Make dart2js String.split match the VM version.

sdk/lib/core/string.dart

 // Copyright (c) 2012, the Dart project authors.  Please see the AUTHORS file
 // for details. All rights reserved. Use of this source code is governed by a
 // BSD-style license that can be found in the LICENSE file.
 
 part of dart.core;
 
 /**
  * A sequence of characters.
  *
  * A string can be either single or multiline. Single line strings are
@@ skipping 445 matching lines @@
    *
    *     pigLatin(String words) => words.replaceAllMapped(
    *         new RegExp(r'\b(\w*?)([aeiou]\w*)', caseSensitive: false),
    *         (Match m) => "${m[2]}${m[1]}${m[1].isEmpty ? 'way' : 'ay'}");
    *
    *     pigLatin('I have a secret now!'); // 'Iway avehay away ecretsay ownay!'
    */
   String replaceAllMapped(Pattern from, String replace(Match match));
 
   /**
-   * Splits the string at matches of [pattern]. Returns
-   * a list of substrings.
+   * Splits the string at matches of [pattern] and returns a list of substrings.
    *
-   * Splitting with an empty string pattern (`''`) splits at UTF-16 code unit
-   * boundaries and not at rune boundaries:
+   * Finds all the matches of `pattern` in this string,
+   * and returns the list of the substrings between the matches.
+   *
+   *     var string = "Hello world!";
+   *     string.split(" ");                      // ['Hello', 'world!'];
+   *
+   * Empty matches at the beginning and end of the strings are ignored,
+   * and so are empty matches right after another match.
+   *
+   *     var string = "abba";
+   *     string.split(new RegExp(r"b*"));        // ['a', 'a']
+   *                                             // not ['', 'a', 'a', '']
+   *
+   * If this string is empty, the result is an empty list if `pattern` matches
+   * the empty string, and it is `[""]` if the pattern doesn't match.
+   *
+   *     var string = '';
+   *     string.split('');                       // []
+   *     string.split("a");                      // ['']
+   *
+   * Splitting with an empty pattern splits the string into single-code unit
+   * strings.
    *
    *     var string = 'Pub';
    *     string.split('');                       // ['P', 'u', 'b']
    *
    *     string.codeUnits.map((unit) {
    *       return new String.fromCharCode(unit);
    *     }).toList();                            // ['P', 'u', 'b']
    *
+   * Splitting happens at UTF-16 code unit boundaries,
+   * and not at rune boundaries:
+   *
    *     // String made up of two code units, but one rune.
    *     string = '\u{1D11E}';
-   *     string.split('').length;                 // 2
+   *     string.split('').length;                 // 2 surrogate values
    *
-   * You should [map] the runes unless you are certain that the string is in
-   * the basic multilingual plane (meaning that each code unit represents a
-   * rune):
+   * To get a list of strings containing the individual runes of a string,
+   * you should not use split. You can instead map each rune to a string
+   * as follows:
    *
-   *     string.runes.map((rune) => new String.fromCharCode(rune));
+   *     string.runes.map((rune) => new String.fromCharCode(rune)).toList();
    */
   List<String> split(Pattern pattern);
 
   /**
    * Splits the string, converts its parts, and combines them into a new
    * string.
    *
    * [pattern] is used to split the string into parts and separating matches.
    *
    * Each match is converted to a string by calling [onMatch]. If [onMatch]
@@ skipping 244 matching lines @@
         _position = position - 1;
         _currentCodePoint = _combineSurrogatePair(prevCodeUnit, codeUnit);
         return true;
       }
     }
     _position = position;
     _currentCodePoint = codeUnit;
     return true;
   }
 }