-
-
Notifications
You must be signed in to change notification settings - Fork 18.4k
CI/DOC: fix clipboard on travis #26103
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
Conversation
saurav2608
commented
Apr 15, 2019
- closes DOC: clipboard library missing in doc env on travis #26075
this is more complex than what I thought. I will relook tomorrow. |
Codecov Report
@@ Coverage Diff @@
## master #26103 +/- ##
==========================================
- Coverage 91.96% 91.95% -0.01%
==========================================
Files 175 175
Lines 52405 52405
==========================================
- Hits 48193 48189 -4
- Misses 4212 4216 +4
Continue to review full report at Codecov.
|
Codecov Report
@@ Coverage Diff @@
## master #26103 +/- ##
==========================================
+ Coverage 91.69% 91.72% +0.03%
==========================================
Files 174 174
Lines 50739 50741 +2
==========================================
+ Hits 46523 46544 +21
+ Misses 4216 4197 -19
Continue to review full report at Codecov.
|
I am unable to replicate this in any of my environments. Thus I am create PRs to investigate. |
@saurav2608 meaning the doc-build does not show any pyperclip error? |
closing as stale |
(reopened to try something out, based on the diff I noticed the use of |
@jreback OK? (if this passes, will add a comment explaining why / referencing this PR) |
so these are actually run? can u show that somewhere |
What do you mean with "so these are actually run?" ? |
@jorisvandenbossche we didn't notice that we should be running these (and weren't); I dont' think this has changed. We need a highly visible test that we can simply look in the logs and inspect (maybe just add a print statement to it). https://travis-ci.org/pandas-dev/pandas/jobs/532738283
|
Ah, so those tests have not been running on master for a while? (since the CI env refactor PR #24632), ok, I see now. But it seems that this PR actually solves that. They are running again (just not on the first travis build, but if you look eg at the second travis build: https://travis-ci.org/pandas-dev/pandas/jobs/532738284: no message about skipped tests anymore. On the first travis build, they have been skipped even before the CI env refactor). But not sure I understand why the first travis build (in the build matrix) would skip those and the second not. I don't see anything in their definition that can explain that (the clipboard tests are marked as 'single'). |
Ah, maybe because the first build has no qt installed and the second does (through matplotlib), travis-37.yml vs travis-36-locale.yaml. |
right I don't care that they run on all builds, but at least one. I think just add a print statement in the basic clip test might be enough here to provide us an easy visual reference. |
@jreback I added some print statements, but they don't show up in the travis logs (although we have |
@jorisvandenbossche then how can we be sure these are actually run? This is the entire problem. |
Well, you can say that for all other tests as well... |
maybe, but this is a continual issue, and we need a foolproof way to say these tests are actually run |
Will open an issue about that, but merging this since it is actually solving a problem |
@saurav2608 thanks for putting me on the right track with the DISPLAY change ! |
@jorisvandenbossche Absolutely. I wish I could have contributed more. I am totally swamped at work. |