prettify.js

/**
 * @license
 * Copyright (C) 2006 Google Inc.
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

/**
 * @fileoverview
 * some functions for browser-side pretty printing of code contained in html.
 *
 * <p>
 * For a fairly comprehensive set of languages see the
 * <a href="https://github.com/google/code-prettify#for-which-languages-does-it-work">README</a>
 * file that came with this source.  At a minimum, the lexer should work on a
 * number of languages including C and friends, Java, Python, Bash, SQL, HTML,
 * XML, CSS, Javascript, and Makefiles.  It works passably on Ruby, PHP and Awk
 * and a subset of Perl, but, because of commenting conventions, doesn't work on
 * Smalltalk, Lisp-like, or CAML-like languages without an explicit lang class.
 * <p>
 * Usage: <ol>
 * <li> include this source file in an html page via
 *   {@code <script type="text/javascript" src="/path/to/prettify.js"></script>}
 * <li> define style rules.  See the example page for examples.
 * <li> mark the {@code <pre>} and {@code <code>} tags in your source with
 *    {@code class=prettyprint.}
 *    You can also use the (html deprecated) {@code <xmp>} tag, but the pretty
 *    printer needs to do more substantial DOM manipulations to support that, so
 *    some css styles may not be preserved.
 * </ol>
 * That's it.  I wanted to keep the API as simple as possible, so there's no
 * need to specify which language the code is in, but if you wish, you can add
 * another class to the {@code <pre>} or {@code <code>} element to specify the
 * language, as in {@code <pre class="prettyprint lang-java">}.  Any class that
 * starts with "lang-" followed by a file extension, specifies the file type.
 * See the "lang-*.js" files in this directory for code that implements
 * per-language file handlers.
 * <p>
 * Change log:<br>
 * cbeust, 2006/08/22
 * <blockquote>
 *   Java annotations (start with "@") are now captured as literals ("lit")
 * </blockquote>
 * @requires console
 */

// JSLint declarations
/*global console, document, navigator, setTimeout, window, define */

/**
 * @typedef {!Array.<number|string>}
 * Alternating indices and the decorations that should be inserted there.
 * The indices are monotonically increasing.
 */
var DecorationsT;

/**
 * @typedef {!{
 *   sourceNode: !Element,
 *   pre: !(number|boolean),
 *   langExtension: ?string,
 *   numberLines: ?(number|boolean),
 *   sourceCode: ?string,
 *   spans: ?(Array.<number|Node>),
 *   basePos: ?number,
 *   decorations: ?DecorationsT
 * }}
 * <dl>
 *  <dt>sourceNode<dd>the element containing the source
 *  <dt>sourceCode<dd>source as plain text
 *  <dt>pre<dd>truthy if white-space in text nodes
 *     should be considered significant.
 *  <dt>spans<dd> alternating span start indices into source
 *     and the text node or element (e.g. {@code <BR>}) corresponding to that
 *     span.
 *  <dt>decorations<dd>an array of style classes preceded
 *     by the position at which they start in job.sourceCode in order
 *  <dt>basePos<dd>integer position of this.sourceCode in the larger chunk of
 *     source.
 * </dl>
 */
var JobT;

/**
 * @typedef {!{
 *   sourceCode: string,
 *   spans: !(Array.<number|Node>)
 * }}
 * <dl>
 *  <dt>sourceCode<dd>source as plain text
 *  <dt>spans<dd> alternating span start indices into source
 *     and the text node or element (e.g. {@code <BR>}) corresponding to that
 *     span.
 * </dl>
 */
var SourceSpansT;

/** @define {boolean} */
var IN_GLOBAL_SCOPE = true;


/**
 * {@type !{
 *   'createSimpleLexer': function (Array, Array): (function (JobT)),
 *   'registerLangHandler': function (function (JobT), Array.<string>),
 *   'PR_ATTRIB_NAME': string,
 *   'PR_ATTRIB_NAME': string,
 *   'PR_ATTRIB_VALUE': string,
 *   'PR_COMMENT': string,
 *   'PR_DECLARATION': string,
 *   'PR_KEYWORD': string,
 *   'PR_LITERAL': string,
 *   'PR_NOCODE': string,
 *   'PR_PLAIN': string,
 *   'PR_PUNCTUATION': string,
 *   'PR_SOURCE': string,
 *   'PR_STRING': string,
 *   'PR_TAG': string,
 *   'PR_TYPE': string,
 *   'prettyPrintOne': function (string, string, number|boolean),
 *   'prettyPrint': function (?function, ?(HTMLElement|HTMLDocument))
 * }}
 * @const
 */
var PR;

/**
 * Split {@code prettyPrint} into multiple timeouts so as not to interfere with
 * UI events.
 * If set to {@code false}, {@code prettyPrint()} is synchronous.
 */
var PR_SHOULD_USE_CONTINUATION = true
if (typeof window !== 'undefined') {
  window['PR_SHOULD_USE_CONTINUATION'] = PR_SHOULD_USE_CONTINUATION;
}

/**
 * Pretty print a chunk of code.
 * @param {string} sourceCodeHtml The HTML to pretty print.
 * @param {string} opt_langExtension The language name to use.
 *     Typically, a filename extension like 'cpp' or 'java'.
 * @param {number|boolean} opt_numberLines True to number lines,
 *     or the 1-indexed number of the first line in sourceCodeHtml.
 * @return {string} code as html, but prettier
 */
var prettyPrintOne;
/**
 * Find all the {@code <pre>} and {@code <code>} tags in the DOM with
 * {@code class=prettyprint} and prettify them.
 *
 * @param {Function} opt_whenDone called when prettifying is done.
 * @param {HTMLElement|HTMLDocument} opt_root an element or document
 *   containing all the elements to pretty print.
 *   Defaults to {@code document.body}.
 */
var prettyPrint;


(function () {
  var win = (typeof window !== 'undefined') ? window : {};
  // Keyword lists for various languages.
  // We use things that coerce to strings to make them compact when minified
  // and to defeat aggressive optimizers that fold large string constants.
  var FLOW_CONTROL_KEYWORDS = ["break,continue,do,else,for,if,return,while"];
  var C_KEYWORDS = [FLOW_CONTROL_KEYWORDS,"auto,case,char,const,default," +
      "double,enum,extern,float,goto,inline,int,long,register,restrict,short,signed," +
      "sizeof,static,struct,switch,typedef,union,unsigned,void,volatile"];
  var COMMON_KEYWORDS = [C_KEYWORDS,"catch,class,delete,false,import," +
      "new,operator,private,protected,public,this,throw,true,try,typeof"];
  var CPP_KEYWORDS = [COMMON_KEYWORDS,"alignas,alignof,align_union,asm,axiom,bool," +
      "concept,concept_map,const_cast,constexpr,decltype,delegate," +
      "dynamic_cast,explicit,export,friend,generic,late_check," +
      "mutable,namespace,noexcept,noreturn,nullptr,property,reinterpret_cast,static_assert," +
      "static_cast,template,typeid,typename,using,virtual,where"];
  var JAVA_KEYWORDS = [COMMON_KEYWORDS,
      "abstract,assert,boolean,byte,extends,finally,final,implements,import," +
      "instanceof,interface,null,native,package,strictfp,super,synchronized," +
      "throws,transient"];
  var CSHARP_KEYWORDS = [COMMON_KEYWORDS,
      "abstract,add,alias,as,ascending,async,await,base,bool,by,byte,checked,decimal,delegate,descending," +
      "dynamic,event,finally,fixed,foreach,from,get,global,group,implicit,in,interface," +
      "internal,into,is,join,let,lock,null,object,out,override,orderby,params," +
      "partial,readonly,ref,remove,sbyte,sealed,select,set,stackalloc,string,select,uint,ulong," +
      "unchecked,unsafe,ushort,value,var,virtual,where,yield"];
  var COFFEE_KEYWORDS = "all,and,by,catch,class,else,extends,false,finally," +
      "for,if,in,is,isnt,loop,new,no,not,null,of,off,on,or,return,super,then," +
      "throw,true,try,unless,until,when,while,yes";
  var JSCRIPT_KEYWORDS = [COMMON_KEYWORDS,
      "abstract,async,await,constructor,debugger,enum,eval,export,from,function," +
      "get,import,implements,instanceof,interface,let,null,of,set,undefined," +
      "var,with,yield,Infinity,NaN"];
  var PERL_KEYWORDS = "caller,delete,die,do,dump,elsif,eval,exit,foreach,for," +
      "goto,if,import,last,local,my,next,no,our,print,package,redo,require," +
      "sub,undef,unless,until,use,wantarray,while,BEGIN,END";
  var PYTHON_KEYWORDS = [FLOW_CONTROL_KEYWORDS, "and,as,assert,class,def,del," +
      "elif,except,exec,finally,from,global,import,in,is,lambda," +
      "nonlocal,not,or,pass,print,raise,try,with,yield," +
      "False,True,None"];
  var RUBY_KEYWORDS = [FLOW_CONTROL_KEYWORDS, "alias,and,begin,case,class," +
      "def,defined,elsif,end,ensure,false,in,module,next,nil,not,or,redo," +
      "rescue,retry,self,super,then,true,undef,unless,until,when,yield," +
      "BEGIN,END"];
  var SH_KEYWORDS = [FLOW_CONTROL_KEYWORDS, "case,done,elif,esac,eval,fi," +
      "function,in,local,set,then,until"];
  var ALL_KEYWORDS = [
      CPP_KEYWORDS, CSHARP_KEYWORDS, JAVA_KEYWORDS, JSCRIPT_KEYWORDS,
      PERL_KEYWORDS, PYTHON_KEYWORDS, RUBY_KEYWORDS, SH_KEYWORDS];
  var C_TYPES = /^(DIR|FILE|array|vector|(de|priority_)?queue|(forward_)?list|stack|(const_)?(reverse_)?iterator|(unordered_)?(multi)?(set|map)|bitset|u?(int|float)\d*)\b/;

