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

[closed] Re: The 'content-type' attribute is not optional on c:data

From: Norman Walsh <ndw@nwalsh.com>
Date: Thu, 21 Aug 2008 07:36:45 -0400
To: public-xml-processing-model-wg@w3.org
Message-ID: <m2r68izjjm.fsf@nwalsh.com>
/ Toman_Vojtech@emc.com was heard to say:
| Hi all,
| I am a bit confused by the text about the content-type attribute in
| section 5.14 (p:data). The text contains the following sentences:
| "If no wrapper element is specified, the default is c:data"
| "The wrapper element should have a content-type attribute"
| "If a content-type attribute is specified on a c:data wrapper, ..."
| But, since @content-type is not optional in c:data, I don't think these
| sentences work together very well.

I think that's the bug. I've made content-type optional on c:data (it's
only a SHOULD in the spec.)

| If I remember it correctly, c:data was originally only used in p:xquery,
| where it was required to always specify a content type (the p:xquery
| section still says: "If the document root element is c:data or has a
| c:content-type attribute that specifies..."). 
| I think we should either make @content-type optional in c:data (which
| could make problems in p:xquery) or reword the text about p:data in such

I don't think it makes any problems for XQuery. If you use c:data and pass
some binary resource (which will be base64 encoded), your query will fail.
But that's a feature not a bug :-)

| a way that is mandates that @content-type must be specified if c:data is
| used. I am not sure which option is better/worse, though.

                                        Be seeing you,

Norman Walsh <ndw@nwalsh.com> | In the life of saints, technically so
http://nwalsh.com/            | called, the spiritual facilities are
                              | strong, but what gives the impression
                              | of extravagance proves usually on
                              | examination to be a relative deficiency
                              | of intellect.--William James

Received on Thursday, 21 August 2008 11:38:03 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:32:46 UTC