Author tim.peters
Recipients Mark.Shannon, benjamin.peterson, christian.heimes, inada.naoki, jdemeyer, larry, lukasz.langa, miss-islington, nascheme, ned.deily, pablogsal, petr.viktorin, pitrou, tim.peters, vstinner
Date 2019-10-01.18:29:30
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1569954570.37.0.419140675844.issue38006@roundup.psfhosted.org>
In-reply-to
Content
Neil, about this comment:

#   - ct is not yet trash (it actually is but the GC doesn't know because of
#     the missing tp_traverse method).

I believe gc should know ct is trash.  ct is in the cf list, and the latter does have tp_traverse.

What gc won't know is that `a` is trash, because `a` is attached to ct, and ct doesn't have tp_traverse.

It should blow up anyway :-)

Maybe with a simpler structure it would be easier to rearrange code to nudge the callback into getting cleared before use?

Z <- Y <- A <-> B -> WZ -> C

where WZ is a weakref to Z with callback C, and Y doesn't implement tp_traverse.  The only cycle is between A and B, which could just as well be the same object.  All the other stuff hangs off that cycle.

It's all trash, but we won't know in advance that Z is part of it.
History
Date User Action Args
2019-10-01 18:29:30tim.peterssetrecipients: + tim.peters, nascheme, pitrou, vstinner, larry, christian.heimes, benjamin.peterson, ned.deily, petr.viktorin, inada.naoki, lukasz.langa, Mark.Shannon, jdemeyer, pablogsal, miss-islington
2019-10-01 18:29:30tim.peterssetmessageid: <1569954570.37.0.419140675844.issue38006@roundup.psfhosted.org>
2019-10-01 18:29:30tim.peterslinkissue38006 messages
2019-10-01 18:29:30tim.peterscreate