Getting the right displayed number of articles

Getting the right displayed number of articles

[Harry Putnam]

Having moved some 900 messages out of a ding-list group, leaving some
300 in there.  The number of articles displayed is still 1200
somehting.

I tried the trick posted here a few times where you move the remaining
messages to the same group ie in ding-list 'B m' 300 articles to
'ding-list', but still the displayed number is 1200+. Even after a
full restart.

-- 

Harry Putnam  reader@newsguy.com

Re: Getting the right displayed number of articles

[Francisco Solsona]

Harry Putnam <reader@newsguy.com> writes:

[...]

> messages to the same group ie in ding-list 'B m' 300 articles to
> 'ding-list', but still the displayed number is 1200+. Even after a
> full restart.

	You must've forgotten at least one of the small-numbered
articles. For instance, I have this "auctex" group with only four
articles in it, so I go:

~/Mail % ls auctex
1   12  37  61

and then I do (group buffer), 'B m' (or 'B r') on those four articles
and I get:

~/Mail % ls auctex
62  63  64  65 

and Gnus now thinks that there are only 4 messages in the group, which 
is correct, instead of (1+ (- 61 1)) ==> 61. Try entering the group
with 'C-u RET' so you don't miss any of the articles, mark them all
and do the 'B m' thing. No need to restart Gnus, just re-entering the
group works for me.

	I just did this using Gnus v5.6.33, because the patch to
make it v5.6.34 told me:

solsona@deprof ...lisp/gnus/gnus-beta % patch -p 2 < gnus-5.6.33-5.6.34.diff.gz
patching file `lisp/gnus-msg.el'
patching file `lisp/gnus-score.el'
patching file `lisp/gnus.el'
patching file `lisp/ChangeLog'
patching file `Makefile.in'
patching file `Makefile.in'
Hunk #1 FAILED at 16.
Hunk #2 FAILED at 74.
Hunk #3 FAILED at 90.
3 out of 3 hunks FAILED -- saving rejects to Makefile.in.rej
patching file `texi/gnus.texi' 

:-(... downloading the whole tarball. HTH, though.
Francisco
-- 
A thing I learned from videogames:
There is no problem that cannot be overcome by violence.

Re: Getting the right displayed number of articles

[Harry Putnam]

Francisco Solsona <solsona@deprof.fciencias.unam.mx> writes:

> > full restart.
> 
> 	You must've forgotten at least one of the small-numbered
> articles. For instance, I have this "auctex" group with only four
> articles in it, so I go:
> 
> ~/Mail % ls auctex
> 1   12  37  61
> 
> and then I do (group buffer), 'B m' (or 'B r') on those four articles
> and I get:
> 
> ~/Mail % ls auctex
> 62  63  64  65 
> 
> and Gnus now thinks that there are only 4 messages in the group, which 
> is correct, instead of (1+ (- 61 1)) ==> 61. Try entering the group
> with 'C-u RET' so you don't miss any of the articles, mark them all
> and do the 'B m' thing. No need to restart Gnus, just re-entering the
> group works for me.

Afraid that isn't doing it here.  The first time around I did use 'C-u'
to enter.  But have done it again now -- used this sequence:

1) C-u enter wrongly shows 1231, mark all with #  (214 articles)
2) 'B-m' move to same group (ding-list)
3) 'ZZ' to exit, C-u to re-enter now shows 1438, there are actually
   214

Tried the same sequence again now it shows 1645 -- I'm either getting
a lot of mail awful quick or something is wacky.

Did a full restart -- no change

-- 

Harry Putnam  reader@newsguy.com

Re: Getting the right displayed number of articles

[Eze Ogwuma]

Harry Putnam <reader@newsguy.com> writes:

[...]

> Afraid that isn't doing it here.  The first time around I did use 'C-u'
> to enter.  But have done it again now -- used this sequence:
> 
> 1) C-u enter wrongly shows 1231, mark all with #  (214 articles)
> 2) 'B-m' move to same group (ding-list)
> 3) 'ZZ' to exit, C-u to re-enter now shows 1438, there are actually
>    214
> 
> Tried the same sequence again now it shows 1645 -- I'm either getting
> a lot of mail awful quick or something is wacky.
> 
> Did a full restart -- no change

Possibly you have some duplicates which were suppressed until you moved
the first lot else where.

One other thing I've found is that if I type C-u 100 RET to enter a
group from which I've moved articles I only see 30 or so
articles. Presumably these articles numbers cover the last 100
messages. 

