v1.10.0

Important change: the bind_address of the sACNreceiver is ignored on Linux when binding the socket to an address. This is part of a bugfix, but might alter behavior in certain cases. (Thanks to @andrewyager! See #51 for more information)

Full Changelog: v1.9.1...v1.10.0

Fixed excpetion silently dropped

When a sACN packet could not be send out, the exception is raised instead of silently dropped. (See #48 for more information)

Internal: updated GitHub workflows to current Python versions and the publish workflow also runs on edited releases.

Unified Multicast Behavior

The behavior of multicast sending and receiving is now unified across most operating systems. This means Windows no longer requires to set a bind_address to be able to use multicast or universe discovery. (Thanks to @mthespian ! See #42 for more information)

Close Sockets on Stop

Calling stop on a sender or receiver now closes the underlying socket too. Note: after stopping a sender or receiver, it can not be started again with start. (See #39 for more information)

Added function for removing listener on receiver by universe

Added function for removing a listener on a receiver by specifing the universe. See sACNreceiver.remove_listener_from_universe(<universe>), #34, #35 and #37 for more information.

Minor timing improvements

Small changes that might improve timing on the sender. (Thanks to @mthespian ! See #36 for more information)

Added new function for removing listener on receiver

Added function for removing a listener on a receiver. See sACNreceiver.remove_listener(<callback>), #34 and #35 for more information.

Tests for sACNsender

Functionality related to sending of sACN data is now mostly covered by tests. Removed undocumented parameters for sACNsender.start().

Bugfix: Exception on special first received packet

Test coverage for functionality related to receiving sACN data is now mostly covered by tests.
Fixed a bug, where an exception was thrown on the first DataPacket when the stream-termination option was set.

(Additional thanks to @mthespian ! See #31 for more information)

Bugfixes of missing checks

Test coverage of sub-module messages is now 100%.

• Fixed a bug where a too long source name did not throw an exception.
• Fixed a bug where invalid DMX data could be set on the DataPacket.

(Thanks to mthespian! See #30 for more information)