = require('./helpers').escapeHtml;
+var escapeHtmlKeepEntities = require('./helpers').escapeHtmlKeepEntities;
+var unescapeMd = require('./helpers').unescapeMd;
 
 
 // check if we need to hide '\n' before next token
@@ -40,9 +42,11 @@ rules.fence = function (tokens, idx, options) {
   var token = tokens[idx];
   var langMark = '';
   var langPrefix = options.codeLangPrefix || '';
+  var params;
 
-  if (token.params.length) {
+  if (token.params) {
-    langMark = ' class="' + langPrefix + escapeHtml(token.params[0]) + '"';
+    params = token.params.split(/ +/g);
+    langMark = ' class="' + langPrefix + escapeHtmlKeepEntities(unescapeMd(params[0])) + '"';
   }
 
   return  '<pre><code' + langMark + '>'
@@ -96,7 +100,7 @@ rules.paragraph_close = function (tokens, idx /*, options*/) {
 
 
 rules.link_open = function (tokens, idx /*, options*/) {
-  return '<a href="' + tokens[idx].href + '">';
+  return '<a href="' + escapeHtmlKeepEntities(tokens[idx].href) + '">';
 };
 rules.link_close = function (/*tokens, idx, options*/) {
   return '</a>';

lib/state_inline.js

@@ -16,11 +16,7 @@ function StateInline(src, lexer, options) {
 
 StateInline.prototype.pushText = function () {
   var pending = this.pending;
-  // strip leading spaces from the first token.
+
-  // others will be stripped by logic in `newline` rule
-  if (this.tokens.length === 0 && pending.charCodeAt(0) === 0x20) {
-    pending = pending.replace(/^ +/, '');
-  }
   this.tokens.push({
     type: 'text',
     content: pending

test/fixtures/stmd/bad.txt

@@ -718,24 +718,6 @@ error:
 <p>[foo]: /bar* &quot;ti*tle&quot;</p>
 
 
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-src line: 3718
-
-.
-``` foo\+bar
-foo
-```
-.
-<pre><code class="language-foo+bar">foo
-</code></pre>
-.
-
-error:
-
-<pre><code class="language-foo\+bar">foo
-</code></pre>
-
-
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 src line: 3791
 
@@ -781,24 +763,6 @@ error:
 <p>[foo]: /f&ouml;&ouml; &quot;f&ouml;&ouml;&quot;</p>
 
 
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-src line: 3811
-
-.
-``` f&ouml;&ouml;
-foo
-```
-.
-<pre><code class="language-f&ouml;&ouml;">foo
-</code></pre>
-.
-
-error:
-
-<pre><code class="language-f&amp;ouml;&amp;ouml;">foo
-</code></pre>
-
-
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 src line: 3954
 

test/fixtures/stmd/good.txt

@@ -2668,6 +2668,18 @@ src line: 3688
 <p><a href="http://google.com?find=\*">http://google.com?find=\*</a></p>
 .
 
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+src line: 3718
+
+.
+``` foo\+bar
+foo
+```
+.
+<pre><code class="language-foo+bar">foo
+</code></pre>
+.
+
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 src line: 3736
 
@@ -2722,6 +2734,18 @@ src line: 3781
 <p>&MadeUpEntity;</p>
 .
 
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+src line: 3811
+
+.
+``` f&ouml;&ouml;
+foo
+```
+.
+<pre><code class="language-f&ouml;&ouml;">foo
+</code></pre>
+.
+
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 src line: 3822