New groups placing

New groups placing

[Toni Drabik]

Is there a way to control in which topic will newly created groups be
placed?

I don't mean groups created with `G m' (they will end up in a topic under
the point, I believe), but those which are a product of
`gnus-message-archive-group', `B m'-ing articles to nonexistent group(s),
and similar procedures.

It would also be nice if user could somehow say which newly created groups
should be subscribed, and which should not.

-- 
Toni Drabik <tdrabik@public.srce.hr>
Warning: This article may be fatal if swallowed.

Re: New groups placing

[Lars Magne Ingebrigtsen]

Toni Drabik <tdrabik@public.srce.hr> writes:

> I don't mean groups created with `G m' (they will end up in a topic under
> the point, I believe), but those which are a product of
> `gnus-message-archive-group', `B m'-ing articles to nonexistent group(s),
> and similar procedures.

And since the new nnslashdot backend generates groups like crazy, I've 
added a `gnus-subscribe-topic' function that you can set
`gnus-subscribe-newsgroup-method' to.  See the manual in the CVS for
details. 

> It would also be nice if user could somehow say which newly created groups
> should be subscribed, and which should not.

See `gnus-subscribe-interactively'.

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: New groups placing

[Jan Vroonhof]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> And since the new nnslashdot backend generates groups like crazy,

Shouldn't it really create just one group and let each item appear as
threads.  (Or multiple groups based on the subject group this article
is in). i.e. features appear in nnslashot:features, there is an
nnslashdot:ask-slashot, nnslashot:debian, etc.

Jan

Re: New groups placing

[Lars Magne Ingebrigtsen]

Jan Vroonhof <vroonhof@math.ethz.ch> writes:

> Shouldn't it really create just one group and let each item appear as
> threads.

That would require doing something new and complex.

For instance, how would it keep track of which comments you've read,
and tick marks?  And how would an Agentized version of that work?  And
stuff?

Doing it in the traditional Gnus framework means that all those
newsreaderly things Just Work like they should.  

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: New groups placing

[Hrvoje Niksic]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Jan Vroonhof <vroonhof@math.ethz.ch> writes:
> 
> > Shouldn't it really create just one group and let each item appear as
> > threads.
> 
> That would require doing something new and complex.
> 
> For instance, how would it keep track of which comments you've read,
> and tick marks?

Uh?  Like, the same way it's done for NNTP (.newsrc.eld)?  What am I missing?

> Doing it in the traditional Gnus framework means that all those
> newsreaderly things Just Work like they should.

Creating new groups out of the blue doesn't look very newsreaderly to
me.  Easier to implement it may well be, but to me it seems very
weird.

Re: New groups placing

[Per Abrahamsen]

Hrvoje Niksic <hniksic@iskon.hr> writes:

> Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> 
> > For instance, how would it keep track of which comments you've read,
> > and tick marks?
> 
> Uh?  Like, the same way it's done for NNTP (.newsrc.eld)?  What am I missing?

./ comments have unique number, but only within each article.
I.e. the "first post" for each article is numbered 1 (and usually has
the interesting text "first post"), the second comment is numbered 2,
etc.

Re: New groups placing

[Kai Großjohann]

Per Abrahamsen <abraham@dina.kvl.dk> writes:

> ./ comments have unique number, but only within each article.
> I.e. the "first post" for each article is numbered 1 (and usually has
> the interesting text "first post"), the second comment is numbered 2,
> etc.

Wibni Gnus grokked a non-integer article `number'?  Then nn/. could
return (cons article-id comment-number)...

kai
-- 
This gubblick contains many nonsklarkish English flutzpahs,
but the overall pluggandisp can be glorked from context. -- David Moser

Re: New groups placing

[Lars Magne Ingebrigtsen]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> Wibni Gnus grokked a non-integer article `number'?  Then nn/. could
> return (cons article-id comment-number)...

Gnus is rather attached to the article number paradigm.  :-)

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen

Re: New groups placing

[Kai Großjohann]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:
> 
> > Wibni Gnus grokked a non-integer article `number'?  Then nn/. could
> > return (cons article-id comment-number)...
> 
> Gnus is rather attached to the article number paradigm.  :-)

Well, yes.  But what does it do with article numbers?  WIBNI there was
an abstract data type with operations = and <, say?  (Are any more ops
needed?  Maybe - for estimating the number of articles?)  Then, you
could change all the existing code to invoke the ADT ops and then
changing from numbers to conses would be a piece of cake...

Sounds like boring work, though.

kai
-- 
This gubblick contains many nonsklarkish English flutzpahs,
but the overall pluggandisp can be glorked from context. -- David Moser

Re: New groups placing

[Lars Magne Ingebrigtsen]

Kai.Grossjohann@CS.Uni-Dortmund.DE (Kai Großjohann) writes:

> Sounds like boring work, though.

And lots of it.  :-)

-- 
(domestic pets only, the antidote for overdose, milk.)
   larsi@gnus.org * Lars Magne Ingebrigtsen