Re: Merging many different MH and mbox mail archives

Re: Merging many different MH and mbox mail archives

[Lars Magne Ingebrigtsen]

Peter Davis <pd@world.std.com> writes:

> 1) I'm assuming I would have to open each existing mbox of MH folder
>    and respool all the messages to get them all filtered according to
>    my split-fancy rules.  Is there any way to automate that, so I can
>    just point to a top level directory and go away for a few hours
>    (days?), and have all the mail in that tree get re-spooled?

Hm...  I don't think so.  Perhaps the, er, easiest way would be to
write a shell script that copies all the files from the entire
hierarchy into one biiiig directory, where you name the files with
numerical names that increase with the date.  (That is, sort the
files by date first.)  And then open that directory as an nnmh group,
and then respool.

> 2) When I respool messages, will messages with duplicate message IDs
>    get eliminated?  I would be respooling into nnml groups.  Will new
>    messages being split into a group be checked for matching ID
>    against messages already in that group?

You can set `nnmail-treat-duplicates' to control that.  You should
also probably increase `nnmail-message-id-cache-length'.

Hm...  but when respooling, those variables aren't really heeded.

Ok, scratch that.  Instead of respooling, use something like this:

(setq mail-sources
      '((directory :path "/home/pd/that-directory-where-all-the-messages-are/"
                   :suffix "")))

and just press `g' to fetch the mail as if it were new, which will
lead to it getting run through the splitting methods, with duplicate
suppression, etc.

> 3) For some high-volume groups, I'd like to automatically split into
>    different sub-groups by year ... nnml:this.that.1979,
>    nnml:this.that.1980, etc.  Is there a way to set that up, and to
>    have gnus create the groups automatically as needed?

There is no pre-defined functionality in Gnus for doing this.

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

Re: Merging many different MH and mbox mail archives

[Peter Davis]

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

> Peter Davis <pd@world.std.com> writes:
> 
> > 1) I'm assuming I would have to open each existing mbox of MH folder
> >    and respool all the messages to get them all filtered according to
> >    my split-fancy rules.  Is there any way to automate that, so I can
> >    just point to a top level directory and go away for a few hours
> >    (days?), and have all the mail in that tree get re-spooled?
> 
> Hm...  I don't think so.  Perhaps the, er, easiest way would be to
> write a shell script that copies all the files from the entire
> hierarchy into one biiiig directory, where you name the files with
> numerical names that increase with the date.  (That is, sort the
> files by date first.)  And then open that directory as an nnmh group,
> and then respool.

Hmm.  Interesting possibility.  The regular MH 'refile' command should
work for moving the messages, keeping numbers in sequence, etc.  I'd
really like to get the message numbers sorted by message date, but I
don't think refile can do that.

> > 2) When I respool messages, will messages with duplicate message IDs
> >    get eliminated?  I would be respooling into nnml groups.  Will new
> >    messages being split into a group be checked for matching ID
> >    against messages already in that group?
> 
> You can set `nnmail-treat-duplicates' to control that.  You should
> also probably increase `nnmail-message-id-cache-length'.
> 
> Hm...  but when respooling, those variables aren't really heeded.
> 
> Ok, scratch that.  Instead of respooling, use something like this:
> 
> (setq mail-sources
>       '((directory :path "/home/pd/that-directory-where-all-the-messages-are/"
>                    :suffix "")))
> 
> and just press `g' to fetch the mail as if it were new, which will
> lead to it getting run through the splitting methods, with duplicate
> suppression, etc.

Interesting idea.  Thanks!

> > 3) For some high-volume groups, I'd like to automatically split into
> >    different sub-groups by year ... nnml:this.that.1979,
> >    nnml:this.that.1980, etc.  Is there a way to set that up, and to
> >    have gnus create the groups automatically as needed?
> 
> There is no pre-defined functionality in Gnus for doing this.

I didn't think so.  Anyone have some lisp code that will do this?

Thanks very much!

-pd


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