  // token style names.  correspond to css classes
  /**
   * token style for a string literal
   * @const
   */
  var PR_STRING = 'str';
  /**
   * token style for a keyword
   * @const
   */
  var PR_KEYWORD = 'kwd';
  /**
   * token style for a comment
   * @const
   */
  var PR_COMMENT = 'com';
  /**
   * token style for a type
   * @const
   */
  var PR_TYPE = 'typ';
  /**
   * token style for a literal value.  e.g. 1, null, true.
   * @const
   */
  var PR_LITERAL = 'lit';
  /**
   * token style for a punctuation string.
   * @const
   */
  var PR_PUNCTUATION = 'pun';
  /**
   * token style for plain text.
   * @const
   */
  var PR_PLAIN = 'pln';

  /**
   * token style for an sgml tag.
   * @const
   */
  var PR_TAG = 'tag';
  /**
   * token style for a markup declaration such as a DOCTYPE.
   * @const
   */
  var PR_DECLARATION = 'dec';
  /**
   * token style for embedded source.
   * @const
   */
  var PR_SOURCE = 'src';
  /**
   * token style for an sgml attribute name.
   * @const
   */
  var PR_ATTRIB_NAME = 'atn';
  /**
   * token style for an sgml attribute value.
   * @const
   */
  var PR_ATTRIB_VALUE = 'atv';

  /**
   * A class that indicates a section of markup that is not code, e.g. to allow
   * embedding of line numbers within code listings.
   * @const
   */
  var PR_NOCODE = 'nocode';


  // Regex pattern below is automatically generated by regexpPrecederPatterns.pl
  // Do not modify, your changes will be erased.

  // CAVEAT: this does not properly handle the case where a regular
  // expression immediately follows another since a regular expression may
  // have flags for case-sensitivity and the like.  Having regexp tokens
  // adjacent is not valid in any language I'm aware of, so I'm punting.
  // TODO: maybe style special characters inside a regexp as punctuation.

  /**
   * A set of tokens that can precede a regular expression literal in
   * javascript
   * http://web.archive.org/web/20070717142515/http://www.mozilla.org/js/language/js20/rationale/syntax.html
   * has the full list, but I've removed ones that might be problematic when
   * seen in languages that don't support regular expression literals.
   *
   * Specifically, I've removed any keywords that can't precede a regexp
   * literal in a syntactically legal javascript program, and I've removed the
   * "in" keyword since it's not a keyword in many languages, and might be used
   * as a count of inches.
   *
   * The link above does not accurately describe EcmaScript rules since
   * it fails to distinguish between (a=++/b/i) and (a++/b/i) but it works
   * very well in practice.
   *
   * @private
   * @const
   */
  var REGEXP_PRECEDER_PATTERN = '(?:^^\\.?|[+-]|[!=]=?=?|\\#|%=?|&&?=?|\\(|\\*=?|[+\\-]=|->|\\/=?|::?|<<?=?|>>?>?=?|,|;|\\?|@|\\[|~|{|\\^\\^?=?|\\|\\|?=?|break|case|continue|delete|do|else|finally|instanceof|return|throw|try|typeof)\\s*';


  /**
   * Given a group of {@link RegExp}s, returns a {@code RegExp} that globally
   * matches the union of the sets of strings matched by the input RegExp.
   * Since it matches globally, if the input strings have a start-of-input
   * anchor (/^.../), it is ignored for the purposes of unioning.
   * @param {Array.<RegExp>} regexs non multiline, non-global regexs.
   * @return {RegExp} a global regex.
   */
  function combinePrefixPatterns(regexs) {
    var capturedGroupIndex = 0;

    var needToFoldCase = false;
    var ignoreCase = false;
    for (var i = 0, n = regexs.length; i < n; ++i) {
      var regex = regexs[i];
      if (regex.ignoreCase) {
        ignoreCase = true;
      } else if (/[a-z]/i.test(regex.source.replace(
                     /\\u[0-9a-f]{4}|\\x[0-9a-f]{2}|\\[^ux]/gi, ''))) {
        needToFoldCase = true;
        ignoreCase = false;
        break;
      }
    }

    var escapeCharToCodeUnit = {
      'b': 8,
      't': 9,
      'n': 0xa,
      'v': 0xb,
      'f': 0xc,
      'r': 0xd
    };

    function decodeEscape(charsetPart) {
      var cc0 = charsetPart.charCodeAt(0);
      if (cc0 !== 92 /* \\ */) {
        return cc0;
      }
      var c1 = charsetPart.charAt(1);
      cc0 = escapeCharToCodeUnit[c1];
      if (cc0) {
        return cc0;
      } else if ('0' <= c1 && c1 <= '7') {
        return parseInt(charsetPart.substring(1), 8);
      } else if (c1 === 'u' || c1 === 'x') {
        return parseInt(charsetPart.substring(2), 16);
      } else {
        return charsetPart.charCodeAt(1);
      }
    }

    function encodeEscape(charCode) {
      if (charCode < 0x20) {
        return (charCode < 0x10 ? '\\x0' : '\\x') + charCode.toString(16);
      }
      var ch = String.fromCharCode(charCode);
      return (ch === '\\' || ch === '-' || ch === ']' || ch === '^')
          ? "\\" + ch : ch;
    }

    function caseFoldCharset(charSet) {
      var charsetParts = charSet.substring(1, charSet.length - 1).match(
          new RegExp(
              '\\\\u[0-9A-Fa-f]{4}'
              + '|\\\\x[0-9A-Fa-f]{2}'
              + '|\\\\[0-3][0-7]{0,2}'
              + '|\\\\[0-7]{1,2}'
              + '|\\\\[\\s\\S]'
              + '|-'
              + '|[^-\\\\]',
              'g'));
      var ranges = [];
      var inverse = charsetParts[0] === '^';

      var out = ['['];
      if (inverse) { out.push('^'); }

      for (var i = inverse ? 1 : 0, n = charsetParts.length; i < n; ++i) {
        var p = charsetParts[i];
        if (/\\[bdsw]/i.test(p)) {  // Don't muck with named groups.
          out.push(p);
        } else {
          var start = decodeEscape(p);
          var end;
          if (i + 2 < n && '-' === charsetParts[i + 1]) {
            end = decodeEscape(charsetParts[i + 2]);
            i += 2;
          } else {
            end = start;
          }
          ranges.push([start, end]);
          // If the range might intersect letters, then expand it.
          // This case handling is too simplistic.
          // It does not deal with non-latin case folding.
          // It works for latin source code identifiers though.
          if (!(end < 65 || start > 122)) {
            if (!(end < 65 || start > 90)) {
              ranges.push([Math.max(65, start) | 32, Math.min(end, 90) | 32]);
            }
            if (!(end < 97 || start > 122)) {
              ranges.push([Math.max(97, start) & ~32, Math.min(end, 122) & ~32]);
            }
          }
        }
      }

