Bug 87679 – Vertical writing mode can overflow fixed size grandparent container

• Status: RESOLVED
• Resolution: FIXED
• Component: CSS

[Victor Carbune]

Created attachment 144401 [details]
html page

The height of the parent containing block is not respected and the text overflows. (-webkit-flex: 1)

There are multiple nested layouts in the example (using display:-webkit-flex)

[Victor Carbune]

Created attachment 144402 [details]
screenshot of how it renders

[Ojan Vafai]

This test case is very complicated. Just to be clear, the only bug here is that verticalGrowingLeftContainer grows outside of verticalCueContainer, right? Or are there multiple bugs?

[Ojan Vafai]

Created attachment 145115 [details]
reduced test case

[Ojan Vafai]

Created attachment 145134 [details]
Patch

[Ojan Vafai]

Hyatt, this should be a simple 2-line review. Mind taking a quick look? I don't think there's anyone but you and I familiar with vertical writing mode. :)

[Victor Carbune]

(In reply to comment #2)
> This test case is very complicated. Just to be clear, the only bug here is that verticalGrowingLeftContainer grows outside of verticalCueContainer, right? Or are there multiple bugs?

Right, that should be it. Sorry about the fairly complicated csae, but that's what I was working on.

[Ojan Vafai]

(In reply to comment #6)
> (In reply to comment #2)
> > This test case is very complicated. Just to be clear, the only bug here is that verticalGrowingLeftContainer grows outside of verticalCueContainer, right? Or are there multiple bugs?
> 
> Right, that should be it. Sorry about the fairly complicated csae, but that's what I was working on.

No worries. Just wanted to make sure I wasn't missing something.

[WebKit Review Bot]

Comment on attachment 145134 [details]
Patch

Attachment 145134 [details] did not pass chromium-ews (chromium-xvfb):
Output: http://queues.webkit.org/results/12862446

New failing tests:
perf/object-keys.html
fast/html/details-writing-mode.html

[WebKit Review Bot]

Created attachment 145191 [details]
Archive of layout-test-results from ec2-cr-linux-01

The attached test failures were seen while running run-webkit-tests on the chromium-ews.
Bot: ec2-cr-linux-01  Port: <class 'webkitpy.common.config.ports.ChromiumXVFBPort'>  Platform: Linux-2.6.35-28-virtual-x86_64-with-Ubuntu-10.10-maverick

[Julien Chaffraix]

Comment on attachment 145134 [details]
Patch

The fix looks OK, but it seems to make us regress a table test. It's a shot in the dark but RenderTableSection::calcRowLogicalHeight has some special case for overrideHeight that may be impacted.

[Ojan Vafai]

Comment on attachment 145134 [details]
Patch

Yeah, I saw that. I should have removed the r?. I was working on it last night. It turns out to be tricky to fix. Different uses of overrideHeight seem to include or not incude border+padding. Making the same causes more tests to fail for reasons I don't understand. :(

[Ojan Vafai]

Created attachment 145398 [details]
Patch

[Ojan Vafai]

Comment on attachment 145398 [details]
Patch

This patch depends on the one in bug 88116.

[Gustavo Noronha (kov)]

Comment on attachment 145398 [details]
Patch

Attachment 145398 [details] did not pass gtk-ews (gtk):
Output: http://queues.webkit.org/results/12863678

[Early Warning System Bot]

Comment on attachment 145398 [details]
Patch

Attachment 145398 [details] did not pass qt-wk2-ews (qt):
Output: http://queues.webkit.org/results/12872581

[Early Warning System Bot]

Comment on attachment 145398 [details]
Patch

Attachment 145398 [details] did not pass qt-ews (qt):
Output: http://queues.webkit.org/results/12869636

[Build Bot]

Comment on attachment 145398 [details]
Patch

Attachment 145398 [details] did not pass mac-ews (mac):
Output: http://queues.webkit.org/results/12868664

[Build Bot]

Comment on attachment 145398 [details]
Patch

Attachment 145398 [details] did not pass win-ews (win):
Output: http://queues.webkit.org/results/12860823

[Gyuyoung Kim]

Comment on attachment 145398 [details]
Patch

Attachment 145398 [details] did not pass efl-ews (efl):
Output: http://queues.webkit.org/results/12888105

[WebKit Review Bot]

Comment on attachment 145398 [details]
Patch

Attachment 145398 [details] did not pass chromium-ews (chromium-xvfb):
Output: http://queues.webkit.org/results/12888139

[Ojan Vafai]

Created attachment 145848 [details]
Patch

[Julien Chaffraix]

Comment on attachment 145848 [details]
Patch

r=me assuming the tests are passing.

[WebKit Review Bot]

Comment on attachment 145848 [details]
Patch

Clearing flags on attachment: 145848

Committed r119562: <http://trac.webkit.org/changeset/119562>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.

[Victor Carbune]

The initial html page attached still isn't rendered at all properly, but I guess there are other related bugs filed?

[Tony Chang]

Created attachment 146615 [details]
reduction

Here's a new reduction.  This doesn't seem to be specific to flexbox, but with vertical writing mode.

A few comments:
- If you remove #middle, the text no longer overflows.
- If you keep #middle and set the height to 300px, the text no longer overflows.
- If you keep #middle and set the height to 100%, the text overflow.

It seems like setting the height of #middle to 100% should work.

[Tony Chang]

Focusing the reduction.

[Ojan Vafai]

This bug has too much noise with the flexbox patch in it. Moving over to bug 93655 to focus on the remaining writing-mode issue.