On 2007-11-18 02:13:04 -0400, Juan Manuel Palacios wrote:
I still have to find where that header is being set, since the relevant field on the mailing list administration page doesn't have any value (i.e. Mailman itself doesn't seem to be setting the Reply-To we currently have on those messages, so I'm guessing it's the post-commit hook itself doing it).
Yes, the "Reply-To:" header is not a header a mailing-list manager should set. -- Vincent Lefèvre <vincent@vinc17.org> - Web: <http://www.vinc17.org/> 100% accessible validated (X)HTML - Blog: <http://www.vinc17.org/blog/> Work: CR INRIA - computer arithmetic / Arenaire project (LIP, ENS-Lyon)
On Nov 18, 2007, at 7:14 AM, Vincent Lefevre wrote:
On 2007-11-18 02:13:04 -0400, Juan Manuel Palacios wrote:
I still have to find where that header is being set, since the relevant field on the mailing list administration page doesn't have any value (i.e. Mailman itself doesn't seem to be setting the Reply-To we currently have on those messages, so I'm guessing it's the post-commit hook itself doing it).
Yes, the "Reply-To:" header is not a header a mailing-list manager should set.
What difference does it make to impose a "Reply-To" header at the post-commit hook level or at the mailing list level? The overall effect is the same. In any case, Bill will set author & macports-dev as the Reply-To of commit mails in the coming days. Regards,... -jmpp
On 2007-11-19 19:00:45 -0400, Juan Manuel Palacios wrote:
What difference does it make to impose a "Reply-To" header at the post-commit hook level or at the mailing list level? The overall effect is the same.
Well, this was a general rule. If someone (who is not the post-commit hook) sends a message to the mailing-list, I think this should be treated as a normal mailing-list message and should not have a "Reply-To:" (except the possible one added by the user). And in some cases, the post-commit hook may have more information than the mailing-list manager to decide what should go to the Reply-To. -- Vincent Lefèvre <vincent@vinc17.org> - Web: <http://www.vinc17.org/> 100% accessible validated (X)HTML - Blog: <http://www.vinc17.org/blog/> Work: CR INRIA - computer arithmetic / Arenaire project (LIP, ENS-Lyon)
On Nov 19, 2007, at 17:19, Vincent Lefevre wrote:
On 2007-11-19 19:00:45 -0400, Juan Manuel Palacios wrote:
What difference does it make to impose a "Reply-To" header at the post-commit hook level or at the mailing list level? The overall effect is the same.
Well, this was a general rule. If someone (who is not the post-commit hook) sends a message to the mailing-list, I think this should be treated as a normal mailing-list message and should not have a "Reply-To:" (except the possible one added by the user). And in some cases, the post-commit hook may have more information than the mailing-list manager to decide what should go to the Reply-To.
True, but nobody other than the post-commit hook should ever be sending anything to the macports-changes list, right?
On Nov 19, 2007, at 11:16 PM, Ryan Schmidt wrote:
On Nov 19, 2007, at 17:19, Vincent Lefevre wrote:
On 2007-11-19 19:00:45 -0400, Juan Manuel Palacios wrote:
What difference does it make to impose a "Reply-To" header at the post-commit hook level or at the mailing list level? The overall effect is the same.
Well, this was a general rule. If someone (who is not the post-commit hook) sends a message to the mailing-list, I think this should be treated as a normal mailing-list message and should not have a "Reply-To:" (except the possible one added by the user). And in some cases, the post-commit hook may have more information than the mailing-list manager to decide what should go to the Reply-To.
True, but nobody other than the post-commit hook should ever be sending anything to the macports-changes list, right?
Yes, no-one else should send mail to that list, it is read-only to the effects of everyone but the post-commit hook. So even though Vincent makes some good points, in this case the overall result is still the same regardless of where the Reply-To's are set. But well, done with the dissertation exercise, Bill will cover for us at the post-commit hook as soon as he gets a chance. Regards,... -jmpp
participants (3)
-
Juan Manuel Palacios
-
Ryan Schmidt
-
Vincent Lefevre