      // [[1, 10], [3, 4], [8, 12], [14, 14], [16, 16], [17, 17]]
      // -> [[1, 12], [14, 14], [16, 17]]
      ranges.sort(function (a, b) { return (a[0] - b[0]) || (b[1]  - a[1]); });
      var consolidatedRanges = [];
      var lastRange = [];
      for (var i = 0; i < ranges.length; ++i) {
        var range = ranges[i];
        if (range[0] <= lastRange[1] + 1) {
          lastRange[1] = Math.max(lastRange[1], range[1]);
        } else {
          consolidatedRanges.push(lastRange = range);
        }
      }

      for (var i = 0; i < consolidatedRanges.length; ++i) {
        var range = consolidatedRanges[i];
        out.push(encodeEscape(range[0]));
        if (range[1] > range[0]) {
          if (range[1] + 1 > range[0]) { out.push('-'); }
          out.push(encodeEscape(range[1]));
        }
      }
      out.push(']');
      return out.join('');
    }

    function allowAnywhereFoldCaseAndRenumberGroups(regex) {
      // Split into character sets, escape sequences, punctuation strings
      // like ('(', '(?:', ')', '^'), and runs of characters that do not
      // include any of the above.
      var parts = regex.source.match(
          new RegExp(
              '(?:'
              + '\\[(?:[^\\x5C\\x5D]|\\\\[\\s\\S])*\\]'  // a character set
              + '|\\\\u[A-Fa-f0-9]{4}'  // a unicode escape
              + '|\\\\x[A-Fa-f0-9]{2}'  // a hex escape
              + '|\\\\[0-9]+'  // a back-reference or octal escape
              + '|\\\\[^ux0-9]'  // other escape sequence
              + '|\\(\\?[:!=]'  // start of a non-capturing group
              + '|[\\(\\)\\^]'  // start/end of a group, or line start
              + '|[^\\x5B\\x5C\\(\\)\\^]+'  // run of other characters
              + ')',
              'g'));
      var n = parts.length;

      // Maps captured group numbers to the number they will occupy in
      // the output or to -1 if that has not been determined, or to
      // undefined if they need not be capturing in the output.
      var capturedGroups = [];

      // Walk over and identify back references to build the capturedGroups
      // mapping.
      for (var i = 0, groupIndex = 0; i < n; ++i) {
        var p = parts[i];
        if (p === '(') {
          // groups are 1-indexed, so max group index is count of '('
          ++groupIndex;
        } else if ('\\' === p.charAt(0)) {
          var decimalValue = +p.substring(1);
          if (decimalValue) {
            if (decimalValue <= groupIndex) {
              capturedGroups[decimalValue] = -1;
            } else {
              // Replace with an unambiguous escape sequence so that
              // an octal escape sequence does not turn into a backreference
              // to a capturing group from an earlier regex.
              parts[i] = encodeEscape(decimalValue);
            }
          }
        }
      }

      // Renumber groups and reduce capturing groups to non-capturing groups
      // where possible.
      for (var i = 1; i < capturedGroups.length; ++i) {
        if (-1 === capturedGroups[i]) {
          capturedGroups[i] = ++capturedGroupIndex;
        }
      }
      for (var i = 0, groupIndex = 0; i < n; ++i) {
        var p = parts[i];
        if (p === '(') {
          ++groupIndex;
          if (!capturedGroups[groupIndex]) {
            parts[i] = '(?:';
          }
        } else if ('\\' === p.charAt(0)) {
          var decimalValue = +p.substring(1);
          if (decimalValue && decimalValue <= groupIndex) {
            parts[i] = '\\' + capturedGroups[decimalValue];
          }
        }
      }

      // Remove any prefix anchors so that the output will match anywhere.
      // ^^ really does mean an anchored match though.
      for (var i = 0; i < n; ++i) {
        if ('^' === parts[i] && '^' !== parts[i + 1]) { parts[i] = ''; }
      }

      // Expand letters to groups to handle mixing of case-sensitive and
      // case-insensitive patterns if necessary.
      if (regex.ignoreCase && needToFoldCase) {
        for (var i = 0; i < n; ++i) {
          var p = parts[i];
          var ch0 = p.charAt(0);
          if (p.length >= 2 && ch0 === '[') {
            parts[i] = caseFoldCharset(p);
          } else if (ch0 !== '\\') {
            // TODO: handle letters in numeric escapes.
            parts[i] = p.replace(
                /[a-zA-Z]/g,
                function (ch) {
                  var cc = ch.charCodeAt(0);
                  return '[' + String.fromCharCode(cc & ~32, cc | 32) + ']';
                });
          }
        }
      }

      return parts.join('');
    }

    var rewritten = [];
    for (var i = 0, n = regexs.length; i < n; ++i) {
      var regex = regexs[i];
      if (regex.global || regex.multiline) { throw new Error('' + regex); }
      rewritten.push(
          '(?:' + allowAnywhereFoldCaseAndRenumberGroups(regex) + ')');
    }

    return new RegExp(rewritten.join('|'), ignoreCase ? 'gi' : 'g');
  }


  /**
   * Split markup into a string of source code and an array mapping ranges in
   * that string to the text nodes in which they appear.
   *
   * <p>
   * The HTML DOM structure:</p>
   * <pre>
   * (Element   "p"
   *   (Element "b"
   *     (Text  "print "))       ; #1
   *   (Text    "'Hello '")      ; #2
   *   (Element "br")            ; #3
   *   (Text    "  + 'World';")) ; #4
   * </pre>
   * <p>
   * corresponds to the HTML
   * {@code <p><b>print </b>'Hello '<br>  + 'World';</p>}.</p>
   *
   * <p>
   * It will produce the output:</p>
   * <pre>
   * {
   *   sourceCode: "print 'Hello '\n  + 'World';",
   *   //                     1          2
   *   //           012345678901234 5678901234567
   *   spans: [0, #1, 6, #2, 14, #3, 15, #4]
   * }
   * </pre>
   * <p>
   * where #1 is a reference to the {@code "print "} text node above, and so
   * on for the other text nodes.
   * </p>
   *
   * <p>
   * The {@code} spans array is an array of pairs.  Even elements are the start
   * indices of substrings, and odd elements are the text nodes (or BR elements)
   * that contain the text for those substrings.
   * Substrings continue until the next index or the end of the source.
   * </p>
   *
   * @param {Node} node an HTML DOM subtree containing source-code.
   * @param {boolean|number} isPreformatted truthy if white-space in
   *    text nodes should be considered significant.
   * @return {SourceSpansT} source code and the nodes in which they occur.
   */
  function extractSourceSpans(node, isPreformatted) {
    var nocode = /(?:^|\s)nocode(?:\s|$)/;

    var chunks = [];
    var length = 0;
    var spans = [];
    var k = 0;

    function walk(node) {
      var type = node.nodeType;
      if (type == 1) {  // Element
        if (nocode.test(node.className)) { return; }
        for (var child = node.firstChild; child; child = child.nextSibling) {
          walk(child);
        }
        var nodeName = node.nodeName.toLowerCase();
        if ('br' === nodeName || 'li' === nodeName) {
          chunks[k] = '\n';
          spans[k << 1] = length++;
          spans[(k++ << 1) | 1] = node;
        }
      } else if (type == 3 || type == 4) {  // Text
        var text = node.nodeValue;
        if (text.length) {
          if (!isPreformatted) {
            text = text.replace(/[ \t\r\n]+/g, ' ');
          } else {
            text = text.replace(/\r\n?/g, '\n');  // Normalize newlines.
          }
          // TODO: handle tabs here?
          chunks[k] = text;
          spans[k << 1] = length;
          length += text.length;
          spans[(k++ << 1) | 1] = node;
        }
      }
    }

    walk(node);

    return {
      sourceCode: chunks.join('').replace(/\n$/, ''),
      spans: spans
    };
  }


