Add support for surrogates when serializing and deserializing for native ports

runtime/vm/unicode.cc

-// Copyright (c) 2011, the Dart project authors.  Please see the AUTHORS file
+// 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.
 
 #include "vm/unicode.h"
 
 #include "vm/allocation.h"
 #include "vm/globals.h"
 #include "vm/object.h"
 
 namespace dart {
@@ skipping 12 matching lines @@
   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
   0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
   2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
   3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
   4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 6, 6, 0, 0
 };
 
 
 const uint32_t Utf8::kMagicBits[7] = {
-  0,  // padding
+  0,  // Padding.
   0x00000000,
   0x00003080,
   0x000E2080,
   0x03C82080,
   0xFA082080,
   0x82082080
 };
 
 
 // Minimum values of code points used to check shortest form.
 const uint32_t Utf8::kOverlongMinimum[7] = {
-  0,  // padding
+  0,  // Padding.
   0x0,
   0x80,
   0x800,
   0x10000,
   0xFFFFFFFF,
   0xFFFFFFFF
 };
 
 
 // Returns a count of the number of UTF-8 trail bytes.
@@ skipping 72 matching lines @@
   String::CodePointIterator it(str);
   while (it.Next()) {
     int32_t ch = it.Current();
     length += Utf8::Length(ch);
   }
   return length;
 }
 
 
 intptr_t Utf8::Encode(int32_t ch, char* dst) {
+  ASSERT(!Utf16::IsSurrogate(ch));
   static const int kMask = ~(1 << 6);
   if (ch <= kMaxOneByteChar) {
     dst[0] = ch;
     return 1;
   }
   if (ch <= kMaxTwoByteChar) {
     dst[0] = 0xC0 | (ch >> 6);
     dst[1] = 0x80 | (ch & kMask);
     return 2;
   }
@@ skipping 66 matching lines @@
                           uint8_t* dst,
                           intptr_t len) {
   intptr_t i = 0;
   intptr_t j = 0;
   intptr_t num_bytes;
   for (; (i < array_len) && (j < len); i += num_bytes, ++j) {
     int32_t ch;
     ASSERT(IsLatin1SequenceStart(utf8_array[i]));
     num_bytes = Utf8::Decode(&utf8_array[i], (array_len - i), &ch);
     if (ch == -1) {
-      return false;  // invalid input
+      return false;  // Invalid input.
     }
     ASSERT(Utf::IsLatin1(ch));
     dst[j] = ch;
   }
   if ((i < array_len) && (j == len)) {
-    return false;  // output overflow
+    return false;  // Output overflow.
   }
-  return true;  // success
+  return true;  // Success.
 }
 
 
 bool Utf8::DecodeToUTF16(const uint8_t* utf8_array,
                          intptr_t array_len,
                          uint16_t* dst,
                          intptr_t len) {
   intptr_t i = 0;
   intptr_t j = 0;
   intptr_t num_bytes;
   for (; (i < array_len) && (j < len); i += num_bytes, ++j) {
     int32_t ch;
     bool is_supplementary = IsSupplementarySequenceStart(utf8_array[i]);
     num_bytes = Utf8::Decode(&utf8_array[i], (array_len - i), &ch);
     if (ch == -1) {
-      return false;  // invalid input
+      return false;  // Invalid input.
     }
     if (is_supplementary) {
       Utf16::Encode(ch, &dst[j]);
       j = j + 1;
     } else {
       dst[j] = ch;
     }
   }
   if ((i < array_len) && (j == len)) {
-    return false;  // output overflow
+    return false;  // Output overflow.
   }
-  return true;  // success
+  return true;  // Success.
 }
 
 
 bool Utf8::DecodeToUTF32(const uint8_t* utf8_array,
                          intptr_t array_len,
                          int32_t* dst,
                          intptr_t len) {
   intptr_t i = 0;
   intptr_t j = 0;
   intptr_t num_bytes;
   for (; (i < array_len) && (j < len); i += num_bytes, ++j) {
     int32_t ch;
     num_bytes = Utf8::Decode(&utf8_array[i], (array_len - i), &ch);
     if (ch == -1) {
-      return false;  // invalid input
+      return false;  // Invalid input.
     }
     dst[j] = ch;
   }
   if ((i < array_len) && (j == len)) {
-    return false;  // output overflow
+    return false;  // Output overflow.
   }
-  return true;  // success
+  return true;  // Success.
 }
 
 
 bool Utf8::DecodeCStringToUTF32(const char* str, int32_t* dst, intptr_t len) {
   ASSERT(str != NULL);
   intptr_t array_len = strlen(str);
   const uint8_t* utf8_array = reinterpret_cast<const uint8_t*>(str);
   return Utf8::DecodeToUTF32(utf8_array, array_len, dst, len);
 }
 
 
 void Utf16::Encode(int32_t codepoint, uint16_t* dst) {
   ASSERT(codepoint > Utf16::kMaxCodeUnit);
   ASSERT(dst != NULL);
   dst[0] = (Utf16::kLeadSurrogateOffset + (codepoint >> 10));
   dst[1] = (0xDC00 + (codepoint & 0x3FF));
 }
 
 }  // namespace dart