Re: New Version Notification for draft-pardue-http-identity-digest-01.txt

Hi @all,

first of all, thanks Lucas for drafting this out, and thanks Roy for
your feedback!
I just wanted to add a couple of bits to Lucas' reply.

On Wed, 8 Mar 2023 at 19:28, Lucas Pardue <lucaspardue.24.7@gmail.com> wrote:
> On Wed, Mar 8, 2023 at 5:39 PM Roy T. Fielding <fielding@gbiv.com> wrote:
>> Multiple digests should not be sent in the same message unless their
>> digests are different.

Different digests could be processed at
different levels (eg, one by a proxy, another by an application, ..).
Do you think there are specific issues in having
different digest fields, with different semantics (e.g.
Content-Digest, Repr-Digest, Identity-Digest)
conveying the same value in specific cases? (see Lucas' example below,
where for an unencoded content, C-Digest == I-Digest)

>> Transforming the message content to remove the
>> outer coding would necessitate replacing the field values as well,
>> [..] in my
>> experience the readers tend to implement the examples in practice
>> even when they are nonsensical and a waste of network bandwidth.

On this, I think that in Digest fields
https://httpwg.org/http-extensions/draft-ietf-httpbis-digest-headers.html#name-end-to-end-integrity
we warn intermediaries about the effects of mangling Digest fields.
Clearly there's always the risk
of "creative" usage of the specifications.
Tuning up examples that might suggest this behavior in the Id-Digest
is ok, though.

>> Also, you have not considered multiple content codings yet, [,,]
>> I would expect, in that case, for both encodings to be decoded before the
>> identity-digest is applicable. YMMV.
Yes, the idea is that the Id-Digest can be applied to the fully
decoded content stored on a data store (e.g. as a file).

Thanks for all your time and have a nice day,
R

Received on Tuesday, 14 March 2023 01:03:42 UTC