Bug 200237 – Web Inspector: Styles: variable swatch not shown for var() with a fallback

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

[Devin Rousso]

var(--test) shows a swatch

var(--test, 0) does NOT show a swatch

[Devin Rousso]

Created attachment 375111 [details]
Patch

[EWS Watchlist]

Comment on attachment 375111 [details]
Patch

Attachment 375111 [details] did not pass mac-wk2-ews (mac-wk2):
Output: https://webkit-queues.webkit.org/results/12832490

New failing tests:
fast/workers/worker-document-leak.html

[EWS Watchlist]

Created attachment 375124 [details]
Archive of layout-test-results from ews104 for mac-highsierra-wk2

The attached test failures were seen while running run-webkit-tests on the mac-wk2-ews.
Bot: ews104  Port: mac-highsierra-wk2  Platform: Mac OS X 10.13.6

[Joseph Pecoraro]

Comment on attachment 375111 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=375111&action=review

> Source/WebInspectorUI/UserInterface/Models/CSSKeywordCompletions.js:98
> +WI.CSSKeywordCompletions.isTimingFunctionAwareProperty = function(name)
> +{
> +    if (name in WI.CSSKeywordCompletions._timingFunctionAwareProperties)
> +        return true;
> +
> +    let isNotPrefixed = name.charAt(0) !== "-";
> +    if (isNotPrefixed && ("-webkit-" + name) in WI.CSSKeywordCompletions._timingFunctionAwareProperties)
> +        return true;

I don't think this works at all. `_timingFunctionAwareProperties` is a Set and JavaScript's ` in ` syntax does not work with Sets.

They should be like the other code that was updated:

    if (WI.CSSKeywordCompletions._timingFunctionAwareProperties.has(name))
    ...
    if (isNotPrefixed && WI.CSSKeywordCompletions._timingFunctionAwareProperties.has("-webkit-" + name))
    ...

> Source/WebInspectorUI/UserInterface/Views/SpreadsheetStyleProperty.js:453
> +            tokens = this._addVariableTokens(tokens);

Did the order matter? This now creates an inline swatch for the variable earlier, before colors. I didn't follow this logic all the way to see if that would be good or bad.

[Devin Rousso]

Comment on attachment 375111 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=375111&action=review

>> Source/WebInspectorUI/UserInterface/Models/CSSKeywordCompletions.js:98
>> +        return true;
> 
> I don't think this works at all. `_timingFunctionAwareProperties` is a Set and JavaScript's ` in ` syntax does not work with Sets.
> 
> They should be like the other code that was updated:
> 
>     if (WI.CSSKeywordCompletions._timingFunctionAwareProperties.has(name))
>     ...
>     if (isNotPrefixed && WI.CSSKeywordCompletions._timingFunctionAwareProperties.has("-webkit-" + name))
>     ...

Oops!  I changed `_colorAwareProperties`, but forgot about `_timingFunctionAwareProperties` 😅

>> Source/WebInspectorUI/UserInterface/Views/SpreadsheetStyleProperty.js:453
>> +            tokens = this._addVariableTokens(tokens);
> 
> Did the order matter? This now creates an inline swatch for the variable earlier, before colors. I didn't follow this logic all the way to see if that would be good or bad.

Yes, the order does have an effect, because the various `_add*Tokens` replaces the token with an `WI.InlineSwatch`, meaning we no longer get access to the token data from CodeMirror.  I actually need to rework this a tiny bit so we get swatches for fallback values.

[Devin Rousso]

Created attachment 375491 [details]
Patch

[Joseph Pecoraro]

Comment on attachment 375491 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=375491&action=review

r=me

> Source/WebInspectorUI/ChangeLog:33
> +        Follow the variable "path" until an ultimate value is reached.

Follow the variable chain?

[Devin Rousso]

Created attachment 375571 [details]
Patch

[WebKit Commit Bot]

Comment on attachment 375571 [details]
Patch

Clearing flags on attachment: 375571

Committed r248279: <https://trac.webkit.org/changeset/248279>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Radar WebKit Bug Importer]

<rdar://problem/53964395>