Fix clim_percentile

[maximlt]

The clim_percentile option was added in #4712 but not tested and contained two small bugs, the first one making the option never available basically. I've tested clim_percentile as I was trying to run the plotting guide of xarray and was wondering if we had something similar to their robust=True option.

I'd like to write a test for this now. I'm not sure how to obtain the ranges so here's what I came up with:

import holoviews as hv
import numpy as np

hv.extension('bokeh')

arr = np.random.rand(10,10)
arr[0, 0] = -100
arr[-1, -1] = 100
im = hv.Image(arr).opts(clim_percentile=True)

bk_renderer = hv.Store.renderers['bokeh']
plot = bk_renderer.get_plot(im)

low, high = plot.ranges[('Image',)]['z']['robust']

assert low > 0
assert high < 1

The second issue I have is, if this test is alright, where to save it?

[codecov-commenter]

Codecov Report

Merging #5495 (79f521a) into main (55ddbcb) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##             main    #5495   +/-   ##
=======================================
  Coverage   88.23%   88.23%           
=======================================
  Files         302      302           
  Lines       62211    62234   +23     
=======================================
+ Hits        54889    54914   +25     
+ Misses       7322     7320    -2     

Impacted Files	Coverage Δ
holoviews/plotting/plot.py	93.52% <100.00%> (+0.26%)	⬆️
holoviews/tests/plotting/bokeh/test_elementplot.py	96.82% <100.00%> (+0.03%)	⬆️
holoviews/ipython/display_hooks.py	59.70% <0.00%> (-0.61%)	⬇️
holoviews/tests/plotting/bokeh/test_layoutplot.py	100.00% <0.00%> (ø)
holoviews/tests/operation/test_operation.py	99.19% <0.00%> (+0.01%)	⬆️
holoviews/plotting/bokeh/element.py	89.39% <0.00%> (+0.06%)	⬆️
holoviews/operation/element.py	71.79% <0.00%> (+0.10%)	⬆️

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[jlstevens]

Looks good to me! Does look like the original implementation was rather buggy...

@maximlt Ready to merge? All tests are passing except on macos...

[jlstevens]

Ah, I guess you want to add some docs first!

[maximlt]

This was previously untested and I'd like to add at least a test before merging. The thing is that I'm not quite sure where to write the test, and if the test I suggest (in the OP) is right.

Sure why not documenting that properly. Where would you recommend adding that?

[jlstevens]

I'm ok with testings this as part of the bokeh plotting tests only. Technically the change applies to all backends, so arguably there should a similar test for matplotlib and plotly but I would say that is entirely optional.

The approach you've suggested above looks ok to me. Happy to see this fix merged as soon as the tests goes in!

[philippjfr]

The second issue I have is, if this test is alright, where to save it?

I'd suggest putting it in holoviews/tests/plotting/bokeh/test_elementplot.py:TestColorbarPlot

[hoxbro]

I will merge this as I want this fix into the next release.

[github-actions]

This pull request has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.