Re: [Cbor] Chunks with tags inside indefinite-length string (major type 2 and 3)

[Carsten Bormann <[e-mail]>]

On Dec 10, 2019, at 02:46, Dale R. Worley <[e-mail]> wrote:
> 
> Faye Amacker <[e-mail]> writes:
>> And the pseudocode in Appendix C allows tags for chunks inside
>> indefinite-length strings.

As Jeffrey noted, the pseudocode actually catches chunks that are of a different major type than the indefinite string they are contained in.

> It seems to me that the safe option, the one that will cause the least
> trouble in the future, is to define tags on items inside
> indefinite-length structure as syntactically well-structured but
> semantically having no defined meaning.  

Why would we allow tags here?  Indefinite strings are a very limited syntactical structure; there is no need to allow flexibility here (in particular if there is no semantic gain from that).

> Actually, that's the same as we
> treat use of tags whose values haven’t been defined.

I don’t follow.

> Functionally, that means that no current application should generate
> them (and thus an application for the same universe of values should
> expect to never see them), but "generic" decoders should be prepared to
> see such tags.

I don’t think so.  Both the text and the pseudocode disallow tags on chunks.

Grüße, Carsten