See at http://upcase.inf.upol.cz. You will see search input field, which is not layouted according CCS. When you click on input filed, the page is rearranged and rolled to right. But input field is after click arranged according CCS! I see this bug in other pages too, but there is the most clear.

Chris, here is the interesting HTML: <td class="regular"><small><small> <form action="/cgi-bin/webglimpse/home/services/www/webglimpse" method="get"> <input style="width: 90%" size="10" name="query"><input value="0" name="errors" type="hidden"><input value="50" name="maxfiles" type="hidden"><input value="30" name="maxlines" type="hidden"><input value="10000" name="maxchars" type="hidden"><input value="on" name="whole" type="hidden"> </form> </small></small></td> The input's width is set to 90%. When the page loads the input gets an initial reflow with everything being unconstrained. When you click on it is gets an Incremental (unconstrained) then a Resize reflow with the available widht and computed width set correctly. Do table currently work with percentages? Or is that still work in progress? This seems like a table bug to me, wouldn't surprise me if it is a dup.

changed summary

As it turns out, when I reduced it, it looks like the two <small> tags around the text field are what are causing the problem. When they are removed it lays out correctly. When I put this inside of a div instead of a table it works fine. Reassigning to chris

[It appears I added the 2nd attachment a second after Rod added the 1st one]. The problem looks like it is either in blocks/lines or text fields, probably the former. Notice that the cell containing the text field always reflows the cell with the same width, so it not a table problem. Rod, Steve - I have gotten a flood of bugs recently where clicking in a text field causes pages to shift, break, etc. If you fix this, I would greatly appreciate it if you looked at 27390, 28212, 28811, 29688, 29723, 29787, 29877, 30047, 30628. I can't remember if all of these involved text fields but some of them did. TO::Rfl en 01D31434 rea=0 av=(8940,UC) comp=(8910,UC) count=0 T::Rfl en 01D31488 rea=0 av=(8940,UC) comp=(8910,UC) count=1 TRG::Rfl 01D314F4 rea=0 av=(UC,UC) comp=(UC,UC) count=2 TR::Rfl en 01D31534 rea=0 av=(UC,UC) comp=(UC,UC) count=3 TC::Rfl 01D3157C rea=0 av=(UC,UC) comp=(UC,UC) count=4 Area::Rfl en 01D315D8 rea=0 av=(UC,UC) comp=(UC,UC) count=5 Area::Rfl ex 01D315D8 des=(1080,510) maxElem=(1080,270) TC::Rfl ex 01D3157C des=(1140,570) maxElem=(1140,330) TR::Rfl ex 01D31534 des=(1200,630) maxElem=(1140,330) TRG::Rfl ex 01D314F4 des=(UC,630) maxElem=(1140,330) TRG::Rfl 01D314F4 rea=2 av=(1200,UC) comp=(1200,UC) count=6 TR::Rfl en 01D31534 rea=2 av=(1200,UC) comp=(1200,UC) count=7 TC::Rfl 01D3157C rea=2 av=(1140,UC) comp=(1080,UC) count=8 Area::Rfl en 01D315D8 rea=2 av=(1080,UC) comp=(1080,UC) count=9 Area::Rfl ex 01D315D8 des=(1080,510) TC::Rfl ex 01D3157C des=(1140,570) TR::Rfl ex 01D31534 des=(1170,630) TRG::Rfl ex 01D314F4 des=(1200,630) T::Rfl ex 01D31488 des=(1230,660) TO::Rfl ex 01D31434 des=(1230,660) WEBSHELL+ = 2 TO::Rfl en 01D31434 rea=1 av=(8940,UC) comp=(8910,UC) count=10 T::Rfl en 01D31488 rea=1 av=(8940,UC) comp=(8910,UC) count=11 TRG::Rfl 01D314F4 rea=1 av=(1200,UC) comp=(1200,UC) count=12 TR::Rfl en 01D31534 rea=1 av=(1200,UC) comp=(1200,UC) count=13 TC::Rfl 01D3157C rea=1 av=(1140,UC) comp=(1080,UC) count=14 Area::Rfl en 01D315D8 rea=1 av=(1080,UC) comp=(1080,UC) count=15 Area::Rfl ex 01D315D8 des=(1080,510) maxElem=(1080,270)max=1080 TC::Rfl ex 01D3157C des=(1140,570) maxElem=(1140,330)max=1140 TR::Rfl ex 01D31534 des=(1200,630) max=0 TRG::Rfl ex 01D314F4 des=(1200,0) max=0 TRG::Rfl 01D314F4 rea=2 av=(1200,UC) comp=(1200,UC) count=16 TR::Rfl en 01D31534 rea=2 av=(1200,UC) comp=(1200,UC) count=17 TR::Rfl ex 01D31534 des=(1170,630) TRG::Rfl ex 01D314F4 des=(1200,630) T::Rfl ex 01D31488 des=(1230,660) max=1230 TO::Rfl ex 01D31434 des=(1230,660) TO::Rfl en 01D31434 rea=1 av=(8940,UC) comp=(8910,UC) count=18 T::Rfl en 01D31488 rea=1 av=(8940,UC) comp=(8910,UC) count=19 TRG::Rfl 01D314F4 rea=1 av=(1200,UC) comp=(1200,UC) count=20 TR::Rfl en 01D31534 rea=1 av=(1200,UC) comp=(1200,UC) count=21 TC::Rfl 01D3157C rea=1 av=(1140,UC) comp=(1080,UC) count=22 Area::Rfl en 01D315D8 rea=1 av=(1080,UC) comp=(1080,UC) count=23 Area::Rfl ex 01D315D8 des=(1080,510) maxElem=(1080,270)max=1080 TC::Rfl ex 01D3157C des=(1140,570) maxElem=(1140,330)max=1140 TR::Rfl ex 01D31534 des=(1200,630) max=0 TRG::Rfl ex 01D314F4 des=(1200,0) max=0 TRG::Rfl 01D314F4 rea=2 av=(1200,UC) comp=(1200,UC) count=24 TR::Rfl en 01D31534 rea=2 av=(1200,UC) comp=(1200,UC) count=25 TR::Rfl ex 01D31534 des=(1170,630) TRG::Rfl ex 01D314F4 des=(1200,630) T::Rfl ex 01D31488 des=(1230,660) max=1230 TO::Rfl ex 01D31434 des=(1230,660)

I suspected line layout, but thought maybe the div experiemnt ruled it out. reassigning to Buster

Fixed typo in Summary: with-->width. css1.

redistributing bugs across future milestones, sorry for the spam

I think this is fixed now. I do see a layout problem on the second attachment, but that is a separate bug. Chris, could you file that please?

I just filed bug 41806 for the layout problem related to the 2nd attachment.

Updating QA contact.

Verified Build 2001080603 os:win98,95,NT