Releases: microsoft/pyright
Releases Β· microsoft/pyright
Published 1.1.369
Behavior Changes:
- Improved consistency of unreachable code. Previously, unreachable code was not supported for
if
orelse
suites when the condition type was narrowed toNever
. - Changed hover provider output for type aliases to conform more closely with the new Python 3.12
type
syntax.
Enhancements:
- Added support for detecting mismatched property types in base classes used for multiple inheritance.
- Improved
reportUninitializedInstanceVariable
check to handle dataclass variables that are initialized implicitly by the synthesized__init__
method. - Added support for walrus (assignment) operator within set literal expressions if the python version is >= 3.10. The grammar was changed in 3.10 to support this.
Bug Fixes:
- Fixed recent regression in
reportUninitializedInstanceVariable
check relating toNamedTuple
instance variables. - Fixed bug that leads to a false negative when a subscript object used in an index expression supports
__index__
but the indexed object specifies that it requiresint
. - Fixed recent regression that results in a runtime assert (and therefore a crash) in certain circumstances when computing the MRO for a class.
- Fixed bug that results in an incorrect overload implementation diagnostic when using
tuple[()]
in overloads. - Improved diagnostic messages for type errors detected during
__set__
and__delete__
calls to a descriptor object. - Fixed bug that results in a false positive error when using a callback protocol with a function-scoped type variable.
- Fixed bug that incorrectly evaluates a constructor call to a constructor that infers a class-scoped ParamSpec when the passed function is generic.
- Fixed bug that leads to a spurious "unbound variable" diagnostic when a variable is assigned a value using a walrus operator and is later used in a
**
call argument. - Fixed bug that results in incorrect type narrowing in the negative (fall-through) case of a match expression when the subject expression is an unbounded tuple.
Published 1.1.368
Bug Fixes:
- Fixed a bug that causes
warnings.deprecated
not to work if the python version is 3.13. This was caused by recent changes to typeshed stubs. - Fixed bug that results in inconsistent diagnostics when surrounding an argument expression with parentheses.
- Fixed bug that causes a "bare"
type
(with no type argument) ortype[Any]
to be treated as a possible descriptor object. - Fixed regression that results in an internal assertion (and crash) related to TypeVar variance inference.
- Fixed a bug in the "type printer" (the component responsible for translating a type into its textual form) for type aliases defined using PEP 695 syntax.
- Fixed bug that resulted in the incorrect synthesis of
__dataclass_fields__
for named tuples. - (Contributed by @insync) Fixed
stubPath
's default value in configuration schema. - Fixed bug in pattern matching for
Callable()
in the negative (fall-through) case. - Fixed bug that results in unions consisting of different specialized forms of a generic TypedDict to "lose" all but one of these subtypes.
Behavior Changes:
- Changed inference behavior for generator functions with no reachable
yield
oryield from
statements. The "yield type" (the first type argument toGenerator
) is now inferred asNever
in this case rather thanNone
. - Changed the behavior of double underscored symbols that are part of a module's namespace when that module is not a stub or in a "py.typed" package. Such symbols are no longer ignored, although they are considered private.
- Changed the way that PEP 695 type aliases are displayed when hovering over a reference to the type alias. Previously, the symbol was displayed as
TypeAliasType
rather than the expanded type alias type.
Enhancements:
- Added support for
deprecated
objects that are instantiated prior to being used as a decorator. This allows for a factory usage pattern. - Added check for illegal use of
Protocol
type argument that is not a type parameter. The runtime generates an exception for this condition.
Published 1.1.367
Bug Fixes:
- Fixed bug that results in incorrect truthy/falsy type narrowing for an enum that overrides
__bool__
. - Fixed bug that results in a false positive when accessing a generic attribute in a base class from a subclass that explicitly specializes the generic type.
- Fixed bug that results in duplicate and (contradictory) diagnostic outputs in certain cases where
reveal_type
is used in a loop. - Fixed a bug that results in incorrect (local) type evaluation of an instance variable with a declared type when assigned within a loop that uses an augmented assignment.
- Fixed bug that causes
Final
annotation not to be honored when wrapped in anAnnotated
annotation. - Fixed bug that results in a false positive error under certain circumstances involving tuple type narrowing in a loop.
- Fixed bug that led to divergent behaviors between pyrightconfig.json and pyproject.toml configurations. Specifically, default exclude patterns were not added in the case of pyproject.toml. Thanks to @augustelalande for contributing to this fix.
- Fixed bug that results in parameter types being converted to
Any
when converting aNewType
or dataclass constructor to a callable.
