> Actually, I got the (|) sequence to work, but never worked was having
> the block after the initial filter
>
> The verbose log looks like this on a failure
> procmail: Assigning "SUBJ=Fri Nov 1 16:23:31 PST 1996"
> procmail: Match on ! "^Subject: *\[testing\]"
> procmail: Match on ! "^Subject: *re:"
> procmail: Executing " formail -I "Subject: [$list] $SUBJ""
> procmail: [15442] Fri Nov 1 16:23:35 1996
> procmail: Assigning "LASTFOLDER= formail -I "Subject: [$list] $SUBJ""
> Subject: Fri Nov 1 16:23:31 PST 1996
> Folder: formail -I "Subject: [$list] $SUBJ"
394
>
> SUBJ=`formail -zx Subject:` # extract the subject
> :0 w
> * $! ^Subject: (Re:|*\[$list\])
> {
> :0 w
> | formail -I "Subject: [$list] $SUBJ"
> }
Jason,
This recipe is wrong:
1. the '*' is misplaced; it should be after the blank following the
colon, meaning "an arbitrary number of blanks". I'm sorry if I had
originally typed it this way.
2. the flags on the second recipe should be "fhW", where "fh" means
"filter the headers", and "W" means wait for a successful result.
Also, If you have procmail 3.10 or 3.11, you can combine the subject
extraction by using the match pattern "\/" and then the MATCH variable
like this:
:0 fhW
* $!^Subject: *(Re:|\[$list\])
* ^Subject: *\/[^ ].*
| formail -I"Subject: [$list] $MATCH"
I hope this helps.
___________________________________________________________
Alan Stebbens <aks(_at_)sgi(_dot_)com> http://reality.sgi.com/aks