restructure string decoding to support iterable use and include benchmarks for UTF-8 decoding.

utils/tests/string_encoding/utf16_tests.dart

Index: utils/tests/string_encoding/utf16_tests.dart
diff --git a/utils/tests/string_encoding/utf16_tests.dart b/utils/tests/string_encoding/utf16_tests.dart
index 436ef21fa24e994aa8550f63734417c238ae816f..2743a9c77c1d3cfad2545886a26764b3ad9761d4 100755
--- a/utils/tests/string_encoding/utf16_tests.dart
+++ b/utils/tests/string_encoding/utf16_tests.dart
@@ -5,6 +5,7 @@
 
 #library("utf16_tests");
 #import("dunit.dart");
+#import("../../string_encoding/unicode_core.dart");
 #import("../../string_encoding/utf16.dart");
 
 void main() {
@@ -20,7 +21,7 @@ class Utf16Tests extends TestClass {
 갸갹갺갻갼갽갾갿걀걁걂걃걄걅걆걇걈걉걊걋걌걍걎걏걐걑걒걓""";
   static final String testHanWater = "水";
 
-  static final List<int> testKoreanCharSubsetUtf16be = const<int>[
+  static final List<int> testKoreanCharSubsetUtf16beBom = const<int>[
       0xfe, 0xff, 0xac, 0x00, 0xac, 0x01, 0xac, 0x02,
       0xac, 0x03, 0xac, 0x04, 0xac, 0x05, 0xac, 0x06,
       0xac, 0x07, 0xac, 0x08, 0xac, 0x09, 0xac, 0x0a,
@@ -44,63 +45,92 @@ class Utf16Tests extends TestClass {
       0xac, 0x4d, 0xac, 0x4e, 0xac, 0x4f, 0xac, 0x50,
       0xac, 0x51, 0xac, 0x52, 0xac, 0x53];
 
+  static final List<int> testKoreanCharSubsetUtf16le = const<int>    [
+      0x00, 0xac, 0x01, 0xac, 0x02, 0xac, 0x03, 0xac,
+      0x04, 0xac, 0x05, 0xac, 0x06, 0xac, 0x07, 0xac,
+      0x08, 0xac, 0x09, 0xac, 0x0a, 0xac, 0x0b, 0xac,
+      0x0c, 0xac, 0x0d, 0xac, 0x0e, 0xac, 0x0f, 0xac,
+      0x10, 0xac, 0x11, 0xac, 0x12, 0xac, 0x13, 0xac,
+      0x14, 0xac, 0x15, 0xac, 0x16, 0xac, 0x17, 0xac,
+      0x18, 0xac, 0x19, 0xac, 0x1a, 0xac, 0x1b, 0xac,
+      0x0a, 0x00, 0x1c, 0xac, 0x1d, 0xac, 0x1e, 0xac,
+      0x1f, 0xac, 0x20, 0xac, 0x21, 0xac, 0x22, 0xac,
+      0x23, 0xac, 0x24, 0xac, 0x25, 0xac, 0x26, 0xac,
+      0x27, 0xac, 0x28, 0xac, 0x29, 0xac, 0x2a, 0xac,
+      0x2b, 0xac, 0x2c, 0xac, 0x2d, 0xac, 0x2e, 0xac,
+      0x2f, 0xac, 0x30, 0xac, 0x31, 0xac, 0x32, 0xac,
+      0x33, 0xac, 0x34, 0xac, 0x35, 0xac, 0x36, 0xac,
+      0x37, 0xac, 0x0a, 0x00, 0x38, 0xac, 0x39, 0xac,
+      0x3a, 0xac, 0x3b, 0xac, 0x3c, 0xac, 0x3d, 0xac,
+      0x3e, 0xac, 0x3f, 0xac, 0x40, 0xac, 0x41, 0xac,
+      0x42, 0xac, 0x43, 0xac, 0x44, 0xac, 0x45, 0xac,
+      0x46, 0xac, 0x47, 0xac, 0x48, 0xac, 0x49, 0xac,
+      0x4a, 0xac, 0x4b, 0xac, 0x4c, 0xac, 0x4d, 0xac,
+      0x4e, 0xac, 0x4f, 0xac, 0x50, 0xac, 0x51, 0xac,
+      0x52, 0xac, 0x53, 0xac];
+
   void registerTests(TestSuite suite) {
-    register("testEncodeToUtf16", testEncodeToUtf16, suite);
-    register("testUtf16BytesToString", testUtf16BytesToString, suite);
+    register("Utf16Tests.testEncodeToUtf16", testEncodeToUtf16, suite);
+    register("Utf16Tests.testUtf16BytesToString", testUtf16BytesToString, suite);
+    register("Utf16Tests.testIterableMethods", testIterableMethods, suite);
   }
 
   void testEncodeToUtf16() {
-    Expect.listEquals(testKoreanCharSubsetUtf16be,
-        encodeAsUtf16(testKoreanCharSubset),
+    Expect.listEquals([], encodeUtf16be("")); // TODO(dcarlson) should we skip bom if empty?
+    Expect.listEquals(testKoreanCharSubsetUtf16beBom,
+        encodeUtf16(testKoreanCharSubset),
         "encode UTF-16(BE by default) Korean");
 
