nnimap-split-fancy broken

nnimap-split-fancy broken

[Niklas Morberg]

There's something weird with nnimap-split-fancy, possibly
due to the recent introduction of nnimap-split-download-body.

The problem is that `to' now catches occurences in the body
as well. Here's an example:

(setq nnimap-split-rule 'nnimap-split-fancy
	nnimap-split-fancy
	'(|
	  (to "ding" "lists/ding")))

If a message is sent directly to me and contains `To: ding'
in the body, the message will be split to lists/ding. This
happens quite often since Outlook includes the original
message with some headers when replying.

nnimap-split-download-body's value is default

Niklas

Re: nnimap-split-fancy broken

[Simon Josefsson]

Niklas Morberg <niklas.morberg@axis.com> writes:

> There's something weird with nnimap-split-fancy, possibly
> due to the recent introduction of nnimap-split-download-body.
>
> The problem is that `to' now catches occurences in the body
> as well. Here's an example:
>
> (setq nnimap-split-rule 'nnimap-split-fancy
> 	nnimap-split-fancy
> 	'(|
> 	  (to "ding" "lists/ding")))
>
> If a message is sent directly to me and contains `To: ding'
> in the body, the message will be split to lists/ding. This
> happens quite often since Outlook includes the original
> message with some headers when replying.

Hm, the code tries to narrow to headers.  Could you M-x debug-on-entry
RET nnimap-split-fancy RET and look in the nnimap server buffer ("
*nnimap* SERVER") and check whether it is narrowed to the headers?  If
so, try widening to see if it works.

> nnimap-split-download-body's value is default

I assume n-s-d-b-d's value was set to non-nil by spam.el.

Re: nnimap-split-fancy broken

[Niklas Morberg]

Simon Josefsson <jas@extundo.com> writes:

> Hm, the code tries to narrow to headers. Could you M-x
> debug-on-entry RET nnimap-split-fancy RET and look in the
> nnimap server buffer (" *nnimap* SERVER") and check whether
> it is narrowed to the headers?

That buffer seems to be empty here (it does exist, though)...

> If so, try widening to see if it works.

Is `C-x n w' what I should do? Doing it in the empty buffer
reveals nothing.

Niklas