nodelet_core Do not subscribe /clock from `nodelet load`

Depends on https://github.com/ros/ros_comm/pull/2342 .

See discussion in https://robotics.stackexchange.com/questions/96165/nodelet-load-process-does-message-deserialization .

When running in nodelet load mode, the node doesn't really do anything that would require sim time. It should only run the bond, one service call and nothing else. This PR makes sure that it is the case even if /use_sim_time is set to true.

Aug 16 '23 12:08 peci1

@ros-pull-request-builder retest this please

Sep 04 '24 16:09 sloretz

Hmm, maybe that was too fast and we actually need to wait for ros_comm release?

Sep 04 '24 17:09 peci1

The tests in this PR have passed after the required changes in ros_comm have been merged.

This is now ready for review.

Sep 14 '24 11:09 peci1

@sloretz it would be great to have this PR in the larger ros_comm test you called for. Could you please have a look at it?

Sep 20 '24 20:09 peci1

Thanks! Do you intend to make a release of nodelet_core in the period when ros_comm 1.17 is being tested?

Sep 25 '24 18:09 peci1

This pull request has been mentioned on ROS Discourse. There might be relevant details there:

https://discourse.ros.org/t/preparing-for-noetic-sync-2024-10-31/40332/5

Nov 02 '24 13:11 ros-discourse

This pull request has been mentioned on ROS Discourse. There might be relevant details there:

https://discourse.ros.org/t/preparing-for-noetic-sync-2024-12-05/40942/2

Dec 04 '24 08:12 ros-discourse

This pull request has been mentioned on ROS Discourse. There might be relevant details there:

https://discourse.ros.org/t/new-packages-for-noetic-2024-12-06/40995/2

Dec 06 '24 18:12 ros-discourse