Re: DR 09-0286 — OPC: The syntax of "references"

MURATA Makoto eb2m-mrt at asahi-net.or.jp
Thu Jun 16 16:14:10 CEST 2011


Dear colleagues,

To close this DR, we have to add a sentence in 9.2

     References from one part to another are specified as relative
LEIRI references.

and also an entry in the bibliography.

   Legacy extended IRIs for XML resource identification, W3C Note, 3
November 2008


As part of the revision process, I think that we should rewrite the
content of Annex using
LEIRIs and make the rewrite as part of 9.2.

Regards,
Makoto

2011/2/8 MURATA Makoto <eb2m-mrt at asahi-net.or.jp>:
> Chris,
>
> Sorry for my belated reply.  This issue causes too much brain damage!
>
> You think that Shawn's wording, which relies on Annex B, is fine.  But
> 9.1.1.1.1  Part IRI Syntax  and 9.1.1.1.2  Part URI Syntax have something
> different from Annex B about the definition of part names.  Depending on
> which we use as a basis, the answer becomes different.
>
> Let me ask a question.  Which of the strings shown below is
> allowed as a reference?
>
> a.xml
> ./a.xml
> /a.xml
> /../a.xml
> \a.xml
> %a.xml
> ц.xml
> ./ц.xml
> %61.xml
> %5Ca.xml
> %25a.xml
> %61.xml
> %5C%2e/a.xml
> %5C%41.xml
> /%2541.xml
> /%2e/%2e/a.xml
> /%41/%61.xml
> /%D1%86.xml
> /./%D1%86.xml
> /a/%D1%86.xm
> \%2e/a.xml
> \%41.xml
>
>
>
> 2010/9/1 Chris Rae <Chris.Rae at microsoft.com>:
>> Hi Murata-san - as the standard states right now, these can be "any URI and nothing else". I am guessing from your response that you're not going to be wild with excitement about that one.
>>
>> I think this specific DR can be resolved as it was created to request clarifying the syntax of references, and the current solution from Shawn seems to do that. You may want to submit a separate DR requesting the broadening of these syntaxes - I'd be very interested to look at other possible solutions although I'd be keen for the existing URI schemes still to work, just in order to allow current IS 29500 documents to also be compatible with future IS 29500s.
>>
>> You're right, I was talking nonsense about the "pack:" requirement, they don't have to include the scheme specifier. Apologies.
>>
>> Chris
>>
>> -----Original Message-----
>> From: eb2mmrt at gmail.com [mailto:eb2mmrt at gmail.com] On Behalf Of MURATA Makoto
>> Sent: 26 August 2010 14:49
>> To: Chris Rae
>> Cc: e-SC34-WG4 at ecma-international.org
>> Subject: Re: DR 09-0286 -- OPC: The syntax of "references"
>>
>> Chris,
>>
>> I do not think your answer is clear yet.
>>
>> I would like to hear answers such as
>>
>>  "any IRI and nothing else"
>>  "any LEIRI  and nothing else"
>>  "any IRI having the pack scheme and nothing else"
>>  "any LEIRI having the pack scheme and nothing else"
>>
>> I do not want to percent-encode non-ascii when I reference to another OPC part.  So, I did not list "any URI and nothing else"
>>
>> And are you sure that what you call references is required to begin with "pack"?
>>
>> Cheers,
>> Makoto
>



-- 

Praying for the victims of the Japan Tohoku earthquake

Makoto


More information about the sc34wg4 mailing list