[Webkit-unassigned] [Bug 55684] New: visibility:hidden on top of duplicate use of clipPath hides both uses

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=55684

           Summary: visibility:hidden on top of duplicate use of clipPath
                    hides both uses
           Product: WebKit
           Version: 528+ (Nightly build)
          Platform: PC
        OS/Version: Mac OS X 10.5
            Status: UNCONFIRMED
          Severity: Normal
          Priority: P2
         Component: SVG
        AssignedTo: webkit-unassigned at lists.webkit.org
        ReportedBy: robinmessage+webkit at google.com
                CC: zimmermann at kde.org


Created an attachment (id=84573)
 --> (https://bugs.webkit.org/attachment.cgi?id=84573&action=review)
Test case which should show 4 squares

When a group with visibility:hidden style contains a g node or a use node with a clip-path, and that group is on top of another group with the same clip-path in the same position, the bottom group is also invisible.

Changing the visibility of the top group makes the bottom group visible. Changing it back to hidden makes it invisible again. However, manipulating the bottom group in the DOM (e.g. changing the clip-path away and back again) makes it visible, and it remains visible, even if the top group visibility is flipped.

Expected result of attachment: 4 black squares across the top of the screen.

Actual result: 2 black squares show up, cases c and d.

Description of the cases:
a. clipPath node inside the second group, which has visibility:hidden. Not sure if this is legal SVG, but if it isn't, then the first group should render.
b. A group in the defs section contains the clipPath and a group that is clipped. This is used in two groups, the second of which is visibilty:hidden.
c. Similar group in the defs section, but only the clipped group is used. This one works.
d. clipPath and two groups defined as siblings. Groups are clipped directly. Second group is visibility:hidden. This also works.

It seems like case a might be incorrect SVG. Certainly c is very similar to b and works, so this is a suitable work-around for me, but b ought to work.

(Tested on variety of WebKit versions, including r80210 nightly)

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.