Reverse preset color order #2568
Conversation
rosteen
requested review from
pllim,
bmorris3,
haticekaratay and
cshanahan1
as code owners
Add changelog
rosteen
force-pushed
the
reverse-rgb-order
branch
from
ff4d96f to
103acaa
Compare
| preset_colors = [preset_colors[0], preset_colors[3]] | ||
| preset_colors = [preset_colors[1], preset_colors[4]] | ||
| elif n_visible == 3: | ||
| preset_colors = [preset_colors[0], preset_colors[2], preset_colors[3]] | ||
| preset_colors = [preset_colors[1], preset_colors[2], preset_colors[4]] |
There was a problem hiding this comment.
this surprised me at first (I expected just a change to preset_colors), but I guess this is just to avoid the purple on the end and the purple switched sides?
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #2568 +/- ##
==========================================
- Coverage 90.84% 90.79% -0.05%
==========================================
Files 160 160
Lines 19357 19356 -1
==========================================
- Hits 17584 17575 -9
- Misses 1773 1781 +8 ☔ View full report in Codecov by Sentry. |
|
I am surprised that it does not go to red now... |
Updated. |
|
The changleog failure seems to be due to a docker build failure in the CI infrastructure, not related. @kecnry cool if I hit merge? |
| if n_visible >= 2 and n_visible < len(preset_colors): | ||
| preset_colors = [preset_colors[i] for i in preset_inds[n_visible]] | ||
| elif n_visible > len(preset_colors): | ||
| cmap = matplotlib.colormaps['gist_rainbow'].resampled(n_visible) | ||
| preset_colors = [matplotlib.colors.to_hex(cmap(i), keep_alpha=True) for |
There was a problem hiding this comment.
this might now read better and be harder to break in the opposite order (if n_visible > len(preset_color): ... elif n_visible > 2: ...)
There was a problem hiding this comment.
ah, I wasn't thinking about the == case, so doesn't really make a difference I guess 🤷♂️
There was a problem hiding this comment.
Well, I'm not changing it back!
As requested in #2558 (comment).