387866 - table-layout:fixed fixed width col should win over pct width?

Status: NEW

(Core :: Layout: Tables, defect)

Description

[Amitava Bhattacharyya]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a6) Gecko/20070629 GranParadiso/3.0a6
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a6) Gecko/20070629 GranParadiso/3.0a6

The ExtJS DatePicker doesn't render properly. It only shows a very wide table-type structure (the title bar etc.) but no text/buttons, i.e., there are no dates to pick. It works correctly in FireFox 2, IE 6/7, Opera etc., so I suspect this is a bug with Gran Paradiso. 

Reproducible: Always

Steps to Reproduce:
1. Goto http://extjs.com/deploy/ext/docs/index.html and select 'Examples and Demos' > 'Toolbar and Menus' > 'Toolbar and Menus'
2. Click on 'Button w/ Menus' > 'Choose a Date'

Actual Results:  
A very wide table structure is rendered. The rows (title bar, footer bar, etc) are styled properly, but no text anywhere. No dates, no buttons, nothing.

Expected Results:  
Should display a calendar date picker component.

Default Install w/ Adblock Plus.

Comment 1

[Marco Bonardo [:mak]]

Yes, i can cofirm this works on Branch, not on Trunk

Comment 2

[:Gavin Sharp [email: [e-mail]]]

Could you find out when this regressed, using the builds at http://archive.mozilla.org/pub/firefox/nightly and/or http://ftp.mozilla.org/pub/mozilla.org/firefox/nightly ?

Comment 3

[Marco Bonardo [:mak]]

it regressed between 04 December 2006 (works) and 10 December 2006 (don't work)

is this enough?

Comment 4

[:Gavin Sharp [email: [e-mail]]]

Seems likely to be a regression from the reflow branch landing, in that case. A minimized testcase would be nice :). You might also want to look through the dependency list for that bug to see if there are any duplicates.

Comment 5

[Marco Bonardo [:mak]]

there seem to be an issue with

table-layout:fixed;

removing it from the css solves the problem...

Comment 6

[Marco Bonardo [:mak]]

Attachment: not-so reduced testcase

this is not so reduced, but it's at least self-contained

the only thing i could reproduce in another way is that table-layout:fixed; has different behaviuor between brach & trunk, in trunk it expands to 100% width, in branch it does not

Comment 7

[Biju]

Attachment: fx_ie_menu_button.png

I dont know this is also same issue.
There is a problem in Menu Button 
See 'Action Menu' button at
http://extjs.com/deploy/dev/examples/menu/actions.html

Comment 8

[Brian Moeskau]

Here's an additional, reduced example.  I'm an Ext JS core developer, so if you need any assistance, please let me know.

http://localhost/ext/playpen/browser-bugs/FF/Beta4/datepicker.html

Comment 9

[Jeff Walden [:Waldo]]

Um, localhost?  ;-)

Comment 10

[Brian Moeskau]

D'oh!  Wrong tab ;)

http://extjs.com/playpen/browser-bugs/FF/Beta4/datepicker.html

Comment 11

[Bernd]

I went to http://extjs.com/playpen/browser-bugs/FF/Beta4/datepicker.htmland the firebug shows that there is width set on the table style as on the containing div.

I am skeptical that is valid.

Comment 12

[Bernd]

we need a reduced testcase to judge this.

Comment 13

[Saint Wesonga]

Attachment: Reduced Testcase

Renders differently only on Firefox (in comparison to IE6-8, Safari, Opera)

Comment 14

[Bernd]

Attachment: same testcase with reference redenring

Comment 15

[Saint Wesonga]

Attachment: Patch

Comment 16

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

I think this is going to behave badly if a table-layout:fixed table has a column with a percentage width, and perhaps other cases as well.  In those cases the GetMinWidth code is designed to return only a minimum, and not a preferred width, which would need to be different.  That said, for the percentage case, we could probably fix it using a similar approach to the one we use for BasicTableLayoutStrategy, but I'm not sure (after a quick glance) if the other cases are things we can ever hit.

Comment 17

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Comment on attachment 416514 [details] [diff] [review]
Patch

It might also be possible to do something that returns fixed widths when possible and otherwise returns nscoord_MAX.

But I don't think we want to do what this patch does.

Comment 18

[BugBot [:suhaib / :marco/ :calixte]]

The bug assignee didn't login in Bugzilla in the last 7 months, so the assignee is being reset.