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

Re: Unable to resolve select/match decisions for 17 Nov draft

From: Alessandro Vernet <avernet@orbeon.com>
Date: Tue, 14 Nov 2006 17:54:21 -0800
Message-ID: <4828ceec0611141754t46fd3586kc996b7249ee1b995@mail.gmail.com>
To: public-xml-processing-model-wg <public-xml-processing-model-wg@w3.org>

On 11/14/06, Norman Walsh <Norman.Walsh@sun.com> wrote:
> I realize now that I cannot see how to implement these decisions for
> the 17 November draft. Saying that the semantics of the 'select'
> attribute on p:input is different when the parent is p:viewport just
> seems horrible. Adding a match attribute to the p:input element for
> p:viewport seems equally horrible.

As mentioned earlier, I also think that we are better off using
'select' everywhere.

If I remember correctly, we arrived at the conclusion that from an
implementation perspective, 'select' and 'match' were similar: it is
not the case that one is better that the other for streaming or
simpler to implement.

So it seems arguments center around simplicity and intuitiveness for
the pipeline author. Using 'match' in viewport is better than 'select'
as it makes it clear that the attribute behaves differently that the
'select' in for-each. But it does so at the cost of introducing a new
concept to XProc (patterns). And most people are not as familiar with
patterns as they are with XPath. I think that the benefit is not worth
the cost. If we are just trying to avoid a confusion with selects on
for-each, we could name that attribute 'select-outer' and keep the
XPath semantic.

Alex
-- 
Blog (XML, Web apps, Open Source):
http://www.orbeon.com/blog/
Received on Wednesday, 15 November 2006 01:54:31 GMT

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