Rework ascii and unicode helpers.

lodash.js

@@ -150,7 +150,7 @@
       reSplitDetails = /,? & /;
 
   /** Used to match non-compound words composed of alphanumeric characters. */
-  var reBasicWord = /[a-zA-Z0-9]+/g;
+  var reAsciiWord = /[a-zA-Z0-9]+/g;
 
   /** Used to match backslashes in property paths. */
   var reEscapeChar = /\\(\\)?/g;
@@ -244,10 +244,10 @@
   var reComboMark = RegExp(rsCombo, 'g');
 
   /** Used to match [string symbols](https://mathiasbynens.be/notes/javascript-unicode). */
-  var reComplexSymbol = RegExp(rsFitz + '(?=' + rsFitz + ')|' + rsSymbol + rsSeq, 'g');
+  var reUnicode = RegExp(rsFitz + '(?=' + rsFitz + ')|' + rsSymbol + rsSeq, 'g');
 
   /** Used to match complex or compound words. */
-  var reComplexWord = RegExp([
+  var reUnicodeWord = RegExp([
     rsUpper + '?' + rsLower + '+' + rsOptLowerContr + '(?=' + [rsBreak, rsUpper, '$'].join('|') + ')',
     rsUpperMisc + '+' + rsOptUpperContr + '(?=' + [rsBreak, rsUpper + rsLowerMisc, '$'].join('|') + ')',
     rsUpper + '?' + rsLowerMisc + '+' + rsOptLowerContr,
@@ -257,10 +257,10 @@
   ].join('|'), 'g');
 
   /** Used to detect strings with [zero-width joiners or code points from the astral planes](http://eev.ee/blog/2015/09/12/dark-corners-of-unicode/). */
-  var reHasComplexSymbol = RegExp('[' + rsZWJ + rsAstralRange  + rsComboMarksRange + rsComboSymbolsRange + rsVarRange + ']');
+  var reHasUnicode = RegExp('[' + rsZWJ + rsAstralRange  + rsComboMarksRange + rsComboSymbolsRange + rsVarRange + ']');
 
   /** Used to detect strings that need a more robust regexp to match words. */
-  var reHasComplexWord = /[a-z][A-Z]|[A-Z]{2,}[a-z]|[0-9][a-zA-Z]|[a-zA-Z][0-9]|[^a-zA-Z0-9 ]/;
+  var reHasUnicodeWord = /[a-z][A-Z]|[A-Z]{2,}[a-z]|[0-9][a-zA-Z]|[a-zA-Z][0-9]|[^a-zA-Z0-9 ]/;
 
   /** Used to assign default `context` object properties. */
   var contextProps = [
@@ -560,7 +560,7 @@
    * specifying an index to search from.
    *
    * @private
-   * @param {Array} [array] The array to search.
+   * @param {Array} [array] The array to inspect.
    * @param {*} target The value to search for.
    * @returns {boolean} Returns `true` if `target` is found, else `false`.
    */
@@ -573,7 +573,7 @@
    * This function is like `arrayIncludes` except that it accepts a comparator.
    *
    * @private
-   * @param {Array} [array] The array to search.
+   * @param {Array} [array] The array to inspect.
    * @param {*} target The value to search for.
    * @param {Function} comparator The comparator invoked per element.
    * @returns {boolean} Returns `true` if `target` is found, else `false`.
@@ -699,13 +699,44 @@
     return false;
   }
 