  /**
   * Apply the given language handler to sourceCode and add the resulting
   * decorations to out.
   * @param {!Element} sourceNode
   * @param {number} basePos the index of sourceCode within the chunk of source
   *    whose decorations are already present on out.
   * @param {string} sourceCode
   * @param {function(JobT)} langHandler
   * @param {DecorationsT} out
   */
  function appendDecorations(
      sourceNode, basePos, sourceCode, langHandler, out) {
    if (!sourceCode) { return; }
    /** @type {JobT} */
    var job = {
      sourceNode: sourceNode,
      pre: 1,
      langExtension: null,
      numberLines: null,
      sourceCode: sourceCode,
      spans: null,
      basePos: basePos,
      decorations: null
    };
    langHandler(job);
    out.push.apply(out, job.decorations);
  }

  var notWs = /\S/;

  /**
   * Given an element, if it contains only one child element and any text nodes
   * it contains contain only space characters, return the sole child element.
   * Otherwise returns undefined.
   * <p>
   * This is meant to return the CODE element in {@code <pre><code ...>} when
   * there is a single child element that contains all the non-space textual
   * content, but not to return anything where there are multiple child elements
   * as in {@code <pre><code>...</code><code>...</code></pre>} or when there
   * is textual content.
   */
  function childContentWrapper(element) {
    var wrapper = undefined;
    for (var c = element.firstChild; c; c = c.nextSibling) {
      var type = c.nodeType;
      wrapper = (type === 1)  // Element Node
          ? (wrapper ? element : c)
          : (type === 3)  // Text Node
          ? (notWs.test(c.nodeValue) ? element : wrapper)
          : wrapper;
    }
    return wrapper === element ? undefined : wrapper;
  }

  /** Given triples of [style, pattern, context] returns a lexing function,
    * The lexing function interprets the patterns to find token boundaries and
    * returns a decoration list of the form
    * [index_0, style_0, index_1, style_1, ..., index_n, style_n]
    * where index_n is an index into the sourceCode, and style_n is a style
    * constant like PR_PLAIN.  index_n-1 <= index_n, and style_n-1 applies to
    * all characters in sourceCode[index_n-1:index_n].
    *
    * The stylePatterns is a list whose elements have the form
    * [style : string, pattern : RegExp, DEPRECATED, shortcut : string].
    *
    * Style is a style constant like PR_PLAIN, or can be a string of the
    * form 'lang-FOO', where FOO is a language extension describing the
    * language of the portion of the token in $1 after pattern executes.
    * E.g., if style is 'lang-lisp', and group 1 contains the text
    * '(hello (world))', then that portion of the token will be passed to the
    * registered lisp handler for formatting.
    * The text before and after group 1 will be restyled using this decorator
    * so decorators should take care that this doesn't result in infinite
    * recursion.  For example, the HTML lexer rule for SCRIPT elements looks
    * something like ['lang-js', /<[s]cript>(.+?)<\/script>/].  This may match
    * '<script>foo()<\/script>', which would cause the current decorator to
    * be called with '<script>' which would not match the same rule since
    * group 1 must not be empty, so it would be instead styled as PR_TAG by
    * the generic tag rule.  The handler registered for the 'js' extension would
    * then be called with 'foo()', and finally, the current decorator would
    * be called with '<\/script>' which would not match the original rule and
    * so the generic tag rule would identify it as a tag.
    *
    * Pattern must only match prefixes, and if it matches a prefix, then that
    * match is considered a token with the same style.
    *
    * Context is applied to the last non-whitespace, non-comment token
    * recognized.
    *
    * Shortcut is an optional string of characters, any of which, if the first
    * character, gurantee that this pattern and only this pattern matches.
    *
    * @param {Array} shortcutStylePatterns patterns that always start with
    *   a known character.  Must have a shortcut string.
    * @param {Array} fallthroughStylePatterns patterns that will be tried in
    *   order if the shortcut ones fail.  May have shortcuts.
    *
    * @return {function (JobT)} a function that takes an undecorated job and
    *   attaches a list of decorations.
    */
  function createSimpleLexer(shortcutStylePatterns, fallthroughStylePatterns) {
    var shortcuts = {};
    var tokenizer;
    (function () {
      var allPatterns = shortcutStylePatterns.concat(fallthroughStylePatterns);
      var allRegexs = [];
      var regexKeys = {};
      for (var i = 0, n = allPatterns.length; i < n; ++i) {
        var patternParts = allPatterns[i];
        var shortcutChars = patternParts[3];
        if (shortcutChars) {
          for (var c = shortcutChars.length; --c >= 0;) {
            shortcuts[shortcutChars.charAt(c)] = patternParts;
          }
        }
        var regex = patternParts[1];
        var k = '' + regex;
        if (!regexKeys.hasOwnProperty(k)) {
          allRegexs.push(regex);
          regexKeys[k] = null;
        }
      }
      allRegexs.push(/[\0-\uffff]/);
      tokenizer = combinePrefixPatterns(allRegexs);
    })();

    var nPatterns = fallthroughStylePatterns.length;

    /**
     * Lexes job.sourceCode and attaches an output array job.decorations of
     * style classes preceded by the position at which they start in
     * job.sourceCode in order.
     *
     * @type{function (JobT)}
     */
    var decorate = function (job) {
      var sourceCode = job.sourceCode, basePos = job.basePos;
      var sourceNode = job.sourceNode;
      /** Even entries are positions in source in ascending order.  Odd enties
        * are style markers (e.g., PR_COMMENT) that run from that position until
        * the end.
        * @type {DecorationsT}
        */
      var decorations = [basePos, PR_PLAIN];
      var pos = 0;  // index into sourceCode
      var tokens = sourceCode.match(tokenizer) || [];
      var styleCache = {};

      for (var ti = 0, nTokens = tokens.length; ti < nTokens; ++ti) {
        var token = tokens[ti];
        var style = styleCache[token];
        var match = void 0;

        var isEmbedded;
        if (typeof style === 'string') {
          isEmbedded = false;
        } else {
          var patternParts = shortcuts[token.charAt(0)];
          if (patternParts) {
            match = token.match(patternParts[1]);
            style = patternParts[0];
          } else {
            for (var i = 0; i < nPatterns; ++i) {
              patternParts = fallthroughStylePatterns[i];
              match = token.match(patternParts[1]);
              if (match) {
                style = patternParts[0];
                break;
              }
            }

            if (!match) {  // make sure that we make progress
              style = PR_PLAIN;
            }
          }

          isEmbedded = style.length >= 5 && 'lang-' === style.substring(0, 5);
          if (isEmbedded && !(match && typeof match[1] === 'string')) {
            isEmbedded = false;
            style = PR_SOURCE;
          }

          if (!isEmbedded) { styleCache[token] = style; }
        }

        var tokenStart = pos;
        pos += token.length;

        if (!isEmbedded) {
          decorations.push(basePos + tokenStart, style);
        } else {  // Treat group 1 as an embedded block of source code.
          var embeddedSource = match[1];
          var embeddedSourceStart = token.indexOf(embeddedSource);
          var embeddedSourceEnd = embeddedSourceStart + embeddedSource.length;
          if (match[2]) {
            // If embeddedSource can be blank, then it would match at the
            // beginning which would cause us to infinitely recurse on the
            // entire token, so we catch the right context in match[2].
            embeddedSourceEnd = token.length - match[2].length;
            embeddedSourceStart = embeddedSourceEnd - embeddedSource.length;
          }
          var lang = style.substring(5);
          // Decorate the left of the embedded source
          appendDecorations(
              sourceNode,
              basePos + tokenStart,
              token.substring(0, embeddedSourceStart),
              decorate, decorations);
          // Decorate the embedded source
          appendDecorations(
              sourceNode,
              basePos + tokenStart + embeddedSourceStart,
              embeddedSource,
              langHandlerForExtension(lang, embeddedSource),
              decorations);
          // Decorate the right of the embedded section
          appendDecorations(
              sourceNode,
              basePos + tokenStart + embeddedSourceEnd,
              token.substring(embeddedSourceEnd),
              decorate, decorations);
        }
      }
      job.decorations = decorations;
    };
    return decorate;
  }

