62304 - [CASCADE] Can't style form controls

Status: VERIFIED FIXED

(Core :: Layout: Form Controls, defect, P3)

Description

[Pierre Saslawsky]

This bug is extracted from bug 45850.  See the comments from [Jeremy Wrinkle 
2000-12-06 16:58]: userContent.css can no longer style the form controls.

This is a regression introduced by the latest changes to implement forms.css.

Comment 1

[Pierre Saslawsky]

Attachment: proposed patch

Comment 2

[David Hyatt]

a=hyatt, although we really should fix user sheets so that they aren't appended
as backstop sheets. :)

Comment 3

[David Baron :dbaron: (⌚️UTC-5, no longer working on Mozilla)]

Well, user stylesheets should really be a middle level between the UA stylesheet
and the author(=page) stylesheet, as the CSS spec says...

Comment 4

[Marc Attinasi]

Pierre, I cannot apply the patch, and from looking at it I am confused. Is the
patch intact? If so, I don't understand what happened to the loop over the
chrome registry stylesheets - I probably need the full context but, like I said,
I cannot apply it. Can you attach another with 'diff -u'? thanks.

Comment 5

[buster]

Looks good, with one simple question:

In this code:

> sheets->Count(&count);
> for(PRUint32 i=0; i<count; i++) {
>   sheets->GetElementAt(i, getter_AddRefs(sheet));

I just want to verify that "count" is valid in the loop.  There is nothing in
the code that follows in the loop that can change the count of items in
"sheets", right?

If true, sr=buster.

Comment 6

[Pierre Saslawsky]

Marc, the patch consists in appending the chrome sheets after the user sheets 
(instead of before).  I'm going to attach new diffs and the function before and 
after the changes.

In response to buster (just for the record: he's not copied on this bug and he 
didn't seem to care): correct, the count of stysheets cannot be changed from 
within the loop.  The sheets are read from the chromeRegistry and appended to the 
styleset.  The styleset cannot change the chromeRegistry. 

Comment 7

[Pierre Saslawsky]

Attachment: diffs -u

Comment 8

[Pierre Saslawsky]

Attachment: the function before the changes

Comment 9

[Pierre Saslawsky]

Attachment: the function after the changes

Comment 10

[Marc Attinasi]

Thanks Pierre - looks great, r=attinasi

Comment 11

[Hixie (not reading bugmail)]

Netscape's standard compliance QA team reorganised itself once again, so taking 
remaining non-tables style bugs. Sorry about the spam. I tried to get this done 
directly at the database level, but apparently that is "not easy because of the 
shadow db", "plus it screws up the audit trail", so no can do...

Comment 12

[Pierre Saslawsky]

Shoot!  I forgot to checkin that fix in the last 2 milestones.  Moved to 0.9.1.

Comment 13

[Pierre Saslawsky]

fixed

Comment 14

[Hixie (not reading bugmail)]

QA -> HTML Form Controls

Comment 15

[Madhur Bhatia]

marking verifed as per developers comments