package: debbugs severity: normal Just noticed this today. I sent a message to 534466-close@bugs.debian.org including a command with the new "control:" syntax, but it was ignored: http://bugs.debian.org/534466 Best wishes, Mike
Control: severity -1 wishlist Control: retitle -1 consider allowing control commands at -done and -forwarded time Right, this is as documented: http://www.donarmstrong.com/posts/control_at_submit/ Don Armstrong
The brute fact “[implemented] with the exception of messages received at nnn-done and nnn-forwarded” is documented, true. I don't see a rationale for *why* that exclusion is needed, though. Why exclude those addresses from this feature? It's surprising, because IMO those addresses are part of the set “sending messages to the bug report” and should parse the message content the same way. To answer an implied question at that document: Because a message to a bug report is an excellent time to update the bug report's metadata with BTS control commands, and ‘NNN-done’ / ‘NNN-forwarded’ are no exception to that. For example, <URL:https://bugs.debian.org/323015#51>. I hope that provides a use case, so that you *do* know why this is wanted.
Control: retitle -1 allow Control: commands at -done and -forwarded I've become convinced that this is useful in the time since I documented it. [Indeed, if it was trivial to implement, I would have added it right then too.] I just haven't had a chance to do the legwork to implement it properly.
Thanks, good to know :-)
Shouldn't these bugs be merged? 685264: debbugs: obey “Control” field in pseudoheader of messages to ‘NNN-done@bugs.debian.org’ 705155: allow Control: commands at -done and -forwarded 746206: Control: pseudoheaders do not work for -done bugs; finish() called too early
forcemerge 685264 705155 746206 thanks Yep. Merging them.