223122 - Mozilla accepts CSS statements with errors in the selector

Status: VERIFIED DUPLICATE of bug 145968

(Core :: CSS Parsing and Computation, defect)

Description

[Robert Brindell]

User-Agent:       Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.5) Gecko/20031007 Firebird/0.7
Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.5) Gecko/20031007 Firebird/0.7

When a class is defined as:

.small,.small:input, .small:textarea, .small:select
{
	font-family: verdana;
	font-size: 8pt;
}

Code such as <td class="small"> will not render properly.  if the class is
broken out as such:

.small
{
	font-family: verdana;
	font-size: 8pt;
}

.small:input
{
	font-family: verdana;
	font-size: 8pt;
}

Then they will all be rendered properly.  

Reproducible: Always

Steps to Reproduce:
1.  Create a web page that links an external CSS class definition.
2.  Create an external CSS class definition using the definitions shown in
'details' section.
3.  Use the definitions mentioned in the 'details' section (<td
class="small">test</td>) on the page and view them in the Firebird browser.

Actual Results:  
The defined class "small" will not render.

Expected Results:  
Render the style as defined above.

My computer has only common configurations that would not affect this discussion.

Comment 1

[Bill Mason]

".small,.small:input, .small:textarea, .small:select"

This syntax isn't even valid.  It's not a surprise it's not doing what you want.

Try .small, input.small, textarea.small, select.small

Or for that matter, if small is styled the same for each of those tags, all you
need is
.small

Anyway.

Comment 2

[Bill Mason]

*** Bug 223124 has been marked as a duplicate of this bug. ***

Comment 3

[Robert Brindell]

http://www.w3.org/TR/css-style-attr

The above document references the ":" syntax as valid representation of a 
pseudo-class.

Comment 4

[Bill Mason]

You can't use HTML tags as pseudoclasses.  Reference
http://www.w3.org/TR/CSS21/selector.html#pseudo-elements

Comment 5

[Robert Brindell]

Bill,

My intent is not to be difficult but it seems you are not willing to 
test/accept this as a bug.  I would like to hear from some other people before 
you change the status of this bug.  I understand that the code referenced is 
not the best syntactical usage (I have unfortunately inherited it).  It is 
however valid.  The document referenced as 
http://www.w3.org/TR/CSS21/selector.html#pseudo-elements does not specifically 
disallow tag names as pseudo-classes.  My biggest contention with Firebird 
regarding this is that the same code usage will render properly in Mozilla 1.5 
and 1.4 (and of course IE5 and IE6).  It seems to me that there should be some 
consistency between the two, especially since the child does not allow all the 
things that the parent will.  Without an answer to why there is a difference 
between the two, I feel this is still a bug.

Comment 6

[Bill Mason]

OK, here we go.

This syntax:

.small,.small:input, .small:textarea, .small:select { }

is not valid and should be ignored.  Why?


http://www.w3.org/TR/CSS21/selector.html#pseudo-elements does not say you can
make up your own pseudo-elements and/or pseudo-classes, and in fact you cannot.
 The exact allowable psuedo-elements/pseudo-classes are defined in that section.

For Firebird to not render anything that you are attempting to style with such a
rule is correct.  The specification requires that if any part of a selector is
not valid, the entire rule/statement is to be ignored.  Reference
http://www.w3.org/TR/CSS21/syndata.html#q10

The bottom line is the syntax you are using is not valid.  There is nothing in
the specification to support your position.

Now, if you can produce a URL or a testcase where your invalid syntax is
(correctly) ignored by Firebird and (incorrectly) accepted by Mozilla, that
would constitute a Mozilla bug.  Moving this bug to Browser/Style System pending
such an example.

Comment 7

[Bill Mason]

Attachment: testcase

Comment 8

[Bill Mason]

In the attached testcase, the CSS using your syntax is correctly ignored in
Firebird 20031019 and Mozilla 20031021 PC/Win2000.

The CSS validator confirms that your syntax is invalid.

Please attach a URL or testcase where you can demonstrate this syntax working,
or this bug will be re-resolved invalid.

Comment 9

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

I'm not that patient.  I don't see anything in this bug that suggests there's a
valid bug here, so marking invalid.  If you disagree, please reopen.  Note
specifically:

http://www.w3.org/TR/CSS21/selector.html#pseudo-elements  (sections 5.10-5.12)
http://www.w3.org/TR/CSS21/syndata.html#q10 ("When a user agent can't parse...")

Comment 10

[Bill Mason]

Verifying, then.

Comment 11

[Robert Brindell]

Well I can not give you guys access to a url to show you because of the nature 
of the system.  I am able to reproduce this every time.  I aggree it is invalid 
syntax and is most likely a bug in Mozilla.  All browsers except Firebird will 
render the example properly.  It is no longer that important to me to have my 
for hundred thousand users migrate to Firebird. Thanks for your time guys.

Comment 12

[Hixie (not reading bugmail)]

This is not a bug in Mozilla. If your CSS is invalid, the CSS spec says the 
browser MUST ignore it. If other browsers don't, then they are buggy. Opera is 
buggy in this respect. I will be getting Opera fixed so that it will also ignore 
invalid CSS like this.

Comment 13

[Robert Brindell]

My comment is that Mozilla DOES render it so it IS a bug in Mozilla.  As you 
put it - "buggy".  No big deal though.

Comment 14

[Hixie (not reading bugmail)]

It does? Ah. Can you attach a testcase? What version of Mozilla are you using?
This is correctly ignored for me using the following test cases:
   http://www.hixie.ch/tests/adhoc/css/selectors/parsing/001.xml
   http://bugzilla.mozilla.org/attachment.cgi?id=133784&action=view

Comment 15

[Robert Brindell]

Attachment: the styles

Comment 16

[Robert Brindell]

Attachment: the html

Comment 17

[Robert Brindell]

Ok.  Here is the code example.  You guys are right.  My test was being 
conducted comparing firebird and mozilla 1.4  When I correctly started testing 
using Mozilla 1.5 the code is ignored.  1.4 alowed the sloppy code but 1.5 does 
not and that is consistant with Firebird.  Thanks again for all of your 
professional comments.

Comment 18

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

Ah, so really...

Comment 19

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

...this is a duplicate of a bug that was fixed a few months ago.

It's helpful to give the version number of what you're testing.  From comment 0
I was under the impression that you were seeing the bug in a 1.5-based Firebird.

*** This bug has been marked as a duplicate of 145968 ***

Comment 20

[Bill Mason]

V. again.

Comment 21

[Robert Brindell]

That is correct.  I thought it was 1.5 from the start but was misinformed.