remarkable - npm Package Compare versions

Comparing version 1.3.0 to 1.4.0

CHANGELOG.md

@@ -0,1 +1,14 @@
+1.4.0 / WIP
+------------------
+
+- Added `core` chain, to better organize code and improve pluggability.
+- Added `renderInline()` and `parseInline()` methods.
+- Added abbreviations support.
+- Fixed problem with tables, having single column.
+- Fixed rendered rules rewrite for inline tags.
+- Changed internal api (ruler, inline, block classes).
+- Removed typographer chain (rules moved to `core`).
+- Removed all typographer options. Quote chars defs moved to `options.quotes`.
+
+
 1.3.0 / 2014-10-29
@@ -2,0 +15,0 @@ ------------------

lib/common/utils.js

@@ -14,17 +14,15 @@ // Utilities
   var sources = Array.prototype.slice.call(arguments, 1);
-  while (sources.length) {
-    var source = sources.shift();
-    if (!source) { continue; }
 
-    if (typeof(source) !== 'object') {
-      throw new TypeError(source + 'must be non-object');
-    }
+  sources.forEach(function (source) {
+    if (!source) { return; }
 
-    for (var p in source) {
-      if (source.hasOwnProperty(p)) {
-        obj[p] = source[p];
-      }
+    if (typeof source !== 'object') {
+      throw new TypeError(source + 'must be object');
     }
-  }
 
+    Object.keys(source).forEach(function (key) {
+      obj[key] = source[key];
+    });
+  });
+
   return obj;
@@ -70,14 +68,27 @@ }
 
-var NAMED_ENTITY_RE   = /&([a-z][a-z0-9]{1,31});/gi;
+var NAMED_ENTITY_RE   = /&([a-z#][a-z0-9]{1,31});/gi;
+var DIGITAL_ENTITY_TEST_RE = /^#((?:x[a-f0-9]{1,8}|[0-9]{1,8}))/i;
 var entities = require('./entities');
 
+function replaceEntityPattern(match, name) {
+  var code = 0;
+
+  if (entities.hasOwnProperty(name)) {
+    return entities[name];
+  } else if (name.charCodeAt(0) === 0x23/* # */ && DIGITAL_ENTITY_TEST_RE.test(name)) {
+    code = name[1].toLowerCase() === 'x' ?
+      parseInt(name.slice(2), 16)
+    :
+      parseInt(name.slice(1), 10);
+    if (isValidEntityCode(code)) {
+      return fromCodePoint(code);
+    }
+  }
+  return match;
+}
+
 function replaceEntities(str) {
   if (str.indexOf('&') < 0) { return str; }
 
-  return str.replace(NAMED_ENTITY_RE, function(match, name) {
-    if (entities.hasOwnProperty(name)) {
-      return entities[name];
-    }
-    return match;
-  });
+  return str.replace(NAMED_ENTITY_RE, replaceEntityPattern);
 }
@@ -84,0 +95,0 @@

lib/configs/commonmark.js

@@ -13,7 +13,16 @@ // Commonmark default options
     linkify:      false,        // autoconvert URL-like texts to links
-    typographer:  false,        // Enable smartypants and other sweet transforms
 
+    // Enable some language-neutral replacements + quotes beautification
+    typographer:  false,
+
+    // Double + single quotes replacement pairs, when typographer enabled,
+    // and smartquotes on. Set doubles to '«»' for Russian, '„“' for German.
+    quotes: '“”‘’',
+
     // Highlighter function. Should return escaped HTML,
     // or '' if input not changed
-    highlight: function (/*str, lang*/) { return ''; },
+    //
+    // function (/*str, lang*/) { return ''; }
+    //
+    highlight: null,
 
@@ -25,2 +34,11 @@ maxNesting:   20            // Internal protection, recursion limit
 
+    core: {
+      rules: [
+        'block',
+        'inline',
+        'references',
+        'abbr2'
+      ]
+    },
+
     block: {
@@ -52,19 +70,4 @@ rules: [
       ]
-    },
-
-    typographer: {
-      options: {
-        singleQuotes: '‘’', // set empty to disable
-        doubleQuotes: '“”', // set '«»' for Russian, '„“' for German, empty to disable
-        copyright:    true, // (c) (C) → ©
-        trademark:    true, // (tm) (TM) → ™
-        registered:   true, // (r) (R) → ®
-        plusminus:    true, // +- → ±
-        paragraph:    true, // (p) (P) → §
-        ellipsis:     true, // ... → …
-        dupes:        true, // ???????? → ???, !!!!! → !!!, `,,` → `,`
-        dashes:       true  // -- → —
-      }
     }
   }
 };

lib/configs/default.js

@@ -13,7 +13,16 @@ // Remarkable default options
     linkify:      false,        // autoconvert URL-like texts to links
-    typographer:  false,        // Enable smartypants and other sweet transforms
 
+    // Enable some language-neutral replacements + quotes beautification
+    typographer:  false,
+
+    // Double + single quotes replacement pairs, when typographer enabled,
+    // and smartquotes on. Set doubles to '«»' for Russian, '„“' for German.
+    quotes: '“”‘’',
+
     // Highlighter function. Should return escaped HTML,
     // or '' if input not changed
-    highlight: function (/*str, lang*/) { return ''; },
+    //
+    // function (/*str, lang*/) { return ''; }
+    //
+    highlight: null,
 
@@ -25,2 +34,15 @@ maxNesting:   20            // Internal protection, recursion limit
 
+    core: {
+      rules: [
+        'block',
+        'inline',
+        'references',
+        'replacements',
+        'linkify',
+        'smartquotes',
+        'references',
+        'abbr2'
+      ]
+    },
+
     block: {
@@ -54,19 +76,4 @@ rules: [
       ]
-    },
-
-    typographer: {
-      options: {
-        singleQuotes: '‘’', // set empty to disable
-        doubleQuotes: '“”', // set '«»' for Russian, '„“' for German, empty to disable
-        copyright:    true, // (c) (C) → ©
-        trademark:    true, // (tm) (TM) → ™
-        registered:   true, // (r) (R) → ®
-        plusminus:    true, // +- → ±
-        paragraph:    true, // (p) (P) → §
-        ellipsis:     true, // ... → …
-        dupes:        true, // ???????? → ???, !!!!! → !!!, `,,` → `,`
-        dashes:       true  // -- → —
-      }
     }
   }
 };

lib/configs/full.js

@@ -13,7 +13,16 @@ // Remarkable default options
     linkify:      false,        // autoconvert URL-like texts to links
-    typographer:  false,        // Enable smartypants and other sweet transforms
 
