53891 - table explodes/reflows horizonatally after page loads

Status: VERIFIED FIXED

(Core :: Layout: Tables, defect, P2)

Description

[rob cherny]

BuildID:    2000091508

sorry if this is a duplicate, was unable to find another quite this extreme.  
also, not sure if it belongs to HTML Tables or not ... 

go to URL ... page loads and appears to load correctly - yet when finishes 
loading the page table content explodes outward/horizontally.

Reproducible: Always
Steps to Reproduce:
1. go to url
2. watch as page loads
3. page finishes loading
4. page reflows as table expands horizontally (make sure your screen res is a 
large one - i'm running 1280x1024)

Actual Results:  page reflows onload

Expected Results:  table should stick to defined, fixed width			
			
not sure what in the HTML is doing this

Comment 1

[Eric S. Smith]

Not seeing this w. 2000092212 on Win95.  Page may have changed, of course,
between the reporter's visit and my own.

Comment 2

[Matthias Versen [:Matti]]

WFM on Windows 2000 , Build :2000091920
also wfm in Windows 98 with build : 2000081008

Comment 3

[rob cherny]

Downloaded 2000092212 on my Win98 box.  Still does it for me.  Does it now both 
on WinNT4 and Win98 for me. Odd it isn't doing it on other's machines.

Will create an attachment - screen shot.

Comment 4

[rob cherny]

Attachment: squeezed the life out of this GIF, but you get the point on the screenshot - see how it exploded outwards.

Comment 5

[Christine Hoffman]

Created a smaller testcase. Image is causing the reflow problem. Need to see if 
this has already been reported - will confirm bug.

Comment 6

[Christine Hoffman]

Attachment: simplified testcase

Comment 7

[Christine Hoffman]

Attachment: incorrect testcase above

Comment 8

[Christine Hoffman]

This testcase is behaving differently than on my local machine but if you 
reload, the image placeholder expands after a minute and forces the table 
movement.

Comment 9

[Christine Hoffman]

Dup of #40892?

Comment 10

[rob cherny]

it looks to me like the ALT text is doing it on the above attachment, am I wrong?

this would be after an image that was supposed to load didn't load ... but
neither the star wars site nor a page i have in development (which i can't show
as it is for a client and unfinished - still might be able to distill test case
from it though, maybe tomorrow) has a broken image.  Am i making that up
(misunderstanding)?

The test case doesn't have HEIGHT and WIDTH defined, however I just saved the
file off and it seems to do the same thing with the H/W attributes defined as
well.  Then I took out the ALT text and the page reflowed, however my guess is
that the spec would read that the table should reflow if the image does not
exist (help?).  With the ALT text it reflows to fit the words (i made the alt
text a long run on sentence and it did just that), and without the ALT text it
reflows to simply collapse.  That may indeed be correct behavior.

I can look at it further, but the question is - are these two different bugs?  I
could be mistaken, but I think we need a test case with legit graphics.

Comment 11

[Christine Hoffman]

The testcase was broken down from the 'starwars' page. On that page, the image I 
broke that down to (img/new.gif) had no dimensions or alt tag. I will try to 
provide a revised testcsae.

Comment 12

[Christine Hoffman]

I am attaching another testcase (essentially the same but with an image that has 
dimensions and 'alt'). When I first load this page, the second column 'jumps' to 
the right. But if I reload, it 'jumps' back. This is the behavior I saw on the 
starwars page. The difference is that after the reload, it does not jump to the 
right again - in order to see that behavior you need to open the page anew. 

So this may or may not be the same problem. The problem is clearly evident on 
the 'startwars' page.

Comment 13

[Christine Hoffman]

Attachment: testcase with image below

Comment 14

[Christine Hoffman]

Attachment: image to go with testcase

Comment 15

[Christine Hoffman]

Nominating this for RTM. This is a serious reflow problem.

Comment 16

[karnaze (gone)]

Attachment: slightly reduced test case from previous

Comment 17

[karnaze (gone)]

Attachment: patch to fix the bug

Comment 18

[karnaze (gone)]

Adding rmt+. PDT, this really needs to be fixed for rtm. It is basic stuff and 
could affect any nested table that has a width specified.

Comment 19

[karnaze (gone)]

Changing to rtm+. 

Comment 20

[karnaze (gone)]

rtm++ per discussion with PDT. Must land on the trunk one day ahead of the 
branch.

Comment 21

[karnaze (gone)]

Fixed.

Comment 22

[Christine Hoffman]

Verified fixed on Win, Mac and Linux with 10_11 branch build. Added vtrunk 
keyword.

Comment 23

[Asa Dotzler [:asa]]

Verified Fixed on trunk builds.  Testcases and starwars site look good and don't
"explode" at end of load.
linux 101808 RedHat 6.2
win32 101804 NT 4
mac 101804 Mac OS9
Setting bug to Verified and removing vtrunk keyword