pyanalyze.name_check_visitor

The core of the pyanalyze type checker.

NameCheckVisitor is the AST visitor that powers pyanalyze’s type inference. It is the central object that invokes other parts of the system.

class pyanalyze.name_check_visitor.ComprehensionLengthInferenceLimit(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

If we iterate over something longer than this, we don’t try to infer precise types for comprehensions. Increasing this can hurt performance.

class pyanalyze.name_check_visitor.UnionSimplificationLimit(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

We may simplify unions with more than this many values.

class pyanalyze.name_check_visitor.DisallowCallsToDunders(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

Set of dunder methods (e.g., ‘{“__lshift__”}’) that pyanalyze is not allowed to call on objects.

class pyanalyze.name_check_visitor.DisallowedImports(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

List of imports that will trigger an error.

Entries may be top-level modules (e.g., “os”) or dotted submodule paths (e.g., “os.path”).

class pyanalyze.name_check_visitor.ForLoopAlwaysEntered(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

If True, we assume that for loops are always entered at least once, which affects the potentially_undefined_name check. This will miss some bugs but also remove some annoying false positives.

class pyanalyze.name_check_visitor.IgnoreNoneAttributes(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

If True, we ignore None when type checking attribute access on a Union type.

class pyanalyze.name_check_visitor.UnimportableModules(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

Do not attempt to import these modules if they are imported within a function.

class pyanalyze.name_check_visitor.ExtraBuiltins(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

Even if these variables are undefined, no errors are shown.

class pyanalyze.name_check_visitor.IgnoredPaths(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

Attribute accesses on these do not result in errors.

class pyanalyze.name_check_visitor.IgnoredEndOfReference(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

When these attributes are accessed but they don’t exist, the error is ignored.

class pyanalyze.name_check_visitor.IgnoredForIncompatibleOverride(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

These attributes are not checked for incompatible overrides.

class pyanalyze.name_check_visitor.IgnoredUnusedAttributes(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

When these attributes are unused, they are not listed as such by the unused attribute finder.

class pyanalyze.name_check_visitor.IgnoredUnusedClassAttributes(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

List of pairs of (class, set of attribute names). When these attribute names are seen as unused on a child or base class of the class, they are not listed.

class pyanalyze.name_check_visitor.CheckForDuplicateValues(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

For subclasses of these classes, we error if multiple attributes have the same value. This is used for the duplicate_enum check.

class pyanalyze.name_check_visitor.AllowDuplicateValues(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

For subclasses of these classes, we do not error if multiple attributes have the same value. This overrides CheckForDuplicateValues.

class pyanalyze.name_check_visitor.IgnoredTypesForAttributeChecking(value: T, applicable_to: Tuple[str, ...] = (), from_command_line: bool = False, priority: int = 0)

Used in the check for object attributes that are accessed but not set. In general, the check will only alert about attributes that don’t exist when it has visited all the base classes of the class with the possibly missing attribute. However, these classes are never going to be visited (since they’re builtin), but they don’t set any attributes that we rely on.

class pyanalyze.name_check_visitor.ClassAttributeChecker(*, enabled: bool = True, should_check_unused_attributes: bool = False, should_serialize: bool = False, options: Options = Options(options={}, module_path=()), ts_finder: TypeshedFinder | None = None)

Helper class to keep track of attributes that are read and set on instances.

record_attribute_read(typ: type, attr_name: str, node: AST, visitor: NameCheckVisitor) → None

Records that attribute attr_name was accessed on type typ.

record_attribute_set(typ: type, attr_name: str, node: AST, value: Value) → None

Records that attribute attr_name was set on type typ.

record_class_examined(cls: type) → None

Records that we examined the attributes of class cls.

serialize_type(typ: type) → object

Serialize a type so it is pickleable.

We do this to make it possible to pass ClassAttributeChecker objects around to parallel workers.

get_attribute_value(typ: type, attr_name: str) → Value

Gets the current recorded value of the attribute.

check_attribute_reads() → None

Checks that all recorded attribute reads refer to valid attributes.

This is done by checking for each read whether the class has the attribute or whether any code sets the attribute on a class instance, among other conditions.

check_unused_attributes() → None

Attempts to find attributes.

This relies on comparing the set of attributes read on each class with the attributes in the class’s __dict__. It has many false positives and should be considered experimental.

Some known causes of false positives:

• Methods called in base classes of children (mixins)

• Special methods like __eq__

• Insufficiently powerful type inference

class pyanalyze.name_check_visitor.StackedContexts

Object to keep track of a stack of states.

This is used to indicate all the AST node types that are parents of the node being examined.

add(value: AST) → Iterator[None]

Context manager to add a context to the stack.

class pyanalyze.name_check_visitor.CallSiteCollector

Class to record function calls with their origin.

class pyanalyze.name_check_visitor.NameCheckVisitor(filename: str, contents: str, tree: Module, *, settings: Mapping[Error, bool] | None = None, fail_after_first: bool = False, verbosity: int = 50, unused_finder: UnusedObjectFinder | None = None, module: ModuleType | None = None, attribute_checker: ClassAttributeChecker | None = None, collector: CallSiteCollector | None = None, annotate: bool = False, add_ignores: bool = False, checker: Checker, is_code_only: bool = False)

Visitor class that infers the type and value of Python objects and detects errors.

config_filename: ClassVar[str | None] = None

Path (relative to this class’s file) to a pyproject.toml config file.

has_used_any_match() → bool

Whether Any was used to secure a match.

record_any_used() → None

Record that Any was used to secure a match.

reset_any_used() → ContextManager[None]

Context that resets the value used by has_used_any_match() and record_any_match().

set_exclude_any() → ContextManager[None]

Within this context, Any is compatible only with itself.

should_exclude_any() → bool

Whether Any should be compatible only with itself.

check(ignore_missing_module: bool = False) → List[Failure]

Run the visitor on this module.

visit(node: AST) → Value

Visit a node and return the pyanalyze.value.Value corresponding to the node.

resolve_name(node: Name, error_node: AST | None = None, suppress_errors: bool = False) → Tuple[Value, FrozenSet[object | None]]

Resolves a Name node to a value.

Parameters:

• node (ast.AST) – Node to resolve the name from

• error_node (Optional[ast.AST]) – If given, this AST node is used instead of node for displaying errors.

• suppress_errors (bool) – If True, do not produce errors if the name is undefined.

compute_function_info(node: FunctionDef | AsyncFunctionDef | Lambda, *, is_nested_in_class: bool = False, enclosing_class: TypedValue | None = None, potential_function: object | None = None) → Generator[FunctionInfo, None, None]

Visits a function’s decorator list.

get_attribute(root_composite: Composite, attr: str, node: AST | None = None, *, ignore_none: bool = False, use_fallback: bool = False, prefer_typeshed: bool = False) → Value

Get an attribute of this value.

Returns pyanalyze.value.UNINITIALIZED_VALUE if the attribute cannot be found.

varname_for_constraint(node: AST) → VarnameWithOrigin | None

Given a node, returns a variable name that could be used in a local scope.

varname_for_self_constraint(node: AST) → VarnameWithOrigin | None

Helper for constraints on self from method calls.

Given an ast.Call node representing a method call, return the variable name to be used for a constraint on the self object.