blob: e2fa3b5371db5213cb889f99f030916a22aba083 [file] [log] [blame]
// Copyright 2006-2009 the V8 project authors. All rights reserved.
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are
// met:
//
// * Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
// * Redistributions in binary form must reproduce the above
// copyright notice, this list of conditions and the following
// disclaimer in the documentation and/or other materials provided
// with the distribution.
// * Neither the name of Google Inc. nor the names of its
// contributors may be used to endorse or promote products derived
// from this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#include "v8.h"
#include "api.h"
#include "arguments.h"
#include "bootstrapper.h"
#include "debug.h"
#include "execution.h"
#include "objects-inl.h"
#include "macro-assembler.h"
#include "scanner.h"
#include "scopeinfo.h"
#include "string-stream.h"
#ifdef ENABLE_DISASSEMBLER
#include "disassembler.h"
#endif
namespace v8 {
namespace internal {
// Getters and setters are stored in a fixed array property. These are
// constants for their indices.
const int kGetterIndex = 0;
const int kSetterIndex = 1;
static Object* CreateJSValue(JSFunction* constructor, Object* value) {
Object* result = Heap::AllocateJSObject(constructor);
if (result->IsFailure()) return result;
JSValue::cast(result)->set_value(value);
return result;
}
Object* Object::ToObject(Context* global_context) {
if (IsNumber()) {
return CreateJSValue(global_context->number_function(), this);
} else if (IsBoolean()) {
return CreateJSValue(global_context->boolean_function(), this);
} else if (IsString()) {
return CreateJSValue(global_context->string_function(), this);
}
ASSERT(IsJSObject());
return this;
}
Object* Object::ToObject() {
Context* global_context = Top::context()->global_context();
if (IsJSObject()) {
return this;
} else if (IsNumber()) {
return CreateJSValue(global_context->number_function(), this);
} else if (IsBoolean()) {
return CreateJSValue(global_context->boolean_function(), this);
} else if (IsString()) {
return CreateJSValue(global_context->string_function(), this);
}
// Throw a type error.
return Failure::InternalError();
}
Object* Object::ToBoolean() {
if (IsTrue()) return Heap::true_value();
if (IsFalse()) return Heap::false_value();
if (IsSmi()) {
return Heap::ToBoolean(Smi::cast(this)->value() != 0);
}
if (IsUndefined() || IsNull()) return Heap::false_value();
// Undetectable object is false
if (IsUndetectableObject()) {
return Heap::false_value();
}
if (IsString()) {
return Heap::ToBoolean(String::cast(this)->length() != 0);
}
if (IsHeapNumber()) {
return HeapNumber::cast(this)->HeapNumberToBoolean();
}
return Heap::true_value();
}
void Object::Lookup(String* name, LookupResult* result) {
if (IsJSObject()) return JSObject::cast(this)->Lookup(name, result);
Object* holder = NULL;
Context* global_context = Top::context()->global_context();
if (IsString()) {
holder = global_context->string_function()->instance_prototype();
} else if (IsNumber()) {
holder = global_context->number_function()->instance_prototype();
} else if (IsBoolean()) {
holder = global_context->boolean_function()->instance_prototype();
}
ASSERT(holder != NULL); // Cannot handle null or undefined.
JSObject::cast(holder)->Lookup(name, result);
}
Object* Object::GetPropertyWithReceiver(Object* receiver,
String* name,
PropertyAttributes* attributes) {
LookupResult result;
Lookup(name, &result);
Object* value = GetProperty(receiver, &result, name, attributes);
ASSERT(*attributes <= ABSENT);
return value;
}
Object* Object::GetPropertyWithCallback(Object* receiver,
Object* structure,
String* name,
Object* holder) {
// To accommodate both the old and the new api we switch on the
// data structure used to store the callbacks. Eventually proxy
// callbacks should be phased out.
if (structure->IsProxy()) {
AccessorDescriptor* callback =
reinterpret_cast<AccessorDescriptor*>(Proxy::cast(structure)->proxy());
Object* value = (callback->getter)(receiver, callback->data);
RETURN_IF_SCHEDULED_EXCEPTION();
return value;
}
// api style callbacks.
if (structure->IsAccessorInfo()) {
AccessorInfo* data = AccessorInfo::cast(structure);
Object* fun_obj = data->getter();
v8::AccessorGetter call_fun = v8::ToCData<v8::AccessorGetter>(fun_obj);
HandleScope scope;
JSObject* self = JSObject::cast(receiver);
JSObject* holder_handle = JSObject::cast(holder);
Handle<String> key(name);
LOG(ApiNamedPropertyAccess("load", self, name));
CustomArguments args(data->data(), self, holder_handle);
v8::AccessorInfo info(args.end());
v8::Handle<v8::Value> result;
{
// Leaving JavaScript.
VMState state(EXTERNAL);
result = call_fun(v8::Utils::ToLocal(key), info);
}
RETURN_IF_SCHEDULED_EXCEPTION();
if (result.IsEmpty()) return Heap::undefined_value();
return *v8::Utils::OpenHandle(*result);
}
// __defineGetter__ callback
if (structure->IsFixedArray()) {
Object* getter = FixedArray::cast(structure)->get(kGetterIndex);
if (getter->IsJSFunction()) {
return Object::GetPropertyWithDefinedGetter(receiver,
JSFunction::cast(getter));
}
// Getter is not a function.
return Heap::undefined_value();
}
UNREACHABLE();
return 0;
}
Object* Object::GetPropertyWithDefinedGetter(Object* receiver,
JSFunction* getter) {
HandleScope scope;
Handle<JSFunction> fun(JSFunction::cast(getter));
Handle<Object> self(receiver);
#ifdef ENABLE_DEBUGGER_SUPPORT
// Handle stepping into a getter if step into is active.
if (Debug::StepInActive()) {
Debug::HandleStepIn(fun, Handle<Object>::null(), 0, false);
}
#endif
bool has_pending_exception;
Handle<Object> result =
Execution::Call(fun, self, 0, NULL, &has_pending_exception);
// Check for pending exception and return the result.
if (has_pending_exception) return Failure::Exception();
return *result;
}
// Only deal with CALLBACKS and INTERCEPTOR
Object* JSObject::GetPropertyWithFailedAccessCheck(
Object* receiver,
LookupResult* result,
String* name,
PropertyAttributes* attributes) {
if (result->IsValid()) {
switch (result->type()) {
case CALLBACKS: {
// Only allow API accessors.
Object* obj = result->GetCallbackObject();
if (obj->IsAccessorInfo()) {
AccessorInfo* info = AccessorInfo::cast(obj);
if (info->all_can_read()) {
*attributes = result->GetAttributes();
return GetPropertyWithCallback(receiver,
result->GetCallbackObject(),
name,
result->holder());
}
}
break;
}
case NORMAL:
case FIELD:
case CONSTANT_FUNCTION: {
// Search ALL_CAN_READ accessors in prototype chain.
LookupResult r;
result->holder()->LookupRealNamedPropertyInPrototypes(name, &r);
if (r.IsValid()) {
return GetPropertyWithFailedAccessCheck(receiver,
&r,
name,
attributes);
}
break;
}
case INTERCEPTOR: {
// If the object has an interceptor, try real named properties.
// No access check in GetPropertyAttributeWithInterceptor.
LookupResult r;
result->holder()->LookupRealNamedProperty(name, &r);
if (r.IsValid()) {
return GetPropertyWithFailedAccessCheck(receiver,
&r,
name,
attributes);
}
}
default: {
break;
}
}
}
// No accessible property found.
*attributes = ABSENT;
Top::ReportFailedAccessCheck(this, v8::ACCESS_GET);
return Heap::undefined_value();
}
PropertyAttributes JSObject::GetPropertyAttributeWithFailedAccessCheck(
Object* receiver,
LookupResult* result,
String* name,
bool continue_search) {
if (result->IsValid()) {
switch (result->type()) {
case CALLBACKS: {
// Only allow API accessors.
Object* obj = result->GetCallbackObject();
if (obj->IsAccessorInfo()) {
AccessorInfo* info = AccessorInfo::cast(obj);
if (info->all_can_read()) {
return result->GetAttributes();
}
}
break;
}
case NORMAL:
case FIELD:
case CONSTANT_FUNCTION: {
if (!continue_search) break;
// Search ALL_CAN_READ accessors in prototype chain.
LookupResult r;
result->holder()->LookupRealNamedPropertyInPrototypes(name, &r);
if (r.IsValid()) {
return GetPropertyAttributeWithFailedAccessCheck(receiver,
&r,
name,
continue_search);
}
break;
}
case INTERCEPTOR: {
// If the object has an interceptor, try real named properties.
// No access check in GetPropertyAttributeWithInterceptor.
LookupResult r;
if (continue_search) {
result->holder()->LookupRealNamedProperty(name, &r);
} else {
result->holder()->LocalLookupRealNamedProperty(name, &r);
}
if (r.IsValid()) {
return GetPropertyAttributeWithFailedAccessCheck(receiver,
&r,
name,
continue_search);
}
break;
}
default: {
break;
}
}
}
Top::ReportFailedAccessCheck(this, v8::ACCESS_HAS);
return ABSENT;
}
Object* JSObject::GetLazyProperty(Object* receiver,
LookupResult* result,
String* name,
PropertyAttributes* attributes) {
HandleScope scope;
Handle<Object> this_handle(this);
Handle<Object> receiver_handle(receiver);
Handle<String> name_handle(name);
bool pending_exception;
LoadLazy(Handle<JSObject>(JSObject::cast(result->GetLazyValue())),
&pending_exception);
if (pending_exception) return Failure::Exception();
return this_handle->GetPropertyWithReceiver(*receiver_handle,
*name_handle,
attributes);
}
Object* JSObject::SetLazyProperty(LookupResult* result,
String* name,
Object* value,
PropertyAttributes attributes) {
ASSERT(!IsJSGlobalProxy());
HandleScope scope;
Handle<JSObject> this_handle(this);
Handle<String> name_handle(name);
Handle<Object> value_handle(value);
bool pending_exception;
LoadLazy(Handle<JSObject>(JSObject::cast(result->GetLazyValue())),
&pending_exception);
if (pending_exception) return Failure::Exception();
return this_handle->SetProperty(*name_handle, *value_handle, attributes);
}
Object* JSObject::DeleteLazyProperty(LookupResult* result,
String* name,
DeleteMode mode) {
HandleScope scope;
Handle<JSObject> this_handle(this);
Handle<String> name_handle(name);
bool pending_exception;
LoadLazy(Handle<JSObject>(JSObject::cast(result->GetLazyValue())),
&pending_exception);
if (pending_exception) return Failure::Exception();
return this_handle->DeleteProperty(*name_handle, mode);
}
Object* JSObject::GetNormalizedProperty(LookupResult* result) {
ASSERT(!HasFastProperties());
Object* value = property_dictionary()->ValueAt(result->GetDictionaryEntry());
if (IsGlobalObject()) {
value = JSGlobalPropertyCell::cast(value)->value();
}
ASSERT(!value->IsJSGlobalPropertyCell());
return value;
}
Object* JSObject::SetNormalizedProperty(LookupResult* result, Object* value) {
ASSERT(!HasFastProperties());
if (IsGlobalObject()) {
JSGlobalPropertyCell* cell =
JSGlobalPropertyCell::cast(
property_dictionary()->ValueAt(result->GetDictionaryEntry()));
cell->set_value(value);
} else {
property_dictionary()->ValueAtPut(result->GetDictionaryEntry(), value);
}
return value;
}
Object* JSObject::SetNormalizedProperty(String* name,
Object* value,
PropertyDetails details) {
ASSERT(!HasFastProperties());
int entry = property_dictionary()->FindEntry(name);
if (entry == StringDictionary::kNotFound) {
Object* store_value = value;
if (IsGlobalObject()) {
store_value = Heap::AllocateJSGlobalPropertyCell(value);
if (store_value->IsFailure()) return store_value;
}
Object* dict = property_dictionary()->Add(name, store_value, details);
if (dict->IsFailure()) return dict;
set_properties(StringDictionary::cast(dict));
return value;
}
// Preserve enumeration index.
details = PropertyDetails(details.attributes(),
details.type(),
property_dictionary()->DetailsAt(entry).index());
if (IsGlobalObject()) {
JSGlobalPropertyCell* cell =
JSGlobalPropertyCell::cast(property_dictionary()->ValueAt(entry));
cell->set_value(value);
// Please note we have to update the property details.
property_dictionary()->DetailsAtPut(entry, details);
} else {
property_dictionary()->SetEntry(entry, name, value, details);
}
return value;
}
Object* JSObject::DeleteNormalizedProperty(String* name, DeleteMode mode) {
ASSERT(!HasFastProperties());
StringDictionary* dictionary = property_dictionary();
int entry = dictionary->FindEntry(name);
if (entry != StringDictionary::kNotFound) {
// If we have a global object set the cell to the hole.
if (IsGlobalObject()) {
PropertyDetails details = dictionary->DetailsAt(entry);
if (details.IsDontDelete()) {
if (mode != FORCE_DELETION) return Heap::false_value();
// When forced to delete global properties, we have to make a
// map change to invalidate any ICs that think they can load
// from the DontDelete cell without checking if it contains
// the hole value.
Object* new_map = map()->CopyDropDescriptors();
if (new_map->IsFailure()) return new_map;
set_map(Map::cast(new_map));
}
JSGlobalPropertyCell* cell =
JSGlobalPropertyCell::cast(dictionary->ValueAt(entry));
cell->set_value(Heap::the_hole_value());
dictionary->DetailsAtPut(entry, details.AsDeleted());
} else {
return dictionary->DeleteProperty(entry, mode);
}
}
return Heap::true_value();
}
bool JSObject::IsDirty() {
Object* cons_obj = map()->constructor();
if (!cons_obj->IsJSFunction())
return true;
JSFunction* fun = JSFunction::cast(cons_obj);
if (!fun->shared()->function_data()->IsFunctionTemplateInfo())
return true;
// If the object is fully fast case and has the same map it was
// created with then no changes can have been made to it.
return map() != fun->initial_map()
|| !HasFastElements()
|| !HasFastProperties();
}
Object* Object::GetProperty(Object* receiver,
LookupResult* result,
String* name,
PropertyAttributes* attributes) {
// Make sure that the top context does not change when doing
// callbacks or interceptor calls.
AssertNoContextChange ncc;
// Traverse the prototype chain from the current object (this) to
// the holder and check for access rights. This avoid traversing the
// objects more than once in case of interceptors, because the
// holder will always be the interceptor holder and the search may
// only continue with a current object just after the interceptor
// holder in the prototype chain.
Object* last = result->IsValid() ? result->holder() : Heap::null_value();
for (Object* current = this; true; current = current->GetPrototype()) {
if (current->IsAccessCheckNeeded()) {
// Check if we're allowed to read from the current object. Note
// that even though we may not actually end up loading the named
// property from the current object, we still check that we have
// access to it.
JSObject* checked = JSObject::cast(current);
if (!Top::MayNamedAccess(checked, name, v8::ACCESS_GET)) {
return checked->GetPropertyWithFailedAccessCheck(receiver,
result,
name,
attributes);
}
}
// Stop traversing the chain once we reach the last object in the
// chain; either the holder of the result or null in case of an
// absent property.
if (current == last) break;
}
if (!result->IsProperty()) {
*attributes = ABSENT;
return Heap::undefined_value();
}
*attributes = result->GetAttributes();
if (!result->IsLoaded()) {
return JSObject::cast(this)->GetLazyProperty(receiver,
result,
name,
attributes);
}
Object* value;
JSObject* holder = result->holder();
switch (result->type()) {
case NORMAL:
value = holder->GetNormalizedProperty(result);
ASSERT(!value->IsTheHole() || result->IsReadOnly());
return value->IsTheHole() ? Heap::undefined_value() : value;
case FIELD:
value = holder->FastPropertyAt(result->GetFieldIndex());
ASSERT(!value->IsTheHole() || result->IsReadOnly());
return value->IsTheHole() ? Heap::undefined_value() : value;
case CONSTANT_FUNCTION:
return result->GetConstantFunction();
case CALLBACKS:
return GetPropertyWithCallback(receiver,
result->GetCallbackObject(),
name,
holder);
case INTERCEPTOR: {
JSObject* recvr = JSObject::cast(receiver);
return holder->GetPropertyWithInterceptor(recvr, name, attributes);
}
default:
UNREACHABLE();
return NULL;
}
}
Object* Object::GetElementWithReceiver(Object* receiver, uint32_t index) {
// Non-JS objects do not have integer indexed properties.
if (!IsJSObject()) return Heap::undefined_value();
return JSObject::cast(this)->GetElementWithReceiver(JSObject::cast(receiver),
index);
}
Object* Object::GetPrototype() {
// The object is either a number, a string, a boolean, or a real JS object.
if (IsJSObject()) return JSObject::cast(this)->map()->prototype();
Context* context = Top::context()->global_context();
if (IsNumber()) return context->number_function()->instance_prototype();
if (IsString()) return context->string_function()->instance_prototype();
if (IsBoolean()) {
return context->boolean_function()->instance_prototype();
} else {
return Heap::null_value();
}
}
void Object::ShortPrint() {
HeapStringAllocator allocator;
StringStream accumulator(&allocator);
ShortPrint(&accumulator);
accumulator.OutputToStdOut();
}
void Object::ShortPrint(StringStream* accumulator) {
if (IsSmi()) {
Smi::cast(this)->SmiPrint(accumulator);
} else if (IsFailure()) {
Failure::cast(this)->FailurePrint(accumulator);
} else {
HeapObject::cast(this)->HeapObjectShortPrint(accumulator);
}
}
void Smi::SmiPrint() {
PrintF("%d", value());
}
void Smi::SmiPrint(StringStream* accumulator) {
accumulator->Add("%d", value());
}
void Failure::FailurePrint(StringStream* accumulator) {
accumulator->Add("Failure(%d)", value());
}
void Failure::FailurePrint() {
PrintF("Failure(%d)", value());
}
Failure* Failure::RetryAfterGC(int requested_bytes, AllocationSpace space) {
ASSERT((space & ~kSpaceTagMask) == 0);
// TODO(X64): Stop using Smi validation for non-smi checks, even if they
// happen to be identical at the moment.
int requested = requested_bytes >> kObjectAlignmentBits;
int value = (requested << kSpaceTagSize) | space;
// We can't very well allocate a heap number in this situation, and if the
// requested memory is so large it seems reasonable to say that this is an
// out of memory situation. This fixes a crash in
// js1_5/Regress/regress-303213.js.
if (value >> kSpaceTagSize != requested ||
!Smi::IsValid(value) ||
value != ((value << kFailureTypeTagSize) >> kFailureTypeTagSize) ||
!Smi::IsValid(value << kFailureTypeTagSize)) {
Top::context()->mark_out_of_memory();
return Failure::OutOfMemoryException();
}
return Construct(RETRY_AFTER_GC, value);
}
// Should a word be prefixed by 'a' or 'an' in order to read naturally in
// English? Returns false for non-ASCII or words that don't start with
// a capital letter. The a/an rule follows pronunciation in English.
// We don't use the BBC's overcorrect "an historic occasion" though if
// you speak a dialect you may well say "an 'istoric occasion".
static bool AnWord(String* str) {
if (str->length() == 0) return false; // A nothing.
int c0 = str->Get(0);
int c1 = str->length() > 1 ? str->Get(1) : 0;
if (c0 == 'U') {
if (c1 > 'Z') {
return true; // An Umpire, but a UTF8String, a U.
}
} else if (c0 == 'A' || c0 == 'E' || c0 == 'I' || c0 == 'O') {
return true; // An Ape, an ABCBook.
} else if ((c1 == 0 || (c1 >= 'A' && c1 <= 'Z')) &&
(c0 == 'F' || c0 == 'H' || c0 == 'M' || c0 == 'N' || c0 == 'R' ||
c0 == 'S' || c0 == 'X')) {
return true; // An MP3File, an M.
}
return false;
}
Object* String::TryFlatten() {
#ifdef DEBUG
// Do not attempt to flatten in debug mode when allocation is not
// allowed. This is to avoid an assertion failure when allocating.
// Flattening strings is the only case where we always allow
// allocation because no GC is performed if the allocation fails.
if (!Heap::IsAllocationAllowed()) return this;
#endif
switch (StringShape(this).representation_tag()) {
case kSlicedStringTag: {
SlicedString* ss = SlicedString::cast(this);
// The SlicedString constructor should ensure that there are no
// SlicedStrings that are constructed directly on top of other
// SlicedStrings.
String* buf = ss->buffer();
ASSERT(!buf->IsSlicedString());
Object* ok = buf->TryFlatten();
if (ok->IsFailure()) return ok;
// Under certain circumstances (TryFlattenIfNotFlat fails in
// String::Slice) we can have a cons string under a slice.
// In this case we need to get the flat string out of the cons!
if (StringShape(String::cast(ok)).IsCons()) {
ss->set_buffer(ConsString::cast(ok)->first());
}
return this;
}
case kConsStringTag: {
ConsString* cs = ConsString::cast(this);
if (cs->second()->length() == 0) {
return this;
}
// There's little point in putting the flat string in new space if the
// cons string is in old space. It can never get GCed until there is
// an old space GC.
PretenureFlag tenure = Heap::InNewSpace(this) ? NOT_TENURED : TENURED;
int len = length();
Object* object;
String* result;
if (IsAsciiRepresentation()) {
object = Heap::AllocateRawAsciiString(len, tenure);
if (object->IsFailure()) return object;
result = String::cast(object);
String* first = cs->first();
int first_length = first->length();
char* dest = SeqAsciiString::cast(result)->GetChars();
WriteToFlat(first, dest, 0, first_length);
String* second = cs->second();
WriteToFlat(second,
dest + first_length,
0,
len - first_length);
} else {
object = Heap::AllocateRawTwoByteString(len, tenure);
if (object->IsFailure()) return object;
result = String::cast(object);
uc16* dest = SeqTwoByteString::cast(result)->GetChars();
String* first = cs->first();
int first_length = first->length();
WriteToFlat(first, dest, 0, first_length);
String* second = cs->second();
WriteToFlat(second,
dest + first_length,
0,
len - first_length);
}
cs->set_first(result);
cs->set_second(Heap::empty_string());
return this;
}
default:
return this;
}
}
bool String::MakeExternal(v8::String::ExternalStringResource* resource) {
#ifdef DEBUG
{ // NOLINT (presubmit.py gets confused about if and braces)
// Assert that the resource and the string are equivalent.
ASSERT(static_cast<size_t>(this->length()) == resource->length());
SmartPointer<uc16> smart_chars = this->ToWideCString();
ASSERT(memcmp(*smart_chars,
resource->data(),
resource->length() * sizeof(**smart_chars)) == 0);
}
#endif // DEBUG
int size = this->Size(); // Byte size of the original string.
if (size < ExternalString::kSize) {
// The string is too small to fit an external String in its place. This can
// only happen for zero length strings.
return false;
}
ASSERT(size >= ExternalString::kSize);
bool is_symbol = this->IsSymbol();
int length = this->length();
// Morph the object to an external string by adjusting the map and
// reinitializing the fields.
this->set_map(ExternalTwoByteString::StringMap(length));
ExternalTwoByteString* self = ExternalTwoByteString::cast(this);
self->set_length(length);
self->set_resource(resource);
// Additionally make the object into an external symbol if the original string
// was a symbol to start with.
if (is_symbol) {
self->Hash(); // Force regeneration of the hash value.
// Now morph this external string into a external symbol.
self->set_map(ExternalTwoByteString::SymbolMap(length));
}
// Fill the remainder of the string with dead wood.
int new_size = this->Size(); // Byte size of the external String object.
Heap::CreateFillerObjectAt(this->address() + new_size, size - new_size);
return true;
}
bool String::MakeExternal(v8::String::ExternalAsciiStringResource* resource) {
#ifdef DEBUG
{ // NOLINT (presubmit.py gets confused about if and braces)
// Assert that the resource and the string are equivalent.
ASSERT(static_cast<size_t>(this->length()) == resource->length());
SmartPointer<char> smart_chars = this->ToCString();
ASSERT(memcmp(*smart_chars,
resource->data(),
resource->length()*sizeof(**smart_chars)) == 0);
}
#endif // DEBUG
int size = this->Size(); // Byte size of the original string.
if (size < ExternalString::kSize) {
// The string is too small to fit an external String in its place. This can
// only happen for zero length strings.
return false;
}
ASSERT(size >= ExternalString::kSize);
bool is_symbol = this->IsSymbol();
int length = this->length();
// Morph the object to an external string by adjusting the map and
// reinitializing the fields.
this->set_map(ExternalAsciiString::StringMap(length));
ExternalAsciiString* self = ExternalAsciiString::cast(this);
self->set_length(length);
self->set_resource(resource);
// Additionally make the object into an external symbol if the original string
// was a symbol to start with.
if (is_symbol) {
self->Hash(); // Force regeneration of the hash value.
// Now morph this external string into a external symbol.
self->set_map(ExternalAsciiString::SymbolMap(length));
}
// Fill the remainder of the string with dead wood.
int new_size = this->Size(); // Byte size of the external String object.
Heap::CreateFillerObjectAt(this->address() + new_size, size - new_size);
return true;
}
void String::StringShortPrint(StringStream* accumulator) {
int len = length();
if (len > kMaxMediumStringSize) {
accumulator->Add("<Very long string[%u]>", len);
return;
}
if (!LooksValid()) {
accumulator->Add("<Invalid String>");
return;
}
StringInputBuffer buf(this);
bool truncated = false;
if (len > kMaxShortPrintLength) {
len = kMaxShortPrintLength;
truncated = true;
}
bool ascii = true;
for (int i = 0; i < len; i++) {
int c = buf.GetNext();
if (c < 32 || c >= 127) {
ascii = false;
}
}
buf.Reset(this);
if (ascii) {
accumulator->Add("<String[%u]: ", length());
for (int i = 0; i < len; i++) {
accumulator->Put(buf.GetNext());
}
accumulator->Put('>');
} else {
// Backslash indicates that the string contains control
// characters and that backslashes are therefore escaped.
accumulator->Add("<String[%u]\\: ", length());
for (int i = 0; i < len; i++) {
int c = buf.GetNext();
if (c == '\n') {
accumulator->Add("\\n");
} else if (c == '\r') {
accumulator->Add("\\r");
} else if (c == '\\') {
accumulator->Add("\\\\");
} else if (c < 32 || c > 126) {
accumulator->Add("\\x%02x", c);
} else {
accumulator->Put(c);
}
}
if (truncated) {
accumulator->Put('.');
accumulator->Put('.');
accumulator->Put('.');
}
accumulator->Put('>');
}
return;
}
void JSObject::JSObjectShortPrint(StringStream* accumulator) {
switch (map()->instance_type()) {
case JS_ARRAY_TYPE: {
double length = JSArray::cast(this)->length()->Number();
accumulator->Add("<JS array[%u]>", static_cast<uint32_t>(length));
break;
}
case JS_REGEXP_TYPE: {
accumulator->Add("<JS RegExp>");
break;
}
case JS_FUNCTION_TYPE: {
Object* fun_name = JSFunction::cast(this)->shared()->name();
bool printed = false;
if (fun_name->IsString()) {
String* str = String::cast(fun_name);
if (str->length() > 0) {
accumulator->Add("<JS Function ");
accumulator->Put(str);
accumulator->Put('>');
printed = true;
}
}
if (!printed) {
accumulator->Add("<JS Function>");
}
break;
}
// All other JSObjects are rather similar to each other (JSObject,
// JSGlobalProxy, JSGlobalObject, JSUndetectableObject, JSValue).
default: {
Object* constructor = map()->constructor();
bool printed = false;
if (constructor->IsHeapObject() &&
!Heap::Contains(HeapObject::cast(constructor))) {
accumulator->Add("!!!INVALID CONSTRUCTOR!!!");
} else {
bool global_object = IsJSGlobalProxy();
if (constructor->IsJSFunction()) {
if (!Heap::Contains(JSFunction::cast(constructor)->shared())) {
accumulator->Add("!!!INVALID SHARED ON CONSTRUCTOR!!!");
} else {
Object* constructor_name =
JSFunction::cast(constructor)->shared()->name();
if (constructor_name->IsString()) {
String* str = String::cast(constructor_name);
if (str->length() > 0) {
bool vowel = AnWord(str);
accumulator->Add("<%sa%s ",
global_object ? "Global Object: " : "",
vowel ? "n" : "");
accumulator->Put(str);
accumulator->Put('>');
printed = true;
}
}
}
}
if (!printed) {
accumulator->Add("<JS %sObject", global_object ? "Global " : "");
}
}
if (IsJSValue()) {
accumulator->Add(" value = ");
JSValue::cast(this)->value()->ShortPrint(accumulator);
}
accumulator->Put('>');
break;
}
}
}
void HeapObject::HeapObjectShortPrint(StringStream* accumulator) {
// if (!Heap::InNewSpace(this)) PrintF("*", this);
if (!Heap::Contains(this)) {
accumulator->Add("!!!INVALID POINTER!!!");
return;
}
if (!Heap::Contains(map())) {
accumulator->Add("!!!INVALID MAP!!!");
return;
}
accumulator->Add("%p ", this);
if (IsString()) {
String::cast(this)->StringShortPrint(accumulator);
return;
}
if (IsJSObject()) {
JSObject::cast(this)->JSObjectShortPrint(accumulator);
return;
}
switch (map()->instance_type()) {
case MAP_TYPE:
accumulator->Add("<Map>");
break;
case FIXED_ARRAY_TYPE:
accumulator->Add("<FixedArray[%u]>", FixedArray::cast(this)->length());
break;
case BYTE_ARRAY_TYPE:
accumulator->Add("<ByteArray[%u]>", ByteArray::cast(this)->length());
break;
case PIXEL_ARRAY_TYPE:
accumulator->Add("<PixelArray[%u]>", PixelArray::cast(this)->length());
break;
case SHARED_FUNCTION_INFO_TYPE:
accumulator->Add("<SharedFunctionInfo>");
break;
#define MAKE_STRUCT_CASE(NAME, Name, name) \
case NAME##_TYPE: \
accumulator->Put('<'); \
accumulator->Add(#Name); \
accumulator->Put('>'); \
break;
STRUCT_LIST(MAKE_STRUCT_CASE)
#undef MAKE_STRUCT_CASE
case CODE_TYPE:
accumulator->Add("<Code>");
break;
case ODDBALL_TYPE: {
if (IsUndefined())
accumulator->Add("<undefined>");
else if (IsTheHole())
accumulator->Add("<the hole>");
else if (IsNull())
accumulator->Add("<null>");
else if (IsTrue())
accumulator->Add("<true>");
else if (IsFalse())
accumulator->Add("<false>");
else
accumulator->Add("<Odd Oddball>");
break;
}
case HEAP_NUMBER_TYPE:
accumulator->Add("<Number: ");
HeapNumber::cast(this)->HeapNumberPrint(accumulator);
accumulator->Put('>');
break;
case PROXY_TYPE:
accumulator->Add("<Proxy>");
break;
case JS_GLOBAL_PROPERTY_CELL_TYPE:
accumulator->Add("Cell for ");
JSGlobalPropertyCell::cast(this)->value()->ShortPrint(accumulator);
break;
default:
accumulator->Add("<Other heap object (%d)>", map()->instance_type());
break;
}
}
int HeapObject::SlowSizeFromMap(Map* map) {
// Avoid calling functions such as FixedArray::cast during GC, which
// read map pointer of this object again.
