gh-144316: Fix missing exception in _remote_debugging with debug=False #144442
Conversation
taegyunkim
force-pushed
the
gh-144316-get_stack_trace
branch
from
7d1bb63 to
0f2d244
Compare
taegyunkim
force-pushed
the
gh-144316-get_stack_trace
branch
from
cffe58f to
66b060d
Compare
Misc/NEWS.d/next/Library/2026-02-03-19-57-41.gh-issue-144316.wop870.rst
Outdated
Show resolved
Hide resolved
|
This macro is intended to add context to an exception not to be the actual cause of the error. Doing this will mask actual bugs in the code where we are not setting exceptions properly in places that should catch it and fail. Unfortunately this is not only is a partial fix but it will hide errors. The correct solution is finding the error paths that are not properly handling errors and fix there (and perhaps chain). |
|
@pablogsal Found |
picnixz
left a comment
picnixz
left a comment
There was a problem hiding this comment.
I think we should instead add assert(PyErr_Occurred()) in _set_debug_exception_cause instead.
|
Please, stop updating the branch when there is nothing to do. It wastes CI resources and notifies everyone subscribed to the thread. See https://devguide.python.org/getting-started/pull-request-lifecycle/#update-branch-button. |
3 participants
PyErr_NoMemory()call whenPyMem_RawMalloc()returnedNULL.PyErr_Format()calls before callingset_exception_causemacroset_exception_causemacro to assert an error is set.RemoteUnwinder.get_stack_traceis missing exceptions on some error code paths #144316