237090 - TR height attribute ignored

Status: RESOLVED FIXED

(Core :: Layout: Tables, defect)

Description

[Tim De Pauw]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7b) Gecko/20040310 Firefox/0.8.0+
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7b) Gecko/20040310 Firefox/0.8.0+

The height of the single-cell table used in the demo URL used to be 18 pixels,
but recent builds seem to be ignoring the row's height attribute and setting the
row height to the height of the  . The problem is more visible in the
navigation at the top of http://www.gratiz.be/, where the 3 bars used to be
adjacent; now they're just adjacent if they all contain navigation items, which
its have maximum height.

Reproducible: Always
Steps to Reproduce:

Comment 1

[Boris Zbarsky [:bzbarsky]]

> The height of the single-cell table used in the demo URL used to be 18 pixels,

When?  That's certainly not been the behavior in 1.6, 1.5, 1.4, 1.3 (I just
checked).  Further, using   in a table cell to force it to the height of
the font is a commonly used web authoring hack.  So what we're doing now is not
only what we've been doing for over a year, but what authors expect to
happen....  Opera and Konqueror do exactly what we do.

Comment 2

[Tim De Pauw]

Then how do you explain the height of 18px at
http://www.gratiz.be/v3/forumbar.html for instance?

I just put the   there so IE would render it as a non-empty cell, but
that's not the issue. It's worked like that for years. I'm not criticizing here,
I could easily have changed my code and never spoken of it again, but I just
noticed it had suddenly stopped working _that way_ in recent builds. Actually,
with Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7a) Gecko/20040219
it's still 18px high and with the latest Fx, it isn't. What you do with that
information is up to you, I'll change the code soon anyway. :)

Comment 3

[Boris Zbarsky [:bzbarsky]]

> Then how do you explain the height of 18px at
> http://www.gratiz.be/v3/forumbar.html for instance?

What height of 18px?  The table has a height of 100%, and that forces the table
row to the height of the table.  The rendering on that page is the same in a
2004-03-10-08 build of SeaMonkey, as well as builds from November, June, and
February of last year.

> I just put the   there so IE would render it as a non-empty cell, but
> that's not the issue.

Maybe I'm misunderstanding?  What _is_ the issue?  If you could post screenshots
of both renderings involved, that may help.

> What you do with that information is up to you, I'll change the code soon
> anyway. :)

I'd appreciate it if you didn't change anything till we figure out what's going
on with this bug... ;)

Comment 4

[Tim De Pauw]

> What height of 18px?  The table has a height of 100%

Yeah, sorry, I got confused there.

> Maybe I'm misunderstanding?  What _is_ the issue?  If you could post screenshots
> of both renderings involved, that may help.

Okay, I put 2 screenshots up:
http://www.gratiz.be/~tim/bugzilla/237090/mozilla.png (desired behavior)
http://www.gratiz.be/~tim/bugzilla/237090/fx.png (behavior in Gecko/20040310
Firefox/0.8.0+)

(Never mind the purple.)

> I'd appreciate it if you didn't change anything till we figure out what's going
> on with this bug... ;)

No prob, those were my intentions all along. :)

Comment 5

[Boris Zbarsky [:bzbarsky]]

Ah, there we go.  This changed between 2004-03-03-08 and 2004-03-04-08. 
Sicking, this is all you.  See
http://lxr.mozilla.org/seamonkey/source/content/html/content/src/nsHTMLTableRowElement.cpp#523

Comment 6

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

Attachment: patch to fix

Comment 7

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

I looked through all elements to make sure this didn't appear anywhere else.

Comment 8

[Johnny Stenback (:jst)]

Comment on attachment 143654 [details] [diff] [review]
patch to fix

r+sr=jst

Comment 9

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

checked in