W3C home > Mailing lists > Public > xproc-dev@w3.org > September 2009

RE: make-absolute-uris may match strings which are not of type anyURI ...

From: <Toman_Vojtech@emc.com>
Date: Mon, 21 Sep 2009 08:48:50 -0400
Message-ID: <6E216CCE0679B5489A61125D0EFEC78710DBAFF5@CORPUSMX10A.corp.emc.com>
To: <xproc-dev@w3.org>
> Thanks for clarifying this! So, if the string-value of the matched
> item (element or attribute) is _not_ of type anyURI, what should the
> XProc spec mandate for p:make-absolute-uris ?

I think p:make-absolute-uris should work with any node you point it to.
Whether a node is of type anyURI or not is not important at all in this
case, as long as the node contains a string value that can be treated as
a URI/IRI.

I was trying to understand whether the input for p:make-absolute-uris
must be URI escaped in advance, or not. In my view, the requirement
should be the same as for @xml:base values, and if I understand it
correctly, @xml:base values do not have to be escaped.

> a) the result is implementation defined
> b) in case XPath 2.0 is supported, fn:encode-for-uri must be applied
> to the string-value of the matched item, in case only XPath 1.0 is
> supported the result is implementation defined.
> c) the step should fail

In my opinion:

d) because the URI may contain characters not allowed in URIs,
processors are required to perform URI escaping themselves when they
need to further process the URI information.

> Currently the XProc spec does not cover this case.

I am still not sure what the right behavior is. I will bring this up to
the WG for discussion.

Regards,
Vojtech
Received on Monday, 21 September 2009 12:49:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 21 September 2009 12:49:33 GMT