﻿if (!window.twttr) {
    window.twttr = {};
}

(function () {
    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);
    }

    // simple string interpolation
    function stringSupplant(str, values) {
        return str.replace(/#\{(\w+)\}/g, function (match, name) {
            return values[name] || "";
        });
    }

    function addCharsToCharClass(charClass, start, end) {
        var s = String.fromCharCode(start);
        if (end !== start) {
            s += "-" + String.fromCharCode(end);
        }
        charClass.push(s);
        return charClass;
    }

    // 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.atSigns = /[@＠]/;
    twttr.txt.regexen.extractMentions = regexSupplant(/(^|[^a-zA-Z0-9_])(#{atSigns})([a-zA-Z0-9_]{1,20})/g);
    twttr.txt.regexen.extractReply = regexSupplant(/^(?:#{spaces})*#{atSigns}([a-zA-Z0-9_]{1,20})/);
    twttr.txt.regexen.listName = /[a-zA-Z][a-zA-Z0-9_\-\u0080-\u00ff]{0,24}/;
    twttr.txt.regexen.extractMentionsOrLists = regexSupplant(/(^|[^a-zA-Z0-9_])(#{atSigns})([a-zA-Z0-9_]{1,20})(\/[a-zA-Z][a-zA-Z0-9_\-]{0,24})?/g);

    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
    // 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, 0x3005, 0x3005); // Kanji iteration mark
    addCharsToCharClass(nonLatinHashtagChars, 0x303B, 0x303B); // Han iteration mark

    twttr.txt.regexen.nonLatinHashtagChars = regexSupplant(nonLatinHashtagChars.join(""));
    // Latin accented characters (subtracted 0xD7 from the range, it's a confusable multiplication sign. Looks like "x")
    twttr.txt.regexen.latinAccentChars = regexSupplant("ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõöøùúûüýþş\\303\\277");

    twttr.txt.regexen.endScreenNameMatch = regexSupplant(/^(?:#{atSigns}|[#{latinAccentChars}]|:\/\/)/);

    // A hashtag must contain characters, numbers and underscores, but not all numbers.
    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 = /^(?:[#＃]|:\/\/)/;
    twttr.txt.regexen.hashtagBoundary = regexSupplant(/(?:^|$|[^&\/a-z0-9_#{latinAccentChars}#{nonLatinHashtagChars}])/);
    twttr.txt.regexen.autoLinkHashtags = regexSupplant(/(#{hashtagBoundary})(#|＃)(#{hashtagAlphaNumeric}*#{hashtagAlpha}#{hashtagAlphaNumeric}*)/gi);
    twttr.txt.regexen.autoLinkUsernamesOrLists = /(^|[^a-zA-Z0-9_]|RT:?)([@＠]+)([a-zA-Z0-9_]{1,20})(\/[a-zA-Z][a-zA-Z0-9_\-]{0,24})?/g;
    twttr.txt.regexen.autoLinkEmoticon = /(8\-\#|8\-E|\+\-\(|\`\@|\`O|\&lt;\|:~\(|\}:o\{|:\-\[|\&gt;o\&lt;|X\-\/|\[:-\]\-I\-|\/\/\/\/Ö\\\\\\\\|\(\|:\|\/\)|∑:\*\)|\( \| \))/g;

    // URL related hash regex collection
    twttr.txt.regexen.validPrecedingChars = regexSupplant(/(?:[^-\/"'!=A-Za-z0-9_@＠$#＃\.#{invalid_chars_group}]|^)/);