InstanceType instance_type = map->instance_type();
uint32_t type = static_cast<uint32_t>(instance_type);
if (instance_type < FIRST_NONSTRING_TYPE
&& (StringShape(instance_type).IsSequential())) {
if ((type & kStringEncodingMask) == kAsciiStringTag) {
SeqAsciiString* seq_ascii_this = reinterpret_cast<SeqAsciiString*>(this);
return seq_ascii_this->SeqAsciiStringSize(instance_type);
} else {
SeqTwoByteString* self = reinterpret_cast<SeqTwoByteString*>(this);
return self->SeqTwoByteStringSize(instance_type);
}
}
switch (instance_type) {
case FIXED_ARRAY_TYPE:
return reinterpret_cast<FixedArray*>(this)->FixedArraySize();
case BYTE_ARRAY_TYPE:
return reinterpret_cast<ByteArray*>(this)->ByteArraySize();
case CODE_TYPE:
return reinterpret_cast<Code*>(this)->CodeSize();
case MAP_TYPE:
return Map::kSize;
default:
return map->instance_size();
}
}
void HeapObject::Iterate(ObjectVisitor* v) {
// Handle header
IteratePointer(v, kMapOffset);
// Handle object body
Map* m = map();
IterateBody(m->instance_type(), SizeFromMap(m), v);
}
void HeapObject::IterateBody(InstanceType type, int object_size,
ObjectVisitor* v) {
// Avoiding <Type>::cast(this) because it accesses the map pointer field.
// During GC, the map pointer field is encoded.
if (type < FIRST_NONSTRING_TYPE) {
switch (type & kStringRepresentationMask) {
case kSeqStringTag:
break;
case kConsStringTag:
reinterpret_cast<ConsString*>(this)->ConsStringIterateBody(v);
break;
case kSlicedStringTag:
reinterpret_cast<SlicedString*>(this)->SlicedStringIterateBody(v);
break;
}
return;
}
switch (type) {
case FIXED_ARRAY_TYPE:
reinterpret_cast<FixedArray*>(this)->FixedArrayIterateBody(v);
break;
case JS_OBJECT_TYPE:
case JS_CONTEXT_EXTENSION_OBJECT_TYPE:
case JS_VALUE_TYPE:
case JS_ARRAY_TYPE:
case JS_REGEXP_TYPE:
case JS_FUNCTION_TYPE:
case JS_GLOBAL_PROXY_TYPE:
case JS_GLOBAL_OBJECT_TYPE:
case JS_BUILTINS_OBJECT_TYPE:
reinterpret_cast<JSObject*>(this)->JSObjectIterateBody(object_size, v);
break;
case ODDBALL_TYPE:
reinterpret_cast<Oddball*>(this)->OddballIterateBody(v);
break;
case PROXY_TYPE:
reinterpret_cast<Proxy*>(this)->ProxyIterateBody(v);
break;
case MAP_TYPE:
reinterpret_cast<Map*>(this)->MapIterateBody(v);
break;
case CODE_TYPE:
reinterpret_cast<Code*>(this)->CodeIterateBody(v);
break;
case JS_GLOBAL_PROPERTY_CELL_TYPE:
reinterpret_cast<JSGlobalPropertyCell*>(this)
->JSGlobalPropertyCellIterateBody(v);
break;
case HEAP_NUMBER_TYPE:
case FILLER_TYPE:
case BYTE_ARRAY_TYPE:
case PIXEL_ARRAY_TYPE:
break;
case SHARED_FUNCTION_INFO_TYPE: {
SharedFunctionInfo* shared = reinterpret_cast<SharedFunctionInfo*>(this);
shared->SharedFunctionInfoIterateBody(v);
break;
}
#define MAKE_STRUCT_CASE(NAME, Name, name) \
case NAME##_TYPE:
STRUCT_LIST(MAKE_STRUCT_CASE)
#undef MAKE_STRUCT_CASE
IterateStructBody(object_size, v);
break;
default:
PrintF("Unknown type: %d\n", type);
UNREACHABLE();
}
}
void HeapObject::IterateStructBody(int object_size, ObjectVisitor* v) {
IteratePointers(v, HeapObject::kHeaderSize, object_size);
}
Object* HeapNumber::HeapNumberToBoolean() {
// NaN, +0, and -0 should return the false object
switch (fpclassify(value())) {
case FP_NAN: // fall through
case FP_ZERO: return Heap::false_value();
default: return Heap::true_value();
}
}
void HeapNumber::HeapNumberPrint() {
PrintF("%.16g", Number());
}
void HeapNumber::HeapNumberPrint(StringStream* accumulator) {
// The Windows version of vsnprintf can allocate when printing a %g string
// into a buffer that may not be big enough. We don't want random memory
// allocation when producing post-crash stack traces, so we print into a
// buffer that is plenty big enough for any floating point number, then
// print that using vsnprintf (which may truncate but never allocate if
// there is no more space in the buffer).
EmbeddedVector<char, 100> buffer;
OS::SNPrintF(buffer, "%.16g", Number());
accumulator->Add("%s", buffer.start());
}
String* JSObject::class_name() {
if (IsJSFunction()) {
return Heap::function_class_symbol();
}
if (map()->constructor()->IsJSFunction()) {
JSFunction* constructor = JSFunction::cast(map()->constructor());
return String::cast(constructor->shared()->instance_class_name());
}
// If the constructor is not present, return "Object".
return Heap::Object_symbol();
}
String* JSObject::constructor_name() {
if (IsJSFunction()) {
return Heap::function_class_symbol();
}
if (map()->constructor()->IsJSFunction()) {
JSFunction* constructor = JSFunction::cast(map()->constructor());
String* name = String::cast(constructor->shared()->name());
return name->length() > 0 ? name : constructor->shared()->inferred_name();
}
// If the constructor is not present, return "Object".
return Heap::Object_symbol();
}
void JSObject::JSObjectIterateBody(int object_size, ObjectVisitor* v) {
// Iterate over all fields in the body. Assumes all are Object*.
IteratePointers(v, kPropertiesOffset, object_size);
}
Object* JSObject::AddFastPropertyUsingMap(Map* new_map,
String* name,
Object* value) {
int index = new_map->PropertyIndexFor(name);
if (map()->unused_property_fields() == 0) {
ASSERT(map()->unused_property_fields() == 0);
int new_unused = new_map->unused_property_fields();
Object* values =
properties()->CopySize(properties()->length() + new_unused + 1);
if (values->IsFailure()) return values;
set_properties(FixedArray::cast(values));
}
set_map(new_map);
return FastPropertyAtPut(index, value);
}
Object* JSObject::AddFastProperty(String* name,
Object* value,
PropertyAttributes attributes) {
// Normalize the object if the name is an actual string (not the
// hidden symbols) and is not a real identifier.
StringInputBuffer buffer(name);
if (!Scanner::IsIdentifier(&buffer) && name != Heap::hidden_symbol()) {
Object* obj = NormalizeProperties(CLEAR_INOBJECT_PROPERTIES, 0);
if (obj->IsFailure()) return obj;
return AddSlowProperty(name, value, attributes);
}
DescriptorArray* old_descriptors = map()->instance_descriptors();
// Compute the new index for new field.
int index = map()->NextFreePropertyIndex();
// Allocate new instance descriptors with (name, index) added
FieldDescriptor new_field(name, index, attributes);
Object* new_descriptors =
old_descriptors->CopyInsert(&new_field, REMOVE_TRANSITIONS);
if (new_descriptors->IsFailure()) return new_descriptors;
// Only allow map transition if the object's map is NOT equal to the
// global object_function's map and there is not a transition for name.
bool allow_map_transition =
!old_descriptors->Contains(name) &&
(Top::context()->global_context()->object_function()->map() != map());
ASSERT(index < map()->inobject_properties() ||
(index - map()->inobject_properties()) < properties()->length() ||
map()->unused_property_fields() == 0);
// Allocate a new map for the object.
Object* r = map()->CopyDropDescriptors();
if (r->IsFailure()) return r;
Map* new_map = Map::cast(r);
if (allow_map_transition) {
// Allocate new instance descriptors for the old map with map transition.
MapTransitionDescriptor d(name, Map::cast(new_map), attributes);
Object* r = old_descriptors->CopyInsert(&d, KEEP_TRANSITIONS);
if (r->IsFailure()) return r;
old_descriptors = DescriptorArray::cast(r);
}
if (map()->unused_property_fields() == 0) {
if (properties()->length() > kMaxFastProperties) {
Object* obj = NormalizeProperties(CLEAR_INOBJECT_PROPERTIES, 0);
if (obj->IsFailure()) return obj;
return AddSlowProperty(name, value, attributes);
}
// Make room for the new value
Object* values =
properties()->CopySize(properties()->length() + kFieldsAdded);
if (values->IsFailure()) return values;
set_properties(FixedArray::cast(values));
new_map->set_unused_property_fields(kFieldsAdded - 1);
} else {
new_map->set_unused_property_fields(map()->unused_property_fields() - 1);
}
// We have now allocated all the necessary objects.
// All the changes can be applied at once, so they are atomic.
map()->set_instance_descriptors(old_descriptors);
new_map->set_instance_descriptors(DescriptorArray::cast(new_descriptors));
set_map(new_map);
return FastPropertyAtPut(index, value);
}
Object* JSObject::AddConstantFunctionProperty(String* name,
JSFunction* function,
PropertyAttributes attributes) {
// Allocate new instance descriptors with (name, function) added
ConstantFunctionDescriptor d(name, function, attributes);
Object* new_descriptors =
map()->instance_descriptors()->CopyInsert(&d, REMOVE_TRANSITIONS);
if (new_descriptors->IsFailure()) return new_descriptors;
// Allocate a new map for the object.
Object* new_map = map()->CopyDropDescriptors();
if (new_map->IsFailure()) return new_map;
DescriptorArray* descriptors = DescriptorArray::cast(new_descriptors);
Map::cast(new_map)->set_instance_descriptors(descriptors);
Map* old_map = map();
set_map(Map::cast(new_map));
// If the old map is the global object map (from new Object()),
// then transitions are not added to it, so we are done.
if (old_map == Top::context()->global_context()->object_function()->map()) {
return function;
}
// Do not add CONSTANT_TRANSITIONS to global objects
if (IsGlobalObject()) {
return function;
}
// Add a CONSTANT_TRANSITION descriptor to the old map,
// so future assignments to this property on other objects
// of the same type will create a normal field, not a constant function.
// Don't do this for special properties, with non-trival attributes.
if (attributes != NONE) {
return function;
}
ConstTransitionDescriptor mark(name);
new_descriptors =
old_map->instance_descriptors()->CopyInsert(&mark, KEEP_TRANSITIONS);
if (new_descriptors->IsFailure()) {
return function; // We have accomplished the main goal, so return success.
}
old_map->set_instance_descriptors(DescriptorArray::cast(new_descriptors));
return function;
}
// Add property in slow mode
Object* JSObject::AddSlowProperty(String* name,
Object* value,
PropertyAttributes attributes) {
ASSERT(!HasFastProperties());
StringDictionary* dict = property_dictionary();
Object* store_value = value;
if (IsGlobalObject()) {
// In case name is an orphaned property reuse the cell.
int entry = dict->FindEntry(name);
if (entry != StringDictionary::kNotFound) {
store_value = dict->ValueAt(entry);
JSGlobalPropertyCell::cast(store_value)->set_value(value);
// Assign an enumeration index to the property and update
// SetNextEnumerationIndex.
int index = dict->NextEnumerationIndex();
PropertyDetails details = PropertyDetails(attributes, NORMAL, index);
dict->SetNextEnumerationIndex(index + 1);
dict->SetEntry(entry, name, store_value, details);
return value;
}
store_value = Heap::AllocateJSGlobalPropertyCell(value);
if (store_value->IsFailure()) return store_value;
JSGlobalPropertyCell::cast(store_value)->set_value(value);
}
PropertyDetails details = PropertyDetails(attributes, NORMAL);
Object* result = dict->Add(name, store_value, details);
if (result->IsFailure()) return result;
if (dict != result) set_properties(StringDictionary::cast(result));
return value;
}
Object* JSObject::AddProperty(String* name,
Object* value,
PropertyAttributes attributes) {
ASSERT(!IsJSGlobalProxy());
if (HasFastProperties()) {
// Ensure the descriptor array does not get too big.
if (map()->instance_descriptors()->number_of_descriptors() <
DescriptorArray::kMaxNumberOfDescriptors) {
if (value->IsJSFunction()) {
return AddConstantFunctionProperty(name,
JSFunction::cast(value),
attributes);
} else {
return AddFastProperty(name, value, attributes);
}
} else {
// Normalize the object to prevent very large instance descriptors.
// This eliminates unwanted N^2 allocation and lookup behavior.
Object* obj = NormalizeProperties(CLEAR_INOBJECT_PROPERTIES, 0);
if (obj->IsFailure()) return obj;
}
}
return AddSlowProperty(name, value, attributes);
}
Object* JSObject::SetPropertyPostInterceptor(String* name,
Object* value,
PropertyAttributes attributes) {
// Check local property, ignore interceptor.
LookupResult result;
LocalLookupRealNamedProperty(name, &result);
if (result.IsValid()) return SetProperty(&result, name, value, attributes);
// Add real property.
return AddProperty(name, value, attributes);
}
Object* JSObject::ReplaceSlowProperty(String* name,
Object* value,
PropertyAttributes attributes) {
StringDictionary* dictionary = property_dictionary();
int old_index = dictionary->FindEntry(name);
int new_enumeration_index = 0; // 0 means "Use the next available index."
if (old_index != -1) {
// All calls to ReplaceSlowProperty have had all transitions removed.
ASSERT(!dictionary->DetailsAt(old_index).IsTransition());
new_enumeration_index = dictionary->DetailsAt(old_index).index();
}
PropertyDetails new_details(attributes, NORMAL, new_enumeration_index);
return SetNormalizedProperty(name, value, new_details);
}
Object* JSObject::ConvertDescriptorToFieldAndMapTransition(
String* name,
Object* new_value,
PropertyAttributes attributes) {
Map* old_map = map();
Object* result = ConvertDescriptorToField(name, new_value, attributes);
if (result->IsFailure()) return result;
// If we get to this point we have succeeded - do not return failure
// after this point. Later stuff is optional.
if (!HasFastProperties()) {
return result;
}
// Do not add transitions to the map of "new Object()".
if (map() == Top::context()->global_context()->object_function()->map()) {
return result;
}
MapTransitionDescriptor transition(name,
map(),
attributes);
Object* new_descriptors =
old_map->instance_descriptors()->
CopyInsert(&transition, KEEP_TRANSITIONS);
if (new_descriptors->IsFailure()) return result; // Yes, return _result_.
old_map->set_instance_descriptors(DescriptorArray::cast(new_descriptors));
return result;
}
Object* JSObject::ConvertDescriptorToField(String* name,
Object* new_value,
PropertyAttributes attributes) {
if (map()->unused_property_fields() == 0 &&
properties()->length() > kMaxFastProperties) {
Object* obj = NormalizeProperties(CLEAR_INOBJECT_PROPERTIES, 0);
if (obj->IsFailure()) return obj;
return ReplaceSlowProperty(name, new_value, attributes);
}
int index = map()->NextFreePropertyIndex();
FieldDescriptor new_field(name, index, attributes);
// Make a new DescriptorArray replacing an entry with FieldDescriptor.
Object* descriptors_unchecked = map()->instance_descriptors()->
CopyInsert(&new_field, REMOVE_TRANSITIONS);
if (descriptors_unchecked->IsFailure()) return descriptors_unchecked;
DescriptorArray* new_descriptors =
DescriptorArray::cast(descriptors_unchecked);
// Make a new map for the object.
Object* new_map_unchecked = map()->CopyDropDescriptors();
if (new_map_unchecked->IsFailure()) return new_map_unchecked;
Map* new_map = Map::cast(new_map_unchecked);
new_map->set_instance_descriptors(new_descriptors);
// Make new properties array if necessary.
FixedArray* new_properties = 0; // Will always be NULL or a valid pointer.
int new_unused_property_fields = map()->unused_property_fields() - 1;
if (map()->unused_property_fields() == 0) {
new_unused_property_fields = kFieldsAdded - 1;
Object* new_properties_unchecked =
properties()->CopySize(properties()->length() + kFieldsAdded);
if (new_properties_unchecked->IsFailure()) return new_properties_unchecked;
new_properties = FixedArray::cast(new_properties_unchecked);
}
// Update pointers to commit changes.
// Object points to the new map.
new_map->set_unused_property_fields(new_unused_property_fields);
set_map(new_map);
if (new_properties) {
set_properties(FixedArray::cast(new_properties));
}
return FastPropertyAtPut(index, new_value);
}
Object* JSObject::SetPropertyWithInterceptor(String* name,
Object* value,
PropertyAttributes attributes) {
HandleScope scope;
Handle<JSObject> this_handle(this);
Handle<String> name_handle(name);
Handle<Object> value_handle(value);
Handle<InterceptorInfo> interceptor(GetNamedInterceptor());
if (!interceptor->setter()->IsUndefined()) {
LOG(ApiNamedPropertyAccess("interceptor-named-set", this, name));
CustomArguments args(interceptor->data(), this, this);
v8::AccessorInfo info(args.end());
v8::NamedPropertySetter setter =
v8::ToCData<v8::NamedPropertySetter>(interceptor->setter());
v8::Handle<v8::Value> result;
{
// Leaving JavaScript.
VMState state(EXTERNAL);
Handle<Object> value_unhole(value->IsTheHole() ?
Heap::undefined_value() :
value);
result = setter(v8::Utils::ToLocal(name_handle),
v8::Utils::ToLocal(value_unhole),
info);
}
RETURN_IF_SCHEDULED_EXCEPTION();
if (!result.IsEmpty()) return *value_handle;
}
Object* raw_result = this_handle->SetPropertyPostInterceptor(*name_handle,
*value_handle,
attributes);
RETURN_IF_SCHEDULED_EXCEPTION();
return raw_result;
}
Object* JSObject::SetProperty(String* name,
Object* value,
PropertyAttributes attributes) {
LookupResult result;
LocalLookup(name, &result);
return SetProperty(&result, name, value, attributes);
}
Object* JSObject::SetPropertyWithCallback(Object* structure,
String* name,
Object* value,
JSObject* holder) {
HandleScope scope;
// We should never get here to initialize a const with the hole
// value since a const declaration would conflict with the setter.
ASSERT(!value->IsTheHole());
Handle<Object> value_handle(value);
// To accommodate both the old and the new api we switch on the
// data structure used to store the callbacks. Eventually proxy
// callbacks should be phased out.
if (structure->IsProxy()) {
AccessorDescriptor* callback =
reinterpret_cast<AccessorDescriptor*>(Proxy::cast(structure)->proxy());
Object* obj = (callback->setter)(this, value, callback->data);
RETURN_IF_SCHEDULED_EXCEPTION();
if (obj->IsFailure()) return obj;
return *value_handle;
}
if (structure->IsAccessorInfo()) {
// api style callbacks
AccessorInfo* data = AccessorInfo::cast(structure);
Object* call_obj = data->setter();
v8::AccessorSetter call_fun = v8::ToCData<v8::AccessorSetter>(call_obj);
if (call_fun == NULL) return value;
Handle<String> key(name);
LOG(ApiNamedPropertyAccess("store", this, name));
CustomArguments args(data->data(), this, JSObject::cast(holder));
v8::AccessorInfo info(args.end());
{
// Leaving JavaScript.
VMState state(EXTERNAL);
call_fun(v8::Utils::ToLocal(key),
v8::Utils::ToLocal(value_handle),
info);
}
RETURN_IF_SCHEDULED_EXCEPTION();
return *value_handle;
}
if (structure->IsFixedArray()) {
Object* setter = FixedArray::cast(structure)->get(kSetterIndex);
if (setter->IsJSFunction()) {
return SetPropertyWithDefinedSetter(JSFunction::cast(setter), value);
} else {
Handle<String> key(name);
Handle<Object> holder_handle(holder);
Handle<Object> args[2] = { key, holder_handle };
return Top::Throw(*Factory::NewTypeError("no_setter_in_callback",
HandleVector(args, 2)));
}
}
UNREACHABLE();
return 0;
}
Object* JSObject::SetPropertyWithDefinedSetter(JSFunction* setter,
Object* value) {
Handle<Object> value_handle(value);
Handle<JSFunction> fun(JSFunction::cast(setter));
Handle<JSObject> self(this);
#ifdef ENABLE_DEBUGGER_SUPPORT
// Handle stepping into a setter if step into is active.
if (Debug::StepInActive()) {
Debug::HandleStepIn(fun, Handle<Object>::null(), 0, false);
}
#endif
bool has_pending_exception;
Object** argv[] = { value_handle.location() };
Execution::Call(fun, self, 1, argv, &has_pending_exception);
// Check for pending exception and return the result.
if (has_pending_exception) return Failure::Exception();
return *value_handle;
}
void JSObject::LookupCallbackSetterInPrototypes(String* name,
LookupResult* result) {
for (Object* pt = GetPrototype();
pt != Heap::null_value();
pt = pt->GetPrototype()) {
JSObject::cast(pt)->LocalLookupRealNamedProperty(name, result);
if (result->IsValid()) {
if (!result->IsTransitionType() && result->IsReadOnly()) {
result->NotFound();
return;
}
if (result->type() == CALLBACKS) {
return;
}
}
}
result->NotFound();
}
Object* JSObject::LookupCallbackSetterInPrototypes(uint32_t index) {
for (Object* pt = GetPrototype();
pt != Heap::null_value();
pt = pt->GetPrototype()) {
if (!JSObject::cast(pt)->HasDictionaryElements()) {
continue;
}
NumberDictionary* dictionary = JSObject::cast(pt)->element_dictionary();
int entry = dictionary->FindEntry(index);
if (entry != NumberDictionary::kNotFound) {
Object* element = dictionary->ValueAt(entry);
PropertyDetails details = dictionary->DetailsAt(entry);
if (details.type() == CALLBACKS) {
// Only accessors allowed as elements.
return FixedArray::cast(element)->get(kSetterIndex);
}
}
}
return Heap::undefined_value();
}
void JSObject::LookupInDescriptor(String* name, LookupResult* result) {
DescriptorArray* descriptors = map()->instance_descriptors();
int number = DescriptorLookupCache::Lookup(descriptors, name);
if (number == DescriptorLookupCache::kAbsent) {
number = descriptors->Search(name);
DescriptorLookupCache::Update(descriptors, name, number);
}
if (number != DescriptorArray::kNotFound) {
result->DescriptorResult(this, descriptors->GetDetails(number), number);
} else {
result->NotFound();
}
}
void JSObject::LocalLookupRealNamedProperty(String* name,
LookupResult* result) {
if (IsJSGlobalProxy()) {
Object* proto = GetPrototype();
if (proto->IsNull()) return result->NotFound();
ASSERT(proto->IsJSGlobalObject());
return JSObject::cast(proto)->LocalLookupRealNamedProperty(name, result);
}
if (HasFastProperties()) {
LookupInDescriptor(name, result);
if (result->IsValid()) {
ASSERT(result->holder() == this && result->type() != NORMAL);
// Disallow caching for uninitialized constants. These can only
// occur as fields.
if (result->IsReadOnly() && result->type() == FIELD &&
FastPropertyAt(result->GetFieldIndex())->IsTheHole()) {
result->DisallowCaching();
}
return;
}
} else {
int entry = property_dictionary()->FindEntry(name);
if (entry != StringDictionary::kNotFound) {
Object* value = property_dictionary()->ValueAt(entry);
if (IsGlobalObject()) {
PropertyDetails d = property_dictionary()->DetailsAt(entry);
if (d.IsDeleted()) {
result->NotFound();
return;
}
value = JSGlobalPropertyCell::cast(value)->value();
ASSERT(result->IsLoaded());
}
// Make sure to disallow caching for uninitialized constants
// found in the dictionary-mode objects.
if (value->IsTheHole()) result->DisallowCaching();
result->DictionaryResult(this, entry);
return;
}
// Slow case object skipped during lookup. Do not use inline caching.
if (!IsGlobalObject()) result->DisallowCaching();
}
result->NotFound();
}
void JSObject::LookupRealNamedProperty(String* name, LookupResult* result) {
LocalLookupRealNamedProperty(name, result);
if (result->IsProperty()) return;
LookupRealNamedPropertyInPrototypes(name, result);
}
void JSObject::LookupRealNamedPropertyInPrototypes(String* name,
LookupResult* result) {
for (Object* pt = GetPrototype();
pt != Heap::null_value();
pt = JSObject::cast(pt)->GetPrototype()) {
JSObject::cast(pt)->LocalLookupRealNamedProperty(name, result);
if (result->IsValid()) {
switch (result->type()) {
case NORMAL:
case FIELD:
case CONSTANT_FUNCTION:
case CALLBACKS:
return;
default: break;
}
}
}
result->NotFound();
}
// We only need to deal with CALLBACKS and INTERCEPTORS
Object* JSObject::SetPropertyWithFailedAccessCheck(LookupResult* result,
String* name,
Object* value) {
if (!result->IsProperty()) {
LookupCallbackSetterInPrototypes(name, result);
}
if (result->IsProperty()) {
if (!result->IsReadOnly()) {
switch (result->type()) {
case CALLBACKS: {
Object* obj = result->GetCallbackObject();
if (obj->IsAccessorInfo()) {
AccessorInfo* info = AccessorInfo::cast(obj);
if (info->all_can_write()) {
return SetPropertyWithCallback(result->GetCallbackObject(),
name,
value,
result->holder());
}
}
break;
}
case INTERCEPTOR: {
// Try lookup real named properties. Note that only property can be
// set is callbacks marked as ALL_CAN_WRITE on the prototype chain.
LookupResult r;
LookupRealNamedProperty(name, &r);
if (r.IsProperty()) {
return SetPropertyWithFailedAccessCheck(&r, name, value);
}
break;
}
default: {
break;
}
}
}
}
Top::ReportFailedAccessCheck(this, v8::ACCESS_SET);
return value;
}
Object* JSObject::SetProperty(LookupResult* result,
String* name,
Object* value,
PropertyAttributes attributes) {
// Make sure that the top context does not change when doing callbacks or
// interceptor calls.
AssertNoContextChange ncc;
// Check access rights if needed.
if (IsAccessCheckNeeded()
&& !Top::MayNamedAccess(this, name, v8::ACCESS_SET)) {
return SetPropertyWithFailedAccessCheck(result, name, value);
}
if (IsJSGlobalProxy()) {
Object* proto = GetPrototype();
if (proto->IsNull()) return value;
ASSERT(proto->IsJSGlobalObject());
return JSObject::cast(proto)->SetProperty(result, name, value, attributes);
}
if (!result->IsProperty() && !IsJSContextExtensionObject()) {
// We could not find a local property so let's check whether there is an
// accessor that wants to handle the property.
LookupResult accessor_result;
LookupCallbackSetterInPrototypes(name, &accessor_result);
if (accessor_result.IsValid()) {
return SetPropertyWithCallback(accessor_result.GetCallbackObject(),
name,
value,
accessor_result.holder());
}
}
if (result->IsNotFound()) {
return AddProperty(name, value, attributes);
}
if (!result->IsLoaded()) {
return SetLazyProperty(result, name, value, attributes);
}
if (result->IsReadOnly() && result->IsProperty()) return value;
// This is a real property that is not read-only, or it is a
// transition or null descriptor and there are no setters in the prototypes.
switch (result->type()) {
case NORMAL:
return SetNormalizedProperty(result, value);
case FIELD:
return FastPropertyAtPut(result->GetFieldIndex(), value);
case MAP_TRANSITION:
if (attributes == result->GetAttributes()) {
// Only use map transition if the attributes match.
return AddFastPropertyUsingMap(result->GetTransitionMap(),
name,
value);
}
return ConvertDescriptorToField(name, value, attributes);
case CONSTANT_FUNCTION:
// Only replace the function if necessary.
if (value == result->GetConstantFunction()) return value;
// Preserve the attributes of this existing property.
attributes = result->GetAttributes();
return ConvertDescriptorToField(name, value, attributes);
case CALLBACKS:
return SetPropertyWithCallback(result->GetCallbackObject(),
name,
value,
result->holder());
case INTERCEPTOR:
return SetPropertyWithInterceptor(name, value, attributes);
case CONSTANT_TRANSITION:
// Replace with a MAP_TRANSITION to a new map with a FIELD, even
// if the value is a function.
return ConvertDescriptorToFieldAndMapTransition(name, value, attributes);
case NULL_DESCRIPTOR:
return ConvertDescriptorToFieldAndMapTransition(name, value, attributes);
default:
UNREACHABLE();
}
UNREACHABLE();
return value;
}
// Set a real local property, even if it is READ_ONLY. If the property is not
// present, add it with attributes NONE. This code is an exact clone of
// SetProperty, with the check for IsReadOnly and the check for a
// callback setter removed. The two lines looking up the LookupResult
// result are also added. If one of the functions is changed, the other
// should be.
Object* JSObject::IgnoreAttributesAndSetLocalProperty(
String* name,
Object* value,
PropertyAttributes attributes) {
// Make sure that the top context does not change when doing callbacks or
// interceptor calls.
AssertNoContextChange ncc;
// ADDED TO CLONE
LookupResult result_struct;
LocalLookup(name, &result_struct);
LookupResult* result = &result_struct;
// END ADDED TO CLONE
// Check access rights if needed.
if (IsAccessCheckNeeded()
&& !Top::MayNamedAccess(this, name, v8::ACCESS_SET)) {
return SetPropertyWithFailedAccessCheck(result, name, value);
}
if (IsJSGlobalProxy()) {
Object* proto = GetPrototype();
if (proto->IsNull()) return value;
ASSERT(proto->IsJSGlobalObject());
return JSObject::cast(proto)->IgnoreAttributesAndSetLocalProperty(
name,
value,
attributes);
}
// Check for accessor in prototype chain removed here in clone.
if (result->IsNotFound()) {
return AddProperty(name, value, attributes);
}
if (!result->IsLoaded()) {
return SetLazyProperty(result, name, value, attributes);
}
// Check of IsReadOnly removed from here in clone.
switch (result->type()) {
case NORMAL:
return SetNormalizedProperty(result, value);
case FIELD:
return FastPropertyAtPut(result->GetFieldIndex(), value);
case MAP_TRANSITION:
if (attributes == result->GetAttributes()) {
// Only use map transition if the attributes match.
return AddFastPropertyUsingMap(result->GetTransitionMap(),
name,
value);
}
return ConvertDescriptorToField(name, value, attributes);
case CONSTANT_FUNCTION:
// Only replace the function if necessary.
if (value == result->GetConstantFunction()) return value;
// Preserve the attributes of this existing property.
attributes = result->GetAttributes();
return ConvertDescriptorToField(name, value, attributes);
case CALLBACKS:
case INTERCEPTOR:
// Override callback in clone
return ConvertDescriptorToField(name, value, attributes);
case CONSTANT_TRANSITION:
// Replace with a MAP_TRANSITION to a new map with a FIELD, even
// if the value is a function.
return ConvertDescriptorToFieldAndMapTransition(name, value, attributes);
case NULL_DESCRIPTOR:
return ConvertDescriptorToFieldAndMapTransition(name, value, attributes);
default:
UNREACHABLE();
}
UNREACHABLE();
return value;
}
PropertyAttributes JSObject::GetPropertyAttributePostInterceptor(
JSObject* receiver,
String* name,
bool continue_search) {
// Check local property, ignore interceptor.
