chat/protocols/twitter/twitter-text.jsm
author aleth <aleth@instantbird.org>
Sat, 08 Aug 2015 00:47:37 +0200
changeset 23061 655dcbac8fae72185edcfbbe15b4792d1d40ab91
parent 19292 f81a23bfefcd0f6ec6215fbde12f8af2bc7749f1
child 23489 7a2c243821cc79b3a11bc8fae12bc0c6135f35d5
permissions -rw-r--r--
Keep build/ in sync following bug 1191212. rs=bustage-fix

/* Copyright 2011 Twitter, 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.
 */

const EXPORTED_SYMBOLS = ["twttr"];

var window = {};

// The code below is imported from Twitter's JavaScript utility for parsing
// tweets. The original version of this file can be found at
// https://github.com/twitter/twitter-text-js/blob/master/twitter-text.js

(function() {
  if (typeof twttr === "undefined" || twttr === null) {
    var twttr = {};
  }

  twttr.txt = {};
  twttr.txt.regexen = {};

  var HTML_ENTITIES = {
    '&': '&amp;',
    '>': '&gt;',
    '<': '&lt;',
    '"': '&quot;',
    "'": '&#39;'
  };

  // HTML escaping
  twttr.txt.htmlEscape = function(text) {
    return text && text.replace(/[&"'><]/g, function(character) {
      return HTML_ENTITIES[character];
    });
  };

  // Builds a RegExp
  function regexSupplant(regex, flags) {
    flags = flags || "";
    if (typeof regex !== "string") {
      if (regex.global && flags.indexOf("g") < 0) {
        flags += "g";
      }
      if (regex.ignoreCase && flags.indexOf("i") < 0) {
        flags += "i";
      }
      if (regex.multiline && flags.indexOf("m") < 0) {
        flags += "m";
      }

      regex = regex.source;
    }

    return new RegExp(regex.replace(/#\{(\w+)\}/g, function(match, name) {
      var newRegex = twttr.txt.regexen[name] || "";
      if (typeof newRegex !== "string") {
        newRegex = newRegex.source;
      }
      return newRegex;
    }), flags);
  }

  twttr.txt.regexSupplant = regexSupplant;

  // simple string interpolation
  function stringSupplant(str, values) {
    return str.replace(/#\{(\w+)\}/g, function(match, name) {
      return values[name] || "";
    });
  }

  twttr.txt.stringSupplant = stringSupplant;

  function addCharsToCharClass(charClass, start, end) {
    var s = String.fromCharCode(start);
    if (end !== start) {
      s += "-" + String.fromCharCode(end);
    }
    charClass.push(s);
    return charClass;
  }

  twttr.txt.addCharsToCharClass = addCharsToCharClass;

  // Space is more than %20, U+3000 for example is the full-width space used with Kanji. Provide a short-hand
  // to access both the list of characters and a pattern suitible for use with String#split
  // Taken from: ActiveSupport::Multibyte::Handlers::UTF8Handler::UNICODE_WHITESPACE
  var fromCode = String.fromCharCode;
  var UNICODE_SPACES = [
    fromCode(0x0020), // White_Space # Zs       SPACE
    fromCode(0x0085), // White_Space # Cc       <control-0085>
    fromCode(0x00A0), // White_Space # Zs       NO-BREAK SPACE
    fromCode(0x1680), // White_Space # Zs       OGHAM SPACE MARK
    fromCode(0x180E), // White_Space # Zs       MONGOLIAN VOWEL SEPARATOR
    fromCode(0x2028), // White_Space # Zl       LINE SEPARATOR
    fromCode(0x2029), // White_Space # Zp       PARAGRAPH SEPARATOR
    fromCode(0x202F), // White_Space # Zs       NARROW NO-BREAK SPACE
    fromCode(0x205F), // White_Space # Zs       MEDIUM MATHEMATICAL SPACE
    fromCode(0x3000)  // White_Space # Zs       IDEOGRAPHIC SPACE
  ];
  addCharsToCharClass(UNICODE_SPACES, 0x009, 0x00D); // White_Space # Cc   [5] <control-0009>..<control-000D>
  addCharsToCharClass(UNICODE_SPACES, 0x2000, 0x200A); // White_Space # Zs  [11] EN QUAD..HAIR SPACE

  var INVALID_CHARS = [
    fromCode(0xFFFE),
    fromCode(0xFEFF), // BOM
    fromCode(0xFFFF) // Special
  ];
  addCharsToCharClass(INVALID_CHARS, 0x202A, 0x202E); // Directional change

  twttr.txt.regexen.spaces_group = regexSupplant(UNICODE_SPACES.join(""));
  twttr.txt.regexen.spaces = regexSupplant("[" + UNICODE_SPACES.join("") + "]");
  twttr.txt.regexen.invalid_chars_group = regexSupplant(INVALID_CHARS.join(""));
  twttr.txt.regexen.punct = /\!'#%&'\(\)*\+,\\\-\.\/:;<=>\?@\[\]\^_{|}~\$/;
  twttr.txt.regexen.rtl_chars = /[\u0600-\u06FF]|[\u0750-\u077F]|[\u0590-\u05FF]|[\uFE70-\uFEFF]/mg;
  twttr.txt.regexen.non_bmp_code_pairs = /[\uD800-\uDBFF][\uDC00-\uDFFF]/mg;

