[NNTP] LISTGROUP

[Matthias Andree]

"Charles Lindsey" <chl at clerew.man.ac.uk> writes:

> In <87k6nzkz0p.fsf at windlord.stanford.edu> Russ Allbery <rra at stanford.edu> writes:
>
>>Second, LISTGROUP would be much more useful for some purposes if it took a
>>range as an optional second argument.  It's pretty trivial to specify that
>>argument:  it would be identical to the ranges that OVER and HDR take, and
>>LISTGROUP would function exactly as before but would limit the article
>>numbers returned to articles within that range.  If there are no articles
>>in the group within that range, an empty multiline response would be
>>returned but everything else about LISTGROUP would remain unchanged.
>
> I think we are a bit late for changing whether LISTGROUP is optional or
> not, but I like the idea of a range parameter. Is that currently implemented
> anywhere to anyone's knowledge?

Will the current article number (pointer) be set to the first article in
the group, as GROUP demands, or will it be set to the first article in
the range? I'm inclined to suggest the latter.

After that question is answered, the implementation will take fifteen
minutes in leafnode-2 as proof-of-concept and some minutes to document
it.

Note that the new syntax should contain the same prefix as RFC-2980 laid
out, so if a range will be given, the group parameter must be made
mandatory.

I suggest that the standard DO NOT define a special "current group"
pointer, as forking code into regular and special case increases
implementation effort and doubles the testing effort. Keep it simple.
If the newsreader cannot remember the most recently selected newsgroup,
it's broken, the standard should not take an interest in this SOEP(*).

BTW, another round of discussion about such features doesn't make a
difference. What are a few weeks versus two decades? NNTPv2 needs to
last another 19 years... so why not do it right? :-)


(*) SOEP = some-one else's problem

-- 
Matthias Andree