LookupResult result;
LocalLookupRealNamedProperty(name, &result);
if (result.IsProperty()) return result.GetAttributes();
if (continue_search) {
// Continue searching via the prototype chain.
Object* pt = GetPrototype();
if (pt != Heap::null_value()) {
return JSObject::cast(pt)->
GetPropertyAttributeWithReceiver(receiver, name);
}
}
return ABSENT;
}
PropertyAttributes JSObject::GetPropertyAttributeWithInterceptor(
JSObject* receiver,
String* name,
bool continue_search) {
// Make sure that the top context does not change when doing
// callbacks or interceptor calls.
AssertNoContextChange ncc;
HandleScope scope;
Handle<InterceptorInfo> interceptor(GetNamedInterceptor());
Handle<JSObject> receiver_handle(receiver);
Handle<JSObject> holder_handle(this);
Handle<String> name_handle(name);
CustomArguments args(interceptor->data(), receiver, this);
v8::AccessorInfo info(args.end());
if (!interceptor->query()->IsUndefined()) {
v8::NamedPropertyQuery query =
v8::ToCData<v8::NamedPropertyQuery>(interceptor->query());
LOG(ApiNamedPropertyAccess("interceptor-named-has", *holder_handle, name));
v8::Handle<v8::Boolean> result;
{
// Leaving JavaScript.
VMState state(EXTERNAL);
result = query(v8::Utils::ToLocal(name_handle), info);
}
if (!result.IsEmpty()) {
// Convert the boolean result to a property attribute
// specification.
return result->IsTrue() ? NONE : ABSENT;
}
} else if (!interceptor->getter()->IsUndefined()) {
v8::NamedPropertyGetter getter =
v8::ToCData<v8::NamedPropertyGetter>(interceptor->getter());
LOG(ApiNamedPropertyAccess("interceptor-named-get-has", this, name));
v8::Handle<v8::Value> result;
{
// Leaving JavaScript.
VMState state(EXTERNAL);
result = getter(v8::Utils::ToLocal(name_handle), info);
}
if (!result.IsEmpty()) return NONE;
}
return holder_handle->GetPropertyAttributePostInterceptor(*receiver_handle,
*name_handle,
continue_search);
}
PropertyAttributes JSObject::GetPropertyAttributeWithReceiver(
JSObject* receiver,
String* key) {
uint32_t index = 0;
if (key->AsArrayIndex(&index)) {
if (HasElementWithReceiver(receiver, index)) return NONE;
return ABSENT;
}
// Named property.
LookupResult result;
Lookup(key, &result);
return GetPropertyAttribute(receiver, &result, key, true);
}
PropertyAttributes JSObject::GetPropertyAttribute(JSObject* receiver,
LookupResult* result,
String* name,
bool continue_search) {
// Check access rights if needed.
if (IsAccessCheckNeeded() &&
!Top::MayNamedAccess(this, name, v8::ACCESS_HAS)) {
return GetPropertyAttributeWithFailedAccessCheck(receiver,
result,
name,
continue_search);
}
if (result->IsValid()) {
switch (result->type()) {
case NORMAL: // fall through
case FIELD:
case CONSTANT_FUNCTION:
case CALLBACKS:
return result->GetAttributes();
case INTERCEPTOR:
return result->holder()->
GetPropertyAttributeWithInterceptor(receiver, name, continue_search);
case MAP_TRANSITION:
case CONSTANT_TRANSITION:
case NULL_DESCRIPTOR:
return ABSENT;
default:
UNREACHABLE();
break;
}
}
return ABSENT;
}
PropertyAttributes JSObject::GetLocalPropertyAttribute(String* name) {
// Check whether the name is an array index.
uint32_t index = 0;
if (name->AsArrayIndex(&index)) {
if (HasLocalElement(index)) return NONE;
return ABSENT;
}
// Named property.
LookupResult result;
LocalLookup(name, &result);
return GetPropertyAttribute(this, &result, name, false);
}
Object* JSObject::NormalizeProperties(PropertyNormalizationMode mode,
int expected_additional_properties) {
if (!HasFastProperties()) return this;
// The global object is always normalized.
ASSERT(!IsGlobalObject());
// Allocate new content.
int property_count = map()->NumberOfDescribedProperties();
if (expected_additional_properties > 0) {
property_count += expected_additional_properties;
} else {
property_count += 2; // Make space for two more properties.
}
Object* obj =
StringDictionary::Allocate(property_count * 2);
if (obj->IsFailure()) return obj;
StringDictionary* dictionary = StringDictionary::cast(obj);
DescriptorArray* descs = map()->instance_descriptors();
for (int i = 0; i < descs->number_of_descriptors(); i++) {
PropertyDetails details = descs->GetDetails(i);
switch (details.type()) {
case CONSTANT_FUNCTION: {
PropertyDetails d =
PropertyDetails(details.attributes(), NORMAL, details.index());
Object* value = descs->GetConstantFunction(i);
Object* result = dictionary->Add(descs->GetKey(i), value, d);
if (result->IsFailure()) return result;
dictionary = StringDictionary::cast(result);
break;
}
case FIELD: {
PropertyDetails d =
PropertyDetails(details.attributes(), NORMAL, details.index());
Object* value = FastPropertyAt(descs->GetFieldIndex(i));
Object* result = dictionary->Add(descs->GetKey(i), value, d);
if (result->IsFailure()) return result;
dictionary = StringDictionary::cast(result);
break;
}
case CALLBACKS: {
PropertyDetails d =
PropertyDetails(details.attributes(), CALLBACKS, details.index());
Object* value = descs->GetCallbacksObject(i);
Object* result = dictionary->Add(descs->GetKey(i), value, d);
if (result->IsFailure()) return result;
dictionary = StringDictionary::cast(result);
break;
}
case MAP_TRANSITION:
case CONSTANT_TRANSITION:
case NULL_DESCRIPTOR:
case INTERCEPTOR:
break;
default:
UNREACHABLE();
}
}
// Copy the next enumeration index from instance descriptor.
int index = map()->instance_descriptors()->NextEnumerationIndex();
dictionary->SetNextEnumerationIndex(index);
// Allocate new map.
obj = map()->CopyDropDescriptors();
if (obj->IsFailure()) return obj;
Map* new_map = Map::cast(obj);
// Clear inobject properties if needed by adjusting the instance size and
// putting in a filler object instead of the inobject properties.
if (mode == CLEAR_INOBJECT_PROPERTIES && map()->inobject_properties() > 0) {
int instance_size_delta = map()->inobject_properties() * kPointerSize;
int new_instance_size = map()->instance_size() - instance_size_delta;
new_map->set_inobject_properties(0);
new_map->set_instance_size(new_instance_size);
Heap::CreateFillerObjectAt(this->address() + new_instance_size,
instance_size_delta);
}
new_map->set_unused_property_fields(0);
// We have now successfully allocated all the necessary objects.
// Changes can now be made with the guarantee that all of them take effect.
set_map(new_map);
map()->set_instance_descriptors(Heap::empty_descriptor_array());
set_properties(dictionary);
Counters::props_to_dictionary.Increment();
#ifdef DEBUG
if (FLAG_trace_normalization) {
PrintF("Object properties have been normalized:\n");
Print();
}
#endif
return this;
}
Object* JSObject::TransformToFastProperties(int unused_property_fields) {
if (HasFastProperties()) return this;
ASSERT(!IsGlobalObject());
return property_dictionary()->
TransformPropertiesToFastFor(this, unused_property_fields);
}
Object* JSObject::NormalizeElements() {
ASSERT(!HasPixelElements());
if (HasDictionaryElements()) return this;
// Get number of entries.
FixedArray* array = FixedArray::cast(elements());
// Compute the effective length.
int length = IsJSArray() ?
Smi::cast(JSArray::cast(this)->length())->value() :
array->length();
Object* obj = NumberDictionary::Allocate(length);
if (obj->IsFailure()) return obj;
NumberDictionary* dictionary = NumberDictionary::cast(obj);
// Copy entries.
for (int i = 0; i < length; i++) {
Object* value = array->get(i);
if (!value->IsTheHole()) {
PropertyDetails details = PropertyDetails(NONE, NORMAL);
Object* result = dictionary->AddNumberEntry(i, array->get(i), details);
if (result->IsFailure()) return result;
dictionary = NumberDictionary::cast(result);
}
}
// Switch to using the dictionary as the backing storage for elements.
set_elements(dictionary);
Counters::elements_to_dictionary.Increment();
#ifdef DEBUG
if (FLAG_trace_normalization) {
PrintF("Object elements have been normalized:\n");
Print();
}
#endif
return this;
}
Object* JSObject::DeletePropertyPostInterceptor(String* name, DeleteMode mode) {
// Check local property, ignore interceptor.
LookupResult result;
LocalLookupRealNamedProperty(name, &result);
if (!result.IsValid()) return Heap::true_value();
// Normalize object if needed.
Object* obj = NormalizeProperties(CLEAR_INOBJECT_PROPERTIES, 0);
if (obj->IsFailure()) return obj;
return DeleteNormalizedProperty(name, mode);
}
Object* JSObject::DeletePropertyWithInterceptor(String* name) {
HandleScope scope;
Handle<InterceptorInfo> interceptor(GetNamedInterceptor());
Handle<String> name_handle(name);
Handle<JSObject> this_handle(this);
if (!interceptor->deleter()->IsUndefined()) {
v8::NamedPropertyDeleter deleter =
v8::ToCData<v8::NamedPropertyDeleter>(interceptor->deleter());
LOG(ApiNamedPropertyAccess("interceptor-named-delete", *this_handle, name));
CustomArguments args(interceptor->data(), this, this);
v8::AccessorInfo info(args.end());
v8::Handle<v8::Boolean> result;
{
// Leaving JavaScript.
VMState state(EXTERNAL);
result = deleter(v8::Utils::ToLocal(name_handle), info);
}
RETURN_IF_SCHEDULED_EXCEPTION();
if (!result.IsEmpty()) {
ASSERT(result->IsBoolean());
return *v8::Utils::OpenHandle(*result);
}
}
Object* raw_result =
this_handle->DeletePropertyPostInterceptor(*name_handle, NORMAL_DELETION);
RETURN_IF_SCHEDULED_EXCEPTION();
return raw_result;
}
Object* JSObject::DeleteElementPostInterceptor(uint32_t index,
DeleteMode mode) {
ASSERT(!HasPixelElements());
switch (GetElementsKind()) {
case FAST_ELEMENTS: {
uint32_t length = IsJSArray() ?
static_cast<uint32_t>(Smi::cast(JSArray::cast(this)->length())->value()) :
static_cast<uint32_t>(FixedArray::cast(elements())->length());
if (index < length) {
FixedArray::cast(elements())->set_the_hole(index);
}
break;
}
case DICTIONARY_ELEMENTS: {
NumberDictionary* dictionary = element_dictionary();
int entry = dictionary->FindEntry(index);
if (entry != NumberDictionary::kNotFound) {
return dictionary->DeleteProperty(entry, mode);
}
break;
}
default:
UNREACHABLE();
break;
}
return Heap::true_value();
}
Object* JSObject::DeleteElementWithInterceptor(uint32_t index) {
// Make sure that the top context does not change when doing
// callbacks or interceptor calls.
AssertNoContextChange ncc;
HandleScope scope;
Handle<InterceptorInfo> interceptor(GetIndexedInterceptor());
if (interceptor->deleter()->IsUndefined()) return Heap::false_value();
v8::IndexedPropertyDeleter deleter =
v8::ToCData<v8::IndexedPropertyDeleter>(interceptor->deleter());
Handle<JSObject> this_handle(this);
LOG(ApiIndexedPropertyAccess("interceptor-indexed-delete", this, index));
CustomArguments args(interceptor->data(), this, this);
v8::AccessorInfo info(args.end());
v8::Handle<v8::Boolean> result;
{
// Leaving JavaScript.
VMState state(EXTERNAL);
result = deleter(index, info);
}
RETURN_IF_SCHEDULED_EXCEPTION();
if (!result.IsEmpty()) {
ASSERT(result->IsBoolean());
return *v8::Utils::OpenHandle(*result);
}
Object* raw_result =
this_handle->DeleteElementPostInterceptor(index, NORMAL_DELETION);
RETURN_IF_SCHEDULED_EXCEPTION();
return raw_result;
}
Object* JSObject::DeleteElement(uint32_t index, DeleteMode mode) {
// Check access rights if needed.
if (IsAccessCheckNeeded() &&
!Top::MayIndexedAccess(this, index, v8::ACCESS_DELETE)) {
Top::ReportFailedAccessCheck(this, v8::ACCESS_DELETE);
return Heap::false_value();
}
if (IsJSGlobalProxy()) {
Object* proto = GetPrototype();
if (proto->IsNull()) return Heap::false_value();
ASSERT(proto->IsJSGlobalObject());
return JSGlobalObject::cast(proto)->DeleteElement(index, mode);
}
if (HasIndexedInterceptor()) {
// Skip interceptor if forcing deletion.
if (mode == FORCE_DELETION) {
return DeleteElementPostInterceptor(index, mode);
}
return DeleteElementWithInterceptor(index);
}
switch (GetElementsKind()) {
case FAST_ELEMENTS: {
uint32_t length = IsJSArray() ?
static_cast<uint32_t>(Smi::cast(JSArray::cast(this)->length())->value()) :
static_cast<uint32_t>(FixedArray::cast(elements())->length());
if (index < length) {
FixedArray::cast(elements())->set_the_hole(index);
}
break;
}
case PIXEL_ELEMENTS: {
// Pixel elements cannot be deleted. Just silently ignore here.
break;
}
case DICTIONARY_ELEMENTS: {
NumberDictionary* dictionary = element_dictionary();
int entry = dictionary->FindEntry(index);
if (entry != NumberDictionary::kNotFound) {
return dictionary->DeleteProperty(entry, mode);
}
break;
}
default:
UNREACHABLE();
break;
}
return Heap::true_value();
}
Object* JSObject::DeleteProperty(String* name, DeleteMode mode) {
// ECMA-262, 3rd, 8.6.2.5
ASSERT(name->IsString());
// Check access rights if needed.
if (IsAccessCheckNeeded() &&
!Top::MayNamedAccess(this, name, v8::ACCESS_DELETE)) {
Top::ReportFailedAccessCheck(this, v8::ACCESS_DELETE);
return Heap::false_value();
}
if (IsJSGlobalProxy()) {
Object* proto = GetPrototype();
if (proto->IsNull()) return Heap::false_value();
ASSERT(proto->IsJSGlobalObject());
return JSGlobalObject::cast(proto)->DeleteProperty(name, mode);
}
uint32_t index = 0;
if (name->AsArrayIndex(&index)) {
return DeleteElement(index, mode);
} else {
LookupResult result;
LocalLookup(name, &result);
if (!result.IsValid()) return Heap::true_value();
// Ignore attributes if forcing a deletion.
if (result.IsDontDelete() && mode != FORCE_DELETION) {
return Heap::false_value();
}
// Check for interceptor.
if (result.type() == INTERCEPTOR) {
// Skip interceptor if forcing a deletion.
if (mode == FORCE_DELETION) {
return DeletePropertyPostInterceptor(name, mode);
}
return DeletePropertyWithInterceptor(name);
}
if (!result.IsLoaded()) {
return JSObject::cast(this)->DeleteLazyProperty(&result,
name,
mode);
}
// Normalize object if needed.
Object* obj = NormalizeProperties(CLEAR_INOBJECT_PROPERTIES, 0);
if (obj->IsFailure()) return obj;
// Make sure the properties are normalized before removing the entry.
return DeleteNormalizedProperty(name, mode);
}
}
// Check whether this object references another object.
bool JSObject::ReferencesObject(Object* obj) {
AssertNoAllocation no_alloc;
// Is the object the constructor for this object?
if (map()->constructor() == obj) {
return true;
}
// Is the object the prototype for this object?
if (map()->prototype() == obj) {
return true;
}
// Check if the object is among the named properties.
Object* key = SlowReverseLookup(obj);
if (key != Heap::undefined_value()) {
return true;
}
// Check if the object is among the indexed properties.
switch (GetElementsKind()) {
case PIXEL_ELEMENTS:
// Raw pixels do not reference other objects.
break;
case FAST_ELEMENTS: {
int length = IsJSArray() ?
Smi::cast(JSArray::cast(this)->length())->value() :
FixedArray::cast(elements())->length();
for (int i = 0; i < length; i++) {
Object* element = FixedArray::cast(elements())->get(i);
if (!element->IsTheHole() && element == obj) {
return true;
}
}
break;
}
case DICTIONARY_ELEMENTS: {
key = element_dictionary()->SlowReverseLookup(obj);
if (key != Heap::undefined_value()) {
return true;
}
break;
}
default:
UNREACHABLE();
break;
}
// For functions check the context. Boilerplate functions do
// not have to be traversed since they have no real context.
if (IsJSFunction() && !JSFunction::cast(this)->IsBoilerplate()) {
// Get the constructor function for arguments array.
JSObject* arguments_boilerplate =
Top::context()->global_context()->arguments_boilerplate();
JSFunction* arguments_function =
JSFunction::cast(arguments_boilerplate->map()->constructor());
// Get the context and don't check if it is the global context.
JSFunction* f = JSFunction::cast(this);
Context* context = f->context();
if (context->IsGlobalContext()) {
return false;
}
// Check the non-special context slots.
for (int i = Context::MIN_CONTEXT_SLOTS; i < context->length(); i++) {
// Only check JS objects.
if (context->get(i)->IsJSObject()) {
JSObject* ctxobj = JSObject::cast(context->get(i));
// If it is an arguments array check the content.
if (ctxobj->map()->constructor() == arguments_function) {
if (ctxobj->ReferencesObject(obj)) {
return true;
}
} else if (ctxobj == obj) {
return true;
}
}
}
// Check the context extension if any.
if (context->has_extension()) {
return context->extension()->ReferencesObject(obj);
}
}
// No references to object.
return false;
}
// Tests for the fast common case for property enumeration:
// - this object has an enum cache
// - this object has no elements
// - no prototype has enumerable properties/elements
// - neither this object nor any prototype has interceptors
bool JSObject::IsSimpleEnum() {
JSObject* arguments_boilerplate =
Top::context()->global_context()->arguments_boilerplate();
JSFunction* arguments_function =
JSFunction::cast(arguments_boilerplate->map()->constructor());
if (IsAccessCheckNeeded()) return false;
if (map()->constructor() == arguments_function) return false;
for (Object* o = this;
o != Heap::null_value();
o = JSObject::cast(o)->GetPrototype()) {
JSObject* curr = JSObject::cast(o);
if (!curr->HasFastProperties()) return false;
if (!curr->map()->instance_descriptors()->HasEnumCache()) return false;
if (curr->NumberOfEnumElements() > 0) return false;
if (curr->HasNamedInterceptor()) return false;
if (curr->HasIndexedInterceptor()) return false;
if (curr != this) {
FixedArray* curr_fixed_array =
FixedArray::cast(curr->map()->instance_descriptors()->GetEnumCache());
if (curr_fixed_array->length() > 0) {
return false;
}
}
}
return true;
}
int Map::NumberOfDescribedProperties() {
int result = 0;
DescriptorArray* descs = instance_descriptors();
for (int i = 0; i < descs->number_of_descriptors(); i++) {
if (descs->IsProperty(i)) result++;
}
return result;
}
int Map::PropertyIndexFor(String* name) {
DescriptorArray* descs = instance_descriptors();
for (int i = 0; i < descs->number_of_descriptors(); i++) {
if (name->Equals(descs->GetKey(i)) && !descs->IsNullDescriptor(i)) {
return descs->GetFieldIndex(i);
}
}
return -1;
}
int Map::NextFreePropertyIndex() {
int max_index = -1;
DescriptorArray* descs = instance_descriptors();
for (int i = 0; i < descs->number_of_descriptors(); i++) {
if (descs->GetType(i) == FIELD) {
int current_index = descs->GetFieldIndex(i);
if (current_index > max_index) max_index = current_index;
}
}
return max_index + 1;
}
AccessorDescriptor* Map::FindAccessor(String* name) {
DescriptorArray* descs = instance_descriptors();
for (int i = 0; i < descs->number_of_descriptors(); i++) {
if (name->Equals(descs->GetKey(i)) && descs->GetType(i) == CALLBACKS) {
return descs->GetCallbacks(i);
}
}
return NULL;
}
void JSObject::LocalLookup(String* name, LookupResult* result) {
ASSERT(name->IsString());
if (IsJSGlobalProxy()) {
Object* proto = GetPrototype();
if (proto->IsNull()) return result->NotFound();
ASSERT(proto->IsJSGlobalObject());
return JSObject::cast(proto)->LocalLookup(name, result);
}
// Do not use inline caching if the object is a non-global object
// that requires access checks.
if (!IsJSGlobalProxy() && IsAccessCheckNeeded()) {
result->DisallowCaching();
}
// Check __proto__ before interceptor.
if (name->Equals(Heap::Proto_symbol()) && !IsJSContextExtensionObject()) {
result->ConstantResult(this);
return;
}
// Check for lookup interceptor except when bootstrapping.
if (HasNamedInterceptor() && !Bootstrapper::IsActive()) {
result->InterceptorResult(this);
return;
}
LocalLookupRealNamedProperty(name, result);
}
void JSObject::Lookup(String* name, LookupResult* result) {
// Ecma-262 3rd 8.6.2.4
for (Object* current = this;
current != Heap::null_value();
current = JSObject::cast(current)->GetPrototype()) {
JSObject::cast(current)->LocalLookup(name, result);
if (result->IsValid() && !result->IsTransitionType()) return;
}
result->NotFound();
}
// Search object and it's prototype chain for callback properties.
void JSObject::LookupCallback(String* name, LookupResult* result) {
for (Object* current = this;
current != Heap::null_value();
current = JSObject::cast(current)->GetPrototype()) {
JSObject::cast(current)->LocalLookupRealNamedProperty(name, result);
if (result->IsValid() && result->type() == CALLBACKS) return;
}
result->NotFound();
}
Object* JSObject::DefineGetterSetter(String* name,
PropertyAttributes attributes) {
// Make sure that the top context does not change when doing callbacks or
// interceptor calls.
AssertNoContextChange ncc;
// Check access rights if needed.
if (IsAccessCheckNeeded() &&
!Top::MayNamedAccess(this, name, v8::ACCESS_SET)) {
Top::ReportFailedAccessCheck(this, v8::ACCESS_SET);
return Heap::undefined_value();
}
// Try to flatten before operating on the string.
name->TryFlattenIfNotFlat();
// Check if there is an API defined callback object which prohibits
// callback overwriting in this object or it's prototype chain.
// This mechanism is needed for instance in a browser setting, where
// certain accessors such as window.location should not be allowed
// to be overwritten because allowing overwriting could potentially
// cause security problems.
LookupResult callback_result;
LookupCallback(name, &callback_result);
if (callback_result.IsValid()) {
Object* obj = callback_result.GetCallbackObject();
if (obj->IsAccessorInfo() &&
AccessorInfo::cast(obj)->prohibits_overwriting()) {
return Heap::undefined_value();
}
}
uint32_t index;
bool is_element = name->AsArrayIndex(&index);
if (is_element && IsJSArray()) return Heap::undefined_value();
if (is_element) {
switch (GetElementsKind()) {
case FAST_ELEMENTS:
break;
case PIXEL_ELEMENTS:
// Ignore getters and setters on pixel elements.
return Heap::undefined_value();
case DICTIONARY_ELEMENTS: {
// Lookup the index.
NumberDictionary* dictionary = element_dictionary();
int entry = dictionary->FindEntry(index);
if (entry != NumberDictionary::kNotFound) {
Object* result = dictionary->ValueAt(entry);
PropertyDetails details = dictionary->DetailsAt(entry);
if (details.IsReadOnly()) return Heap::undefined_value();
if (details.type() == CALLBACKS) {
// Only accessors allowed as elements.
ASSERT(result->IsFixedArray());
return result;
}
}
break;
}
default:
UNREACHABLE();
break;
}
} else {
// Lookup the name.
LookupResult result;
LocalLookup(name, &result);
if (result.IsValid()) {
if (result.IsReadOnly()) return Heap::undefined_value();
if (result.type() == CALLBACKS) {
Object* obj = result.GetCallbackObject();
if (obj->IsFixedArray()) return obj;
}
}
}
// Allocate the fixed array to hold getter and setter.
Object* structure = Heap::AllocateFixedArray(2, TENURED);
if (structure->IsFailure()) return structure;
PropertyDetails details = PropertyDetails(attributes, CALLBACKS);
if (is_element) {
// Normalize object to make this operation simple.
Object* ok = NormalizeElements();
if (ok->IsFailure()) return ok;
// Update the dictionary with the new CALLBACKS property.
Object* dict =
element_dictionary()->Set(index, structure, details);
if (dict->IsFailure()) return dict;
// If name is an index we need to stay in slow case.
NumberDictionary* elements = NumberDictionary::cast(dict);
elements->set_requires_slow_elements();
// Set the potential new dictionary on the object.
set_elements(NumberDictionary::cast(dict));
} else {
// Normalize object to make this operation simple.
Object* ok = NormalizeProperties(CLEAR_INOBJECT_PROPERTIES, 0);
if (ok->IsFailure()) return ok;
// For the global object allocate a new map to invalidate the global inline
// caches which have a global property cell reference directly in the code.
if (IsGlobalObject()) {
Object* new_map = map()->CopyDropDescriptors();
if (new_map->IsFailure()) return new_map;
set_map(Map::cast(new_map));
}
// Update the dictionary with the new CALLBACKS property.
return SetNormalizedProperty(name, structure, details);
}
return structure;
}
Object* JSObject::DefineAccessor(String* name, bool is_getter, JSFunction* fun,
PropertyAttributes attributes) {
// Check access rights if needed.
if (IsAccessCheckNeeded() &&
!Top::MayNamedAccess(this, name, v8::ACCESS_HAS)) {
Top::ReportFailedAccessCheck(this, v8::ACCESS_HAS);
return Heap::undefined_value();
}
if (IsJSGlobalProxy()) {
Object* proto = GetPrototype();
if (proto->IsNull()) return this;
ASSERT(proto->IsJSGlobalObject());
return JSObject::cast(proto)->DefineAccessor(name, is_getter,
fun, attributes);
}
Object* array = DefineGetterSetter(name, attributes);
if (array->IsFailure() || array->IsUndefined()) return array;
FixedArray::cast(array)->set(is_getter ? 0 : 1, fun);
return this;
}
Object* JSObject::LookupAccessor(String* name, bool is_getter) {
// Make sure that the top context does not change when doing callbacks or
// interceptor calls.
AssertNoContextChange ncc;
// Check access rights if needed.
if (IsAccessCheckNeeded() &&
!Top::MayNamedAccess(this, name, v8::ACCESS_HAS)) {
Top::ReportFailedAccessCheck(this, v8::ACCESS_HAS);
return Heap::undefined_value();
}
// Make the lookup and include prototypes.
int accessor_index = is_getter ? kGetterIndex : kSetterIndex;
uint32_t index;
if (name->AsArrayIndex(&index)) {
for (Object* obj = this;
obj != Heap::null_value();
obj = JSObject::cast(obj)->GetPrototype()) {
JSObject* js_object = JSObject::cast(obj);
if (js_object->HasDictionaryElements()) {
NumberDictionary* dictionary = js_object->element_dictionary();
int entry = dictionary->FindEntry(index);
if (entry != NumberDictionary::kNotFound) {
Object* element = dictionary->ValueAt(entry);
PropertyDetails details = dictionary->DetailsAt(entry);
if (details.type() == CALLBACKS) {
// Only accessors allowed as elements.
return FixedArray::cast(element)->get(accessor_index);
}
}
}
}
} else {
for (Object* obj = this;
obj != Heap::null_value();
obj = JSObject::cast(obj)->GetPrototype()) {
LookupResult result;
JSObject::cast(obj)->LocalLookup(name, &result);
if (result.IsValid()) {
if (result.IsReadOnly()) return Heap::undefined_value();
if (result.type() == CALLBACKS) {
Object* obj = result.GetCallbackObject();
if (obj->IsFixedArray()) {
return FixedArray::cast(obj)->get(accessor_index);
}
}
}
}
}
return Heap::undefined_value();
}
Object* JSObject::SlowReverseLookup(Object* value) {
if (HasFastProperties()) {
DescriptorArray* descs = map()->instance_descriptors();
for (int i = 0; i < descs->number_of_descriptors(); i++) {
if (descs->GetType(i) == FIELD) {
if (FastPropertyAt(descs->GetFieldIndex(i)) == value) {
return descs->GetKey(i);
}
} else if (descs->GetType(i) == CONSTANT_FUNCTION) {
if (descs->GetConstantFunction(i) == value) {
return descs->GetKey(i);
}
}
}
return Heap::undefined_value();
} else {
return property_dictionary()->SlowReverseLookup(value);
}
}
Object* Map::CopyDropDescriptors() {
Object* result = Heap::AllocateMap(instance_type(), instance_size());
if (result->IsFailure()) return result;
Map::cast(result)->set_prototype(prototype());
Map::cast(result)->set_constructor(constructor());
// Don't copy descriptors, so map transitions always remain a forest.
// If we retained the same descriptors we would have two maps
// pointing to the same transition which is bad because the garbage
// collector relies on being able to reverse pointers from transitions
// to maps. If properties need to be retained use CopyDropTransitions.
Map::cast(result)->set_instance_descriptors(Heap::empty_descriptor_array());
// Please note instance_type and instance_size are set when allocated.
Map::cast(result)->set_inobject_properties(inobject_properties());
Map::cast(result)->set_unused_property_fields(unused_property_fields());
// If the map has pre-allocated properties always start out with a descriptor
// array describing these properties.
if (pre_allocated_property_fields() > 0) {
ASSERT(constructor()->IsJSFunction());
JSFunction* ctor = JSFunction::cast(constructor());
Object* descriptors =
ctor->initial_map()->instance_descriptors()->RemoveTransitions();
if (descriptors->IsFailure()) return descriptors;
Map::cast(result)->set_instance_descriptors(
DescriptorArray::cast(descriptors));
Map::cast(result)->set_pre_allocated_property_fields(
pre_allocated_property_fields());
}
Map::cast(result)->set_bit_field(bit_field());
Map::cast(result)->set_bit_field2(bit_field2());
Map::cast(result)->ClearCodeCache();
return result;
}
Object* Map::CopyDropTransitions() {
Object* new_map = CopyDropDescriptors();
if (new_map->IsFailure()) return new_map;
Object* descriptors = instance_descriptors()->RemoveTransitions();
if (descriptors->IsFailure()) return descriptors;
cast(new_map)->set_instance_descriptors(DescriptorArray::cast(descriptors));
return cast(new_map);
}
Object* Map::UpdateCodeCache(String* name, Code* code) {
ASSERT(code->ic_state() == MONOMORPHIC);
FixedArray* cache = code_cache();
// When updating the code cache we disregard the type encoded in the
// flags. This allows call constant stubs to overwrite call field
// stubs, etc.