  /** returns a function that produces a list of decorations from source text.
    *
    * This code treats ", ', and ` as string delimiters, and \ as a string
    * escape.  It does not recognize perl's qq() style strings.
    * It has no special handling for double delimiter escapes as in basic, or
    * the tripled delimiters used in python, but should work on those regardless
    * although in those cases a single string literal may be broken up into
    * multiple adjacent string literals.
    *
    * It recognizes C, C++, and shell style comments.
    *
    * @param {Object} options a set of optional parameters.
    * @return {function (JobT)} a function that examines the source code
    *     in the input job and builds a decoration list which it attaches to
    *     the job.
    */
  function sourceDecorator(options) {
    var shortcutStylePatterns = [], fallthroughStylePatterns = [];
    if (options['tripleQuotedStrings']) {
      // '''multi-line-string''', 'single-line-string', and double-quoted
      shortcutStylePatterns.push(
          [PR_STRING,  /^(?:\'\'\'(?:[^\'\\]|\\[\s\S]|\'{1,2}(?=[^\']))*(?:\'\'\'|$)|\"\"\"(?:[^\"\\]|\\[\s\S]|\"{1,2}(?=[^\"]))*(?:\"\"\"|$)|\'(?:[^\\\']|\\[\s\S])*(?:\'|$)|\"(?:[^\\\"]|\\[\s\S])*(?:\"|$))/,
           null, '\'"']);
    } else if (options['multiLineStrings']) {
      // 'multi-line-string', "multi-line-string"
      shortcutStylePatterns.push(
          [PR_STRING,  /^(?:\'(?:[^\\\']|\\[\s\S])*(?:\'|$)|\"(?:[^\\\"]|\\[\s\S])*(?:\"|$)|\`(?:[^\\\`]|\\[\s\S])*(?:\`|$))/,
           null, '\'"`']);
    } else {
      // 'single-line-string', "single-line-string"
      shortcutStylePatterns.push(
          [PR_STRING,
           /^(?:\'(?:[^\\\'\r\n]|\\.)*(?:\'|$)|\"(?:[^\\\"\r\n]|\\.)*(?:\"|$))/,
           null, '"\'']);
    }
    if (options['verbatimStrings']) {
      // verbatim-string-literal production from the C# grammar.  See issue 93.
      fallthroughStylePatterns.push(
          [PR_STRING, /^@\"(?:[^\"]|\"\")*(?:\"|$)/, null]);
    }
    var hc = options['hashComments'];
    if (hc) {
      if (options['cStyleComments']) {
        if (hc > 1) {  // multiline hash comments
          shortcutStylePatterns.push(
              [PR_COMMENT, /^#(?:##(?:[^#]|#(?!##))*(?:###|$)|.*)/, null, '#']);
        } else {
          // Stop C preprocessor declarations at an unclosed open comment
          shortcutStylePatterns.push(
              [PR_COMMENT, /^#(?:(?:define|e(?:l|nd)if|else|error|ifn?def|include|line|pragma|undef|warning)\b|[^\r\n]*)/,
               null, '#']);
        }
        // #include <stdio.h>
        fallthroughStylePatterns.push(
            [PR_STRING,
             /^<(?:(?:(?:\.\.\/)*|\/?)(?:[\w-]+(?:\/[\w-]+)+)?[\w-]+\.h(?:h|pp|\+\+)?|[a-z]\w*)>/,
             null]);
      } else {
        shortcutStylePatterns.push([PR_COMMENT, /^#[^\r\n]*/, null, '#']);
      }
    }
    if (options['cStyleComments']) {
      fallthroughStylePatterns.push([PR_COMMENT, /^\/\/[^\r\n]*/, null]);
      fallthroughStylePatterns.push(
          [PR_COMMENT, /^\/\*[\s\S]*?(?:\*\/|$)/, null]);
    }
    var regexLiterals = options['regexLiterals'];
    if (regexLiterals) {
      /**
       * @const
       */
      var regexExcls = regexLiterals > 1
        ? ''  // Multiline regex literals
        : '\n\r';
      /**
       * @const
       */
      var regexAny = regexExcls ? '.' : '[\\S\\s]';
      /**
       * @const
       */
      var REGEX_LITERAL = (
          // A regular expression literal starts with a slash that is
          // not followed by * or / so that it is not confused with
          // comments.
          '/(?=[^/*' + regexExcls + '])'
          // and then contains any number of raw characters,
          + '(?:[^/\\x5B\\x5C' + regexExcls + ']'
          // escape sequences (\x5C),
          +    '|\\x5C' + regexAny
          // or non-nesting character sets (\x5B\x5D);
          +    '|\\x5B(?:[^\\x5C\\x5D' + regexExcls + ']'
          +             '|\\x5C' + regexAny + ')*(?:\\x5D|$))+'
          // finally closed by a /.
          + '/');
      fallthroughStylePatterns.push(
          ['lang-regex',
           RegExp('^' + REGEXP_PRECEDER_PATTERN + '(' + REGEX_LITERAL + ')')
           ]);
    }

    var types = options['types'];
    if (types) {
      fallthroughStylePatterns.push([PR_TYPE, types]);
    }

    var keywords = ("" + options['keywords']).replace(/^ | $/g, '');
    if (keywords.length) {
      fallthroughStylePatterns.push(
          [PR_KEYWORD,
           new RegExp('^(?:' + keywords.replace(/[\s,]+/g, '|') + ')\\b'),
           null]);
    }

    shortcutStylePatterns.push([PR_PLAIN,       /^\s+/, null, ' \r\n\t\xA0']);

    var punctuation =
      // The Bash man page says

      // A word is a sequence of characters considered as a single
      // unit by GRUB. Words are separated by metacharacters,
      // which are the following plus space, tab, and newline: { }
      // | & $ ; < >
      // ...

      // A word beginning with # causes that word and all remaining
      // characters on that line to be ignored.

      // which means that only a '#' after /(?:^|[{}|&$;<>\s])/ starts a
      // comment but empirically
      // $ echo {#}
      // {#}
      // $ echo \$#
      // $#
      // $ echo }#
      // }#

      // so /(?:^|[|&;<>\s])/ is more appropriate.

      // http://gcc.gnu.org/onlinedocs/gcc-2.95.3/cpp_1.html#SEC3
      // suggests that this definition is compatible with a
      // default mode that tries to use a single token definition
      // to recognize both bash/python style comments and C
      // preprocessor directives.

      // This definition of punctuation does not include # in the list of
      // follow-on exclusions, so # will not be broken before if preceeded
      // by a punctuation character.  We could try to exclude # after
      // [|&;<>] but that doesn't seem to cause many major problems.
      // If that does turn out to be a problem, we should change the below
      // when hc is truthy to include # in the run of punctuation characters
      // only when not followint [|&;<>].
      '^.[^\\s\\w.$@\'"`/\\\\]*';
    if (options['regexLiterals']) {
      punctuation += '(?!\s*\/)';
    }

    fallthroughStylePatterns.push(
        // TODO(mikesamuel): recognize non-latin letters and numerals in idents
        [PR_LITERAL,     /^@[a-z_$][a-z_$@0-9]*/i, null],
        [PR_TYPE,        /^(?:[@_]?[A-Z]+[a-z][A-Za-z_$@0-9]*|\w+_t\b)/, null],
        [PR_PLAIN,       /^[a-z_$][a-z_$@0-9]*/i, null],
        [PR_LITERAL,
         new RegExp(
             '^(?:'
             // A hex number
             + '0x[a-f0-9]+'
             // or an octal or decimal number,
             + '|(?:\\d(?:_\\d+)*\\d*(?:\\.\\d*)?|\\.\\d\\+)'
             // possibly in scientific notation
             + '(?:e[+\\-]?\\d+)?'
             + ')'
             // with an optional modifier like UL for unsigned long
             + '[a-z]*', 'i'),
         null, '0123456789'],
        // Don't treat escaped quotes in bash as starting strings.
        // See issue 144.
        [PR_PLAIN,       /^\\[\s\S]?/, null],
        [PR_PUNCTUATION, new RegExp(punctuation), null]);

    return createSimpleLexer(shortcutStylePatterns, fallthroughStylePatterns);
  }

  var decorateSource = sourceDecorator({
        'keywords': ALL_KEYWORDS,
        'hashComments': true,
        'cStyleComments': true,
        'multiLineStrings': true,
        'regexLiterals': true
      });