  var nonLatinHashtagChars = [];
  // Cyrillic
  addCharsToCharClass(nonLatinHashtagChars, 0x0400, 0x04ff); // Cyrillic
  addCharsToCharClass(nonLatinHashtagChars, 0x0500, 0x0527); // Cyrillic Supplement
  addCharsToCharClass(nonLatinHashtagChars, 0x2de0, 0x2dff); // Cyrillic Extended A
  addCharsToCharClass(nonLatinHashtagChars, 0xa640, 0xa69f); // Cyrillic Extended B
  // Hebrew
  addCharsToCharClass(nonLatinHashtagChars, 0x0591, 0x05bf); // Hebrew
  addCharsToCharClass(nonLatinHashtagChars, 0x05c1, 0x05c2);
  addCharsToCharClass(nonLatinHashtagChars, 0x05c4, 0x05c5);
  addCharsToCharClass(nonLatinHashtagChars, 0x05c7, 0x05c7);
  addCharsToCharClass(nonLatinHashtagChars, 0x05d0, 0x05ea);
  addCharsToCharClass(nonLatinHashtagChars, 0x05f0, 0x05f4);
  addCharsToCharClass(nonLatinHashtagChars, 0xfb12, 0xfb28); // Hebrew Presentation Forms
  addCharsToCharClass(nonLatinHashtagChars, 0xfb2a, 0xfb36);
  addCharsToCharClass(nonLatinHashtagChars, 0xfb38, 0xfb3c);
  addCharsToCharClass(nonLatinHashtagChars, 0xfb3e, 0xfb3e);
  addCharsToCharClass(nonLatinHashtagChars, 0xfb40, 0xfb41);
  addCharsToCharClass(nonLatinHashtagChars, 0xfb43, 0xfb44);
  addCharsToCharClass(nonLatinHashtagChars, 0xfb46, 0xfb4f);
  // Arabic
  addCharsToCharClass(nonLatinHashtagChars, 0x0610, 0x061a); // Arabic
  addCharsToCharClass(nonLatinHashtagChars, 0x0620, 0x065f);
  addCharsToCharClass(nonLatinHashtagChars, 0x066e, 0x06d3);
  addCharsToCharClass(nonLatinHashtagChars, 0x06d5, 0x06dc);
  addCharsToCharClass(nonLatinHashtagChars, 0x06de, 0x06e8);
  addCharsToCharClass(nonLatinHashtagChars, 0x06ea, 0x06ef);
  addCharsToCharClass(nonLatinHashtagChars, 0x06fa, 0x06fc);
  addCharsToCharClass(nonLatinHashtagChars, 0x06ff, 0x06ff);
  addCharsToCharClass(nonLatinHashtagChars, 0x0750, 0x077f); // Arabic Supplement
  addCharsToCharClass(nonLatinHashtagChars, 0x08a0, 0x08a0); // Arabic Extended A
  addCharsToCharClass(nonLatinHashtagChars, 0x08a2, 0x08ac);
  addCharsToCharClass(nonLatinHashtagChars, 0x08e4, 0x08fe);
  addCharsToCharClass(nonLatinHashtagChars, 0xfb50, 0xfbb1); // Arabic Pres. Forms A
  addCharsToCharClass(nonLatinHashtagChars, 0xfbd3, 0xfd3d);
  addCharsToCharClass(nonLatinHashtagChars, 0xfd50, 0xfd8f);
  addCharsToCharClass(nonLatinHashtagChars, 0xfd92, 0xfdc7);
  addCharsToCharClass(nonLatinHashtagChars, 0xfdf0, 0xfdfb);
  addCharsToCharClass(nonLatinHashtagChars, 0xfe70, 0xfe74); // Arabic Pres. Forms B
  addCharsToCharClass(nonLatinHashtagChars, 0xfe76, 0xfefc);
  addCharsToCharClass(nonLatinHashtagChars, 0x200c, 0x200c); // Zero-Width Non-Joiner
  // Thai
  addCharsToCharClass(nonLatinHashtagChars, 0x0e01, 0x0e3a);
  addCharsToCharClass(nonLatinHashtagChars, 0x0e40, 0x0e4e);
  // Hangul (Korean)
  addCharsToCharClass(nonLatinHashtagChars, 0x1100, 0x11ff); // Hangul Jamo
  addCharsToCharClass(nonLatinHashtagChars, 0x3130, 0x3185); // Hangul Compatibility Jamo
  addCharsToCharClass(nonLatinHashtagChars, 0xA960, 0xA97F); // Hangul Jamo Extended-A
  addCharsToCharClass(nonLatinHashtagChars, 0xAC00, 0xD7AF); // Hangul Syllables
  addCharsToCharClass(nonLatinHashtagChars, 0xD7B0, 0xD7FF); // Hangul Jamo Extended-B
  addCharsToCharClass(nonLatinHashtagChars, 0xFFA1, 0xFFDC); // half-width Hangul
  // Japanese and Chinese
  addCharsToCharClass(nonLatinHashtagChars, 0x30A1, 0x30FA); // Katakana (full-width)
  addCharsToCharClass(nonLatinHashtagChars, 0x30FC, 0x30FE); // Katakana Chouon and iteration marks (full-width)
  addCharsToCharClass(nonLatinHashtagChars, 0xFF66, 0xFF9F); // Katakana (half-width)
  addCharsToCharClass(nonLatinHashtagChars, 0xFF70, 0xFF70); // Katakana Chouon (half-width)
  addCharsToCharClass(nonLatinHashtagChars, 0xFF10, 0xFF19); // \
  addCharsToCharClass(nonLatinHashtagChars, 0xFF21, 0xFF3A); //  - Latin (full-width)
  addCharsToCharClass(nonLatinHashtagChars, 0xFF41, 0xFF5A); // /
  addCharsToCharClass(nonLatinHashtagChars, 0x3041, 0x3096); // Hiragana
  addCharsToCharClass(nonLatinHashtagChars, 0x3099, 0x309E); // Hiragana voicing and iteration mark
  addCharsToCharClass(nonLatinHashtagChars, 0x3400, 0x4DBF); // Kanji (CJK Extension A)
  addCharsToCharClass(nonLatinHashtagChars, 0x4E00, 0x9FFF); // Kanji (Unified)
  // -- Disabled as it breaks the Regex.
  //addCharsToCharClass(nonLatinHashtagChars, 0x20000, 0x2A6DF); // Kanji (CJK Extension B)
  addCharsToCharClass(nonLatinHashtagChars, 0x2A700, 0x2B73F); // Kanji (CJK Extension C)
  addCharsToCharClass(nonLatinHashtagChars, 0x2B740, 0x2B81F); // Kanji (CJK Extension D)
  addCharsToCharClass(nonLatinHashtagChars, 0x2F800, 0x2FA1F); // Kanji (CJK supplement)
  addCharsToCharClass(nonLatinHashtagChars, 0x3003, 0x3003); // Kanji iteration mark
  addCharsToCharClass(nonLatinHashtagChars, 0x3005, 0x3005); // Kanji iteration mark
  addCharsToCharClass(nonLatinHashtagChars, 0x303B, 0x303B); // Han iteration mark

  twttr.txt.regexen.nonLatinHashtagChars = regexSupplant(nonLatinHashtagChars.join(""));

  var latinAccentChars = [];
  // Latin accented characters (subtracted 0xD7 from the range, it's a confusable multiplication sign. Looks like "x")
  addCharsToCharClass(latinAccentChars, 0x00c0, 0x00d6);
  addCharsToCharClass(latinAccentChars, 0x00d8, 0x00f6);
  addCharsToCharClass(latinAccentChars, 0x00f8, 0x00ff);
  // Latin Extended A and B
  addCharsToCharClass(latinAccentChars, 0x0100, 0x024f);
  // assorted IPA Extensions
  addCharsToCharClass(latinAccentChars, 0x0253, 0x0254);
  addCharsToCharClass(latinAccentChars, 0x0256, 0x0257);
  addCharsToCharClass(latinAccentChars, 0x0259, 0x0259);
  addCharsToCharClass(latinAccentChars, 0x025b, 0x025b);
  addCharsToCharClass(latinAccentChars, 0x0263, 0x0263);
  addCharsToCharClass(latinAccentChars, 0x0268, 0x0268);
  addCharsToCharClass(latinAccentChars, 0x026f, 0x026f);
  addCharsToCharClass(latinAccentChars, 0x0272, 0x0272);
  addCharsToCharClass(latinAccentChars, 0x0289, 0x0289);
  addCharsToCharClass(latinAccentChars, 0x028b, 0x028b);
  // Okina for Hawaiian (it *is* a letter character)
  addCharsToCharClass(latinAccentChars, 0x02bb, 0x02bb);
  // Combining diacritics
  addCharsToCharClass(latinAccentChars, 0x0300, 0x036f);
  // Latin Extended Additional
  addCharsToCharClass(latinAccentChars, 0x1e00, 0x1eff);
  twttr.txt.regexen.latinAccentChars = regexSupplant(latinAccentChars.join(""));

  // A hashtag must contain characters, numbers and underscores, but not all numbers.
  twttr.txt.regexen.hashSigns = /[##]/;
  twttr.txt.regexen.hashtagAlpha = regexSupplant(/[a-z_#{latinAccentChars}#{nonLatinHashtagChars}]/i);
  twttr.txt.regexen.hashtagAlphaNumeric = regexSupplant(/[a-z0-9_#{latinAccentChars}#{nonLatinHashtagChars}]/i);
  twttr.txt.regexen.endHashtagMatch = regexSupplant(/^(?:#{hashSigns}|:\/\/)/);
  twttr.txt.regexen.hashtagBoundary = regexSupplant(/(?:^|$|[^&a-z0-9_#{latinAccentChars}#{nonLatinHashtagChars}])/);
  twttr.txt.regexen.validHashtag = regexSupplant(/(#{hashtagBoundary})(#{hashSigns})(#{hashtagAlphaNumeric}*#{hashtagAlpha}#{hashtagAlphaNumeric}*)/gi);

  // Mention related regex collection
  twttr.txt.regexen.validMentionPrecedingChars = /(?:^|[^a-zA-Z0-9_!#$%&*@@]|RT:?)/;
  twttr.txt.regexen.atSigns = /[@@]/;
  twttr.txt.regexen.validMentionOrList = regexSupplant(
    '(#{validMentionPrecedingChars})' +  // $1: Preceding character
    '(#{atSigns})' +                     // $2: At mark
    '([a-zA-Z0-9_]{1,20})' +             // $3: Screen name
    '(\/[a-zA-Z][a-zA-Z0-9_\-]{0,24})?'  // $4: List (optional)
  , 'g');
  twttr.txt.regexen.validReply = regexSupplant(/^(?:#{spaces})*#{atSigns}([a-zA-Z0-9_]{1,20})/);
  twttr.txt.regexen.endMentionMatch = regexSupplant(/^(?:#{atSigns}|[#{latinAccentChars}]|:\/\/)/);

