Write a proper `rmw_take_sequence` test

[clalancette]

While investigating the causes behind https://github.com/ros2/rmw_cyclonedds/issues/279 , we discovered that there is no test within ROS 2 that actually does a rmw_take_sequence and then checks on the results. We do have a test that is pretty close: https://github.com/ros2/rcl/blob/7a25a74cf911840224fb5b992d118c22c825f961/rcl/test/rcl/test_subscription.cpp#L388 . That initializes the sequence, fills it with messages, and then calls rcl_take_sequence.

It would be nice to have another variant of this test in this repository which specifically deals with rmw_take_sequence.

[Aswin0305]

Hi maintainers 👋,

I’m new to ROS 2 and open-source contributions, and I’m looking for a well-scoped first issue to work on and learn more about RMW behavior and testing.

I came across this issue (#176) and wanted to check whether it is still relevant and available to be worked on. If so, I’d be happy to take it up.

Are there any pointers on where to start (e.g., existing tests to look at, expected behavior, or preferred rmw implementations to target)?

Thanks for your time!

[fujitatomoya]

this issue is still valid and IMO nobody is working on this right at this moment.

probably you can take a look at around https://github.com/ros2/rmw_implementation/blob/8bcc72770ff638b816d723660dafda82fccc28ff/test_rmw_implementation/test/test_subscription.cpp#L775 to understand.

these tests are running on each rmw implementation which includes rmw_cyclonedds.

the current test cases are not enough to detect https://github.com/ros2/rmw_cyclonedds/issues/279, because it does not actually put the message in the topic endpoint and take it with calling rmw_take_sequence call.

so the issue report says that, if we can do the similar test https://github.com/ros2/rcl/blob/7a25a74cf911840224fb5b992d118c22c825f961/rcl/test/rcl/test_subscription.cpp#L388 in https://github.com/ros2/rmw_implementation/tree/rolling/test_rmw_implementation, that can detect the issues like https://github.com/ros2/rmw_cyclonedds/issues/279. (we should have been able to detect that issue.)