- Add CO_NESTED flag (0x10) for nested function scopes
- Emit LOAD_SMALL_INT for integers 0..=255 instead of LOAD_CONST
- Eliminate dead constant expression statements (no side effects)
- Ensure None in co_consts for functions with no other constants
- Add code.__replace__() for copy.replace() support
- Mark test_co_lnotab and test_invalid_bytecode as expectedFailure
When -O flag removes assert statements, any nested scopes
(generators, comprehensions, lambdas) inside the assert
expression still have symbol tables in the sub_tables list.
Without consuming them, the next_sub_table index gets
misaligned, causing later scopes to use wrong symbol tables.
Walk the skipped assert expression with an AST visitor to
find and consume nested scope symbol tables, keeping the
index aligned with AST traversal order.
* gc: add CollectResult, stats fields, get_referrers, and fix count reset
- Add CollectResult struct with collected/uncollectable/candidates/duration
- Add candidates and duration fields to GcStats and gc.get_stats()
- Pass CollectResult to gc.callbacks info dict
- Reset generation counts for all collected generations (0..=N)
- Return 0 for third value in gc.get_threshold() (3.13+)
- Implement gc.get_referrers() by scanning all tracked objects
- Add DEBUG_COLLECTABLE output for collectable objects
- Update test_gc.py to expect candidates/duration in stats
* Update test_gc from v3.14.3
* Update test_gc.py from CPython v3.15.0a5
Taken from v3.15 (not v3.14.3) because get_stats() candidates/duration
fields were added in 3.13+ and the corresponding test assertions only
exist in 3.15.
* Fix gc_state build on wasm32: skip Instant timing
* Add candidates/duration to gc callback info, mark v3.15 test failures
* Fix gc.get_referrers to exclude executing frames, fix Future cancelled exc leak
- get_referrers: skip frame objects on the execution stack, since
they are not GC-tracked in CPython (_PyInterpreterFrame)
- _asyncio Future/Task make_cancelled_error_impl: clear the stored
cancelled exception after returning it, matching the Python
_make_cancelled_error behavior
* Fix gc.get_threshold to return actual gen2 threshold value
* Fix inconsistent GC count reset in early-return paths
Use the same reset_end formula in unreachable-empty early returns
as in the main collection path and collecting-empty path.
* Accept keyword arguments in socket.__init__
Use a FromArgs struct instead of a positional-only tuple so that
family, type, proto, and fileno can be passed as keyword arguments.
* Disable comp_inlined in symbol table to match compiler
The compiler does not yet implement PEP 709 inlined comprehensions
(is_inlined_comprehension_context always returns false), but the
symbol table was marking comprehensions as inlined. This mismatch
could cause comprehension-local symbols to be merged into the parent
scope while the compiler still looks them up in a separate scope.
---------
Co-authored-by: CPython Developers <>
* Fix symbol table sub_table desync for non-simple annotation targets
Non-simple annotations (subscript/attribute/parenthesized targets like
`a[0]: expr`) were scanned in the annotation scope during symbol table
analysis, creating sub_tables for any comprehensions. But codegen only
compiles simple name annotations into __annotate__, so those sub_tables
were never consumed. This caused subsequent simple annotations'
comprehension sub_tables to get the wrong index, resulting in
"the symbol 'X' must be present in the symbol table" errors.
Fix: skip entering annotation scope for non-simple annotations since
they are never compiled into __annotate__.
* Validate forbidden expressions in non-simple annotation targets
Fix cspell "desynchronize" warning and validate yield/await/named/async
comprehension expressions in non-simple annotations without creating
annotation scopes.
* Restore in_annotation flag before propagating error
* Implement LOAD_ATTR inline caching with adaptive specialization
Add type version counter (tp_version_tag) to PyType with subclass
invalidation cascade. Add cache read/write methods (u16/u32/u64)
to CodeUnits. Implement adaptive specialization in load_attr that
replaces the opcode with specialized variants on first execution:
- LoadAttrMethodNoDict: cached method lookup for slotted types
- LoadAttrMethodWithValues: cached method with dict shadow check
- LoadAttrInstanceValue: direct dict lookup skipping descriptors
Specialized opcodes guard on type_version_tag and deoptimize back
to generic LOAD_ATTR with backoff counter on cache miss.
* Add BINARY_OP and CALL adaptive specialization
BINARY_OP: Specialize int add/subtract/multiply and float
add/subtract/multiply with type guards and deoptimization.