+    // Enable some language-neutral replacements + quotes beautification
+    typographer:  false,
+
+    // Double + single quotes replacement pairs, when typographer enabled,
+    // and smartquotes on. Set doubles to '«»' for Russian, '„“' for German.
+    quotes: '“”‘’',
+
     // Highlighter function. Should return escaped HTML,
     // or '' if input not changed
-    highlight: function (/*str, lang*/) { return ''; },
+    //
+    // function (/*str, lang*/) { return ''; }
+    //
+    highlight: null,
 
@@ -25,21 +34,7 @@ maxNesting:   20            // Internal protection, recursion limit
 
-    // Don't restrict block/inline rules
+    // Don't restrict core/block/inline rules
+    core: {},
     block: {},
-    inline: {},
-
-    typographer: {
-      options: {
-        singleQuotes: '‘’', // set empty to disable
-        doubleQuotes: '“”', // set '«»' for Russian, '„“' for German, empty to disable
-        copyright:    true, // (c) (C) → ©
-        trademark:    true, // (tm) (TM) → ™
-        registered:   true, // (r) (R) → ®
-        plusminus:    true, // +- → ±
-        paragraph:    true, // (p) (P) → §
-        ellipsis:     true, // ... → …
-        dupes:        true, // ???????? → ???, !!!!! → !!!, `,,` → `,`
-        dashes:       true  // -- → —
-      }
-    }
+    inline: {}
   }
 };

lib/index.js

@@ -9,8 +9,7 @@ // Main perser class
 var Renderer     = require('./renderer');
+var ParserCore   = require('./parser_core');
 var ParserBlock  = require('./parser_block');
 var ParserInline = require('./parser_inline');
-var Typographer  = require('./typographer');
-var Linkifier    = require('./linkifier');
+var Ruler        = require('./ruler');
 
-
 var config = {
@@ -23,2 +22,15 @@ 'default': require('./configs/default'),
 
+function StateCore(self, src, env) {
+  this.src = src;
+  this.env = env;
+  this.options = self.options;
+  this.tokens = [];
+  this.inlineMode = false;
+
+  this.inline = self.inline;
+  this.block = self.block;
+  this.renderer = self.renderer;
+  this.typographer = self.typographer;
+}
+
 // Main class
@@ -34,16 +46,9 @@ //
 
-  this.options  = {};
-  this.state    = null;
-
   this.inline   = new ParserInline();
   this.block    = new ParserBlock();
+  this.core     = new ParserCore();
   this.renderer = new Renderer();
-  this.typographer = new Typographer();
-  this.linkifier   = new Linkifier();
+  this.ruler    = new Ruler();
 
-  // Cross-references to simplify code (a bit dirty, but easy).
-  this.block.inline       = this.inline;
-  this.inline.typographer = this.typographer;
-  this.inline.linkifier   = this.linkifier;
-
+  this.options  = {};
   this.configure(config[presetName]);
@@ -67,3 +72,3 @@
 
-  if (!presets) { throw new Error('Wrong config name'); }
+  if (!presets) { throw new Error('Wrong `remarkable` preset, check name/content'); }
 
@@ -77,5 +82,2 @@ if (presets.options) { self.set(presets.options); }
       }
-      if (presets.components[name].options) {
-        self[name].set(presets.components[name].options);
-      }
     });
@@ -104,15 +106,7 @@ }
 Remarkable.prototype.parse = function (src, env) {
-  var tokens, tok, i, l;
-  // Parse blocks
-  tokens = this.block.parse(src, this.options, env);
+  var state = new StateCore(this, src, env);
 
-  // Parse inlines
-  for (i = 0, l = tokens.length; i < l; i++) {
-    tok = tokens[i];
-    if (tok.type === 'inline') {
-      tok.children = this.inline.parse(tok.content, this.options, env);
-    }
-  }
+  this.core.process(state);
 
-  return tokens;
+  return state.tokens;
 };
@@ -122,4 +116,4 @@
 //
-Remarkable.prototype.render = function (src) {
-  var env = { references: {} };
+Remarkable.prototype.render = function (src, env) {
+  env = env || {};
 
@@ -130,2 +124,22 @@ return this.renderer.render(this.parse(src, env), this.options, env);
 
+// Parse content as single string
+//
+Remarkable.prototype.parseInline = function (src, env) {
+  var state = new StateCore(this, src, env);
+
+  state.inlineMode = true;
+  this.core.process(state);
+
+  return state.tokens;
+};
+
+// Render single string, without wrapping it to paragraphs
+//
+Remarkable.prototype.renderInline = function (src, env) {
+  env = env || {};
+
+  return this.renderer.render(this.parseInline(src, env), this.options, env);
+};
+
+
 module.exports = Remarkable;

lib/parser_block.js

@@ -8,3 +8,3 @@ // Block parser
 var Ruler           = require('./ruler');
-var State           = require('./rules_block/state_block');
+var StateBlock      = require('./rules_block/state_block');
 
@@ -29,9 +29,4 @@
 function ParserBlock() {
-  this._rules = [];
-  this._rulesParagraphTerm  = [];
-  this._rulesBlockquoteTerm = [];
-  this._rulesListTerm       = [];
+  this.ruler = new Ruler();
 
-  this.ruler = new Ruler(this.rulesUpdate.bind(this));
-
   for (var i = 0; i < _rules.length; i++) {
@@ -43,10 +38,2 @@ this.ruler.push(_rules[i][0], _rules[i][1], { alt: (_rules[i][2] || []).slice() });
 