  // URL related regex collection
  twttr.txt.regexen.validUrlPrecedingChars = regexSupplant(/(?:[^A-Za-z0-9@@$###{invalid_chars_group}]|^)/);
  twttr.txt.regexen.invalidUrlWithoutProtocolPrecedingChars = /[-_.\/]$/;
  twttr.txt.regexen.invalidDomainChars = stringSupplant("#{punct}#{spaces_group}#{invalid_chars_group}", twttr.txt.regexen);
  twttr.txt.regexen.validDomainChars = regexSupplant(/[^#{invalidDomainChars}]/);
  twttr.txt.regexen.validSubdomain = regexSupplant(/(?:(?:#{validDomainChars}(?:[_-]|#{validDomainChars})*)?#{validDomainChars}\.)/);
  twttr.txt.regexen.validDomainName = regexSupplant(/(?:(?:#{validDomainChars}(?:-|#{validDomainChars})*)?#{validDomainChars}\.)/);
  twttr.txt.regexen.validGTLD = regexSupplant(/(?:(?:aero|asia|biz|cat|com|coop|edu|gov|info|int|jobs|mil|mobi|museum|name|net|org|pro|tel|travel|xxx)(?=[^0-9a-zA-Z]|$))/);
  twttr.txt.regexen.validCCTLD = regexSupplant(RegExp(
        "(?:(?:ac|ad|ae|af|ag|ai|al|am|an|ao|aq|ar|as|at|au|aw|ax|az|ba|bb|bd|be|bf|bg|bh|bi|bj|bm|bn|bo|br|bs|bt|bv|bw|by|bz|" +
        "ca|cc|cd|cf|cg|ch|ci|ck|cl|cm|cn|co|cr|cs|cu|cv|cx|cy|cz|dd|de|dj|dk|dm|do|dz|ec|ee|eg|eh|er|es|et|eu|fi|fj|fk|fm|fo|fr|" +
        "ga|gb|gd|ge|gf|gg|gh|gi|gl|gm|gn|gp|gq|gr|gs|gt|gu|gw|gy|hk|hm|hn|hr|ht|hu|id|ie|il|im|in|io|iq|ir|is|it|je|jm|jo|jp|" +
        "ke|kg|kh|ki|km|kn|kp|kr|kw|ky|kz|la|lb|lc|li|lk|lr|ls|lt|lu|lv|ly|ma|mc|md|me|mg|mh|mk|ml|mm|mn|mo|mp|mq|mr|ms|mt|mu|mv|mw|mx|my|mz|" +
        "na|nc|ne|nf|ng|ni|nl|no|np|nr|nu|nz|om|pa|pe|pf|pg|ph|pk|pl|pm|pn|pr|ps|pt|pw|py|qa|re|ro|rs|ru|rw|" +
        "sa|sb|sc|sd|se|sg|sh|si|sj|sk|sl|sm|sn|so|sr|ss|st|su|sv|sx|sy|sz|tc|td|tf|tg|th|tj|tk|tl|tm|tn|to|tp|tr|tt|tv|tw|tz|" +
        "ua|ug|uk|us|uy|uz|va|vc|ve|vg|vi|vn|vu|wf|ws|ye|yt|za|zm|zw)(?=[^0-9a-zA-Z]|$))"));
  twttr.txt.regexen.validPunycode = regexSupplant(/(?:xn--[0-9a-z]+)/);
  twttr.txt.regexen.validDomain = regexSupplant(/(?:#{validSubdomain}*#{validDomainName}(?:#{validGTLD}|#{validCCTLD}|#{validPunycode}))/);
  twttr.txt.regexen.validAsciiDomain = regexSupplant(/(?:(?:[\-a-z0-9#{latinAccentChars}]+)\.)+(?:#{validGTLD}|#{validCCTLD}|#{validPunycode})/gi);
  twttr.txt.regexen.invalidShortDomain = regexSupplant(/^#{validDomainName}#{validCCTLD}$/);

  twttr.txt.regexen.validPortNumber = regexSupplant(/[0-9]+/);

  twttr.txt.regexen.validGeneralUrlPathChars = regexSupplant(/[a-z0-9!\*';:=\+,\.\$\/%#\[\]\-_~@|&#{latinAccentChars}]/i);
  // Allow URL paths to contain balanced parens
  //  1. Used in Wikipedia URLs like /Primer_(film)
  //  2. Used in IIS sessions like /S(dfd346)/
  twttr.txt.regexen.validUrlBalancedParens = regexSupplant(/\(#{validGeneralUrlPathChars}+\)/i);
  // Valid end-of-path chracters (so /foo. does not gobble the period).
  // 1. Allow =&# for empty URL parameters and other URL-join artifacts
  twttr.txt.regexen.validUrlPathEndingChars = regexSupplant(/[\+\-a-z0-9=_#\/#{latinAccentChars}]|(?:#{validUrlBalancedParens})/i);
  // Allow @ in a url, but only in the middle. Catch things like http://example.com/@user/
  twttr.txt.regexen.validUrlPath = regexSupplant('(?:' +
    '(?:' +
      '#{validGeneralUrlPathChars}*' +
        '(?:#{validUrlBalancedParens}#{validGeneralUrlPathChars}*)*' +
        '#{validUrlPathEndingChars}'+
      ')|(?:@#{validGeneralUrlPathChars}+\/)'+
    ')', 'i');

  twttr.txt.regexen.validUrlQueryChars = /[a-z0-9!?\*'@\(\);:&=\+\$\/%#\[\]\-_\.,~|]/i;
  twttr.txt.regexen.validUrlQueryEndingChars = /[a-z0-9_&=#\/]/i;
  twttr.txt.regexen.extractUrl = regexSupplant(
    '('                                                            + // $1 total match
      '(#{validUrlPrecedingChars})'                                + // $2 Preceeding chracter
      '('                                                          + // $3 URL
        '(https?:\\/\\/)?'                                         + // $4 Protocol (optional)
        '(#{validDomain})'                                         + // $5 Domain(s)
        '(?::(#{validPortNumber}))?'                               + // $6 Port number (optional)
        '(\\/#{validUrlPath}*)?'                                   + // $7 URL Path
        '(\\?#{validUrlQueryChars}*#{validUrlQueryEndingChars})?'  + // $8 Query String
      ')'                                                          +
    ')'
  , 'gi');

  twttr.txt.regexen.validTcoUrl = /^https?:\/\/t\.co\/[a-z0-9]+/i;
  twttr.txt.regexen.urlHasProtocol = /^https?:\/\//i;
  twttr.txt.regexen.urlHasHttps = /^https:\/\//i;

  // cashtag related regex
  twttr.txt.regexen.cashtag = /[a-z]{1,6}(?:[._][a-z]{1,2})?/i;
  twttr.txt.regexen.validCashtag = regexSupplant('(^|#{spaces})(\\$)(#{cashtag})(?=$|\\s|[#{punct}])', 'gi');