CALL: Add func_version to PyFunction, specialize simple
function calls (CallPyExactArgs, CallBoundMethodExactArgs)
with invoke_exact_args fast path that skips FuncArgs
allocation and fill_locals_from_args.
* Lazy quickening for adaptive specialization counters
Move counter initialization from compile-time to RESUME execution,
matching CPython's _PyCode_Quicken pattern. Store counter in CACHE
entry's arg byte to preserve op=Instruction::Cache for dis/JIT.
Add PyCode.quickened flag for one-time initialization.
* Add Instruction::deoptimize() and CodeUnits::original_bytes()
- deoptimize() maps specialized opcodes back to their base adaptive variant
- original_bytes() produces deoptimized bytecode with zeroed CACHE entries
- co_code now returns deoptimized bytes, _co_code_adaptive returns current bytes
- Marshal serialization uses original_bytes() instead of raw transmute
* Fix monitoring and specialization interaction
- cache_entries() returns correct count for instrumented opcodes
- deoptimize() maps instrumented opcodes back to base
- quicken() skips adaptive counter for instrumented opcodes
- instrument_code Phase 3 deoptimizes specialized opcodes and
clears CACHE entries to prevent stale pointer dereferences
* Address review: bounds checks, UB fix, version overflow, error handling
- Add bounds checks to read_cache_u16/u32/u64
- Fix quicken() aliasing UB by using &mut directly
- Add JumpBackwardJit/JumpBackwardNoJit to deoptimize()
- Guard can_specialize_call with NEWLOCALS flag check
- Use compare_exchange_weak for version tag to prevent wraparound
- Propagate dict lookup errors in LoadAttrMethodWithValues
- Apply adaptive backoff on version tag assignment failure
- Remove duplicate imports in frame.rs
* Emit CACHE code units in bytecode to match CPython 3.14
- Add cache_entries() method to Instruction enum
- Emit CACHE code units after opcodes in finalize_code
- Handle NO_LOCATION (line=-1) in linetable for CACHE entries
- Account for CACHE entries in exception table generation
- Skip CACHE entries in VM execution loop (with jump detection)
- Handle CACHE in InstrumentedLine/InstrumentedInstruction/InstrumentedForIter/InstrumentedNotTaken
- Skip CACHE in monitoring instrumentation phases
- Update co_branches() for cache-adjusted offsets
- Restore _cache_format in Lib/opcode.py
- Remove expectedFailure from test_c_call, test_start_offset
* Use relative jump offsets and fix bytecode layout
- Convert jump arguments from absolute to relative offsets
in frame.rs, monitoring.rs, and stack_analysis
- Add jump_relative_forward/backward helpers to ExecutingFrame
- Resolve pseudo jump instructions before offset fixpoint loop
- Emit NOP for break, continue, pass to match line-tracing
- Fix async for: emit EndAsyncFor with correct target, add NotTaken
- Fix comprehension if-cleanup to use separate block
- Fix super() source range for multi-line calls
- Fix NOP removal to preserve line-marker NOPs
- Fix InstrumentedLine cache skipping after re-dispatch
- Match InstrumentedResume/YieldValue in yield_from_target
- Remove CALL_FUNCTION_EX cache entry from opcode.py
- Remove resolved expectedFailure markers
* Align CPython 3.14 LOAD_GLOBAL null-bit and RERAISE semantics
* Remove redundant CPython-referencing comments
Clean up comments that unnecessarily mention CPython per project
convention. Replace with concise descriptions of the behavior itself.
- Implement set_f_lineno with stack analysis and deferred unwinding
- Add Frame::set_lasti() for trace callback line jumps
- Implement co_branches() on code objects
- Clear _cache_format in opcode.py (no inline caches)
- Fix getattro slot inheritance: preserve native slot from inherit_slots
- Fix BRANCH_RIGHT src_offset in InstrumentedPopJumpIf*
- Move lasti increment before line event for correct f_lineno
- Skip RESUME instruction from generating line events
- Defer stack pops via pending_stack_pops/pending_unwind_from_stack
to avoid deadlock with state mutex
- Fix ForIter exhaust target in mark_stacks to skip END_FOR
- Prevent exception handler paths from overwriting normal-flow stacks
- Replace #[cfg(feature = "threading")] duplication with PyAtomic<T>
from rustpython_common::atomic (Radium-based unified API)
- Remove expectedFailure from 31 now-passing jump tests
* cold block reordering and jump normalization
Add mark_cold, push_cold_blocks_to_end, and normalize_jumps
passes to the codegen CFG pipeline. Use JumpNoInterrupt for
exception handler exit paths in try-except-finally compilation.
