Re: [Json] Proposal: JSON text sequence MIME type and Proposed Standard

Yeah, the options are:

1. zero or more whitespace characters
2. 1 or more whitespace characters
3. A specific required set of whitespace characters, for example \s*\n\s*
or just \n

On Tue, Mar 11, 2014 at 8:12 PM, Manger, James <
James.H.Manger@team.telstra.com> wrote:

> > > I agree that a newline is the best separator for creators to include,
> > and always including it is best.
> > >
> > > Some (most?) receivers will be more lenient and accept values
> > separated by any whitespace, or with no separator (for objects, arrays,
> > and/or strings). Perhaps it is best for interop to say receivers MUST
> > be this lenient.
>
> > Good point.  Sequence parsers MUST be able to handle newline text
> > separators and SHOULD be able to handle other whitespace, as well as
> > no separator in the unambiguous cases.  Sequence encoders SHOULD use a
> > newline separator and MAY use other whitespace (with some guidance as
> > to why newline is best); no need to mention when the separator may be
> > omitted.
>
> You don't quite guarantee interop with "encoders MAY use other whitespace"
> and "parsers SHOULD be able to handle other whitespace".
>
> The spec needs to say: either "encoders MUST use a newline separator"; or
> "decoders MUST be able to handle any whitespace (including none in the
> unambiguous cases)".
>
> --
> James Manger
> _______________________________________________
> json mailing list
> json@ietf.org
> https://www.ietf.org/mailman/listinfo/json
>