InferenceContext

class astroid.context.InferenceContext(path: set[tuple[nodes.NodeNG, str | None]] | None = None, nodes_inferred: list[int] | None = None)

Bases: object

Provide context for inference.

Store already inferred nodes to save time Account for already visited nodes to stop infinite recursion

boundnode: SuccessfulInferenceResult | None

The bound node of the given context.

e.g. the bound node of object.__new__(cls) is the object node

callcontext: CallContext | None

The call arguments and keywords for the given context.

clone() → InferenceContext

Clone inference path.

For example, each side of a binary operation (BinOp) starts with the same context but diverge as each side is inferred so the InferenceContext will need be cloned

constraints: dict[str, dict[nodes.NodeNG, set[constraint.Constraint]]]

The constraints on nodes.

extra_context: dict[SuccessfulInferenceResult, InferenceContext]

Context that needs to be passed down through call stacks for call arguments.

property inferred: _InferenceCache

Inferred node contexts to their mapped results.

Currently the key is (node, lookupname, callcontext, boundnode) and the value is tuple of the inferred results

is_empty() → bool

lookupname: str | None

The original name of the node.

e.g. foo = 1 The inference of ‘foo’ is nodes.Const(1) but the lookup name is ‘foo’

max_inferred = 100

property nodes_inferred: int

Number of nodes inferred in this context and all its clones/descendents.

Wrap inner value in a mutable cell to allow for mutating a class variable in the presence of __slots__

path

Path of visited nodes and their lookupname.

Currently this key is (node, context.lookupname)

push(node: nodes.NodeNG) → bool

Push node into inference path.

Allows one to see if the given node has already been looked at for this inference context

restore_path() → Iterator[None]