Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Normative: specify for-in enumeration order in more cases #1791

Merged
merged 1 commit into from
Jan 3, 2020
Merged

Normative: specify for-in enumeration order in more cases #1791

merged 1 commit into from
Jan 3, 2020
+135 −1

Conversation

bakkot
Copy link
Contributor

@bakkot bakkot commented Nov 23, 2019

This PR implements the for-in-order proposal, currently at stage 3.

I intend to request stage 4 at the December 2019 meeting.

Here is the PR adding tests to test262.

@bakkot bakkot added normative change Affects behavior required to correctly evaluate some ECMAScript source text pending stage 4 This proposal has not yet achieved stage 4, but may otherwise be ready to merge. has test262 tests proposal This is related to a specific proposal, and will be closed/merged when the proposal reaches stage 4. labels Nov 23, 2019
@ljharb ljharb added needs test262 tests The proposal should specify how to test an implementation. Ideally via github.com/tc39/test262 and removed has test262 tests labels Nov 23, 2019
@ljharb
Copy link
Member

ljharb commented Nov 23, 2019

("has tests" applies after the test262 tests are merged)

@ljharb ljharb requested review from syg, michaelficarra and a team November 23, 2019 06:21
ljharb
ljharb approved these changes Nov 23, 2019
View reviewed changes
spec.html Outdated Show resolved Hide resolved
spec.html Outdated Show resolved Hide resolved
bakkot
bakkot commented Nov 27, 2019
View reviewed changes
spec.html Outdated Show resolved Hide resolved
syg
syg approved these changes Nov 28, 2019
View reviewed changes
Copy link
Contributor

@syg syg left a comment •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Editor review: lgtm with nits and wordsmithing.

spec.html Outdated Show resolved Hide resolved
spec.html Outdated Show resolved Hide resolved
spec.html Outdated Show resolved Hide resolved
@bakkot
Copy link
Contributor Author

bakkot commented Nov 28, 2019

Addressed comments; thanks for the feedback.

@ljharb ljharb added has stage 4 This PR represents a proposal that has achieved stage 4, and is ready to merge. and removed pending stage 4 This proposal has not yet achieved stage 4, but may otherwise be ready to merge. labels Dec 4, 2019
@ljharb ljharb added has test262 tests and removed needs test262 tests The proposal should specify how to test an implementation. Ideally via github.com/tc39/test262 labels Dec 18, 2019
@ljharb ljharb self-assigned this Dec 18, 2019
@ljharb ljharb mentioned this pull request Dec 21, 2019
Closed
jmdyck
jmdyck reviewed Jan 2, 2020
View reviewed changes
spec.html Outdated
<emu-note>
<p>The following is an informative definition of an ECMAScript generator function that conforms to these rules:</p>
<p>Hosts are not required to implement the algorithm in <emu-xref href="#sec-%foriniteratorprototype%.next"></emu-xref> directly. They may choose any implementation whose behavior will not deviate from that algorithm unless one of the constraints in the previous paragraph is violated.</p>
<p>The following is an informative definition of an ECMAScript generator function that conforms to these rules:</ins></p>
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
<p>The following is an informative definition of an ECMAScript generator function that conforms to these rules:</ins></p>
<p>The following is an informative definition of an ECMAScript generator function that conforms to these rules:</p>

@ljharb ljharb merged commit 1bed13a into master Jan 3, 2020
@ljharb ljharb deleted the proposal-for-in-order branch January 3, 2020 00:00
ljharb added a commit that referenced this pull request Jan 3, 2020
@ljharb
Markup: remove stray </ins> tag (#1791)
cb73c69 
This correction was noted here: #1791 (comment)
but omitted in the PR merge.
@azu azu mentioned this pull request Jun 20, 2020
Closed
2 tasks
@azu azu mentioned this pull request Jun 28, 2020
Merged
bertogg pushed a commit to Igalia/webkit that referenced this pull request Dec 16, 2020
Non-enumerable property fails to shadow inherited enumerable property…
d208111 
… from for-in

https://bugs.webkit.org/show_bug.cgi?id=38970

Reviewed by Keith Miller.

