-
-
Notifications
You must be signed in to change notification settings - Fork 4.4k
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
Remove pattern
dependency
#3012
Conversation
Trying to work around "has no attribute '__reduce_cython__'" problem
Why was it removed? Parts of the code still need it.
Is there anything left to do here? I think this may be ready to merge. |
Yes, looks ready, but all tests are failing. |
Those failures are unrelated to this PR. In between me making this PR one week ago, and the build triggered by your changes to segment_wiki.py yesterday, there has been a new sklearn release and they deprecated one of the modules we were relying on.
I'll disable the failing tests and open a ticket to deal with the problem properly. |
This is another case where for the sake of maintaining the tests a trustworthy indicator of whether the code is ready for release, I would not be so quick to disable them, even with a note-linking-to-a-pending-issue. If previous functionality is now broken due to outside library changes – and also broken on If test-disablement is chosen, because that functionality can be ignored for a while, it'd be best done in a standalone PR that's applied to |
Sure, I'll have a look at it, most likely in the New Year. I disabled the tests because it was least bad alternative out of the following list:
If you disagree with my decision, I can re-enable the tests in a separate PR. It's simple. |
If this PR's changes caused the test-failures (even via bumping dependency versions), I'd agree that merging it would be a problem. But, I'm pretty sure Given that, merging this PR, with that known-but-unrelated problem, wouldn't create any new problems for others, so I think (2) would be OK. With regard to passing tests, I think the precise policy could be: "any PR shouldn't break any tests that were working before the PR - but it doesn't have to fix everything that's currently broken". |
nltk
package is installed #2697