-ParserBlock.prototype.rulesUpdate = function () {
-  this._rules = this.ruler.getRules();
-  this._rulesParagraphTerm  = this.ruler.getRules('paragraph');
-  this._rulesBlockquoteTerm = this.ruler.getRules('blockquote');
-  this._rulesListTerm       = this.ruler.getRules('list');
-};
-
-
 // Generate tokens for input range
@@ -56,4 +43,4 @@ //
   var ok, i,
-      rules = this._rules,
-      len = this._rules.length,
+      rules = this.ruler.getRules(''),
+      len = rules.length,
       line = startLine,
@@ -82,8 +69,2 @@ hasEmptyLines = false;
 
-    if (!ok) { throw new Error('No matching rules found'); }
-
-    if (line === state.line) {
-      throw new Error('None of rules updated state.line');
-    }
-
     // set state.tight iff we had an empty line before current tag
@@ -115,3 +96,3 @@ // i.e. latest empty line should not count
 
-ParserBlock.prototype.parse = function (src, options, env) {
+ParserBlock.prototype.parse = function (src, options, env, outTokens) {
   var state, lineStart = 0, lastTabPos = 0;
@@ -142,13 +123,11 @@
 
-  state = new State(
+  state = new StateBlock(
     src,
     this,
-    [],
     options,
-    env
+    env,
+    outTokens
   );
 
   this.tokenize(state, state.line, state.lineMax);
-
-  return state.tokens;
 };
@@ -155,0 +134,0 @@

lib/parser_inline.js

@@ -33,10 +33,4 @@ // Inline parser
 function validateLink(url) {
-  var str = '';
+  var str = decodeURI(url).trim().toLowerCase();
 
-  try {
-    str = decodeURI(url).trim().toLowerCase();
-  } catch (_) {}
-
-  if (!str) { return false; }
-
   if (str.indexOf(':') >= 0 && BAD_PROTOCOLS.indexOf(str.split(':')[0]) >= 0) {
@@ -51,8 +45,2 @@ return false;
 function ParserInline() {
-  this._rules = [];
-
-  // Rule to skip pure text
-  // - '{}$%@+=:' reserved for extentions
-  this.textMatch = /[\n\\`*_^\[\]!&<{}$%@~+=:]/;
-
   // By default CommonMark allows too much in links
@@ -62,3 +50,3 @@ // If you need to restrict it - override this with your validator.
 
-  this.ruler = new Ruler(this.rulesUpdate.bind(this));
+  this.ruler = new Ruler();
 
@@ -71,7 +59,2 @@ for (var i = 0; i < _rules.length; i++) {
 
-ParserInline.prototype.rulesUpdate = function () {
-  this._rules = this.ruler.getRules();
-};
-
-
 // Skip single token by running all rules in validation mode;
@@ -82,3 +65,4 @@ // returns `true` if any rule reported success
   var i, cached_pos, pos = state.pos,
-      len = this._rules.length;
+      rules = this.ruler.getRules(''),
+      len = rules.length;
 
@@ -91,3 +75,3 @@ if ((cached_pos = state.cacheGet(pos)) > 0) {
   for (i = 0; i < len; i++) {
-    if (this._rules[i](state, true)) {
+    if (rules[i](state, true)) {
       state.cacheSet(pos, state.pos);
@@ -107,3 +91,4 @@ return;
   var ok, i,
-      len = this._rules.length,
+      rules = this.ruler.getRules(''),
+      len = rules.length,
       end = state.posMax;
@@ -121,3 +106,3 @@
     for (i = 0; i < len; i++) {
-      ok = this._rules[i](state, false);
+      ok = rules[i](state, false);
       if (ok) { break; }
@@ -137,4 +122,2 @@ }
   }
-
-  return state.tokens;
 };
@@ -145,15 +128,6 @@
 //
-ParserInline.prototype.parse = function (str, options, env) {
-  var state = new StateInline(str, this, options, env);
+ParserInline.prototype.parse = function (str, options, env, outTokens) {
+  var state = new StateInline(str, this, options, env, outTokens);
 
   this.tokenize(state);
-
-  if (options.linkify) {
-    this.linkifier.process(state);
-  }
-  if (options.typographer) {
-    this.typographer.process(state);
-  }
-
-  return state.tokens;
 };
@@ -160,0 +134,0 @@

lib/renderer.js

@@ -12,15 +12,2 @@ 'use strict';
 
-function escapeUrl(str) {
-  try {
-    return encodeURI(str);
-  } catch (__) {}
-  return '';
-}
-function unescapeUrl(str) {
-  try {
-    return decodeURI(str);
-  } catch (__) {}
-  return '';
-}
-
 var HTML_ESCAPE_TEST_RE = /[&<>"]/;
@@ -46,6 +33,17 @@ var HTML_ESCAPE_REPLACE_RE = /[&<>"]/g;
 
+function nextToken(tokens, idx) {
+  if (++idx >= tokens.length - 2) { return idx; }
+  if ((tokens[idx].type === 'paragraph_open' && tokens[idx].tight) &&
+      (tokens[idx + 1].type === 'inline' && tokens[idx + 1].content.length === 0) &&
+      (tokens[idx + 2].type === 'paragraph_close' && tokens[idx + 2].tight)) {
+    return nextToken(tokens, idx + 2);
+  }
+  return idx;
+}
 
+
 // check if we need to hide '\n' before next token
 function getBreak(tokens, idx) {
-  if (++idx < tokens.length &&
+  idx = nextToken(tokens, idx);
+  if (idx < tokens.length &&
       tokens[idx].type === 'list_item_close') {
@@ -63,6 +61,7 @@ return '';
 
-rules.blockquote_open = function (/*tokens, idx, options*/) {
+
+rules.blockquote_open = function (/* tokens, idx, options, env */) {
   return '<blockquote>\n';
 };
-rules.blockquote_close = function (tokens, idx /*, options*/) {
+rules.blockquote_close = function (tokens, idx /*, options, env */) {
   return '</blockquote>' + getBreak(tokens, idx);
@@ -72,3 +71,3 @@ };
 
-rules.code = function (tokens, idx /*, options*/) {
+rules.code = function (tokens, idx /*, options, env */) {
   if (tokens[idx].block) {
@@ -82,6 +81,6 @@ return '<pre><code>' + escapeHtml(tokens[idx].content) + '</code></pre>' + getBreak(tokens, idx);
 