Code::Flags flags = Code::RemoveTypeFromFlags(code->flags());
// First check whether we can update existing code cache without
// extending it.
int length = cache->length();
int deleted_index = -1;
for (int i = 0; i < length; i += 2) {
Object* key = cache->get(i);
if (key->IsNull()) {
if (deleted_index < 0) deleted_index = i;
continue;
}
if (key->IsUndefined()) {
if (deleted_index >= 0) i = deleted_index;
cache->set(i + 0, name);
cache->set(i + 1, code);
return this;
}
if (name->Equals(String::cast(key))) {
Code::Flags found = Code::cast(cache->get(i + 1))->flags();
if (Code::RemoveTypeFromFlags(found) == flags) {
cache->set(i + 1, code);
return this;
}
}
}
// Reached the end of the code cache. If there were deleted
// elements, reuse the space for the first of them.
if (deleted_index >= 0) {
cache->set(deleted_index + 0, name);
cache->set(deleted_index + 1, code);
return this;
}
// Extend the code cache with some new entries (at least one).
int new_length = length + ((length >> 1) & ~1) + 2;
ASSERT((new_length & 1) == 0); // must be a multiple of two
Object* result = cache->CopySize(new_length);
if (result->IsFailure()) return result;
// Add the (name, code) pair to the new cache.
cache = FixedArray::cast(result);
cache->set(length + 0, name);
cache->set(length + 1, code);
set_code_cache(cache);
return this;
}
Object* Map::FindInCodeCache(String* name, Code::Flags flags) {
FixedArray* cache = code_cache();
int length = cache->length();
for (int i = 0; i < length; i += 2) {
Object* key = cache->get(i);
// Skip deleted elements.
if (key->IsNull()) continue;
if (key->IsUndefined()) return key;
if (name->Equals(String::cast(key))) {
Code* code = Code::cast(cache->get(i + 1));
if (code->flags() == flags) return code;
}
}
return Heap::undefined_value();
}
int Map::IndexInCodeCache(Code* code) {
FixedArray* array = code_cache();
int len = array->length();
for (int i = 0; i < len; i += 2) {
if (array->get(i + 1) == code) return i + 1;
}
return -1;
}
void Map::RemoveFromCodeCache(int index) {
FixedArray* array = code_cache();
ASSERT(array->length() >= index && array->get(index)->IsCode());
// Use null instead of undefined for deleted elements to distinguish
// deleted elements from unused elements. This distinction is used
// when looking up in the cache and when updating the cache.
array->set_null(index - 1); // key
array->set_null(index); // code
}
void FixedArray::FixedArrayIterateBody(ObjectVisitor* v) {
IteratePointers(v, kHeaderSize, kHeaderSize + length() * kPointerSize);
}
static bool HasKey(FixedArray* array, Object* key) {
int len0 = array->length();
for (int i = 0; i < len0; i++) {
Object* element = array->get(i);
if (element->IsSmi() && key->IsSmi() && (element == key)) return true;
if (element->IsString() &&
key->IsString() && String::cast(element)->Equals(String::cast(key))) {
return true;
}
}
return false;
}
Object* FixedArray::AddKeysFromJSArray(JSArray* array) {
ASSERT(!array->HasPixelElements());
switch (array->GetElementsKind()) {
case JSObject::FAST_ELEMENTS:
return UnionOfKeys(FixedArray::cast(array->elements()));
case JSObject::DICTIONARY_ELEMENTS: {
NumberDictionary* dict = array->element_dictionary();
int size = dict->NumberOfElements();
// Allocate a temporary fixed array.
Object* object = Heap::AllocateFixedArray(size);
if (object->IsFailure()) return object;
FixedArray* key_array = FixedArray::cast(object);
int capacity = dict->Capacity();
int pos = 0;
// Copy the elements from the JSArray to the temporary fixed array.
for (int i = 0; i < capacity; i++) {
if (dict->IsKey(dict->KeyAt(i))) {
key_array->set(pos++, dict->ValueAt(i));
}
}
// Compute the union of this and the temporary fixed array.
return UnionOfKeys(key_array);
}
default:
UNREACHABLE();
}
UNREACHABLE();
return Heap::null_value(); // Failure case needs to "return" a value.
}
Object* FixedArray::UnionOfKeys(FixedArray* other) {
int len0 = length();
int len1 = other->length();
// Optimize if either is empty.
if (len0 == 0) return other;
if (len1 == 0) return this;
// Compute how many elements are not in this.
int extra = 0;
for (int y = 0; y < len1; y++) {
Object* value = other->get(y);
if (!value->IsTheHole() && !HasKey(this, value)) extra++;
}
if (extra == 0) return this;
// Allocate the result
Object* obj = Heap::AllocateFixedArray(len0 + extra);
if (obj->IsFailure()) return obj;
// Fill in the content
FixedArray* result = FixedArray::cast(obj);
WriteBarrierMode mode = result->GetWriteBarrierMode();
for (int i = 0; i < len0; i++) {
result->set(i, get(i), mode);
}
// Fill in the extra keys.
int index = 0;
for (int y = 0; y < len1; y++) {
Object* value = other->get(y);
if (!value->IsTheHole() && !HasKey(this, value)) {
result->set(len0 + index, other->get(y), mode);
index++;
}
}
ASSERT(extra == index);
return result;
}
Object* FixedArray::CopySize(int new_length) {
if (new_length == 0) return Heap::empty_fixed_array();
Object* obj = Heap::AllocateFixedArray(new_length);
if (obj->IsFailure()) return obj;
FixedArray* result = FixedArray::cast(obj);
// Copy the content
int len = length();
if (new_length < len) len = new_length;
result->set_map(map());
WriteBarrierMode mode = result->GetWriteBarrierMode();
for (int i = 0; i < len; i++) {
result->set(i, get(i), mode);
}
return result;
}
void FixedArray::CopyTo(int pos, FixedArray* dest, int dest_pos, int len) {
WriteBarrierMode mode = dest->GetWriteBarrierMode();
for (int index = 0; index < len; index++) {
dest->set(dest_pos+index, get(pos+index), mode);
}
}
#ifdef DEBUG
bool FixedArray::IsEqualTo(FixedArray* other) {
if (length() != other->length()) return false;
for (int i = 0 ; i < length(); ++i) {
if (get(i) != other->get(i)) return false;
}
return true;
}
#endif
Object* DescriptorArray::Allocate(int number_of_descriptors) {
if (number_of_descriptors == 0) {
return Heap::empty_descriptor_array();
}
// Allocate the array of keys.
Object* array = Heap::AllocateFixedArray(ToKeyIndex(number_of_descriptors));
if (array->IsFailure()) return array;
// Do not use DescriptorArray::cast on incomplete object.
FixedArray* result = FixedArray::cast(array);
// Allocate the content array and set it in the descriptor array.
array = Heap::AllocateFixedArray(number_of_descriptors << 1);
if (array->IsFailure()) return array;
result->set(kContentArrayIndex, array);
result->set(kEnumerationIndexIndex,
Smi::FromInt(PropertyDetails::kInitialIndex),
SKIP_WRITE_BARRIER);
return result;
}
void DescriptorArray::SetEnumCache(FixedArray* bridge_storage,
FixedArray* new_cache) {
ASSERT(bridge_storage->length() >= kEnumCacheBridgeLength);
if (HasEnumCache()) {
FixedArray::cast(get(kEnumerationIndexIndex))->
set(kEnumCacheBridgeCacheIndex, new_cache);
} else {
if (IsEmpty()) return; // Do nothing for empty descriptor array.
FixedArray::cast(bridge_storage)->
set(kEnumCacheBridgeCacheIndex, new_cache);
fast_set(FixedArray::cast(bridge_storage),
kEnumCacheBridgeEnumIndex,
get(kEnumerationIndexIndex));
set(kEnumerationIndexIndex, bridge_storage);
}
}
Object* DescriptorArray::CopyInsert(Descriptor* descriptor,
TransitionFlag transition_flag) {
// Transitions are only kept when inserting another transition.
// This precondition is not required by this function's implementation, but
// is currently required by the semantics of maps, so we check it.
// Conversely, we filter after replacing, so replacing a transition and
// removing all other transitions is not supported.
bool remove_transitions = transition_flag == REMOVE_TRANSITIONS;
ASSERT(remove_transitions == !descriptor->GetDetails().IsTransition());
ASSERT(descriptor->GetDetails().type() != NULL_DESCRIPTOR);
// Ensure the key is a symbol.
Object* result = descriptor->KeyToSymbol();
if (result->IsFailure()) return result;
int transitions = 0;
int null_descriptors = 0;
if (remove_transitions) {
for (int i = 0; i < number_of_descriptors(); i++) {
if (IsTransition(i)) transitions++;
if (IsNullDescriptor(i)) null_descriptors++;
}
} else {
for (int i = 0; i < number_of_descriptors(); i++) {
if (IsNullDescriptor(i)) null_descriptors++;
}
}
int new_size = number_of_descriptors() - transitions - null_descriptors;
// If key is in descriptor, we replace it in-place when filtering.
// Count a null descriptor for key as inserted, not replaced.
int index = Search(descriptor->GetKey());
const bool inserting = (index == kNotFound);
const bool replacing = !inserting;
bool keep_enumeration_index = false;
if (inserting) {
++new_size;
}
if (replacing) {
// We are replacing an existing descriptor. We keep the enumeration
// index of a visible property.
PropertyType t = PropertyDetails(GetDetails(index)).type();
if (t == CONSTANT_FUNCTION ||
t == FIELD ||
t == CALLBACKS ||
t == INTERCEPTOR) {
keep_enumeration_index = true;
} else if (remove_transitions) {
// Replaced descriptor has been counted as removed if it is
// a transition that will be replaced. Adjust count in this case.
++new_size;
}
}
result = Allocate(new_size);
if (result->IsFailure()) return result;
DescriptorArray* new_descriptors = DescriptorArray::cast(result);
// Set the enumeration index in the descriptors and set the enumeration index
// in the result.
int enumeration_index = NextEnumerationIndex();
if (!descriptor->GetDetails().IsTransition()) {
if (keep_enumeration_index) {
descriptor->SetEnumerationIndex(
PropertyDetails(GetDetails(index)).index());
} else {
descriptor->SetEnumerationIndex(enumeration_index);
++enumeration_index;
}
}
new_descriptors->SetNextEnumerationIndex(enumeration_index);
// Copy the descriptors, filtering out transitions and null descriptors,
// and inserting or replacing a descriptor.
uint32_t descriptor_hash = descriptor->GetKey()->Hash();
int from_index = 0;
int to_index = 0;
for (; from_index < number_of_descriptors(); from_index++) {
String* key = GetKey(from_index);
if (key->Hash() > descriptor_hash || key == descriptor->GetKey()) {
break;
}
if (IsNullDescriptor(from_index)) continue;
if (remove_transitions && IsTransition(from_index)) continue;
new_descriptors->CopyFrom(to_index++, this, from_index);
}
new_descriptors->Set(to_index++, descriptor);
if (replacing) from_index++;
for (; from_index < number_of_descriptors(); from_index++) {
if (IsNullDescriptor(from_index)) continue;
if (remove_transitions && IsTransition(from_index)) continue;
new_descriptors->CopyFrom(to_index++, this, from_index);
}
ASSERT(to_index == new_descriptors->number_of_descriptors());
SLOW_ASSERT(new_descriptors->IsSortedNoDuplicates());
return new_descriptors;
}
Object* DescriptorArray::RemoveTransitions() {
// Remove all transitions and null descriptors. Return a copy of the array
// with all transitions removed, or a Failure object if the new array could
// not be allocated.
// Compute the size of the map transition entries to be removed.
int num_removed = 0;
for (int i = 0; i < number_of_descriptors(); i++) {
if (!IsProperty(i)) num_removed++;
}
// Allocate the new descriptor array.
Object* result = Allocate(number_of_descriptors() - num_removed);
if (result->IsFailure()) return result;
DescriptorArray* new_descriptors = DescriptorArray::cast(result);
// Copy the content.
int next_descriptor = 0;
for (int i = 0; i < number_of_descriptors(); i++) {
if (IsProperty(i)) new_descriptors->CopyFrom(next_descriptor++, this, i);
}
ASSERT(next_descriptor == new_descriptors->number_of_descriptors());
return new_descriptors;
}
void DescriptorArray::Sort() {
// In-place heap sort.
int len = number_of_descriptors();
// Bottom-up max-heap construction.
for (int i = 1; i < len; ++i) {
int child_index = i;
while (child_index > 0) {
int parent_index = ((child_index + 1) >> 1) - 1;
uint32_t parent_hash = GetKey(parent_index)->Hash();
uint32_t child_hash = GetKey(child_index)->Hash();
if (parent_hash < child_hash) {
Swap(parent_index, child_index);
} else {
break;
}
child_index = parent_index;
}
}
// Extract elements and create sorted array.
for (int i = len - 1; i > 0; --i) {
// Put max element at the back of the array.
Swap(0, i);
// Sift down the new top element.
int parent_index = 0;
while (true) {
int child_index = ((parent_index + 1) << 1) - 1;
if (child_index >= i) break;
uint32_t child1_hash = GetKey(child_index)->Hash();
uint32_t child2_hash = GetKey(child_index + 1)->Hash();
uint32_t parent_hash = GetKey(parent_index)->Hash();
if (child_index + 1 >= i || child1_hash > child2_hash) {
if (parent_hash > child1_hash) break;
Swap(parent_index, child_index);
parent_index = child_index;
} else {
if (parent_hash > child2_hash) break;
Swap(parent_index, child_index + 1);
parent_index = child_index + 1;
}
}
}
SLOW_ASSERT(IsSortedNoDuplicates());
}
int DescriptorArray::BinarySearch(String* name, int low, int high) {
uint32_t hash = name->Hash();
while (low <= high) {
int mid = (low + high) / 2;
String* mid_name = GetKey(mid);
uint32_t mid_hash = mid_name->Hash();
if (mid_hash > hash) {
high = mid - 1;
continue;
}
if (mid_hash < hash) {
low = mid + 1;
continue;
}
// Found an element with the same hash-code.
ASSERT(hash == mid_hash);
// There might be more, so we find the first one and
// check them all to see if we have a match.
if (name == mid_name && !is_null_descriptor(mid)) return mid;
while ((mid > low) && (GetKey(mid - 1)->Hash() == hash)) mid--;
for (; (mid <= high) && (GetKey(mid)->Hash() == hash); mid++) {
if (GetKey(mid)->Equals(name) && !is_null_descriptor(mid)) return mid;
}
break;
}
return kNotFound;
}
int DescriptorArray::LinearSearch(String* name, int len) {
uint32_t hash = name->Hash();
for (int number = 0; number < len; number++) {
String* entry = GetKey(number);
if ((entry->Hash() == hash) &&
name->Equals(entry) &&
!is_null_descriptor(number)) {
return number;
}
}
return kNotFound;
}
#ifdef DEBUG
bool DescriptorArray::IsEqualTo(DescriptorArray* other) {
if (IsEmpty()) return other->IsEmpty();
if (other->IsEmpty()) return false;
if (length() != other->length()) return false;
for (int i = 0; i < length(); ++i) {
if (get(i) != other->get(i) && i != kContentArrayIndex) return false;
}
return GetContentArray()->IsEqualTo(other->GetContentArray());
}
#endif
static StaticResource<StringInputBuffer> string_input_buffer;
bool String::LooksValid() {
if (!Heap::Contains(this)) return false;
return true;
}
int String::Utf8Length() {
if (IsAsciiRepresentation()) return length();
// Attempt to flatten before accessing the string. It probably
// doesn't make Utf8Length faster, but it is very likely that
// the string will be accessed later (for example by WriteUtf8)
// so it's still a good idea.
TryFlattenIfNotFlat();
Access<StringInputBuffer> buffer(&string_input_buffer);
buffer->Reset(0, this);
int result = 0;
while (buffer->has_more())
result += unibrow::Utf8::Length(buffer->GetNext());
return result;
}
Vector<const char> String::ToAsciiVector() {
ASSERT(IsAsciiRepresentation());
ASSERT(IsFlat());
int offset = 0;
int length = this->length();
StringRepresentationTag string_tag = StringShape(this).representation_tag();
String* string = this;
if (string_tag == kSlicedStringTag) {
SlicedString* sliced = SlicedString::cast(string);
offset += sliced->start();
string = sliced->buffer();
string_tag = StringShape(string).representation_tag();
} else if (string_tag == kConsStringTag) {
ConsString* cons = ConsString::cast(string);
ASSERT(cons->second()->length() == 0);
string = cons->first();
string_tag = StringShape(string).representation_tag();
}
if (string_tag == kSeqStringTag) {
SeqAsciiString* seq = SeqAsciiString::cast(string);
char* start = seq->GetChars();
return Vector<const char>(start + offset, length);
}
ASSERT(string_tag == kExternalStringTag);
ExternalAsciiString* ext = ExternalAsciiString::cast(string);
const char* start = ext->resource()->data();
return Vector<const char>(start + offset, length);
}
Vector<const uc16> String::ToUC16Vector() {
ASSERT(IsTwoByteRepresentation());
ASSERT(IsFlat());
int offset = 0;
int length = this->length();
StringRepresentationTag string_tag = StringShape(this).representation_tag();
String* string = this;
if (string_tag == kSlicedStringTag) {
SlicedString* sliced = SlicedString::cast(string);
offset += sliced->start();
string = String::cast(sliced->buffer());
string_tag = StringShape(string).representation_tag();
} else if (string_tag == kConsStringTag) {
ConsString* cons = ConsString::cast(string);
ASSERT(cons->second()->length() == 0);
string = cons->first();
string_tag = StringShape(string).representation_tag();
}
if (string_tag == kSeqStringTag) {
SeqTwoByteString* seq = SeqTwoByteString::cast(string);
return Vector<const uc16>(seq->GetChars() + offset, length);
}
ASSERT(string_tag == kExternalStringTag);
ExternalTwoByteString* ext = ExternalTwoByteString::cast(string);
const uc16* start =
reinterpret_cast<const uc16*>(ext->resource()->data());
return Vector<const uc16>(start + offset, length);
}
SmartPointer<char> String::ToCString(AllowNullsFlag allow_nulls,
RobustnessFlag robust_flag,
int offset,
int length,
int* length_return) {
ASSERT(NativeAllocationChecker::allocation_allowed());
if (robust_flag == ROBUST_STRING_TRAVERSAL && !LooksValid()) {
return SmartPointer<char>(NULL);
}
// Negative length means the to the end of the string.
if (length < 0) length = kMaxInt - offset;
// Compute the size of the UTF-8 string. Start at the specified offset.
Access<StringInputBuffer> buffer(&string_input_buffer);
buffer->Reset(offset, this);
int character_position = offset;
int utf8_bytes = 0;
while (buffer->has_more()) {
uint16_t character = buffer->GetNext();
if (character_position < offset + length) {
utf8_bytes += unibrow::Utf8::Length(character);
}
character_position++;
}
if (length_return) {
*length_return = utf8_bytes;
}
char* result = NewArray<char>(utf8_bytes + 1);
// Convert the UTF-16 string to a UTF-8 buffer. Start at the specified offset.
buffer->Rewind();
buffer->Seek(offset);
character_position = offset;
int utf8_byte_position = 0;
while (buffer->has_more()) {
uint16_t character = buffer->GetNext();
if (character_position < offset + length) {
if (allow_nulls == DISALLOW_NULLS && character == 0) {
character = ' ';
}
utf8_byte_position +=
unibrow::Utf8::Encode(result + utf8_byte_position, character);
}
character_position++;
}
result[utf8_byte_position] = 0;
return SmartPointer<char>(result);
}
SmartPointer<char> String::ToCString(AllowNullsFlag allow_nulls,
RobustnessFlag robust_flag,
int* length_return) {
return ToCString(allow_nulls, robust_flag, 0, -1, length_return);
}
const uc16* String::GetTwoByteData() {
return GetTwoByteData(0);
}
const uc16* String::GetTwoByteData(unsigned start) {
ASSERT(!IsAsciiRepresentation());
switch (StringShape(this).representation_tag()) {
case kSeqStringTag:
return SeqTwoByteString::cast(this)->SeqTwoByteStringGetData(start);
case kExternalStringTag:
return ExternalTwoByteString::cast(this)->
ExternalTwoByteStringGetData(start);
case kSlicedStringTag: {
SlicedString* sliced_string = SlicedString::cast(this);
String* buffer = sliced_string->buffer();
if (StringShape(buffer).IsCons()) {
ConsString* cs = ConsString::cast(buffer);
// Flattened string.
ASSERT(cs->second()->length() == 0);
buffer = cs->first();
}
return buffer->GetTwoByteData(start + sliced_string->start());
}
case kConsStringTag:
UNREACHABLE();
return NULL;
}
UNREACHABLE();
return NULL;
}
SmartPointer<uc16> String::ToWideCString(RobustnessFlag robust_flag) {
ASSERT(NativeAllocationChecker::allocation_allowed());
if (robust_flag == ROBUST_STRING_TRAVERSAL && !LooksValid()) {
return SmartPointer<uc16>();
}
Access<StringInputBuffer> buffer(&string_input_buffer);
buffer->Reset(this);
uc16* result = NewArray<uc16>(length() + 1);
int i = 0;
while (buffer->has_more()) {
uint16_t character = buffer->GetNext();
result[i++] = character;
}
result[i] = 0;
return SmartPointer<uc16>(result);
}
const uc16* SeqTwoByteString::SeqTwoByteStringGetData(unsigned start) {
return reinterpret_cast<uc16*>(
reinterpret_cast<char*>(this) - kHeapObjectTag + kHeaderSize) + start;
}
void SeqTwoByteString::SeqTwoByteStringReadBlockIntoBuffer(ReadBlockBuffer* rbb,
unsigned* offset_ptr,
unsigned max_chars) {
unsigned chars_read = 0;
unsigned offset = *offset_ptr;
while (chars_read < max_chars) {
uint16_t c = *reinterpret_cast<uint16_t*>(
reinterpret_cast<char*>(this) -
kHeapObjectTag + kHeaderSize + offset * kShortSize);
if (c <= kMaxAsciiCharCode) {
// Fast case for ASCII characters. Cursor is an input output argument.
if (!unibrow::CharacterStream::EncodeAsciiCharacter(c,
rbb->util_buffer,
rbb->capacity,
rbb->cursor)) {
break;
}
} else {
if (!unibrow::CharacterStream::EncodeNonAsciiCharacter(c,
rbb->util_buffer,
rbb->capacity,
rbb->cursor)) {
break;
}
}
offset++;
chars_read++;
}
*offset_ptr = offset;
rbb->remaining += chars_read;
}
const unibrow::byte* SeqAsciiString::SeqAsciiStringReadBlock(
unsigned* remaining,
unsigned* offset_ptr,
unsigned max_chars) {
const unibrow::byte* b = reinterpret_cast<unibrow::byte*>(this) -
kHeapObjectTag + kHeaderSize + *offset_ptr * kCharSize;
*remaining = max_chars;
*offset_ptr += max_chars;
return b;
}
// This will iterate unless the block of string data spans two 'halves' of
// a ConsString, in which case it will recurse. Since the block of string
// data to be read has a maximum size this limits the maximum recursion
// depth to something sane. Since C++ does not have tail call recursion
// elimination, the iteration must be explicit. Since this is not an
// -IntoBuffer method it can delegate to one of the efficient
// *AsciiStringReadBlock routines.
const unibrow::byte* ConsString::ConsStringReadBlock(ReadBlockBuffer* rbb,
unsigned* offset_ptr,
unsigned max_chars) {
ConsString* current = this;
unsigned offset = *offset_ptr;
int offset_correction = 0;
while (true) {
String* left = current->first();
unsigned left_length = (unsigned)left->length();
if (left_length > offset &&
(max_chars <= left_length - offset ||
(rbb->capacity <= left_length - offset &&
(max_chars = left_length - offset, true)))) { // comma operator!
// Left hand side only - iterate unless we have reached the bottom of
// the cons tree. The assignment on the left of the comma operator is
// in order to make use of the fact that the -IntoBuffer routines can
// produce at most 'capacity' characters. This enables us to postpone
// the point where we switch to the -IntoBuffer routines (below) in order
// to maximize the chances of delegating a big chunk of work to the
// efficient *AsciiStringReadBlock routines.
if (StringShape(left).IsCons()) {
current = ConsString::cast(left);
continue;
} else {
const unibrow::byte* answer =
String::ReadBlock(left, rbb, &offset, max_chars);
*offset_ptr = offset + offset_correction;
return answer;
}
} else if (left_length <= offset) {
// Right hand side only - iterate unless we have reached the bottom of
// the cons tree.
String* right = current->second();
offset -= left_length;
offset_correction += left_length;
if (StringShape(right).IsCons()) {
current = ConsString::cast(right);
continue;
} else {
const unibrow::byte* answer =
String::ReadBlock(right, rbb, &offset, max_chars);
*offset_ptr = offset + offset_correction;
return answer;
}
} else {
// The block to be read spans two sides of the ConsString, so we call the
// -IntoBuffer version, which will recurse. The -IntoBuffer methods
// are able to assemble data from several part strings because they use
// the util_buffer to store their data and never return direct pointers
// to their storage. We don't try to read more than the buffer capacity
// here or we can get too much recursion.
ASSERT(rbb->remaining == 0);
ASSERT(rbb->cursor == 0);
current->ConsStringReadBlockIntoBuffer(
rbb,
&offset,
max_chars > rbb->capacity ? rbb->capacity : max_chars);
*offset_ptr = offset + offset_correction;
return rbb->util_buffer;
}
}
}
const unibrow::byte* SlicedString::SlicedStringReadBlock(ReadBlockBuffer* rbb,
unsigned* offset_ptr,
unsigned max_chars) {
String* backing = buffer();
unsigned offset = start() + *offset_ptr;
unsigned length = backing->length();
if (max_chars > length - offset) {
max_chars = length - offset;
}
const unibrow::byte* answer =
String::ReadBlock(backing, rbb, &offset, max_chars);
*offset_ptr = offset - start();
return answer;
}
uint16_t ExternalAsciiString::ExternalAsciiStringGet(int index) {
ASSERT(index >= 0 && index < length());
return resource()->data()[index];
}
const unibrow::byte* ExternalAsciiString::ExternalAsciiStringReadBlock(
unsigned* remaining,
unsigned* offset_ptr,
unsigned max_chars) {
// Cast const char* to unibrow::byte* (signedness difference).
const unibrow::byte* b =
reinterpret_cast<const unibrow::byte*>(resource()->data()) + *offset_ptr;
*remaining = max_chars;
*offset_ptr += max_chars;
return b;
}
const uc16* ExternalTwoByteString::ExternalTwoByteStringGetData(
unsigned start) {
return resource()->data() + start;
}
uint16_t ExternalTwoByteString::ExternalTwoByteStringGet(int index) {
ASSERT(index >= 0 && index < length());
return resource()->data()[index];
}
void ExternalTwoByteString::ExternalTwoByteStringReadBlockIntoBuffer(
ReadBlockBuffer* rbb,
unsigned* offset_ptr,
unsigned max_chars) {
unsigned chars_read = 0;
unsigned offset = *offset_ptr;
const uint16_t* data = resource()->data();
while (chars_read < max_chars) {
uint16_t c = data[offset];
if (c <= kMaxAsciiCharCode) {
// Fast case for ASCII characters. Cursor is an input output argument.
if (!unibrow::CharacterStream::EncodeAsciiCharacter(c,
rbb->util_buffer,
rbb->capacity,
rbb->cursor))
break;
} else {
if (!unibrow::CharacterStream::EncodeNonAsciiCharacter(c,
rbb->util_buffer,
rbb->capacity,
rbb->cursor))
break;
}
offset++;
chars_read++;
}
*offset_ptr = offset;
rbb->remaining += chars_read;
}
void SeqAsciiString::SeqAsciiStringReadBlockIntoBuffer(ReadBlockBuffer* rbb,
unsigned* offset_ptr,
unsigned max_chars) {
unsigned capacity = rbb->capacity - rbb->cursor;
if (max_chars > capacity) max_chars = capacity;
memcpy(rbb->util_buffer + rbb->cursor,
reinterpret_cast<char*>(this) - kHeapObjectTag + kHeaderSize +
*offset_ptr * kCharSize,
max_chars);
rbb->remaining += max_chars;
*offset_ptr += max_chars;
rbb->cursor += max_chars;
}
void ExternalAsciiString::ExternalAsciiStringReadBlockIntoBuffer(
ReadBlockBuffer* rbb,
unsigned* offset_ptr,
unsigned max_chars) {
unsigned capacity = rbb->capacity - rbb->cursor;
if (max_chars > capacity) max_chars = capacity;
memcpy(rbb->util_buffer + rbb->cursor,
resource()->data() + *offset_ptr,
max_chars);
rbb->remaining += max_chars;
*offset_ptr += max_chars;
rbb->cursor += max_chars;
}
// This method determines the type of string involved and then copies
// a whole chunk of characters into a buffer, or returns a pointer to a buffer
// where they can be found. The pointer is not necessarily valid across a GC
// (see AsciiStringReadBlock).
const unibrow::byte* String::ReadBlock(String* input,
ReadBlockBuffer* rbb,
unsigned* offset_ptr,
unsigned max_chars) {
ASSERT(*offset_ptr <= static_cast<unsigned>(input->length()));
if (max_chars == 0) {
rbb->remaining = 0;
return NULL;
}
switch (StringShape(input).representation_tag()) {
case kSeqStringTag:
if (input->IsAsciiRepresentation()) {
SeqAsciiString* str = SeqAsciiString::cast(input);
return str->SeqAsciiStringReadBlock(&rbb->remaining,
offset_ptr,
max_chars);
} else {
SeqTwoByteString* str = SeqTwoByteString::cast(input);
str->SeqTwoByteStringReadBlockIntoBuffer(rbb,
offset_ptr,
max_chars);
return rbb->util_buffer;
}
case kConsStringTag:
return ConsString::cast(input)->ConsStringReadBlock(rbb,
offset_ptr,
max_chars);
case kSlicedStringTag:
return SlicedString::cast(input)->SlicedStringReadBlock(rbb,
offset_ptr,
max_chars);
case kExternalStringTag:
if (input->IsAsciiRepresentation()) {
return ExternalAsciiString::cast(input)->ExternalAsciiStringReadBlock(
&rbb->remaining,
offset_ptr,
max_chars);
} else {
ExternalTwoByteString::cast(input)->
ExternalTwoByteStringReadBlockIntoBuffer(rbb,
offset_ptr,
max_chars);
return rbb->util_buffer;
}
default:
break;
}
UNREACHABLE();
return 0;
}
Relocatable* Relocatable::top_ = NULL;
void Relocatable::PostGarbageCollectionProcessing() {
Relocatable* current = top_;
while (current != NULL) {
current->PostGarbageCollection();
current = current->prev_;
}
}
// Reserve space for statics needing saving and restoring.
int Relocatable::ArchiveSpacePerThread() {
return sizeof(top_);
}
// Archive statics that are thread local.
char* Relocatable::ArchiveState(char* to) {
*reinterpret_cast<Relocatable**>(to) = top_;
top_ = NULL;
return to + ArchiveSpacePerThread();
}
// Restore statics that are thread local.
