[NNTP] draft-ietf-nntpext-streaming-02

[Russ Allbery]

Ken Murchison <ken at oceana.com> writes:
> Clive D.W. Feather wrote:

>>> As I see it, the 432 response to TAKETHIS has exactly the same meaning
>>> as the 436 response to IHAVE. We explicitly list 436 as being possible
>>> after the article body has been sent. So when do existing servers
>>> generate this response? If they do, why shouldn't they generate 432 in
>>> the same situation?

>> Oh, and what about 441 to POST. Should this be replaced by 400 as well?

> I am also confused.  I think we need Andrew to help us (or at least me)
> understand this how/why TAKETHIS is vastly different from IHAVE/POST.

I don't understand why people are confused.  441 is a rejection, not a
deferral.  All of IHAVE, TAKETHIS, and POST have return codes they
generate after the body to reject the message (namely 437, 439, and 441
respectively).  IHAVE, alone among these three, also has a deferral code
at present because some servers do this.  Andrew's point is that this
should be removed, and I'm inclined to agree after reading his messages.

There is, distinct from this, a *different* set of codes that are used as
responses to CHECK and *initial* responses to IHAVE, but that's a
completely different situation.

BTW, I just noted in streaming-01 that the 439 TAKETHIS code is described
as "Transfer failed" in the Syntax summary.  I don't know if this has
already been fixed or not (I'm grabbing the latest version but haven't
gotten that far in my catching up yet), but this is misleading.  A better
description would be "Article rejected".

-- 
Russ Allbery (rra at stanford.edu)             <http://www.eyrie.org/~eagle/>