Re: Accessing all incoming mails, checking for network connection, changes in message status

[Ted Zlatanov <tzz@lifelogs.com>]

On Tue, 27 Jul 2010 14:47:05 +0200 Manuel Carro <mcarro@fi.upm.es> wrote: 

>> Virtual groups should do this.  Otherwise you're reinventing the same
>> functionality.  So if the only problem is that the GCC field is wrong,
>> let's fix that.  Or is there another reason why virtual groups are not
>> what you want?

MC> I have a virtual group which puts together all mail, so that when I
MC> enter the group I see all unread and ticked articles.  But these
MC> articles have already been spooled to the corresponding groups.  What I
MC> would like is to receive all mail in a single group (that is easy) and,
MC> as I read the messages, respool them to other groups using a set of
MC> rules and "B t" (basically the ones I have now).  It is the second part I did not
MC> manage to find how to do, as it seems that at the moment I set up
MC> splitting rules the mail start to be split.

Yeah, splitting rules are not meant to be optional.  But they can call
ELisp!  So make a function that normally returns the "main" group, but
make it operate differently (returning nil) when invoked interactively.

(defvar pass-through nil)
(defun my-special-splitter ()
  (if pass-through
      nil
    "main group name here"))

Now you just have to a) insert this function in your split rules before
anything else, and b) advise or override `B t' inside a lambda that sets
pass-through to t before the call and sets it back to nil afterwards.

Look in the splitting rules section of the manual for more details.
It's not too bad.

>> This indicates article marks are inconsistent.  Is this over IMAP or
>> another protocol?  Could the server be modifying article numbers behind
>> Gnus' back?

MC> It is over IMAP, but use IMAP to copy the mail messages over to my
MC> machine and I the processing is done locally after my mail has been
MC> downloaded.  I do not think the article numbers are being modified,
MC> then.

OK.  I'm not sure what's going on then.

Ted