char* Relocatable::RestoreState(char* from) {
top_ = *reinterpret_cast<Relocatable**>(from);
return from + ArchiveSpacePerThread();
}
char* Relocatable::Iterate(ObjectVisitor* v, char* thread_storage) {
Relocatable* top = *reinterpret_cast<Relocatable**>(thread_storage);
Iterate(v, top);
return thread_storage + ArchiveSpacePerThread();
}
void Relocatable::Iterate(ObjectVisitor* v) {
Iterate(v, top_);
}
void Relocatable::Iterate(ObjectVisitor* v, Relocatable* top) {
Relocatable* current = top;
while (current != NULL) {
current->IterateInstance(v);
current = current->prev_;
}
}
FlatStringReader::FlatStringReader(Handle<String> str)
: str_(str.location()),
length_(str->length()) {
PostGarbageCollection();
}
FlatStringReader::FlatStringReader(Vector<const char> input)
: str_(0),
is_ascii_(true),
length_(input.length()),
start_(input.start()) { }
void FlatStringReader::PostGarbageCollection() {
if (str_ == NULL) return;
Handle<String> str(str_);
ASSERT(str->IsFlat());
is_ascii_ = str->IsAsciiRepresentation();
if (is_ascii_) {
start_ = str->ToAsciiVector().start();
} else {
start_ = str->ToUC16Vector().start();
}
}
void StringInputBuffer::Seek(unsigned pos) {
Reset(pos, input_);
}
void SafeStringInputBuffer::Seek(unsigned pos) {
Reset(pos, input_);
}
// This method determines the type of string involved and then copies
// a whole chunk of characters into a buffer. It can be used with strings
// that have been glued together to form a ConsString and which must cooperate
// to fill up a buffer.
void String::ReadBlockIntoBuffer(String* input,
ReadBlockBuffer* rbb,
unsigned* offset_ptr,
unsigned max_chars) {
ASSERT(*offset_ptr <= (unsigned)input->length());
if (max_chars == 0) return;
switch (StringShape(input).representation_tag()) {
case kSeqStringTag:
if (input->IsAsciiRepresentation()) {
SeqAsciiString::cast(input)->SeqAsciiStringReadBlockIntoBuffer(rbb,
offset_ptr,
max_chars);
return;
} else {
SeqTwoByteString::cast(input)->SeqTwoByteStringReadBlockIntoBuffer(rbb,
offset_ptr,
max_chars);
return;
}
case kConsStringTag:
ConsString::cast(input)->ConsStringReadBlockIntoBuffer(rbb,
offset_ptr,
max_chars);
return;
case kSlicedStringTag:
SlicedString::cast(input)->SlicedStringReadBlockIntoBuffer(rbb,
offset_ptr,
max_chars);
return;
case kExternalStringTag:
if (input->IsAsciiRepresentation()) {
ExternalAsciiString::cast(input)->
ExternalAsciiStringReadBlockIntoBuffer(rbb, offset_ptr, max_chars);
} else {
ExternalTwoByteString::cast(input)->
ExternalTwoByteStringReadBlockIntoBuffer(rbb,
offset_ptr,
max_chars);
}
return;
default:
break;
}
UNREACHABLE();
return;
}
const unibrow::byte* String::ReadBlock(String* input,
unibrow::byte* util_buffer,
unsigned capacity,
unsigned* remaining,
unsigned* offset_ptr) {
ASSERT(*offset_ptr <= (unsigned)input->length());
unsigned chars = input->length() - *offset_ptr;
ReadBlockBuffer rbb(util_buffer, 0, capacity, 0);
const unibrow::byte* answer = ReadBlock(input, &rbb, offset_ptr, chars);
ASSERT(rbb.remaining <= static_cast<unsigned>(input->length()));
*remaining = rbb.remaining;
return answer;
}
const unibrow::byte* String::ReadBlock(String** raw_input,
unibrow::byte* util_buffer,
unsigned capacity,
unsigned* remaining,
unsigned* offset_ptr) {
Handle<String> input(raw_input);
ASSERT(*offset_ptr <= (unsigned)input->length());
unsigned chars = input->length() - *offset_ptr;
if (chars > capacity) chars = capacity;
ReadBlockBuffer rbb(util_buffer, 0, capacity, 0);
ReadBlockIntoBuffer(*input, &rbb, offset_ptr, chars);
ASSERT(rbb.remaining <= static_cast<unsigned>(input->length()));
*remaining = rbb.remaining;
return rbb.util_buffer;
}
// This will iterate unless the block of string data spans two 'halves' of
// a ConsString, in which case it will recurse. Since the block of string
// data to be read has a maximum size this limits the maximum recursion
// depth to something sane. Since C++ does not have tail call recursion
// elimination, the iteration must be explicit.
void ConsString::ConsStringReadBlockIntoBuffer(ReadBlockBuffer* rbb,
unsigned* offset_ptr,
unsigned max_chars) {
ConsString* current = this;
unsigned offset = *offset_ptr;
int offset_correction = 0;
while (true) {
String* left = current->first();
unsigned left_length = (unsigned)left->length();
if (left_length > offset &&
max_chars <= left_length - offset) {
// Left hand side only - iterate unless we have reached the bottom of
// the cons tree.
if (StringShape(left).IsCons()) {
current = ConsString::cast(left);
continue;
} else {
String::ReadBlockIntoBuffer(left, rbb, &offset, max_chars);
*offset_ptr = offset + offset_correction;
return;
}
} else if (left_length <= offset) {
// Right hand side only - iterate unless we have reached the bottom of
// the cons tree.
offset -= left_length;
offset_correction += left_length;
String* right = current->second();
if (StringShape(right).IsCons()) {
current = ConsString::cast(right);
continue;
} else {
String::ReadBlockIntoBuffer(right, rbb, &offset, max_chars);
*offset_ptr = offset + offset_correction;
return;
}
} else {
// The block to be read spans two sides of the ConsString, so we recurse.
// First recurse on the left.
max_chars -= left_length - offset;
String::ReadBlockIntoBuffer(left, rbb, &offset, left_length - offset);
// We may have reached the max or there may not have been enough space
// in the buffer for the characters in the left hand side.
if (offset == left_length) {
// Recurse on the right.
String* right = String::cast(current->second());
offset -= left_length;
offset_correction += left_length;
String::ReadBlockIntoBuffer(right, rbb, &offset, max_chars);
}
*offset_ptr = offset + offset_correction;
return;
}
}
}
void SlicedString::SlicedStringReadBlockIntoBuffer(ReadBlockBuffer* rbb,
unsigned* offset_ptr,
unsigned max_chars) {
String* backing = buffer();
unsigned offset = start() + *offset_ptr;
unsigned length = backing->length();
if (max_chars > length - offset) {
max_chars = length - offset;
}
String::ReadBlockIntoBuffer(backing, rbb, &offset, max_chars);
*offset_ptr = offset - start();
}
void ConsString::ConsStringIterateBody(ObjectVisitor* v) {
IteratePointers(v, kFirstOffset, kSecondOffset + kPointerSize);
}
void JSGlobalPropertyCell::JSGlobalPropertyCellIterateBody(ObjectVisitor* v) {
IteratePointers(v, kValueOffset, kValueOffset + kPointerSize);
}
uint16_t ConsString::ConsStringGet(int index) {
ASSERT(index >= 0 && index < this->length());
// Check for a flattened cons string
if (second()->length() == 0) {
String* left = first();
return left->Get(index);
}
String* string = String::cast(this);
while (true) {
if (StringShape(string).IsCons()) {
ConsString* cons_string = ConsString::cast(string);
String* left = cons_string->first();
if (left->length() > index) {
string = left;
} else {
index -= left->length();
string = cons_string->second();
}
} else {
return string->Get(index);
}
}
UNREACHABLE();
return 0;
}
template <typename sinkchar>
void String::WriteToFlat(String* src,
sinkchar* sink,
int f,
int t) {
String* source = src;
int from = f;
int to = t;
while (true) {
ASSERT(0 <= from && from <= to && to <= source->length());
switch (StringShape(source).full_representation_tag()) {
case kAsciiStringTag | kExternalStringTag: {
CopyChars(sink,
ExternalAsciiString::cast(source)->resource()->data() + from,
to - from);
return;
}
case kTwoByteStringTag | kExternalStringTag: {
const uc16* data =
ExternalTwoByteString::cast(source)->resource()->data();
CopyChars(sink,
data + from,
to - from);
return;
}
case kAsciiStringTag | kSeqStringTag: {
CopyChars(sink,
SeqAsciiString::cast(source)->GetChars() + from,
to - from);
return;
}
case kTwoByteStringTag | kSeqStringTag: {
CopyChars(sink,
SeqTwoByteString::cast(source)->GetChars() + from,
to - from);
return;
}
case kAsciiStringTag | kSlicedStringTag:
case kTwoByteStringTag | kSlicedStringTag: {
SlicedString* sliced_string = SlicedString::cast(source);
int start = sliced_string->start();
from += start;
to += start;
source = String::cast(sliced_string->buffer());
break;
}
case kAsciiStringTag | kConsStringTag:
case kTwoByteStringTag | kConsStringTag: {
ConsString* cons_string = ConsString::cast(source);
String* first = cons_string->first();
int boundary = first->length();
if (to - boundary >= boundary - from) {
// Right hand side is longer. Recurse over left.
if (from < boundary) {
WriteToFlat(first, sink, from, boundary);
sink += boundary - from;
from = 0;
} else {
from -= boundary;
}
to -= boundary;
source = cons_string->second();
} else {
// Left hand side is longer. Recurse over right.
if (to > boundary) {
String* second = cons_string->second();
WriteToFlat(second,
sink + boundary - from,
0,
to - boundary);
to = boundary;
}
source = first;
}
break;
}
}
}
}
void SlicedString::SlicedStringIterateBody(ObjectVisitor* v) {
IteratePointer(v, kBufferOffset);
}
uint16_t SlicedString::SlicedStringGet(int index) {
ASSERT(index >= 0 && index < this->length());
// Delegate to the buffer string.
String* underlying = buffer();
return underlying->Get(start() + index);
}
template <typename IteratorA, typename IteratorB>
static inline bool CompareStringContents(IteratorA* ia, IteratorB* ib) {
// General slow case check. We know that the ia and ib iterators
// have the same length.
while (ia->has_more()) {
uc32 ca = ia->GetNext();
uc32 cb = ib->GetNext();
if (ca != cb)
return false;
}
return true;
}
// Compares the contents of two strings by reading and comparing
// int-sized blocks of characters.
template <typename Char>
static inline bool CompareRawStringContents(Vector<Char> a, Vector<Char> b) {
int length = a.length();
ASSERT_EQ(length, b.length());
const Char* pa = a.start();
const Char* pb = b.start();
int i = 0;
#ifndef V8_HOST_CAN_READ_UNALIGNED
// If this architecture isn't comfortable reading unaligned ints
// then we have to check that the strings are aligned before
// comparing them blockwise.
const int kAlignmentMask = sizeof(uint32_t) - 1; // NOLINT
uint32_t pa_addr = reinterpret_cast<uint32_t>(pa);
uint32_t pb_addr = reinterpret_cast<uint32_t>(pb);
if (((pa_addr & kAlignmentMask) | (pb_addr & kAlignmentMask)) == 0) {
#endif
const int kStepSize = sizeof(int) / sizeof(Char); // NOLINT
int endpoint = length - kStepSize;
// Compare blocks until we reach near the end of the string.
for (; i <= endpoint; i += kStepSize) {
uint32_t wa = *reinterpret_cast<const uint32_t*>(pa + i);
uint32_t wb = *reinterpret_cast<const uint32_t*>(pb + i);
if (wa != wb) {
return false;
}
}
#ifndef V8_HOST_CAN_READ_UNALIGNED
}
#endif
// Compare the remaining characters that didn't fit into a block.
for (; i < length; i++) {
if (a[i] != b[i]) {
return false;
}
}
return true;
}
static StringInputBuffer string_compare_buffer_b;
template <typename IteratorA>
static inline bool CompareStringContentsPartial(IteratorA* ia, String* b) {
if (b->IsFlat()) {
if (b->IsAsciiRepresentation()) {
VectorIterator<char> ib(b->ToAsciiVector());
return CompareStringContents(ia, &ib);
} else {
VectorIterator<uc16> ib(b->ToUC16Vector());
return CompareStringContents(ia, &ib);
}
} else {
string_compare_buffer_b.Reset(0, b);
return CompareStringContents(ia, &string_compare_buffer_b);
}
}
static StringInputBuffer string_compare_buffer_a;
bool String::SlowEquals(String* other) {
// Fast check: negative check with lengths.
int len = length();
if (len != other->length()) return false;
if (len == 0) return true;
// Fast check: if hash code is computed for both strings
// a fast negative check can be performed.
if (HasHashCode() && other->HasHashCode()) {
if (Hash() != other->Hash()) return false;
}
if (StringShape(this).IsSequentialAscii() &&
StringShape(other).IsSequentialAscii()) {
const char* str1 = SeqAsciiString::cast(this)->GetChars();
const char* str2 = SeqAsciiString::cast(other)->GetChars();
return CompareRawStringContents(Vector<const char>(str1, len),
Vector<const char>(str2, len));
}
if (this->IsFlat()) {
if (IsAsciiRepresentation()) {
Vector<const char> vec1 = this->ToAsciiVector();
if (other->IsFlat()) {
if (other->IsAsciiRepresentation()) {
Vector<const char> vec2 = other->ToAsciiVector();
return CompareRawStringContents(vec1, vec2);
} else {
VectorIterator<char> buf1(vec1);
VectorIterator<uc16> ib(other->ToUC16Vector());
return CompareStringContents(&buf1, &ib);
}
} else {
VectorIterator<char> buf1(vec1);
string_compare_buffer_b.Reset(0, other);
return CompareStringContents(&buf1, &string_compare_buffer_b);
}
} else {
Vector<const uc16> vec1 = this->ToUC16Vector();
if (other->IsFlat()) {
if (other->IsAsciiRepresentation()) {
VectorIterator<uc16> buf1(vec1);
VectorIterator<char> ib(other->ToAsciiVector());
return CompareStringContents(&buf1, &ib);
} else {
Vector<const uc16> vec2(other->ToUC16Vector());
return CompareRawStringContents(vec1, vec2);
}
} else {
VectorIterator<uc16> buf1(vec1);
string_compare_buffer_b.Reset(0, other);
return CompareStringContents(&buf1, &string_compare_buffer_b);
}
}
} else {
string_compare_buffer_a.Reset(0, this);
return CompareStringContentsPartial(&string_compare_buffer_a, other);
}
}
bool String::MarkAsUndetectable() {
if (StringShape(this).IsSymbol()) return false;
Map* map = this->map();
if (map == Heap::short_string_map()) {
this->set_map(Heap::undetectable_short_string_map());
return true;
} else if (map == Heap::medium_string_map()) {
this->set_map(Heap::undetectable_medium_string_map());
return true;
} else if (map == Heap::long_string_map()) {
this->set_map(Heap::undetectable_long_string_map());
return true;
} else if (map == Heap::short_ascii_string_map()) {
this->set_map(Heap::undetectable_short_ascii_string_map());
return true;
} else if (map == Heap::medium_ascii_string_map()) {
this->set_map(Heap::undetectable_medium_ascii_string_map());
return true;
} else if (map == Heap::long_ascii_string_map()) {
this->set_map(Heap::undetectable_long_ascii_string_map());
return true;
}
// Rest cannot be marked as undetectable
return false;
}
bool String::IsEqualTo(Vector<const char> str) {
int slen = length();
Access<Scanner::Utf8Decoder> decoder(Scanner::utf8_decoder());
decoder->Reset(str.start(), str.length());
int i;
for (i = 0; i < slen && decoder->has_more(); i++) {
uc32 r = decoder->GetNext();
if (Get(i) != r) return false;
}
return i == slen && !decoder->has_more();
}
uint32_t String::ComputeAndSetHash() {
// Should only be called if hash code has not yet been computed.
ASSERT(!(length_field() & kHashComputedMask));
// Compute the hash code.
StringInputBuffer buffer(this);
uint32_t field = ComputeLengthAndHashField(&buffer, length());
// Store the hash code in the object.
set_length_field(field);
// Check the hash code is there.
ASSERT(length_field() & kHashComputedMask);
uint32_t result = field >> kHashShift;
ASSERT(result != 0); // Ensure that the hash value of 0 is never computed.
return result;
}
bool String::ComputeArrayIndex(unibrow::CharacterStream* buffer,
uint32_t* index,
int length) {
if (length == 0 || length > kMaxArrayIndexSize) return false;
uc32 ch = buffer->GetNext();
// If the string begins with a '0' character, it must only consist
// of it to be a legal array index.
if (ch == '0') {
*index = 0;
return length == 1;
}
// Convert string to uint32 array index; character by character.
int d = ch - '0';
if (d < 0 || d > 9) return false;
uint32_t result = d;
while (buffer->has_more()) {
d = buffer->GetNext() - '0';
if (d < 0 || d > 9) return false;
// Check that the new result is below the 32 bit limit.
if (result > 429496729U - ((d > 5) ? 1 : 0)) return false;
result = (result * 10) + d;
}
*index = result;
return true;
}
bool String::SlowAsArrayIndex(uint32_t* index) {
if (length() <= kMaxCachedArrayIndexLength) {
Hash(); // force computation of hash code
uint32_t field = length_field();
if ((field & kIsArrayIndexMask) == 0) return false;
*index = (field & ((1 << kShortLengthShift) - 1)) >> kLongLengthShift;
return true;
} else {
StringInputBuffer buffer(this);
return ComputeArrayIndex(&buffer, index, length());
}
}
static inline uint32_t HashField(uint32_t hash, bool is_array_index) {
uint32_t result =
(hash << String::kLongLengthShift) | String::kHashComputedMask;
if (is_array_index) result |= String::kIsArrayIndexMask;
return result;
}
uint32_t StringHasher::GetHashField() {
ASSERT(is_valid());
if (length_ <= String::kMaxShortStringSize) {
uint32_t payload;
if (is_array_index()) {
payload = v8::internal::HashField(array_index(), true);
} else {
payload = v8::internal::HashField(GetHash(), false);
}
return (payload & ((1 << String::kShortLengthShift) - 1)) |
(length_ << String::kShortLengthShift);
} else if (length_ <= String::kMaxMediumStringSize) {
uint32_t payload = v8::internal::HashField(GetHash(), false);
return (payload & ((1 << String::kMediumLengthShift) - 1)) |
(length_ << String::kMediumLengthShift);
} else {
return v8::internal::HashField(length_, false);
}
}
uint32_t String::ComputeLengthAndHashField(unibrow::CharacterStream* buffer,
int length) {
StringHasher hasher(length);
// Very long strings have a trivial hash that doesn't inspect the
// string contents.
if (hasher.has_trivial_hash()) {
return hasher.GetHashField();
}
// Do the iterative array index computation as long as there is a
// chance this is an array index.
while (buffer->has_more() && hasher.is_array_index()) {
hasher.AddCharacter(buffer->GetNext());
}
// Process the remaining characters without updating the array
// index.
while (buffer->has_more()) {
hasher.AddCharacterNoIndex(buffer->GetNext());
}
return hasher.GetHashField();
}
Object* String::Slice(int start, int end) {
if (start == 0 && end == length()) return this;
if (StringShape(this).representation_tag() == kSlicedStringTag) {
// Translate slices of a SlicedString into slices of the
// underlying string buffer.
SlicedString* str = SlicedString::cast(this);
String* buf = str->buffer();
return Heap::AllocateSlicedString(buf,
str->start() + start,
str->start() + end);
}
Object* result = Heap::AllocateSlicedString(this, start, end);
if (result->IsFailure()) {
return result;
}
// Due to the way we retry after GC on allocation failure we are not allowed
// to fail on allocation after this point. This is the one-allocation rule.
// Try to flatten a cons string that is under the sliced string.
// This is to avoid memory leaks and possible stack overflows caused by
// building 'towers' of sliced strings on cons strings.
// This may fail due to an allocation failure (when a GC is needed), but it
// will succeed often enough to avoid the problem. We only have to do this
// if Heap::AllocateSlicedString actually returned a SlicedString. It will
// return flat strings for small slices for efficiency reasons.
String* answer = String::cast(result);
if (StringShape(answer).IsSliced() &&
StringShape(this).representation_tag() == kConsStringTag) {
TryFlatten();
// If the flatten succeeded we might as well make the sliced string point
// to the flat string rather than the cons string.
String* second = ConsString::cast(this)->second();
if (second->length() == 0) {
SlicedString::cast(answer)->set_buffer(ConsString::cast(this)->first());
}
}
return answer;
}
void String::PrintOn(FILE* file) {
int length = this->length();
for (int i = 0; i < length; i++) {
fprintf(file, "%c", Get(i));
}
}
void Map::CreateBackPointers() {
DescriptorArray* descriptors = instance_descriptors();
for (int i = 0; i < descriptors->number_of_descriptors(); i++) {
if (descriptors->GetType(i) == MAP_TRANSITION) {
// Get target.
Map* target = Map::cast(descriptors->GetValue(i));
#ifdef DEBUG
// Verify target.
Object* source_prototype = prototype();
Object* target_prototype = target->prototype();
ASSERT(source_prototype->IsJSObject() ||
source_prototype->IsMap() ||
source_prototype->IsNull());
ASSERT(target_prototype->IsJSObject() ||
target_prototype->IsNull());
ASSERT(source_prototype->IsMap() ||
source_prototype == target_prototype);
#endif
// Point target back to source. set_prototype() will not let us set
// the prototype to a map, as we do here.
*RawField(target, kPrototypeOffset) = this;
}
}
}
void Map::ClearNonLiveTransitions(Object* real_prototype) {
// Live DescriptorArray objects will be marked, so we must use
// low-level accessors to get and modify their data.
DescriptorArray* d = reinterpret_cast<DescriptorArray*>(
*RawField(this, Map::kInstanceDescriptorsOffset));
if (d == Heap::raw_unchecked_empty_descriptor_array()) return;
Smi* NullDescriptorDetails =
PropertyDetails(NONE, NULL_DESCRIPTOR).AsSmi();
FixedArray* contents = reinterpret_cast<FixedArray*>(
d->get(DescriptorArray::kContentArrayIndex));
ASSERT(contents->length() >= 2);
for (int i = 0; i < contents->length(); i += 2) {
// If the pair (value, details) is a map transition,
// check if the target is live. If not, null the descriptor.
// Also drop the back pointer for that map transition, so that this
// map is not reached again by following a back pointer from a
// non-live object.
PropertyDetails details(Smi::cast(contents->get(i + 1)));
if (details.type() == MAP_TRANSITION) {
Map* target = reinterpret_cast<Map*>(contents->get(i));
ASSERT(target->IsHeapObject());
if (!target->IsMarked()) {
ASSERT(target->IsMap());
contents->set(i + 1, NullDescriptorDetails, SKIP_WRITE_BARRIER);
contents->set(i, Heap::null_value(), SKIP_WRITE_BARRIER);
ASSERT(target->prototype() == this ||
target->prototype() == real_prototype);
// Getter prototype() is read-only, set_prototype() has side effects.
*RawField(target, Map::kPrototypeOffset) = real_prototype;
}
}
}
}
void Map::MapIterateBody(ObjectVisitor* v) {
// Assumes all Object* members are contiguously allocated!
IteratePointers(v, kPrototypeOffset, kCodeCacheOffset + kPointerSize);
}
Object* JSFunction::SetInstancePrototype(Object* value) {
ASSERT(value->IsJSObject());
if (has_initial_map()) {
initial_map()->set_prototype(value);
} else {
// Put the value in the initial map field until an initial map is
// needed. At that point, a new initial map is created and the
// prototype is put into the initial map where it belongs.
set_prototype_or_initial_map(value);
}
return value;
}
Object* JSFunction::SetPrototype(Object* value) {
Object* construct_prototype = value;
// If the value is not a JSObject, store the value in the map's
// constructor field so it can be accessed. Also, set the prototype
// used for constructing objects to the original object prototype.
// See ECMA-262 13.2.2.
if (!value->IsJSObject()) {
// Copy the map so this does not affect unrelated functions.
// Remove map transitions because they point to maps with a
// different prototype.
Object* new_map = map()->CopyDropTransitions();
if (new_map->IsFailure()) return new_map;
set_map(Map::cast(new_map));
map()->set_constructor(value);
map()->set_non_instance_prototype(true);
construct_prototype =
Top::context()->global_context()->initial_object_prototype();
} else {
map()->set_non_instance_prototype(false);
}
return SetInstancePrototype(construct_prototype);
}
Object* JSFunction::SetInstanceClassName(String* name) {
shared()->set_instance_class_name(name);
return this;
}
Context* JSFunction::GlobalContextFromLiterals(FixedArray* literals) {
return Context::cast(literals->get(JSFunction::kLiteralGlobalContextIndex));
}
void Oddball::OddballIterateBody(ObjectVisitor* v) {
// Assumes all Object* members are contiguously allocated!
IteratePointers(v, kToStringOffset, kToNumberOffset + kPointerSize);
}
Object* Oddball::Initialize(const char* to_string, Object* to_number) {
Object* symbol = Heap::LookupAsciiSymbol(to_string);
if (symbol->IsFailure()) return symbol;
set_to_string(String::cast(symbol));
set_to_number(to_number);
return this;
}
bool SharedFunctionInfo::HasSourceCode() {
return !script()->IsUndefined() &&
!Script::cast(script())->source()->IsUndefined();
}
Object* SharedFunctionInfo::GetSourceCode() {
HandleScope scope;
if (script()->IsUndefined()) return Heap::undefined_value();
Object* source = Script::cast(script())->source();
if (source->IsUndefined()) return Heap::undefined_value();
return *SubString(Handle<String>(String::cast(source)),
start_position(), end_position());
}
int SharedFunctionInfo::CalculateInstanceSize() {
int instance_size =
JSObject::kHeaderSize +
expected_nof_properties() * kPointerSize;
if (instance_size > JSObject::kMaxInstanceSize) {
instance_size = JSObject::kMaxInstanceSize;
}
return instance_size;
}
int SharedFunctionInfo::CalculateInObjectProperties() {
return (CalculateInstanceSize() - JSObject::kHeaderSize) / kPointerSize;
}
void SharedFunctionInfo::SetThisPropertyAssignmentsInfo(
bool only_this_property_assignments,
bool only_simple_this_property_assignments,
FixedArray* assignments) {
set_compiler_hints(BooleanBit::set(compiler_hints(),
kHasOnlyThisPropertyAssignments,
only_this_property_assignments));
set_compiler_hints(BooleanBit::set(compiler_hints(),
kHasOnlySimpleThisPropertyAssignments,
only_simple_this_property_assignments));
set_this_property_assignments(assignments);
set_this_property_assignments_count(assignments->length() / 3);
}
void SharedFunctionInfo::ClearThisPropertyAssignmentsInfo() {
set_compiler_hints(BooleanBit::set(compiler_hints(),
kHasOnlyThisPropertyAssignments,
false));
set_compiler_hints(BooleanBit::set(compiler_hints(),
kHasOnlySimpleThisPropertyAssignments,
false));
set_this_property_assignments(Heap::undefined_value());
set_this_property_assignments_count(0);
}
String* SharedFunctionInfo::GetThisPropertyAssignmentName(int index) {
Object* obj = this_property_assignments();
ASSERT(obj->IsFixedArray());
ASSERT(index < this_property_assignments_count());
obj = FixedArray::cast(obj)->get(index * 3);
ASSERT(obj->IsString());
return String::cast(obj);
}
bool SharedFunctionInfo::IsThisPropertyAssignmentArgument(int index) {
Object* obj = this_property_assignments();
ASSERT(obj->IsFixedArray());
ASSERT(index < this_property_assignments_count());
obj = FixedArray::cast(obj)->get(index * 3 + 1);
return Smi::cast(obj)->value() != -1;
}
int SharedFunctionInfo::GetThisPropertyAssignmentArgument(int index) {
ASSERT(IsThisPropertyAssignmentArgument(index));
Object* obj =
FixedArray::cast(this_property_assignments())->get(index * 3 + 1);
return Smi::cast(obj)->value();
}
Object* SharedFunctionInfo::GetThisPropertyAssignmentConstant(int index) {
ASSERT(!IsThisPropertyAssignmentArgument(index));
Object* obj =
FixedArray::cast(this_property_assignments())->get(index * 3 + 2);
return obj;
}
// Support function for printing the source code to a StringStream
// without any allocation in the heap.
void SharedFunctionInfo::SourceCodePrint(StringStream* accumulator,
int max_length) {
// For some native functions there is no source.
if (script()->IsUndefined() ||
Script::cast(script())->source()->IsUndefined()) {
accumulator->Add("<No Source>");
return;
}
// Get the slice of the source for this function.
// Don't use String::cast because we don't want more assertion errors while
// we are already creating a stack dump.
String* script_source =
reinterpret_cast<String*>(Script::cast(script())->source());
if (!script_source->LooksValid()) {
accumulator->Add("<Invalid Source>");
return;
}
if (!is_toplevel()) {
accumulator->Add("function ");
Object* name = this->name();
if (name->IsString() && String::cast(name)->length() > 0) {
accumulator->PrintName(name);
}
}
int len = end_position() - start_position();
if (len > max_length) {
accumulator->Put(script_source,
start_position(),
start_position() + max_length);
accumulator->Add("...\n");
} else {
accumulator->Put(script_source, start_position(), end_position());
}
}
void SharedFunctionInfo::SharedFunctionInfoIterateBody(ObjectVisitor* v) {
IteratePointers(v, kNameOffset, kConstructStubOffset + kPointerSize);
IteratePointers(v, kInstanceClassNameOffset, kScriptOffset + kPointerSize);
IteratePointers(v, kDebugInfoOffset, kInferredNameOffset + kPointerSize);
IteratePointers(v, kThisPropertyAssignmentsOffset,
kThisPropertyAssignmentsOffset + kPointerSize);
}
void ObjectVisitor::VisitCodeTarget(RelocInfo* rinfo) {
ASSERT(RelocInfo::IsCodeTarget(rinfo->rmode()));
Object* target = Code::GetCodeFromTargetAddress(rinfo->target_address());
Object* old_target = target;
VisitPointer(&target);
CHECK_EQ(target, old_target); // VisitPointer doesn't change Code* *target.
}
void ObjectVisitor::VisitDebugTarget(RelocInfo* rinfo) {
ASSERT(RelocInfo::IsJSReturn(rinfo->rmode()) && rinfo->IsCallInstruction());
Object* target = Code::GetCodeFromTargetAddress(rinfo->call_address());
Object* old_target = target;
VisitPointer(&target);
CHECK_EQ(target, old_target); // VisitPointer doesn't change Code* *target.