    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)(?=[^a-zA-Z]|$))/);
    twttr.txt.regexen.validCCTLD = regexSupplant(/(?:(?:ac|ad|ae|af|ag|ai|al|am|an|ao|aq|ar|as|at|au|aw|ax|az|ba|bb|bd|be|bf|bg|bh|bi|bj|bm|bn|bo|br|bs|bt|bv|bw|by|bz|ca|cc|cd|cf|cg|ch|ci|ck|cl|cm|cn|co|cr|cs|cu|cv|cx|cy|cz|dd|de|dj|dk|dm|do|dz|ec|ee|eg|eh|er|es|et|eu|fi|fj|fk|fm|fo|fr|ga|gb|gd|ge|gf|gg|gh|gi|gl|gm|gn|gp|gq|gr|gs|gt|gu|gw|gy|hk|hm|hn|hr|ht|hu|id|ie|il|im|in|io|iq|ir|is|it|je|jm|jo|jp|ke|kg|kh|ki|km|kn|kp|kr|kw|ky|kz|la|lb|lc|li|lk|lr|ls|lt|lu|lv|ly|ma|mc|md|me|mg|mh|mk|ml|mm|mn|mo|mp|mq|mr|ms|mt|mu|mv|mw|mx|my|mz|na|nc|ne|nf|ng|ni|nl|no|np|nr|nu|nz|om|pa|pe|pf|pg|ph|pk|pl|pm|pn|pr|ps|pt|pw|py|qa|re|ro|rs|ru|rw|sa|sb|sc|sd|se|sg|sh|si|sj|sk|sl|sm|sn|so|sr|ss|st|su|sv|sy|sz|tc|td|tf|tg|th|tj|tk|tl|tm|tn|to|tp|tr|tt|tv|tw|tz|ua|ug|uk|us|uy|uz|va|vc|ve|vg|vi|vn|vu|wf|ws|ye|yt|za|zm|zw)(?=[^a-zA-Z]|$))/);
    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
      '(#{validPrecedingChars})' + // $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;

    // 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 URLs
    var DEFAULT_URL_CLASS = "tweet-url";
    // Default CSS class for auto-linked lists (along with the url class)
    var DEFAULT_LIST_CLASS = "list-slug";
    // Default CSS class for auto-linked usernames (along with the url class)
    var DEFAULT_USERNAME_CLASS = "username";
    // Default CSS class for auto-linked hashtags (along with the url class)
    var DEFAULT_HASHTAG_CLASS = "hashtag";
    // HTML attribute for robot nofollow behavior (default)
    var HTML_ATTR_NO_FOLLOW = " rel=\"nofollow\"";

    // 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.autoLink = function (text, options) {
        options = clone(options || {});
        return twttr.txt.autoLinkUsernamesOrLists(
      twttr.txt.autoLinkUrlsCustom(
        twttr.txt.autoLinkHashtags(text, options),
      options),
    options);
    };


    twttr.txt.autoLinkUsernamesOrLists = function (text, options) {
        options = clone(options || {});

        options.urlClass = options.urlClass || DEFAULT_URL_CLASS;
        options.listClass = options.listClass || DEFAULT_LIST_CLASS;
        options.usernameClass = options.usernameClass || DEFAULT_USERNAME_CLASS;
        options.usernameUrlBase = options.usernameUrlBase || "http://twitter.com/";
        options.listUrlBase = options.listUrlBase || "http://twitter.com/";
        if (!options.suppressNoFollow) {
            var extraHtml = HTML_ATTR_NO_FOLLOW;
        }

        var newText = "",
        splitText = twttr.txt.splitTags(text);

        for (var index = 0; index < splitText.length; index++) {
            var chunk = splitText[index];

            if (index !== 0) {
                newText += ((index % 2 === 0) ? ">" : "<");
            }

            if (index % 4 !== 0) {
                newText += chunk;
            } else {
                newText += chunk.replace(twttr.txt.regexen.autoLinkUsernamesOrLists, function (match, before, at, user, slashListname, offset, chunk) {
                    var after = chunk.slice(offset + match.length);

                    var d = {
                        before: before,
                        at: at,
                        user: twttr.txt.htmlEscape(user),
                        slashListname: twttr.txt.htmlEscape(slashListname),
                        extraHtml: extraHtml,
                        preChunk: "",
                        chunk: twttr.txt.htmlEscape(chunk),
                        postChunk: ""
                    };
                    for (var k in options) {
                        if (options.hasOwnProperty(k)) {
                            d[k] = options[k];
                        }
                    }