Enhancements:
- Updated typeshed stubs to the latest version.
- Implemented provisional support for PEP 746, which provides consistency checks for metadata used in an
Annotated
annotation. - Added check for
except*
to enforce that exception class does not derive fromBaseExceptionGroup
. - Added the ability to override type checking configuration settings for each execution environment. This allows, for example, a test directory to use different settings than the directories that contain production code.
Behavior Changes:
- Added extra blank line to JSON output to assist with parsing when used in watch mode.
- Changed the import resolution order to better reflect runtime behavior and match a proposed change to the typing spec. In particular, stdlib typeshed stubs are now resolved prior to
site_packages
.
Published 1.1.366
Bug Fixes:
- Fixed bug that results in a false positive error under certain circumstances involving a call to a higher-order generic function with another function that involves a class with a contravariant type variable.
- Fixed bug that results in a confusing error message when stub generation fails.
- Fixed bug that results in false positive when evaluating the call to a generic function that involves a type variable used in both a covariant and contravariant (or invariant) position and both literal and non-literal types are involved.
- Fixed bug that results in a false negative when solving a type variable that involves an invariant context.
- Fixed a bug that can result in incorrect type inference for a subexpression that is part of an assignment expression if executed with an inference context (bidirectional type inference).
- Fixed bug in
x is E
type narrowing pattern whereE
is an enum literal andx
is a supertype ofE
likeobject
.
Behavior Changes:
- Changed behavior of constructor call evaluation when
__new__
evaluation results in type evaluation errors. Previously, pyright skipped the__init__
evaluation in this case. It now proceeds to evaluate__init__
, potentially generating redundant errors. - Removed support for inlined TypedDict annotations. This was an experimental feature that never gained enough support to make it into a PEP.
- Changed behavior of pyright to match the latest typing spec when it encounters an attribute with a type annotation within an
Enum
class body in a stub. These are now treated as non-member attributes rather than members. Typeshed stubs have been updated to conform to the new standard. - Added diagnostic check for an enum member with a type annotation. The typing spec says that this should be considered a typing error.
Enhancements:
- Updated typeshed stubs to the latest version.
Published 1.1.365
Bug Fixes:
- Fixed regression that results in the incorrect type evaluation for the
value
attribute of aStrEnum
orIntEnum
class. - Fixed a bug that results in incorrect type compatibility checks for a callable that uses
*args: *tuple[T, *tuple[S, ...]]
. - Fixed a bug that results in an unsolved TypeVar when calling
list(tuple())
. - Fixed bug that leads to inconsistent narrowing-on-assignment behavior when a
tuple
instance with unknown type arguments is assigned to a target symbol with a declaredtuple
type. - Fixed bug that results in a false negative when a
Final
class variable is assigned a value in the class body and reassigned a value in the__init__
method.
Behavior Changes:
- Changed tuple expression inference behavior to not preserve literal entry types if the tuple expression is embedded within another tuple, set, list, or dictionary expression.
- [Contribution by @ujan-r] Use typeshed version info for stdlib module list.
Enhancements:
- Improved diagnostic messages for parameter mismatch.
- Added support for configuration inheritance through an "extends" configuration option.
- Added support for custom ".toml" files specified via the
-p
option on the command line.
Published 1.1.364
Bug Fixes:
- Fixed bug that results in a false negative syntax error when a PEP-695 type alias uses a keyword as a name.
- Fixed a bug that leads to a false positive error when first argument to
super
call is an instance of a metaclass. - Fixed recent regression that results in a false positive when accessing an enum member from an instance of an enum class.
- Fixed bug that results in a false negative when determining if a callable type is compatible with another callable type and the first has a
*args
parameter and the second has a single positional+keyword parameter. - Fixed a bug in the code flow engine that results in incorrect type evaluation in certain cases involving codependent variables in a loop.
- Fixed a bug that results in incorrect type narrowing for a type guard function that uses
TypeIs[type[T]]
. - Fixed regression related to "converter" field parameter for dataclasses when the converter refers to a generic class constructor.
- Fixed a bug that results in a false negative when a generic function has a parameter with a generic type and a default argument value. This could lead to a typing hole when considering subtyping rules for callables.
- Fixed a bug in the dataclass "converter" support that resulted in an error if the converter is
tuple
. - Fixed regression that results in incorrect type evaluation for annotations that involve nested protocols (such as
P[P[T]]
). - Fixed a bug that resulted in an infinite loop when a function that uses an
*args: *Ts
parameter is passed as an argument to itself. - Fixed bug that leads to a false negative when calling a callback function that returns a callable with a generic parameter type.
