Add optimized String.fromCharCodes path for Uint8List and Int8List.

runtime/lib/string_patch.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.
 
 patch class String {
   /* patch */ factory String.fromCharCodes(Iterable<int> charCodes) {
     return _StringBase.createFromCharCodes(charCodes);
   }
 
   /* patch */ const factory String.fromEnvironment(String name,
@@ skipping 16 matching lines @@
   Type get runtimeType => String;
 
   int get hashCode native "String_getHashCode";
 
   /**
    *  Create the most efficient string representation for specified
    *  [codePoints].
    */
   static String createFromCharCodes(Iterable<int> charCodes) {
     if (charCodes != null) {
+      if (charCodes is Uint8List || charCodes is Int8List) {

[Anders Johnsen, 2013/11/18 20:26:31]

Should we use _classId here as well?

[srdjan, 2013/11/18 20:45:15]

Yes, using cid-s is much quicker.

[Anders Johnsen, 2013/11/19 07:38:43]

I encountered an issue here. These classes are in typed_data-patch, so:

- _Uint8List is not accessible from dart:core-patch.
- _cid is not accessible within dart:typed_data-patch.
- Uint8List._privateHelper is not accessible

What's your recommendation here?

[srdjan, 2013/11/19 16:47:38]

This would not slow down existing code:

if ((ccid != _List._classId) && (ccid != _GrowableList._classId) && (ccid !=
_ImmutableList._classId)) {
  if (charCodes is Uint8List || charCodes is Int8List) {
    return _OneByteString._allocateFromOneByteList(charCodes);
  }
  charCodes = new List<int>.from(charCodes, growable: false);
}

You may want to benefit from skipping the call to native for small 8-bit lists
and skip the isOneByteString test.

You could also check first the charCodes length:

if (charCodes.length >128) {
  return _OneByteString._allocateFromOneByteList(charCodes);
}

We plan to make all cid's available through an internal VM class. Until then use
the instanceOf test is OK.

+        return _OneByteString._allocateFromOneByteList(charCodes);
+      }
       // TODO(srdjan): Also skip copying of typed arrays.
       final ccid = charCodes._cid;
       if ((ccid != _List._classId) &&
           (ccid != _GrowableList._classId) &&
           (ccid != _ImmutableList._classId)) {
         charCodes = new List<int>.from(charCodes, growable: false);
       }
 
       bool isOneByteString = true;
       for (int i = 0; i < charCodes.length; i++) {
         int e = charCodes[i];
         if (e is! _Smi) throw new ArgumentError(e);
         // Is e Latin1?
         if ((e < 0) || (e > 0xFF)) {
           isOneByteString = false;
           break;
         }
       }
       if (isOneByteString) {
-        var s = _OneByteString._allocate(charCodes.length);
-        for (int i = 0; i < charCodes.length; i++) {
-          s._setAt(i, charCodes[i]);
-        }
-        return s;
+        return _OneByteString._allocateFromOneByteList(charCodes);

[srdjan, 2013/11/18 20:45:15]

Calling to native is quite slow, i.e., for small lists this will be quite more
expensive. Could you benchmark? Any golem numbers on existing benchmarks?

[Anders Johnsen, 2013/11/18 21:10:51]

Before, we already did the runtime-call to allocate the list. So, doing this is
actually faster (~25% for a small local test). Will upload golem numbers
tomorrow at the office.

       }
     }
     return _createFromCodePoints(charCodes);
   }
 
   static String _createFromCodePoints(List<int> codePoints)
       native "StringBase_createFromCodePoints";
 
   String operator [](int index) native "String_charAt";
 
@@ skipping 550 matching lines @@
         return false;
       }
     }
     return super.contains(pattern, start);
   }
 
   // Allocates a string of given length, expecting its content to be
   // set using _setAt.
   static _OneByteString _allocate(int length) native "OneByteString_allocate";
 
+
+  static _OneByteString _allocateFromOneByteList(List<int> list)
+      native "OneByteString_allocateFromOneByteList";
+
   // This is internal helper method. Code point value must be a valid
   // Latin1 value (0..0xFF), index must be valid.
   void _setAt(int index, int codePoint) native "OneByteString_setAt";
 }
 
 
 class _TwoByteString extends _StringBase implements String {
   static final int _classId = "\u{FFFF}"._cid;
 
   factory _TwoByteString._uninstantiable() {
@@ skipping 82 matching lines @@
 class _CodeUnits extends Object with ListMixin<int>,
                                      UnmodifiableListMixin<int> {
   /** The string that this is the code units of. */
   String _string;
 
   _CodeUnits(this._string);
 
   int get length => _string.length;
   int operator[](int i) => _string.codeUnitAt(i);
 }