pytest: 4.4.0 regression for pdbcls
In my code I’m trying to define my own debugger class, which in this case is a bit of a hack to use ipdb I have a file with:
class Debugger(object):
quitting = None
def set_trace(self, frame):
import ipdb # noqa
return ipdb.set_trace(frame) # noqa
def reset(self, *args, **kwargs):
from ipdb.__main__ import _init_pdb
pdb_obj = _init_pdb()
pdb_obj.botframe = None # not sure. exception otherwise at quit
return pdb_obj.reset(*args, **kwargs)
def interaction(self, *args, **kwargs):
from ipdb.__main__ import _init_pdb
pdb_obj = _init_pdb()
pdb_obj.botframe = None # not sure. exception otherwise at quit
return pdb_obj.interaction(*args, **kwargs)
The code is now unable to import any class in my codebase at that step. I just get No module name X for every module in my local repository.
About this issue
- Original URL
- State: closed
- Created 5 years ago
- Comments: 17 (10 by maintainers)
Commits related to this issue
- pdb: try to import --pdbcls in pytest_configure only Fixes https://github.com/pytest-dev/pytest/issues/5039. — committed to blueyed/pytest by blueyed 5 years ago
- pdb: try to import --pdbcls in pytest_configure only Fixes https://github.com/pytest-dev/pytest/issues/5039. — committed to blueyed/pytest by blueyed 5 years ago
- pdb: try to import --pdbcls in pytest_configure only Fixes https://github.com/pytest-dev/pytest/issues/5039. — committed to blueyed/pytest by blueyed 5 years ago
- pdb: try to import --pdbcls in pytest_configure only Fixes https://github.com/pytest-dev/pytest/issues/5039. — committed to blueyed/pytest by blueyed 5 years ago
- Update pytest to 4.4.1 (#337) This PR updates [pytest](https://pypi.org/project/pytest) from **4.4.0** to **4.4.1**. <details> <summary>Changelog</summary> ### 4.4.1 ``` ... — committed to zdict/zdict by pyup-bot 5 years ago
- Merge #67 67: Update pytest to 4.4.1 r=rehandalal a=pyup-bot This PR updates [pytest](https://pypi.org/project/pytest) from **4.4.0** to **4.4.1**. <details> <summary>Changelog</summary> ... — committed to rehandalal/therapist by bors[bot] 5 years ago
- Merge #1869 1869: Scheduled weekly dependency update for week 16 r=rehandalal a=pyup-bot ### Update [botocore](https://pypi.org/project/botocore) from **1.12.130** to **1.12.134**. <details> ... — committed to mozilla/normandy by bors[bot] 5 years ago
- Merge #1869 1869: Scheduled weekly dependency update for week 16 r=mythmon a=pyup-bot ### Update [botocore](https://pypi.org/project/botocore) from **1.12.130** to **1.12.134**. <details> <... — committed to mozilla/normandy by bors[bot] 5 years ago
- Merge #32 32: Pin pytest to latest version 5.1.3 r=duckinator a=pyup-bot This PR pins [pytest](https://pypi.org/project/pytest) to the latest release **5.1.3**. <details> <summary>Changelog</s... — committed to duckinator/bork by bors[bot] 5 years ago
- Merge #1 1: Initial Update r=aragilar a=pyup-bot This PR sets up pyup.io on this repo and updates all dependencies at once, in a single branch. Subsequent pull requests will update one dependency... — committed to aragilar/pytest-info-collector by bors[bot] 4 years ago
- Merge #1 1: Initial Update r=aragilar a=pyup-bot This PR sets up pyup.io on this repo and updates all dependencies at once, in a single branch. Subsequent pull requests will update one dependency... — committed to aragilar/spaceplot by bors[bot] 4 years ago
- Merge #1 1: Initial Update r=aragilar a=pyup-bot This PR sets up pyup.io on this repo and updates all dependencies at once, in a single branch. Subsequent pull requests will update one dependency... — committed to aragilar/venv_tools by bors[bot] 4 years ago
- Merge #5 5: Initial Update r=aragilar a=pyup-bot This PR sets up pyup.io on this repo and updates all dependencies at once, in a single branch. Subsequent pull requests will update one dependency... — committed to aragilar/root-solver by bors[bot] 4 years ago
Yeah, I’m glad the new feature works for you - and somehow glad that it broke your current setup, since you might not have noticed otherwise… 😃 #5041 should fix it. I have some stashed change to make this more lazily even, but that is for later.
Thanks, and thanks for the suggestion about using
ipdbdirectly instead based on the new version. In this case it just simplified what I was already doing.