gh-118605: Fix reference leak in FrameLocalsProxy #118607
Conversation
| @@ -306,7 +308,10 @@ framelocalsproxy_visit(PyObject *self, visitproc visit, void *arg) | |||
| static PyObject * | |||
| framelocalsproxy_iter(PyObject *self) | |||
| { | |||
| return PyObject_GetIter(framelocalsproxy_keys(self, NULL)); | |||
| PyObject* keys = framelocalsproxy_keys(self, NULL); | |||
There was a problem hiding this comment.
Probably should error-check this too, and the PyList_Append in that function could fail so should clean up and return NULL.
There was a problem hiding this comment.
Also, there is a lack of error checking in some new "frameobject.c" functions. I will create a separate issue for this.
There was a problem hiding this comment.
I fixed the framelocalsproxy_keys related checks. We can address the others in a seperate PR?
|
A Python core developer has requested some changes be made to your pull request before we can consider merging it. If you could please address their requests along with any other requests in other reviews from core developers that would be appreciated. Once you have made the requested changes, please leave a comment on this pull request containing the phrase |
|
Right, I actually caught that while I was working on the PR to add proper checks, but I'm happy to make it in this PR. |
I put it after to make the declarations stick together because the other two declarations is quick and without any memory allocation. I also understand why people would put the check immediately after the declaration, but in this specific case I'm leaning slightly more towards the current way. |
|
Doing it immediately has some advantages for people used to reading through this code base. But since I already set up auto-commit, let's keep it as is. |
|
I can polish it up in the next PR that does a bunch of fixes everywhere. |
Also add some error checks.
There are a couple of reference leaks in the implementation of PEP 667. Fix those and hopefully that's all.
test_bdbleaks references #118605