RE: Base URI of the output of p:data

Hi,

Good question. I think it is the absolute IRI used to retrieve the data, similar to p:load and p:document. That is, at least, what our implementation does. Although I must admit the specification is not particularly clear on that.

Another way of looking at this is to take p:data as a more or less simplified, static variant of p:http-request. The base URI of the output of p:http-request follows the href value (“[...] the base URI of the output document is the resolved value of the href attribute from the input c:request.”), so that’s another argument for p:data to behave the same way.

Regarding your question about p:document, I think you are right - although again, I don’t think the specification is clear enough. If you look at the specification of p:load, then p:load step is just a more powerful extension of p:document. I think the last sentence of 7.1.14 (p:load) - “The base URI of the result is the (absolute) IRI used to retrieve it.” - should actually have been in 5.13 (p:document).

Regards,
Vojtech

From: temp [mailto:temp44@szm.sk]
Sent: Friday, July 13, 2012 2:53 PM
To: xproc-dev@w3.org
Subject: Base URI of the output of p:data

Hello,
I have a question regarding the output of the p:data connection.

The document that is constructed by p:data has which base URI? Is it the absolute IRI used to retrieve the binary data (analogy of p:load), or rather the base URI of the p:data itself? Or is it simply empty?

As for p:document, I suppose it is clearly the absolute IRI used to retrieve the document... am I right on this one?
Thanks
Mio


=== reklama ======================================
http://mail.szm.com - e-mail a priestor na www stranku zadarmo
http://webhosting.szm.com - domény a webhosting za najnižšie ceny