-rules.fence = function (tokens, idx, options) {
+rules.fence = function (tokens, idx, options /*, env */) {
   var token = tokens[idx];
   var langClass = '';
-  var langPrefix = options.langPrefix || '';
+  var langPrefix = options.langPrefix;
   var params, langName = '';
@@ -96,4 +95,9 @@ var highlighted;
 
-  highlighted = options.highlight(token.content, langName) || escapeHtml(token.content);
+  if (options.highlight) {
+    highlighted = options.highlight(token.content, langName) || escapeHtml(token.content);
+  } else {
+    highlighted = escapeHtml(token.content);
+  }
 
+
   return  '<pre><code' + langClass + '>'
@@ -105,6 +109,6 @@ + highlighted
 
-rules.heading_open = function (tokens, idx /*, options*/) {
+rules.heading_open = function (tokens, idx /*, options, env */) {
   return '<h' + tokens[idx].hLevel + '>';
 };
-rules.heading_close = function (tokens, idx /*, options*/) {
+rules.heading_close = function (tokens, idx /*, options, env */) {
   return '</h' + tokens[idx].hLevel + '>\n';
@@ -114,3 +118,3 @@ };
 
-rules.hr = function (tokens, idx, options) {
+rules.hr = function (tokens, idx, options /*, env */) {
   return (options.xhtmlOut ? '<hr />' : '<hr>') + getBreak(tokens, idx);
@@ -120,15 +124,15 @@ };
 
-rules.bullet_list_open = function (/*tokens, idx, options*/) {
+rules.bullet_list_open = function (/* tokens, idx, options, env */) {
   return '<ul>\n';
 };
-rules.bullet_list_close = function (tokens, idx /*, options*/) {
+rules.bullet_list_close = function (tokens, idx /*, options, env */) {
   return '</ul>' + getBreak(tokens, idx);
 };
-rules.list_item_open = function (/*tokens, idx, options*/) {
+rules.list_item_open = function (/* tokens, idx, options, env */) {
   return '<li>';
 };
-rules.list_item_close = function (/*tokens, idx, options*/) {
+rules.list_item_close = function (/* tokens, idx, options, env */) {
   return '</li>\n';
 };
-rules.ordered_list_open = function (tokens, idx /*, options*/) {
+rules.ordered_list_open = function (tokens, idx /*, options, env */) {
   var token = tokens[idx];
@@ -139,3 +143,3 @@ return '<ol'
 };
-rules.ordered_list_close = function (tokens, idx /*, options*/) {
+rules.ordered_list_close = function (tokens, idx /*, options, env */) {
   return '</ol>' + getBreak(tokens, idx);
@@ -145,15 +149,16 @@ };
 
-rules.paragraph_open = function (tokens, idx/*, options*/) {
+rules.paragraph_open = function (tokens, idx /*, options, env */) {
   return tokens[idx].tight ? '' : '<p>';
 };
-rules.paragraph_close = function (tokens, idx /*, options*/) {
-  return (tokens[idx].tight ? '' : '</p>') + getBreak(tokens, idx);
+rules.paragraph_close = function (tokens, idx /*, options, env */) {
+  var addBreak = !(tokens[idx].tight && idx && tokens[idx - 1].type === 'inline' && !tokens[idx - 1].content);
+  return (tokens[idx].tight ? '' : '</p>') + (addBreak ? getBreak(tokens, idx) : '');
 };
 
 
-rules.link_open = function (tokens, idx /*, options*/) {
+rules.link_open = function (tokens, idx /*, options, env */) {
   var title = tokens[idx].title ? (' title="' + escapeHtml(replaceEntities(tokens[idx].title)) + '"') : '';
-  return '<a href="' + escapeHtml(escapeUrl(unescapeUrl(replaceEntities(tokens[idx].href)))) + '"' + title + '>';
+  return '<a href="' + escapeHtml(encodeURI(decodeURI(replaceEntities(tokens[idx].href)))) + '"' + title + '>';
 };
-rules.link_close = function (/*tokens, idx, options*/) {
+rules.link_close = function (/* tokens, idx, options, env */) {
   return '</a>';
@@ -163,4 +168,4 @@ };
 
-rules.image = function (tokens, idx, options) {
-  var src = ' src="' + escapeHtml(escapeUrl(tokens[idx].src)) + '"';
+rules.image = function (tokens, idx, options /*, env */) {
+  var src = ' src="' + escapeHtml(encodeURI(tokens[idx].src)) + '"';
   var title = tokens[idx].title ? (' title="' + escapeHtml(replaceEntities(tokens[idx].title)) + '"') : '';
@@ -173,27 +178,27 @@ var alt = ' alt="' + (tokens[idx].alt ? escapeHtml(replaceEntities(tokens[idx].alt)) : '') + '"';
 
-rules.table_open = function (/*tokens, idx, options*/) {
+rules.table_open = function (/* tokens, idx, options, env */) {
   return '<table>\n';
 };
-rules.table_close = function (/*tokens, idx, options*/) {
+rules.table_close = function (/* tokens, idx, options, env */) {
   return '</table>\n';
 };
-rules.thead_open = function (/*tokens, idx, options*/) {
+rules.thead_open = function (/* tokens, idx, options, env */) {
   return '<thead>\n';
 };
-rules.thead_close = function (/*tokens, idx, options*/) {
+rules.thead_close = function (/* tokens, idx, options, env */) {
   return '</thead>\n';
 };
-rules.tbody_open = function (/*tokens, idx, options*/) {
+rules.tbody_open = function (/* tokens, idx, options, env */) {
   return '<tbody>\n';
 };
-rules.tbody_close = function (/*tokens, idx, options*/) {
+rules.tbody_close = function (/* tokens, idx, options, env */) {
   return '</tbody>\n';
 };
-rules.tr_open = function (/*tokens, idx, options*/) {
+rules.tr_open = function (/* tokens, idx, options, env */) {
   return '<tr>';
 };
-rules.tr_close = function (/*tokens, idx, options*/) {
+rules.tr_close = function (/* tokens, idx, options, env */) {
   return '</tr>\n';
 };
-rules.th_open = function (tokens, idx /*, options*/) {
+rules.th_open = function (tokens, idx /*, options, env */) {
   var token = tokens[idx];
@@ -204,6 +209,6 @@ return '<th'
 };
-rules.th_close = function (/*tokens, idx, options*/) {
+rules.th_close = function (/* tokens, idx, options, env */) {
   return '</th>';
 };
-rules.td_open = function (tokens, idx /*, options*/) {
+rules.td_open = function (tokens, idx /*, options, env */) {
   var token = tokens[idx];
@@ -214,3 +219,3 @@ return '<td'
 };
-rules.td_close = function (/*tokens, idx, options*/) {
+rules.td_close = function (/* tokens, idx, options, env */) {
   return '</td>';
@@ -220,12 +225,12 @@ };
 
