Solution: now if the first frame in a multipart message is not subscribe/unsubscribe,
the rest of the parts are also considered to be not subscribe/unsubscribe.
Now only the first part in a multipart message will be treated as subscribe/unsubscribe.
The rest will be considered user messages regardless of the first byte.
Solution: remove the pipe from the real trie when a peer disconnects.
Also add a unit test that exercises the behaviour by reconnecting
a different socket and sending a message that matches.
Fixes#2601 and introduced by #2042
Solution: use ZMQ_LAST_ENDPOINT in most places. This alllows running
tests in paralle, and on over-booked shared machines where many of
the ports would be already in use.
Keep 3 tests with an hardcoded port, as there are some code paths that
require it (eg: connect before bind), but list those ports in
tests/testutil.hpp as macros so that they do not overlap and still
allow parallel runs.
These changes were inspired by a patch uploaded to Ubuntu by the
package maintainer, Steve Langasek <steve.langasek@ubuntu.com>.
Thank you Steve!
Solution:
- Adjust test_xpub_proxy_unsubscribe_disconnect() to support different
protocol types
- Exclude the IPC tests on Windows and OpenVMS
H/T: @somdoron
Solution: use msleep (SETTLE_TIME) everywhere when waiting for the
connections/sockets to be settled instead of a variety of patterns
and functions to make tests more coherent.
Tests were failing, because some deque calls were causing undefined
behavior: calling front() or pop_front() on an empty deque. Such
calls are now safeguarded.
Of course people still "can" distributed the sources under the
LGPLv3. However we provide COPYING.LESSER with additional grants.
Solution: specify these grants in the header of each source file.