* mark test_peepholer
* Add compile_bool_op_inner and optimize nested opposite-operator BoolOps to avoid redundant __bool__ calls
When a nested BoolOp has the opposite operator (e.g., `And` inside `Or`),
the inner BoolOp's short-circuit exits are redirected to skip the outer
BoolOp's redundant truth test. This avoids calling `__bool__()` twice on
the same value (e.g., `Test() and False or False` previously called
`Test().__bool__()` twice instead of once).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* Add snapshot test for nested BoolOp bytecode
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* Add runtime test for redundant __bool__ check (issue #3567)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* Apply clippy and rustfmt
* Apply ruff format
* Refactor compile_bool_op: extract emit_short_circuit_test and unify with compile_bool_op_inner
Reduce code duplication by:
- Extracting the repeated Copy + conditional jump pattern into emit_short_circuit_test
- Merging compile_bool_op and compile_bool_op_inner into a single
compile_bool_op_with_target with an optional short_circuit_target parameter
- Keeping compile_bool_op as a thin wrapper for the public interface
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* Relocate redundant __bool__ check test snippet
* Update extra_tests/snippets/syntax_short_circuit_bool.py
* Fix assertion in syntax_short_circuit_bool
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Jeong, YunWon <69878+youknowone@users.noreply.github.com>
- TypeVar/ParamSpec repr: use infer_variance flag
- ParamSpec: add type_check on bound argument
- ParamSpecArgs/Kwargs: use equality instead of identity
- NoDefault: change to IMMUTABLETYPE flag
- subscript_generic: wrap TypeVarTuple in Unpack
- symboltable: selective name mangling in type param scopes
- symboltable: fix double default scanning for non-generic fns
- Unmark 4 passing tests in test_type_params
- Rewrite compile_dict to preserve insertion order with ** unpacking
- Fix function call error messages to include function name and
keyword-only argument counts
- Fix missing keyword-only argument error to list all missing args
- Fix starred expression error messages to match CPython
- Fix "cannot unpack non-iterable" error message
- Remove resolved expected failure markers in test_extcall and
test_unpack_ex
Annotation scope:
- Remove module-level annotation re-scan that created phantom
sub_tables, breaking annotation closure for comprehensions
- Add async comprehension check in symbol table with
is_in_async_context(); annotation/type-params scopes are
always non-async
- Save/restore CompileContext in enter/exit_annotation_scope
to reset in_async_scope
Deadlock prevention:
- Fix TypeVar/ParamSpec/TypeVarTuple __default__ and
evaluate_default by cloning lock contents before acquiring
a second lock or calling Python
Other fixes:
- Add HEAPTYPE flag to Generic for correct pickle behavior
- Guard heaptype_ext access in name_inner/set___name__/
set___qualname__ with safe checks instead of unwrap
- Fix MRO error message to include base class names
- Add "format" to varnames in TypeAlias annotation scopes
- Fix single-element tuple repr to include trailing comma
Annotation scopes now inherit can_see_class_scope from parent scopes
(not just direct Class parents), allowing annotations in generic
methods inside classes to access class variables via __classdict__.
Remove 3 expectedFailure markers from test_type_params.
