Re: [saag] ASN.1 vs. DER Encoding

[Nico Williams <[e-mail]>]

On Wed, Mar 27, 2019 at 09:45:16AM +0000, Peter Gutmann wrote:
> Nico Williams <[e-mail]> writes:
> 
> I realise this is the standard ASN.1-vs-everything-else debate that comes up
> every year or so, without wanting to dig up the standard responses from every
> previous time it's occurred I'll try and make just one comment:

It does.

> >Now to back up that assertion:
> >
> >1) TLV encodings are bloated by nature due to being highly redundant.
> >
> >2) That redundancy is a source of errors when manually coding a codec.
> 
> It's actually the opposite, with a proper TLV encoding you can create a
> simple, universal recursive-descent parser that will take any arbitrary
> encoded blob and report "valid" or "not valid".  It's with the non-redundant,
> or at least non-self-describing, encodings where you need to hand-roll a
> parser each time any field anywhere is updated, and which can't be statically
> checked like ASN.1 (meaning BER/DER) can.

I covered that.  I understand that one can built BER/DER/CER dumpers
that know nothing of the schema, with some lossage of type information
when using IMPLICIT tagging (none otherwise).

I don't think that's important because as long as you know the outermost
type, you can always parse PER/OER/XDR/whatever by reference to the
actual schema.  The price to pay for self-describing data is bloat and
more chances for hand-rolled codecs to have security vulnerabilities.

> >Thus there is almost zero benefit to self-describing encodings.
> 
> ... apart from the fact that they can be statically analysed to check whether
> they're well-formed or not, unlike the encodings in PGP, TLS, IPsec, SSH, ...

The protocols you list don't use a formal syntax, which instantly makes
validity checking harder (can't generate the code!).  But if they had
used XDR, or ASN.1 with PER/OER/..., you could in fact automatically
check the validity of the encoding of a message.

A protocol using ASN.1 BER/DER/CER with IMPLICT tagging will lose some
type information, so while you can check a lot of an encoded message's
validity without reference to its schema, you cannot check all of it.
Whereas if you have reference to its schema, then you can check all of
it regardless of whether the encoding rules are TLV or not.

Nico
--