Re: [apps-discuss] draft-pbryan-zyp-json-pointer: name syntax for non-ASCII

RFC 3986 prescribes encoding characters in UTF-8 and percent-encoding
non-unreserved characters. Furthermore, JSON by definition uses Unicode
for all string values (most often encoding in UTF-8). Do these not
address the issue?

Paul

On Mon, 2011-11-21 at 15:12 +0100, Julian Reschke wrote:

> Hi there.
> 
> In 
> <https://tools.ietf.org/html/draft-pbryan-zyp-json-pointer-02#section-3> 
> we have:
> 
>     A JSON Pointer is a sequence of zero or more reference tokens, each
>     prefixed by a "/" (%x2F) character.  Each reference token is a
>     sequence of unreserved and/or percent-encoded characters, per
>     [RFC3986].
> 
>     json-pointer = *( "/" reference-token )
>     reference-token = *( unreserved / pct-encoded )
> 
>     Characters in reference tokens that are not unreserved SHOULD be
>     percent-encoded, per [RFC3986], and MUST be so encoded as "%2F" if
>     the character is "/" to avoid being interpreted as a reference token
>     prefix.
> 
>     It is an error condition if a JSON Pointer does not conform to this
>     syntax.
> 
> This doesn't seem to consider the case where the reference token 
> contains non-ASCII characters, which can happen with JSON.
> 
> There seem to be to obvious ways to address this:
> 
> (1) Allow non-ASCII characters in the pointer (which would make the 
> pointers be more IRI-like, and I think that's consistent with XPath), or
> 
> (2) Require UTF-8 encoding.
> 
> I believe (1) makes more sense here.
> 
> Best regards, Julian
> _______________________________________________
> apps-discuss mailing list
> apps-discuss@ietf.org
> https://www.ietf.org/mailman/listinfo/apps-discuss