-
Notifications
You must be signed in to change notification settings - Fork 384
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Fix cassette context decorator nesting issues #108
Fix cassette context decorator nesting issues #108
Conversation
…tte can be called multiple times.
This reverts commit 366e2b7. Conflicts: tests/unit/test_cassettes.py
… that changes to the default settings on the vcr properly propogate.
2442115
to
5edc58f
Compare
lol @ three builds in a row failing. I have a good feeling about e1e08c7 though. I think it might actually pass. |
…ent attribute with hasattr.
76c185e
to
4868a63
Compare
…each cassette context instead of replacing the cassette on the existing mock connection.
…re_first_nesting. by using a single class and patching cassette on that class. Not a great solution :\
No need to apologize, it's great. In fact, are you interested in helping me maintain vcrpy?
Wow, that's another use case I never thought about before. Yeah, I guess having it work like a stack is the right way to do it.
OK, I could see this being useful. You would need to add a section to the README about how to use this feature, and the difference from
Definitely a minor version bump, we are adding new features. |
Yeah I'd love to help you out with the maintenance of vcrpy! What exactly would that entail? Good call on adding a section to the README for before_record_reponse. I'm going to take a swing at #109, but I sort of suspect that it could be a bit tricky so I might give up on getting it into this version. In that case I'd like to leave the test that I wrote for it in, so is it cool if I disable it for now with the reason set to the issue number? I don't think the issue is THAT serious because the use case of using a session across contexts is somewhat obscure, but the behavior is pretty surprising and I can contrive a few situations where you might want to do something like that. |
Also, how do you feel about adding https://github.com/mverteuil/pytest-ipdb (or just raw ipdb) as a py.test plugin. ipdb is just so much better than raw pdb. |
I just gave you commit access, just go ahead and fix bugs. If it's a new feature or a backwards incompatible change, run it by me first though. You'll probably have to ping me to upload to pypi if you want to do a new release.
Yes, sounds good to me, you can mark it xfail
Yup, it's a pretty bizarre bug, but I agree, it would be nice to have a fix. |
Cool, thanks! Do you still want me to merge with --no-ff for the sake of record keeping or are you okay with changes pushed directly to master? |
You can just push to master if you want, the changes in this pull request aren't exactly related anyway (new before_save_request feature, bugfixes, etc) |
… another failing test for kevin1024#109
…ted_before_first_nesting. by using a single class and patching cassette on that class. Not a great solution :\" This reverts commit 2bf23b2.
Cool. that sounds good. I think I have to push all of this together just because I've been touching a lot of the same files in the various commits ive had. I'm hoping to try to get everything done somewhat soon. If you could help me get the new version into pypy that would be great. Is there anything else that you'd want to get into a new version? |
Sure, just ping me when you're ready. A few things I like to do for releases:
I think this version has plenty (thanks for all your work!) - some bugfixes and also a new feature. |
9d54e96
to
2fd21e4
Compare
…md with description of `before_record_response`
2fd21e4
to
83211a1
Compare
…ven if import fails. Make _recursively_apply_get_cassette_subclass actually work with dictionaries.
…onnections in ConnectionPools." This reverts commit dc249b0. Conflicts: vcr/patch.py
55f3fa3
to
113c95f
Compare
…_nesting_issues Fix cassette context decorator nesting issues
Hey. I wasn't able to push to pypi because I don't have access. Can you do this for me? |
OK, pushed to pypi. Thanks for your work! |
Sorry for spamming you with pull requests.
This is getting kind of big (it looks works than it is, I promise) so just to make sure that we are keeping track, here is a list of all of the changes that are included in this pull request:
closes #102, #107, #109