JSTests:

* stress/arguments-bizarre-behaviour-disable-enumerability.js:
* stress/for-in-redefine-enumerable.js: Added.
* stress/for-in-shadow-non-enumerable.js: Added.
* test262/expectations.yaml: Mark 4 test cases as passing.

Source/JavaScriptCore:

While for/in was initially specified with notion of "shadowing", it wasn't clarified
until ES5 that [[Enumerable]] attributes are ignored when determining if a property
has already been processed. Recently, for/in spec was expanded [1] to pin down common
case enumeration as it's currently implemented by V8 and SpiderMonkey.

Since keeping track of DontEnum properties is a massive slowdown for uncached runs
(with any data structure used), this patch simply adds [[Enumerable]] check to
has_{indexed,structure,generic}_property bytecode ops and does renaming chores.

Common code is now shared between HasIndexedProperty (emitted for `0 in arr`) and
HasEnumerableIndexedProperty DFG nodes via passing different slow path ops rather
than having OpInfo with PropertySlot::InternalMethodType, which is a nice refactor.

While this change aligns common case for/in enumeration with the spec and other
engines, it also introduces a few observable discrepancies from V8 and SpiderMonkey,
which are permitted by the spec [2]:
a) properties that have been redefined as DontEnum within loop body are skipped,
   which matches the spec [3] and seems like expected behavior;
b) "shadowing" is broken if a DontEnum property of already visited object is
   added / deleted / redefined within loop body, which (pretty much) never happens.

This patch introduces a new invariant: all properties getOwn*PropertyNames() returns
in DontEnumPropertiesMode::Exclude should be reported as [[Enumerable]] by
getOwnPropertySlot(). JSCallbackObject and RuntimeArray are fixed to follow it.

for/in and Object.keys microbenchmarks are neutral. This change does not affect
JSPropertyNameEnumerator caching, nor fast paths of its bytecodes.

[1]: tc39/ecma262#1791
[2]: https://tc39.es/ecma262/#sec-enumerate-object-properties (last paragraph)
[3]: https://tc39.es/ecma262/#sec-%foriniteratorprototype%.next (step 7.b.iii)

