OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

uddi-spec message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Groups - Action Item Modified: #0039 Submission of a TN discussing implementation and usage issues related with the use of Unicode in elements and attributed of the anyURI datatype



OASIS UDDI Specification TC member,

Luc Clement (luc.clement@systinet.com) has modified this action item.

Number: #0039
Description: Submission of a TN discussing implementation and usage issues related with the use of Unicode in elements and attributed of the anyURI datatype
Owner: Andrew Hately (hately@us.ibm.com)
Status: Open
Due: 16 Jul 2004

Comments:
Luc Clement  2004-03-17 00:11 GMT
As a follow-up to [1], review Draft RFC (draft-duerst-iri-06.txt) and identify:
1.	impact on spec; and 
2.	next steps and ARs

Background:
The XML Working Group, I18N and URI activity leads have been contacted on the interop matter identified by Andrew [1] relating to internationalized URIs and its use with the anyURI datatype. 

In response, the I18N Activity Lead stated that draft-duerst-iri-06.txt is close to submitting it to the IETF and to watch http://www.w3.org/International/iri-edit/ for status. At the time of the meeting, the draft RFC had not been reviewed by any TC member, so we could not determine what if any guidance we could obtain from it. 

[1] http://lists.oasis-open.org/archives/uddi-spec/200403/msg00007.html
[2] http://www.oasis-open.org/apps/org/workgroup/uddi-spec/download.php/5649/TC_FTF_Minutes-V1.7-20040210-12.htm#_Toc65400705

Luc Clement  2004-05-12 15:10 GMT
Claus communicated with Martin DŁerst on this matter Ė [2] below refers. The response he received is that our interpretation was incorrect; in fact internationalized (i.e. Unicode) chars are allowed in the anyURI data type. XML Schema Part 2: Datatypes, 3.2.17.1, Lexical representation (http://www.w3.org/TR/2001/REC-xmlschema-2-20010502/#anyURI) states: "The .lexical space. of anyURI is finite-length character sequences which, when the algorithm defined in Section 5.4 of [XML Linking Language] is applied to them, result in strings which are legal URIs according to [RFC 2396], as amended by [RFC 2732]."

Per 3.2.17, the mapping from anyURI values to URIs is as defined in Section 5.4 Locator Attribute of the XML Linking Language which describes how a URI reference is to be handled by URI resolvers and what escaping procedure resolvers need to carry out. As such, it is entirely legal to use Unicode chars in element and attributes defined as anyURI. 

So our existing schema is correct, and will not need change. As such, tools like Xerxes that currently reject use of Unicode chars in attribute/elements of type anyURI are doing so erroneously.

There is still a need for a TN discussing implementation and usage issues that would be encountered by developers, implementers and users. Claus agreed to produce a TN describing this matter.

Andrew suggested that a CR is forthcoming describing v2 to v3 migration issues in that there are fields in UDDI V2 which are defined as a string which become anyURI in V3, and hence there are characters which are valid in V2 (such as spaces, and formatting characters) which would not be valid elements/attributes when converted to V3 (overviewURL, for example). Andrew will author this TN.

Luc Clement  2004-07-05 19:51 GMT


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]