  /**
   * Given a DOM subtree, wraps it in a list, and puts each line into its own
   * list item.
   *
   * @param {Node} node modified in place.  Its content is pulled into an
   *     HTMLOListElement, and each line is moved into a separate list item.
   *     This requires cloning elements, so the input might not have unique
   *     IDs after numbering.
   * @param {number|null|boolean} startLineNum
   *     If truthy, coerced to an integer which is the 1-indexed line number
   *     of the first line of code.  The number of the first line will be
   *     attached to the list.
   * @param {boolean} isPreformatted true iff white-space in text nodes should
   *     be treated as significant.
   */
  function numberLines(node, startLineNum, isPreformatted) {
    var nocode = /(?:^|\s)nocode(?:\s|$)/;
    var lineBreak = /\r\n?|\n/;

    var document = node.ownerDocument;

    var li = document.createElement('li');
    while (node.firstChild) {
      li.appendChild(node.firstChild);
    }
    // An array of lines.  We split below, so this is initialized to one
    // un-split line.
    var listItems = [li];

    function walk(node) {
      var type = node.nodeType;
      if (type == 1 && !nocode.test(node.className)) {  // Element
        if ('br' === node.nodeName.toLowerCase()) {
          breakAfter(node);
          // Discard the <BR> since it is now flush against a </LI>.
          if (node.parentNode) {
            node.parentNode.removeChild(node);
          }
        } else {
          for (var child = node.firstChild; child; child = child.nextSibling) {
            walk(child);
          }
        }
      } else if ((type == 3 || type == 4) && isPreformatted) {  // Text
        var text = node.nodeValue;
        var match = text.match(lineBreak);
        if (match) {
          var firstLine = text.substring(0, match.index);
          node.nodeValue = firstLine;
          var tail = text.substring(match.index + match[0].length);
          if (tail) {
            var parent = node.parentNode;
            parent.insertBefore(
              document.createTextNode(tail), node.nextSibling);
          }
          breakAfter(node);
          if (!firstLine) {
            // Don't leave blank text nodes in the DOM.
            node.parentNode.removeChild(node);
          }
        }
      }
    }

    // Split a line after the given node.
    function breakAfter(lineEndNode) {
      // If there's nothing to the right, then we can skip ending the line
      // here, and move root-wards since splitting just before an end-tag
      // would require us to create a bunch of empty copies.
      while (!lineEndNode.nextSibling) {
        lineEndNode = lineEndNode.parentNode;
        if (!lineEndNode) { return; }
      }

      function breakLeftOf(limit, copy) {
        // Clone shallowly if this node needs to be on both sides of the break.
        var rightSide = copy ? limit.cloneNode(false) : limit;
        var parent = limit.parentNode;
        if (parent) {
          // We clone the parent chain.
          // This helps us resurrect important styling elements that cross lines.
          // E.g. in <i>Foo<br>Bar</i>
          // should be rewritten to <li><i>Foo</i></li><li><i>Bar</i></li>.
          var parentClone = breakLeftOf(parent, 1);
          // Move the clone and everything to the right of the original
          // onto the cloned parent.
          var next = limit.nextSibling;
          parentClone.appendChild(rightSide);
          for (var sibling = next; sibling; sibling = next) {
            next = sibling.nextSibling;
            parentClone.appendChild(sibling);
          }
        }
        return rightSide;
      }

      var copiedListItem = breakLeftOf(lineEndNode.nextSibling, 0);

      // Walk the parent chain until we reach an unattached LI.
      for (var parent;
           // Check nodeType since IE invents document fragments.
           (parent = copiedListItem.parentNode) && parent.nodeType === 1;) {
        copiedListItem = parent;
      }
      // Put it on the list of lines for later processing.
      listItems.push(copiedListItem);
    }

    // Split lines while there are lines left to split.
    for (var i = 0;  // Number of lines that have been split so far.
         i < listItems.length;  // length updated by breakAfter calls.
         ++i) {
      walk(listItems[i]);
    }

    // Make sure numeric indices show correctly.
    if (startLineNum === (startLineNum|0)) {
      listItems[0].setAttribute('value', startLineNum);
    }

    var ol = document.createElement('ol');
    ol.className = 'linenums';
    var offset = Math.max(0, ((startLineNum - 1 /* zero index */)) | 0) || 0;
    for (var i = 0, n = listItems.length; i < n; ++i) {
      li = listItems[i];
      // Stick a class on the LIs so that stylesheets can
      // color odd/even rows, or any other row pattern that
      // is co-prime with 10.
      li.className = 'L' + ((i + offset) % 10);
      if (!li.firstChild) {
        li.appendChild(document.createTextNode('\xA0'));
      }
      ol.appendChild(li);
    }

    node.appendChild(ol);
  }


  /**
   * Breaks {@code job.sourceCode} around style boundaries in
   * {@code job.decorations} and modifies {@code job.sourceNode} in place.
   * @param {JobT} job
   * @private
   */
  function recombineTagsAndDecorations(job) {
    var isIE8OrEarlier = /\bMSIE\s(\d+)/.exec(navigator.userAgent);
    isIE8OrEarlier = isIE8OrEarlier && +isIE8OrEarlier[1] <= 8;
    var newlineRe = /\n/g;

    var source = job.sourceCode;
    var sourceLength = source.length;
    // Index into source after the last code-unit recombined.
    var sourceIndex = 0;

    var spans = job.spans;
    var nSpans = spans.length;
    // Index into spans after the last span which ends at or before sourceIndex.
    var spanIndex = 0;

    var decorations = job.decorations;
    var nDecorations = decorations.length;
    // Index into decorations after the last decoration which ends at or before
    // sourceIndex.
    var decorationIndex = 0;

    // Remove all zero-length decorations.
    decorations[nDecorations] = sourceLength;
    var decPos, i;
    for (i = decPos = 0; i < nDecorations;) {
      if (decorations[i] !== decorations[i + 2]) {
        decorations[decPos++] = decorations[i++];
        decorations[decPos++] = decorations[i++];
      } else {
        i += 2;
      }
    }
    nDecorations = decPos;

    // Simplify decorations.
    for (i = decPos = 0; i < nDecorations;) {
      var startPos = decorations[i];
      // Conflate all adjacent decorations that use the same style.
      var startDec = decorations[i + 1];
      var end = i + 2;
      while (end + 2 <= nDecorations && decorations[end + 1] === startDec) {
        end += 2;
      }
      decorations[decPos++] = startPos;
      decorations[decPos++] = startDec;
      i = end;
    }

    nDecorations = decorations.length = decPos;

    var sourceNode = job.sourceNode;
    var oldDisplay = "";
    if (sourceNode) {
      oldDisplay = sourceNode.style.display;
      sourceNode.style.display = 'none';
    }
    try {
      var decoration = null;
      while (spanIndex < nSpans) {
        var spanStart = spans[spanIndex];
        var spanEnd = /** @type{number} */ (spans[spanIndex + 2])
            || sourceLength;

        var decEnd = decorations[decorationIndex + 2] || sourceLength;

        var end = Math.min(spanEnd, decEnd);

        var textNode = /** @type{Node} */ (spans[spanIndex + 1]);
        var styledText;
        if (textNode.nodeType !== 1  // Don't muck with <BR>s or <LI>s
            // Don't introduce spans around empty text nodes.
            && (styledText = source.substring(sourceIndex, end))) {
          // This may seem bizarre, and it is.  Emitting LF on IE causes the
          // code to display with spaces instead of line breaks.
          // Emitting Windows standard issue linebreaks (CRLF) causes a blank
          // space to appear at the beginning of every line but the first.
          // Emitting an old Mac OS 9 line separator makes everything spiffy.
          if (isIE8OrEarlier) {
            styledText = styledText.replace(newlineRe, '\r');
          }
          textNode.nodeValue = styledText;
          var document = textNode.ownerDocument;
          var span = document.createElement('span');
          span.className = decorations[decorationIndex + 1];
          var parentNode = textNode.parentNode;
          parentNode.replaceChild(span, textNode);
          span.appendChild(textNode);
          if (sourceIndex < spanEnd) {  // Split off a text node.
            spans[spanIndex + 1] = textNode
                // TODO: Possibly optimize by using '' if there's no flicker.
                = document.createTextNode(source.substring(end, spanEnd));
            parentNode.insertBefore(textNode, span.nextSibling);
          }
        }

        sourceIndex = end;

        if (sourceIndex >= spanEnd) {
          spanIndex += 2;
        }
        if (sourceIndex >= decEnd) {
          decorationIndex += 2;
        }
      }
    } finally {
      if (sourceNode) {
        sourceNode.style.display = oldDisplay;
      }
    }
  }