}
void Code::CodeIterateBody(ObjectVisitor* v) {
int mode_mask = RelocInfo::kCodeTargetMask |
RelocInfo::ModeMask(RelocInfo::EMBEDDED_OBJECT) |
RelocInfo::ModeMask(RelocInfo::EXTERNAL_REFERENCE) |
RelocInfo::ModeMask(RelocInfo::JS_RETURN) |
RelocInfo::ModeMask(RelocInfo::RUNTIME_ENTRY);
for (RelocIterator it(this, mode_mask); !it.done(); it.next()) {
RelocInfo::Mode rmode = it.rinfo()->rmode();
if (rmode == RelocInfo::EMBEDDED_OBJECT) {
v->VisitPointer(it.rinfo()->target_object_address());
} else if (RelocInfo::IsCodeTarget(rmode)) {
v->VisitCodeTarget(it.rinfo());
} else if (rmode == RelocInfo::EXTERNAL_REFERENCE) {
v->VisitExternalReference(it.rinfo()->target_reference_address());
#ifdef ENABLE_DEBUGGER_SUPPORT
} else if (Debug::has_break_points() &&
RelocInfo::IsJSReturn(rmode) &&
it.rinfo()->IsCallInstruction()) {
v->VisitDebugTarget(it.rinfo());
#endif
} else if (rmode == RelocInfo::RUNTIME_ENTRY) {
v->VisitRuntimeEntry(it.rinfo());
}
}
ScopeInfo<>::IterateScopeInfo(this, v);
}
void Code::Relocate(int delta) {
for (RelocIterator it(this, RelocInfo::kApplyMask); !it.done(); it.next()) {
it.rinfo()->apply(delta);
}
CPU::FlushICache(instruction_start(), instruction_size());
}
void Code::CopyFrom(const CodeDesc& desc) {
// copy code
memmove(instruction_start(), desc.buffer, desc.instr_size);
// fill gap with zero bytes
{ byte* p = instruction_start() + desc.instr_size;
byte* q = relocation_start();
while (p < q) {
*p++ = 0;
}
}
// copy reloc info
memmove(relocation_start(),
desc.buffer + desc.buffer_size - desc.reloc_size,
desc.reloc_size);
// unbox handles and relocate
int delta = instruction_start() - desc.buffer;
int mode_mask = RelocInfo::kCodeTargetMask |
RelocInfo::ModeMask(RelocInfo::EMBEDDED_OBJECT) |
RelocInfo::kApplyMask;
for (RelocIterator it(this, mode_mask); !it.done(); it.next()) {
RelocInfo::Mode mode = it.rinfo()->rmode();
if (mode == RelocInfo::EMBEDDED_OBJECT) {
Object** p = reinterpret_cast<Object**>(it.rinfo()->target_object());
it.rinfo()->set_target_object(*p);
} else if (RelocInfo::IsCodeTarget(mode)) {
// rewrite code handles in inline cache targets to direct
// pointers to the first instruction in the code object
Object** p = reinterpret_cast<Object**>(it.rinfo()->target_object());
Code* code = Code::cast(*p);
it.rinfo()->set_target_address(code->instruction_start());
} else {
it.rinfo()->apply(delta);
}
}
CPU::FlushICache(instruction_start(), instruction_size());
}
// Locate the source position which is closest to the address in the code. This
// is using the source position information embedded in the relocation info.
// The position returned is relative to the beginning of the script where the
// source for this function is found.
int Code::SourcePosition(Address pc) {
int distance = kMaxInt;
int position = RelocInfo::kNoPosition; // Initially no position found.
// Run through all the relocation info to find the best matching source
// position. All the code needs to be considered as the sequence of the
// instructions in the code does not necessarily follow the same order as the
// source.
RelocIterator it(this, RelocInfo::kPositionMask);
while (!it.done()) {
// Only look at positions after the current pc.
if (it.rinfo()->pc() < pc) {
// Get position and distance.
int dist = pc - it.rinfo()->pc();
int pos = it.rinfo()->data();
// If this position is closer than the current candidate or if it has the
// same distance as the current candidate and the position is higher then
// this position is the new candidate.
if ((dist < distance) ||
(dist == distance && pos > position)) {
position = pos;
distance = dist;
}
}
it.next();
}
return position;
}
// Same as Code::SourcePosition above except it only looks for statement
// positions.
int Code::SourceStatementPosition(Address pc) {
// First find the position as close as possible using all position
// information.
int position = SourcePosition(pc);
// Now find the closest statement position before the position.
int statement_position = 0;
RelocIterator it(this, RelocInfo::kPositionMask);
while (!it.done()) {
if (RelocInfo::IsStatementPosition(it.rinfo()->rmode())) {
int p = it.rinfo()->data();
if (statement_position < p && p <= position) {
statement_position = p;
}
}
it.next();
}
return statement_position;
}
#ifdef ENABLE_DISASSEMBLER
// Identify kind of code.
const char* Code::Kind2String(Kind kind) {
switch (kind) {
case FUNCTION: return "FUNCTION";
case STUB: return "STUB";
case BUILTIN: return "BUILTIN";
case LOAD_IC: return "LOAD_IC";
case KEYED_LOAD_IC: return "KEYED_LOAD_IC";
case STORE_IC: return "STORE_IC";
case KEYED_STORE_IC: return "KEYED_STORE_IC";
case CALL_IC: return "CALL_IC";
}
UNREACHABLE();
return NULL;
}
const char* Code::ICState2String(InlineCacheState state) {
switch (state) {
case UNINITIALIZED: return "UNINITIALIZED";
case PREMONOMORPHIC: return "PREMONOMORPHIC";
case MONOMORPHIC: return "MONOMORPHIC";
case MONOMORPHIC_PROTOTYPE_FAILURE: return "MONOMORPHIC_PROTOTYPE_FAILURE";
case MEGAMORPHIC: return "MEGAMORPHIC";
case DEBUG_BREAK: return "DEBUG_BREAK";
case DEBUG_PREPARE_STEP_IN: return "DEBUG_PREPARE_STEP_IN";
}
UNREACHABLE();
return NULL;
}
const char* Code::PropertyType2String(PropertyType type) {
switch (type) {
case NORMAL: return "NORMAL";
case FIELD: return "FIELD";
case CONSTANT_FUNCTION: return "CONSTANT_FUNCTION";
case CALLBACKS: return "CALLBACKS";
case INTERCEPTOR: return "INTERCEPTOR";
case MAP_TRANSITION: return "MAP_TRANSITION";
case CONSTANT_TRANSITION: return "CONSTANT_TRANSITION";
case NULL_DESCRIPTOR: return "NULL_DESCRIPTOR";
}
UNREACHABLE();
return NULL;
}
void Code::Disassemble(const char* name) {
PrintF("kind = %s\n", Kind2String(kind()));
if (is_inline_cache_stub()) {
PrintF("ic_state = %s\n", ICState2String(ic_state()));
PrintF("ic_in_loop = %d\n", ic_in_loop() == IN_LOOP);
if (ic_state() == MONOMORPHIC) {
PrintF("type = %s\n", PropertyType2String(type()));
}
}
if ((name != NULL) && (name[0] != '\0')) {
PrintF("name = %s\n", name);
}
PrintF("Instructions (size = %d)\n", instruction_size());
Disassembler::Decode(NULL, this);
PrintF("\n");
PrintF("RelocInfo (size = %d)\n", relocation_size());
for (RelocIterator it(this); !it.done(); it.next())
it.rinfo()->Print();
PrintF("\n");
}
#endif // ENABLE_DISASSEMBLER
void JSObject::SetFastElements(FixedArray* elems) {
// We should never end in here with a pixel array.
ASSERT(!HasPixelElements());
#ifdef DEBUG
// Check the provided array is filled with the_hole.
uint32_t len = static_cast<uint32_t>(elems->length());
for (uint32_t i = 0; i < len; i++) ASSERT(elems->get(i)->IsTheHole());
#endif
WriteBarrierMode mode = elems->GetWriteBarrierMode();
switch (GetElementsKind()) {
case FAST_ELEMENTS: {
FixedArray* old_elements = FixedArray::cast(elements());
uint32_t old_length = static_cast<uint32_t>(old_elements->length());
// Fill out the new array with this content and array holes.
for (uint32_t i = 0; i < old_length; i++) {
elems->set(i, old_elements->get(i), mode);
}
break;
}
case DICTIONARY_ELEMENTS: {
NumberDictionary* dictionary = NumberDictionary::cast(elements());
for (int i = 0; i < dictionary->Capacity(); i++) {
Object* key = dictionary->KeyAt(i);
if (key->IsNumber()) {
uint32_t entry = static_cast<uint32_t>(key->Number());
elems->set(entry, dictionary->ValueAt(i), mode);
}
}
break;
}
default:
UNREACHABLE();
break;
}
set_elements(elems);
}
Object* JSObject::SetSlowElements(Object* len) {
// We should never end in here with a pixel array.
ASSERT(!HasPixelElements());
uint32_t new_length = static_cast<uint32_t>(len->Number());
switch (GetElementsKind()) {
case FAST_ELEMENTS: {
// Make sure we never try to shrink dense arrays into sparse arrays.
ASSERT(static_cast<uint32_t>(FixedArray::cast(elements())->length()) <=
new_length);
Object* obj = NormalizeElements();
if (obj->IsFailure()) return obj;
// Update length for JSArrays.
if (IsJSArray()) JSArray::cast(this)->set_length(len);
break;
}
case DICTIONARY_ELEMENTS: {
if (IsJSArray()) {
uint32_t old_length =
static_cast<uint32_t>(JSArray::cast(this)->length()->Number());
element_dictionary()->RemoveNumberEntries(new_length, old_length),
JSArray::cast(this)->set_length(len);
}
break;
}
default:
UNREACHABLE();
break;
}
return this;
}
Object* JSArray::Initialize(int capacity) {
ASSERT(capacity >= 0);
set_length(Smi::FromInt(0), SKIP_WRITE_BARRIER);
FixedArray* new_elements;
if (capacity == 0) {
new_elements = Heap::empty_fixed_array();
} else {
Object* obj = Heap::AllocateFixedArrayWithHoles(capacity);
if (obj->IsFailure()) return obj;
new_elements = FixedArray::cast(obj);
}
set_elements(new_elements);
return this;
}
void JSArray::Expand(int required_size) {
Handle<JSArray> self(this);
Handle<FixedArray> old_backing(FixedArray::cast(elements()));
int old_size = old_backing->length();
// Doubling in size would be overkill, but leave some slack to avoid
// constantly growing.
int new_size = required_size + (required_size >> 3);
Handle<FixedArray> new_backing = Factory::NewFixedArray(new_size);
// Can't use this any more now because we may have had a GC!
for (int i = 0; i < old_size; i++) new_backing->set(i, old_backing->get(i));
self->SetContent(*new_backing);
}
// Computes the new capacity when expanding the elements of a JSObject.
static int NewElementsCapacity(int old_capacity) {
// (old_capacity + 50%) + 16
return old_capacity + (old_capacity >> 1) + 16;
}
static Object* ArrayLengthRangeError() {
HandleScope scope;
return Top::Throw(*Factory::NewRangeError("invalid_array_length",
HandleVector<Object>(NULL, 0)));
}
Object* JSObject::SetElementsLength(Object* len) {
// We should never end in here with a pixel array.
ASSERT(!HasPixelElements());
Object* smi_length = len->ToSmi();
if (smi_length->IsSmi()) {
int value = Smi::cast(smi_length)->value();
if (value < 0) return ArrayLengthRangeError();
switch (GetElementsKind()) {
case FAST_ELEMENTS: {
int old_capacity = FixedArray::cast(elements())->length();
if (value <= old_capacity) {
if (IsJSArray()) {
int old_length = FastD2I(JSArray::cast(this)->length()->Number());
// NOTE: We may be able to optimize this by removing the
// last part of the elements backing storage array and
// setting the capacity to the new size.
for (int i = value; i < old_length; i++) {
FixedArray::cast(elements())->set_the_hole(i);
}
JSArray::cast(this)->set_length(smi_length, SKIP_WRITE_BARRIER);
}
return this;
}
int min = NewElementsCapacity(old_capacity);
int new_capacity = value > min ? value : min;
if (new_capacity <= kMaxFastElementsLength ||
!ShouldConvertToSlowElements(new_capacity)) {
Object* obj = Heap::AllocateFixedArrayWithHoles(new_capacity);
if (obj->IsFailure()) return obj;
if (IsJSArray()) JSArray::cast(this)->set_length(smi_length,
SKIP_WRITE_BARRIER);
SetFastElements(FixedArray::cast(obj));
return this;
}
break;
}
case DICTIONARY_ELEMENTS: {
if (IsJSArray()) {
if (value == 0) {
// If the length of a slow array is reset to zero, we clear
// the array and flush backing storage. This has the added
// benefit that the array returns to fast mode.
initialize_elements();
} else {
// Remove deleted elements.
uint32_t old_length =
static_cast<uint32_t>(JSArray::cast(this)->length()->Number());
element_dictionary()->RemoveNumberEntries(value, old_length);
}
JSArray::cast(this)->set_length(smi_length, SKIP_WRITE_BARRIER);
}
return this;
}
default:
UNREACHABLE();
break;
}
}
// General slow case.
if (len->IsNumber()) {
uint32_t length;
if (Array::IndexFromObject(len, &length)) {
return SetSlowElements(len);
} else {
return ArrayLengthRangeError();
}
}
// len is not a number so make the array size one and
// set only element to len.
Object* obj = Heap::AllocateFixedArray(1);
if (obj->IsFailure()) return obj;
FixedArray::cast(obj)->set(0, len);
if (IsJSArray()) JSArray::cast(this)->set_length(Smi::FromInt(1),
SKIP_WRITE_BARRIER);
set_elements(FixedArray::cast(obj));
return this;
}
bool JSObject::HasElementPostInterceptor(JSObject* receiver, uint32_t index) {
switch (GetElementsKind()) {
case FAST_ELEMENTS: {
uint32_t length = IsJSArray() ?
static_cast<uint32_t>
(Smi::cast(JSArray::cast(this)->length())->value()) :
static_cast<uint32_t>(FixedArray::cast(elements())->length());
if ((index < length) &&
!FixedArray::cast(elements())->get(index)->IsTheHole()) {
return true;
}
break;
}
case PIXEL_ELEMENTS: {
// TODO(iposva): Add testcase.
PixelArray* pixels = PixelArray::cast(elements());
if (index < static_cast<uint32_t>(pixels->length())) {
return true;
}
break;
}
case DICTIONARY_ELEMENTS: {
if (element_dictionary()->FindEntry(index)
!= NumberDictionary::kNotFound) {
return true;
}
break;
}
default:
UNREACHABLE();
break;
}
// Handle [] on String objects.
if (this->IsStringObjectWithCharacterAt(index)) return true;
Object* pt = GetPrototype();
if (pt == Heap::null_value()) return false;
return JSObject::cast(pt)->HasElementWithReceiver(receiver, index);
}
bool JSObject::HasElementWithInterceptor(JSObject* receiver, uint32_t index) {
// Make sure that the top context does not change when doing
// callbacks or interceptor calls.
AssertNoContextChange ncc;
HandleScope scope;
Handle<InterceptorInfo> interceptor(GetIndexedInterceptor());
Handle<JSObject> receiver_handle(receiver);
Handle<JSObject> holder_handle(this);
CustomArguments args(interceptor->data(), receiver, this);
v8::AccessorInfo info(args.end());
if (!interceptor->query()->IsUndefined()) {
v8::IndexedPropertyQuery query =
v8::ToCData<v8::IndexedPropertyQuery>(interceptor->query());
LOG(ApiIndexedPropertyAccess("interceptor-indexed-has", this, index));
v8::Handle<v8::Boolean> result;
{
// Leaving JavaScript.
VMState state(EXTERNAL);
result = query(index, info);
}
if (!result.IsEmpty()) return result->IsTrue();
} else if (!interceptor->getter()->IsUndefined()) {
v8::IndexedPropertyGetter getter =
v8::ToCData<v8::IndexedPropertyGetter>(interceptor->getter());
LOG(ApiIndexedPropertyAccess("interceptor-indexed-has-get", this, index));
v8::Handle<v8::Value> result;
{
// Leaving JavaScript.
VMState state(EXTERNAL);
result = getter(index, info);
}
if (!result.IsEmpty()) return true;
}
return holder_handle->HasElementPostInterceptor(*receiver_handle, index);
}
bool JSObject::HasLocalElement(uint32_t index) {
// Check access rights if needed.
if (IsAccessCheckNeeded() &&
!Top::MayIndexedAccess(this, index, v8::ACCESS_HAS)) {
Top::ReportFailedAccessCheck(this, v8::ACCESS_HAS);
return false;
}
// Check for lookup interceptor
if (HasIndexedInterceptor()) {
return HasElementWithInterceptor(this, index);
}
// Handle [] on String objects.
if (this->IsStringObjectWithCharacterAt(index)) return true;
switch (GetElementsKind()) {
case FAST_ELEMENTS: {
uint32_t length = IsJSArray() ?
static_cast<uint32_t>
(Smi::cast(JSArray::cast(this)->length())->value()) :
static_cast<uint32_t>(FixedArray::cast(elements())->length());
return (index < length) &&
!FixedArray::cast(elements())->get(index)->IsTheHole();
}
case PIXEL_ELEMENTS: {
PixelArray* pixels = PixelArray::cast(elements());
return (index < static_cast<uint32_t>(pixels->length()));
}
case DICTIONARY_ELEMENTS: {
return element_dictionary()->FindEntry(index)
!= NumberDictionary::kNotFound;
}
default:
UNREACHABLE();
break;
}
UNREACHABLE();
return Heap::null_value();
}
bool JSObject::HasElementWithReceiver(JSObject* receiver, uint32_t index) {
// Check access rights if needed.
if (IsAccessCheckNeeded() &&
!Top::MayIndexedAccess(this, index, v8::ACCESS_HAS)) {
Top::ReportFailedAccessCheck(this, v8::ACCESS_HAS);
return false;
}
// Check for lookup interceptor
if (HasIndexedInterceptor()) {
return HasElementWithInterceptor(receiver, index);
}
switch (GetElementsKind()) {
case FAST_ELEMENTS: {
uint32_t length = IsJSArray() ?
static_cast<uint32_t>
(Smi::cast(JSArray::cast(this)->length())->value()) :
static_cast<uint32_t>(FixedArray::cast(elements())->length());
if ((index < length) &&
!FixedArray::cast(elements())->get(index)->IsTheHole()) return true;
break;
}
case PIXEL_ELEMENTS: {
PixelArray* pixels = PixelArray::cast(elements());
if (index < static_cast<uint32_t>(pixels->length())) {
return true;
}
break;
}
case DICTIONARY_ELEMENTS: {
if (element_dictionary()->FindEntry(index)
!= NumberDictionary::kNotFound) {
return true;
}
break;
}
default:
UNREACHABLE();
break;
}
// Handle [] on String objects.
if (this->IsStringObjectWithCharacterAt(index)) return true;
Object* pt = GetPrototype();
if (pt == Heap::null_value()) return false;
return JSObject::cast(pt)->HasElementWithReceiver(receiver, index);
}
Object* JSObject::SetElementWithInterceptor(uint32_t index, Object* value) {
// Make sure that the top context does not change when doing
// callbacks or interceptor calls.
AssertNoContextChange ncc;
HandleScope scope;
Handle<InterceptorInfo> interceptor(GetIndexedInterceptor());
Handle<JSObject> this_handle(this);
Handle<Object> value_handle(value);
if (!interceptor->setter()->IsUndefined()) {
v8::IndexedPropertySetter setter =
v8::ToCData<v8::IndexedPropertySetter>(interceptor->setter());
LOG(ApiIndexedPropertyAccess("interceptor-indexed-set", this, index));
CustomArguments args(interceptor->data(), this, this);
v8::AccessorInfo info(args.end());
v8::Handle<v8::Value> result;
{
// Leaving JavaScript.
VMState state(EXTERNAL);
result = setter(index, v8::Utils::ToLocal(value_handle), info);
}
RETURN_IF_SCHEDULED_EXCEPTION();
if (!result.IsEmpty()) return *value_handle;
}
Object* raw_result =
this_handle->SetElementWithoutInterceptor(index, *value_handle);
RETURN_IF_SCHEDULED_EXCEPTION();
return raw_result;
}
// Adding n elements in fast case is O(n*n).
// Note: revisit design to have dual undefined values to capture absent
// elements.
Object* JSObject::SetFastElement(uint32_t index, Object* value) {
ASSERT(HasFastElements());
FixedArray* elms = FixedArray::cast(elements());
uint32_t elms_length = static_cast<uint32_t>(elms->length());
if (!IsJSArray() && (index >= elms_length || elms->get(index)->IsTheHole())) {
Object* setter = LookupCallbackSetterInPrototypes(index);
if (setter->IsJSFunction()) {
return SetPropertyWithDefinedSetter(JSFunction::cast(setter), value);
}
}
// Check whether there is extra space in fixed array..
if (index < elms_length) {
elms->set(index, value);
if (IsJSArray()) {
// Update the length of the array if needed.
uint32_t array_length = 0;
CHECK(Array::IndexFromObject(JSArray::cast(this)->length(),
&array_length));
if (index >= array_length) {
JSArray::cast(this)->set_length(Smi::FromInt(index + 1),
SKIP_WRITE_BARRIER);
}
}
return value;
}
// Allow gap in fast case.
if ((index - elms_length) < kMaxGap) {
// Try allocating extra space.
int new_capacity = NewElementsCapacity(index+1);
if (new_capacity <= kMaxFastElementsLength ||
!ShouldConvertToSlowElements(new_capacity)) {
ASSERT(static_cast<uint32_t>(new_capacity) > index);
Object* obj = Heap::AllocateFixedArrayWithHoles(new_capacity);
if (obj->IsFailure()) return obj;
SetFastElements(FixedArray::cast(obj));
if (IsJSArray()) JSArray::cast(this)->set_length(Smi::FromInt(index + 1),
SKIP_WRITE_BARRIER);
FixedArray::cast(elements())->set(index, value);
return value;
}
}
// Otherwise default to slow case.
Object* obj = NormalizeElements();
if (obj->IsFailure()) return obj;
ASSERT(HasDictionaryElements());
return SetElement(index, value);
}
Object* JSObject::SetElement(uint32_t index, Object* value) {
// Check access rights if needed.
if (IsAccessCheckNeeded() &&
!Top::MayIndexedAccess(this, index, v8::ACCESS_SET)) {
Top::ReportFailedAccessCheck(this, v8::ACCESS_SET);
return value;
}
if (IsJSGlobalProxy()) {
Object* proto = GetPrototype();
if (proto->IsNull()) return value;
ASSERT(proto->IsJSGlobalObject());
return JSObject::cast(proto)->SetElement(index, value);
}
// Check for lookup interceptor
if (HasIndexedInterceptor()) {
return SetElementWithInterceptor(index, value);
}
return SetElementWithoutInterceptor(index, value);
}
Object* JSObject::SetElementWithoutInterceptor(uint32_t index, Object* value) {
switch (GetElementsKind()) {
case FAST_ELEMENTS:
// Fast case.
return SetFastElement(index, value);
case PIXEL_ELEMENTS: {
PixelArray* pixels = PixelArray::cast(elements());
return pixels->SetValue(index, value);
}
case DICTIONARY_ELEMENTS: {
// Insert element in the dictionary.
FixedArray* elms = FixedArray::cast(elements());
NumberDictionary* dictionary = NumberDictionary::cast(elms);
int entry = dictionary->FindEntry(index);
if (entry != NumberDictionary::kNotFound) {
Object* element = dictionary->ValueAt(entry);
PropertyDetails details = dictionary->DetailsAt(entry);
if (details.type() == CALLBACKS) {
// Only accessors allowed as elements.
FixedArray* structure = FixedArray::cast(element);
if (structure->get(kSetterIndex)->IsJSFunction()) {
JSFunction* setter = JSFunction::cast(structure->get(kSetterIndex));
return SetPropertyWithDefinedSetter(setter, value);
} else {
Handle<Object> self(this);
Handle<Object> key(Factory::NewNumberFromUint(index));
Handle<Object> args[2] = { key, self };
return Top::Throw(*Factory::NewTypeError("no_setter_in_callback",
HandleVector(args, 2)));
}
} else {
dictionary->UpdateMaxNumberKey(index);
dictionary->ValueAtPut(entry, value);
}
} else {
// Index not already used. Look for an accessor in the prototype chain.
if (!IsJSArray()) {
Object* setter = LookupCallbackSetterInPrototypes(index);
if (setter->IsJSFunction()) {
return SetPropertyWithDefinedSetter(JSFunction::cast(setter),
value);
}
}
Object* result = dictionary->AtNumberPut(index, value);
if (result->IsFailure()) return result;
if (elms != FixedArray::cast(result)) {
set_elements(FixedArray::cast(result));
}
}
// Update the array length if this JSObject is an array.
if (IsJSArray()) {
JSArray* array = JSArray::cast(this);
Object* return_value = array->JSArrayUpdateLengthFromIndex(index,
value);
if (return_value->IsFailure()) return return_value;
}
// Attempt to put this object back in fast case.
if (ShouldConvertToFastElements()) {
uint32_t new_length = 0;
if (IsJSArray()) {
CHECK(Array::IndexFromObject(JSArray::cast(this)->length(),
&new_length));
JSArray::cast(this)->set_length(Smi::FromInt(new_length));
} else {
new_length = NumberDictionary::cast(elements())->max_number_key() + 1;
}
Object* obj = Heap::AllocateFixedArrayWithHoles(new_length);
if (obj->IsFailure()) return obj;
SetFastElements(FixedArray::cast(obj));
#ifdef DEBUG
if (FLAG_trace_normalization) {
PrintF("Object elements are fast case again:\n");
Print();
}
#endif
}
return value;
}
default:
UNREACHABLE();
break;
}
// All possible cases have been handled above. Add a return to avoid the
// complaints from the compiler.
UNREACHABLE();
return Heap::null_value();
}
Object* JSArray::JSArrayUpdateLengthFromIndex(uint32_t index, Object* value) {
uint32_t old_len = 0;
CHECK(Array::IndexFromObject(length(), &old_len));
// Check to see if we need to update the length. For now, we make
// sure that the length stays within 32-bits (unsigned).
if (index >= old_len && index != 0xffffffff) {
Object* len =
Heap::NumberFromDouble(static_cast<double>(index) + 1);
if (len->IsFailure()) return len;
set_length(len);
}
return value;
}
Object* JSObject::GetElementPostInterceptor(JSObject* receiver,
uint32_t index) {
// Get element works for both JSObject and JSArray since
// JSArray::length cannot change.
switch (GetElementsKind()) {
case FAST_ELEMENTS: {
FixedArray* elms = FixedArray::cast(elements());
if (index < static_cast<uint32_t>(elms->length())) {
Object* value = elms->get(index);
if (!value->IsTheHole()) return value;
}
break;
}
case PIXEL_ELEMENTS: {
// TODO(iposva): Add testcase and implement.
UNIMPLEMENTED();
break;
}
case DICTIONARY_ELEMENTS: {
NumberDictionary* dictionary = element_dictionary();
int entry = dictionary->FindEntry(index);
if (entry != NumberDictionary::kNotFound) {
Object* element = dictionary->ValueAt(entry);
PropertyDetails details = dictionary->DetailsAt(entry);
if (details.type() == CALLBACKS) {
// Only accessors allowed as elements.
FixedArray* structure = FixedArray::cast(element);
Object* getter = structure->get(kGetterIndex);
if (getter->IsJSFunction()) {
return GetPropertyWithDefinedGetter(receiver,
JSFunction::cast(getter));
} else {
// Getter is not a function.
return Heap::undefined_value();
}
}
return element;
}
break;
}
default:
UNREACHABLE();
break;
}
// Continue searching via the prototype chain.
Object* pt = GetPrototype();
if (pt == Heap::null_value()) return Heap::undefined_value();
return pt->GetElementWithReceiver(receiver, index);
}
Object* JSObject::GetElementWithInterceptor(JSObject* receiver,
uint32_t index) {
// Make sure that the top context does not change when doing
// callbacks or interceptor calls.
AssertNoContextChange ncc;
HandleScope scope;
Handle<InterceptorInfo> interceptor(GetIndexedInterceptor());
Handle<JSObject> this_handle(receiver);
Handle<JSObject> holder_handle(this);
if (!interceptor->getter()->IsUndefined()) {
v8::IndexedPropertyGetter getter =
v8::ToCData<v8::IndexedPropertyGetter>(interceptor->getter());
LOG(ApiIndexedPropertyAccess("interceptor-indexed-get", this, index));
CustomArguments args(interceptor->data(), receiver, this);
v8::AccessorInfo info(args.end());
v8::Handle<v8::Value> result;
{
// Leaving JavaScript.
VMState state(EXTERNAL);
result = getter(index, info);
}
RETURN_IF_SCHEDULED_EXCEPTION();
if (!result.IsEmpty()) return *v8::Utils::OpenHandle(*result);
}
Object* raw_result =
holder_handle->GetElementPostInterceptor(*this_handle, index);
RETURN_IF_SCHEDULED_EXCEPTION();
return raw_result;
}
Object* JSObject::GetElementWithReceiver(JSObject* receiver, uint32_t index) {
// Check access rights if needed.
if (IsAccessCheckNeeded() &&
!Top::MayIndexedAccess(this, index, v8::ACCESS_GET)) {
Top::ReportFailedAccessCheck(this, v8::ACCESS_GET);
return Heap::undefined_value();
}
if (HasIndexedInterceptor()) {
return GetElementWithInterceptor(receiver, index);
}
// Get element works for both JSObject and JSArray since
// JSArray::length cannot change.
switch (GetElementsKind()) {
case FAST_ELEMENTS: {
FixedArray* elms = FixedArray::cast(elements());
if (index < static_cast<uint32_t>(elms->length())) {
Object* value = elms->get(index);
if (!value->IsTheHole()) return value;
}
break;
}
case PIXEL_ELEMENTS: {
PixelArray* pixels = PixelArray::cast(elements());
if (index < static_cast<uint32_t>(pixels->length())) {
uint8_t value = pixels->get(index);
return Smi::FromInt(value);
}
break;
}
case DICTIONARY_ELEMENTS: {
NumberDictionary* dictionary = element_dictionary();
int entry = dictionary->FindEntry(index);
if (entry != NumberDictionary::kNotFound) {
Object* element = dictionary->ValueAt(entry);
PropertyDetails details = dictionary->DetailsAt(entry);
if (details.type() == CALLBACKS) {
// Only accessors allowed as elements.
FixedArray* structure = FixedArray::cast(element);
Object* getter = structure->get(kGetterIndex);
if (getter->IsJSFunction()) {
return GetPropertyWithDefinedGetter(receiver,
JSFunction::cast(getter));
} else {
// Getter is not a function.
return Heap::undefined_value();
}
}
return element;
}
break;
}
}
Object* pt = GetPrototype();
if (pt == Heap::null_value()) return Heap::undefined_value();
return pt->GetElementWithReceiver(receiver, index);
}
bool JSObject::HasDenseElements() {
int capacity = 0;
int number_of_elements = 0;
switch (GetElementsKind()) {
case FAST_ELEMENTS: {
FixedArray* elms = FixedArray::cast(elements());
capacity = elms->length();
for (int i = 0; i < capacity; i++) {
if (!elms->get(i)->IsTheHole()) number_of_elements++;
}
break;
}
case PIXEL_ELEMENTS: {
return true;
}
case DICTIONARY_ELEMENTS: {
NumberDictionary* dictionary = NumberDictionary::cast(elements());
capacity = dictionary->Capacity();
number_of_elements = dictionary->NumberOfElements();
break;
}
default:
UNREACHABLE();
break;
}
if (capacity == 0) return true;
return (number_of_elements > (capacity / 2));
}
bool JSObject::ShouldConvertToSlowElements(int new_capacity) {
ASSERT(HasFastElements());
// Keep the array in fast case if the current backing storage is
// almost filled and if the new capacity is no more than twice the
// old capacity.
int elements_length = FixedArray::cast(elements())->length();
return !HasDenseElements() || ((new_capacity / 2) > elements_length);
}
bool JSObject::ShouldConvertToFastElements() {
ASSERT(HasDictionaryElements());
NumberDictionary* dictionary = NumberDictionary::cast(elements());
// If the elements are sparse, we should not go back to fast case.
if (!HasDenseElements()) return false;
// If an element has been added at a very high index in the elements
// dictionary, we cannot go back to fast case.
if (dictionary->requires_slow_elements()) return false;
// An object requiring access checks is never allowed to have fast
// elements. If it had fast elements we would skip security checks.
if (IsAccessCheckNeeded()) return false;
// If the dictionary backing storage takes up roughly half as much
// space as a fast-case backing storage would the array should have
// fast elements.
uint32_t length = 0;
if (IsJSArray()) {
CHECK(Array::IndexFromObject(JSArray::cast(this)->length(), &length));
} else {
length = dictionary->max_number_key();
}
return static_cast<uint32_t>(dictionary->Capacity()) >=
(length / (2 * NumberDictionary::kEntrySize));
}
// Certain compilers request function template instantiation when they
// see the definition of the other template functions in the
// class. This requires us to have the template functions put
// together, so even though this function belongs in objects-debug.cc,
// we keep it here instead to satisfy certain compilers.
