[Mailman-Users] bin/rmlist too bold removing entries from postfix aliases file

[Todd]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi all,

First off, thanks to Barry and all the MM developers for such a fine piece
of software.

I was testing out MM2.1 yesterday and created a few test lists.  I was
feeling especially creative so I named them test and test2.  When I used
bin/rmlist to remove "test" I found that it had also deleted the test2
entries from data/aliases (I use postfix as my MTA).

I didn't see any mention of this in the list archives, but I could have
missed it.  Can anyone else confirm this behavior or am I doing something
wrong?

It looks like the problem is in Mailman/MTA/Postfix.py's _do_remove().  It
simply checks that the line startswith '# STANZA START: ' + listname.  This
matches any lists that start with the same string as another listname.  So
If I have the lists test, test-users, and test-announce, running 'bin/rmlist
test' will remove all three lists from data/aliases.  This makes it hard for
postfix to deliver mail to the list. :)

I'm not much of a coder (and even less of a python coder), but I managed to
patch Postfix.py to use a regexp instead of the simpler startswith to make
the match more accurate.  File this under WORKSFORME.  I'm sure it can be
improved upon, assuming this is really a problem and not a feature or some
mis-usage on my part

Here's the several line diff (let me know if there's a better place to post
this):

###########################################################################
- --- Mailman/MTA/Postfix.py.orig	Tue Dec 31 20:53:42 2002
+++ Mailman/MTA/Postfix.py	Tue Dec 31 20:48:47 2002
@@ -22,6 +22,7 @@
 import errno
 import pwd
 import grp
+import re
 from stat import *
 
 from Mailman import mm_cfg
@@ -248,8 +249,8 @@
         finally:
             os.umask(omask)
         filteroutp = 0
- -        start = '# STANZA START: ' + listname
- -        end = '# STANZA END: ' + listname
+        start = re.compile('^# STANZA START: ' + listname + ' *$')
+        end = re.compile('^# STANZA END: ' + listname + ' *$')
         while 1:
             line = infp.readline()
             if not line:
@@ -259,14 +260,14 @@
             # filtering out a stanza, we're just looking for the proper begin
             # marker.
             if filteroutp:
- -                if line.startswith(end):
+                if end.search(line):
                     filteroutp = 0
                     # Discard the trailing blank line, but don't worry if
                     # we're at the end of the file.
                     infp.readline()
                 # Otherwise, ignore the line
             else:
- -                if line.startswith(start):
+                if start.search(line):
                     # Filter out this stanza
                     filteroutp = 1
                 else:
###########################################################################

Thanks,

- -- 
Todd                  OpenPGP -> KeyID: 0xD654075A | URL: www.pobox.com/~tmz
============================================================================
Liberty is always dangerous, but it is the safest thing we have.
    -- Harry Emerson Fosdick

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)
Comment: When crypto is outlawed bayl bhgynjf jvyy unir cevinpl.

iD8DBQE+E4Rxuv+09NZUB1oRAsJ+AKCGCAOP7WVboICb/3iBDlzOAs85TACgkX3/
J0TijmI86Puluia21G+E7VM=
=t3kL
-----END PGP SIGNATURE-----