Add some support for the code-point code-unit distinction.

runtime/vm/object.h

Index: runtime/vm/object.h
diff --git a/runtime/vm/object.h b/runtime/vm/object.h
index 2836dda969ab3435a680294b02d436ae91c882f3..97d4f5c708d0d5e7ec5f4fee602326794fa4f2ed 100644
--- a/runtime/vm/object.h
+++ b/runtime/vm/object.h
@@ -17,6 +17,7 @@
 #include "vm/os.h"
 #include "vm/raw_object.h"
 #include "vm/scanner.h"
+#include "vm/unicode.h"
 
 namespace dart {
 
@@ -3684,7 +3685,8 @@ class String : public Instance {
   static intptr_t Hash(const uint16_t* characters, intptr_t len);
   static intptr_t Hash(const uint32_t* characters, intptr_t len);
 
-  int32_t CharAt(intptr_t index) const;
+  uint32_t CharAt(intptr_t index) const;
+  uint32_t CodeUnitAt(intptr_t index) const;
 
   intptr_t CharSize() const;
 
@@ -3850,7 +3852,11 @@ class String : public Instance {
 
 class OneByteString : public AllStatic {
  public:
-  static int32_t CharAt(const String& str, intptr_t index) {
+  static uint32_t CharAt(const String& str, intptr_t index) {
+    return *CharAddr(str, index);
+  }
+
+  static uint32_t CodeUnitAt(const String& str, intptr_t index) {
     return *CharAddr(str, index);
   }
 
@@ -3907,6 +3913,7 @@ class OneByteString : public AllStatic {
 
   static RawOneByteString* Transform(int32_t (*mapping)(int32_t ch),
                                      const String& str,
+                                     int out_length,
                                      Heap::Space space);
 
   static const ClassId kClassId = kOneByteStringCid;
@@ -3944,7 +3951,19 @@ class OneByteString : public AllStatic {
 
 class TwoByteString : public AllStatic {
  public:
-  static int32_t CharAt(const String& str, intptr_t index) {
+  static uint32_t CharAt(const String& str, intptr_t index) {
+    int32_t unit = *CharAddr(str, index);
+    // For non-surrogate values or incorrect trailing surrogates we just return
+    // the value.
+    if (!Utf16::IsLeadSurrogate(unit)) return unit;
+
+    // If the string ends with a lead surrogate we just return that.
+    if (index + 1 >= str.Length()) return unit;
+
+    return Utf16::CodePointFromCodeUnits(unit, *CharAddr(str, index + 1));
+  }
+
+  static uint32_t CodeUnitAt(const String& str, intptr_t index) {
     return *CharAddr(str, index);
   }

[siva, 2012/11/08 19:06:47]

I don't mind changing the  name 'CharAt' to 'CodeUnitAt" but
I don't think your new 'CharAt' functionality belongs
in the VM code.

Iteration of characters in a string should be done in an
Iterator implementation in the library.

[erikcorry, 2012/11/08 22:09:34]

Do you want to move the C++ support for toUpperCase and toLowerCase into dart
too?  They make use of charAt.

[siva, 2012/11/09 02:40:42]

We could fix String::Transform/TwoByteString::Transform
to locally compute the unicode values as it iterates over
the string.

On 2012/11/08 22:09:34, erikcorry wrote:

[erikcorry, 2012/11/15 13:28:25]

I have fixed the UTF8 and the Transform (toUpper/Lower) routines to use a helper
from unicode.h and removed this from the String classes.

 
@@ -3990,6 +4009,7 @@ class TwoByteString : public AllStatic {
 
   static RawTwoByteString* Transform(int32_t (*mapping)(int32_t ch),
                                      const String& str,
+                                     int out_length,
                                      Heap::Space space);
 
   static RawTwoByteString* null() {
@@ -4028,7 +4048,11 @@ class TwoByteString : public AllStatic {
 
 class ExternalOneByteString : public AllStatic {
  public:
-  static int32_t CharAt(const String& str, intptr_t index) {
+  static uint32_t CharAt(const String& str, intptr_t index) {
+    return *CharAddr(str, index);
+  }
+
+  static uint32_t CodeUnitAt(const String& str, intptr_t index) {
     return *CharAddr(str, index);
   }
 
@@ -4094,7 +4118,19 @@ class ExternalOneByteString : public AllStatic {
 
 class ExternalTwoByteString : public AllStatic {
  public:
-  static int32_t CharAt(const String& str, intptr_t index) {
+  static uint32_t CharAt(const String& str, intptr_t index) {
+    int32_t unit = *CharAddr(str, index);

[floitsch, 2012/11/08 15:28:21]

uint32_t unit =

[erikcorry, 2012/11/15 13:28:25]

This code is gone.

+    // For non-surrogate values or incorrect trailing surrogates we just return
+    // the value.
+    if (!Utf16::IsLeadSurrogate(unit)) return unit;
+
+    // If the string ends with a lead surrogate we just return that.
+    if (index + 1 >= str.Length()) return unit;
+
+    return Utf16::CodePointFromCodeUnits(unit, *CharAddr(str, index + 1));
+  }
+
+  static uint32_t CodeUnitAt(const String& str, intptr_t index) {
     return *CharAddr(str, index);
   }
 
@@ -5858,7 +5894,7 @@ bool String::Equals(const String& str,
     return false;  // Lengths don't match.
   }
   for (intptr_t i = 0; i < len; i++) {
-    if (this->CharAt(i) != str.CharAt(begin_index + i)) {
+    if (this->CodeUnitAt(i) != str.CodeUnitAt(begin_index + i)) {
       return false;
     }
   }