On Thu, Feb 09, 2023 at 12:39:04PM +0100, Erik Skultety wrote: > On Thu, Feb 09, 2023 at 10:00:27AM +0100, Martin Kletzander wrote: > ... > > > diff --git a/tools/virt-qemu-qmp-proxy b/tools/virt-qemu-qmp-proxy > > > index dfbaa1ff0c..2d9dd6495d 100755 > > > --- a/tools/virt-qemu-qmp-proxy > > > +++ b/tools/virt-qemu-qmp-proxy > > > @@ -335,7 +335,7 @@ def main(): > > > sock.bind(args.sockpath) > > > sock.listen(1) > > > > > > - _ = QMPProxy(conn, dom, sock, args.verbose) > > > + QMPProxy(conn, dom, sock, args.verbose) > > > > > > > I don't think you can do that because the object could be garbage > > collected, but I'm not sure how much havoc would that cause... > > Would it? Not advocating to accept the patch, but looking at the object it > passes a reference to itself to qemuMonitorEventRegister() so as long as the > monitor callback map exists the object should not be garbage collected. However > the usage of _ in this case is IMO quite unfortunate in that it's simply there > to really make sure Python holds one more explicit reference to the object. In > cases like these Python's context managers are used instead. Yep, using a context manager pattern there would have been more sane. With regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
