UTF-8 codec can't decode byte

[martfrau]

I'm using the ntrip_client (branch ros2 for ROS2 Humble) to connect to http://rtk2go.com:2101.

The connection is successful in about 90% of the cases, but sometimes the following error occurs:

[ntrip_ros.py-1] [ERROR] [1690972588.743143226] [ntrip_client.ntrip_client_node]: Unable to read response from server at http://rtk2go.com:2101
[ntrip_ros.py-1] [ERROR] [1690972588.743503821] [ntrip_client.ntrip_client_node]: Exception: 'utf-8' codec can't decode byte 0xd3 in position 12: invalid continuation byte
[ntrip_ros.py-1] [ERROR] [1690972588.743955650] [ntrip_client.ntrip_client_node]: Unable to connect to NTRIP server

After some debugging I found out that this is because of this strange server response:
b'ICY 200 OK\r\n\xd3\x00\x13>\xd0\x00\x03\x89\x97\th\xf7\x82q\x11\xf9\x97\x0b\tS\xa4O\xd3\x8b2'

When the connection is successful, the server response looks like this:
b'ICY 200 OK\r\n'

This means that sometimes there is a trailing byte sequence (which is not always the same).

The issue can be fixed by using ISO-8859-1 instead of utf-8 decoding in line 132 of the ntrip_client.py:
response = self._server_socket.recv(_CHUNK_SIZE).decode('ISO-8859-1')
(see https://github.com/LORD-MicroStrain/ntrip_client/blob/ros2/src/ntrip_client/ntrip_client.py#L132)

This fix would make the connection request more robust (although it is just a workaround for the strange server response).

What do you think? Have you any ideas why this server response occurs or how to further improve the fix?