start_backtracking suggestion #81
Conversation
| backtracking_causes = run_resolver([("a", {1, 2}), ("b", {1})]) | ||
| exception_causes = run_resolver([("a", {2}), ("b", {1})]) | ||
| assert exception_causes == backtracking_causes |
There was a problem hiding this comment.
I wanted to make it obvious that the backtracking causes are exactly the same as the failure when the "best" version is selected.
@uranusjr do you prefer I give an explicit value instead?
There was a problem hiding this comment.
This is better IMO if you want to express the causes should be the same.
|
I’m +1 to the idea, but we should use a different function name. It’s not at all clear what the difference between Maybe we can call this something like |
Co-authored-by: Tzu-ping Chung <uranusjr@gmail.com>
Co-authored-by: Tzu-ping Chung <uranusjr@gmail.com>
|
Uh, Travis is dead somehow. We should migrate to GitHub Action. |
|
Could you rebase to or merge the latest master branch so we can apply the new GitHub Action checks? |
…tart-backtracking
|
Linter error! |
sorry for that... :/ |
|
Nothing to be sorry about, the purpose of the CI being there to catch our errors 😛 |
|
Now test failures |
If I had run nox properly before committing, none of those problems would have occurred... anyway, I ran nox now and it passes locally, I hope that CI agrees with me. 🤞 |
|
@uranusjr Thanks for the merge! |
|
Either would do, but first we'll need a new release here since pip's vendoring mechanism only allows public release. Since this does not introduce any backward incompatibility, I'll find some time to do a release this week. |
|
0.8.1 released with this change. |
|
opened pypa/pip#10600 to incorporate the new version into pip. |
resolvelib changes to solve pypa/pip#10210.