-rules.strong_open = function(/*tokens, idx, options*/) {
+rules.strong_open = function (/* tokens, idx, options, env */) {
   return '<strong>';
 };
-rules.strong_close = function(/*tokens, idx, options*/) {
+rules.strong_close = function (/* tokens, idx, options, env */) {
   return '</strong>';
 };
-rules.em_open = function(/*tokens, idx, options*/) {
+rules.em_open = function (/* tokens, idx, options, env */) {
   return '<em>';
 };
-rules.em_close = function(/*tokens, idx, options*/) {
+rules.em_close = function (/* tokens, idx, options, env */) {
   return '</em>';
@@ -235,6 +240,6 @@ };
 
-rules.del_open = function(/*tokens, idx, options*/) {
+rules.del_open = function (/* tokens, idx, options, env */) {
   return '<del>';
 };
-rules.del_close = function(/*tokens, idx, options*/) {
+rules.del_close = function (/* tokens, idx, options, env */) {
   return '</del>';
@@ -244,6 +249,6 @@ };
 
-rules.ins_open = function(/*tokens, idx, options*/) {
+rules.ins_open = function (/* tokens, idx, options, env */) {
   return '<ins>';
 };
-rules.ins_close = function(/*tokens, idx, options*/) {
+rules.ins_close = function (/* tokens, idx, options, env */) {
   return '</ins>';
@@ -253,6 +258,6 @@ };
 
-rules.mark_open = function(/*tokens, idx, options*/) {
+rules.mark_open = function (/* tokens, idx, options, env */) {
   return '<mark>';
 };
-rules.mark_close = function(/*tokens, idx, options*/) {
+rules.mark_close = function (/* tokens, idx, options, env */) {
   return '</mark>';
@@ -262,6 +267,6 @@ };
 
-rules.sub = function(tokens, idx/*, options*/) {
+rules.sub = function (tokens, idx /*, options, env */) {
   return '<sub>' + escapeHtml(tokens[idx].content) + '</sub>';
 };
-rules.sup = function(tokens, idx/*, options*/) {
+rules.sup = function (tokens, idx /*, options, env */) {
   return '<sup>' + escapeHtml(tokens[idx].content) + '</sup>';
@@ -271,6 +276,6 @@ };
 
-rules.hardbreak = function (tokens, idx, options) {
+rules.hardbreak = function (tokens, idx, options /*, env */) {
   return options.xhtmlOut ? '<br />\n' : '<br>\n';
 };
-rules.softbreak = function (tokens, idx, options) {
+rules.softbreak = function (tokens, idx, options /*, env */) {
   return options.breaks ? (options.xhtmlOut ? '<br />\n' : '<br>\n') : '\n';
@@ -280,3 +285,3 @@ };
 
-rules.text = function (tokens, idx /*, options*/) {
+rules.text = function (tokens, idx /*, options, env */) {
   return escapeHtml(tokens[idx].content);
@@ -286,6 +291,6 @@ };
 
-rules.htmlblock = function (tokens, idx /*, options*/) {
+rules.htmlblock = function (tokens, idx /*, options, env */) {
   return tokens[idx].content;
 };
-rules.htmltag = function (tokens, idx /*, options*/) {
+rules.htmltag = function (tokens, idx /*, options, env */) {
   return tokens[idx].content;
@@ -295,2 +300,10 @@ };
 
+rules.abbr_open = function (tokens, idx /*, options, env */) {
+  return '<abbr title="' + escapeHtml(replaceEntities(tokens[idx].title)) + '">';
+};
+rules.abbr_close = function (/* tokens, idx, options, env */) {
+  return '</abbr>';
+};
+
+
 // Renderer class
@@ -303,7 +316,8 @@ function Renderer() {
 
-Renderer.prototype.renderInline = function (tokens, options) {
-  var result = '';
+Renderer.prototype.renderInline = function (tokens, options, env) {
+  var result = '',
+      _rules = this.rules;
 
   for (var i = 0, len = tokens.length; i < len; i++) {
-    result += rules[tokens[i].type](tokens, i, options);
+    result += _rules[tokens[i].type](tokens, i, options, env);
   }
@@ -315,3 +329,3 @@
 
-Renderer.prototype.render = function (tokens, options) {
+Renderer.prototype.render = function (tokens, options, env) {
   var i, len,
@@ -323,5 +337,5 @@ result = '',
     if (tokens[i].type === 'inline') {
-      result += this.renderInline(tokens[i].children, options);
+      result += this.renderInline(tokens[i].children, options, env);
     } else {
-      result += _rules[tokens[i].type](tokens, i, options);
+      result += _rules[tokens[i].type](tokens, i, options, env);
     }
@@ -328,0 +342,0 @@ }

lib/ruler.js

@@ -12,5 +12,3 @@ // Ruler is helper class to build responsibility chains from parse rules.
 
