Update test for Bokeh 3.4

[hoxbro]

Noticed here:

https://github.com/holoviz/holoviews/actions/runs/8020821711/job/21924068272

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 85.71429% with 2 lines in your changes are missing coverage. Please review.

Project coverage is 23.79%. Comparing base (a66eb55) to head (e98bef2).

Files	Patch %	Lines
...oviews/tests/plotting/bokeh/test_annotationplot.py	75.00%	1 Missing ⚠️
holoviews/tests/ui/bokeh/test_callback.py	88.88%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##             main    #6130       +/-   ##
===========================================
- Coverage   88.56%   23.79%   -64.77%     
===========================================
  Files         316      316               
  Lines       66010    66013        +3     
===========================================
- Hits        58460    15710    -42750     
- Misses       7550    50303    +42753     

Flag	Coverage Δ
ui-tests	23.79% <85.71%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[hoxbro]

It won't solve correctly with the dev version of panel + bokeh for unit test cases. This will likely solve itself with the official release of both.

I have run the examples and unit test locally and got no error with this PR.

The reason for the difference in the lasso select UI test in Bokeh 3.4 is unclear to me. It could be related to the floating UI pr.

I would like to merge this PR, as the UI test is failing.

[hoxbro]

Now, we also test the unit test with the latest panel and Bokeh dev release.

Will add the check latest release in a follow up PR, when the next version of Panels dev release have been made.

[hoxbro]

@mattpap, do you know what causes the change in lasso_select between 3.3 and 3.4? And if the change is expected and therefore okay?

[mattpap]

do you know what causes the change in lasso_select between 3.3 and 3.4? And if the change is expected and therefore okay?

Gesture and UI event handling was redesigned, among other things. It seems OK and aligned with the difference between old and new approaches.

[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.