Package: bugs.debian.org On Mon, Apr 12, 2010 at 01:45:11AM +0000, Debian Bug Tracking System wrote: > Processing commands for [email protected]: > > > clone 576944 -1 > Bug#576944: python-xapian: Perform unaligned memory access on ARM > Bug is marked as being merged with others. Use an existing clone. > > > reassign -1 g++ > Failed to clear fixed versions and reopen on -1: The 'bug' parameter ("-1") > to Debbugs::Control::set_package did not pass regex check > > Debbugs::Control::set_package('transcript', > 'IO::Scalar=GLOB(0x27ab730)', 'requester', 'Olly Betts <[email protected]>', > 'request_addr', '[email protected]', 'request_msgid', > '<[email protected]>', 'request_subject', ...) called at > /usr/lib/debbugs/service line 556 > eval {...} called at /usr/lib/debbugs/service line 555 > > > tag 576944 + pending > Bug #576944 [python-xapian] python-xapian: Perform unaligned memory access on > ARM > Bug #576945 [python-xapian] python-xapian: performs unaligned memory access > on ARM > Added tag(s) pending. > Added tag(s) pending. > > thanks > Stopping processing here.
I was using the common idiom, documented here: http://www.debian.org/Bugs/server-control#clone There's no caveat about merged bugs, and even if this limitation was mentioned, it's tedious to have to consider the history of the ticket when you just want to assign a clone to another package. So documenting this limitation would certainly be an improvement, but doesn't see a good way to fix this issue. Forcing an existing merged bug number to be reused here seems unhelpful - external references to that bug number will now rather arbitrarily lead to a bug in a different package. It's especially annoying that any further commands get executed on the uncloned bug number, meaning that these changes then need to be undone once you unmerge the ticket. I'd have been less annoyed by this if the clone failure had ended processing. Cheers, Olly -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
