WeakRefs, callbacks and GC when program quits

[Martin v. Löwis]

daniel_shane_eicon_junk at hotmail.com (Daniel Shane) writes:

> I agree. The reason I belive that the memory content is "uncertain" is
> because I am under the impression that python is freeing resources
> that still have strong links to them, therefore causing errors.

That is certainly not the case. Python will never ever release an
object that still has an outside reference to it.

> I will send in an example of the bug to sourceforge, but if this
> error is not considered a bug, I hardly see how executing callbacks
> at program termination can be considered anything but reliable.

The error is certainly a bug, but the bug is not that the callbacks
are called.

Regards,
Martin