-function Ruler(compileFn) {
-  this.compile = compileFn; // callback to call after each change
-
+function Ruler() {
   // List of added rules. Each element is:
@@ -26,2 +24,9 @@ //
   this.rules = [];
+
+  // Cached rule chains.
+  //
+  // First level - chain name, '' for default.
+  // Second level - diginal anchor for fast filtering by charcodes.
+  //
+  this.cache = null;
 }
@@ -52,3 +57,3 @@
   this.rules[index].alt = opt.alt || [];
-  this.compile();
+  this.cache = null;
 };
@@ -72,3 +77,3 @@
 
-  this.compile();
+  this.cache = null;
 };
@@ -92,3 +97,3 @@
 
-  this.compile();
+  this.cache = null;
 };
@@ -108,29 +113,6 @@
 
-  this.compile();
+  this.cache = null;
 };
 
 
-// Get rules list as array of functions. By default returns main chain
-//
-Ruler.prototype.getRules = function (chainName) {
-  var result = [];
-
-  if (!chainName) {
-    this.rules.forEach(function (rule) {
-      if (rule.enabled) {
-        result.push(rule.fn);
-      }
-    });
-    return result;
-  }
-
-  this.rules.forEach(function (rule) {
-    if (rule.alt.indexOf(chainName) >= 0 && rule.enabled) {
-      result.push(rule.fn);
-    }
-  });
-  return result;
-};
-
-
 // Enable list of rules by names. If `strict` is true, then all non listed
@@ -160,3 +142,3 @@ // rules will be disabled.
 
-  this.compile();
+  this.cache = null;
 };
@@ -181,6 +163,48 @@
 
-  this.compile();
+  this.cache = null;
 };
 
 
+// Build rules lookup cache
+//
+Ruler.prototype.compile = function () {
+  var self = this;
+  var chains = [ '' ];
+
+  // collect unique names
+  self.rules.forEach(function (rule) {
+    if (!rule.enabled) { return; }
+
+    rule.alt.forEach(function (altName) {
+      if (chains.indexOf(altName) < 0) {
+        chains.push(altName);
+      }
+    });
+  });
+
+  self.cache = {};
+
+  chains.forEach(function (chain) {
+    self.cache[chain] = [];
+    self.rules.forEach(function (rule) {
+      if (!rule.enabled) { return; }
+
+      if (chain && rule.alt.indexOf(chain) < 0) { return; }
+
+      self.cache[chain].push(rule.fn);
+    });
+  });
+};
+
+
+// Get rules list as array of functions.
+//
+Ruler.prototype.getRules = function (chainName) {
+  if (this.cache === null) {
+    this.compile();
+  }
+
+  return this.cache[chainName];
+};
+
 module.exports = Ruler;

lib/rules_block/blockquote.js

@@ -8,3 +8,4 @@ // Block quotes
   var nextLine, lastLineEmpty, oldTShift, oldBMarks, oldIndent, oldParentType, lines,
-      terminatorRules = state.parser._rulesBlockquoteTerm, i, l, terminate,
+      terminatorRules,
+      i, l, terminate,
       pos = state.bMarks[startLine] + state.tShift[startLine],
@@ -40,2 +41,4 @@ max = state.eMarks[startLine];
 
+  terminatorRules = state.parser.ruler.getRules('blockquote');
+
   // Search the end of the block
@@ -42,0 +45,0 @@ //

lib/rules_block/code.js

@@ -6,3 +6,3 @@ // Code block (4 spaces padded)
 
-module.exports = function code(state, startLine, endLine, silent) {
+module.exports = function code(state, startLine, endLine/*, silent*/) {
   var nextLine, last;
@@ -27,4 +27,2 @@
 
-  if (silent) { return true; }
-
   state.line = nextLine;
@@ -31,0 +29,0 @@ state.tokens.push({

lib/rules_block/lheading.js

@@ -6,3 +6,3 @@ // lheading (---, ===)
 
-module.exports = function lheading(state, startLine, endLine, silent) {
+module.exports = function lheading(state, startLine, endLine/*, silent*/) {
   var marker, pos, max,
@@ -33,4 +33,2 @@ next = startLine + 1;
 
-  if (silent) { return true; }
-
   pos = state.bMarks[startLine] + state.tShift[startLine];
@@ -37,0 +35,0 @@

lib/rules_block/list.js

@@ -105,3 +105,3 @@ // Lists
       tight = true,
-      terminatorRules = state.parser._rulesListTerm,
+      terminatorRules,
       i, l, terminate;
@@ -154,2 +154,3 @@
   prevEmptyEnd = false;
+  terminatorRules = state.parser.ruler.getRules('list');
 
@@ -156,0 +157,0 @@ while (nextLine < endLine) {

lib/rules_block/paragraph.js

@@ -6,9 +6,6 @@ // Paragraph
 
-var parseRef = require('../parser_ref');
-
-
 module.exports = function paragraph(state, startLine/*, endLine*/) {
-  var endLine, content, pos, terminate, i, l,
+  var endLine, content, terminate, i, l,
       nextLine = startLine + 1,
-      terminatorRules = state.parser._rulesParagraphTerm;
+      terminatorRules;
 
@@ -18,16 +15,20 @@ endLine = state.lineMax;
   // jump line-by-line until empty one or EOF
-  for (; nextLine < endLine && !state.isEmpty(nextLine); nextLine++) {
-    // this would be a code block normally, but after paragraph
-    // it's considered a lazy continuation regardless of what's there
-    if (state.tShift[nextLine] - state.blkIndent > 3) { continue; }
+  if (nextLine < endLine && !state.isEmpty(nextLine)) {
+    terminatorRules = state.parser.ruler.getRules('paragraph');
 
-    // Some tags can terminate paragraph without empty line.
-    terminate = false;
-    for (i = 0, l = terminatorRules.length; i < l; i++) {
-      if (terminatorRules[i](state, nextLine, endLine, true)) {
-        terminate = true;
-        break;
+    for (; nextLine < endLine && !state.isEmpty(nextLine); nextLine++) {
+      // this would be a code block normally, but after paragraph
+      // it's considered a lazy continuation regardless of what's there
+      if (state.tShift[nextLine] - state.blkIndent > 3) { continue; }
+
+      // Some tags can terminate paragraph without empty line.
+      terminate = false;
+      for (i = 0, l = terminatorRules.length; i < l; i++) {
+        if (terminatorRules[i](state, nextLine, endLine, true)) {
+          terminate = true;
+          break;
+        }
       }
+      if (terminate) { break; }
     }
-    if (terminate) { break; }
   }
@@ -37,8 +38,2 @@
 