                    if (slashListname && !options.suppressLists) {
                        // the link is a list
                        var list = d.chunk = stringSupplant("#{user}#{slashListname}", d);
                        d.list = twttr.txt.htmlEscape(list.toLowerCase());
                        return stringSupplant("#{before}#{at}<a class=\"#{urlClass} #{listClass}\" href=\"#{listUrlBase}#{list}\"#{extraHtml}>#{preChunk}#{chunk}#{postChunk}</a>", d);
                    } else {
                        if (after && after.match(twttr.txt.regexen.endScreenNameMatch)) {
                            // Followed by something that means we don't autolink
                            return match;
                        } else {
                            // this is a screen name
                            d.chunk = twttr.txt.htmlEscape(user);
                            d.dataScreenName = !options.suppressDataScreenName ? stringSupplant("data-screen-name=\"#{chunk}\" ", d) : "";
                            return stringSupplant("#{before}#{at}<a class=\"#{urlClass} #{usernameClass}\" #{dataScreenName}href=\"#{usernameUrlBase}#{chunk}\"#{extraHtml}>#{preChunk}#{chunk}#{postChunk}</a>", d);
                        }
                    }
                });
            }
        }

        return newText;
    };

    twttr.txt.autoLinkHashtags = function (text, options) {
        options = clone(options || {});
        options.urlClass = options.urlClass || DEFAULT_URL_CLASS;
        options.hashtagClass = options.hashtagClass || DEFAULT_HASHTAG_CLASS;
        options.hashtagUrlBase = options.hashtagUrlBase || "http://twitter.com/search?q=%23";
        if (!options.suppressNoFollow) {
            var extraHtml = HTML_ATTR_NO_FOLLOW;
        }

        return text.replace(twttr.txt.regexen.autoLinkHashtags, function (match, before, hash, text, offset, chunk) {
            var after = chunk.slice(offset + match.length);
            if (after.match(twttr.txt.regexen.endHashtagMatch))
                return match;

            var d = {
                before: before,
                hash: twttr.txt.htmlEscape(hash),
                preText: "",
                text: twttr.txt.htmlEscape(text),
                postText: "",
                extraHtml: extraHtml
            };

            for (var k in options) {
                if (options.hasOwnProperty(k)) {
                    d[k] = options[k];
                }
            }

            return stringSupplant("#{before}<a href=\"#{hashtagUrlBase}#{text}\" title=\"##{text}\" class=\"#{urlClass} #{hashtagClass}\"#{extraHtml}>#{hash}#{preText}#{text}#{postText}</a>", d);
        });
    };


    twttr.txt.autoLinkUrlsCustom = function (text, options) {
        options = clone(options || {});
        if (!options.suppressNoFollow) {
            options.rel = "nofollow";
        }
        if (options.urlClass) {
            options["class"] = options.urlClass;
            delete options.urlClass;
        }

        // 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];
            }
        }

        delete options.suppressNoFollow;
        delete options.suppressDataScreenName;
        delete options.listClass;
        delete options.usernameClass;
        delete options.usernameUrlBase;
        delete options.listUrlBase;

        return text.replace(twttr.txt.regexen.extractUrl, function (match, all, before, url, protocol, port, domain, path, queryString) {
            var tldComponents;

            if (protocol) {
                var htmlAttrs = "";
                var after = "";
                for (var k in options) {
                    htmlAttrs += stringSupplant(" #{k}=\"#{v}\" ", { k: k, v: options[k].toString().replace(/"/, "&quot;").replace(/</, "&lt;").replace(/>/, "&gt;") });
                }