  // These URL validation pattern strings are based on the ABNF from RFC 3986
  twttr.txt.regexen.validateUrlUnreserved = /[a-z0-9\-._~]/i;
  twttr.txt.regexen.validateUrlPctEncoded = /(?:%[0-9a-f]{2})/i;
  twttr.txt.regexen.validateUrlSubDelims = /[!$&'()*+,;=]/i;
  twttr.txt.regexen.validateUrlPchar = regexSupplant('(?:' +
    '#{validateUrlUnreserved}|' +
    '#{validateUrlPctEncoded}|' +
    '#{validateUrlSubDelims}|' +
    '[:|@]' +
  ')', 'i');

  twttr.txt.regexen.validateUrlScheme = /(?:[a-z][a-z0-9+\-.]*)/i;
  twttr.txt.regexen.validateUrlUserinfo = regexSupplant('(?:' +
    '#{validateUrlUnreserved}|' +
    '#{validateUrlPctEncoded}|' +
    '#{validateUrlSubDelims}|' +
    ':' +
  ')*', 'i');

  twttr.txt.regexen.validateUrlDecOctet = /(?:[0-9]|(?:[1-9][0-9])|(?:1[0-9]{2})|(?:2[0-4][0-9])|(?:25[0-5]))/i;
  twttr.txt.regexen.validateUrlIpv4 = regexSupplant(/(?:#{validateUrlDecOctet}(?:\.#{validateUrlDecOctet}){3})/i);

  // Punting on real IPv6 validation for now
  twttr.txt.regexen.validateUrlIpv6 = /(?:\[[a-f0-9:\.]+\])/i;

  // Also punting on IPvFuture for now
  twttr.txt.regexen.validateUrlIp = regexSupplant('(?:' +
    '#{validateUrlIpv4}|' +
    '#{validateUrlIpv6}' +
  ')', 'i');

  // This is more strict than the rfc specifies
  twttr.txt.regexen.validateUrlSubDomainSegment = /(?:[a-z0-9](?:[a-z0-9_\-]*[a-z0-9])?)/i;
  twttr.txt.regexen.validateUrlDomainSegment = /(?:[a-z0-9](?:[a-z0-9\-]*[a-z0-9])?)/i;
  twttr.txt.regexen.validateUrlDomainTld = /(?:[a-z](?:[a-z0-9\-]*[a-z0-9])?)/i;
  twttr.txt.regexen.validateUrlDomain = regexSupplant(/(?:(?:#{validateUrlSubDomainSegment]}\.)*(?:#{validateUrlDomainSegment]}\.)#{validateUrlDomainTld})/i);

  twttr.txt.regexen.validateUrlHost = regexSupplant('(?:' +
    '#{validateUrlIp}|' +
    '#{validateUrlDomain}' +
  ')', 'i');

  // Unencoded internationalized domains - this doesn't check for invalid UTF-8 sequences
  twttr.txt.regexen.validateUrlUnicodeSubDomainSegment = /(?:(?:[a-z0-9]|[^\u0000-\u007f])(?:(?:[a-z0-9_\-]|[^\u0000-\u007f])*(?:[a-z0-9]|[^\u0000-\u007f]))?)/i;
  twttr.txt.regexen.validateUrlUnicodeDomainSegment = /(?:(?:[a-z0-9]|[^\u0000-\u007f])(?:(?:[a-z0-9\-]|[^\u0000-\u007f])*(?:[a-z0-9]|[^\u0000-\u007f]))?)/i;
  twttr.txt.regexen.validateUrlUnicodeDomainTld = /(?:(?:[a-z]|[^\u0000-\u007f])(?:(?:[a-z0-9\-]|[^\u0000-\u007f])*(?:[a-z0-9]|[^\u0000-\u007f]))?)/i;
  twttr.txt.regexen.validateUrlUnicodeDomain = regexSupplant(/(?:(?:#{validateUrlUnicodeSubDomainSegment}\.)*(?:#{validateUrlUnicodeDomainSegment}\.)#{validateUrlUnicodeDomainTld})/i);

  twttr.txt.regexen.validateUrlUnicodeHost = regexSupplant('(?:' +
    '#{validateUrlIp}|' +
    '#{validateUrlUnicodeDomain}' +
  ')', 'i');

  twttr.txt.regexen.validateUrlPort = /[0-9]{1,5}/;

  twttr.txt.regexen.validateUrlUnicodeAuthority = regexSupplant(
    '(?:(#{validateUrlUserinfo})@)?'  + // $1 userinfo
    '(#{validateUrlUnicodeHost})'     + // $2 host
    '(?::(#{validateUrlPort}))?'        //$3 port
  , "i");

  twttr.txt.regexen.validateUrlAuthority = regexSupplant(
    '(?:(#{validateUrlUserinfo})@)?' + // $1 userinfo
    '(#{validateUrlHost})'           + // $2 host
    '(?::(#{validateUrlPort}))?'       // $3 port
  , "i");

  twttr.txt.regexen.validateUrlPath = regexSupplant(/(\/#{validateUrlPchar}*)*/i);
  twttr.txt.regexen.validateUrlQuery = regexSupplant(/(#{validateUrlPchar}|\/|\?)*/i);
  twttr.txt.regexen.validateUrlFragment = regexSupplant(/(#{validateUrlPchar}|\/|\?)*/i);

  // Modified version of RFC 3986 Appendix B
  twttr.txt.regexen.validateUrlUnencoded = regexSupplant(
    '^'                               + // Full URL
    '(?:'                             +
      '([^:/?#]+):\\/\\/'             + // $1 Scheme
    ')?'                              +
    '([^/?#]*)'                       + // $2 Authority
    '([^?#]*)'                        + // $3 Path
    '(?:'                             +
      '\\?([^#]*)'                    + // $4 Query
    ')?'                              +
    '(?:'                             +
      '#(.*)'                         + // $5 Fragment
    ')?$'
  , "i");


  // Default CSS class for auto-linked lists (along with the url class)
  var DEFAULT_LIST_CLASS = "tweet-url list-slug";
  // Default CSS class for auto-linked usernames (along with the url class)
  var DEFAULT_USERNAME_CLASS = "tweet-url username";
  // Default CSS class for auto-linked hashtags (along with the url class)
  var DEFAULT_HASHTAG_CLASS = "tweet-url hashtag";
  // Default CSS class for auto-linked cashtags (along with the url class)
  var DEFAULT_CASHTAG_CLASS = "tweet-url cashtag";
  // Options which should not be passed as HTML attributes
  var OPTIONS_NOT_ATTRIBUTES = {'urlClass':true, 'listClass':true, 'usernameClass':true, 'hashtagClass':true, 'cashtagClass':true,
                            'usernameUrlBase':true, 'listUrlBase':true, 'hashtagUrlBase':true, 'cashtagUrlBase':true,
                            'usernameUrlBlock':true, 'listUrlBlock':true, 'hashtagUrlBlock':true, 'linkUrlBlock':true,
                            'usernameIncludeSymbol':true, 'suppressLists':true, 'suppressNoFollow':true, 'targetBlank':true,
                            'suppressDataScreenName':true, 'urlEntities':true, 'symbolTag':true, 'textWithSymbolTag':true, 'urlTarget':true,
                            'invisibleTagAttrs':true, 'linkAttributeBlock':true, 'linkTextBlock': true, 'htmlEscapeNonEntities': true
                            };

  var BOOLEAN_ATTRIBUTES = {'disabled':true, 'readonly':true, 'multiple':true, 'checked':true};

