Repository navigation
Added test for tricontour - #27545
Added test for tricontour #27545szhukovska wants to merge 1 commit into
Conversation
Sorry, something went wrong.
There was a problem hiding this comment.
Thank you for opening your first PR into Matplotlib!
If you have not heard from us in a week or so, please leave a new comment below and that should bring it to our attention. Most of our reviewers are volunteers and sometimes things fall through the cracks.
You can also join us on gitter for real-time discussion.
For details on testing, writing docs, and our review process, please see the developer guide
We strive to be a welcoming and open project. Please follow our Code of Conduct.
Sorry, something went wrong.
|
Thank you for your interest in improving Matplotlib! However, this portion of that issue has already received a PR (#27494). And in looking into it we discovered that the tri* plots are not actually properly handling unit-ful inputs, so we raised #27499. Your plot is hiding these failures because it a) calls plot so the axis units get set, when they are not properly set by tricontour alone (but should be) and b) happen to "work" because they use datetime64[D] dtype which coincides with matplotlib's floating point representation of dates for axes, but it does not actually pass through the units code so if you used a different scale (e.g. datetime64[s]) then you would get the wrong thing plotted. (or indeed changed the epoch date for mpl from 1970-01-01) |
Sorry, something went wrong.
|
Alright, thanks for the quick feedback! |
Sorry, something went wrong.
PR summary
Added 3 tests for tricontour (issue #26864): datetime on y only, on x only and on x and y axes
PR checklist