W3C home > Mailing lists > Public > public-xml-processing-model-wg@w3.org > November 2006

RE: Match Pattern Proposal

From: Grosso, Paul <pgrosso@ptc.com>
Date: Thu, 2 Nov 2006 10:23:19 -0500
Message-ID: <CF83BAA719FD2C439D25CBB1C9D1D3020533C013@HQ-MAIL4.ptcnet.ptc.com>
To: "public-xml-processing-model-wg" <public-xml-processing-model-wg@w3.org>

And predicates can contain select patterns, so any
analysis of (predicates in) match patterns must 
include the same code that would analyze select
patterns in general.

So I fail to see how using match patterns instead
of select patterns simplifies the implementation.

Besides, if a given pipeline can be written using
just predicate-less match patterns, then when it
is written using select patterns, it will be just
as streamable.

So I still don't see how restricting to match patterns
will make pipelines any more streamable.

paul 

> -----Original Message-----
> From: public-xml-processing-model-wg-request@w3.org 
> [mailto:public-xml-processing-model-wg-request@w3.org] On 
> Behalf Of Alex Milowski
> Sent: Thursday, 2006 November 02 09:05
> To: public-xml-processing-model-wg
> Subject: Re: Match Pattern Proposal
> 
> 
> Grosso, Paul wrote:
> > I prefer select semantics all around.
> > 
> > Given (per Richard's comment) match semantics don't
> > help with the streaming issue, what (other than
> > personal preference) are the advantages of match
> > over select?
> 
> They do help in a big way.  There are predicates that
> can prevent streaming without caching or, in the worst
> case, having the whole document.
> 
> Match patterns limit this analysis to predicates.
> 
> --Alex Milowski
> 
> 
Received on Thursday, 2 November 2006 15:23:41 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:21:49 GMT