[XSL-LIST Mailing List Archive Home] [By Thread] [By Date]

Re: [xsl] Nicer way to change context-node than xsl:for-each ?


Subject: Re: [xsl] Nicer way to change context-node than xsl:for-each ?
From: Andrew Welch <andrew.j.welch@xxxxxxxxx>
Date: Mon, 7 Dec 2009 14:31:59 +0000

2009/12/7 Mukul Gandhi <gandhi.mukul@xxxxxxxxx>:
> On Mon, Dec 7, 2009 at 4:15 PM, Andrew Welch <andrew.j.welch@xxxxxxxxx> wrote:
>> I can't see how the number of templates involved, the modes or
>> namespaces have any affect on whether its pull or push...?
>
> I think, Dimitre is somewhat right. A pure push processing, is natural
> without modes and namespace qualification in select/match patterns.
>
> If we add modes and namespaces to templates, then we deviate from
> natural template processing, which is built into XSLT engine, and that
> makes it pull processing (either complete pull, or partial pull).

Push is simply where the document order drives the processing... not
sure where modes or namespaces come into it.


>> <xsl:apply-templates mode="foo"/>
>>
>> ...is pull when only a single template exists with that mode, but push
>> otherwise?
>
> I think, if there is only one template, and that has a mode, then it's
> same as a template without a mode (and, that's not a correct design
> according to me! Here a template, defined with a mode is redundant
> design. But I think, you just specified this as an example, and not a
> desirable solution!). And the example you've given above is push
> according to me (since there is only one template, with a mode).

It's fine to a have just one template with a particular mode, not
redundant at all, but again I've no idea what that has to do with
push/pull.


-- 
Andrew Welch
http://andrewjwelch.com
Kernow: http://kernowforsaxon.sf.net/


Current Thread
Keywords