+  /**
+   * Gets the size of an ASCII `string`.
+   *
+   * @private
+   * @param {string} string The string inspect.
+   * @returns {number} Returns the string size.
+   */
+  var asciiSize = baseProperty('length');
+
+  /**
+   * Converts an ASCII `string` to an array.
+   *
+   * @private
+   * @param {string} string The string to convert.
+   * @returns {Array} Returns the converted array.
+   */
+  function asciiToArray(string) {
+    return string.split('');
+  }
+
+  /**
+   * Splits an ASCII `string` into an array of its words.
+   *
+   * @private
+   * @param {string} The string to inspect.
+   * @returns {Array} Returns the words of `string`.
+   */
+  function asciiWords(string) {
+    return string.match(reAsciiWord) || [];
+  }
+
   /**
    * The base implementation of methods like `_.findKey` and `_.findLastKey`,
    * without support for iteratee shorthands, which iterates over `collection`
    * using `eachFunc`.
    *
    * @private
-   * @param {Array|Object} collection The collection to search.
+   * @param {Array|Object} collection The collection to inspect.
    * @param {Function} predicate The function invoked per iteration.
    * @param {Function} eachFunc The function to iterate over `collection`.
    * @returns {*} Returns the found element or its key, else `undefined`.
@@ -726,7 +757,7 @@
    * support for iteratee shorthands.
    *
    * @private
-   * @param {Array} array The array to search.
+   * @param {Array} array The array to inspect.
    * @param {Function} predicate The function invoked per iteration.
    * @param {number} fromIndex The index to search from.
    * @param {boolean} [fromRight] Specify iterating from right to left.
@@ -748,7 +779,7 @@
    * The base implementation of `_.indexOf` without `fromIndex` bounds checks.
    *
    * @private
-   * @param {Array} array The array to search.
+   * @param {Array} array The array to inspect.
    * @param {*} value The value to search for.
    * @param {number} fromIndex The index to search from.
    * @returns {number} Returns the index of the matched value, else `-1`.
@@ -772,7 +803,7 @@
    * This function is like `baseIndexOf` except that it accepts a comparator.
    *
    * @private
-   * @param {Array} array The array to search.
+   * @param {Array} array The array to inspect.
    * @param {*} value The value to search for.
    * @param {number} fromIndex The index to search from.
    * @param {Function} comparator The comparator invoked per element.
@@ -1075,6 +1106,28 @@
     return object == null ? undefined : object[key];
   }
 
+  /**
+   * Checks if `string` contains Unicode symbols.
+   *
+   * @private
+   * @param {string} string The string to inspect.
+   * @returns {boolean} Returns `true` if a symbol is found, else `false`.
+   */
+  function hasUnicode(string) {
+    return reHasUnicode.test(string);
+  }
+
+  /**
+   * Checks if `string` contains a word composed of Unicode symbols.
+   *
+   * @private
+   * @param {string} string The string to inspect.
+   * @returns {boolean} Returns `true` if a word is found, else `false`.
+   */
+  function hasUnicodeWord(string) {
+    return reHasUnicodeWord.test(string);
+  }
+
   /**
    * Checks if `value` is a host object in IE < 9.
    *
@@ -1209,14 +1262,9 @@
    * @returns {number} Returns the string size.
    */
   function stringSize(string) {
-    if (!(string && reHasComplexSymbol.test(string))) {
+    return hasUnicode(string)
-      return string.length;
+      ? unicodeSize(string)
-    }
+      : asciiSize(string);
-    var result = reComplexSymbol.lastIndex = 0;
-    while (reComplexSymbol.test(string)) {
-      result++;
-    }
-    return result;
   }
 
   /**
@@ -1227,7 +1275,9 @@
    * @returns {Array} Returns the converted array.
    */
   function stringToArray(string) {
-    return string.match(reComplexSymbol);
+    return hasUnicode(string)
+      ? unicodeToArray(string)
+      : asciiToArray(string);
   }
 
   /**
@@ -1239,6 +1289,43 @@
    */
   var unescapeHtmlChar = basePropertyOf(htmlUnescapes);
 
