Conversation
…add test_unary + test_pow Track A fixes (6 un-skips): - isinstance() arg2 validation: raises TypeError for non-type args - issubclass() arg1/arg2 validation: raises TypeError for non-class args - issubclass() tuple arg support: issubclass(cls, (type1, type2)) - Accept type_type, user_type_metatype, and exc_metatype as valid metatypes - list.__setitem__ with string key: raise TypeError instead of segfault - Default tp_base to object_type for classes without explicit base Track B (2 new CPython tests, 0 skips): - test_unary.py: unary +, -, ~ operators (4 tests) - test_pow.py: pow() and ** operator (9 tests) Infrastructure fixes: - builtin pow() delegates to int_power for GMP overflow correctness - 3-arg pow() negative modulus: Python-compatible remainder adjustment - 3-arg pow() exp=0: apply final modulus (pow(x,0,mod) = 1 % mod) - assertAlmostEqual/assertNotAlmostEqual added to unittest Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…_contains Fixes: - CONTAINS_OP tp_iter fallback: iterate via tp_iter when sq_contains and __contains__ dunder are unavailable (enables 'in' for range, etc.) - CONTAINS_OP __getitem__ fallback: iterate via __getitem__(0,1,2,...) catching IndexError (enables 'in' for classes with only __getitem__) - str_contains: validate substr is a string before accessing data (fixes segfault on `None in 'abc'`) New test (0 skips): - test_contains.py: 'in' operator for custom classes, builtins, strings (4 tests) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…kips New exception type: - UnboundLocalError (subclass of NameError) for LOAD_FAST_CHECK and LOAD_DEREF New function attribute: - func.__qualname__ reads co_qualname from code object Formatted error messages: - "name 'X' is not defined" (was "name not found") in LOAD_GLOBAL/LOAD_NAME - "f() takes from N to M positional arguments but K were given" (was "function takes too many positional arguments") with qualname, min/max arg counts, and actual count Un-skipped tests: - test_scope.testUnboundLocal — catchable UnboundLocalError - test_keywordonlyarg.testTooManyPositionalErrorMessage — __qualname__ + message format - test_keywordonlyarg.test_default_evaluation_order — NameError message format Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Bug fixes: - CALL_INTRINSIC_1 LIST_TO_TUPLE: remove stale mov rbx,rax that clobbered bytecode IP register, causing crash after star-unpacking in calls - repr_push: raise RecursionError when repr stack depth exceeds 64 (was silently ignoring, leading to stack overflow segfault) Un-skipped tests: - test_keywordonlyarg.testFunctionCall — star-unpacking now works - test_augassign.testSequences — slice augmented assignment works (stale skip) - list_tests.test_repr_deep — RecursionError on deeply nested list repr Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…comps, raise, class Track B imports (all passing, minimal skips): - test_exception_variations.py: 15 tests, 0 skips — try/except/else/finally combos - test_genexps.py: 14 tests, 0 skips — generator expressions - test_listcomps.py: 15 tests, 0 skips — list/dict/set comprehensions - test_raise.py: 16 tests, 2 skips — raise statement, reraise, except filtering - test_class.py: 24 tests, 0 skips — class creation, dunders, inheritance Total CPython test suites: 25 (up from 20) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
… skips) Bug fix: - COMPARE_OP identity fallback: check both payload AND tag for equality. None (payload=0, TAG_NONE) was comparing equal to SmallInt 0 (payload=0, TAG_SMALLINT) because only payload was checked. New test: - test_compare.py: int, float, string, list, tuple, set, dict, None, bool comparisons plus custom __eq__/__ne__/__lt__/__gt__ (15 tests) Total CPython test suites: 26 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…st_opcodes, test_baseexception Infrastructure fixes: - PyExceptionObject.exc_dict: new field for arbitrary instance attributes on exception objects (both built-in and user-defined exception types) - exc_setattr: store custom attributes in exc_dict (creates dict on demand) - exc_getattr: check exc_dict after type dict (fix: skip to exc_dict when type dict is NULL instead of jumping to not_found) - BaseException.tp_base = object_type (enables issubclass(Exception, object)) - exc_subclass_call: call user-defined __init__ after exc_type_call - PyExceptionGroupObject: add exc_dict field (keeps struct alignment) - Wire exc_getattr/exc_setattr on user-defined exception subclasses New tests (0 skips except noted): - test_with.py: 12 tests — with statement, context managers, nesting - test_opcodes.py: 11 tests — try/except loops, unpacking, format, augmented assign - test_baseexception.py: 15 tests, 2 skips — exception hierarchy, args, custom exceptions Total CPython test suites: 29 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…roperty, test_string Infrastructure: - builtin next(iterator, default): return default on StopIteration instead of raising (2-arg form) New tests: - test_extcall.py: 12 tests — star args, kwargs, mixed calls, defaults - test_iter.py: 25 tests — iter protocol, for loops, enumerate, zip, map, filter, reversed, sorted, sum, min/max, any/all, custom iterators - test_lambda.py: 12 tests — closures, defaults, varargs, callbacks - test_property.py: 10 tests, 1 skip — property, staticmethod, classmethod - test_string.py: 21 tests — f-strings, % format, methods, slicing Total CPython test suites: 34 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
