Bug 227941 – Web Inspector: REGRESSION(?): alpha slider doesn't match alpha input for `transparent` in color picker

• Status: RESOLVED
• Resolution: FIXED
• Component: Web Inspector

[Devin Rousso]

# STEPS TO REPRODUCE
1. inspect any page
2. add CSS `color: transparent`
3. click the color swatch to show the color picker

# EXPECTED
the alpha slider would be at the bottom and the alpha input would say `0`

# ACTUAL
the alpha slider is at the top and the alpha input says `0`

[Radar WebKit Bug Importer]

<rdar://problem/80557438>

[Nikita Vasilyev]

Reduction:
data:text/html,<p%20style="color:transparent">color

[Nikita Vasilyev]

Created attachment 433799 [details]
Patch

[Devin Rousso]

Comment on attachment 433799 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=433799&action=review

> Source/WebInspectorUI/ChangeLog:11
> +        When the initial value was set to 0, the `set value` was exiting early when the passed value was 0.

Based on this, it seems like the real issue is that the initial value of `_knobY` is really the incorrect value?  Changing to `NaN` seems a bit scary since there's no handling of `isNaN` anywhere in this file.  Perhaps we just need to add a `recalculateKnobY` call inside the early return in `set value`?

[Nikita Vasilyev]

This works as well.

[Nikita Vasilyev]

Created attachment 433801 [details]
Patch

[Devin Rousso]

Comment on attachment 433801 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=433801&action=review

> Source/WebInspectorUI/UserInterface/Views/Slider.js:61
> +        this.recalculateKnobY();

We should only need to do this inside the `if (value === this._value)`.  As you have it now, we'd call `set knobY` twice in the non-early-return path.

[Nikita Vasilyev]

Created attachment 433804 [details]
Patch

[Devin Rousso]

Comment on attachment 433804 [details]
Patch

r=me

[EWS]

Committed r280045 (239780@main): <https://commits.webkit.org/239780@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 433804 [details].