ISSUE-59: Whether to require or not of processors a certain behaviour in lexical preservation mode when xsi:type occurs to indicate grammar switch
Whether to require or not of processors a certain behaviour in lexical preservation mode when xsi:type occurs to indicate grammar switch
- State:
- CLOSED
- Product:
- EXI Spec
- Raised by:
- Takuki Kamiya
- Opened on:
- 2010-08-23
- Description:
- There appear to be a factor of asymmetry between an encoder and a decoder
in the ability to parse texural xsi:type values. - Related Actions Items:
- No related actions
- Related emails:
- RE: Agenda for 02 Nov EXI Telecon (from Youenn.Fablet@crf.canon.fr on 2010-11-04)
- Agenda for 02 Nov EXI Telecon (from tkamiya@us.fujitsu.com on 2010-11-02)
- Re: Agenda for 27 Oct EXI Telecon (from carine@w3.org on 2010-10-27)
- RE: Agenda for 27 Oct EXI Telecon (from Youenn.Fablet@crf.canon.fr on 2010-10-27)
- AW: Agenda for 27 Oct EXI Telecon (from daniel.peintner.ext@siemens.com on 2010-10-27)
- Agenda for 27 Oct EXI Telecon (from tkamiya@us.fujitsu.com on 2010-10-26)
- Agenda for 13 Oct *Extended* EXI Telecon (from tkamiya@us.fujitsu.com on 2010-10-19)
- Agenda for 13 Oct EXI Telecon (from tkamiya@us.fujitsu.com on 2010-10-12)
- ISSUE-59: Whether to require or not of processors a certain behaviour in lexical preservation mode when xsi:type occurs to indicate grammar switch [EXI Spec] (from sysbot+tracker@w3.org on 2010-08-23)
Related notes:
Note that preservation of namespace declaration is not possible
in strict mode.
http://lists.w3.org/Archives/Member/member-exi-wg/2010Jul/0044.html
The participants of 2010-07-14 telecon noted that:
"Strict mode SHOULD NOT be used with QName value content,
including such cases in which xsi:type attributes are used
with lexical preservation on, and the spec ought to clearly
state that."
http://lists.w3.org/Archives/Member/member-exi-wg/2010Jul/0096.html
The above note does not seem to address the issue present in
non-strict mode.
TK suggested to describe the implications clearer in section
6.3 Fidelity Options.
http://lists.w3.org/Archives/Member/member-exi-wg/2010Aug/0026.html
In 2010-10-06 telecon, the above suggestion was considered. The specification
already seems to be clear enough about the hitch. Therefore, no further
clarification is needed in that respect. However, this issue neither has diminished nor has grown as a result of that.
The WG noted it was up to implementations to decide whether to or
not to come up with ways for making those streams deviating from
the "SHOULD" norms described in the spec usable. The spec quietly
leaves that out with doors open for ingenuity that suit a particular
needs.
Display change log