W3C home > Mailing lists > Public > public-xml-processing-model-comments@w3.org > December 2008

Re: 5.4 p:output. Informative para request

From: Norman Walsh <ndw@nwalsh.com>
Date: Sun, 21 Dec 2008 15:10:21 -0500
To: public-xml-processing-model-comments@w3.org
Message-ID: <m27i5tcms2.fsf@nwalsh.com>
"Vasil Rangelov" <boen.robot@gmail.com> writes:

> To me, it's not clear from this how should the processor behave if the step
> actually produces an output on that port. Does the binding only get used as
> a default, or does it override anything the step might have otherwise
> produced on that port? Whatever the case, it should be said explicitly.
> (Personally, I'd prefer for the binding to be used as a default)

You can only put a binding inside a p:output on a compound step, in
which case the result of that binding *is* the output, there's no
other way for the step to produce anything on that port. Well, you could
make a sequence:

     <p:document .../>
     <p:pipe step="somechildstep" port="result"/>

                                        Be seeing you,

Norman Walsh <ndw@nwalsh.com> | Design and programming are human
http://nwalsh.com/            | activities; forget that and all is
                              | lost.--B. Stroustrup

Received on Sunday, 21 December 2008 20:11:08 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:28:26 UTC