  // Simple object cloning function for simple objects
  function clone(o) {
    var r = {};
    for (var k in o) {
      if (o.hasOwnProperty(k)) {
        r[k] = o[k];
      }
    }

    return r;
  }

  twttr.txt.tagAttrs = function(attributes) {
    var htmlAttrs = "";
    for (var k in attributes) {
      var v = attributes[k];
      if (BOOLEAN_ATTRIBUTES[k]) {
        v = v ? k : null;
      }
      if (v == null) continue;
      htmlAttrs += " " + twttr.txt.htmlEscape(k) + "=\"" + twttr.txt.htmlEscape(v.toString()) + "\"";
    }
    return htmlAttrs;
  };

  twttr.txt.linkToText = function(entity, text, attributes, options) {
    if (!options.suppressNoFollow) {
      attributes.rel = "nofollow";
    }
    // if linkAttributeBlock is specified, call it to modify the attributes
    if (options.linkAttributeBlock) {
      options.linkAttributeBlock(entity, attributes);
    }
    // if linkTextBlock is specified, call it to get a new/modified link text
    if (options.linkTextBlock) {
      text = options.linkTextBlock(entity, text);
    }
    var d = {
      text: text,
      attr: twttr.txt.tagAttrs(attributes)
    };
    return stringSupplant("<a#{attr}>#{text}</a>", d);
  };

  twttr.txt.linkToTextWithSymbol = function(entity, symbol, text, attributes, options) {
    var taggedSymbol = options.symbolTag ? "<" + options.symbolTag + ">" + symbol + "</"+ options.symbolTag + ">" : symbol;
    text = twttr.txt.htmlEscape(text);
    var taggedText = options.textWithSymbolTag ? "<" + options.textWithSymbolTag + ">" + text + "</"+ options.textWithSymbolTag + ">" : text;

    if (options.usernameIncludeSymbol || !symbol.match(twttr.txt.regexen.atSigns)) {
      return twttr.txt.linkToText(entity, taggedSymbol + taggedText, attributes, options);
    } else {
      return taggedSymbol + twttr.txt.linkToText(entity, taggedText, attributes, options);
    }
  };

  twttr.txt.linkToHashtag = function(entity, text, options) {
    var hash = text.substring(entity.indices[0], entity.indices[0] + 1);
    var hashtag = twttr.txt.htmlEscape(entity.hashtag);
    var attrs = clone(options.htmlAttrs || {});
    attrs.href = options.hashtagUrlBase + hashtag;
    attrs.title = "#" + hashtag;
    attrs["class"] = options.hashtagClass;
    if (hashtag[0].match(twttr.txt.regexen.rtl_chars)){
      attrs["class"] += " rtl";
    }
    if (options.targetBlank) {
      attrs.target = '_blank';
    }

    return twttr.txt.linkToTextWithSymbol(entity, hash, hashtag, attrs, options);
  };

  twttr.txt.linkToCashtag = function(entity, text, options) {
    var cashtag = twttr.txt.htmlEscape(entity.cashtag);
    var attrs = clone(options.htmlAttrs || {});
    attrs.href = options.cashtagUrlBase + cashtag;
    attrs.title = "$" + cashtag;
    attrs["class"] =  options.cashtagClass;
    if (options.targetBlank) {
      attrs.target = '_blank';
    }

    return twttr.txt.linkToTextWithSymbol(entity, "$", cashtag, attrs, options);
  };

  twttr.txt.linkToMentionAndList = function(entity, text, options) {
    var at = text.substring(entity.indices[0], entity.indices[0] + 1);
    var user = twttr.txt.htmlEscape(entity.screenName);
    var slashListname = twttr.txt.htmlEscape(entity.listSlug);
    var isList = entity.listSlug && !options.suppressLists;
    var attrs = clone(options.htmlAttrs || {});
    attrs["class"] = (isList ? options.listClass : options.usernameClass);
    attrs.href = isList ? options.listUrlBase + user + slashListname : options.usernameUrlBase + user;
    if (!isList && !options.suppressDataScreenName) {
      attrs['data-screen-name'] = user;
    }
    if (options.targetBlank) {
      attrs.target = '_blank';
    }

    return twttr.txt.linkToTextWithSymbol(entity, at, isList ? user + slashListname : user, attrs, options);
  };

  twttr.txt.linkToUrl = function(entity, text, options) {
    var url = entity.url;
    var displayUrl = url;
    var linkText = twttr.txt.htmlEscape(displayUrl);

    // If the caller passed a urlEntities object (provided by a Twitter API
    // response with include_entities=true), we use that to render the display_url
    // for each URL instead of it's underlying t.co URL.
    var urlEntity = (options.urlEntities && options.urlEntities[url]) || entity;
    if (urlEntity.display_url) {
      linkText = twttr.txt.linkTextWithEntity(urlEntity, options);
    }

    var attrs = clone(options.htmlAttrs || {});

    if (!url.match(twttr.txt.regexen.urlHasProtocol)) {
      url = "http://" + url;
    }
    attrs.href = url;

    if (options.targetBlank) {
      attrs.target = '_blank';
    }

    // set class only if urlClass is specified.
    if (options.urlClass) {
      attrs["class"] = options.urlClass;
    }

    // set target only if urlTarget is specified.
    if (options.urlTarget) {
      attrs.target = options.urlTarget;
    }

    if (!options.title && urlEntity.display_url) {
      attrs.title = urlEntity.expanded_url;
    }

    return twttr.txt.linkToText(entity, linkText, attrs, options);
  };

  twttr.txt.linkTextWithEntity = function (entity, options) {
    var displayUrl = entity.display_url;
    var expandedUrl = entity.expanded_url;

    // Goal: If a user copies and pastes a tweet containing t.co'ed link, the resulting paste
    // should contain the full original URL (expanded_url), not the display URL.
    //
    // Method: Whenever possible, we actually emit HTML that contains expanded_url, and use
    // font-size:0 to hide those parts that should not be displayed (because they are not part of display_url).
    // Elements with font-size:0 get copied even though they are not visible.
    // Note that display:none doesn't work here. Elements with display:none don't get copied.
    //
    // Additionally, we want to *display* ellipses, but we don't want them copied.  To make this happen we
    // wrap the ellipses in a tco-ellipsis class and provide an onCopy handler that sets display:none on
    // everything with the tco-ellipsis class.
    //
    // Exception: pic.twitter.com images, for which expandedUrl = "https://twitter.com/#!/username/status/1234/photo/1
    // For those URLs, display_url is not a substring of expanded_url, so we don't do anything special to render the elided parts.
    // For a pic.twitter.com URL, the only elided part will be the "https://", so this is fine.

    var displayUrlSansEllipses = displayUrl.replace(/…/g, ""); // We have to disregard ellipses for matching
    // Note: we currently only support eliding parts of the URL at the beginning or the end.
    // Eventually we may want to elide parts of the URL in the *middle*.  If so, this code will
    // become more complicated.  We will probably want to create a regexp out of display URL,
    // replacing every ellipsis with a ".*".
    if (expandedUrl.indexOf(displayUrlSansEllipses) != -1) {
      var displayUrlIndex = expandedUrl.indexOf(displayUrlSansEllipses);
      var v = {
        displayUrlSansEllipses: displayUrlSansEllipses,
        // Portion of expandedUrl that precedes the displayUrl substring
        beforeDisplayUrl: expandedUrl.substr(0, displayUrlIndex),
        // Portion of expandedUrl that comes after displayUrl
        afterDisplayUrl: expandedUrl.substr(displayUrlIndex + displayUrlSansEllipses.length),
        precedingEllipsis: displayUrl.match(/^…/) ? "…" : "",
        followingEllipsis: displayUrl.match(/…$/) ? "…" : ""
      };
      for (var k in v) {
        if (v.hasOwnProperty(k)) {
          v[k] = twttr.txt.htmlEscape(v[k]);
        }
      }
      // As an example: The user tweets "hi http://longdomainname.com/foo"
      // This gets shortened to "hi http://t.co/xyzabc", with display_url = "…nname.com/foo"
      // This will get rendered as:
      // <span class='tco-ellipsis'> <!-- This stuff should get displayed but not copied -->
      //   …
      //   <!-- There's a chance the onCopy event handler might not fire. In case that happens,
      //        we include an &nbsp; here so that the … doesn't bump up against the URL and ruin it.
      //        The &nbsp; is inside the tco-ellipsis span so that when the onCopy handler *does*
      //        fire, it doesn't get copied.  Otherwise the copied text would have two spaces in a row,
      //        e.g. "hi  http://longdomainname.com/foo".
      //   <span style='font-size:0'>&nbsp;</span>
      // </span>
      // <span style='font-size:0'>  <!-- This stuff should get copied but not displayed -->
      //   http://longdomai
      // </span>
      // <span class='js-display-url'> <!-- This stuff should get displayed *and* copied -->
      //   nname.com/foo
      // </span>
      // <span class='tco-ellipsis'> <!-- This stuff should get displayed but not copied -->
      //   <span style='font-size:0'>&nbsp;</span>
      //   …
      // </span>
      v['invisible'] = options.invisibleTagAttrs;
      return stringSupplant("<span class='tco-ellipsis'>#{precedingEllipsis}<span #{invisible}>&nbsp;</span></span><span #{invisible}>#{beforeDisplayUrl}</span><span class='js-display-url'>#{displayUrlSansEllipses}</span><span #{invisible}>#{afterDisplayUrl}</span><span class='tco-ellipsis'><span #{invisible}>&nbsp;</span>#{followingEllipsis}</span>", v);
    }
    return displayUrl;
  };