- Reject nonlocal binding for type parameters in symboltable
- Enable __class_getitem__ on memoryview, Template, Interpolation
- Fix __set_name__ error note to quote attribute name
- Set re.Match and re.Pattern module to "re" instead of "_sre"
- Migrate parking_lot::Mutex to PyMutex in typevar.rs
- Collapse nested if-let chains in typing_type_repr (clippy)
- Remove 5 expectedFailure markers from test files
- Implement _ConstEvaluator type in _typing module with STRING
format support via typing_type_repr (port of _Py_typing_type_repr)
- Add evaluate_bound, evaluate_constraints, evaluate_default
pygetset properties to TypeVar, ParamSpec, TypeVarTuple
- Emit format validation in evaluator scopes
(compile_type_param_bound_or_default, TypeAlias value scopes)
so evaluators raise NotImplementedError for unsupported formats
- Add non-default-after-default SyntaxError in scan_type_params
- Fix ParamSpec default_value to use Mutex for proper caching
- Fix TypeVar constructor: evaluate_constraints set to None
instead of constraints tuple for eager-constructed TypeVars
- Pass format=1 (FORMAT_VALUE) to all lazy evaluator calls
- Remove 6 expectedFailure markers from test_type_params
- Add interactive REPL mode: auto-print expression results in single mode
- Implement Destructor for PyGenerator and PyCoroutine
- Add locals_dirty tracking and locals_to_fast() for frame sync
- Add per-line tracing with prev_line tracking in execution loop
- Fix gen_throw to close sub-iterator on GeneratorExit (gen_close_iter)
- Pass callable object as arg in c_call/c_return/c_exception trace events
- Distinguish [Errno] vs [WinError] for CRT vs Win32 API errors
- Fix tee thread safety with AtomicBool running flag
- Fix division error messages to match expected format
For example in the following code:
def foo(x: int, y: float):
def bar(q: int):
return q
pass
Make sure that `foo` type annotations are correctly propogated to
it's `__annotate__` and `__annotations__` attributes.
With this chage, we'll get:
>>>>> foo.__annotations__
{'x': <class 'int'>, 'y': <class 'float'>}
Previously annotations where 'lost', and we would get:
>>>>> foo.__annotations__
{}
* `std` feature for common
- Gate OS-dependent modules behind `#[cfg(feature = "std")]`
- Replace `std::f64` with `core::f64` in float_ops
- Replace `std::process::abort` with panic in refcount
- Remove `thread_local` from levenshtein (stack buffer)
- Split static_cell into threading/non_threading/no_std
* `std` for codegen
* `no_std` for pylib
* Fix test_import: import machinery and errors
- Emit IMPORT_FROM instead of LOAD_ATTR for `import a.b.c as m`
- Add "partially initialized module" error for circular imports
- Add "cannot access submodule" error for initializing submodules
- Implement script shadowing detection with "consider renaming" hint
- Detect user scripts shadowing stdlib/third-party modules
- Compute original sys.path[0] from sys.argv[0]
- Check sys.stdlib_module_names for stdlib detection
- Respect safe_path setting
- Implement _imp._fix_co_filename for code source_path rewriting
- Add data parameter to _imp.get_frozen_object with marshal deser
- Fix import_from: check __spec__.has_location before using origin
- Set ImportError.path attribute on import failures
- Fix import_star error messages for non-str __all__/__dict__ items
- Always call builtins.__import__ in import_inner
* Propagate __bool__ errors from fromlist in import
Previously, try_to_bool errors were silently swallowed via .ok(),
causing fromlist with broken __bool__ to default to false.
* Generate optimized oparg enums
* No need to match on 255
* Remove `num_enum` crate from `compiler-core`
* Update `Cargo.lock`
* macro fmt
* Rename macro vars
* Match without `,`
* Support alternative values
* Fix alternatives
* Improve docs
* Add const assert
* Don't use `as u32`
* Make only ComparisonOperator unoptimized
* Fix test
* cleanup
* All opargs are optimized
* Remove comment
* Update codeop from v3.14.3
* Fix traceback, syntax errors, and exception handling
- Improve unclosed bracket detection with "'(' was never closed" message
- Fix IndentationError location to point to end of line
- Implement frame.clear() with proper checks for executing/suspended frames
- Fix exception context chaining for propagated exceptions
- Add traceback.__dir__() and prevent tb_next deletion
- Fix subscript operation source range restoration in compiler
- Change "duplicate parameter" to "duplicate argument" error message
- Refactor duplicate code in asyncgenerator.rs and frame.rs
---------
Co-authored-by: CPython Developers <>
- Rewrite faulthandler with live frame walking via
Frame.previous AtomicPtr chain and thread-local
CURRENT_FRAME (AtomicPtr) instead of frame snapshots
- Add signal-safe traceback dumping (dump_live_frames,
dump_frame_from_raw) walking the Frame.previous chain
- Add safe_truncate/dump_ascii for UTF-8 safe string
truncation in signal handlers
- Refactor write_thread_id to accept thread_id parameter
- Add SA_RESTART for user signal registration, SA_NODEFER
only when chaining
- Save/restore errno in faulthandler_user_signal
- Add signal re-entrancy guard in trigger_signals to
prevent recursive handler invocation
- Add thread frame tracking (push/pop/cleanup/reinit)
with force_unlock fallback for post-fork recovery
- Remove expectedFailure markers for now-passing tests
