Update GenericSubscription's handle_message signature

[emersonknapp]

Fixes the build since ros2/rclcpp#1047

Signed-off-by: Emerson Knapp emerson.b.knapp@gmail.com

[emersonknapp]

Unblocks #372

[Karsten1987]

Just stumbled upon this myself here: ros2/rmw_iceoryx@52a4bb3

Do you think we can actually shift this generic subscription to rclcpp? I feel like it deserves a more central place than rosbag to be reusable.

[emersonknapp]

Maybe generally valuable, we could port it over after the distro release.

[emersonknapp]

• Linux
• Linux-aarch64
• macOS
• Windows

[emersonknapp]

We're still trying to figure out why the test is failing in the Action CI... And that Linux failure on ROS CI is very confusing, some symbol thing in Connext.

Bit hard to keep on top of all the changes happening in the core

[Karsten1987]

do you have an idea why the GitHub Actions fails?

  26: [INFO] [1586831337.052728306] [rosbag2_transport]: Subscribed to topic '/string_topic'
  26: [ERROR] [1586831337.053618210] [rosbag2_transport]: Failed to record: Failed to create interrupt guard condition in Executor constructor: the given context is not valid, either rcl_init() was not called or rcl_shutdown() was called., at /home/runner/work/rosbag2/rosbag2/ros_ws/src/ros2/rcl/rcl/src/rcl/guard_condition.c:69
  26: [       OK ] RecordIntegrationTestFixture.mixed_qos_subscribes (18 ms)
  26: [ RUN      ] RecordIntegrationTestFixture.duration_and_noncompatibility_policies_mixed
  26: [INFO] [1586831337.070318775] [rosbag2_transport]: Listening for topics...
  26: [INFO] [1586831337.072540884] [rosbag2_transport]: Subscribed to topic '/mixed_nondelivery_policies'
  26: [ERROR] [1586831337.073290787] [rosbag2_transport]: Failed to record: Failed to create interrupt guard condition in Executor constructor: the given context is not valid, either rcl_init() was not called or rcl_shutdown() was called., at /home/runner/work/rosbag2/rosbag2/ros_ws/src/ros2/rcl/rcl/src/rcl/guard_condition.c:69

[emersonknapp]

What you're pointing at there is actually an error message in shutdown after a successful test. It doesn't mark that test as failed.

The failing test is timing out after 60 seconds -

2020-04-14T02:30:03.1573398Z 27: [ RUN      ] RosBag2PlayQosOverrideTestFixture.topic_qos_profiles_overriden
2020-04-14T02:30:03.1573530Z 27: [INFO] [1586831337.995065607] [rosbag2_transport]: Overriding QoS profile for topic /test_topic
2020-04-14T02:30:03.1573807Z 27/34 Test #27: test_play__rmw_fastrtps_dynamic_cpp ......................***Timeout  60.06 sec

2020-04-14T02:30:03.1627329Z 
2020-04-14T02:30:03.1627422Z The following tests FAILED:
2020-04-14T02:30:03.1627704Z 	 18 - test_play__rmw_fastrtps_cpp (Timeout)
2020-04-14T02:30:03.1627958Z 	 27 - test_play__rmw_fastrtps_dynamic_cpp (Timeout)

It's passing locally so I'm having a hard time telling what's going on. It will be nice once we switch over to containers, then I could theoretically pull the CI container locally to run the exact environment.

[emersonknapp]

I'm looking at the test that is timing out and it doesn't look like it's doing what it says it does. I have to dig in further. Probably won't get to it until tomorrow morning, though.

[emersonknapp]

My guess is that the playback portion of the test is publishing some/all of its messages before our subscription connects to it, then waiting until the timeout for messages which do not arrive. It needs a rework to be timing-insensitive, I can do this first thing tomorrow.

[wjwwood]

lgtm, sorry folks, I had this change locally, forgot to pr it in the chaos. My fault.

[piraka9011]

• Linux

[wjwwood]

CI lgtm, you can wait on the CI in ros2/rclcpp#1047 (comment) or merge this now, either is fine I believe.