                // In the case of t.co URLs, don't allow additional path characters.
                if (url.match(twttr.txt.regexen.validTcoUrl)) {
                    url = RegExp.lastMatch;
                    after = RegExp.rightContext;
                }

                var d = {
                    before: before,
                    htmlAttrs: htmlAttrs,
                    url: twttr.txt.htmlEscape(url),
                    after: after
                };
                if (urlEntities && urlEntities[url] && urlEntities[url].display_url) {
                    d.displayUrl = twttr.txt.htmlEscape(urlEntities[url].display_url);
                } else {
                    d.displayUrl = d.url;
                }

                return stringSupplant("#{before}<a href=\"#{url}\"#{htmlAttrs}>#{displayUrl}</a>#{after}", d);
            } else {
                return all;
            }
        });
    };

    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) {
        if (!text) {
            return [];
        }

        var possibleScreenNames = [],
        position = 0;

        text.replace(twttr.txt.regexen.extractMentions, function (match, before, atSign, screenName, offset, chunk) {
            var after = chunk.slice(offset + match.length);
            if (!after.match(twttr.txt.regexen.endScreenNameMatch)) {
                var startPosition = text.indexOf(atSign + screenName, position);
                position = startPosition + screenName.length + 1;
                possibleScreenNames.push({
                    screenName: screenName,
                    indices: [startPosition, position]
                });
            }
        });

        return possibleScreenNames;
    };

    /**
    * Extract list or user mentions.
    * (Presence of listSlug indicates a list)
    */
    twttr.txt.extractMentionsOrListsWithIndices = function (text) {
        if (!text) {
            return [];
        }

        var possibleNames = [],
        position = 0;

        text.replace(twttr.txt.regexen.extractMentionsOrLists, function (match, before, atSign, screenName, slashListname, offset, chunk) {
            var after = chunk.slice(offset + match.length);
            if (!after.match(twttr.txt.regexen.endScreenNameMatch)) {
                slashListname = slashListname || '';
                var startPosition = text.indexOf(atSign + screenName + slashListname, position);
                position = startPosition + screenName.length + slashListname.length + 1;
                possibleNames.push({
                    screenName: screenName,
                    listSlug: slashListname,
                    indices: [startPosition, position]
                });
            }
        });

        return possibleNames;
    };


    twttr.txt.extractReplies = function (text) {
        if (!text) {
            return null;
        }

        var possibleScreenName = text.match(twttr.txt.regexen.extractReply);
        if (!possibleScreenName ||
        RegExp.rightContext.match(twttr.txt.regexen.endScreenNameMatch)) {
            return null;
        }

        return possibleScreenName[1];
    };

    twttr.txt.extractUrls = function (text) {
        var urlsOnly = [],
        urlsWithIndices = twttr.txt.extractUrlsWithIndices(text);

        for (var i = 0; i < urlsWithIndices.length; i++) {
            urlsOnly.push(urlsWithIndices[i].url);
        }

        return urlsOnly;
    };

    twttr.txt.extractUrlsWithIndices = function (text) {
        if (!text) {
            return [];
        }

        var urls = [],
        position = 0;

        text.replace(twttr.txt.regexen.extractUrl, function (match, all, before, url, protocol, domain, port, path, query) {
            var startPosition = text.indexOf(url, position),
          endPosition = startPosition + url.length;

            // if protocol is missing and domain contains non-ASCII characters,
            // extract ASCII-only domains.
            if (!protocol) {
                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) {
                    return;
                }

                // 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) {
        if (!text) {
            return [];
        }

        var tags = [],
        position = 0;

        text.replace(twttr.txt.regexen.autoLinkHashtags, function (match, before, hash, hashText, offset, chunk) {
            var after = chunk.slice(offset + match.length);
            if (after.match(twttr.txt.regexen.endHashtagMatch))
                return;
            var startPosition = text.indexOf(hash + hashText, position);
            position = startPosition + hashText.length + 1;
            tags.push({
                hashtag: hashText,
                indices: [startPosition, position]
            });
        });

        return tags;
    };

    // 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),
        split,
        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)
  ];

    // 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";
        }

        if (text.length > 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(/^#{autoLinkUsernamesOrLists}$/);

    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;
    }

} ());

