Hit-testing of boxes over scrollers should account for border-radius

• Component: Scrolling
• Status: RESOLVED FIXED

[Simon Fraser (smfr)]

Created attachment 363784 [details]
Testcase

Need to take border-radius into account for hit testing.

[Radar WebKit Bug Importer]

<rdar://problem/48649993>

[Antti Koivisto]

Created attachment 366171 [details]
patch

[EWS Watchlist]

Attachment 366171 [details] did not pass style-queue:


ERROR: Source/WebCore/testing/Internals.h:350:  enum members should use InterCaps with an initial capital letter or initial 'k' for C-style enums.  [readability/enum_casing] [4]
Total errors found: 1 in 15 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[EWS Watchlist]

Comment on attachment 366171 [details]
patch

Attachment 366171 [details] did not pass ios-sim-ews (ios-simulator-wk2):
Output: https://webkit-queues.webkit.org/results/11693720

New failing tests:
fast/scrolling/ios/border-radious-event-region.html

[EWS Watchlist]

Created attachment 366200 [details]
Archive of layout-test-results from ews121 for ios-simulator-wk2

The attached test failures were seen while running run-webkit-tests on the ios-sim-ews.
Bot: ews121  Port: ios-simulator-wk2  Platform: Mac OS X 10.13.6

[Antti Koivisto]

Created attachment 366251 [details]
patch

[EWS Watchlist]

Attachment 366251 [details] did not pass style-queue:


ERROR: Source/WebCore/testing/Internals.h:350:  enum members should use InterCaps with an initial capital letter or initial 'k' for C-style enums.  [readability/enum_casing] [4]
Total errors found: 1 in 15 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[Simon Fraser (smfr)]

Comment on attachment 366251 [details]
patch

View in context: https://bugs.webkit.org/attachment.cgi?id=366251&action=review

> Source/WebCore/page/Frame.h:118
>      LayerTreeFlagsIncludeBackingStoreAttached = 1 << 7,
>      LayerTreeFlagsIncludeRootLayerProperties = 1 << 8,
> +    LayerTreeFlagsIncludeEventRegion = 1 << 9,

The cool kids are lining up the "= ..." these days :)

> Source/WebCore/platform/graphics/GraphicsLayer.cpp:935
> +        ts << indent << "(event region\n";
> +        for (auto& rect : m_eventRegion.rects()) {
> +            TextStream::IndentScope indentScope(ts);
> +            ts << indent << "(rect " << rect.x() << " " << rect.y() << " " << rect.width() << " " << rect.height() << ")\n";
> +        }
> +        ts << indent << ")\n";

Please add and use TextStream& operator<<(TextStream&, const Region&)

> Source/WebCore/platform/graphics/RoundedRect.h:117
> +    // Approximate by snipping away half-radii sized rectangles from corners.
> +    WEBCORE_EXPORT Region approximateAsRegion() const;

I think I'd prefer some kind of "stepping" function with a configurable step size at some point.

> LayoutTests/ChangeLog:10
> +        * fast/scrolling/ios/border-radious-event-region-expected.txt: Added.
> +        * fast/scrolling/ios/border-radious-event-region.html: Added.

Bad spelling

[Antti Koivisto]

> > +    // Approximate by snipping away half-radii sized rectangles from corners.

That comment is a lie. We now snip away multiple rectangles based on arc length.

[Antti Koivisto]

Created attachment 366360 [details]
patch

[EWS Watchlist]

Attachment 366360 [details] did not pass style-queue:


ERROR: Source/WebCore/testing/Internals.h:350:  enum members should use InterCaps with an initial capital letter or initial 'k' for C-style enums.  [readability/enum_casing] [4]
Total errors found: 1 in 17 files


If any of these errors are false positives, please file a bug against check-webkit-style.

[WebKit Commit Bot]

Comment on attachment 366360 [details]
patch

Clearing flags on attachment: 366360

Committed r243674: <https://trac.webkit.org/changeset/243674>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.