Re: CSS Mixins proposal

Hi list,

I tested the following 3 patterns for using @mixin declarations with
fallback in a number of browsers:

.finalPosition {
  background: green;
  color: white;
  @mixin includeMe(foo, bar);
}

This style rule tests whether a mixin declaration at the end of a block
affects the rest of the properties in that block.

.nonFinalPosition {
  background: green;
  @mixin includeMe(foo, bar);
  color: white;
}

This style rule tests whether a mixin declaration in the middle of a block
affects the rest of the properties in that block.

.artificialFinalPosition {
  background: green;
  @mixin includeMe(foo, bar);
}

.artificialFinalPosition {
  color: white;
}

This style rule tests whether a mixin declaration at the end of a block
affects properties declared in subsequent blocks.

I'm happy to report that on every browser I could get my hands on, all three
classes result in a green background with white text when applied to a div.
 I tested the following browsers:
* Internet Explorer 6
* Internet Explorer 7
* Internet Explorer 8
* Internet Explorer 9 beta
* Safari 3
* Safari 5
* Firefox 2
* Firefox 3
* Firefox 3.6
* Firefox 4
* Chrome 9
* Chrome 10
* Opera 11

I think it's clear that using an @-rule in this manner is safe on pretty
much every browser likely to be currently in use, or at least the vast
majority.

Cheers,
    -Shane

On Wed, Mar 23, 2011 at 8:34 AM, Sylvain Galineau <sylvaing@microsoft.com>wrote:

> [Tab Atkins:]
> > On Tue, Mar 22, 2011 at 1:41 PM, Sylvain Galineau <
> sylvaing@microsoft.com>
> > wrote:
> > > [Tab Atkins:]
> > >> On Tue, Mar 22, 2011 at 12:46 PM, Alan Gresley <alan@css-class.com>
> > wrote:
> > >> > I will go into detail of parsing in a follow up email.
> > >>
> > >> No offense, but please don't, unless it's relevant to the Mixins
> > proposal.
> > >> If you must detail parsing irregularities, please do so in the form
> > >> of bugs against the relevant browsers instead.
> > >
> > > No offense, but as long as they're related to the syntax you've
> > > proposed I think this is perfectly relevant. If someone is kind enough
> > > to take the time to explain the impact of any new syntax on existing
> > > browsers, they should be most welcome to do so. Compatibility with the
> > > ideas in your head, on your blog and in some WebKit patch are
> secondary.
> >
> > That's precisely why I said "unless it's relevant to the Mixins
> proposal".
> > ^_^  If it's unrelated parsing bugs, like Opera 11 incorrectly parsing
> > character escapes in some situations, that should be reported directly to
> > Opera and/or be brought up in a separate thread; in this thread it would
> > simply be clutter, and be harder to search for later than if it had its
> > own, appropriately titled, thread.
>
> Right, my bad. I saw feedback being shut down before it was produced and
> ran for the gun. (Must. Relax.) This being said, Mr Gresley doesn't exactly
> have a record of bringing random feedback to this mailing list so if he
> thinks browser parsing issues are worth sharing here I'm inclined to hear
> him out.
>
> Whether those issues belong to Mixins or deserve their own thread is now
> his
> call...
>
>
>
>
>