* API/JSCallbackObjectFunctions.h:
(JSC::JSCallbackObject<Parent>::getOwnPropertySlot):
* API/tests/testapi.c:
* API/tests/testapiScripts/testapi.js:
* bytecode/BytecodeList.rb:
* bytecode/BytecodeUseDef.cpp:
(JSC::computeUsesForBytecodeIndexImpl):
(JSC::computeDefsForBytecodeIndexImpl):
* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::finishCreation):
* bytecode/Opcode.h:
* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::emitHasEnumerableIndexedProperty):
(JSC::BytecodeGenerator::emitHasEnumerableStructureProperty):
(JSC::BytecodeGenerator::emitHasEnumerableProperty):
(JSC::BytecodeGenerator::emitHasGenericProperty): Deleted.
(JSC::BytecodeGenerator::emitHasIndexedProperty): Deleted.
(JSC::BytecodeGenerator::emitHasStructureProperty): Deleted.
* bytecompiler/BytecodeGenerator.h:
* bytecompiler/NodesCodegen.cpp:
(JSC::ForInNode::emitBytecode):
* dfg/DFGAbstractInterpreterInlines.h:
(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGCapabilities.cpp:
(JSC::DFG::capabilityLevel):
* dfg/DFGClobberize.h:
(JSC::DFG::clobberize):
* dfg/DFGDoesGC.cpp:
(JSC::DFG::doesGC):
* dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode):
(JSC::DFG::FixupPhase::convertToHasIndexedProperty):
* dfg/DFGNode.h:
(JSC::DFG::Node::hasArrayMode):
(JSC::DFG::Node::hasInternalMethodType const): Deleted.
(JSC::DFG::Node::internalMethodType const): Deleted.
(JSC::DFG::Node::setInternalMethodType): Deleted.
* dfg/DFGNodeType.h:
* dfg/DFGOperations.cpp:
(JSC::DFG::JSC_DEFINE_JIT_OPERATION):
* dfg/DFGOperations.h:
* dfg/DFGPredictionPropagationPhase.cpp:
* dfg/DFGSSALoweringPhase.cpp:
(JSC::DFG::SSALoweringPhase::handleNode):
* dfg/DFGSafeToExecute.h:
(JSC::DFG::safeToExecute):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileHasEnumerableProperty):
(JSC::DFG::SpeculativeJIT::compileHasEnumerableStructureProperty):
(JSC::DFG::SpeculativeJIT::compileHasIndexedProperty):
(JSC::DFG::SpeculativeJIT::compileHasGenericProperty): Deleted.
(JSC::DFG::SpeculativeJIT::compileHasStructureProperty): Deleted.
* dfg/DFGSpeculativeJIT.h:
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* ftl/FTLCapabilities.cpp:
(JSC::FTL::canCompile):
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::compileNode):
(JSC::FTL::DFG::LowerDFGToB3::compileHasIndexedProperty):
(JSC::FTL::DFG::LowerDFGToB3::compileHasEnumerableProperty):
(JSC::FTL::DFG::LowerDFGToB3::compileHasEnumerableStructureProperty):
(JSC::FTL::DFG::LowerDFGToB3::compileHasGenericProperty): Deleted.
(JSC::FTL::DFG::LowerDFGToB3::compileHasStructureProperty): Deleted.
* jit/JIT.cpp:
(JSC::JIT::privateCompileMainPass):
(JSC::JIT::privateCompileSlowCases):
* jit/JIT.h:
* jit/JITOpcodes.cpp:
(JSC::JIT::emit_op_has_enumerable_structure_property):
(JSC::JIT::emit_op_has_enumerable_indexed_property):
(JSC::JIT::emitSlow_op_has_enumerable_indexed_property):
(JSC::JIT::emit_op_has_structure_property): Deleted.
(JSC::JIT::emit_op_has_indexed_property): Deleted.
(JSC::JIT::emitSlow_op_has_indexed_property): Deleted.
* jit/JITOpcodes32_64.cpp:
(JSC::JIT::emit_op_has_enumerable_structure_property):
(JSC::JIT::emit_op_has_enumerable_indexed_property):
(JSC::JIT::emitSlow_op_has_enumerable_indexed_property):
(JSC::JIT::emit_op_has_structure_property): Deleted.
(JSC::JIT::emit_op_has_indexed_property): Deleted.
(JSC::JIT::emitSlow_op_has_indexed_property): Deleted.
* jit/JITOperations.cpp:
(JSC::JSC_DEFINE_JIT_OPERATION):
* jit/JITOperations.h:
* llint/LowLevelInterpreter.asm:
* llint/LowLevelInterpreter64.asm:
* runtime/CommonSlowPaths.cpp:
(JSC::JSC_DEFINE_COMMON_SLOW_PATH):
* runtime/CommonSlowPaths.h:
* runtime/JSObject.cpp:
(JSC::JSObject::hasProperty const):
(JSC::JSObject::hasEnumerableProperty const):
(JSC::JSObject::hasPropertyGeneric const): Deleted.
* runtime/JSObject.h:

Source/WebCore:

Report RuntimeArray indices as [[Enumerable]].

Test: platform/mac/fast/dom/wrapper-classes-objc.html

* bridge/runtime_array.cpp:
(JSC::RuntimeArray::getOwnPropertySlot):
(JSC::RuntimeArray::getOwnPropertySlotByIndex):

LayoutTests:

* platform/mac/fast/dom/wrapper-classes-objc-expected.txt:
* platform/mac/fast/dom/wrapper-classes-objc.html:


git-svn-id: http://svn.webkit.org/repository/webkit/trunk@270874 268f45cc-cd09-0410-ab3c-d52691b4dbfc
ryanhaddad pushed a commit to WebKit/WebKit that referenced this pull request Dec 22, 2020
@shvaikalesh
Non-enumerable property fails to shadow inherited enumerable property…
55e3813 
… from for-in

https://bugs.webkit.org/show_bug.cgi?id=38970

Reviewed by Keith Miller.

JSTests:

* stress/arguments-bizarre-behaviour-disable-enumerability.js:
* stress/for-in-redefine-enumerable.js: Added.
* stress/for-in-shadow-non-enumerable.js: Added.
* test262/expectations.yaml: Mark 4 test cases as passing.