  /** Maps language-specific file extensions to handlers. */
  var langHandlerRegistry = {};
  /** Register a language handler for the given file extensions.
    * @param {function (JobT)} handler a function from source code to a list
    *      of decorations.  Takes a single argument job which describes the
    *      state of the computation and attaches the decorations to it.
    * @param {Array.<string>} fileExtensions
    */
  function registerLangHandler(handler, fileExtensions) {
    for (var i = fileExtensions.length; --i >= 0;) {
      var ext = fileExtensions[i];
      if (!langHandlerRegistry.hasOwnProperty(ext)) {
        langHandlerRegistry[ext] = handler;
      } else if (win['console']) {
        console['warn']('cannot override language handler %s', ext);
      }
    }
  }
  function langHandlerForExtension(extension, source) {
    if (!(extension && langHandlerRegistry.hasOwnProperty(extension))) {
      // Treat it as markup if the first non whitespace character is a < and
      // the last non-whitespace character is a >.
      extension = /^\s*</.test(source)
          ? 'default-markup'
          : 'default-code';
    }
    return langHandlerRegistry[extension];
  }
  registerLangHandler(decorateSource, ['default-code']);
  registerLangHandler(
      createSimpleLexer(
          [],
          [
           [PR_PLAIN,       /^[^<?]+/],
           [PR_DECLARATION, /^<!\w[^>]*(?:>|$)/],
           [PR_COMMENT,     /^<\!--[\s\S]*?(?:-\->|$)/],
           // Unescaped content in an unknown language
           ['lang-',        /^<\?([\s\S]+?)(?:\?>|$)/],
           ['lang-',        /^<%([\s\S]+?)(?:%>|$)/],
           [PR_PUNCTUATION, /^(?:<[%?]|[%?]>)/],
           ['lang-',        /^<xmp\b[^>]*>([\s\S]+?)<\/xmp\b[^>]*>/i],
           // Unescaped content in javascript.  (Or possibly vbscript).
           ['lang-js',      /^<script\b[^>]*>([\s\S]*?)(<\/script\b[^>]*>)/i],
           // Contains unescaped stylesheet content
           ['lang-css',     /^<style\b[^>]*>([\s\S]*?)(<\/style\b[^>]*>)/i],
           ['lang-in.tag',  /^(<\/?[a-z][^<>]*>)/i]
          ]),
      ['default-markup', 'htm', 'html', 'mxml', 'xhtml', 'xml', 'xsl']);
  registerLangHandler(
      createSimpleLexer(
          [
           [PR_PLAIN,        /^[\s]+/, null, ' \t\r\n'],
           [PR_ATTRIB_VALUE, /^(?:\"[^\"]*\"?|\'[^\']*\'?)/, null, '\"\'']
           ],
          [
           [PR_TAG,          /^^<\/?[a-z](?:[\w.:-]*\w)?|\/?>$/i],
           [PR_ATTRIB_NAME,  /^(?!style[\s=]|on)[a-z](?:[\w:-]*\w)?/i],
           ['lang-uq.val',   /^=\s*([^>\'\"\s]*(?:[^>\'\"\s\/]|\/(?=\s)))/],
           [PR_PUNCTUATION,  /^[=<>\/]+/],
           ['lang-js',       /^on\w+\s*=\s*\"([^\"]+)\"/i],
           ['lang-js',       /^on\w+\s*=\s*\'([^\']+)\'/i],
           ['lang-js',       /^on\w+\s*=\s*([^\"\'>\s]+)/i],
           ['lang-css',      /^style\s*=\s*\"([^\"]+)\"/i],
           ['lang-css',      /^style\s*=\s*\'([^\']+)\'/i],
           ['lang-css',      /^style\s*=\s*([^\"\'>\s]+)/i]
           ]),
      ['in.tag']);
  registerLangHandler(
      createSimpleLexer([], [[PR_ATTRIB_VALUE, /^[\s\S]+/]]), ['uq.val']);
  registerLangHandler(sourceDecorator({
          'keywords': CPP_KEYWORDS,
          'hashComments': true,
          'cStyleComments': true,
          'types': C_TYPES
        }), ['c', 'cc', 'cpp', 'cxx', 'cyc', 'm']);
  registerLangHandler(sourceDecorator({
          'keywords': 'null,true,false'
        }), ['json']);
  registerLangHandler(sourceDecorator({
          'keywords': CSHARP_KEYWORDS,
          'hashComments': true,
          'cStyleComments': true,
          'verbatimStrings': true,
          'types': C_TYPES
        }), ['cs']);
  registerLangHandler(sourceDecorator({
          'keywords': JAVA_KEYWORDS,
          'cStyleComments': true
        }), ['java']);
  registerLangHandler(sourceDecorator({
          'keywords': SH_KEYWORDS,
          'hashComments': true,
          'multiLineStrings': true
        }), ['bash', 'bsh', 'csh', 'sh']);
  registerLangHandler(sourceDecorator({
          'keywords': PYTHON_KEYWORDS,
          'hashComments': true,
          'multiLineStrings': true,
          'tripleQuotedStrings': true
        }), ['cv', 'py', 'python']);
  registerLangHandler(sourceDecorator({
          'keywords': PERL_KEYWORDS,
          'hashComments': true,
          'multiLineStrings': true,
          'regexLiterals': 2  // multiline regex literals
        }), ['perl', 'pl', 'pm']);
  registerLangHandler(sourceDecorator({
          'keywords': RUBY_KEYWORDS,
          'hashComments': true,
          'multiLineStrings': true,
          'regexLiterals': true
        }), ['rb', 'ruby']);
  registerLangHandler(sourceDecorator({
          'keywords': JSCRIPT_KEYWORDS,
          'cStyleComments': true,
          'regexLiterals': true
        }), ['javascript', 'js', 'ts', 'typescript']);
  registerLangHandler(sourceDecorator({
          'keywords': COFFEE_KEYWORDS,
          'hashComments': 3,  // ### style block comments
          'cStyleComments': true,
          'multilineStrings': true,
          'tripleQuotedStrings': true,
          'regexLiterals': true
        }), ['coffee']);
  registerLangHandler(
      createSimpleLexer([], [[PR_STRING, /^[\s\S]+/]]), ['regex']);

  /** @param {JobT} job */
  function applyDecorator(job) {
    var opt_langExtension = job.langExtension;

    try {
      // Extract tags, and convert the source code to plain text.
      var sourceAndSpans = extractSourceSpans(job.sourceNode, job.pre);
      /** Plain text. @type {string} */
      var source = sourceAndSpans.sourceCode;
      job.sourceCode = source;
      job.spans = sourceAndSpans.spans;
      job.basePos = 0;

      // Apply the appropriate language handler
      langHandlerForExtension(opt_langExtension, source)(job);

      // Integrate the decorations and tags back into the source code,
      // modifying the sourceNode in place.
      recombineTagsAndDecorations(job);
    } catch (e) {
      if (win['console']) {
        console['log'](e && e['stack'] || e);
      }
    }
  }

  /**
   * Pretty print a chunk of code.
   * @param sourceCodeHtml {string} The HTML to pretty print.
   * @param opt_langExtension {string} The language name to use.
   *     Typically, a filename extension like 'cpp' or 'java'.
   * @param opt_numberLines {number|boolean} True to number lines,
   *     or the 1-indexed number of the first line in sourceCodeHtml.
   */
  function $prettyPrintOne(sourceCodeHtml, opt_langExtension, opt_numberLines) {
    /** @type{number|boolean} */
    var nl = opt_numberLines || false;
    /** @type{string|null} */
    var langExtension = opt_langExtension || null;
    /** @type{!Element} */
    var container = document.createElement('div');
    // This could cause images to load and onload listeners to fire.
    // E.g. <img onerror="alert(1337)" src="nosuchimage.png">.
    // We assume that the inner HTML is from a trusted source.
    // The pre-tag is required for IE8 which strips newlines from innerHTML
    // when it is injected into a <pre> tag.
    // http://stackoverflow.com/questions/451486/pre-tag-loses-line-breaks-when-setting-innerhtml-in-ie
    // http://stackoverflow.com/questions/195363/inserting-a-newline-into-a-pre-tag-ie-javascript
    container.innerHTML = '<pre>' + sourceCodeHtml + '</pre>';
    container = /** @type{!Element} */(container.firstChild);
    if (nl) {
      numberLines(container, nl, true);
    }