+  /**
+   * Gets the size of a Unicode `string`.
+   *
+   * @private
+   * @param {string} string The string inspect.
+   * @returns {number} Returns the string size.
+   */
+  function unicodeSize(string) {
+    var result = reUnicode.lastIndex = 0;
+    while (reUnicode.test(string)) {
+      result++;
+    }
+    return result;
+  }
+
+  /**
+   * Converts a Unicode `string` to an array.
+   *
+   * @private
+   * @param {string} string The string to convert.
+   * @returns {Array} Returns the converted array.
+   */
+  function unicodeToArray(string) {
+    return string.match(reUnicode) || [];
+  }
+
+  /**
+   * Splits a Unicode `string` into an array of its words.
+   *
+   * @private
+   * @param {string} The string to inspect.
+   * @returns {Array} Returns the words of `string`.
+   */
+  function unicodeWords(string) {
+    return string.match(reUnicodeWord) || [];
+  }
+
   /*--------------------------------------------------------------------------*/
 
   /**
@@ -1278,19 +1365,23 @@
    * var defer = _.runInContext({ 'setTimeout': setImmediate }).defer;
    */
   function runInContext(context) {
-    context = context ? _.defaults({}, context, _.pick(root, contextProps)) : root;
+    context = context ? _.defaults(root.Object(), context, _.pick(root, contextProps)) : root;
 
     /** Built-in constructor references. */
     var Array = context.Array,
+        Date = context.Date,
         Error = context.Error,
+        Function = context.Function,
         Math = context.Math,
+        Object = context.Object,
         RegExp = context.RegExp,
         TypeError = context.TypeError;
 
     /** Used for built-in method references. */
-    var arrayProto = context.Array.prototype,
+    var arrayProto = Array.prototype,
-        objectProto = context.Object.prototype,
+        funcProto = Function.prototype,
-        stringProto = context.String.prototype;
+        objectProto = Object.prototype,
+        stringProto = String.prototype;
 
     /** Used to detect overreaching core-js shims. */
     var coreJsData = context['__core-js_shared__'];
@@ -1302,7 +1393,7 @@
     }());
 
     /** Used to resolve the decompiled source of functions. */
-    var funcToString = context.Function.prototype.toString;
+    var funcToString = funcProto.toString;
 
     /** Used to check objects for own properties. */
     var hasOwnProperty = objectProto.hasOwnProperty;
@@ -1335,14 +1426,14 @@
         Uint8Array = context.Uint8Array,
         getPrototype = overArg(Object.getPrototypeOf, Object),
         iteratorSymbol = Symbol ? Symbol.iterator : undefined,
-        objectCreate = context.Object.create,
+        objectCreate = Object.create,
         propertyIsEnumerable = objectProto.propertyIsEnumerable,
         splice = arrayProto.splice,
         spreadableSymbol = Symbol ? Symbol.isConcatSpreadable : undefined;
 
     /** Mocked built-ins. */
     var ctxClearTimeout = context.clearTimeout !== root.clearTimeout && context.clearTimeout,
-        ctxNow = context.Date && context.Date.now !== root.Date.now && context.Date.now,
+        ctxNow = Date && Date.now !== root.Date.now && Date.now,
         ctxSetTimeout = context.setTimeout !== root.setTimeout && context.setTimeout;
 
     /* Built-in method references for those with the same name as other `lodash` methods. */
@@ -1357,9 +1448,7 @@
         nativeMin = Math.min,
         nativeParseInt = context.parseInt,
         nativeRandom = Math.random,
-        nativeReplace = stringProto.replace,
+        nativeReverse = arrayProto.reverse;
-        nativeReverse = arrayProto.reverse,
-        nativeSplit = stringProto.split;
 
     /* Built-in method references that are verified to be native. */
     var DataView = getNative(context, 'DataView'),
@@ -1367,11 +1456,11 @@
         Promise = getNative(context, 'Promise'),
         Set = getNative(context, 'Set'),
         WeakMap = getNative(context, 'WeakMap'),
-        nativeCreate = getNative(context.Object, 'create');
+        nativeCreate = getNative(Object, 'create');
 
