W3C home > Mailing lists > Public > www-style@w3.org > July 2011

Re: [cssom] Unrecognized - request for more information

From: Brian Kardell <bkardell@gmail.com>
Date: Wed, 27 Jul 2011 16:26:17 -0400
Message-ID: <CADC=+jfB+fjhzFXhkvy3gRuFRJ5-K=MXZQNd75OuhF__2ESvmQ@mail.gmail.com>
To: www-style@w3.org
It's true that there are parser generators - but as I explained, this
seems to places an unnecessary / undue burden on a lot of current uses
- beyond just parsing it requires figuring out things that the browser
is actually much better suited to figure out (such as which rules this
particular version of this particular browser can and cannot respect.
Exposing it in a nice simple way opens the door to a lot of
possibilities it seems.
- Show quoted text -

On Tue, Jul 26, 2011 at 8:59 PM, Bjoern Hoehrmann <derhoermi@gmx.net> wrote:
> * Brian Kardell wrote:
>>I think this basically gets into what such a proposal would have to say...
>>We can work that out.   I'm merely looking for a really basic subset parse
>>that gives us something more akin to a syntax tree or some nice intermediate
>>meta structure based on "well formed" data.
>
> You can easily add rules to the core syntax grammar that captures things
> that the prose error recovery rules require to skip or otherwise ignore
> in some structure (and then remove the redundant prose requirements). I
> did this last time I made a parser, but naturally ran into problems with
> the specification, which I reported and which the Working Group ignored.
>
> There are also changes, last I checked, in CSS 2.1, such as the handling
> of nesting constructs in selectors, that at the time nobody implemented,
> so I saw little point in developing this into a proposal to add to the
> specification; if I did, it would most likely be ignored anyway, or I'd
> get "decided not to make these changes" mass mails without rationale, or
> whatever the trend in process violations and treating people badly is.
>
> As for actually exposing such information through browser APIs, I would
> be rather unsure whether that is a good idea. If you have the gammar in
> order, there are any number of parser generators that should make it
> easy to roll your own, and you could offer a better API for the things
> you might care about; the browser can only offer a rather unwieldy in-
> terface (due to resource constraints and lack of very compelling use
> cases to make this usable beyond just providing access).
> --
> Björn Höhrmann · mailto:bjoern@hoehrmann.de · http://bjoern.hoehrmann.de
> Am Badedeich 7 · Telefon: +49(0)160/4415681 · http://www.bjoernsworld.de
> 25899 Dagebüll · PGP Pub. KeyID: 0xA4357E78 · http://www.websitedev.de/
>
Received on Wednesday, 27 July 2011 20:26:45 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:42 GMT