Color edit without tooltip fix

[mstabrin]

Description

The QColorSwatch class had a typo in the assignment of the tooltip if no explicit tooltip was provided.

I fixed the typo by removing the second underscore

self.setToolTip(tooltip or trans.__('click to set color'))
to
self.setToolTip(tooltip or trans._('click to set color'))

Type of change

• Bug-fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

References

None

How has this been tested?

• I wrote two tests creating the respective classes with default values and explicitly provided values
• all tests pass with my change

Final checklist:

• My PR is the minimum possible work for the desired functionality
• I have added tests that prove my fix is effective or that my feature works

[mstabrin]

@Czaki I actually really did not like the inner for loop. Thank you for pointing me in the right direction :)

I changed it to use pytest.mark.parametrize and added all created widgets to the qtbot.

[codecov]

Codecov Report

Merging #4717 (889a954) into main (67e6771) will increase coverage by 0.01%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main    #4717      +/-   ##
==========================================
+ Coverage   87.70%   87.72%   +0.01%     
==========================================
  Files         587      588       +1     
  Lines       50509    50530      +21     
==========================================
+ Hits        44301    44327      +26     
+ Misses       6208     6203       -5     

Impacted Files	Coverage Δ
napari/_qt/widgets/_tests/test_qt_color_swatch.py	100.00% <100.00%> (ø)
napari/_qt/widgets/qt_color_swatch.py	74.79% <100.00%> (+0.81%)	⬆️
napari/components/experimental/chunk/_pool.py	95.23% <0.00%> (+6.34%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 67e6771...889a954. Read the comment docs.

[Czaki]

As parent is only passed to the super class constructor I suggest removing this parameter as we should our code, not Qt itself.

As it looks simple we may also add tests to verify if object is properly initialized? Pleas see what you think about my suggestion

[mstabrin]

@Czaki I think that makes a lot of sense.

I modified your suggestion to check explicitly for the expected value depending on the input for tooltip and color.

[Czaki]

I modified your suggestion to check explicitly

did you know about using the or operator during the evaluation?

a = b or c

if b evaluate to True then assign b to a else assign c to a

Such code is shorter and function (for me will be more readable). But I lest this decision for you and maybe the napari maintainer. The current state is good for me.

[mstabrin]

@Czaki Thanks for the hint, I did not use the or statement this way, yet. I will go with your suggestion :)

[mstabrin]

@Czaki I am still not sure how merging works :) do I need to write someone and ask for the merge and that you reviewed it or do I just need to wait?

[Carreau]

Looks good to me, thanks for the pull request and the reviews.

[Carreau]

I am still not sure how merging works :) do I need to write someone and ask for the merge and that you reviewed it or do I just need to wait?

Usually someone with merge right will check PRs every now and then, and merge if all looks good. Feel free to ping or ask for more-reviews/feedback if you have not got any for a week or so. You can either comment on the issue or ask on the napari zulip.

[Czaki]

@Czaki I am still not sure how merging works :) do I need to write someone and ask for the merge and that you reviewed it or do I just need to wait?

there is a rule in napari that it PR has only one approval the person who gives approval should wait at least 24h before merging to give possibility for other core-devs to review changes.

But GitHub has no option: "remind me after 24h" so it happens that PR is missed by reviewing person. So feel free to ping reviewing person if at least 24 hours have passed and such person does not call another person for review.