#ifdef DEBUG
template<typename Shape, typename Key>
void Dictionary<Shape, Key>::Print() {
int capacity = HashTable<Shape, Key>::Capacity();
for (int i = 0; i < capacity; i++) {
Object* k = HashTable<Shape, Key>::KeyAt(i);
if (HashTable<Shape, Key>::IsKey(k)) {
PrintF(" ");
if (k->IsString()) {
String::cast(k)->StringPrint();
} else {
k->ShortPrint();
}
PrintF(": ");
ValueAt(i)->ShortPrint();
PrintF("\n");
}
}
}
#endif
template<typename Shape, typename Key>
void Dictionary<Shape, Key>::CopyValuesTo(FixedArray* elements) {
int pos = 0;
int capacity = HashTable<Shape, Key>::Capacity();
WriteBarrierMode mode = elements->GetWriteBarrierMode();
for (int i = 0; i < capacity; i++) {
Object* k = Dictionary<Shape, Key>::KeyAt(i);
if (Dictionary<Shape, Key>::IsKey(k)) {
elements->set(pos++, ValueAt(i), mode);
}
}
ASSERT(pos == elements->length());
}
InterceptorInfo* JSObject::GetNamedInterceptor() {
ASSERT(map()->has_named_interceptor());
JSFunction* constructor = JSFunction::cast(map()->constructor());
Object* template_info = constructor->shared()->function_data();
Object* result =
FunctionTemplateInfo::cast(template_info)->named_property_handler();
return InterceptorInfo::cast(result);
}
InterceptorInfo* JSObject::GetIndexedInterceptor() {
ASSERT(map()->has_indexed_interceptor());
JSFunction* constructor = JSFunction::cast(map()->constructor());
Object* template_info = constructor->shared()->function_data();
Object* result =
FunctionTemplateInfo::cast(template_info)->indexed_property_handler();
return InterceptorInfo::cast(result);
}
Object* JSObject::GetPropertyPostInterceptor(JSObject* receiver,
String* name,
PropertyAttributes* attributes) {
// Check local property in holder, ignore interceptor.
LookupResult result;
LocalLookupRealNamedProperty(name, &result);
if (result.IsValid()) return GetProperty(receiver, &result, name, attributes);
// Continue searching via the prototype chain.
Object* pt = GetPrototype();
*attributes = ABSENT;
if (pt == Heap::null_value()) return Heap::undefined_value();
return pt->GetPropertyWithReceiver(receiver, name, attributes);
}
Object* JSObject::GetPropertyWithInterceptor(
JSObject* receiver,
String* name,
PropertyAttributes* attributes) {
InterceptorInfo* interceptor = GetNamedInterceptor();
HandleScope scope;
Handle<JSObject> receiver_handle(receiver);
Handle<JSObject> holder_handle(this);
Handle<String> name_handle(name);
if (!interceptor->getter()->IsUndefined()) {
v8::NamedPropertyGetter getter =
v8::ToCData<v8::NamedPropertyGetter>(interceptor->getter());
LOG(ApiNamedPropertyAccess("interceptor-named-get", *holder_handle, name));
CustomArguments args(interceptor->data(), receiver, this);
v8::AccessorInfo info(args.end());
v8::Handle<v8::Value> result;
{
// Leaving JavaScript.
VMState state(EXTERNAL);
result = getter(v8::Utils::ToLocal(name_handle), info);
}
RETURN_IF_SCHEDULED_EXCEPTION();
if (!result.IsEmpty()) {
*attributes = NONE;
return *v8::Utils::OpenHandle(*result);
}
}
Object* result = holder_handle->GetPropertyPostInterceptor(
*receiver_handle,
*name_handle,
attributes);
RETURN_IF_SCHEDULED_EXCEPTION();
return result;
}
bool JSObject::HasRealNamedProperty(String* key) {
// Check access rights if needed.
if (IsAccessCheckNeeded() &&
!Top::MayNamedAccess(this, key, v8::ACCESS_HAS)) {
Top::ReportFailedAccessCheck(this, v8::ACCESS_HAS);
return false;
}
LookupResult result;
LocalLookupRealNamedProperty(key, &result);
if (result.IsValid()) {
switch (result.type()) {
case NORMAL: // fall through.
case FIELD: // fall through.
case CALLBACKS: // fall through.
case CONSTANT_FUNCTION:
return true;
case INTERCEPTOR:
case MAP_TRANSITION:
case CONSTANT_TRANSITION:
case NULL_DESCRIPTOR:
return false;
default:
UNREACHABLE();
}
}
return false;
}
bool JSObject::HasRealElementProperty(uint32_t index) {
// Check access rights if needed.
if (IsAccessCheckNeeded() &&
!Top::MayIndexedAccess(this, index, v8::ACCESS_HAS)) {
Top::ReportFailedAccessCheck(this, v8::ACCESS_HAS);
return false;
}
// Handle [] on String objects.
if (this->IsStringObjectWithCharacterAt(index)) return true;
switch (GetElementsKind()) {
case FAST_ELEMENTS: {
uint32_t length = IsJSArray() ?
static_cast<uint32_t>(
Smi::cast(JSArray::cast(this)->length())->value()) :
static_cast<uint32_t>(FixedArray::cast(elements())->length());
return (index < length) &&
!FixedArray::cast(elements())->get(index)->IsTheHole();
}
case PIXEL_ELEMENTS: {
PixelArray* pixels = PixelArray::cast(elements());
return index < static_cast<uint32_t>(pixels->length());
}
case DICTIONARY_ELEMENTS: {
return element_dictionary()->FindEntry(index)
!= NumberDictionary::kNotFound;
}
default:
UNREACHABLE();
break;
}
// All possibilities have been handled above already.
UNREACHABLE();
return Heap::null_value();
}
bool JSObject::HasRealNamedCallbackProperty(String* key) {
// Check access rights if needed.
if (IsAccessCheckNeeded() &&
!Top::MayNamedAccess(this, key, v8::ACCESS_HAS)) {
Top::ReportFailedAccessCheck(this, v8::ACCESS_HAS);
return false;
}
LookupResult result;
LocalLookupRealNamedProperty(key, &result);
return result.IsValid() && (result.type() == CALLBACKS);
}
int JSObject::NumberOfLocalProperties(PropertyAttributes filter) {
if (HasFastProperties()) {
DescriptorArray* descs = map()->instance_descriptors();
int result = 0;
for (int i = 0; i < descs->number_of_descriptors(); i++) {
PropertyDetails details = descs->GetDetails(i);
if (details.IsProperty() && (details.attributes() & filter) == 0) {
result++;
}
}
return result;
} else {
return property_dictionary()->NumberOfElementsFilterAttributes(filter);
}
}
int JSObject::NumberOfEnumProperties() {
return NumberOfLocalProperties(static_cast<PropertyAttributes>(DONT_ENUM));
}
void FixedArray::SwapPairs(FixedArray* numbers, int i, int j) {
Object* temp = get(i);
set(i, get(j));
set(j, temp);
if (this != numbers) {
temp = numbers->get(i);
numbers->set(i, numbers->get(j));
numbers->set(j, temp);
}
}
static void InsertionSortPairs(FixedArray* content,
FixedArray* numbers,
int len) {
for (int i = 1; i < len; i++) {
int j = i;
while (j > 0 &&
(NumberToUint32(numbers->get(j - 1)) >
NumberToUint32(numbers->get(j)))) {
content->SwapPairs(numbers, j - 1, j);
j--;
}
}
}
void HeapSortPairs(FixedArray* content, FixedArray* numbers, int len) {
// In-place heap sort.
ASSERT(content->length() == numbers->length());
// Bottom-up max-heap construction.
for (int i = 1; i < len; ++i) {
int child_index = i;
while (child_index > 0) {
int parent_index = ((child_index + 1) >> 1) - 1;
uint32_t parent_value = NumberToUint32(numbers->get(parent_index));
uint32_t child_value = NumberToUint32(numbers->get(child_index));
if (parent_value < child_value) {
content->SwapPairs(numbers, parent_index, child_index);
} else {
break;
}
child_index = parent_index;
}
}
// Extract elements and create sorted array.
for (int i = len - 1; i > 0; --i) {
// Put max element at the back of the array.
content->SwapPairs(numbers, 0, i);
// Sift down the new top element.
int parent_index = 0;
while (true) {
int child_index = ((parent_index + 1) << 1) - 1;
if (child_index >= i) break;
uint32_t child1_value = NumberToUint32(numbers->get(child_index));
uint32_t child2_value = NumberToUint32(numbers->get(child_index + 1));
uint32_t parent_value = NumberToUint32(numbers->get(parent_index));
if (child_index + 1 >= i || child1_value > child2_value) {
if (parent_value > child1_value) break;
content->SwapPairs(numbers, parent_index, child_index);
parent_index = child_index;
} else {
if (parent_value > child2_value) break;
content->SwapPairs(numbers, parent_index, child_index + 1);
parent_index = child_index + 1;
}
}
}
}
// Sort this array and the numbers as pairs wrt. the (distinct) numbers.
void FixedArray::SortPairs(FixedArray* numbers, uint32_t len) {
ASSERT(this->length() == numbers->length());
// For small arrays, simply use insertion sort.
if (len <= 10) {
InsertionSortPairs(this, numbers, len);
return;
}
// Check the range of indices.
uint32_t min_index = NumberToUint32(numbers->get(0));
uint32_t max_index = min_index;
uint32_t i;
for (i = 1; i < len; i++) {
if (NumberToUint32(numbers->get(i)) < min_index) {
min_index = NumberToUint32(numbers->get(i));
} else if (NumberToUint32(numbers->get(i)) > max_index) {
max_index = NumberToUint32(numbers->get(i));
}
}
if (max_index - min_index + 1 == len) {
// Indices form a contiguous range, unless there are duplicates.
// Do an in-place linear time sort assuming distinct numbers, but
// avoid hanging in case they are not.
for (i = 0; i < len; i++) {
uint32_t p;
uint32_t j = 0;
// While the current element at i is not at its correct position p,
// swap the elements at these two positions.
while ((p = NumberToUint32(numbers->get(i)) - min_index) != i &&
j++ < len) {
SwapPairs(numbers, i, p);
}
}
} else {
HeapSortPairs(this, numbers, len);
return;
}
}
// Fill in the names of local properties into the supplied storage. The main
// purpose of this function is to provide reflection information for the object
// mirrors.
void JSObject::GetLocalPropertyNames(FixedArray* storage, int index) {
ASSERT(storage->length() >= (NumberOfLocalProperties(NONE) - index));
if (HasFastProperties()) {
DescriptorArray* descs = map()->instance_descriptors();
for (int i = 0; i < descs->number_of_descriptors(); i++) {
if (descs->IsProperty(i)) storage->set(index++, descs->GetKey(i));
}
ASSERT(storage->length() >= index);
} else {
property_dictionary()->CopyKeysTo(storage);
}
}
int JSObject::NumberOfLocalElements(PropertyAttributes filter) {
return GetLocalElementKeys(NULL, filter);
}
int JSObject::NumberOfEnumElements() {
return NumberOfLocalElements(static_cast<PropertyAttributes>(DONT_ENUM));
}
int JSObject::GetLocalElementKeys(FixedArray* storage,
PropertyAttributes filter) {
int counter = 0;
switch (GetElementsKind()) {
case FAST_ELEMENTS: {
int length = IsJSArray() ?
Smi::cast(JSArray::cast(this)->length())->value() :
FixedArray::cast(elements())->length();
for (int i = 0; i < length; i++) {
if (!FixedArray::cast(elements())->get(i)->IsTheHole()) {
if (storage != NULL) {
storage->set(counter, Smi::FromInt(i), SKIP_WRITE_BARRIER);
}
counter++;
}
}
ASSERT(!storage || storage->length() >= counter);
break;
}
case PIXEL_ELEMENTS: {
int length = PixelArray::cast(elements())->length();
while (counter < length) {
if (storage != NULL) {
storage->set(counter, Smi::FromInt(counter), SKIP_WRITE_BARRIER);
}
counter++;
}
ASSERT(!storage || storage->length() >= counter);
break;
}
case DICTIONARY_ELEMENTS: {
if (storage != NULL) {
element_dictionary()->CopyKeysTo(storage, filter);
}
counter = element_dictionary()->NumberOfElementsFilterAttributes(filter);
break;
}
default:
UNREACHABLE();
break;
}
if (this->IsJSValue()) {
Object* val = JSValue::cast(this)->value();
if (val->IsString()) {
String* str = String::cast(val);
if (storage) {
for (int i = 0; i < str->length(); i++) {
storage->set(counter + i, Smi::FromInt(i), SKIP_WRITE_BARRIER);
}
}
counter += str->length();
}
}
ASSERT(!storage || storage->length() == counter);
return counter;
}
int JSObject::GetEnumElementKeys(FixedArray* storage) {
return GetLocalElementKeys(storage,
static_cast<PropertyAttributes>(DONT_ENUM));
}
bool NumberDictionaryShape::IsMatch(uint32_t key, Object* other) {
ASSERT(other->IsNumber());
return key == static_cast<uint32_t>(other->Number());
}
uint32_t NumberDictionaryShape::Hash(uint32_t key) {
return ComputeIntegerHash(key);
}
uint32_t NumberDictionaryShape::HashForObject(uint32_t key, Object* other) {
ASSERT(other->IsNumber());
return ComputeIntegerHash(static_cast<uint32_t>(other->Number()));
}
Object* NumberDictionaryShape::AsObject(uint32_t key) {
return Heap::NumberFromUint32(key);
}
bool StringDictionaryShape::IsMatch(String* key, Object* other) {
// We know that all entries in a hash table had their hash keys created.
// Use that knowledge to have fast failure.
if (key->Hash() != String::cast(other)->Hash()) return false;
return key->Equals(String::cast(other));
}
uint32_t StringDictionaryShape::Hash(String* key) {
return key->Hash();
}
uint32_t StringDictionaryShape::HashForObject(String* key, Object* other) {
return String::cast(other)->Hash();
}
Object* StringDictionaryShape::AsObject(String* key) {
return key;
}
// StringKey simply carries a string object as key.
class StringKey : public HashTableKey {
public:
explicit StringKey(String* string) :
string_(string),
hash_(HashForObject(string)) { }
bool IsMatch(Object* string) {
// We know that all entries in a hash table had their hash keys created.
// Use that knowledge to have fast failure.
if (hash_ != HashForObject(string)) {
return false;
}
return string_->Equals(String::cast(string));
}
uint32_t Hash() { return hash_; }
uint32_t HashForObject(Object* other) { return String::cast(other)->Hash(); }
Object* AsObject() { return string_; }
String* string_;
uint32_t hash_;
};
// StringSharedKeys are used as keys in the eval cache.
class StringSharedKey : public HashTableKey {
public:
StringSharedKey(String* source, SharedFunctionInfo* shared)
: source_(source), shared_(shared) { }
bool IsMatch(Object* other) {
if (!other->IsFixedArray()) return false;
FixedArray* pair = FixedArray::cast(other);
SharedFunctionInfo* shared = SharedFunctionInfo::cast(pair->get(0));
if (shared != shared_) return false;
String* source = String::cast(pair->get(1));
return source->Equals(source_);
}
static uint32_t StringSharedHashHelper(String* source,
SharedFunctionInfo* shared) {
uint32_t hash = source->Hash();
if (shared->HasSourceCode()) {
// Instead of using the SharedFunctionInfo pointer in the hash
// code computation, we use a combination of the hash of the
// script source code and the start and end positions. We do
// this to ensure that the cache entries can survive garbage
// collection.
Script* script = Script::cast(shared->script());
hash ^= String::cast(script->source())->Hash();
hash += shared->start_position();
}
return hash;
}
uint32_t Hash() {
return StringSharedHashHelper(source_, shared_);
}
uint32_t HashForObject(Object* obj) {
FixedArray* pair = FixedArray::cast(obj);
SharedFunctionInfo* shared = SharedFunctionInfo::cast(pair->get(0));
String* source = String::cast(pair->get(1));
return StringSharedHashHelper(source, shared);
}
Object* AsObject() {
Object* obj = Heap::AllocateFixedArray(2);
if (obj->IsFailure()) return obj;
FixedArray* pair = FixedArray::cast(obj);
pair->set(0, shared_);
pair->set(1, source_);
return pair;
}
private:
String* source_;
SharedFunctionInfo* shared_;
};
// RegExpKey carries the source and flags of a regular expression as key.
class RegExpKey : public HashTableKey {
public:
RegExpKey(String* string, JSRegExp::Flags flags)
: string_(string),
flags_(Smi::FromInt(flags.value())) { }
bool IsMatch(Object* obj) {
FixedArray* val = FixedArray::cast(obj);
return string_->Equals(String::cast(val->get(JSRegExp::kSourceIndex)))
&& (flags_ == val->get(JSRegExp::kFlagsIndex));
}
uint32_t Hash() { return RegExpHash(string_, flags_); }
Object* AsObject() {
// Plain hash maps, which is where regexp keys are used, don't
// use this function.
UNREACHABLE();
return NULL;
}
uint32_t HashForObject(Object* obj) {
FixedArray* val = FixedArray::cast(obj);
return RegExpHash(String::cast(val->get(JSRegExp::kSourceIndex)),
Smi::cast(val->get(JSRegExp::kFlagsIndex)));
}
static uint32_t RegExpHash(String* string, Smi* flags) {
return string->Hash() + flags->value();
}
String* string_;
Smi* flags_;
};
// Utf8SymbolKey carries a vector of chars as key.
class Utf8SymbolKey : public HashTableKey {
public:
explicit Utf8SymbolKey(Vector<const char> string)
: string_(string), length_field_(0) { }
bool IsMatch(Object* string) {
return String::cast(string)->IsEqualTo(string_);
}
uint32_t Hash() {
if (length_field_ != 0) return length_field_ >> String::kHashShift;
unibrow::Utf8InputBuffer<> buffer(string_.start(),
static_cast<unsigned>(string_.length()));
chars_ = buffer.Length();
length_field_ = String::ComputeLengthAndHashField(&buffer, chars_);
uint32_t result = length_field_ >> String::kHashShift;
ASSERT(result != 0); // Ensure that the hash value of 0 is never computed.
return result;
}
uint32_t HashForObject(Object* other) {
return String::cast(other)->Hash();
}
Object* AsObject() {
if (length_field_ == 0) Hash();
return Heap::AllocateSymbol(string_, chars_, length_field_);
}
Vector<const char> string_;
uint32_t length_field_;
int chars_; // Caches the number of characters when computing the hash code.
};
// SymbolKey carries a string/symbol object as key.
class SymbolKey : public HashTableKey {
public:
explicit SymbolKey(String* string) : string_(string) { }
bool IsMatch(Object* string) {
return String::cast(string)->Equals(string_);
}
uint32_t Hash() { return string_->Hash(); }
uint32_t HashForObject(Object* other) {
return String::cast(other)->Hash();
}
Object* AsObject() {
// If the string is a cons string, attempt to flatten it so that
// symbols will most often be flat strings.
if (StringShape(string_).IsCons()) {
ConsString* cons_string = ConsString::cast(string_);
cons_string->TryFlatten();
if (cons_string->second()->length() == 0) {
string_ = cons_string->first();
}
}
// Transform string to symbol if possible.
Map* map = Heap::SymbolMapForString(string_);
if (map != NULL) {
string_->set_map(map);
ASSERT(string_->IsSymbol());
return string_;
}
// Otherwise allocate a new symbol.
StringInputBuffer buffer(string_);
return Heap::AllocateInternalSymbol(&buffer,
string_->length(),
string_->length_field());
}
static uint32_t StringHash(Object* obj) {
return String::cast(obj)->Hash();
}
String* string_;
};
template<typename Shape, typename Key>
void HashTable<Shape, Key>::IteratePrefix(ObjectVisitor* v) {
IteratePointers(v, 0, kElementsStartOffset);
}
template<typename Shape, typename Key>
void HashTable<Shape, Key>::IterateElements(ObjectVisitor* v) {
IteratePointers(v,
kElementsStartOffset,
kHeaderSize + length() * kPointerSize);
}
template<typename Shape, typename Key>
Object* HashTable<Shape, Key>::Allocate(
int at_least_space_for) {
int capacity = RoundUpToPowerOf2(at_least_space_for);
if (capacity < 4) capacity = 4; // Guarantee min capacity.
Object* obj = Heap::AllocateHashTable(EntryToIndex(capacity));
if (!obj->IsFailure()) {
HashTable::cast(obj)->SetNumberOfElements(0);
HashTable::cast(obj)->SetCapacity(capacity);
}
return obj;
}
// Find entry for key otherwise return -1.
template<typename Shape, typename Key>
int HashTable<Shape, Key>::FindEntry(Key key) {
uint32_t nof = NumberOfElements();
if (nof == 0) return kNotFound; // Bail out if empty.
uint32_t capacity = Capacity();
uint32_t hash = Shape::Hash(key);
uint32_t entry = GetProbe(hash, 0, capacity);
Object* element = KeyAt(entry);
uint32_t passed_elements = 0;
if (!element->IsNull()) {
if (!element->IsUndefined() && Shape::IsMatch(key, element)) return entry;
if (++passed_elements == nof) return kNotFound;
}
for (uint32_t i = 1; !element->IsUndefined(); i++) {
entry = GetProbe(hash, i, capacity);
element = KeyAt(entry);
if (!element->IsNull()) {
if (!element->IsUndefined() && Shape::IsMatch(key, element)) return entry;
if (++passed_elements == nof) return kNotFound;
}
}
return kNotFound;
}
template<typename Shape, typename Key>
Object* HashTable<Shape, Key>::EnsureCapacity(int n, Key key) {
int capacity = Capacity();
int nof = NumberOfElements() + n;
// Make sure 50% is free
if (nof + (nof >> 1) <= capacity) return this;
Object* obj = Allocate(nof * 2);
if (obj->IsFailure()) return obj;
HashTable* table = HashTable::cast(obj);
WriteBarrierMode mode = table->GetWriteBarrierMode();
// Copy prefix to new array.
for (int i = kPrefixStartIndex;
i < kPrefixStartIndex + Shape::kPrefixSize;
i++) {
table->set(i, get(i), mode);
}
// Rehash the elements.
for (int i = 0; i < capacity; i++) {
uint32_t from_index = EntryToIndex(i);
Object* k = get(from_index);
if (IsKey(k)) {
uint32_t hash = Shape::HashForObject(key, k);
uint32_t insertion_index =
EntryToIndex(table->FindInsertionEntry(hash));
for (int j = 0; j < Shape::kEntrySize; j++) {
table->set(insertion_index + j, get(from_index + j), mode);
}
}
}
table->SetNumberOfElements(NumberOfElements());
return table;
}
template<typename Shape, typename Key>
uint32_t HashTable<Shape, Key>::FindInsertionEntry(uint32_t hash) {
uint32_t capacity = Capacity();
uint32_t entry = GetProbe(hash, 0, capacity);
Object* element = KeyAt(entry);
for (uint32_t i = 1; !(element->IsUndefined() || element->IsNull()); i++) {
entry = GetProbe(hash, i, capacity);
element = KeyAt(entry);
}
return entry;
}
// Force instantiation of template instances class.
// Please note this list is compiler dependent.
template class HashTable<SymbolTableShape, HashTableKey*>;
template class HashTable<CompilationCacheShape, HashTableKey*>;
template class HashTable<MapCacheShape, HashTableKey*>;
template class Dictionary<StringDictionaryShape, String*>;
template class Dictionary<NumberDictionaryShape, uint32_t>;
template Object* Dictionary<NumberDictionaryShape, uint32_t>::Allocate(
int);
template Object* Dictionary<StringDictionaryShape, String*>::Allocate(
int);
template Object* Dictionary<NumberDictionaryShape, uint32_t>::AtPut(
uint32_t, Object*);
template Object* Dictionary<NumberDictionaryShape, uint32_t>::SlowReverseLookup(
Object*);
template Object* Dictionary<StringDictionaryShape, String*>::SlowReverseLookup(
Object*);
template void Dictionary<NumberDictionaryShape, uint32_t>::CopyKeysTo(
FixedArray*, PropertyAttributes);
template Object* Dictionary<StringDictionaryShape, String*>::DeleteProperty(
int, JSObject::DeleteMode);
template Object* Dictionary<NumberDictionaryShape, uint32_t>::DeleteProperty(
int, JSObject::DeleteMode);
template void Dictionary<StringDictionaryShape, String*>::CopyKeysTo(
FixedArray*);
template int
Dictionary<StringDictionaryShape, String*>::NumberOfElementsFilterAttributes(
PropertyAttributes);
template Object* Dictionary<StringDictionaryShape, String*>::Add(
String*, Object*, PropertyDetails);
template Object*
Dictionary<StringDictionaryShape, String*>::GenerateNewEnumerationIndices();
template int
Dictionary<NumberDictionaryShape, uint32_t>::NumberOfElementsFilterAttributes(
PropertyAttributes);
template Object* Dictionary<NumberDictionaryShape, uint32_t>::Add(
uint32_t, Object*, PropertyDetails);
template Object* Dictionary<NumberDictionaryShape, uint32_t>::EnsureCapacity(
int, uint32_t);
template Object* Dictionary<StringDictionaryShape, String*>::EnsureCapacity(
int, String*);
template Object* Dictionary<NumberDictionaryShape, uint32_t>::AddEntry(
uint32_t, Object*, PropertyDetails, uint32_t);
template Object* Dictionary<StringDictionaryShape, String*>::AddEntry(
String*, Object*, PropertyDetails, uint32_t);
template
int Dictionary<NumberDictionaryShape, uint32_t>::NumberOfEnumElements();
template
int Dictionary<StringDictionaryShape, String*>::NumberOfEnumElements();
// Collates undefined and unexisting elements below limit from position
// zero of the elements. The object stays in Dictionary mode.
Object* JSObject::PrepareSlowElementsForSort(uint32_t limit) {
ASSERT(HasDictionaryElements());
// Must stay in dictionary mode, either because of requires_slow_elements,
// or because we are not going to sort (and therefore compact) all of the
// elements.
NumberDictionary* dict = element_dictionary();
HeapNumber* result_double = NULL;
if (limit > static_cast<uint32_t>(Smi::kMaxValue)) {
// Allocate space for result before we start mutating the object.
Object* new_double = Heap::AllocateHeapNumber(0.0);
if (new_double->IsFailure()) return new_double;
result_double = HeapNumber::cast(new_double);
}
int capacity = dict->Capacity();
Object* obj = NumberDictionary::Allocate(dict->Capacity());
if (obj->IsFailure()) return obj;
NumberDictionary* new_dict = NumberDictionary::cast(obj);
AssertNoAllocation no_alloc;
uint32_t pos = 0;
uint32_t undefs = 0;
for (int i = 0; i < capacity; i++) {
Object* k = dict->KeyAt(i);
if (dict->IsKey(k)) {
ASSERT(k->IsNumber());
ASSERT(!k->IsSmi() || Smi::cast(k)->value() >= 0);
ASSERT(!k->IsHeapNumber() || HeapNumber::cast(k)->value() >= 0);
ASSERT(!k->IsHeapNumber() || HeapNumber::cast(k)->value() <= kMaxUInt32);
Object* value = dict->ValueAt(i);
PropertyDetails details = dict->DetailsAt(i);
if (details.type() == CALLBACKS) {
// Bail out and do the sorting of undefineds and array holes in JS.
return Smi::FromInt(-1);
}
uint32_t key = NumberToUint32(k);
if (key < limit) {
if (value->IsUndefined()) {
undefs++;
} else {
new_dict->AddNumberEntry(pos, value, details);
pos++;
}
} else {
new_dict->AddNumberEntry(key, value, details);
}
}
}
uint32_t result = pos;
PropertyDetails no_details = PropertyDetails(NONE, NORMAL);
while (undefs > 0) {
new_dict->AddNumberEntry(pos, Heap::undefined_value(), no_details);
pos++;
undefs--;
}
set_elements(new_dict);
if (result <= static_cast<uint32_t>(Smi::kMaxValue)) {
return Smi::FromInt(static_cast<int>(result));
}
ASSERT_NE(NULL, result_double);
result_double->set_value(static_cast<double>(result));
return result_double;
}
// Collects all defined (non-hole) and non-undefined (array) elements at
// the start of the elements array.
// If the object is in dictionary mode, it is converted to fast elements
// mode.
Object* JSObject::PrepareElementsForSort(uint32_t limit) {
ASSERT(!HasPixelElements());
if (HasDictionaryElements()) {
// Convert to fast elements containing only the existing properties.
// Ordering is irrelevant, since we are going to sort anyway.
NumberDictionary* dict = element_dictionary();
if (IsJSArray() || dict->requires_slow_elements() ||
dict->max_number_key() >= limit) {
return PrepareSlowElementsForSort(limit);
}
// Convert to fast elements.
PretenureFlag tenure = Heap::InNewSpace(this) ? NOT_TENURED: TENURED;
Object* new_array =
Heap::AllocateFixedArray(dict->NumberOfElements(), tenure);
if (new_array->IsFailure()) {
return new_array;
}
FixedArray* fast_elements = FixedArray::cast(new_array);
dict->CopyValuesTo(fast_elements);
set_elements(fast_elements);
}
ASSERT(HasFastElements());
// Collect holes at the end, undefined before that and the rest at the
// start, and return the number of non-hole, non-undefined values.
FixedArray* elements = FixedArray::cast(this->elements());
uint32_t elements_length = static_cast<uint32_t>(elements->length());
if (limit > elements_length) {
limit = elements_length ;
}
if (limit == 0) {
return Smi::FromInt(0);
}
HeapNumber* result_double = NULL;
if (limit > static_cast<uint32_t>(Smi::kMaxValue)) {
// Pessimistically allocate space for return value before
// we start mutating the array.
Object* new_double = Heap::AllocateHeapNumber(0.0);
if (new_double->IsFailure()) return new_double;
result_double = HeapNumber::cast(new_double);
}
AssertNoAllocation no_alloc;
// Split elements into defined, undefined and the_hole, in that order.
// Only count locations for undefined and the hole, and fill them afterwards.
