ietf-nntp Zero-argument ARTICLE and missing articles

[Russ Allbery]

What is the behavior of the zero-argument form of ARTICLE (and HEAD, BODY,
and STAT for that matter) supposed to be if the current article pointer is
pointing to an article that doesn't exist (such as an article that's been
cancelled since the group was entered)?

The current draft says:

    In the second form, an article number may be specified.  If so, and if
    there is an article with that number in the currently selected group,
    the server MUST set the current article pointer to that number.

    Then, whether or not a number was specified, the article indicated by
    the current article pointer is presented to the client.

and then later says:

    The server MUST NOT change the currently selected group as a result of
    this command.  The server MUST NOT change the current selected article
    except when an article number argument was provided and the article
    exists; in particular, it MUST NOT change it following an unsuccessful
    response.

This seems to indicate that the correct thing to do in this case is to
return a 423 error response.  This isn't, however, what INN does; since
version 1.0, INN has always instead started to advance the current article
pointer if the current article didn't exist, continuing until it found a
valid article (which it would then return) or until the current article
pointer reached the high water mark.

The behavior specified in the draft seems to make slightly more sense in
combination with NEXT, since should one advance the article pointer with
NEXT and then have the article be cancelled before it is retrieved with
ARTICLE, INN's behavior would retrieve some other article instead (with a
different message ID than what NEXT returned), whereas the behavior in the
draft would return 423.  I wanted to double-check before just deciding to
change this in INN, though, because this *has* been the behavior of INN
for nearly a decade and I wasn't sure people realized this would be a
change.

I don't know if any reader could be relying on INN's behavior and be
unable to handle a 423 response to a zero-argument ARTICLE command.

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