diff options
| author | Claudiu Popa <pcmanticore@gmail.com> | 2018-12-30 16:21:10 +0100 |
|---|---|---|
| committer | Claudiu Popa <pcmanticore@gmail.com> | 2018-12-30 16:21:10 +0100 |
| commit | 74d79ef6eb7d5354f63507128fea0ef524e69a01 (patch) | |
| tree | 3051687f38e3f5461cc12327ad703eface89a88b /astroid/context.py | |
| parent | 60bf287a37358792277b5c1a0ba6821e579b97da (diff) | |
| download | astroid-git-74d79ef6eb7d5354f63507128fea0ef524e69a01.tar.gz | |
Revert path copying in InferenceContext()
While copying the path solves some inference issues, it also leads to
horrenduous performance hits on large libraries such as pandas.
We still want to have the path copying, but we need a better understanding
on why that is needed, as well as a change on how we operate with the inference
contexts.
Diffstat (limited to 'astroid/context.py')
| -rw-r--r-- | astroid/context.py | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/astroid/context.py b/astroid/context.py index 931bbf5e..2cd724fd 100644 --- a/astroid/context.py +++ b/astroid/context.py @@ -99,7 +99,7 @@ class InferenceContext: starts with the same context but diverge as each side is inferred so the InferenceContext will need be cloned""" # XXX copy lookupname/callcontext ? - clone = InferenceContext(set(self.path), inferred=self.inferred) + clone = InferenceContext(self.path, inferred=self.inferred) clone.callcontext = self.callcontext clone.boundnode = self.boundnode clone.extra_context = self.extra_context |