Bug 25432 - Simplify SVG hit testing (and make it floating point precise)
Summary: Simplify SVG hit testing (and make it floating point precise)
Alias: None
Product: WebKit
Classification: Unclassified
Component: SVG (show other bugs)
Version: 528+ (Nightly build)
Hardware: Macintosh OS X 10.5
: P2 Normal
Assignee: Nobody
: 21892 (view as bug list)
Depends on: 25431
Blocks: 25433
  Show dependency treegraph
Reported: 2009-04-27 14:39 PDT by Eric Seidel
Modified: 2009-04-28 18:00 PDT (History)
3 users (show)

See Also:

Simplify nodeAtPoint for SVG (29.23 KB, patch)
2009-04-27 14:46 PDT, Eric Seidel
simon.fraser: review+
Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Eric Seidel 2009-04-27 14:39:27 PDT
Simplify SVG hit testing (and make it floating point precise)

SVG hit testing is one of the last clients of absoluteTransform(), so fixing it will bring us one step closer to killing absoluteTransform.

I love that this patch killed at least one "this code is needlessly complicated and probably wrong" statements. :)
Comment 1 Eric Seidel 2009-04-27 14:46:03 PDT
Created attachment 29829 [details]
Simplify nodeAtPoint for SVG

 20 files changed, 185 insertions(+), 95 deletions(-)
Comment 2 Simon Fraser (smfr) 2009-04-28 11:27:27 PDT
Comment on attachment 29829 [details]
Simplify nodeAtPoint for SVG

> +TransformationMatrix RenderForeignObject::localToParentTransform() const
> +{
> +    // FIXME: This trasition is backwards!
> +    // It should be localTransform() * translationForAttributes()
> +    // but leaving it backwards for now for LayoutTest result compatibility
> +    return translationForAttributes() * localTransform();
> +}

Reference the bug you filed in the comment.

> +    // SVG uses a different hit-testing system from CSS
> +    virtual bool nodeAtPoint(const HitTestRequest&, HitTestResult&, const FloatPoint& pointInParent, HitTestAction);

I'd rather the new nodeAtPoint got a different name to make it easier to fix bug 23120.

Looks OK to me but I'd like hyatt to give the final word on whether we should add nodeAtPoint proliferation.
Comment 3 Simon Fraser (smfr) 2009-04-28 11:30:22 PDT
Comment on attachment 29829 [details]
Simplify nodeAtPoint for SVG

Dave says it's OK.
Comment 4 Eric Seidel 2009-04-28 13:56:30 PDT
Committing to http://svn.webkit.org/repository/webkit/trunk ...
	M	WebCore/ChangeLog
	M	WebCore/rendering/RenderForeignObject.cpp
	M	WebCore/rendering/RenderForeignObject.h
	M	WebCore/rendering/RenderObject.cpp
	M	WebCore/rendering/RenderObject.h
	M	WebCore/rendering/RenderPath.cpp
	M	WebCore/rendering/RenderPath.h
	M	WebCore/rendering/RenderSVGContainer.cpp
	M	WebCore/rendering/RenderSVGContainer.h
	M	WebCore/rendering/RenderSVGHiddenContainer.cpp
	M	WebCore/rendering/RenderSVGHiddenContainer.h
	M	WebCore/rendering/RenderSVGImage.cpp
	M	WebCore/rendering/RenderSVGImage.h
	M	WebCore/rendering/RenderSVGModelObject.cpp
	M	WebCore/rendering/RenderSVGModelObject.h
	M	WebCore/rendering/RenderSVGRoot.cpp
	M	WebCore/rendering/RenderSVGText.cpp
	M	WebCore/rendering/RenderSVGText.h
	M	WebCore/rendering/RenderSVGViewportContainer.cpp
	M	WebCore/rendering/RenderSVGViewportContainer.h
Committed r42960
Comment 5 Eric Seidel 2009-04-28 18:00:51 PDT
*** Bug 21892 has been marked as a duplicate of this bug. ***