     /* Used to set `toString` methods. */
     var defineProperty = (function() {
-      var func = getNative(context.Object, 'defineProperty'),
+      var func = getNative(Object, 'defineProperty'),
           name = getNative.name;
 
       return (name && name.length > 2) ? func : undefined;
@@ -2280,7 +2369,7 @@
      * Gets the index at which the `key` is found in `array` of key-value pairs.
      *
      * @private
-     * @param {Array} array The array to search.
+     * @param {Array} array The array to inspect.
      * @param {*} key The key to search for.
      * @returns {number} Returns the index of the matched value, else `-1`.
      */
@@ -3705,7 +3794,7 @@
      * The base implementation of `_.set`.
      *
      * @private
-     * @param {Object} object The object to query.
+     * @param {Object} object The object to modify.
      * @param {Array|string} path The path of the property to set.
      * @param {*} value The value to set.
      * @param {Function} [customizer] The function to customize path creation.
@@ -4035,7 +4124,7 @@
      * The base implementation of `_.update`.
      *
      * @private
-     * @param {Object} object The object to query.
+     * @param {Object} object The object to modify.
      * @param {Array|string} path The path of the property to update.
      * @param {Function} updater The function to produce the updated value.
      * @param {Function} [customizer] The function to customize path creation.
@@ -4646,7 +4735,7 @@
       return function(string) {
         string = toString(string);
 
-        var strSymbols = reHasComplexSymbol.test(string)
+        var strSymbols = hasUnicode(string)
           ? stringToArray(string)
           : undefined;
 
@@ -4989,7 +5078,7 @@
         return charsLength ? baseRepeat(chars, length) : chars;
       }
       var result = baseRepeat(chars, nativeCeil(length / stringSize(chars)));
-      return reHasComplexSymbol.test(chars)
+      return hasUnicode(chars)
         ? castSlice(stringToArray(result), 0, length).join('')
         : result.slice(0, length);
     }
@@ -6729,7 +6818,7 @@
      * @memberOf _
      * @since 1.1.0
      * @category Array
-     * @param {Array} array The array to search.
+     * @param {Array} array The array to inspect.
      * @param {Function} [predicate=_.identity]
      *  The function invoked per iteration.
      * @param {number} [fromIndex=0] The index to search from.
@@ -6777,7 +6866,7 @@
      * @memberOf _
      * @since 2.0.0
      * @category Array
-     * @param {Array} array The array to search.
+     * @param {Array} array The array to inspect.
      * @param {Function} [predicate=_.identity]
      *  The function invoked per iteration.
      * @param {number} [fromIndex=array.length-1] The index to search from.
@@ -6946,7 +7035,7 @@
      * @memberOf _
      * @since 0.1.0
      * @category Array
-     * @param {Array} array The array to search.
+     * @param {Array} array The array to inspect.
      * @param {*} value The value to search for.
      * @param {number} [fromIndex=0] The index to search from.
      * @returns {number} Returns the index of the matched value, else `-1`.
@@ -7131,7 +7220,7 @@
      * @memberOf _
      * @since 0.1.0
      * @category Array
-     * @param {Array} array The array to search.
+     * @param {Array} array The array to inspect.
      * @param {*} value The value to search for.
      * @param {number} [fromIndex=array.length-1] The index to search from.
      * @returns {number} Returns the index of the matched value, else `-1`.
@@ -7509,7 +7598,7 @@
      * @memberOf _
      * @since 4.0.0
      * @category Array
-     * @param {Array} array The array to search.
+     * @param {Array} array The array to inspect.
      * @param {*} value The value to search for.
      * @returns {number} Returns the index of the matched value, else `-1`.
      * @example
@@ -7588,7 +7677,7 @@
      * @memberOf _
      * @since 4.0.0
      * @category Array
-     * @param {Array} array The array to search.
+     * @param {Array} array The array to inspect.
      * @param {*} value The value to search for.
      * @returns {number} Returns the index of the matched value, else `-1`.
      * @example
@@ -8717,7 +8806,7 @@
      * @memberOf _
      * @since 0.1.0
      * @category Collection
-     * @param {Array|Object} collection The collection to search.
+     * @param {Array|Object} collection The collection to inspect.
      * @param {Function} [predicate=_.identity]
      *  The function invoked per iteration.
      * @param {number} [fromIndex=0] The index to search from.
@@ -8755,7 +8844,7 @@
      * @memberOf _
      * @since 2.0.0
      * @category Collection
-     * @param {Array|Object} collection The collection to search.
+     * @param {Array|Object} collection The collection to inspect.
      * @param {Function} [predicate=_.identity]
      *  The function invoked per iteration.
      * @param {number} [fromIndex=collection.length-1] The index to search from.
@@ -8950,7 +9039,7 @@
      * @memberOf _
      * @since 0.1.0
      * @category Collection
-     * @param {Array|Object|string} collection The collection to search.
+     * @param {Array|Object|string} collection The collection to inspect.
      * @param {*} value The value to search for.
      * @param {number} [fromIndex=0] The index to search from.
      * @param- {Object} [guard] Enables use as an iteratee for methods like `_.reduce`.
@@ -9383,7 +9472,7 @@
      * @memberOf _
      * @since 0.1.0
      * @category Collection
-     * @param {Array|Object} collection The collection to inspect.
+     * @param {Array|Object|string} collection The collection to inspect.
      * @returns {number} Returns the collection size.
      * @example
      *
@@ -12374,7 +12463,7 @@
      * @memberOf _
      * @since 1.1.0
      * @category Object
-     * @param {Object} object The object to search.
+     * @param {Object} object The object to inspect.
      * @param {Function} [predicate=_.identity] The function invoked per iteration.
      * @returns {string|undefined} Returns the key of the matched element,
      *  else `undefined`.
@@ -12413,7 +12502,7 @@
      * @memberOf _
      * @since 2.0.0
      * @category Object
-     * @param {Object} object The object to search.
+     * @param {Object} object The object to inspect.
      * @param {Function} [predicate=_.identity] The function invoked per iteration.
      * @returns {string|undefined} Returns the key of the matched element,
      *  else `undefined`.
@@ -13682,7 +13771,7 @@
      * @memberOf _
      * @since 3.0.0
      * @category String
-     * @param {string} [string=''] The string to search.
+     * @param {string} [string=''] The string to inspect.
      * @param {string} [target] The string to search for.
      * @param {number} [position=string.length] The position to search up to.
      * @returns {boolean} Returns `true` if `string` ends with `target`,
@@ -14038,7 +14127,7 @@
       var args = arguments,
           string = toString(args[0]);
 