-  while (content.length) {
-    pos = parseRef(content, state.parser.inline, state.options, state.env);
-    if (pos < 0) { break; }
-    content = content.slice(pos).trim();
-  }
-
   state.line = nextLine;
@@ -45,0 +40,0 @@ if (content.length) {

lib/rules_block/state_block.js

@@ -6,3 +6,3 @@ // Parser state class
 
-function StateBlock(src, parser, tokens, options, env) {
+function StateBlock(src, parser, options, env, tokens) {
   var ch, s, start, pos, len, indent, indent_found;
@@ -9,0 +9,0 @@

lib/rules_block/table.js

@@ -6,7 +6,7 @@ // GFM table, non-standard
 
-function lineMatch(state, line, reg) {
+function getLine(state, line) {
   var pos = state.bMarks[line] + state.blkIndent,
       max = state.eMarks[line];
 
-  return state.src.substr(pos, max - pos).match(reg);
+  return state.src.substr(pos, max - pos);
 }
@@ -16,3 +16,3 @@
 module.exports = function table(state, startLine, endLine, silent) {
-  var ch, firstLineMatch, secondLineMatch, pos, i, nextLine, m, rows,
+  var ch, lineText, pos, i, nextLine, rows,
       aligns, t, tableLines, tbodyLines;
@@ -35,23 +35,33 @@
 
-  secondLineMatch = lineMatch(state, startLine + 1,
-    /^ *\|?(( *[:-]-+[:-] *\|)+( *[:-]-+[:-] *))\|? *$/);
-  if (!secondLineMatch) { return false; }
+  lineText = getLine(state, startLine + 1);
+  if (!/^[-:| ]+$/.test(lineText)) { return false; }
 
-  rows = secondLineMatch[1].split('|');
+  rows = lineText.split('|');
+  if (rows <= 2) { return false; }
   aligns = [];
   for (i = 0; i < rows.length; i++) {
     t = rows[i].trim();
+    if (!t) {
+      // allow empty columns before and after table, but not in between columns;
+      // e.g. allow ` |---| `, disallow ` ---||--- `
+      if (i === 0 || i === rows.length - 1) {
+        continue;
+      } else {
+        return false;
+      }
+    }
+
+    if (!/^:?-+:?$/.test(t)) { return false; }
     if (t.charCodeAt(t.length - 1) === 0x3A/* : */) {
-      aligns[i] = t.charCodeAt(0) === 0x3A/* : */ ? 'center' : 'right';
+      aligns.push(t.charCodeAt(0) === 0x3A/* : */ ? 'center' : 'right');
     } else if (t.charCodeAt(0) === 0x3A/* : */) {
-      aligns[i] = 'left';
+      aligns.push('left');
     } else {
-      aligns[i] = '';
+      aligns.push('');
     }
   }
 
-  firstLineMatch = lineMatch(state, startLine, /^ *\|?(.*?\|.*?)\|? *$/);
-  if (!firstLineMatch) { return false; }
-
-  rows = firstLineMatch[1].split('|');
+  lineText = getLine(state, startLine).trim();
+  if (lineText.indexOf('|') === -1) { return false; }
+  rows = lineText.replace(/^\||\|$/g, '').split('|');
   if (aligns.length !== rows.length) { return false; }
@@ -104,5 +114,5 @@ if (silent) { return true; }
 
-    m = lineMatch(state, nextLine, /^ *\|?(.*?\|.*?)\|? *$/);
-    if (!m) { break; }
-    rows = m[1].split('|');
+    lineText = getLine(state, nextLine).trim();
+    if (lineText.indexOf('|') === -1) { break; }
+    rows = lineText.replace(/^\||\|$/g, '').split('|');
 
@@ -109,0 +119,0 @@ state.tokens.push({ type: 'tr_open', level: state.level++ });

lib/rules_inline/links.js

@@ -5,6 +5,6 @@ // Process [links](<to> "stuff")
 
-var parseLinkLabel       = require('../links').parseLinkLabel;
-var parseLinkDestination = require('../links').parseLinkDestination;
-var parseLinkTitle       = require('../links').parseLinkTitle;
-var normalizeReference   = require('../links').normalizeReference;
+var parseLinkLabel       = require('../helpers/parse_link_label');
+var parseLinkDestination = require('../helpers/parse_link_destination');
+var parseLinkTitle       = require('../helpers/parse_link_title');
+var normalizeReference   = require('../helpers/normalize_reference');
 
@@ -11,0 +11,0 @@

lib/rules_inline/state_inline.js

@@ -6,8 +6,8 @@ // Inline parser state
 
-function StateInline(src, parser, options, env) {
+function StateInline(src, parserInline, options, env, outTokens) {
   this.src = src;
   this.env = env;
   this.options = options;
-  this.parser = parser;
-  this.tokens = [];
+  this.parser = parserInline;
+  this.tokens = outTokens;
   this.pos = 0;
@@ -14,0 +14,0 @@ this.posMax = this.src.length;

lib/rules_inline/text.js

@@ -6,14 +6,49 @@ // Skip text characters for text token, place those to pending buffer
 
+
+// Rule to skip pure text
+// '{}$%@~+=:' reserved for extentions
+
+function isTerminatorChar(ch) {
+  switch (ch) {
+    case 0x0A/* \n */:
+    case 0x5C/* \ */:
+    case 0x60/* ` */:
+    case 0x2A/* * */:
+    case 0x5F/* _ */:
+    case 0x5E/* ^ */:
+    case 0x5B/* [ */:
+    case 0x5D/* ] */:
+    case 0x21/* ! */:
+    case 0x26/* & */:
+    case 0x3C/* < */:
+    case 0x3E/* > */:
+    case 0x7B/* { */:
+    case 0x7D/* } */:
+    case 0x24/* $ */:
+    case 0x25/* % */:
+    case 0x40/* @ */:
+    case 0x7E/* ~ */:
+    case 0x2B/* + */:
+    case 0x3D/* = */:
+    case 0x3A/* : */:
+      return true;
+    default:
+      return false;
+  }
+}
+
 module.exports = function text(state, silent) {
-  var str = state.src.slice(state.pos),
-      next = str.search(state.parser.textMatch);
+  var pos = state.pos;
 
