Landing http://codereview.chromium.org/1539013 for ry@tinyclouds.org.

TBR=ager@chromium.org

Review URL: http://codereview.chromium.org/1629001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4352 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

include/v8.h

@@ -855,12 +855,15 @@ class V8EXPORT String : public Primitive {
    * \param start The starting position within the string at which
    * copying begins.
    * \param length The number of bytes to copy from the string.
-   * \return The number of characters copied to the buffer
+   * \param nchars The number of characters written.
+   * \return The number of bytes copied to the buffer
    * excluding the NULL terminator.
    */
   int Write(uint16_t* buffer, int start = 0, int length = -1) const;  // UTF-16
   int WriteAscii(char* buffer, int start = 0, int length = -1) const;  // ASCII
-  int WriteUtf8(char* buffer, int length = -1) const; // UTF-8
+  int WriteUtf8(char* buffer,
+                int length = -1,
+                int* nchars_ref = NULL) const; // UTF-8
 
   /**
    * A zero length string.

src/api.cc

@@ -2641,7 +2641,7 @@ int String::Utf8Length() const {
 }
 
 
-int String::WriteUtf8(char* buffer, int capacity) const {
+int String::WriteUtf8(char* buffer, int capacity, int* nchars_ref) const {
   if (IsDeadCheck("v8::String::WriteUtf8()")) return 0;
   LOG_API("String::WriteUtf8");
   ENTER_V8;
@@ -2655,10 +2655,12 @@ int String::WriteUtf8(char* buffer, int capacity) const {
   int fast_end = capacity - (unibrow::Utf8::kMaxEncodedSize - 1);
   int i;
   int pos = 0;
+  int nchars = 0;
   for (i = 0; i < len && (capacity == -1 || pos < fast_end); i++) {
     i::uc32 c = write_input_buffer.GetNext();
     int written = unibrow::Utf8::Encode(buffer + pos, c);
     pos += written;
+    nchars++;
   }
   if (i < len) {
     // For the last characters we need to check the length for each one
@@ -2672,12 +2674,14 @@ int String::WriteUtf8(char* buffer, int capacity) const {
         for (int j = 0; j < written; j++)
           buffer[pos + j] = intermediate[j];
         pos += written;
+        nchars++;
       } else {
         // We've reached the end of the buffer
         break;
       }
     }
   }
+  if (nchars_ref != NULL) *nchars_ref = nchars;
   if (i == len && (capacity == -1 || pos < capacity))
     buffer[pos++] = '\0';
   return pos;

test/cctest/test-strings.cc

@@ -323,6 +323,7 @@ TEST(Utf8Conversion) {
       0xE3, 0x81, 0x85, 0x00};
   // The number of bytes expected to be written for each length
   const int lengths[12] = {0, 0, 2, 3, 3, 3, 6, 7, 7, 7, 10, 11};
+  const int char_lengths[12] = {0, 0, 1, 2, 2, 2, 3, 4, 4, 4, 5, 5};
   v8::Handle<v8::String> mixed = v8::String::New(mixed_string, 5);
   CHECK_EQ(10, mixed->Utf8Length());
   // Try encoding the string with all capacities
@@ -332,8 +333,10 @@ TEST(Utf8Conversion) {
     // Clear the buffer before reusing it
     for (int j = 0; j < 11; j++)
       buffer[j] = kNoChar;
-    int written = mixed->WriteUtf8(buffer, i);
+    int chars_written;
+    int written = mixed->WriteUtf8(buffer, i, &chars_written);
     CHECK_EQ(lengths[i], written);
+    CHECK_EQ(char_lengths[i], chars_written);
     // Check that the contents are correct
     for (int j = 0; j < lengths[i]; j++)
       CHECK_EQ(as_utf8[j], static_cast<unsigned char>(buffer[j]));