-      return args.length < 3 ? string : nativeReplace.call(string, args[1], args[2]);
+      return args.length < 3 ? string : string.replace(args[1], args[2]);
     }
 
     /**
@@ -14099,11 +14188,11 @@
             (separator != null && !isRegExp(separator))
           )) {
         separator = baseToString(separator);
-        if (separator == '' && reHasComplexSymbol.test(string)) {
+        if (!separator && hasUnicode(string)) {
           return castSlice(stringToArray(string), 0, limit);
         }
       }
-      return nativeSplit.call(string, separator, limit);
+      return string.split(separator, limit);
     }
 
     /**
@@ -14138,7 +14227,7 @@
      * @memberOf _
      * @since 3.0.0
      * @category String
-     * @param {string} [string=''] The string to search.
+     * @param {string} [string=''] The string to inspect.
      * @param {string} [target] The string to search for.
      * @param {number} [position=0] The position to search from.
      * @returns {boolean} Returns `true` if `string` starts with `target`,
@@ -14575,7 +14664,7 @@
       string = toString(string);
 
       var strLength = string.length;
-      if (reHasComplexSymbol.test(string)) {
+      if (hasUnicode(string)) {
         var strSymbols = stringToArray(string);
         strLength = strSymbols.length;
       }
@@ -14712,7 +14801,7 @@
       pattern = guard ? undefined : pattern;
 
       if (pattern === undefined) {
-        pattern = reHasComplexWord.test(string) ? reComplexWord : reBasicWord;
+        return hasUnicodeWord(string) ? unicodeWords(string) : asciiWords(string);
       }
       return string.match(pattern) || [];
     }