Source/JavaScriptCore:

While for/in was initially specified with notion of "shadowing", it wasn't clarified
until ES5 that [[Enumerable]] attributes are ignored when determining if a property
has already been processed. Recently, for/in spec was expanded [1] to pin down common
case enumeration as it's currently implemented by V8 and SpiderMonkey.

Since keeping track of DontEnum properties is a massive slowdown for uncached runs
(with any data structure used), this patch simply adds [[Enumerable]] check to
has_{indexed,structure,generic}_property bytecode ops and does renaming chores.

Common code is now shared between HasIndexedProperty (emitted for `0 in arr`) and
HasEnumerableIndexedProperty DFG nodes via passing different slow path ops rather
than having OpInfo with PropertySlot::InternalMethodType, which is a nice refactor.

While this change aligns common case for/in enumeration with the spec and other
engines, it also introduces a few observable discrepancies from V8 and SpiderMonkey,
which are permitted by the spec [2]:
a) properties that have been redefined as DontEnum within loop body are skipped,
   which matches the spec [3] and seems like expected behavior;
b) "shadowing" is broken if a DontEnum property of already visited object is
   added / deleted / redefined within loop body, which (pretty much) never happens.

This patch introduces a new invariant: all properties getOwn*PropertyNames() returns
in DontEnumPropertiesMode::Exclude should be reported as [[Enumerable]] by
getOwnPropertySlot(). JSCallbackObject and RuntimeArray are fixed to follow it.

for/in and Object.keys microbenchmarks are neutral. This change does not affect
JSPropertyNameEnumerator caching, nor fast paths of its bytecodes.

[1]: tc39/ecma262#1791
[2]: https://tc39.es/ecma262/#sec-enumerate-object-properties (last paragraph)
[3]: https://tc39.es/ecma262/#sec-%foriniteratorprototype%.next (step 7.b.iii)

