On Friday, September 10, 2004, 14:55, David Elliott wrote:

>> This is by design, see <mid:[EMAIL PROTECTED]>.

> This is one of the massages that I did not have.

I'll PM it to you.

>> Since you have "Continue processing with other filters" checked, the
>> messages are processed with all filters and the last "Move" action is
>> the one deciding where the messages ends up.

> This logic is wrong.

> This means that you have to build the filters the wrong way round.
> i.e. the ones at the top of the list are the last ones that you want
> to use.

Judging from this email and
<mid:[EMAIL PROTECTED]> you have built your
filters in a way they are not intended to be built.

"Continue processing with other filters" decides whether the message
should be carried on to other filters at _the same filter level_.
Unticking this option in a subfilter only means no further subfilters of
that particular parent filter should be processed. Whether the
subsequent parent filters are processed or not is decided by whether
"Continue processing with other filters" is ticked or not in the last
matching filters _at the same filter level_.

> Logic to me states the following.

> Start filtering
> Is message caught by Filter?
> No ..:- Carry on to next filter
> Yes .:- Carry out action as described by filter ?
>         Is there any sub - filters ?
>         No ..:- Stop (unless there is a continue processing flag)
>         Yes .:- Is message caught by Sub filter ?
>                 No ..:- Carry on to next sub filter
>                 Yes .:- Carry out action of sub filter

This is how it does work, with one exception. A subfilter never stops
the processing of subsequent parent filters and it shouldn't. See the
138 messages long discussion named "NFS: Big question for 9Val", I'm
sure you can find the messages you have lost in the archives.

> So what was happening before is that on the TBBeta filter it was
> copying (not moving) and the message was still present in the in-box
> when the Marck filter was hit. So a fresh copy was made into the Marck
> folder and when the filtering was finished the message was deleted.

This was a bug, confirmed and fixed by 9val. The NFS should not loose
track of a message just because it has been moved (and doesn't anymore,
which you have painfully noticed :-) and subsequent move actions moves
the message again.

> 1) When a filter has been activated it should not pass control to
> another filter if the 'Continue processing' option is not checked.

Correct. This is how the NFS is currently working, except that you have
misunderstood the concept of subfilters.

> 2) When I ask for a message to be moved then it should be moved not
> copied and marked for deletion at the end of filtering.

This too is how it works. The message is still there for further moves
though, even if it's not in the Inbox anymore.



Just a final comment on your message
<mid:[EMAIL PROTECTED]>:

> I want each list to go into it's separate folder but each time I join
> a list I don't want to have to change the top level filter to say this
> is in a list and then create one for the list to move it.

But this is how it works and I do believe this suites most of the users.
You apparently do not use parent filters to filter messages but rather
to group filter together. I can see the need for that too, but that is
not what the concept of subfilters is there fore.

Changing the way the NFS works would break it's current functionality.
I'm sorry to say, but I think you will have to adjust.

You could make your parent list filter trigger on a certain address book
group and add all new mailing lists you join to this group. That way,
you could still use the concept of subfilters to group filters together
just like you want it. Not a perfect solution for you, but at least a
solution that won't break the functionality of the NFS.

-- 
Regards,
Marcus Ohlström

Using The Bat! v3.0.0.8 on Windows 2000 5.0 Build 2195 Service Pack 4
PGP Public Key at http://www.canit.se/~marcus/pgp.asc





________________________________________________________
 Current beta is 3.00.11 | 'Using TBBETA' information:
http://www.silverstones.com/thebat/TBUDLInfo.html
IMPORTANT: To register as a Beta tester, use this link first -
http://www.ritlabs.com/en/partners/testers/

Reply via email to