  twttr.txt.autoLinkEntities = function(text, entities, options) {
    options = clone(options || {});

    options.hashtagClass = options.hashtagClass || DEFAULT_HASHTAG_CLASS;
    options.hashtagUrlBase = options.hashtagUrlBase || "https://twitter.com/#!/search?q=%23";
    options.cashtagClass = options.cashtagClass || DEFAULT_CASHTAG_CLASS;
    options.cashtagUrlBase = options.cashtagUrlBase || "https://twitter.com/#!/search?q=%24";
    options.listClass = options.listClass || DEFAULT_LIST_CLASS;
    options.usernameClass = options.usernameClass || DEFAULT_USERNAME_CLASS;
    options.usernameUrlBase = options.usernameUrlBase || "https://twitter.com/";
    options.listUrlBase = options.listUrlBase || "https://twitter.com/";
    options.htmlAttrs = twttr.txt.extractHtmlAttrsFromOptions(options);
    options.invisibleTagAttrs = options.invisibleTagAttrs || "style='position:absolute;left:-9999px;'";

    // remap url entities to hash
    var urlEntities, i, len;
    if(options.urlEntities) {
      urlEntities = {};
      for(i = 0, len = options.urlEntities.length; i < len; i++) {
        urlEntities[options.urlEntities[i].url] = options.urlEntities[i];
      }
      options.urlEntities = urlEntities;
    }

    var result = "";
    var beginIndex = 0;

    // sort entities by start index
    entities.sort(function(a,b){ return a.indices[0] - b.indices[0]; });

    var nonEntity = options.htmlEscapeNonEntities ? twttr.txt.htmlEscape : function(text) {
      return text;
    };

    for (var i = 0; i < entities.length; i++) {
      var entity = entities[i];
      result += nonEntity(text.substring(beginIndex, entity.indices[0]));

      if (entity.url) {
        result += twttr.txt.linkToUrl(entity, text, options);
      } else if (entity.hashtag) {
        result += twttr.txt.linkToHashtag(entity, text, options);
      } else if (entity.screenName) {
        result += twttr.txt.linkToMentionAndList(entity, text, options);
      } else if (entity.cashtag) {
        result += twttr.txt.linkToCashtag(entity, text, options);
      }
      beginIndex = entity.indices[1];
    }
    result += nonEntity(text.substring(beginIndex, text.length));
    return result;
  };

  twttr.txt.autoLinkWithJSON = function(text, json, options) {
    // concatenate all entities
    var entities = [];
    for (var key in json) {
      entities = entities.concat(json[key]);
    }
    // map JSON entity to twitter-text entity
    for (var i = 0; i < entities.length; i++) {
      entity = entities[i];
      if (entity.screen_name) {
        // this is @mention
        entity.screenName = entity.screen_name;
      } else if (entity.text) {
        // this is #hashtag
        entity.hashtag = entity.text;
      }
    }
    // modify indices to UTF-16
    twttr.txt.modifyIndicesFromUnicodeToUTF16(text, entities);

    return twttr.txt.autoLinkEntities(text, entities, options);
  };

  twttr.txt.extractHtmlAttrsFromOptions = function(options) {
    var htmlAttrs = {};
    for (var k in options) {
      var v = options[k];
      if (OPTIONS_NOT_ATTRIBUTES[k]) continue;
      if (BOOLEAN_ATTRIBUTES[k]) {
        v = v ? k : null;
      }
      if (v == null) continue;
      htmlAttrs[k] = v;
    }
    return htmlAttrs;
  };

  twttr.txt.autoLink = function(text, options) {
    var entities = twttr.txt.extractEntitiesWithIndices(text, {extractUrlsWithoutProtocol: false});
    return twttr.txt.autoLinkEntities(text, entities, options);
  };

  twttr.txt.autoLinkUsernamesOrLists = function(text, options) {
    var entities = twttr.txt.extractMentionsOrListsWithIndices(text);
    return twttr.txt.autoLinkEntities(text, entities, options);
  };

  twttr.txt.autoLinkHashtags = function(text, options) {
    var entities = twttr.txt.extractHashtagsWithIndices(text);
    return twttr.txt.autoLinkEntities(text, entities, options);
  };

  twttr.txt.autoLinkCashtags = function(text, options) {
    var entities = twttr.txt.extractCashtagsWithIndices(text);
    return twttr.txt.autoLinkEntities(text, entities, options);
  };

  twttr.txt.autoLinkUrlsCustom = function(text, options) {
    var entities = twttr.txt.extractUrlsWithIndices(text, {extractUrlsWithoutProtocol: false});
    return twttr.txt.autoLinkEntities(text, entities, options);
  };

  twttr.txt.removeOverlappingEntities = function(entities) {
    entities.sort(function(a,b){ return a.indices[0] - b.indices[0]; });

    var prev = entities[0];
    for (var i = 1; i < entities.length; i++) {
      if (prev.indices[1] > entities[i].indices[0]) {
        entities.splice(i, 1);
        i--;
      } else {
        prev = entities[i];
      }
    }
  };

  twttr.txt.extractEntitiesWithIndices = function(text, options) {
    var entities = twttr.txt.extractUrlsWithIndices(text, options)
                    .concat(twttr.txt.extractMentionsOrListsWithIndices(text))
                    .concat(twttr.txt.extractHashtagsWithIndices(text, {checkUrlOverlap: false}))
                    .concat(twttr.txt.extractCashtagsWithIndices(text));

    if (entities.length == 0) {
      return [];
    }

    twttr.txt.removeOverlappingEntities(entities);
    return entities;
  };

  twttr.txt.extractMentions = function(text) {
    var screenNamesOnly = [],
        screenNamesWithIndices = twttr.txt.extractMentionsWithIndices(text);

    for (var i = 0; i < screenNamesWithIndices.length; i++) {
      var screenName = screenNamesWithIndices[i].screenName;
      screenNamesOnly.push(screenName);
    }

    return screenNamesOnly;
  };

  twttr.txt.extractMentionsWithIndices = function(text) {
    var mentions = [],
        mentionOrList,
        mentionsOrLists = twttr.txt.extractMentionsOrListsWithIndices(text);

    for (var i = 0 ; i < mentionsOrLists.length; i++) {
      mentionOrList = mentionsOrLists[i];
      if (mentionOrList.listSlug == '') {
        mentions.push({
          screenName: mentionOrList.screenName,
          indices: mentionOrList.indices
        });
      }
    }

    return mentions;
  };

