I've been testing with the new Qt 5.5 release candidate and have had issues with our application ceasing to get notified of new datagrams.
We have a simple signal handler that follows the example given in the documentation that loops while the socket indicates that there are still datagrams available. To test, I set up a timer to poll the 'hasPendingDatagrams' property and see that when I stop processing datagrams the socket still indicates that there are datagrams available. After investigating QAbstractSocketPrivate::canReadNotification I am wondering if there is an issue there with a recent patch which turns off the socket engine read notification. Is it possible that in the signal handler I read all datagrams and between the time I complete my processing and the time the abstract socket checks to see if there are pending datagrams, the socket could have received a new datagram my signal handler didn't see but now the abstract socket does see? It seems like the read notification should not be disabled if the signal handler read until there was no more data. Does this seem reasonable and is there a way to prevent the disabling of the read notification under this circumstance? Thanks, Robert Daniels _______________________________________________ Interest mailing list Interest@qt-project.org http://lists.qt-project.org/mailman/listinfo/interest