Re: Escaped characters in RDF-1.1 N-Triples literals for Canonical documents

On 12/06/2013 07:11 AM, Andy Seaborne wrote:
> Peter,
> David,
>
> Thank you for your comments regarding the rules for using ECHAR in
> Canonical N-Triples documents.
>
> The wording on use of ECHARS has been clarified:
>
> """
> * Characters MUST NOT be represented by UCHAR.
>
> * Within STRING_LITERAL_QUOTE, only characters not allowed directly in
> STRING_LITERAL_QUOTE (U+0022, U+005C, U+000A, U+000D) should use ECHAR.
> For all other characters, ECHAR MUST NOT be used.
> """

Sorry to bother you again about this, but the phrase "should use ECHAR" 
does not seem like the right conformance phrase to use for *canonical* 
N-Triples.

[[
Within STRING_LITERAL_QUOTE, characters not allowed directly in 
STRING_LITERAL_QUOTE (i.e., U+0022, U+005C, U+000A and U+000D) MUST use 
ECHAR.  For all other characters, ECHAR MUST NOT be used.
]]

Also, there is a run-on sentence:
[[
The whitespace following subject, predicate, and object MUST be a single 
space, (U+0020) all other locations that allow whitespace MUST be empty.
]]
That should be two sentences:
[[
The whitespace following subject, predicate, and object MUST be a single 
space, (U+0020).  All other locations that allow whitespace MUST be empty.
]]

Thanks!
David

>
> https://dvcs.w3.org/hg/rdf/raw-file/default/rdf-turtle/n-triples.html#canonical-ntriples
>
>
> Please could both of you reply with "[RESOLVED]" in the subject line to
> acknowledge handling of this comment.
>
>      Andy
>
>
>
> On 28/11/13 22:28, Peter Ansell wrote:
>> Hi Andy,
>>
>> Thank for working through the ECHAR issue. Having ECHAR consistent
>> across the variations is definitely useful.
>>
>> Could you also clarify the main issue that I am raising here about the
>> Canonical form for N-Triples and which characters are expected to be
>> escaped? It isn't completely clear whether "\" escapes should be used
>> for all 8 represented characters in ECHAR or just the 4 that are
>> disallowed in STRING_LITERAL_QUOTE [1].
>>
>> If only the 4 that are disallowed can be used, then one would need to
>> use the raw character, as the last rule in Canonical N-Triples says
>> that the \u form isn't allowed to be used either, but it isn't
>> completely clear whether that is the intention.
>>
>> If the goal of Canonical N-Triples is to be fairly close to the RDF
>> Test Cases format (with the exception of UTF-8 being preferred against
>> \u|\U) then having at least the 5 escaped ASCII characters that it
>> specifies [2] (and hopefully all 8 from RDF-1.1 N-Triples ECHAR) may
>> make sense.
>>
>> Thanks,
>>
>> Peter
>>
>> [1]
>> http://www.w3.org/TR/2013/CR-n-triples-20131105/#grammar-production-STRING_LITERAL_QUOTE
>>
>> [2] http://www.w3.org/TR/rdf-testcases/#ntrip_strings
>
> On 03/12/13 22:52, David Booth wrote:
>  > 1. http://www.w3.org/TR/n-triples/#dfn-canonical-n-triple-document
>  > says: "Characters MUST be represented directly and not by UCHAR."
>  >
>  > Shouldn't that be something like: "Characters allowed directly in
>  > STRING_LITERAL_QUOTE MUST be represented directly and not by UCHAR." ?
>
>
>
>
>

Received on Friday, 6 December 2013 21:32:40 UTC