W3C home > Mailing lists > Public > xproc-dev@w3.org > September 2008

Re: How does a processor get selected?

From: Florent Georges <fgeorges@fgeorges.org>
Date: Wed, 24 Sep 2008 18:01:02 +0200
Message-ID: <ebaca5bf0809240901p17a74056x8ed7de84f516fda2@mail.gmail.com>
To: "Norman Walsh" <ndw@nwalsh.com>
Cc: "XProc Dev" <xproc-dev@w3.org>

2008/9/24 Norman Walsh wrote:

> "Florent Georges" writes:

>>   I didn't followed XProc really closely for a long time (I think the
>> last time I've read a draft was while flying to XML Prague...)  But is
>> there any mechanism similar to XSLT's import and import precedence,
>> allowing one user to overload a standard stylesheet to plug his own
>> logic and customization?

> No, but I don't think the XProc model really supports that kind of
> interchange. Unlike XSLT, which is a collection of templates, pipeline
> steps are tightly connected.

  Not the same way than in XSLT, but what for instance if we were able
to plug a processor on the link between two processors?  That would
enable one to create a proxy over (in front of or behind are maybe
better) a processor.  For example that would allow to make XInclude
processing at one point in a standard pipeline that didn't support
that initially.

  I am just thinking aloud...

  That's maybe something we'd be better experimenting after the XProc
REC, to get real usage for XProc 1.1...

-- 
Florent Georges
http://www.fgeorges.org/
Received on Wednesday, 24 September 2008 16:01:43 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 24 September 2008 16:01:43 GMT