Trying to direct email based on a regex match.
I know I can do this:
echo | mail -s test1 me-route2
with this recipe:
* ^To:.*me-route2\@
:0:
/home/me/folder-route2/afile
What I'd like to do is something like this:
* ^To:.*me-(route\d)\@
:0:
/home/me/folder-$MATCH/afile
but I am getting:
procmail: Lock failure on "/home/me/folder-/afile.lock"
You have the prologue line and the condition in the wrong order. But in addition, Procmail does not recognize the Perlism \d. (See e.g. here for a bit of regex history and arcana.) Try this instead:
:0:
* ^To:.*me-\/route[0-9]
/home/me/folder-$MATCH/afile
or, if you really want to verify that the matching expression is immediately followed by an @ sign,
:0:
* ^To:.*me-\/route[0-9]@
* MATCH ?? ^\/route[0-9]
/home/me/folder-$MATCH/afile
From the input To: reallynotme-route2@example.com, the first condition captures route2@ and the second condition matches up through route2 in order to drop the trailing @ from MATCH.
My suspicion is that no lock file is necessary, so the second colon should be removed, but this depends on what afile is. If it's a regular Berkely mbox file, you should definitely use locking. If it's a Maildir folder, you should not.
You might also want to look into replacing ^To:.*me with ^TO_me -- it will match on Cc: and other pertinent headers as well as a literal To: header, and avoid matching on reallynotme.
Procmail doesn't use capture groups like normal regular expressions, at least for MATCH. From the man page:
This variable is assigned to by procmail whenever it is told to extract text from a matching regular expression. It will contain all text matching the regular expression past the `/' token.
So the following should work (untested):
* ^To:.*me-\/route\d
MATCH will then contain whatever matched the regex after the \/, so "route\d".
Sadly this will do possibly unexpected things with an address like me-route2-ab@example.com, with MATCH being "route2". But that may be acceptable to you.