-    Expect.listEquals([
-        0x00, 0xac, 0x01, 0xac, 0x02, 0xac, 0x03, 0xac,
-        0x04, 0xac, 0x05, 0xac, 0x06, 0xac, 0x07, 0xac,
-        0x08, 0xac, 0x09, 0xac, 0x0a, 0xac, 0x0b, 0xac,
-        0x0c, 0xac, 0x0d, 0xac, 0x0e, 0xac, 0x0f, 0xac,
-        0x10, 0xac, 0x11, 0xac, 0x12, 0xac, 0x13, 0xac,
-        0x14, 0xac, 0x15, 0xac, 0x16, 0xac, 0x17, 0xac,
-        0x18, 0xac, 0x19, 0xac, 0x1a, 0xac, 0x1b, 0xac,
-        0x0a, 0x00, 0x1c, 0xac, 0x1d, 0xac, 0x1e, 0xac,
-        0x1f, 0xac, 0x20, 0xac, 0x21, 0xac, 0x22, 0xac,
-        0x23, 0xac, 0x24, 0xac, 0x25, 0xac, 0x26, 0xac,
-        0x27, 0xac, 0x28, 0xac, 0x29, 0xac, 0x2a, 0xac,
-        0x2b, 0xac, 0x2c, 0xac, 0x2d, 0xac, 0x2e, 0xac,
-        0x2f, 0xac, 0x30, 0xac, 0x31, 0xac, 0x32, 0xac,
-        0x33, 0xac, 0x34, 0xac, 0x35, 0xac, 0x36, 0xac,
-        0x37, 0xac, 0x0a, 0x00, 0x38, 0xac, 0x39, 0xac,
-        0x3a, 0xac, 0x3b, 0xac, 0x3c, 0xac, 0x3d, 0xac,
-        0x3e, 0xac, 0x3f, 0xac, 0x40, 0xac, 0x41, 0xac,
-        0x42, 0xac, 0x43, 0xac, 0x44, 0xac, 0x45, 0xac,
-        0x46, 0xac, 0x47, 0xac, 0x48, 0xac, 0x49, 0xac,
-        0x4a, 0xac, 0x4b, 0xac, 0x4c, 0xac, 0x4d, 0xac,
-        0x4e, 0xac, 0x4f, 0xac, 0x50, 0xac, 0x51, 0xac,
-        0x52, 0xac, 0x53, 0xac], encodeAsUtf16le(testKoreanCharSubset),
+    Expect.listEquals(testKoreanCharSubsetUtf16le,
+        encodeUtf16le(testKoreanCharSubset),
         "encode UTF-16LE Korean");
   }
 
   void testUtf16BytesToString() {
-    Expect.stringEquals(testHanWater, decodeFromUtf16([0x6C, 0x34]),
+    Expect.stringEquals("", decodeUtf16([]));
+    Expect.stringEquals(testHanWater, decodeUtf16([0x6C, 0x34]),
         "Water variation 1");
-    Expect.stringEquals(testHanWater, decodeFromUtf16([0xFE, 0xFF, 0x6C, 0x34]),
+    Expect.stringEquals(testHanWater, decodeUtf16([0xFE, 0xFF, 0x6C, 0x34]),
         "Water variation 2");
-    Expect.stringEquals(testHanWater, decodeFromUtf16([0xFF, 0xFE, 0x34, 0x6C]),
+    Expect.stringEquals(testHanWater, decodeUtf16([0xFF, 0xFE, 0x34, 0x6C]),
         "Water variation 3");
 
-    Expect.stringEquals(testHanWater, decodeFromUtf16be([0x6C, 0x34]),
+    Expect.stringEquals(testHanWater, decodeUtf16be([0x6C, 0x34]),
         "Water variation 4");
     Expect.stringEquals(testHanWater,
-        decodeFromUtf16be([0xFE, 0xFF, 0x6C, 0x34]),
+        decodeUtf16be([0xFE, 0xFF, 0x6C, 0x34]),
         "Water variation 5");
 
-    Expect.stringEquals(testHanWater, decodeFromUtf16le([0x34, 0x6C]),
+    Expect.stringEquals(testHanWater, decodeUtf16le([0x34, 0x6C]),
         "Water variation 6");
     Expect.stringEquals(testHanWater,
-        decodeFromUtf16le([0xFF, 0xFE, 0x34, 0x6C]),
+        decodeUtf16le([0xFF, 0xFE, 0x34, 0x6C]),
         "Water variation 7");
 
     Expect.stringEquals(testKoreanCharSubset,
-        decodeFromUtf16(testKoreanCharSubsetUtf16be), "UTF-16BE Korean");
+        decodeUtf16(testKoreanCharSubsetUtf16beBom), "UTF-16BE Korean");
+  }
+
+  void testIterableMethods() {
+    // empty input
+    Expect.isFalse(decodeUtf16AsIterable([]).iterator().hasNext());
+
+    IterableUtf16Decoder koreanDecoder =
+      decodeUtf16AsIterable(testKoreanCharSubsetUtf16beBom);
+    // get the first character
+    Expect.equals(testKoreanCharSubset.charCodes()[0],
+        koreanDecoder.iterator().next());
+    // get the whole translation using the Iterable interface
+    Expect.stringEquals(testKoreanCharSubset,
+        new String.fromCharCodes(new List<int>.from(koreanDecoder)));
+
+    // specify types
+    Expect.equals(44032, (new List<int>
+        .from(decodeUtf16beAsIterable(testKoreanCharSubsetUtf16beBom)))[0]);
+    Expect.equals(44032, (new List<int>
+        .from(decodeUtf16leAsIterable(testKoreanCharSubsetUtf16le)))[0]);
+    Expect.equals(UNICODE_BOM, (new List<int>
+        .from(decodeUtf16beAsIterable(testKoreanCharSubsetUtf16beBom,
+              stripBom: false)))[0]);
   }
 }