Behavior Changes:
- Changed the handling of
Enum
subclasses that explicitly overridevalue
orname
to avoid using the special-case logic for computing these types. This is done regardless of whether the class uses a custom metaclass. - Changed support for dataclass "converter" field parameter to be off by default. The "enableExperimentalFeatures" must now be enabled to use this feature. If and when PEP 712 is approved, it will be moved out of experimental.
- Added support for using
Literal
and various other special forms in a value expression.
Enhancements:
- Added support for recent addendum to typing spec that provides a special case for dataclasses so
ClassVar
andFinal
can be used together to distinguish between a regularFinal
instance variable and aClassVar
that is alsoFinal
. - Updated typeshed stubs to the latest version.
- Improved type printer (the component that converts an internal type to textual format for diagnostic messages) so it better handles special forms like
Literal
or unions when used in a value expression.
Published 1.1.363
Bug Fixes:
- Updated manifest to reflect the correct default value for
reportInvalidTypeForm
. - Fixed bug that resulted in diagnostics associated with the
reportUnnecessaryTypeIgnoreComment
check to not include the error code. - Fixed bug that causes a false positive
reportPrivateUsage
error when a global or nonlocal binding is used within a class to access an outer-scoped variable. - Fixed bug that results in a false positive
reportUnnecessaryComparison
when a variable is used in a conditional expression and its type is a union that includes aCoroutine
and a non-coroutine. - Fixed false positive
deprecatedTypingAliases
deprecation forcollections.abc.AbstractSet
due to a recent change in typeshed stubs. - Fixed recent regression that results in
assert_type
failure when a class is parameterized by a ParamSpec with a default value. - Improved handling of constrained TypeVars when used with unary operators.
- Fixed bug that results in incorrect type narrowing for a tuple with determinate length when indexed with an out-of-bounds slice.
- Fixed bug that results in incorrect type narrowing for
isinstance
orissubclass
type guard when the filter is a generic class whose type parameter has a default value. - Fixed bug that results in incorrect type evaluation behavior within class body of an enum class when one enum member is used to define another enum member. This involved a significant rewrite of the logic involving enum symbol evaluation.
- Fixed a bug that results in a false positive error when a decorator is applied to a property.
- Fixed bug that resulted in false positive when evaluating a function call involving a TypeVarTuple under certain circumstances.
- Fixed a bug that results in incorrect type evaluation when a function parameterized with a TypeVarTuple is passed as an argument to another function parameterized with a TypeVarTuple.
Behavior Changes:
- Deferred execution of type annotations (PEP 649) did not make it into Python 3.13 and has been deferred (again) until Python 3.14. Changing pyright's logic accordingly.
- Changed the type evaluation for
type(Any)
ortype(x)
wherex
is a value of typeAny
orUnknown
. These now evaluate totype[Unknown]
. - Changed the TypeVar scoping behavior for TypeVars that appear only within a Callable subtype within a function's declared return type. Previously, pyright scoped these TypeVars to the Callable rather than the function. This was due to issue #4285. I'm reversing this decision because it's not spec-compliant, creates a bunch of unnecessary complexity, and results in buggy behavior.
Published 1.1.362
Bug Fixes:
- Fixed a bug that results in incorrect type narrowing when a
match
statement uses an empty (zero-element) sequence pattern and the subject expression type is a tuple that potentially (but does not always) have a zero length. - Fixed a crashing bug that results from an internal assertion failure when a generic class that uses PEP 695 syntax is located within an unreachable code block.
- Fixed a bug that results in incorrect type narrowing in the negative (fall-through) case when a
match
statement includes a class pattern with a runtime-checkable protocol class. - Fixed recent regression with the
TypeIs
type guard that resulted in incorrect type narrowing in the negative (else) case. - Fixed a bug that results in incorrect type evaluation for a variable that uses a nonlocal or global binding within an inner scope. Such a variable should never honor the narrowed type from the outer scope.
- Fixed false positive when a TypeVar appears within a function type comment.
- Fixed incorrect type narrowing for a class pattern argument if the class is a generic whose type parameters have default values.
- Fixed recent regression that resulted in incorrect type narrowing when a
*args: P.args
parameter was used in the expressionlen(args) >= x
. - Fixed a bug that resulted in a false negative when a
Literal
with multiple arguments is used in a value expression. It should be treated as aUnionType
special form in this case.
