Set minimal shiny version for a snapshots

[llrs-roche]

Pull Request

Fixes #341

• Sets minimal shiny version for a snapshot test as it seems shiny version affects snapshots
• In interactive testing I saw some warning related to modal_content, the second commit on R/verbatim_popup.R is to remove them and simplify a bit the code

[github-actions]

Code Coverage Summary

Filename                      Stmts    Miss  Cover    Missing
--------------------------  -------  ------  -------  ------------------------------------------------------
R/basic_table_args.R             23       0  100.00%
R/draggable_buckets.R            87      12  86.21%   116-121, 151-156
R/get_dt_rows.R                  13       0  100.00%
R/ggplot2_args.R                 49       0  100.00%
R/nested_closeable_modal.R       20      20  0.00%    83-103
R/optionalInput.R               255      79  69.02%   190, 315-389, 402-409, 411-416, 434, 436, 501, 583-596
R/panel_group.R                  39      39  0.00%    50-136
R/plot_with_settings.R          309      16  94.82%   299-305, 327, 364, 373-374, 390, 578-579, 581, 583
R/standard_layout.R              52       4  92.31%   90-93
R/table_with_settings.R         199      24  87.94%   32-36, 82, 127-158
R/utils.R                         7       0  100.00%
R/verbatim_popup.R              104      35  66.35%   112-113, 115, 123-154
R/white_small_well.R              7       0  100.00%
TOTAL                          1164     229  80.33%

Diff against main

Filename              Stmts    Miss  Cover
------------------  -------  ------  -------
R/verbatim_popup.R       -1       0  -0.32%
TOTAL                    -1       0  -0.02%

Results for commit: 9fc7031

Minimum allowed coverage is 80%

♻️ This comment has been updated with latest results

[github-actions]

Unit Tests Summary

  1 files   16 suites   1m 59s ⏱️
172 tests 172 ✅ 0 💤 0 ❌
400 runs  400 ✅ 0 💤 0 ❌

Results for commit 9fc7031.

♻️ This comment has been updated with latest results.

[github-actions]

Unit Test Performance Difference

Test Suite	$Status$	Time on main	$±Time$	$±Tests$	$±Skipped$	$±Failures$	$±Errors$
table_with_settings_ui	💚	$35.46$	$-1.24$	$0$	$0$	$0$	$0$

Results for commit 20602b6

♻️ This comment has been updated with latest results.

[donyunardi]

I never like snapshot test. Should we take this opportunity to change this test to something else? Maybe just regex if the html expression exist?

[osenan]

God job! Only a minor comment

[llrs-roche]

I never like snapshot test. Should we take this opportunity to change this test to something else? Maybe just regex if the html expression exist?

In this case we are testing the UI. We could just use regex that the labels/options are there but that wouldn't test if the html changed or that we get it.
Parsing XML/HTML with regex is a path to disaster but we could test that we get an object of a given class (not sure if this is html or shiny.tag or something else). Or maybe just testing that the label shows up on the resulting UI is enough?

[donyunardi]

I think we need to investigate this a bit deeper. It's not clear why it's failing in the first place as I have other PR in this repo that passes R CMD Check using the latest shiny version.

Putting Request Changes so we don't merge accidentally during investigation.

[llrs-roche]

Not totally sure of what is causing this issue. I've added a couple of checks but I haven't removed the snapshot completely. Let me know what do you think of my analysis on #341 (comment)

[osenan]

The changes look good. In the end it would depend on the content of the button created by shiny.