On 20 Jul 2001, Martin Gasbichler wrote:
>>>>>> "Kai" == Kai Gro�johann <[EMAIL PROTECTED]>
>>>>>> writes:
>
> Kai> On 19 Jul 2001, Martin Gasbichler wrote:
>>> Yes, and this is bad. nnmail-split-fancy-with-parent should always
>>> have highest precedence (as I assume that the parent was split
>>> correctly).
>
> Kai> Yes. Therefore you need something which is similar to
> Kai> bbdb/gnus-split-method, but returns nil when nomatch, rather
> Kai> than calling a specific function for that.
>
> Kai> But maybe you can set bbdb/gnus-split-nomatch-function to a
> Kai> function which returns nil. Yeah, that might work. `ignore'
> Kai> is such a function.
>
> But the nomatch case it not the problem: I want to take the group
> suggested by nnmail-split-fancy-with-parent even if
> bbdb/gnus-split-method would say something different.
I know. So that's why you want to put (:
nnmail-split-fancy-with-parent) first in nnmail-split-fancy, followed
by (: bbdb/gnus-split-method).
> So, I have to call nnmail-split-fancy-with-parent first. This can
> only happen within nnmail-split-fancy. However, calling
> bbdb/gnus-split-method from nnmail-split-fancy yields
> bbdb/gnus-default/nomatch-function in any case and I am lost.
Well, bbdb/gnus-split-method calls the value of
bbdb/gnus-split-nomatch-function, right? And if that function is
ignore, then bbdb/gnus-split-method will return nil. And that's
exactly what nnmail-split-fancy expects.
Please try
(setq nnmail-split-fancy
'(| (: nnmail-split-fancy-with-parent)
(: bbdb/gnus-split-method)
...))
(setq bbdb/gnus-split-nomatch-function 'ignore)
What happens when you do that? It might even work.
> I would like to know, if there is anybody out there who successfully
> calls bbdb/gnus-split-method from nnmail-split-fancy?
Dunno.
kai
--
~/.signature: No such file or directory
_______________________________________________
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/bbdb-info
BBDB Home Page: http://bbdb.sourceforge.net/