Enhancements:
- Addressed a performance issue when indexing into a value that has a type consisting of a union of many (dozens or hundreds) of tuple types.
- Updated typeshed stubs to the latest version.
Behavior Changes:
- Changed interpretation of function definitions with (*args: Any, **kwargs: Any) to be
...
in compliance with latest typing spec. - Added special-case handling for
__init_subclass__
checks for classes that haveABCMeta
as a metaclass. Previously, these were exempted from__init_subclass__
becauseABCMeta
has a custom__new__
method, but we know that this metaclass calls through totype.__new__
which calls__init_subclass__
.
Published 1.1.361
Bug Fixes:
- Fixed a bug that results in a false negative when using bidirectional type inference to evaluate the type of a lambda expression if the expected type is a union of two or more callables.
- Fixed bug caused by incorrect synthesis of
item
andkeys
attributes for named tuple classes. - Fixed bug that resulted in incorrect bidirectional type inference when calling a constructor for a generic class that defines no
__new__
or__init__
method. - Fixed bug that results in a false positive error when a
yield
statement is used within a lambda following a function. - Fixed a bug that led to poor performance (effectively a hang) if an unannotated function is called through many (greater than 8) call expressions within a loop using different literal-value arguments each time.
- Fixed a bug that results in a false positive error when using
__extra_items__
(PEP 728) with aTypedDict
imported fromtyping_extensions
. - Fixed a bug that leads to an infinite loop when performing protocol matching under certain circumstances that involve recursive protocol definitions.
- Fixed a bug that results in a false positive error when
Callable()
is used as a class pattern and the subject type isAny
orUnknown
. - Fixed a false negative that occurs when reassigning a
Final
variable within a class body. - Changed the tokenizer so it is bug-for-bug compatible with the CPython tokenizer in versions 3.10 and newer in the case where a backslash (continuation character) is located by itself on a line.
- Fixed a bug that results in incorrect type evaluation when a global (module-scoped) variable is captured within an inner scope and the variable is not modified anywhere below the inner scope. It's possible in this case for the variable to be modified by code outside of the module.
- Fixed bug that resulted in incorrect evaluation of an identifier used in a method parameter type annotation if it shadows a forward reference in a class scope when the method is using PEP 695 type parameters.
- Fixed a bug that results in a false negative when an Unpack is used in a union. This isn't allowed.
- Fixed bug that produces incorrect output when converting to a textual representation the type
type[Foo]
whereFoo
is a type alias.
Behavior Changes:
- Changed logic that synthesizes the constructor methods for a NamedTuple subclass (using the class syntax) so the
__new__
method contains parameters based on field names and the__init__
method contains a general*args: Any, **kwargs: Any
parameter signature. Previously, this was swapped, but the new way is more faithful to the runtime. - Changed
TypeIs
to use the same logic asisinstance
type narrowing logic for consistency.
Published 1.1.360
Enhancements:
- Added support for
@deprecation
messages for decorators that wrap a function in a callable object. The deprecated message is now propagated through the ParamSpec and the__call__
method. - Updated typeshed stubs to the latest version.
Behavior Changes:
- Moved check for invalid use of class-scoped TypeVar within a
self
annotation in__init__
. It was previously reported underreportGeneralTypeIssues
, but it's now moved toreportInvalidTypeVarUse
. This was done to help typeshed maintainers migrate away from this pattern. - Changed behavior to no longer exempt unguarded access to not-required TypedDict member within a
try
orwith
block. Previously, such errors were exempt, but this is inconsistent with other type checks in pyright which eschew the practice of using exception handling for normal code flow.
Bug Fixes:
- Fixed recent regression that affected unannotated
__call__
methods in a metaclass. This change aligns pyright's behavior to the typing spec. - Fixed recent regression that incorrectly narrowed the type of
kwargs
when used in a type guard of the formif "a" in kwargs
. - Fixed bug in protocol matching that results in a false positive when the subject object is a dataclass that contains a callable. It should be considered an instance member in this case, so it should not be bound to the class.
- Fixed a bug that results in a false positive error when bidirectional type inference is used for a dictionary comprehension when the expected type is a union.
- Redesigned the handling of PEP 695-style type parameter scoping to better match the CPython runtime implementation. This fixes multiple bugs with type parameter symbol resolution. For example, if a type parameter is quoted as a forward reference.
- Fixed bug that leads to false negative when binding an object to an overloaded method where all of the overloads have incompatible
self
parameter types. - Fixed bug that leads to incorrect metatype for a class object that is created by calling the metaclass with a two-argument form.