Re: Relative URIs in p:log

Toman_Vojtech@emc.com writes:
> The spec does not say what happens when a relative URI is used in
> p:log/@href. Shouldn't it behave the same way as p:load, p:store, ...
> That is: "If it is relative, it is made absolute against the base URI of
> the element on which it is specified".

Yes.

> I encountered problems with this when I was running the latest version
> of the test xinclude-002.xml over the net. It contains this line:
>
> <p:log port="result" href="/tmp/out.xml"/>
>
> In our present implementation, we always resolve the value of @href
> against the current base URI (which in this case is equal to
> "http://tests.xproc.org/tests/required/xinclude-002.xml"), and we end up
> attempting to log to: http://tests.xproc.org/tmp/out.xml. Note that in
> this case, the value of @href is not relative, but it does not specify a
> scheme, so the proper fix here would be to set the href to
> "fils:///tmp/out.xml".

Yeah, that looks like a bug. It should be changed to
file:///tmp/out.xml in the test, I think.

> Also, is the value of p:log/@href an URI or an IRI?

All URI are IRI.

                                        Be seeing you,
                                          norm

-- 
Norman Walsh <ndw@nwalsh.com> | We cannot put off living until we are
http://nwalsh.com/            | ready. The most salient characteristic
                              | of life is its coerciveness: it is
                              | always urgent, 'here and now' without
                              | any possible postponement. Life is
                              | fired at us point blank.--José Ortega Y
                              | Gasset