RE: BATH PROFILE - XML RECORDS

I think ISO 8824 is perfectly clear about this.  If the external object is a
single ASN.1 type, then you may encode it with the "single-ASN1-type"
choice.  If it is not then you may not and you are reduced to encoding with
the "octet-aligned" or "arbitrary" choices.  The fact that you may also
encode a single ASN.1 type as "octet-aligned" or "arbitrary" (the part we
never agreed on) is immaterial.

So encoding an external object that is not specified in ASN.1 as a
"single-ASN1-type" is wrong.

j.

> -----Original Message-----
> From: www-zig-request@w3.org [mailto:www-zig-request@w3.org]On Behalf Of
> Ray Denenberg
> Sent: Friday, October 04, 2002 3:52 PM
> To: zig
> Cc: Sebastian Hammer
> Subject: Re: BATH PROFILE - XML RECORDS
>
>
>
> Sebastian Hammer wrote:
>
> > >recordSyntax = XML or SUTRS
> >
> > This target appears to use an invalid encoding for XML records
> -- they come
> > back as a BER-encoded datatype, whereas I am pretty sure that the proper
> > way to represent XML is as an "Octet-aligned" type in the EXTERNAL.
>
> The issue is with XML, not sutrs? In other words, sutrs come as
> single-ASN1-type and that's ok?
>
> > Ray, we looked around a bit but had a hard time finding a hard
> statement to
> > this effect on the Maintenance agency site, or indeed in the
> record syntax
> > list.
>
> There is no recorded agreement on this.  We were never able to come to
> agreement.  (There was a draft agreement,  I've looked everywhere
> and can't
> find it, and unfortunately I don't remember exactly what it was.The list
> archive earlier than 2000 has been lost.) However it's worth
> trying again, as I
> think there is a good chance that the reasons we couldn't agree
> no longer apply
> (i.e.  the "disagreers" might all be gone).
>
> If we distinguish syntaxes that are described in ASN.1 (e.g.
> sutrs, grs-1) from
> those that aren't (e.g. xml and all the marc sytntaxes)  it
> shouldn't be too
> difficult to reach an agreement to use octet aligned for the latter.
>
> --Ray

Received on Friday, 4 October 2002 16:21:00 UTC