-  if (next === 0) { return false; }
+  while (pos < state.posMax && !isTerminatorChar(state.src.charCodeAt(pos))) {
+    pos++;
+  }
 
-  if (next < 0) { next = str.length; }
+  if (pos === state.pos) { return false; }
 
-  if (!silent) { state.pending += str.slice(0, next); }
-  state.pos += next;
+  if (!silent) { state.pending += state.src.slice(state.pos, pos); }
 
+  state.pos = pos;
+
   return true;
 };

package.json

 {
   "name": "remarkable",
-  "version": "1.3.0",
+  "version": "1.4.0",
   "description": "Markdown parser, done right. Commonmark support, extensions, syntax plugins, high speed - all in one.",
@@ -5,0 +5,0 @@ "keywords": [

README.md

@@ -11,6 +11,6 @@ # remarkable
 
-- Supports the [CommonMark](http://commonmark.org/) spec + extentions
-  (URL autolinking, typographer).
+- Supports the [CommonMark](http://commonmark.org/) spec +
+  [syntax extensions](#syntax-extensions) + sugar (URL autolinking, typographer).
 - Configurable syntax! You can add new rules and even replace existing ones.
-- High speed! See the [benchmarks](./benchmark).
+- [High speed](#benchmark)!
 
@@ -67,4 +67,10 @@
   linkify:      false,        // Autoconvert URL-like text to links
-  typographer:  false,        // Enable smartypants and other sweet transforms
 
+  // Enable some language-neutral replacement + quotes beautification
+  typographer:  false,
+
+  // Double + single quotes replacement pairs, when typographer enabled,
+  // and smartquotes on. Set doubles to '«»' for Russian, '„“' for German.
+  quotes: '“”‘’',
+
   // Highlighter function. Should return escaped HTML,
@@ -157,3 +163,3 @@ // or '' if the source string is not changed
 
-### Syntax extentions
+### Syntax extensions
 
@@ -168,8 +174,9 @@ Enabled by default:
 
-- [\<sup](http://johnmacfarlane.net/pandoc/README.html#superscripts-and-subscripts) - `19^th^`
+- [\<sup>](http://johnmacfarlane.net/pandoc/README.html#superscripts-and-subscripts) - `19^th^`
 - [\<sub>](http://johnmacfarlane.net/pandoc/README.html#superscripts-and-subscripts) - `H~2~0`
+- [abbreviations](https://michelf.ca/projects/php-markdown/extra/#abbr)
 - __\<ins>__ - `++inserted text++` (experimental)
 - __\<mark>__ - `==marked text==` (experimental)
 
-__*__ Experimental extentions can be changed later for something like
+__*__ Experimental extensions can be changed later for something like
 [Critic Markup](http://criticmarkup.com/), but you will still be able to use
@@ -202,17 +209,23 @@ old-style rules via external plugins if you prefer.
 var Remarkable = require('remarkable');
-var md = new Remarkable({ typographer: true });
+var md = new Remarkable({
+  typographer: true,
+  quotes: '“”‘’'
+});
 
-// Actual default values
-md.typographer.set({
-  singleQuotes: '‘’', // set empty to disable
-  doubleQuotes: '“”', // set '«»' for Russian, '„“' for German, empty to disable
-  copyright:    true, // (c) (C) → ©
-  trademark:    true, // (tm) (TM) → ™
-  registered:   true, // (r) (R) → ®
-  plusminus:    true, // +- → ±
-  paragraph:    true, // (p) (P) -> §
-  ellipsis:     true, // ... → … (also ?.... → ?.., !.... → !..)
-  dupes:        true, // ???????? → ???, !!!!! → !!!, `,,` → `,`
-  dashes:       true  // -- → &ndash;, --- → &mdash;
-})
+// Disable rules at all:
+md.core.ruler.disable([ 'replacements', 'smartquotes' ]);
+
+// Actual default replacements:
+//
+// '' → ‘’
+// "" → “”. Set '«»' for Russian, '„“' for German, empty to disable
+// (c) (C) → ©
+// (tm) (TM) → ™
+// (r) (R) → ®
+// +- → ±
+// (p) (P) -> §
+// ... → … (also ?.... → ?.., !.... → !..)
+// ???????? → ???, !!!!! → !!!, `,,` → `,`
+// -- → &ndash;, --- → &mdash;
+//
 ```
@@ -258,2 +271,4 @@
 ```js
+Remarkable.core
+Remarkable.core.ruler
 Remarkable.block
@@ -263,6 +278,2 @@ Remarkable.block.ruler
 Remarkable.inline.ruler
-Remarkable.typographer
-Remarkable.typographer.ruler
-Remarkable.linkifier
-Remarkable.linkifier.ruler
 Remarkable.renderer
@@ -272,3 +283,22 @@ Remarkable.renderer.rules
 
+## Benchmark
 
+Here is result of CommonMark spec parse at Core i5 2.4 GHz (i5-4258U):
+
+```bash
+$ benchmark/benchmark.js spec
+Selected samples: (1 of 27)
+ > spec
+
+Sample: spec.txt (110610 bytes)
+ > commonmark-reference x 40.42 ops/sec ±4.07% (51 runs sampled)
+ > current x 74.99 ops/sec ±4.69% (67 runs sampled)
+ > current-commonmark x 93.76 ops/sec ±1.23% (79 runs sampled)
+ > marked-0.3.2 x 22.92 ops/sec ±0.79% (41 runs sampled)
+```
+
+As you can see, `remarkabe` doesn't pay with speed for it's flexibility. Because
+it's written in monomorphyc style and use JIT inline caches effectively.
+
+
 ## Authors
@@ -275,0 +305,0 @@

No alert changes

Improved metrics

Total package byte prevSize

491710

increased by4.34%

Number of package files

62

increased by10.71%

Lines of code

13471

increased by3.5%

Number of lines in readme file

306

increased by10.87%

No dependency changes