  /**
   * Extract list or user mentions.
   * (Presence of listSlug indicates a list)
   */
  twttr.txt.extractMentionsOrListsWithIndices = function(text) {
    if (!text || !text.match(twttr.txt.regexen.atSigns)) {
      return [];
    }

    var possibleNames = [],
        slashListname;

    text.replace(twttr.txt.regexen.validMentionOrList, function(match, before, atSign, screenName, slashListname, offset, chunk) {
      var after = chunk.slice(offset + match.length);
      if (!after.match(twttr.txt.regexen.endMentionMatch)) {
        slashListname = slashListname || '';
        var startPosition = offset + before.length;
        var endPosition = startPosition + screenName.length + slashListname.length + 1;
        possibleNames.push({
          screenName: screenName,
          listSlug: slashListname,
          indices: [startPosition, endPosition]
        });
      }
    });

    return possibleNames;
  };


  twttr.txt.extractReplies = function(text) {
    if (!text) {
      return null;
    }

    var possibleScreenName = text.match(twttr.txt.regexen.validReply);
    if (!possibleScreenName ||
        RegExp.rightContext.match(twttr.txt.regexen.endMentionMatch)) {
      return null;
    }

    return possibleScreenName[1];
  };

  twttr.txt.extractUrls = function(text, options) {
    var urlsOnly = [],
        urlsWithIndices = twttr.txt.extractUrlsWithIndices(text, options);

    for (var i = 0; i < urlsWithIndices.length; i++) {
      urlsOnly.push(urlsWithIndices[i].url);
    }

    return urlsOnly;
  };

  twttr.txt.extractUrlsWithIndices = function(text, options) {
    if (!options) {
      options = {extractUrlsWithoutProtocol: true};
    }

    if (!text || (options.extractUrlsWithoutProtocol ? !text.match(/\./) : !text.match(/:/))) {
      return [];
    }

    var urls = [];

    while (twttr.txt.regexen.extractUrl.exec(text)) {
      var before = RegExp.$2, url = RegExp.$3, protocol = RegExp.$4, domain = RegExp.$5, path = RegExp.$7;
      var endPosition = twttr.txt.regexen.extractUrl.lastIndex,
          startPosition = endPosition - url.length;

      // if protocol is missing and domain contains non-ASCII characters,
      // extract ASCII-only domains.
      if (!protocol) {
        if (!options.extractUrlsWithoutProtocol
            || before.match(twttr.txt.regexen.invalidUrlWithoutProtocolPrecedingChars)) {
          continue;
        }
        var lastUrl = null,
            lastUrlInvalidMatch = false,
            asciiEndPosition = 0;
        domain.replace(twttr.txt.regexen.validAsciiDomain, function(asciiDomain) {
          var asciiStartPosition = domain.indexOf(asciiDomain, asciiEndPosition);
          asciiEndPosition = asciiStartPosition + asciiDomain.length;
          lastUrl = {
            url: asciiDomain,
            indices: [startPosition + asciiStartPosition, startPosition + asciiEndPosition]
          };
          lastUrlInvalidMatch = asciiDomain.match(twttr.txt.regexen.invalidShortDomain);
          if (!lastUrlInvalidMatch) {
            urls.push(lastUrl);
          }
        });

        // no ASCII-only domain found. Skip the entire URL.
        if (lastUrl == null) {
          continue;
        }

        // lastUrl only contains domain. Need to add path and query if they exist.
        if (path) {
          if (lastUrlInvalidMatch) {
            urls.push(lastUrl);
          }
          lastUrl.url = url.replace(domain, lastUrl.url);
          lastUrl.indices[1] = endPosition;
        }
      } else {
        // In the case of t.co URLs, don't allow additional path characters.
        if (url.match(twttr.txt.regexen.validTcoUrl)) {
          url = RegExp.lastMatch;
          endPosition = startPosition + url.length;
        }
        urls.push({
          url: url,
          indices: [startPosition, endPosition]
        });
      }
    }

    return urls;
  };

  twttr.txt.extractHashtags = function(text) {
    var hashtagsOnly = [],
        hashtagsWithIndices = twttr.txt.extractHashtagsWithIndices(text);

    for (var i = 0; i < hashtagsWithIndices.length; i++) {
      hashtagsOnly.push(hashtagsWithIndices[i].hashtag);
    }

    return hashtagsOnly;
  };

  twttr.txt.extractHashtagsWithIndices = function(text, options) {
    if (!options) {
      options = {checkUrlOverlap: true};
    }

    if (!text || !text.match(twttr.txt.regexen.hashSigns)) {
      return [];
    }

    var tags = [];

    text.replace(twttr.txt.regexen.validHashtag, function(match, before, hash, hashText, offset, chunk) {
      var after = chunk.slice(offset + match.length);
      if (after.match(twttr.txt.regexen.endHashtagMatch))
        return;
      var startPosition = offset + before.length;
      var endPosition = startPosition + hashText.length + 1;
      tags.push({
        hashtag: hashText,
        indices: [startPosition, endPosition]
      });
    });

    if (options.checkUrlOverlap) {
      // also extract URL entities
      var urls = twttr.txt.extractUrlsWithIndices(text);
      if (urls.length > 0) {
        var entities = tags.concat(urls);
        // remove overlap
        twttr.txt.removeOverlappingEntities(entities);
        // only push back hashtags
        tags = [];
        for (var i = 0; i < entities.length; i++) {
          if (entities[i].hashtag) {
            tags.push(entities[i]);
          }
        }
      }
    }

    return tags;
  };

  twttr.txt.extractCashtags = function(text) {
    var cashtagsOnly = [],
        cashtagsWithIndices = twttr.txt.extractCashtagsWithIndices(text);

    for (var i = 0; i < cashtagsWithIndices.length; i++) {
      cashtagsOnly.push(cashtagsWithIndices[i].cashtag);
    }

    return cashtagsOnly;
  };

  twttr.txt.extractCashtagsWithIndices = function(text) {
    if (!text || text.indexOf("$") == -1) {
      return [];
    }

    var tags = [];

    text.replace(twttr.txt.regexen.validCashtag, function(match, before, dollar, cashtag, offset, chunk) {
      var startPosition = offset + before.length;
      var endPosition = startPosition + cashtag.length + 1;
      tags.push({
        cashtag: cashtag,
        indices: [startPosition, endPosition]
      });
    });

    return tags;
  };

  twttr.txt.modifyIndicesFromUnicodeToUTF16 = function(text, entities) {
    twttr.txt.convertUnicodeIndices(text, entities, false);
  };

  twttr.txt.modifyIndicesFromUTF16ToUnicode = function(text, entities) {
    twttr.txt.convertUnicodeIndices(text, entities, true);
  };

  twttr.txt.getUnicodeTextLength = function(text) {
    return text.replace(twttr.txt.regexen.non_bmp_code_pairs, ' ').length;
  };

  twttr.txt.convertUnicodeIndices = function(text, entities, indicesInUTF16) {
    if (entities.length == 0) {
      return;
    }

    var charIndex = 0;
    var codePointIndex = 0;

    // sort entities by start index
    entities.sort(function(a,b){ return a.indices[0] - b.indices[0]; });
    var entityIndex = 0;
    var entity = entities[0];

    while (charIndex < text.length) {
      if (entity.indices[0] == (indicesInUTF16 ? charIndex : codePointIndex)) {
        var len = entity.indices[1] - entity.indices[0];
        entity.indices[0] = indicesInUTF16 ? codePointIndex : charIndex;
        entity.indices[1] = entity.indices[0] + len;

        entityIndex++;
        if (entityIndex == entities.length) {
          // no more entity
          break;
        }
        entity = entities[entityIndex];
      }

      var c = text.charCodeAt(charIndex);
      if (0xD800 <= c && c <= 0xDBFF && charIndex < text.length - 1) {
        // Found high surrogate char
        c = text.charCodeAt(charIndex + 1);
        if (0xDC00 <= c && c <= 0xDFFF) {
          // Found surrogate pair
          charIndex++;
        }
      }
      codePointIndex++;
      charIndex++;
    }
  };