    /** @type{JobT} */
    var job = {
      langExtension: langExtension,
      numberLines: nl,
      sourceNode: container,
      pre: 1,
      sourceCode: null,
      basePos: null,
      spans: null,
      decorations: null
    };
    applyDecorator(job);
    return container.innerHTML;
  }

   /**
    * Find all the {@code <pre>} and {@code <code>} tags in the DOM with
    * {@code class=prettyprint} and prettify them.
    *
    * @param {Function} opt_whenDone called when prettifying is done.
    * @param {HTMLElement|HTMLDocument} opt_root an element or document
    *   containing all the elements to pretty print.
    *   Defaults to {@code document.body}.
    */
  function $prettyPrint(opt_whenDone, opt_root) {
    var root = opt_root || document.body;
    var doc = root.ownerDocument || document;
    function byTagName(tn) { return root.getElementsByTagName(tn); }
    // fetch a list of nodes to rewrite
    var codeSegments = [byTagName('pre'), byTagName('code'), byTagName('xmp')];
    var elements = [];
    for (var i = 0; i < codeSegments.length; ++i) {
      for (var j = 0, n = codeSegments[i].length; j < n; ++j) {
        elements.push(codeSegments[i][j]);
      }
    }
    codeSegments = null;

    var clock = Date;
    if (!clock['now']) {
      clock = { 'now': function () { return +(new Date); } };
    }

    // The loop is broken into a series of continuations to make sure that we
    // don't make the browser unresponsive when rewriting a large page.
    var k = 0;

    var langExtensionRe = /\blang(?:uage)?-([\w.]+)(?!\S)/;
    var prettyPrintRe = /\bprettyprint\b/;
    var prettyPrintedRe = /\bprettyprinted\b/;
    var preformattedTagNameRe = /pre|xmp/i;
    var codeRe = /^code$/i;
    var preCodeXmpRe = /^(?:pre|code|xmp)$/i;
    var EMPTY = {};

    function doWork() {
      var endTime = (win['PR_SHOULD_USE_CONTINUATION'] ?
                     clock['now']() + 250 /* ms */ :
                     Infinity);
      for (; k < elements.length && clock['now']() < endTime; k++) {
        var cs = elements[k];

        // Look for a preceding comment like
        // <?prettify lang="..." linenums="..."?>
        var attrs = EMPTY;
        {
          for (var preceder = cs; (preceder = preceder.previousSibling);) {
            var nt = preceder.nodeType;
            // <?foo?> is parsed by HTML 5 to a comment node (8)
            // like <!--?foo?-->, but in XML is a processing instruction
            var value = (nt === 7 || nt === 8) && preceder.nodeValue;
            if (value
                ? !/^\??prettify\b/.test(value)
                : (nt !== 3 || /\S/.test(preceder.nodeValue))) {
              // Skip over white-space text nodes but not others.
              break;
            }
            if (value) {
              attrs = {};
              value.replace(
                  /\b(\w+)=([\w:.%+-]+)/g,
                function (_, name, value) { attrs[name] = value; });
              break;
            }
          }
        }

        var className = cs.className;
        if ((attrs !== EMPTY || prettyPrintRe.test(className))
            // Don't redo this if we've already done it.
            // This allows recalling pretty print to just prettyprint elements
            // that have been added to the page since last call.
            && !prettyPrintedRe.test(className)) {

          // make sure this is not nested in an already prettified element
          var nested = false;
          for (var p = cs.parentNode; p; p = p.parentNode) {
            var tn = p.tagName;
            if (preCodeXmpRe.test(tn)
                && p.className && prettyPrintRe.test(p.className)) {
              nested = true;
              break;
            }
          }
          if (!nested) {
            // Mark done.  If we fail to prettyprint for whatever reason,
            // we shouldn't try again.
            cs.className += ' prettyprinted';

            // If the classes includes a language extensions, use it.
            // Language extensions can be specified like
            //     <pre class="prettyprint lang-cpp">
            // the language extension "cpp" is used to find a language handler
            // as passed to PR.registerLangHandler.
            // HTML5 recommends that a language be specified using "language-"
            // as the prefix instead.  Google Code Prettify supports both.
            // http://dev.w3.org/html5/spec-author-view/the-code-element.html
            var langExtension = attrs['lang'];
            if (!langExtension) {
              langExtension = className.match(langExtensionRe);
              // Support <pre class="prettyprint"><code class="language-c">
              var wrapper;
              if (!langExtension && (wrapper = childContentWrapper(cs))
                  && codeRe.test(wrapper.tagName)) {
                langExtension = wrapper.className.match(langExtensionRe);
              }

              if (langExtension) { langExtension = langExtension[1]; }
            }

            var preformatted;
            if (preformattedTagNameRe.test(cs.tagName)) {
              preformatted = 1;
            } else {
              var currentStyle = cs['currentStyle'];
              var defaultView = doc.defaultView;
              var whitespace = (
                  currentStyle
                  ? currentStyle['whiteSpace']
                  : (defaultView
                     && defaultView.getComputedStyle)
                  ? defaultView.getComputedStyle(cs, null)
                  .getPropertyValue('white-space')
                  : 0);
              preformatted = whitespace
                  && 'pre' === whitespace.substring(0, 3);
            }

            // Look for a class like linenums or linenums:<n> where <n> is the
            // 1-indexed number of the first line.
            var lineNums = attrs['linenums'];
            if (!(lineNums = lineNums === 'true' || +lineNums)) {
              lineNums = className.match(/\blinenums\b(?::(\d+))?/);
              lineNums =
                lineNums
                ? lineNums[1] && lineNums[1].length
                  ? +lineNums[1] : true
                : false;
            }
            if (lineNums) { numberLines(cs, lineNums, preformatted); }

            // do the pretty printing
            var prettyPrintingJob = {
              langExtension: langExtension,
              sourceNode: cs,
              numberLines: lineNums,
              pre: preformatted,
              sourceCode: null,
              basePos: null,
              spans: null,
              decorations: null
            };
            applyDecorator(prettyPrintingJob);
          }
        }
      }
      if (k < elements.length) {
        // finish up in a continuation
        win.setTimeout(doWork, 250);
      } else if ('function' === typeof opt_whenDone) {
        opt_whenDone();
      }
    }

    doWork();
  }

  /**
   * Contains functions for creating and registering new language handlers.
   * @type {Object}
   */
  var PR = win['PR'] = {
        'createSimpleLexer': createSimpleLexer,
        'registerLangHandler': registerLangHandler,
        'sourceDecorator': sourceDecorator,
        'PR_ATTRIB_NAME': PR_ATTRIB_NAME,
        'PR_ATTRIB_VALUE': PR_ATTRIB_VALUE,
        'PR_COMMENT': PR_COMMENT,
        'PR_DECLARATION': PR_DECLARATION,
        'PR_KEYWORD': PR_KEYWORD,
        'PR_LITERAL': PR_LITERAL,
        'PR_NOCODE': PR_NOCODE,
        'PR_PLAIN': PR_PLAIN,
        'PR_PUNCTUATION': PR_PUNCTUATION,
        'PR_SOURCE': PR_SOURCE,
        'PR_STRING': PR_STRING,
        'PR_TAG': PR_TAG,
        'PR_TYPE': PR_TYPE,
        'prettyPrintOne':
           IN_GLOBAL_SCOPE
             ? (win['prettyPrintOne'] = $prettyPrintOne)
             : (prettyPrintOne = $prettyPrintOne),
        'prettyPrint':
           IN_GLOBAL_SCOPE
             ? (win['prettyPrint'] = $prettyPrint)
             : (prettyPrint = $prettyPrint)
      };

  // Make PR available via the Asynchronous Module Definition (AMD) API.
  // Per https://github.com/amdjs/amdjs-api/wiki/AMD:
  // The Asynchronous Module Definition (AMD) API specifies a
  // mechanism for defining modules such that the module and its
  // dependencies can be asynchronously loaded.
  // ...
  // To allow a clear indicator that a global define function (as
  // needed for script src browser loading) conforms to the AMD API,
  // any global define function SHOULD have a property called "amd"
  // whose value is an object. This helps avoid conflict with any
  // other existing JavaScript code that could have defined a define()
  // function that does not conform to the AMD API.
  var define = win['define'];
  if (typeof define === "function" && define['amd']) {
    define("google-code-prettify", [], function () {
      return PR;
    });
  }
})();