Conversation
| // reactivated | ||
| // wait so controller process the third point when reactivated | ||
| std::this_thread::sleep_for(std::chrono::milliseconds(3000)); | ||
| // TODO(anyone) test copied from ROS 1: it fails now! |
There was a problem hiding this comment.
I merged the old reactivation test into that one, should the old trajectory really be processed after reactivation? (it doesn't)
There was a problem hiding this comment.
It shouldn't imo and we should have a test to ensure it
There was a problem hiding this comment.
yes, I agree. Would be strange if it does. Writing the test for that is easy, because this one fails already :D
| expected_actual, expected_desired, executor, rclcpp::Duration(delay * (2 + 2)), 0.1); | ||
| } | ||
|
|
||
| // TODO(destogl) this test fails with errors |
There was a problem hiding this comment.
@destogl maybe you can have a quick look on that?
| } | ||
| #endif | ||
|
|
||
| // TODO(destogl) this test fails |
There was a problem hiding this comment.
@destogl maybe you can have a quick look on that?
| // if using RCL_STEADY_TIME -> | ||
| // C++ exception with description | ||
| // "can't compare times with different time sources" thrown in the test body. | ||
| // traj_controller_->update(clock.now(), clock.now() - start_time); |
There was a problem hiding this comment.
should we create the node_ with NodeOptions, clock_type = RCL_STEADY_TIME?
There was a problem hiding this comment.
Some of these may be from a time we didn't have any clock solution, worth reassessing if needed
|
Confirming here that the only reason I remember tests were disabled was flakiness |
(cherry picked from commit 9908a9c) # Conflicts: # joint_trajectory_controller/test/test_trajectory_controller.cpp
(cherry picked from commit 9908a9c) # Conflicts: # joint_trajectory_controller/test/test_trajectory_controller.cpp
In preparation for writing new tests for the upcoming changes:
If tests were deactivated for other reasons than being flaky, please let me know ;)
Three tests are failing, and I don't know if the test logic or JTC behavior is broken. I added comments below