Re: Re-sorting *lots* of old mail

[Peter Davis <pd@world.std.com>]

Benjamin Rutt <rutt+news@cis.ohio-state.edu> writes:

> "Peter Davis" <pdavis@pageflexinc.com> writes:
> 
> > "Kai Großjohann" <Kai.Grossjohann@CS.Uni-Dortmund.DE> wrote in message
> > news:vafu1lgpqol.fsf@lucy.cs.uni-dortmund.de...
> >> "Peter Davis" <pdavis@pageflexinc.com> writes:
> >>
> >> > I have one nnml group with about 30,000 messages in it.  I'd like to
> > split
> >> > it by year into different groups (e.g., nnml:stuff1998, nnml:stuff1999,
> >> > etc.).  Is there some way to do that which:
> >> >
> >> > 1) I can start running and leave overnight or something?  Unless it's
> > *very*
> >> > fast, it will take a long time to process all these messages.
> >> >
> >> > 2) Will keep the the .overview file(s) correct in all groups?
> >>
> >> If you can set up split rules that work, then it should work to just
> >> `B r' all them messages.
> >
> > Is there a way to 'B r' them from the *Group* buffer?  I can't open the
> > actual group and all the messages, as it would take hours, and probably
> > exhaust memory, to build a summary.
> 
> Then why not just enter with C-u 100?  And then process 100 respools
> at a time.  You could write a keyboard macro which presses something
> like 'C-u 100 M P a B r RET' and then presses 'q' to go back to the
> group buffer.  Then, execute the keyboard macro 300 times, to arrive
> at 30,000.  It will probably still take hours, but at least it
> probably won't exhaust memory.

Not only will it take hours, but it will take hours of human
involvement, as opposed to something I can just start running and
leave it overnight.

No better solutions around?

Thanks,

-pd


-- 
--------
                             Peter Davis
               Funny stuff at http://www.pfdstudio.com
    List of resources for children's writers and illustrators at:
                  http://www.pfdstudio.com/cwrl.html