Refactor colorbar loc handling#304
Conversation
beckermr
left a comment
There was a problem hiding this comment.
Do we want to add a test for this by adding an image to our image comparison tests?
|
Yes. Hold on. It's a bit tricky as the vert bars suffer from looking bad if you open the figure in gtk but look good on save |
|
Was drafting a release and noticed this |
Codecov Report❌ Patch coverage is
📢 Thoughts on this report? Let us know! |
|
Need to fix the framing for the inset |
|
Ok I sat down and went through all the edge cases but I think I got it looking relatively nice now. I am afraid, however, that the tests will indicate a visual difference since the logic is now concerning all the locations for insets. |
|
The resulting tests look good but differ in shape. Not sure why but likely related to the newly added refactors. Otherwise visually looks good. |
|
See above for changes @beckermr |
|
Can we go ahead an merge this @beckermr? I manually checked the errors and there is an image dimension error but visually they look similar. |
|
yes merge away! |
minor adjustment of #301This PR is now a refactor of how the label positioning is handled. It moves the ugly block to a much easier to read helper function with some internal checking.
Important changes:
I will need to add some tests to test this properly and still adjust the bbox handling with this new approach.