ValueSerializer: Support efficiently reading and writing one-byte strings.

src/value-serializer.cc

 // Copyright 2016 the V8 project authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #include "src/value-serializer.h"
 
 #include <type_traits>
 
 #include "src/base/logging.h"
 #include "src/conversions.h"
 #include "src/factory.h"
 #include "src/flags.h"
 #include "src/handles-inl.h"
 #include "src/isolate.h"
 #include "src/objects-inl.h"
 #include "src/objects.h"
 #include "src/snapshot/code-serializer.h"
 #include "src/transitions.h"
 #include "src/wasm/wasm-module.h"
 #include "src/wasm/wasm-objects.h"
 #include "src/wasm/wasm-result.h"
 
 namespace v8 {
 namespace internal {
 
-static const uint32_t kLatestVersion = 9;
+// Version 9: (imported from Blink)
+// Version 10: one-byte (Latin-1) strings
+static const uint32_t kLatestVersion = 10;
+
 static const int kPretenureThreshold = 100 * KB;
 
 template <typename T>
 static size_t BytesNeededForVarint(T value) {
   static_assert(std::is_integral<T>::value && std::is_unsigned<T>::value,
                 "Only unsigned integer types can be written as varints.");
   size_t result = 0;
   do {
     result++;
     value >>= 7;
@@ skipping 17 matching lines @@
   // (like sint32 in protobuf)
   kInt32 = 'I',
   // Number represented as 32-bit unsigned integer, varint-encoded
   // (like uint32 in protobuf)
   kUint32 = 'U',
   // Number represented as a 64-bit double.
   // Host byte order is used (N.B. this makes the format non-portable).
   kDouble = 'N',
   // byteLength:uint32_t, then raw data
   kUtf8String = 'S',
+  kOneByteString = '"',
   kTwoByteString = 'c',
   // Reference to a serialized object. objectID:uint32_t
   kObjectReference = '^',
   // Beginning of a JS object.
   kBeginJSObject = 'o',
   // End of a JS object. numProperties:uint32_t
   kEndJSObject = '{',
   // Beginning of a sparse JS array. length:uint32_t
   // Elements and properties are written as key/value pairs, like objects.
   kBeginSparseJSArray = 'a',
@@ skipping 291 matching lines @@
   WriteTag(SerializationTag::kDouble);
   WriteDouble(number->value());
 }
 
 void ValueSerializer::WriteString(Handle<String> string) {
   string = String::Flatten(string);
   DisallowHeapAllocation no_gc;
   String::FlatContent flat = string->GetFlatContent();
   DCHECK(flat.IsFlat());
   if (flat.IsOneByte()) {
-    // The existing format uses UTF-8, rather than Latin-1. As a result we must
-    // to do work to encode strings that have characters outside ASCII.
-    // TODO(jbroman): In a future format version, consider adding a tag for
-    // Latin-1 strings, so that this can be skipped.
-    WriteTag(SerializationTag::kUtf8String);
     Vector<const uint8_t> chars = flat.ToOneByteVector();
-    if (String::IsAscii(chars.begin(), chars.length())) {
-      WriteOneByteString(chars);
-    } else {
-      v8::Local<v8::String> api_string = Utils::ToLocal(string);
-      uint32_t utf8_length = api_string->Utf8Length();
-      WriteVarint(utf8_length);
-      uint8_t* dest;
-      if (ReserveRawBytes(utf8_length).To(&dest)) {
-        api_string->WriteUtf8(reinterpret_cast<char*>(dest), utf8_length,
-                              nullptr, v8::String::NO_NULL_TERMINATION);
-      }
-    }
+    WriteTag(SerializationTag::kOneByteString);
+    WriteOneByteString(chars);
   } else if (flat.IsTwoByte()) {
     Vector<const uc16> chars = flat.ToUC16Vector();
     uint32_t byte_length = chars.length() * sizeof(uc16);
     // The existing reading code expects 16-byte strings to be aligned.
     if ((buffer_size_ + 1 + BytesNeededForVarint(byte_length)) & 1)
       WriteTag(SerializationTag::kPadding);
     WriteTag(SerializationTag::kTwoByteString);
     WriteTwoByteString(chars);
   } else {
     UNREACHABLE();
@@ skipping 706 matching lines @@
       return isolate_->factory()->NewNumberFromUint(number.FromJust(),
                                                     pretenure_);
     }
     case SerializationTag::kDouble: {
       Maybe<double> number = ReadDouble();
       if (number.IsNothing()) return MaybeHandle<Object>();
       return isolate_->factory()->NewNumber(number.FromJust(), pretenure_);
     }
     case SerializationTag::kUtf8String:
       return ReadUtf8String();
+    case SerializationTag::kOneByteString:
+      return ReadOneByteString();
     case SerializationTag::kTwoByteString:
       return ReadTwoByteString();
     case SerializationTag::kObjectReference: {
       uint32_t id;
       if (!ReadVarint<uint32_t>().To(&id)) return MaybeHandle<Object>();
       return GetObjectWithID(id);
     }
     case SerializationTag::kBeginJSObject:
       return ReadJSObject();
     case SerializationTag::kBeginSparseJSArray:
@@ skipping 39 matching lines @@
   if (!ReadVarint<uint32_t>().To(&utf8_length) ||
       utf8_length >
           static_cast<uint32_t>(std::numeric_limits<int32_t>::max()) ||
       !ReadRawBytes(utf8_length).To(&utf8_bytes)) {
     return MaybeHandle<String>();
   }
   return isolate_->factory()->NewStringFromUtf8(
       Vector<const char>::cast(utf8_bytes), pretenure_);
 }
 
+MaybeHandle<String> ValueDeserializer::ReadOneByteString() {
+  uint32_t byte_length;
+  Vector<const uint8_t> bytes;
+  if (!ReadVarint<uint32_t>().To(&byte_length) ||
+      byte_length >
+          static_cast<uint32_t>(std::numeric_limits<int32_t>::max()) ||
+      !ReadRawBytes(byte_length).To(&bytes)) {
+    return MaybeHandle<String>();
+  }
+  return isolate_->factory()->NewStringFromOneByte(bytes, pretenure_);
+}
+
 MaybeHandle<String> ValueDeserializer::ReadTwoByteString() {
   uint32_t byte_length;
   Vector<const uint8_t> bytes;
   if (!ReadVarint<uint32_t>().To(&byte_length) ||
       byte_length >
           static_cast<uint32_t>(std::numeric_limits<int32_t>::max()) ||
       byte_length % sizeof(uc16) != 0 ||
       !ReadRawBytes(byte_length).To(&bytes)) {
     return MaybeHandle<String>();
   }
@@ skipping 728 matching lines @@
   if (stack.size() != 1) {
     isolate_->Throw(*isolate_->factory()->NewError(
         MessageTemplate::kDataCloneDeserializationError));
     return MaybeHandle<Object>();
   }
   return scope.CloseAndEscape(stack[0]);
 }
 
 }  // namespace internal
 }  // namespace v8