/* This Source Code Form is subject to the terms of the Mozilla Public
 * License, v. 2.0. If a copy of the MPL was not distributed with this
 * file, You can obtain one at http://mozilla.org/MPL/2.0/. */

this.EXPORTED_SYMBOLS = ["GetAttachmentKeywords"];

/**
 * Check whether the character is a CJK character or not.
 *
 * @return true if it is a CJK character.
 */
function IsCJK(code)
{
  if (code >= 0x2000 && code <= 0x9fff) {
    // Hiragana, Katakana and Kanaji
    return true;
  }
  else if (code >= 0xac00 && code <= 0xd7ff) {
    // Hangul
    return true;
  }
  else if (code >= 0xf900 && code <= 0xffff) {
    // Hiragana, Katakana and Kanaji
    return true;
  }
  return false;
}

/**
 * Get the (possibly-empty) list of attachment keywords in this message.
 *
 * @return the (possibly-empty) list of attachment keywords in this message
 **/
function GetAttachmentKeywords(mailData,keywordsInCsv)
{
  // The empty string would get split to an array of size 1.  Avoid that...
  var keywordsArray = (keywordsInCsv) ? keywordsInCsv.split(",") : [];

  function escapeRegxpSpecials(inputString)
  {
    const specials = [".", "\\", "^", "$", "*", "+", "?", "|",
                      "(", ")" , "[", "]", "{", "}"];
    var re = new RegExp("(\\"+specials.join("|\\")+")", "g");
    inputString = inputString.replace(re, "\\$1");
    return inputString.replace(" ", "\\s+");
  }

  // NOT_W is the character class that isn't in the Unicode classes "Ll",
  // "Lu" and "Lt".  It should work like \W, if \W knew about Unicode.
  const NOT_W = "[^\\u0041-\\u005a\\u0061-\\u007a\\u00aa\\u00b5\\u00ba\\u00c0-\\u00d6\\u00d8-\\u00f6\\u00f8-\\u01ba\\u01bc-\\u01bf\\u01c4-\\u02ad\\u0386\\u0388-\\u0481\\u048c-\\u0556\\u0561-\\u0587\\u10a0-\\u10c5\\u1e00-\\u1fbc\\u1fbe\\u1fc2-\\u1fcc\\u1fd0-\\u1fdb\\u1fe0-\\u1fec\\u1ff2-\\u1ffc\\u207f\\u2102\\u2107\\u210a-\\u2113\\u2115\\u2119-\\u211d\\u2124\\u2126\\u2128\\u212a-\\u212d\\u212f-\\u2131\\u2133\\u2134\\u2139\\ufb00-\\ufb17\\uff21-\\uff3a\\uff41-\\uff5a]";

  var keywordsFound = [];
  for (var i = 0; i < keywordsArray.length; i++) {
    var kw = escapeRegxpSpecials(keywordsArray[i]);
    // If the keyword starts (ends) with a CJK character, we don't care
    // what the previous (next) character is, because the words aren't
    // space delimited.
    if (keywordsArray[i].charAt(0) == ".") { // like .pdf
      // For this case we want to match the whole document name.
      var start = "(([^\\s]*)\\b)";
      var end = IsCJK(kw.charCodeAt(kw.length - 1)) ? "" : "(\\s|$)";
      var re = new RegExp(start + kw + end, "ig");
      var matching = mailData.match(re);
      if (matching) {
        for (var j = 0; j < matching.length; j++) {
          // Ignore the match if it was in a URL.
          if (!(/^(https?|ftp):\/\//i.test(matching[j]))) {
            // We can have several *different* matches for one dot-keyword.
            // E.g. foo.pdf and bar.pdf would both match for .pdf.
            var m = matching[j].trim();
            if (!keywordsFound.includes(m))
              keywordsFound.push(m);
          }
        }
      }
    }
    else {
      var start = IsCJK(kw.charCodeAt(0)) ? "" : ("((^|\\s)\\S*)");
      var end = IsCJK(kw.charCodeAt(kw.length - 1)) ? "" : "(" + NOT_W + "|$)";
      var re = new RegExp(start + kw + end, "ig");
      var matching;
      while ((matching = re.exec(mailData)) !== null) {
        // Ignore the match if it was in a URL.
        if (!(/^(https?|ftp):\/\//i.test(matching[0].trim()))) {
          keywordsFound.push(keywordsArray[i]);
          break;
        }
      }
    }
  }
  return keywordsFound;
}

var onmessage = function(event)
{
  var keywordsFound = GetAttachmentKeywords(event.data[0], event.data[1]);
  postMessage(keywordsFound);
};