-- 
Eze Ogwuma

Re: Getting the right displayed number of articles

[Harry Putnam]

Eze Ogwuma <typhoon@dircon.co.uk> writes:

> Harry Putnam <reader@newsguy.com> writes:
> 
> [...]
> 
> > Afraid that isn't doing it here.  The first time around I did use 'C-u'
> > to enter.  But have done it again now -- used this sequence:
> > 
> > 1) C-u enter wrongly shows 1231, mark all with #  (214 articles)
> > 2) 'B-m' move to same group (ding-list)
> > 3) 'ZZ' to exit, C-u to re-enter now shows 1438, there are actually
> >    214
> > 
> > Tried the same sequence again now it shows 1645 -- I'm either getting
> > a lot of mail awful quick or something is wacky.
> > 
> > Did a full restart -- no change
> 
> Possibly you have some duplicates which were suppressed until you moved
> the first lot else where.

Well I don't have duplicate suppressing on so when I have dups they
have the 'Gnus-Warning: ...' on them and I run M-& and give it the
warning string. and delete any that show up.  Nice try though.

Everytime I try to redo the numbering with moving technique, the
number displayed increases by the 200+ that are still there.

Also a quick line count of the files in ~/Mail/ding-mail/* from
'dired' shows 217 files.  The displayed count is now up to 1651.

> 
> One other thing I've found is that if I type C-u 100 RET to enter a
> group from which I've moved articles I only see 30 or so
> articles. Presumably these articles numbers cover the last 100
> messages. 

Yes, I'm not sure how that works either but it can be somewhat
confusing. Especially if you try from inside with 100 'Z R'.

I've used the 'moving to same group' technique before on other groups
successfully so having a bit of trouble seeing what is different here.

-- 

Harry Putnam  reader@newsguy.com

Re: Getting the right displayed number of articles

[Eze Ogwuma]

Harry Putnam <reader@newsguy.com> writes:

> Eze Ogwuma <typhoon@dircon.co.uk> writes:

[...]

> Everytime I try to redo the numbering with moving technique, the
> number displayed increases by the 200+ that are still there.
> 
> Also a quick line count of the files in ~/Mail/ding-mail/* from
> 'dired' shows 217 files.  The displayed count is now up to 1651.

I've seen the same thing too. I gave up when the article numbers
climbed to above 2000.

[...]

> I've used the 'moving to same group' technique before on other groups
> successfully so having a bit of trouble seeing what is different here.

As have I. At some point along the way it seems that something's
changed. Perhaps now's the time to implement renumbering.

-- 
Eze Ogwuma

Re: Getting the right displayed number of articles

[Harry Putnam]

Eze Ogwuma <typhoon@dircon.co.uk> writes:

> Harry Putnam <reader@newsguy.com> writes:
> 
> > Eze Ogwuma <typhoon@dircon.co.uk> writes:
> 
> [...]
> 
> > Everytime I try to redo the numbering with moving technique, the
> > number displayed increases by the 200+ that are still there.
> > 
> > Also a quick line count of the files in ~/Mail/ding-mail/* from
> > 'dired' shows 217 files.  The displayed count is now up to 1651.
> 
> I've seen the same thing too. I gave up when the article numbers
> climbed to above 2000.
> 
> [...]
> 
> > I've used the 'moving to same group' technique before on other groups
> > successfully so having a bit of trouble seeing what is different here.
> 
> As have I. At some point along the way it seems that something's
> changed. Perhaps now's the time to implement renumbering.

I just went ahead and did it the long way round by copying to a temp
group , then 'C-u G del'  Remake group  and copy back.  The numbers are now
right but it seems a bit much.

-- 

Harry Putnam  reader@newsguy.com

Re: Getting the right displayed number of articles

[Eze Ogwuma]

Harry Putnam <reader@newsguy.com> writes:

[...]

> I just went ahead and did it the long way round by copying to a temp
> group , then 'C-u G del'  Remake group  and copy back.  The numbers are now
> right but it seems a bit much.

Do you know what happens to marked and cached articles?

-- 
Eze Ogwuma

Re: Getting the right displayed number of articles

[Harry Putnam]

Eze Ogwuma <typhoon@dircon.co.uk> writes:

> Harry Putnam <reader@newsguy.com> writes:
> 
> [...]
> 
> > I just went ahead and did it the long way round by copying to a temp
> > group , then 'C-u G del'  Remake group  and copy back.  The numbers are now
> > right but it seems a bit much.
> 
> Do you know what happens to marked and cached articles?

Not sure what your asking here Eze.  Do you mean concerning the
numbers they have?  Or do you mean during copying?  Its been my
experience that marks are preserved when copying takes place.


-- 

Harry Putnam  reader@newsguy.com

Re: Getting the right displayed number of articles

[Eze Ogwuma]

Harry Putnam <reader@newsguy.com> writes:

> Eze Ogwuma <typhoon@dircon.co.uk> writes:
> 
> > Harry Putnam <reader@newsguy.com> writes:
> > 
> > [...]
> > 
> > > I just went ahead and did it the long way round by copying to a temp
> > > group , then 'C-u G del'  Remake group  and copy back.  The numbers are now
> > > right but it seems a bit much.
> > 
> > Do you know what happens to marked and cached articles?
> 
> Not sure what your asking here Eze.  Do you mean concerning the
> numbers they have?  Or do you mean during copying?  Its been my
> experience that marks are preserved when copying takes place.