WriteBarrierMode write_barrier = elements->GetWriteBarrierMode();
unsigned int undefs = limit;
unsigned int holes = limit;
// Assume most arrays contain no holes and undefined values, so minimize the
// number of stores of non-undefined, non-the-hole values.
for (unsigned int i = 0; i < undefs; i++) {
Object* current = elements->get(i);
if (current->IsTheHole()) {
holes--;
undefs--;
} else if (current->IsUndefined()) {
undefs--;
} else {
continue;
}
// Position i needs to be filled.
while (undefs > i) {
current = elements->get(undefs);
if (current->IsTheHole()) {
holes--;
undefs--;
} else if (current->IsUndefined()) {
undefs--;
} else {
elements->set(i, current, write_barrier);
break;
}
}
}
uint32_t result = undefs;
while (undefs < holes) {
elements->set_undefined(undefs);
undefs++;
}
while (holes < limit) {
elements->set_the_hole(holes);
holes++;
}
if (result <= static_cast<uint32_t>(Smi::kMaxValue)) {
return Smi::FromInt(static_cast<int>(result));
}
ASSERT_NE(NULL, result_double);
result_double->set_value(static_cast<double>(result));
return result_double;
}
Object* PixelArray::SetValue(uint32_t index, Object* value) {
uint8_t clamped_value = 0;
if (index < static_cast<uint32_t>(length())) {
if (value->IsSmi()) {
int int_value = Smi::cast(value)->value();
if (int_value < 0) {
clamped_value = 0;
} else if (int_value > 255) {
clamped_value = 255;
} else {
clamped_value = static_cast<uint8_t>(int_value);
}
} else if (value->IsHeapNumber()) {
double double_value = HeapNumber::cast(value)->value();
if (!(double_value > 0)) {
// NaN and less than zero clamp to zero.
clamped_value = 0;
} else if (double_value > 255) {
// Greater than 255 clamp to 255.
clamped_value = 255;
} else {
// Other doubles are rounded to the nearest integer.
clamped_value = static_cast<uint8_t>(double_value + 0.5);
}
} else {
// Clamp undefined to zero (default). All other types have been
// converted to a number type further up in the call chain.
ASSERT(value->IsUndefined());
}
set(index, clamped_value);
}
return Smi::FromInt(clamped_value);
}
Object* GlobalObject::GetPropertyCell(LookupResult* result) {
ASSERT(!HasFastProperties());
Object* value = property_dictionary()->ValueAt(result->GetDictionaryEntry());
ASSERT(value->IsJSGlobalPropertyCell());
return value;
}
Object* GlobalObject::EnsurePropertyCell(String* name) {
ASSERT(!HasFastProperties());
int entry = property_dictionary()->FindEntry(name);
if (entry == StringDictionary::kNotFound) {
Object* cell = Heap::AllocateJSGlobalPropertyCell(Heap::the_hole_value());
if (cell->IsFailure()) return cell;
PropertyDetails details(NONE, NORMAL);
details = details.AsDeleted();
Object* dictionary = property_dictionary()->Add(name, cell, details);
if (dictionary->IsFailure()) return dictionary;
set_properties(StringDictionary::cast(dictionary));
return cell;
} else {
Object* value = property_dictionary()->ValueAt(entry);
ASSERT(value->IsJSGlobalPropertyCell());
return value;
}
}
Object* SymbolTable::LookupString(String* string, Object** s) {
SymbolKey key(string);
return LookupKey(&key, s);
}
bool SymbolTable::LookupSymbolIfExists(String* string, String** symbol) {
SymbolKey key(string);
int entry = FindEntry(&key);
if (entry == kNotFound) {
return false;
} else {
String* result = String::cast(KeyAt(entry));
ASSERT(StringShape(result).IsSymbol());
*symbol = result;
return true;
}
}
Object* SymbolTable::LookupSymbol(Vector<const char> str, Object** s) {
Utf8SymbolKey key(str);
return LookupKey(&key, s);
}
Object* SymbolTable::LookupKey(HashTableKey* key, Object** s) {
int entry = FindEntry(key);
// Symbol already in table.
if (entry != kNotFound) {
*s = KeyAt(entry);
return this;
}
// Adding new symbol. Grow table if needed.
Object* obj = EnsureCapacity(1, key);
if (obj->IsFailure()) return obj;
// Create symbol object.
Object* symbol = key->AsObject();
if (symbol->IsFailure()) return symbol;
// If the symbol table grew as part of EnsureCapacity, obj is not
// the current symbol table and therefore we cannot use
// SymbolTable::cast here.
SymbolTable* table = reinterpret_cast<SymbolTable*>(obj);
// Add the new symbol and return it along with the symbol table.
entry = table->FindInsertionEntry(key->Hash());
table->set(EntryToIndex(entry), symbol);
table->ElementAdded();
*s = symbol;
return table;
}
Object* CompilationCacheTable::Lookup(String* src) {
StringKey key(src);
int entry = FindEntry(&key);
if (entry == kNotFound) return Heap::undefined_value();
return get(EntryToIndex(entry) + 1);
}
Object* CompilationCacheTable::LookupEval(String* src, Context* context) {
StringSharedKey key(src, context->closure()->shared());
int entry = FindEntry(&key);
if (entry == kNotFound) return Heap::undefined_value();
return get(EntryToIndex(entry) + 1);
}
Object* CompilationCacheTable::LookupRegExp(String* src,
JSRegExp::Flags flags) {
RegExpKey key(src, flags);
int entry = FindEntry(&key);
if (entry == kNotFound) return Heap::undefined_value();
return get(EntryToIndex(entry) + 1);
}
Object* CompilationCacheTable::Put(String* src, Object* value) {
StringKey key(src);
Object* obj = EnsureCapacity(1, &key);
if (obj->IsFailure()) return obj;
CompilationCacheTable* cache =
reinterpret_cast<CompilationCacheTable*>(obj);
int entry = cache->FindInsertionEntry(key.Hash());
cache->set(EntryToIndex(entry), src);
cache->set(EntryToIndex(entry) + 1, value);
cache->ElementAdded();
return cache;
}
Object* CompilationCacheTable::PutEval(String* src,
Context* context,
Object* value) {
StringSharedKey key(src, context->closure()->shared());
Object* obj = EnsureCapacity(1, &key);
if (obj->IsFailure()) return obj;
CompilationCacheTable* cache =
reinterpret_cast<CompilationCacheTable*>(obj);
int entry = cache->FindInsertionEntry(key.Hash());
Object* k = key.AsObject();
if (k->IsFailure()) return k;
cache->set(EntryToIndex(entry), k);
cache->set(EntryToIndex(entry) + 1, value);
cache->ElementAdded();
return cache;
}
Object* CompilationCacheTable::PutRegExp(String* src,
JSRegExp::Flags flags,
FixedArray* value) {
RegExpKey key(src, flags);
Object* obj = EnsureCapacity(1, &key);
if (obj->IsFailure()) return obj;
CompilationCacheTable* cache =
reinterpret_cast<CompilationCacheTable*>(obj);
int entry = cache->FindInsertionEntry(key.Hash());
cache->set(EntryToIndex(entry), value);
cache->set(EntryToIndex(entry) + 1, value);
cache->ElementAdded();
return cache;
}
// SymbolsKey used for HashTable where key is array of symbols.
class SymbolsKey : public HashTableKey {
public:
explicit SymbolsKey(FixedArray* symbols) : symbols_(symbols) { }
bool IsMatch(Object* symbols) {
FixedArray* o = FixedArray::cast(symbols);
int len = symbols_->length();
if (o->length() != len) return false;
for (int i = 0; i < len; i++) {
if (o->get(i) != symbols_->get(i)) return false;
}
return true;
}
uint32_t Hash() { return HashForObject(symbols_); }
uint32_t HashForObject(Object* obj) {
FixedArray* symbols = FixedArray::cast(obj);
int len = symbols->length();
uint32_t hash = 0;
for (int i = 0; i < len; i++) {
hash ^= String::cast(symbols->get(i))->Hash();
}
return hash;
}
Object* AsObject() { return symbols_; }
private:
FixedArray* symbols_;
};
Object* MapCache::Lookup(FixedArray* array) {
SymbolsKey key(array);
int entry = FindEntry(&key);
if (entry == kNotFound) return Heap::undefined_value();
return get(EntryToIndex(entry) + 1);
}
Object* MapCache::Put(FixedArray* array, Map* value) {
SymbolsKey key(array);
Object* obj = EnsureCapacity(1, &key);
if (obj->IsFailure()) return obj;
MapCache* cache = reinterpret_cast<MapCache*>(obj);
int entry = cache->FindInsertionEntry(key.Hash());
cache->set(EntryToIndex(entry), array);
cache->set(EntryToIndex(entry) + 1, value);
cache->ElementAdded();
return cache;
}
template<typename Shape, typename Key>
Object* Dictionary<Shape, Key>::Allocate(int at_least_space_for) {
Object* obj = HashTable<Shape, Key>::Allocate(at_least_space_for);
// Initialize the next enumeration index.
if (!obj->IsFailure()) {
Dictionary<Shape, Key>::cast(obj)->
SetNextEnumerationIndex(PropertyDetails::kInitialIndex);
}
return obj;
}
template<typename Shape, typename Key>
Object* Dictionary<Shape, Key>::GenerateNewEnumerationIndices() {
int length = HashTable<Shape, Key>::NumberOfElements();
// Allocate and initialize iteration order array.
Object* obj = Heap::AllocateFixedArray(length);
if (obj->IsFailure()) return obj;
FixedArray* iteration_order = FixedArray::cast(obj);
for (int i = 0; i < length; i++) {
iteration_order->set(i, Smi::FromInt(i), SKIP_WRITE_BARRIER);
}
// Allocate array with enumeration order.
obj = Heap::AllocateFixedArray(length);
if (obj->IsFailure()) return obj;
FixedArray* enumeration_order = FixedArray::cast(obj);
// Fill the enumeration order array with property details.
int capacity = HashTable<Shape, Key>::Capacity();
int pos = 0;
for (int i = 0; i < capacity; i++) {
if (Dictionary<Shape, Key>::IsKey(Dictionary<Shape, Key>::KeyAt(i))) {
enumeration_order->set(pos++,
Smi::FromInt(DetailsAt(i).index()),
SKIP_WRITE_BARRIER);
}
}
// Sort the arrays wrt. enumeration order.
iteration_order->SortPairs(enumeration_order, enumeration_order->length());
// Overwrite the enumeration_order with the enumeration indices.
for (int i = 0; i < length; i++) {
int index = Smi::cast(iteration_order->get(i))->value();
int enum_index = PropertyDetails::kInitialIndex + i;
enumeration_order->set(index,
Smi::FromInt(enum_index),
SKIP_WRITE_BARRIER);
}
// Update the dictionary with new indices.
capacity = HashTable<Shape, Key>::Capacity();
pos = 0;
for (int i = 0; i < capacity; i++) {
if (Dictionary<Shape, Key>::IsKey(Dictionary<Shape, Key>::KeyAt(i))) {
int enum_index = Smi::cast(enumeration_order->get(pos++))->value();
PropertyDetails details = DetailsAt(i);
PropertyDetails new_details =
PropertyDetails(details.attributes(), details.type(), enum_index);
DetailsAtPut(i, new_details);
}
}
// Set the next enumeration index.
SetNextEnumerationIndex(PropertyDetails::kInitialIndex+length);
return this;
}
template<typename Shape, typename Key>
Object* Dictionary<Shape, Key>::EnsureCapacity(int n, Key key) {
// Check whether there are enough enumeration indices to add n elements.
if (Shape::kIsEnumerable &&
!PropertyDetails::IsValidIndex(NextEnumerationIndex() + n)) {
// If not, we generate new indices for the properties.
Object* result = GenerateNewEnumerationIndices();
if (result->IsFailure()) return result;
}
return HashTable<Shape, Key>::EnsureCapacity(n, key);
}
void NumberDictionary::RemoveNumberEntries(uint32_t from, uint32_t to) {
// Do nothing if the interval [from, to) is empty.
if (from >= to) return;
int removed_entries = 0;
Object* sentinel = Heap::null_value();
int capacity = Capacity();
for (int i = 0; i < capacity; i++) {
Object* key = KeyAt(i);
if (key->IsNumber()) {
uint32_t number = static_cast<uint32_t>(key->Number());
if (from <= number && number < to) {
SetEntry(i, sentinel, sentinel, Smi::FromInt(0));
removed_entries++;
}
}
}
// Update the number of elements.
SetNumberOfElements(NumberOfElements() - removed_entries);
}
template<typename Shape, typename Key>
Object* Dictionary<Shape, Key>::DeleteProperty(int entry,
JSObject::DeleteMode mode) {
PropertyDetails details = DetailsAt(entry);
// Ignore attributes if forcing a deletion.
if (details.IsDontDelete() && mode == JSObject::NORMAL_DELETION) {
return Heap::false_value();
}
SetEntry(entry, Heap::null_value(), Heap::null_value(), Smi::FromInt(0));
HashTable<Shape, Key>::ElementRemoved();
return Heap::true_value();
}
template<typename Shape, typename Key>
Object* Dictionary<Shape, Key>::AtPut(Key key, Object* value) {
int entry = FindEntry(key);
// If the entry is present set the value;
if (entry != Dictionary<Shape, Key>::kNotFound) {
ValueAtPut(entry, value);
return this;
}
// Check whether the dictionary should be extended.
Object* obj = EnsureCapacity(1, key);
if (obj->IsFailure()) return obj;
Object* k = Shape::AsObject(key);
if (k->IsFailure()) return k;
PropertyDetails details = PropertyDetails(NONE, NORMAL);
return Dictionary<Shape, Key>::cast(obj)->
AddEntry(key, value, details, Shape::Hash(key));
}
template<typename Shape, typename Key>
Object* Dictionary<Shape, Key>::Add(Key key,
Object* value,
PropertyDetails details) {
// Valdate key is absent.
SLOW_ASSERT((FindEntry(key) == Dictionary<Shape, Key>::kNotFound));
// Check whether the dictionary should be extended.
Object* obj = EnsureCapacity(1, key);
if (obj->IsFailure()) return obj;
return Dictionary<Shape, Key>::cast(obj)->
AddEntry(key, value, details, Shape::Hash(key));
}
// Add a key, value pair to the dictionary.
template<typename Shape, typename Key>
Object* Dictionary<Shape, Key>::AddEntry(Key key,
Object* value,
PropertyDetails details,
uint32_t hash) {
// Compute the key object.
Object* k = Shape::AsObject(key);
if (k->IsFailure()) return k;
uint32_t entry = Dictionary<Shape, Key>::FindInsertionEntry(hash);
// Insert element at empty or deleted entry
if (!details.IsDeleted() && details.index() == 0 && Shape::kIsEnumerable) {
// Assign an enumeration index to the property and update
// SetNextEnumerationIndex.
int index = NextEnumerationIndex();
details = PropertyDetails(details.attributes(), details.type(), index);
SetNextEnumerationIndex(index + 1);
}
SetEntry(entry, k, value, details);
ASSERT((Dictionary<Shape, Key>::KeyAt(entry)->IsNumber()
|| Dictionary<Shape, Key>::KeyAt(entry)->IsString()));
HashTable<Shape, Key>::ElementAdded();
return this;
}
void NumberDictionary::UpdateMaxNumberKey(uint32_t key) {
// If the dictionary requires slow elements an element has already
// been added at a high index.
if (requires_slow_elements()) return;
// Check if this index is high enough that we should require slow
// elements.
if (key > kRequiresSlowElementsLimit) {
set_requires_slow_elements();
return;
}
// Update max key value.
Object* max_index_object = get(kMaxNumberKeyIndex);
if (!max_index_object->IsSmi() || max_number_key() < key) {
FixedArray::set(kMaxNumberKeyIndex,
Smi::FromInt(key << kRequiresSlowElementsTagSize),
SKIP_WRITE_BARRIER);
}
}
Object* NumberDictionary::AddNumberEntry(uint32_t key,
Object* value,
PropertyDetails details) {
UpdateMaxNumberKey(key);
SLOW_ASSERT(FindEntry(key) == kNotFound);
return Add(key, value, details);
}
Object* NumberDictionary::AtNumberPut(uint32_t key, Object* value) {
UpdateMaxNumberKey(key);
return AtPut(key, value);
}
Object* NumberDictionary::Set(uint32_t key,
Object* value,
PropertyDetails details) {
int entry = FindEntry(key);
if (entry == kNotFound) return AddNumberEntry(key, value, details);
// Preserve enumeration index.
details = PropertyDetails(details.attributes(),
details.type(),
DetailsAt(entry).index());
SetEntry(entry, NumberDictionaryShape::AsObject(key), value, details);
return this;
}
template<typename Shape, typename Key>
int Dictionary<Shape, Key>::NumberOfElementsFilterAttributes(
PropertyAttributes filter) {
int capacity = HashTable<Shape, Key>::Capacity();
int result = 0;
for (int i = 0; i < capacity; i++) {
Object* k = HashTable<Shape, Key>::KeyAt(i);
if (HashTable<Shape, Key>::IsKey(k)) {
PropertyDetails details = DetailsAt(i);
if (details.IsDeleted()) continue;
PropertyAttributes attr = details.attributes();
if ((attr & filter) == 0) result++;
}
}
return result;
}
template<typename Shape, typename Key>
int Dictionary<Shape, Key>::NumberOfEnumElements() {
return NumberOfElementsFilterAttributes(
static_cast<PropertyAttributes>(DONT_ENUM));
}
template<typename Shape, typename Key>
void Dictionary<Shape, Key>::CopyKeysTo(FixedArray* storage,
PropertyAttributes filter) {
ASSERT(storage->length() >= NumberOfEnumElements());
int capacity = HashTable<Shape, Key>::Capacity();
int index = 0;
for (int i = 0; i < capacity; i++) {
Object* k = HashTable<Shape, Key>::KeyAt(i);
if (HashTable<Shape, Key>::IsKey(k)) {
PropertyDetails details = DetailsAt(i);
if (details.IsDeleted()) continue;
PropertyAttributes attr = details.attributes();
if ((attr & filter) == 0) storage->set(index++, k);
}
}
storage->SortPairs(storage, index);
ASSERT(storage->length() >= index);
}
void StringDictionary::CopyEnumKeysTo(FixedArray* storage,
FixedArray* sort_array) {
ASSERT(storage->length() >= NumberOfEnumElements());
int capacity = Capacity();
int index = 0;
for (int i = 0; i < capacity; i++) {
Object* k = KeyAt(i);
if (IsKey(k)) {
PropertyDetails details = DetailsAt(i);
if (details.IsDeleted() || details.IsDontEnum()) continue;
storage->set(index, k);
sort_array->set(index,
Smi::FromInt(details.index()),
SKIP_WRITE_BARRIER);
index++;
}
}
storage->SortPairs(sort_array, sort_array->length());
ASSERT(storage->length() >= index);
}
template<typename Shape, typename Key>
void Dictionary<Shape, Key>::CopyKeysTo(FixedArray* storage) {
ASSERT(storage->length() >= NumberOfElementsFilterAttributes(
static_cast<PropertyAttributes>(NONE)));
int capacity = HashTable<Shape, Key>::Capacity();
int index = 0;
for (int i = 0; i < capacity; i++) {
Object* k = HashTable<Shape, Key>::KeyAt(i);
if (HashTable<Shape, Key>::IsKey(k)) {
PropertyDetails details = DetailsAt(i);
if (details.IsDeleted()) continue;
storage->set(index++, k);
}
}
ASSERT(storage->length() >= index);
}
// Backwards lookup (slow).
template<typename Shape, typename Key>
Object* Dictionary<Shape, Key>::SlowReverseLookup(Object* value) {
int capacity = HashTable<Shape, Key>::Capacity();
for (int i = 0; i < capacity; i++) {
Object* k = HashTable<Shape, Key>::KeyAt(i);
if (Dictionary<Shape, Key>::IsKey(k)) {
Object* e = ValueAt(i);
if (e->IsJSGlobalPropertyCell()) {
e = JSGlobalPropertyCell::cast(e)->value();
}
if (e == value) return k;
}
}
return Heap::undefined_value();
}
Object* StringDictionary::TransformPropertiesToFastFor(
JSObject* obj, int unused_property_fields) {
// Make sure we preserve dictionary representation if there are too many
// descriptors.
if (NumberOfElements() > DescriptorArray::kMaxNumberOfDescriptors) return obj;
// Figure out if it is necessary to generate new enumeration indices.
int max_enumeration_index =
NextEnumerationIndex() +
(DescriptorArray::kMaxNumberOfDescriptors -
NumberOfElements());
if (!PropertyDetails::IsValidIndex(max_enumeration_index)) {
Object* result = GenerateNewEnumerationIndices();
if (result->IsFailure()) return result;
}
int instance_descriptor_length = 0;
int number_of_fields = 0;
// Compute the length of the instance descriptor.
int capacity = Capacity();
for (int i = 0; i < capacity; i++) {
Object* k = KeyAt(i);
if (IsKey(k)) {
Object* value = ValueAt(i);
PropertyType type = DetailsAt(i).type();
ASSERT(type != FIELD);
instance_descriptor_length++;
if (type == NORMAL && !value->IsJSFunction()) number_of_fields += 1;
}
}
// Allocate the instance descriptor.
Object* descriptors_unchecked =
DescriptorArray::Allocate(instance_descriptor_length);
if (descriptors_unchecked->IsFailure()) return descriptors_unchecked;
DescriptorArray* descriptors = DescriptorArray::cast(descriptors_unchecked);
int inobject_props = obj->map()->inobject_properties();
int number_of_allocated_fields =
number_of_fields + unused_property_fields - inobject_props;
// Allocate the fixed array for the fields.
Object* fields = Heap::AllocateFixedArray(number_of_allocated_fields);
if (fields->IsFailure()) return fields;
// Fill in the instance descriptor and the fields.
int next_descriptor = 0;
int current_offset = 0;
for (int i = 0; i < capacity; i++) {
Object* k = KeyAt(i);
if (IsKey(k)) {
Object* value = ValueAt(i);
// Ensure the key is a symbol before writing into the instance descriptor.
Object* key = Heap::LookupSymbol(String::cast(k));
if (key->IsFailure()) return key;
PropertyDetails details = DetailsAt(i);
PropertyType type = details.type();
if (value->IsJSFunction()) {
ConstantFunctionDescriptor d(String::cast(key),
JSFunction::cast(value),
details.attributes(),
details.index());
descriptors->Set(next_descriptor++, &d);
} else if (type == NORMAL) {
if (current_offset < inobject_props) {
obj->InObjectPropertyAtPut(current_offset,
value,
UPDATE_WRITE_BARRIER);
} else {
int offset = current_offset - inobject_props;
FixedArray::cast(fields)->set(offset, value);
}
FieldDescriptor d(String::cast(key),
current_offset++,
details.attributes(),
details.index());
descriptors->Set(next_descriptor++, &d);
} else if (type == CALLBACKS) {
CallbacksDescriptor d(String::cast(key),
value,
details.attributes(),
details.index());
descriptors->Set(next_descriptor++, &d);
} else {
UNREACHABLE();
}
}
}
ASSERT(current_offset == number_of_fields);
descriptors->Sort();
// Allocate new map.
Object* new_map = obj->map()->CopyDropDescriptors();
if (new_map->IsFailure()) return new_map;
// Transform the object.
obj->set_map(Map::cast(new_map));
obj->map()->set_instance_descriptors(descriptors);
obj->map()->set_unused_property_fields(unused_property_fields);
obj->set_properties(FixedArray::cast(fields));
ASSERT(obj->IsJSObject());
descriptors->SetNextEnumerationIndex(NextEnumerationIndex());
// Check that it really works.
ASSERT(obj->HasFastProperties());
return obj;
}
#ifdef ENABLE_DEBUGGER_SUPPORT
// Check if there is a break point at this code position.
bool DebugInfo::HasBreakPoint(int code_position) {
// Get the break point info object for this code position.
Object* break_point_info = GetBreakPointInfo(code_position);
// If there is no break point info object or no break points in the break
// point info object there is no break point at this code position.
if (break_point_info->IsUndefined()) return false;
return BreakPointInfo::cast(break_point_info)->GetBreakPointCount() > 0;
}
// Get the break point info object for this code position.
Object* DebugInfo::GetBreakPointInfo(int code_position) {
// Find the index of the break point info object for this code position.
int index = GetBreakPointInfoIndex(code_position);
// Return the break point info object if any.
if (index == kNoBreakPointInfo) return Heap::undefined_value();
return BreakPointInfo::cast(break_points()->get(index));
}
// Clear a break point at the specified code position.
void DebugInfo::ClearBreakPoint(Handle<DebugInfo> debug_info,
int code_position,
Handle<Object> break_point_object) {
Handle<Object> break_point_info(debug_info->GetBreakPointInfo(code_position));
if (break_point_info->IsUndefined()) return;
BreakPointInfo::ClearBreakPoint(
Handle<BreakPointInfo>::cast(break_point_info),
break_point_object);
}
void DebugInfo::SetBreakPoint(Handle<DebugInfo> debug_info,
int code_position,
int source_position,
int statement_position,
Handle<Object> break_point_object) {
Handle<Object> break_point_info(debug_info->GetBreakPointInfo(code_position));
if (!break_point_info->IsUndefined()) {
BreakPointInfo::SetBreakPoint(
Handle<BreakPointInfo>::cast(break_point_info),
break_point_object);
return;
}
// Adding a new break point for a code position which did not have any
// break points before. Try to find a free slot.
int index = kNoBreakPointInfo;
for (int i = 0; i < debug_info->break_points()->length(); i++) {
if (debug_info->break_points()->get(i)->IsUndefined()) {
index = i;
break;
}
}
if (index == kNoBreakPointInfo) {
// No free slot - extend break point info array.
Handle<FixedArray> old_break_points =
Handle<FixedArray>(FixedArray::cast(debug_info->break_points()));
debug_info->set_break_points(*Factory::NewFixedArray(
old_break_points->length() +
Debug::kEstimatedNofBreakPointsInFunction));
Handle<FixedArray> new_break_points =
Handle<FixedArray>(FixedArray::cast(debug_info->break_points()));
for (int i = 0; i < old_break_points->length(); i++) {
new_break_points->set(i, old_break_points->get(i));
}
index = old_break_points->length();
}
ASSERT(index != kNoBreakPointInfo);
// Allocate new BreakPointInfo object and set the break point.
Handle<BreakPointInfo> new_break_point_info =
Handle<BreakPointInfo>::cast(Factory::NewStruct(BREAK_POINT_INFO_TYPE));
new_break_point_info->set_code_position(Smi::FromInt(code_position));
new_break_point_info->set_source_position(Smi::FromInt(source_position));
new_break_point_info->
set_statement_position(Smi::FromInt(statement_position));
new_break_point_info->set_break_point_objects(Heap::undefined_value());
BreakPointInfo::SetBreakPoint(new_break_point_info, break_point_object);
debug_info->break_points()->set(index, *new_break_point_info);
}
// Get the break point objects for a code position.
Object* DebugInfo::GetBreakPointObjects(int code_position) {
Object* break_point_info = GetBreakPointInfo(code_position);
if (break_point_info->IsUndefined()) {
return Heap::undefined_value();
}
return BreakPointInfo::cast(break_point_info)->break_point_objects();
}
// Get the total number of break points.
int DebugInfo::GetBreakPointCount() {
if (break_points()->IsUndefined()) return 0;
int count = 0;
for (int i = 0; i < break_points()->length(); i++) {
if (!break_points()->get(i)->IsUndefined()) {
BreakPointInfo* break_point_info =
BreakPointInfo::cast(break_points()->get(i));
count += break_point_info->GetBreakPointCount();
}
}
return count;
}
Object* DebugInfo::FindBreakPointInfo(Handle<DebugInfo> debug_info,
Handle<Object> break_point_object) {
if (debug_info->break_points()->IsUndefined()) return Heap::undefined_value();
for (int i = 0; i < debug_info->break_points()->length(); i++) {
if (!debug_info->break_points()->get(i)->IsUndefined()) {
Handle<BreakPointInfo> break_point_info =
Handle<BreakPointInfo>(BreakPointInfo::cast(
debug_info->break_points()->get(i)));
if (BreakPointInfo::HasBreakPointObject(break_point_info,
break_point_object)) {
return *break_point_info;
}
}
}
return Heap::undefined_value();
}
// Find the index of the break point info object for the specified code
// position.
int DebugInfo::GetBreakPointInfoIndex(int code_position) {
if (break_points()->IsUndefined()) return kNoBreakPointInfo;
for (int i = 0; i < break_points()->length(); i++) {
if (!break_points()->get(i)->IsUndefined()) {
BreakPointInfo* break_point_info =
BreakPointInfo::cast(break_points()->get(i));
if (break_point_info->code_position()->value() == code_position) {
return i;
}
}
}
return kNoBreakPointInfo;
}
// Remove the specified break point object.
void BreakPointInfo::ClearBreakPoint(Handle<BreakPointInfo> break_point_info,
Handle<Object> break_point_object) {
// If there are no break points just ignore.
if (break_point_info->break_point_objects()->IsUndefined()) return;
// If there is a single break point clear it if it is the same.
if (!break_point_info->break_point_objects()->IsFixedArray()) {
if (break_point_info->break_point_objects() == *break_point_object) {
break_point_info->set_break_point_objects(Heap::undefined_value());
}
return;
}
// If there are multiple break points shrink the array
ASSERT(break_point_info->break_point_objects()->IsFixedArray());
Handle<FixedArray> old_array =
Handle<FixedArray>(
FixedArray::cast(break_point_info->break_point_objects()));
Handle<FixedArray> new_array =
Factory::NewFixedArray(old_array->length() - 1);
int found_count = 0;
for (int i = 0; i < old_array->length(); i++) {
if (old_array->get(i) == *break_point_object) {
ASSERT(found_count == 0);
found_count++;
} else {
new_array->set(i - found_count, old_array->get(i));
}
}
// If the break point was found in the list change it.
if (found_count > 0) break_point_info->set_break_point_objects(*new_array);
}
// Add the specified break point object.
void BreakPointInfo::SetBreakPoint(Handle<BreakPointInfo> break_point_info,
Handle<Object> break_point_object) {
// If there was no break point objects before just set it.
if (break_point_info->break_point_objects()->IsUndefined()) {
break_point_info->set_break_point_objects(*break_point_object);
return;
}
// If the break point object is the same as before just ignore.
if (break_point_info->break_point_objects() == *break_point_object) return;
// If there was one break point object before replace with array.
if (!break_point_info->break_point_objects()->IsFixedArray()) {
Handle<FixedArray> array = Factory::NewFixedArray(2);
array->set(0, break_point_info->break_point_objects());
array->set(1, *break_point_object);
break_point_info->set_break_point_objects(*array);
return;
}
// If there was more than one break point before extend array.
Handle<FixedArray> old_array =
Handle<FixedArray>(
FixedArray::cast(break_point_info->break_point_objects()));
Handle<FixedArray> new_array =
Factory::NewFixedArray(old_array->length() + 1);
for (int i = 0; i < old_array->length(); i++) {
// If the break point was there before just ignore.
if (old_array->get(i) == *break_point_object) return;
new_array->set(i, old_array->get(i));
}
// Add the new break point.
new_array->set(old_array->length(), *break_point_object);
break_point_info->set_break_point_objects(*new_array);
}
bool BreakPointInfo::HasBreakPointObject(
Handle<BreakPointInfo> break_point_info,
Handle<Object> break_point_object) {
// No break point.
if (break_point_info->break_point_objects()->IsUndefined()) return false;
// Single beak point.
if (!break_point_info->break_point_objects()->IsFixedArray()) {
return break_point_info->break_point_objects() == *break_point_object;
}
// Multiple break points.
FixedArray* array = FixedArray::cast(break_point_info->break_point_objects());
for (int i = 0; i < array->length(); i++) {
if (array->get(i) == *break_point_object) {
return true;
}
}
return false;
}
// Get the number of break points.
int BreakPointInfo::GetBreakPointCount() {
// No break point.
if (break_point_objects()->IsUndefined()) return 0;
// Single beak point.
if (!break_point_objects()->IsFixedArray()) return 1;
// Multiple break points.
return FixedArray::cast(break_point_objects())->length();
}
#endif
} } // namespace v8::internal