290880 - poor support for CSS: background-repeat property

Status: RESOLVED INVALID

(Core :: Layout, defect)

Description

[Amanda]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.7.7) Gecko/20050414 Firefox/1.0.3
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.7.7) Gecko/20050414 Firefox/1.0.3

The CSS background-repeat property is not well-supported. Does not support
background-repeat along an axis (background-repeat: repeat-x, for example).
Almost incompatible with the IE settings (if settings work in IE, they look
'wrong' in Firefox, and vice versa). Other "background" properties are not
well-supported, such as setting the background image to none for hyperlink
properties. Am a designer, this is noticed while designing sites.

Reproducible: Always

Steps to Reproduce:
1. Code a site using aforementioned properties as styles.
2. Try using a very thin graphic, like a graphical line, as a repeating
background for a <p> or <div>, and notice differences between IE and Firefox.
3.

Actual Results:  
The background either disappears (repeat-x), or or doesn't display correctly,
appearing wrapped and disappearing into upper and lower <div> limits.

Expected Results:  
Displayed a repeating background along an X axis properly or compatibly with
IE's display values (until IE is no longer so popular!)

Comment 1

[Brian]

is it W3C compliant?

Comment 2

[Steve England [:stevee]]

Amanda, can you attach a testcase to show what you're seeing? It will improve
your chances of getting an answer tenfold.

Comment 3

[Josh Birnbaum]

Brian, the spec for this is at
http://www.w3.org/TR/CSS21/colors.html#propdef-background-repeat

over to Core:Layout for further triage

Comment 4

[Amanda]

Attachment: The CSS with code at top pertaining to problem

Comment 5

[Amanda]

Attachment: The HTML sheet with relevent code reference, works only in IE

Comment 6

[Amanda]

Attachment: The line used with code, can be any line gif tho

Comment 7

[Amanda]

(In reply to comment #1)
> is it W3C compliant?

Yes, (I) comply to XHTML and CSS standards.

Comment 8

[Marius Strumyla]

Attachment: Corrected CSS

Comment 9

[Marius Strumyla]

Attachment: Corrected testcase

Tescase with corrected links.

Comment 10

[Marius Strumyla]

Attachment: Corrected testcase

Testcase with corrected links.

Comment 11

[Marius Strumyla]

Attachment: Corrected testcase

Comment 12

[Marius Strumyla]

Attachment: CSS

Comment 13

[Marius Strumyla]

Sorry for the mess :) 

The problem is that you setting "background-attachment: fixed". Do you really
want this? Generally this means that a background does not move when you scroll
a page.

For the exact behaviour, see http://www.hixie.ch/tests/evil/mixed/bgafixed.html

I think you should mark this bug as INVALID.

Comment 14

[Amanda]

(In reply to comment #13)
> Sorry for the mess :) 
> The problem is that you setting "background-attachment: fixed". Do you really
> want this? Generally this means that a background does not move when you 
scroll
> a page.
> For the exact behaviour, see 
http://www.hixie.ch/tests/evil/mixed/bgafixed.html
> I think you should mark this bug as INVALID.

Problem: note what happens when that bit of code is removed and compare it to 
IE - when I do this, the line appears wrapped, or doubled. Are you not seeing 
these results? I can do a print-screen type thing if that would help.

Comment 15

[Marius Strumyla]

Attachment: Screenshot of Corrected testcase

(In reply to comment #14)
> ... the line appears wrapped, or doubled. Are you not seeing 
> these results? I can do a print-screen type thing if that would help.

Nope. For me Firefox 1.0.3 displays the background the same way as IE6 on
win2k.

Comment 16

[Amanda]

(In reply to comment #15)

OK, now it is working...indeed it is because the attachment was fixed.
Originally, image was different so maybe that's why I kept that part of the
code, don't know, apologies. But it's great to know you all were looking into
it, top reason to use FF.

Am curious as to what would happen if a larger image was used, and the coder
wanted those properties of "fixed" and "repeat-x", but guess that rarely comes up.