Hi,
I think I traced a bug in 2.0beta6
in Mailman/Handlers/CookHeaders.py
# # Reply-To: munging. Do not do this if the message is "fast tracked", # meaning it is internally crafted and delivered to a specific user, # or if there is already a reply-to set. If the user has set # one we assume they have a good reason for it, and we don't # second guess them. if not fasttrack or msg.get('reply-to'):
this line should be if not fasttrack and not msg.get('reply-to'):
Sorry but I have no time to use diff.
This bug caused confirm message to go list address in some cases (cf. -users discussion)
Tokio
"TK" == Tokio Kikuchi <tkikuchi@is.kochi-u.ac.jp> writes:
TK> I think I traced a bug in 2.0beta6
TK> in Mailman/Handlers/CookHeaders.py
>> # # Reply-To: munging. Do not do this if the message is "fast
>> tracked", # meaning it is internally crafted and delivered to a
>> specific user, # or if there is already a reply-to set. If the
>> user has set # one we assume they have a good reason for it,
>> and we don't # second guess them. if not fasttrack or
>> msg.get('reply-to'):
| this line should be
| if not fasttrack and not msg.get('reply-to'):
TK> Sorry but I have no time to use diff.
| This bug caused confirm message to go list address
| in some cases (cf. -users discussion)That's in the CVS tree. Given that MM 2.0 final is just a couple of weeks away, is this problem serious enough to warrant an emergency beta 7?
-Barry
At 09:12 AM 9/29/00 -0400, Barry A. Warsaw wrote:
That's in the CVS tree. Given that MM 2.0 final is just a couple of weeks away, is this problem serious enough to warrant an emergency beta 7?
My first reaction was "naah." Then I thought about it for a few minutes, and, actually, probably the answer should be "Yes." Because it's an itty bitty thing that can turn into a *major* hassle for a list admin, particularly a less experienced one who's not going to know why in hell his users have suddenly gone stupid and started sending all their confirms to the list. Heck, *I'm* far from inexperienced, and when I had 6 cases of it in two days after the bug hit the cvs tree *I* was initially wondering what in hell was wrong with my users all of a sudden. Then I finally realized the headers were wrong right about the point the patch hit the tree...
Particularly if the list allows outside subscribers, and the listadmin doesn't actually really read the list, he could have hordes of people subscribing, pestering the list with their confirms, and then wondering why they never get anything from the list they're actually not really subscribed to...
"confirm " + a number should probably be an administrivia filter trap.
participants (3)
-
bwarsaw＠beopen.com -
Ron Jarrell -
Tokio Kikuchi