I was asking whether cached articles (*) are preserved and whether
marks are preserved.

-- 
Eze Ogwuma

Re: Getting the right displayed number of articles

[Kai Grossjohann]

>>>>> Harry Putnam <reader@newsguy.com> writes:

  > Also a quick line count of the files in ~/Mail/ding-mail/* from
  > 'dired' shows 217 files.  The displayed count is now up to 1651.

You should have a look at the actual listing.  Maybe for some reason
or other there's a file around with a small number?

Hm.  And then there's the overview information which might have gotten
hosed.  And the active file.  Hm.

kai
-- 
OOP: object oriented programming;  OOPS: object oriented mistakes

Re: Getting the right displayed number of articles

[Harry Putnam]

Eze Ogwuma <typhoon@dircon.co.uk> writes:

> Harry Putnam <reader@newsguy.com> writes:
> 
> > Eze Ogwuma <typhoon@dircon.co.uk> writes:
> > 
> > > Harry Putnam <reader@newsguy.com> writes:
> > > 
> > > [...]
> > > 
> > > > I just went ahead and did it the long way round by copying to a temp
> > > > group , then 'C-u G del'  Remake group  and copy back.  The numbers are now
> > > > right but it seems a bit much.
> > > 
> > > Do you know what happens to marked and cached articles?
> > 
> > Not sure what your asking here Eze.  Do you mean concerning the
> > numbers they have?  Or do you mean during copying?  Its been my
> > experience that marks are preserved when copying takes place.
> 
> I was asking whether cached articles (*) are preserved and whether
> marks are preserved.

I usually run with caching turned off for nnml groups:
(setq gnus-uncacheable-groups "^nnml")
I turned that on and experimented a little.

Copying  (B c) seems to preserve read/unread, ticked etc.  But not
cache (*)
Moving (B m) exibits the same behavior.


-- 

Harry Putnam  reader@newsguy.com

Re: Getting the right displayed number of articles

[Justin Sheehy]

Harry Putnam <reader@newsguy.com> writes:

> Having moved some 900 messages out of a ding-list group, leaving some
> 300 in there.  The number of articles displayed is still 1200
> somehting.

Hm.  The first thing to note is that the number in the *Group* buffer
is not the number of articles in the group.  It is the numeric
difference between the highest and lowest numbered articles in the
group.  Iff all of your articles are contiguously numbered, then this
is the "number of articles".  Otherwise, it isn't

> Tried the same sequence again now it shows 1645 -- I'm either getting
> a lot of mail awful quick or something is wacky.

Nothing is wacky.  You have a low-numbered article that isn't being
moved, for whatever reason.

> Everytime I try to redo the numbering with moving technique, the
> number displayed increases by the 200+ that are still there.

Yep.

> Also a quick line count of the files in ~/Mail/ding-mail/* from
> 'dired' shows 217 files.  The displayed count is now up to 1651.

Right, but what are the lowest and highest-numbered files?

> I've used the 'moving to same group' technique before on other groups
> successfully so having a bit of trouble seeing what is different here.

Looking at the files in the directory (and the active file) will tell
you for certain.

-- 
Justin Sheehy

In a cloud bones of steel.
  

Re: Getting the right displayed number of articles

[Eze Ogwuma]

Harry Putnam <reader@newsguy.com> writes:

> Eze Ogwuma <typhoon@dircon.co.uk> writes:

[...]

> > I was asking whether cached articles (*) are preserved and whether
> > marks are preserved.
> 
> I usually run with caching turned off for nnml groups:
> (setq gnus-uncacheable-groups "^nnml")
> I turned that on and experimented a little.
> 
> Copying  (B c) seems to preserve read/unread, ticked etc.  But not
> cache (*)
> Moving (B m) exibits the same behavior.

Thanks.
-- 
Eze Ogwuma

Re: Getting the right displayed number of articles

[Harry Putnam]

Justin Sheehy <justin@linus.mitre.org> writes:


> 
> Right, but what are the lowest and highest-numbered files?
> 
> > I've used the 'moving to same group' technique before on other groups
> > successfully so having a bit of trouble seeing what is different here.
> 
> Looking at the files in the directory (and the active file) will tell
> you for certain.

To late to tell on this go around since I've 'B m' ed to a temp group 
ran 'C-u G del' then remade the group and moved the messages back.
They all have orderly numbers now.  Except for some kind of system
where every 12th file is a different sequence number.

Looking closely in the directories involved I was suprised to find the
group that had been nuked with 'C-u G del' to still be there and still
have one lonesome article in it (# 1)  The culprit I guess.  it was a
recent article though -- part of this thread.

If it comes up again though I'll know how to proceed and what is
actually happening -- Thanks

-- 

Harry Putnam  reader@newsguy.com