New tests: - test_bytes.py: 8 tests — literals, indexing, slicing, comparison, decode, methods - test_builtin.py: 23 tests — abs, bool, chr/ord, divmod, hash, hex/oct/bin, id, int, isinstance, len, max/min, pow, range, repr, round, sorted, str, sum, type, zip, enumerate, map/filter, callable - test_types.py: 21 tests, 1 skip — truth values, type checks, conversions (int, float, str, bool, list, tuple, set) Total CPython test suites: 37 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…xtra, test_set_extra, test_list_extra Infrastructure: - assertIsNone/assertIsNotNone added to unittest New tests: - test_closures.py: 11 tests — closures, nonlocal, nested scopes, lambda closures - test_dict_extra.py: 20 tests — constructor, get/set/del, keys/values/items, get/pop/update/setdefault/clear/copy, comprehension, fromkeys, nested - test_tuple_extra.py: 18 tests — indexing, slicing, concat, count, index, unpacking, star-unpacking, constructor, nested, mixed types - test_set_extra.py: 20 tests, 1 skip — add/discard/remove/pop/clear, union/intersection/difference/symmetric_difference, copy, comprehension, frozenset, isdisjoint - test_list_extra.py: 25 tests — append/extend/insert/remove/pop/index/count, reverse/sort/copy/clear, slicing, multiplication, comparison, self-ref repr Total CPython test suites: 42 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…unpacking, test_inheritance New tests: - test_controlflow.py: 23 tests — if/elif/else, while, for, break/continue, while/for-else, ternary, pass, nested loops - test_math_basic.py: 25 tests — int/float/bool arithmetic, bitwise ops, abs, divmod, large ints, mixed-type arithmetic - test_global_nonlocal.py: 12 tests — global/nonlocal statements, enclosing scope, class scope, comprehension scope isolation - test_unpacking.py: 14 tests, 3 skips — tuple/list unpacking, star unpacking, nested unpacking, swap, for-loop star unpacking - test_inheritance.py: 13 tests — single inheritance, super(), method resolution, isinstance/issubclass chains, exception inheritance Total CPython test suites: 47 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…ites Bug fix: - op_send exhausted: guard gi_return_value read with tp_basicsize > 56. Plain iterators (str_iter, list_iter) have smaller objects without gi_return_value field. Was crashing on yield-from-string and silently returning garbage for some iterator types. - Export async_gen_asend_type for use by SEND type checks. New tests (0 skips unless noted): - test_del.py: 7 tests — del local/list/dict/attribute/multiple - test_assert.py: 8 tests — assert true/false/message/expression - test_assignment.py: 28 tests — simple/augmented/subscript/attribute assign - test_exceptions_extra.py: 17 tests — except clauses, finally, raise/reraise - test_generators_extra.py: 19 tests — yield, send, yield-from, genexps Total CPython test suites: 52 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
New tests: - test_format.py: 24 tests — f-strings (expressions, conversions, adjacent), % formatting (string, int, repr, hex, multiple), string methods - test_slice_ops.py: 14 tests — list/tuple/string slicing, step, negative indices, slice assign/delete, extended slicing - test_numeric.py: 17 tests — int edge cases (large, conversion, bitwise), float (special values, NaN, rounding), bool (is-int, arithmetic) - test_comprehensions.py: 22 tests — list/dict/set comps, genexps with filter, nested, closure, empty, max/min/sum/any/all Total CPython test suites: 56 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…tures New tests: - test_decorators_extra.py: 8 tests — function/class/method decorators, stacked, with args, staticmethod, classmethod, property - test_walrus.py: 5 tests — assignment expressions in if/while/listcomp - test_match.py: 8 tests — match/case with literals, strings, capture, or-pattern, guard, tuple pattern, None/True/False patterns - test_datastructures.py: 14 tests — nested structures, frequency count, groupby, sorting with key/reverse, stable sort Total CPython test suites: 60 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…t_conditional New tests: - test_exceptions_builtin.py: 16 tests — exception hierarchy, args, catching, error-raising builtins (TypeError, ValueError, IndexError, etc.) - test_functions.py: 18 tests — defaults, varargs, kwargs, keyword-only, recursion (factorial, fibonacci, mutual), higher-order functions - test_range_extra.py: 13 tests — range constructor, step, negative, len, contains, reversed, enumerate, sum - test_conditional.py: 15 tests — ternary expressions, short-circuit and/or, not, chained booleans, is/is not identity Total CPython test suites: 64 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
…es, stdlib modules Phase 0 — crash/correctness fixes: - Fix raise segfault on non-exception values (strings, ints) - Fix callable() false positives for dict/set/list instances - Fix dict_richcompare r11/r9 clobber bug (5+ entry equality) - Fix UNPACK_SEQUENCE: raise ValueError on count mismatch Phase 1 — core language gaps: - String unpacking: a, b, c = "xyz" - __ne__ auto-derivation from __eq__ for user classes - dict() kwargs and iterable support: dict(x=1), dict([(k,v)]) - Set <=, >=, <, > operators (subset/superset) Phase 2 — 31 new exception types: GeneratorExit, ModuleNotFoundError, SyntaxError, EOFError, UnicodeDecodeError/EncodeError, ConnectionError family, PermissionError, IsADirectoryError, FloatingPointError, BufferError, SystemError, warning subtypes, and more Phase 3 — Ellipsis singleton + breakpoint() stub Phase 4 — pure Python stdlib modules: abc, operator, string, io (StringIO/BytesIO), contextlib, copy (copy/deepcopy), collections (Counter, defaultdict, ChainMap, namedtuple, OrderedDict) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.