Add support for non ASCII strings when communicating with native ports

runtime/vm/dart_api_message.cc

 // 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/dart_api_message.h"
 #include "vm/object.h"
 #include "vm/snapshot_ids.h"
 #include "vm/symbols.h"
+#include "vm/unicode.h"
 
 namespace dart {
 
 static const int kNumInitialReferences = 4;
 
 ApiMessageReader::ApiMessageReader(const uint8_t* buffer,
                                    intptr_t length,
                                    ReAlloc alloc)
     : BaseReader(buffer, length),
       alloc_(alloc),
@@ skipping 329 matching lines @@
       USE(hash);
       Dart_CObject* object = AllocateDartCObjectString(len);
       AddBackRef(object_id, object, kIsDeserialized);
       char* p = object->value.as_string;
       for (intptr_t i = 0; i < len; i++) {
         p[i] = Read<uint8_t>();
       }
       p[len] = '\0';
       return object;
     }
-    case kTwoByteStringCid:
-      // Two byte strings not supported.
-      return AllocateDartCObjectUnsupported();
+    case kTwoByteStringCid: {
+      intptr_t len = ReadSmiValue();
+      intptr_t hash = ReadSmiValue();
+      USE(hash);
+      uint16_t *utf16 =
+          reinterpret_cast<uint16_t*>(::malloc(len * sizeof(uint16_t)));
+      intptr_t utf8_len = 0;
+      for (intptr_t i = 0; i < len; i++) {
+        utf16[i] = Read<uint16_t>();
+        // TODO(sgjesse): Check for surrogate pairs.
+        utf8_len += Utf8::Length(utf16[i]);
+      }
+      Dart_CObject* object = AllocateDartCObjectString(utf8_len);
+      AddBackRef(object_id, object, kIsDeserialized);
+      char* p = object->value.as_string;
+      for (intptr_t i = 0; i < len; i++) {
+        // TODO(sgjesse): Check for surrogate pairs.
+        p += Utf8::Encode(utf16[i], p);
+      }
+      *p = '\0';
+      ASSERT(p == object->value.as_string + utf8_len);
+      ::free(utf16);
+      return object;
+    }
     case kUint8ArrayCid: {
       intptr_t len = ReadSmiValue();
       Dart_CObject* object = AllocateDartCObjectUint8Array(len);
       AddBackRef(object_id, object, kIsDeserialized);
       if (len > 0) {
         uint8_t* p = object->value.as_byte_array.values;
         for (intptr_t i = 0; i < len; i++) {
           p[i] = Read<uint8_t>();
         }
       }
@@ skipping 356 matching lines @@
     case Dart_CObject::kDouble:
       // Write out the serialization header value for this object.
       WriteInlinedHeader(object);
       // Write out the class and tags information.
       WriteIndexedObject(kDoubleCid);
       WriteIntptrValue(0);
       // Write double value.
       Write<double>(object->value.as_double);
       break;
     case Dart_CObject::kString: {
+      const uint8_t* utf8_str =
+          reinterpret_cast<const uint8_t*>(object->value.as_string);
+      intptr_t utf8_len = strlen(object->value.as_string);
+      Utf8::Type type;
+      intptr_t len = Utf8::CodePointCount(utf8_str, utf8_len, &type);
+
       // Write out the serialization header value for this object.
       WriteInlinedHeader(object);
       // Write out the class and tags information.
-      WriteIndexedObject(kOneByteStringCid);
+      WriteIndexedObject(type == Utf8::kAscii ? kOneByteStringCid
+                                              : kTwoByteStringCid);
       WriteIntptrValue(0);
       // Write string length, hash and content
-      char* str = object->value.as_string;
-      intptr_t len = strlen(str);
       WriteSmi(len);
       WriteSmi(0);  // TODO(sgjesse): Hash - not written.
-      for (intptr_t i = 0; i < len; i++) {
-        Write<uint8_t>(str[i]);
+      if (type == Utf8::kAscii) {
+        for (intptr_t i = 0; i < len; i++) {
+          Write<uint8_t>(utf8_str[i]);
+        }
+      } else {
+        // TODO(sgjesse): Make sure surrogate pairs are handled.
+        uint16_t* utf16_str =
+            reinterpret_cast<uint16_t*>(::malloc(len * sizeof(uint16_t)));
+        Utf8::DecodeToUTF16(utf8_str, utf8_len, utf16_str, len);
+        for (intptr_t i = 0; i < len; i++) {
+          Write<uint16_t>(utf16_str[i]);
+        }
+        ::free(utf16_str);
       }
       break;
     }
     case Dart_CObject::kUint8Array: {
       // Write out the serialization header value for this object.
       WriteInlinedHeader(object);
       // Write out the class and tags information.
       WriteIndexedObject(kUint8ArrayCid);
       WriteIntptrValue(0);
       uint8_t* bytes = object->value.as_byte_array.values;
@@ skipping 37 matching lines @@
   // Write out all objects that were added to the forward list and have
   // not been serialized yet. These would typically be fields of arrays.
   // NOTE: The forward list might grow as we process the list.
   for (intptr_t i = 0; i < forward_id_; i++) {
     WriteForwardedCObject(forward_list_[i]);
   }
   UnmarkAllCObjects(object);
 }
 
 }  // namespace dart