  // this essentially does text.split(/<|>/)
  // except that won't work in IE, where empty strings are ommitted
  // so "<>".split(/<|>/) => [] in IE, but is ["", "", ""] in all others
  // but "<<".split("<") => ["", "", ""]
  twttr.txt.splitTags = function(text) {
    var firstSplits = text.split("<"),
        secondSplits,
        allSplits = [],
        split;

    for (var i = 0; i < firstSplits.length; i += 1) {
      split = firstSplits[i];
      if (!split) {
        allSplits.push("");
      } else {
        secondSplits = split.split(">");
        for (var j = 0; j < secondSplits.length; j += 1) {
          allSplits.push(secondSplits[j]);
        }
      }
    }

    return allSplits;
  };

  twttr.txt.hitHighlight = function(text, hits, options) {
    var defaultHighlightTag = "em";

    hits = hits || [];
    options = options || {};

    if (hits.length === 0) {
      return text;
    }

    var tagName = options.tag || defaultHighlightTag,
        tags = ["<" + tagName + ">", "</" + tagName + ">"],
        chunks = twttr.txt.splitTags(text),
        i,
        j,
        result = "",
        chunkIndex = 0,
        chunk = chunks[0],
        prevChunksLen = 0,
        chunkCursor = 0,
        startInChunk = false,
        chunkChars = chunk,
        flatHits = [],
        index,
        hit,
        tag,
        placed,
        hitSpot;

    for (i = 0; i < hits.length; i += 1) {
      for (j = 0; j < hits[i].length; j += 1) {
        flatHits.push(hits[i][j]);
      }
    }

    for (index = 0; index < flatHits.length; index += 1) {
      hit = flatHits[index];
      tag = tags[index % 2];
      placed = false;

      while (chunk != null && hit >= prevChunksLen + chunk.length) {
        result += chunkChars.slice(chunkCursor);
        if (startInChunk && hit === prevChunksLen + chunkChars.length) {
          result += tag;
          placed = true;
        }

        if (chunks[chunkIndex + 1]) {
          result += "<" + chunks[chunkIndex + 1] + ">";
        }

        prevChunksLen += chunkChars.length;
        chunkCursor = 0;
        chunkIndex += 2;
        chunk = chunks[chunkIndex];
        chunkChars = chunk;
        startInChunk = false;
      }

      if (!placed && chunk != null) {
        hitSpot = hit - prevChunksLen;
        result += chunkChars.slice(chunkCursor, hitSpot) + tag;
        chunkCursor = hitSpot;
        if (index % 2 === 0) {
          startInChunk = true;
        } else {
          startInChunk = false;
        }
      } else if(!placed) {
        placed = true;
        result += tag;
      }
    }

    if (chunk != null) {
      if (chunkCursor < chunkChars.length) {
        result += chunkChars.slice(chunkCursor);
      }
      for (index = chunkIndex + 1; index < chunks.length; index += 1) {
        result += (index % 2 === 0 ? chunks[index] : "<" + chunks[index] + ">");
      }
    }

    return result;
  };

  var MAX_LENGTH = 140;

  // Characters not allowed in Tweets
  var INVALID_CHARACTERS = [
    // BOM
    fromCode(0xFFFE),
    fromCode(0xFEFF),

    // Special
    fromCode(0xFFFF),

    // Directional Change
    fromCode(0x202A),
    fromCode(0x202B),
    fromCode(0x202C),
    fromCode(0x202D),
    fromCode(0x202E)
  ];

  // Returns the length of Tweet text with consideration to t.co URL replacement
  // and chars outside the basic multilingual plane that use 2 UTF16 code points
  twttr.txt.getTweetLength = function(text, options) {
    if (!options) {
      options = {
          // These come from https://api.twitter.com/1/help/configuration.json
          // described by https://dev.twitter.com/docs/api/1/get/help/configuration
          short_url_length: 22,
          short_url_length_https: 23
      };
    }
    var textLength = twttr.txt.getUnicodeTextLength(text),
        urlsWithIndices = twttr.txt.extractUrlsWithIndices(text);
    twttr.txt.modifyIndicesFromUTF16ToUnicode(text, urlsWithIndices);

    for (var i = 0; i < urlsWithIndices.length; i++) {
    	// Subtract the length of the original URL
      textLength += urlsWithIndices[i].indices[0] - urlsWithIndices[i].indices[1];

      // Add 23 characters for URL starting with https://
      // Otherwise add 22 characters
      if (urlsWithIndices[i].url.toLowerCase().match(twttr.txt.regexen.urlHasHttps)) {
         textLength += options.short_url_length_https;
      } else {
        textLength += options.short_url_length;
      }
    }

    return textLength;
  };

  // Check the text for any reason that it may not be valid as a Tweet. This is meant as a pre-validation
  // before posting to api.twitter.com. There are several server-side reasons for Tweets to fail but this pre-validation
  // will allow quicker feedback.
  //
  // Returns false if this text is valid. Otherwise one of the following strings will be returned:
  //
  //   "too_long": if the text is too long
  //   "empty": if the text is nil or empty
  //   "invalid_characters": if the text contains non-Unicode or any of the disallowed Unicode characters
  twttr.txt.isInvalidTweet = function(text) {
    if (!text) {
      return "empty";
    }

    // Determine max length independent of URL length
    if (twttr.txt.getTweetLength(text) > MAX_LENGTH) {
      return "too_long";
    }

    for (var i = 0; i < INVALID_CHARACTERS.length; i++) {
      if (text.indexOf(INVALID_CHARACTERS[i]) >= 0) {
        return "invalid_characters";
      }
    }

    return false;
  };

  twttr.txt.isValidTweetText = function(text) {
    return !twttr.txt.isInvalidTweet(text);
  };

  twttr.txt.isValidUsername = function(username) {
    if (!username) {
      return false;
    }

    var extracted = twttr.txt.extractMentions(username);

    // Should extract the username minus the @ sign, hence the .slice(1)
    return extracted.length === 1 && extracted[0] === username.slice(1);
  };

  var VALID_LIST_RE = regexSupplant(/^#{validMentionOrList}$/);

  twttr.txt.isValidList = function(usernameList) {
    var match = usernameList.match(VALID_LIST_RE);

    // Must have matched and had nothing before or after
    return !!(match && match[1] == "" && match[4]);
  };

  twttr.txt.isValidHashtag = function(hashtag) {
    if (!hashtag) {
      return false;
    }

    var extracted = twttr.txt.extractHashtags(hashtag);

    // Should extract the hashtag minus the # sign, hence the .slice(1)
    return extracted.length === 1 && extracted[0] === hashtag.slice(1);
  };

  twttr.txt.isValidUrl = function(url, unicodeDomains, requireProtocol) {
    if (unicodeDomains == null) {
      unicodeDomains = true;
    }

    if (requireProtocol == null) {
      requireProtocol = true;
    }

    if (!url) {
      return false;
    }

    var urlParts = url.match(twttr.txt.regexen.validateUrlUnencoded);

    if (!urlParts || urlParts[0] !== url) {
      return false;
    }

    var scheme = urlParts[1],
        authority = urlParts[2],
        path = urlParts[3],
        query = urlParts[4],
        fragment = urlParts[5];

    if (!(
      (!requireProtocol || (isValidMatch(scheme, twttr.txt.regexen.validateUrlScheme) && scheme.match(/^https?$/i))) &&
      isValidMatch(path, twttr.txt.regexen.validateUrlPath) &&
      isValidMatch(query, twttr.txt.regexen.validateUrlQuery, true) &&
      isValidMatch(fragment, twttr.txt.regexen.validateUrlFragment, true)
    )) {
      return false;
    }

    return (unicodeDomains && isValidMatch(authority, twttr.txt.regexen.validateUrlUnicodeAuthority)) ||
           (!unicodeDomains && isValidMatch(authority, twttr.txt.regexen.validateUrlAuthority));
  };

  function isValidMatch(string, regex, optional) {
    if (!optional) {
      // RegExp["$&"] is the text of the last match
      // blank strings are ok, but are falsy, so we check stringiness instead of truthiness
      return ((typeof string === "string") && string.match(regex) && RegExp["$&"] === string);
    }

    // RegExp["$&"] is the text of the last match
    return (!string || (string.match(regex) && RegExp["$&"] === string));
  }

  if (typeof module != 'undefined' && module.exports) {
    module.exports = twttr.txt;
  }

  if (typeof window != 'undefined') {
    if (window.twttr) {
      for (var prop in twttr) {
        window.twttr[prop] = twttr[prop];
      }
    } else {
      window.twttr = twttr;
    }
  }
})();

// End of the imported code from twitter-text.js.

// Create the exported symbol for the JavaScript module.
var twttr = window.twttr;