Malamute version 1.1 stable, released on 20xx/xx/xx
===================================================

* This release is ABI/API compatible with 1.0.
  All 1.0 APIs have been officially marked as "STABLE", which means they
  will not change anymore. There is no change for users.

* Many general improvements to the malamute service. systemd units and the
  configuration files will now be correctly installed.

* Add bounded mailbox that uses up to a configurable amount of memory.
  The default behaviour is unchanged and will still use as much memory as it
  can, set mailbox size-limit in malamute.cfg to change it.

* Add settings to limit service queue memory usage.
  The default behaviour is unchanged and will still use as much memory as it
  can, set service queue size-limit to change it.

* mlm_server: add CLIENTLIST actor API to return list of connected clients

* mlm_server: add STREAMLIST actor API to return list of known streams

* Added LUA bindings

* Add Python CFFI bindings

* Improve support for Visual Studio

* Fix compilation on Windows via Mingw64

* Fix compilation wih GCC 7

* Add Debian and Redhat packaging

* Use inproc for unit tests rather than IPC, to fix running on Windows

* Fixed #167 - mlm_client: fixed recovery from disconnect state

* Fixed #171 #175 - mlm_client: fixed sporadic hang in destructor

* Fixed #172 - mlm_client: fixed mlm_client_connected sometimes being false
               despite successful connection

* Fixed #174 - mlm_server: send error message back to client when in state
               settling and the wrong message is received to avoid deadlock

* Fixed #176 - mlm_client: fix moving from reconnecting to connected state

* Fixed #177 - mlm_client: solve ambiguity by renaming disconnecting to
               terminating

* Fixed #178 - mlm_client: fix possible set_* hang when receiving PONG

* Fixed #215 - mlm_client: return an error if an invalid pattern is used in
               set_consumer

* Fixed #254 - mlm_server: fix segfault on restart due to incomplete connection

* Fixed #261 - mlm_server: fix memory leak when de-registering clients

* Fixed #268 - mlm_client: fix memory leak in Python binding on receive