* API/JSCallbackObjectFunctions.h:
(JSC::JSCallbackObject<Parent>::getOwnPropertySlot):
* API/tests/testapi.c:
* API/tests/testapiScripts/testapi.js:
* bytecode/BytecodeList.rb:
* bytecode/BytecodeUseDef.cpp:
(JSC::computeUsesForBytecodeIndexImpl):
(JSC::computeDefsForBytecodeIndexImpl):
* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::finishCreation):
* bytecode/Opcode.h:
* bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::emitHasEnumerableIndexedProperty):
(JSC::BytecodeGenerator::emitHasEnumerableStructureProperty):
(JSC::BytecodeGenerator::emitHasEnumerableProperty):
(JSC::BytecodeGenerator::emitHasGenericProperty): Deleted.
(JSC::BytecodeGenerator::emitHasIndexedProperty): Deleted.
(JSC::BytecodeGenerator::emitHasStructureProperty): Deleted.
* bytecompiler/BytecodeGenerator.h:
* bytecompiler/NodesCodegen.cpp:
(JSC::ForInNode::emitBytecode):
* dfg/DFGAbstractInterpreterInlines.h:
(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::parseBlock):
* dfg/DFGCapabilities.cpp:
(JSC::DFG::capabilityLevel):
* dfg/DFGClobberize.h:
(JSC::DFG::clobberize):
* dfg/DFGDoesGC.cpp:
(JSC::DFG::doesGC):
* dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode):
(JSC::DFG::FixupPhase::convertToHasIndexedProperty):
* dfg/DFGNode.h:
(JSC::DFG::Node::hasArrayMode):
(JSC::DFG::Node::hasInternalMethodType const): Deleted.
(JSC::DFG::Node::internalMethodType const): Deleted.
(JSC::DFG::Node::setInternalMethodType): Deleted.
* dfg/DFGNodeType.h:
* dfg/DFGOperations.cpp:
(JSC::DFG::JSC_DEFINE_JIT_OPERATION):
* dfg/DFGOperations.h:
* dfg/DFGPredictionPropagationPhase.cpp:
* dfg/DFGSSALoweringPhase.cpp:
(JSC::DFG::SSALoweringPhase::handleNode):
* dfg/DFGSafeToExecute.h:
(JSC::DFG::safeToExecute):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileHasEnumerableProperty):
(JSC::DFG::SpeculativeJIT::compileHasEnumerableStructureProperty):
(JSC::DFG::SpeculativeJIT::compileHasIndexedProperty):
(JSC::DFG::SpeculativeJIT::compileHasGenericProperty): Deleted.
(JSC::DFG::SpeculativeJIT::compileHasStructureProperty): Deleted.
* dfg/DFGSpeculativeJIT.h:
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* ftl/FTLCapabilities.cpp:
(JSC::FTL::canCompile):
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::compileNode):
(JSC::FTL::DFG::LowerDFGToB3::compileHasIndexedProperty):
(JSC::FTL::DFG::LowerDFGToB3::compileHasEnumerableProperty):
(JSC::FTL::DFG::LowerDFGToB3::compileHasEnumerableStructureProperty):
(JSC::FTL::DFG::LowerDFGToB3::compileHasGenericProperty): Deleted.
(JSC::FTL::DFG::LowerDFGToB3::compileHasStructureProperty): Deleted.
* jit/JIT.cpp:
(JSC::JIT::privateCompileMainPass):
(JSC::JIT::privateCompileSlowCases):
* jit/JIT.h:
* jit/JITOpcodes.cpp:
(JSC::JIT::emit_op_has_enumerable_structure_property):
(JSC::JIT::emit_op_has_enumerable_indexed_property):
(JSC::JIT::emitSlow_op_has_enumerable_indexed_property):
(JSC::JIT::emit_op_has_structure_property): Deleted.
(JSC::JIT::emit_op_has_indexed_property): Deleted.
(JSC::JIT::emitSlow_op_has_indexed_property): Deleted.
* jit/JITOpcodes32_64.cpp:
(JSC::JIT::emit_op_has_enumerable_structure_property):
(JSC::JIT::emit_op_has_enumerable_indexed_property):
(JSC::JIT::emitSlow_op_has_enumerable_indexed_property):
(JSC::JIT::emit_op_has_structure_property): Deleted.
(JSC::JIT::emit_op_has_indexed_property): Deleted.
(JSC::JIT::emitSlow_op_has_indexed_property): Deleted.
* jit/JITOperations.cpp:
(JSC::JSC_DEFINE_JIT_OPERATION):
* jit/JITOperations.h:
* llint/LowLevelInterpreter.asm:
* llint/LowLevelInterpreter64.asm:
* runtime/CommonSlowPaths.cpp:
(JSC::JSC_DEFINE_COMMON_SLOW_PATH):
* runtime/CommonSlowPaths.h:
* runtime/JSObject.cpp:
(JSC::JSObject::hasProperty const):
(JSC::JSObject::hasEnumerableProperty const):
(JSC::JSObject::hasPropertyGeneric const): Deleted.
* runtime/JSObject.h:

Source/WebCore:

Report RuntimeArray indices as [[Enumerable]].

Test: platform/mac/fast/dom/wrapper-classes-objc.html

* bridge/runtime_array.cpp:
(JSC::RuntimeArray::getOwnPropertySlot):
(JSC::RuntimeArray::getOwnPropertySlotByIndex):

LayoutTests:

* platform/mac/fast/dom/wrapper-classes-objc-expected.txt:
* platform/mac/fast/dom/wrapper-classes-objc.html:


Canonical link: https://commits.webkit.org/232510@main
git-svn-id: https://svn.webkit.org/repository/webkit/trunk@270874 268f45cc-cd09-0410-ab3c-d52691b4dbfc
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@jmdyck jmdyck jmdyck left review comments

@syg syg syg approved these changes

@ljharb ljharb ljharb approved these changes

@michaelficarra michaelficarra Awaiting requested review from michaelficarra

Assignees

@ljharb ljharb

Labels
editorial change has stage 4 This PR represents a proposal that has achieved stage 4, and is ready to merge. has test262 tests normative change Affects behavior required to correctly evaluate some ECMAScript source text proposal This is related to a specific proposal, and will be closed/merged when the proposal reaches stage 4.
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@bakkot @ljharb @syg @jmdyck