W3C home > Mailing lists > Public > public-xml-processing-model-wg@w3.org > December 2009

Re: Version attribute issues

From: Norman Walsh <ndw@nwalsh.com>
Date: Tue, 08 Dec 2009 08:12:13 -0500
To: public-xml-processing-model-wg@w3.org
Message-ID: <m2skblzk2a.fsf@nwalsh.com>
"Toman_Vojtech@emc.com" <Toman_Vojtech@emc.com> writes:
> Good catch.

Actually, on closer inspection, I wonder if the intent of 2.13 was to say
that XProc version applies to p:pipeline, p:declare-step, or p:library
and no where else.

Do we really *need* or *want* to allow version mixing at a lower level
than whole pipelines?

> What about simply renaming p:xslt/@version to
> p:xslt/@xslt-version? 

Hmph. We could. Anyone else have a preference for which we rename,
assuming we don't limit XProc version to just whole pipelines (or
libraries, of course).

> Regarding the nested p:declare-step, I think it does inherit the
> version
> information (at least in our implementation, it does). Section 2.13
> currently says:
>
> "The version identified applies to the element on which the version
> attribute appears and all of its descendants, unless or until another
> version is explicitly identified."

Right. I agree. We do reset some things at nested
declare-step/pipeline boundries (variables and options aren't
inherited for example) but I don't think that need or should apply to
version.

                                        Be seeing you,
                                          norm

-- 
Norman Walsh <ndw@nwalsh.com> | For the mental patient's family and
http://nwalsh.com/            | society, mental illness is a 'problem';
                              | for the patient himself it is a
                              | 'solution'.--Thomas Szasz

Received on Tuesday, 8 December 2009 13:12:54 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 December 2009 13:12:54 GMT