769944 - DLBI broke positioning of foreignObjects

Status: RESOLVED WORKSFORME

(Core :: SVG, defect)

Description

[Jonathan Watt [:jwatt]]

The DLBI patches from bug 539356 appear to have broken positioning of foreignObjects. I've not had time to do a lot of testing (leaving on vacation in a few hours), but a cursory test suggests that foreignObject content is being positioned at the top-left of the outer-<svg> element, ignoring any transforms or offsets introduced by the foreignObject itself, or any of its ancestors.

This looks suspiciously like the display list created for foreignObject content is being integrated into the main display list, but I don't actually thing the DLBI patches have done that. Weird.

Comment 1

[Jonathan Watt [:jwatt]]

Attachment: testcase

Comment 2

[Chris Jones [:cjones] inactive; ni?/f?/r? if you need me]

I see a little red square if I load this test in Chrome, but it doesn't show up in FF.  So I think this is still a bug.

Comment 3

[Matt Woodrow (:mattwoodrow)]

I believe all-green is the expected behaviour. DLBI builds match the behaviour of pre-DLBI builds.

I think Chrome is just broken.

Comment 4

[Mayank Bansal]

The red box also appears on IE9 and Opera12.10 beta RC.

Comment 5

[Chris Jones [:cjones] inactive; ni?/f?/r? if you need me]

Also Opera 11.64 on linux-x64 and android-browser in JB.

Comment 6

[Alice0775 White]

Attachment: sample

Just changed testcase as follows:
      <div xmlns="http://www.w3.org/1999/xhtml"
           style="background:lime; display:block; width:100%; height:100%;"/>
to

      <div xmlns="http://www.w3.org/1999/xhtml"
           style="background:blue; display:block; width:100%; height:100%;">111<br />2222</div>


The difference between Firefox and other browsers seems to be height of the div inside the foreignObject.

Comment 7

[Jonathan Watt [:jwatt]]

I can confirm I designed the test with the intention that red==failure. Fully green does indeed mean the test passed.

Comment 8

[Chris Jones [:cjones] inactive; ni?/f?/r? if you need me]

*golf clap*.  We should file bugs on the other browsers (every other one, apparently).

Comment 9

[Robert Longson [:longsonr]]

IE9 does not support foreignObject so it's an expected failure there.