Problem: -1 used for invalid socket

Solution: use retired_fd instead

Problem: testutil build broken with gcc 4.4 on CentOS 6

Problem: Multiple fd_t definitions

Problem: Poller event fd unspecified

Solution: Unify definition

Solution: do not alias sockaddr_un, but take a copy

tests/testutil.cpp: In function 'fd_t bind_socket_resolve_port(const char*, const char*, char*, int, int)':
tests/testutil.cpp:468: error: dereferencing pointer 'un_addr' does break strict-aliasing rules

Solution: remove it and rework the ifdefs to be safe against uninitialised use

Problem: no fuzz testing

Solution: Specify an invalid file descriptor for socket events

Solution: add tests for zmq_bind, zmq_connect and zmq_z85_decode

Solution: add simple, initial tests for client and server

Solution: port the 2 new tests from oss-fuzz, and wire them up to
be ran manually with a static input in normal builds.
Add a specific configure option to use the external fuzzing engine
from oss-fuzz.

Solution: define buffers in common header

Solution: refactor in testutil.lib, so that they can be used for fuzzers too

Solution: add a new TEST_ASSERT_SUCCESS_RAW_ZERO_ERRNO macro so that
it can check explicitly for non-zero values. This will be used
for getaddrinfo().

Solution: create socket in the same thread where it is used

RELICENSE: Grant from laplaceyang.

problem: appveyor is very slow

Solution: move to github actions

Problem: out-of-bounds array access in socket_poller::check_events.

Problem: poller item lookup can be simplified

Solution: Extract generic find function

problem: checking the wrong socket when sending hello msg

Problem: Spinning on atomics can cause hangs

Fix coverage job and appveyor job

Solution: Add a sleep in the loop. Some versions of
valgrind may hang when spinning on atomic variables.

Solution: bump OS to Xenial for newer python libraries

Solution: don't run it

problem: router doesn't know when peer disconnected 

typo fix root README

Problem: No support to query poller size

Solution: Add zmq_poller_size that queries the number
of objects registered, allowing safer usages of poller
to avoid livelock situations.

ZMQ_ROUTER_NOTIFY doesn't have a context and doesn't play nice with protocols. with ZMQ_DISCONNECT_MSG we can set it to a protocol message, like DISCONNECT in majordomo. Router will send it when a peer is disconnected. Another advantage of ZMQ_DISCONNECT_MSG is that it also works on inproc.

Together with ZMQ_HEARTBEAT it allows to build very reliable protocols, and much simpler as well.

problem: ZMQ_HEARTBEAT is not useful without sending an hello message

When using ZMQ_HEARTBEAT one still needs to implement application-level heartbeat in order to know when to send a hello message.
For example, with the majordomo protocol, the worker needs to send a READY message when connecting to a broker. If the connection to the broker drops, and the heartbeat recognizes it the worker won't know about it and won't send the READY msg.
To solve that, the majordomo worker still has to implement heartbeat. With this new option, whenever the connection drops and reconnects the hello message will be sent, greatly simplify the majordomo protocol, as now READY and HEARTBEAT can be handled by zeromq.