Occasional RETCODE_NO_DATA in basic_test

[iguessthislldo]

@mitza-oci this is the issue I mentioned a few weeks ago when I got this before 0.1 release. It happened again: https://github.com/oci-labs/pyopendds/runs/455791295. I will try to see if it's my fault or not (It probably is).

[iguessthislldo]

take_next_sample python implementation is here and that calls a type specific version here where the OpenDDS take_next_sample is called.

[mitza-oci]

On this line, what happens when rc is timeout?
https://github.com/oci-labs/pyopendds/blob/c6d4b8314661b064e68896c515b0e5a047ef8759/pyopendds/dev/include/pyopendds/user.hpp#L193

[iguessthislldo]

Huh, yea it's not handling that. It should be using Errors::check_rc() like I do just below where it's handling the return code from take_next_sample. I looked through the history and it looks like I wrote it like that and just forgot to do anything with rc:

https://github.com/oci-labs/pyopendds/blob/0a3319f01324ba791152a3c292dc70a45994eeeb/pyopendds/dev/itl2py/CppOutput.py#L182

[iguessthislldo]

Okay so that's being checked now, but if the test subscriber is timing out sometimes in the github actions, then there's another issue that's causing it to time out. The publisher only hangs around for 1 second after writing the sample and doesn't use wait_for_acknowledgements(), maybe it needs to?