441267 - Wrong positioning for absolute positioned elements

Status: RESOLVED INVALID

(Core :: Layout, defect)

Description

[Eugen Konkov]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.1; ru; rv:1.8.1.14) Gecko/20080404 Firefox/2.0.0.14
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.2; ru; rv:1.9) Gecko/2008052906 Firefox/3.0

see attached file

Reproducible: Always

Steps to Reproduce:
1.
2.
3.

Comment 1

[Eugen Konkov]

Attachment: test case

Comment 2

[Martijn Wargers (dead)]

Eugen, do you have the source of that page or an url somewhere?

Comment 3

[Eugen Konkov]

Attachment: test case

Comment 4

[Eugen Konkov]

Attachment: test case

Comment 5

[Eugen Konkov]

Comment on attachment 326630 [details]
test case

type something long in input field near Fetch

Comment 6

[Martijn Wargers (dead)]

Right and bottom for absoluted positioned elements that are replaced doesn't work anymore. That is according to CSS2.1. See bug 423397.
A text input is a replaced element, so that is why the text input is wider now with your testcase.

For the rest, I don't really see a difference compared to FF2 with your testcase.

Comment 7

[Terje Rosenlund]

Same problem, different explanation (check testcase):

FF3 do not calculate position relative to its closest positioned ancestor when an absolute positioned table is contained inside another absolute positioned table (using relative to initial containing block => window)

Comment 8

[Terje Rosenlund]

Attachment: Calculates position relative to wrong object

Comment 9

[Terje Rosenlund]

It seems to me that my comment and test-case is a separate case?

Eugen Konkov's testcase may not be a bug and I believe as Martijn Wargers that this was the same in FF2:

FF3 gives INPUT a width: 151px (fontSize=13px) ( and length: 149 (??)) as default 
(INPUT has no size or width attribute and no width or max-width in style)

Confusion around these issues is very common and may be partly due too:

- Setting width or max-width on containing element is not respected (bug ?)
- Setting width-attribute has no effect (bug?)
- Setting style='width: 40px' limits width
- Setting style='max-width: 40px' limits width more (?) (bug?)

Comment 10

[Daniel.S]

(In reply to Terje Rosenlund from comment #7)
> Same problem, different explanation (check testcase):

Completely differen issue. What you describe is bug 10209 and/or bug 63895. There is a patch on the way in another bug that will fix those issues.

(In reply to Terje Rosenlund from comment #9)
> - Setting width or max-width on containing element is not respected (bug ?)
> - Setting width-attribute has no effect (bug?)
> - Setting style='width: 40px' limits width
> - Setting style='max-width: 40px' limits width more (?) (bug?)

All those cases appear to work correctly and identically in Firefox and other browsers.

I'll resolve this report as Invalid because the input element behaves correctly.
The table issue has its bugs and does not really fit in here (so I don't dupe the report).