Re: [cssom] Author-defined at-rules

> Again, I sympathize with everything you say about extensibility, but I am 
> not really moved by this specific proposal. If this is just about storing 
> stuff in CSS files remember that you can have whole at-rules in variables 
> in you feel like it:

While I don't think abusing custom properties to define at-rules-like things 
is a good idea (both performance-wise and conceptually-wise), I'm still 
leaning towards the idea that the best way to deal with this is to 
preprocess some "HyperCSS" files (SASS, LESS, ...) where you can introduce 
any at-rule you may want to generate a browser-understandable CSS file and a 
companion JavaScript file containing user-defined things. There's no point 
in keeping in CSS format data that you'll need to convert to JavaScript 
understandable information just after if you cannot harness the CSS power at 
some point in a way that would be impossible or cumbersome in CSS. As Tab 
and Simon noted, you can already use in JavaScript the conditional features 
of CSS, it's just a matter of rewriting.

As far as extensibility goes, I'd be far more interested in getting a 
"querySelectorLive" function which would fire asynchronously a function when 
an element starts or stops to match a selector. Right now, the only way to 
do it is to make diffs from "qSA" and "qSA" is not even guaranteed to use a 
cache so you may end up reexecuting the query all the times instead of 
making a true diff.

Then, a function that fires when the computed value of a property changes on 
an (displayed) element would be awesome. But both are approximately 
prollyfillable from each others so the easiest should be prioritized and we 
can deemphasize the other.

Received on Saturday, 29 June 2013 09:17:36 UTC