Update realtime containers

[christophfroehlich]

Apply changes from the guidelines added with ros-controls/realtime_tools#347

[codecov]

Codecov Report

Attention: Patch coverage is 97.46835% with 2 lines in your changes missing coverage. Please review.

Project coverage is 85.70%. Comparing base (483d6ac) to head (f093e00).
Report is 3 commits behind head on master.

Files with missing lines	Patch %	Lines
..._drive_controller/src/mecanum_drive_controller.cpp	94.28%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1721   +/-   ##
=======================================
  Coverage   85.69%   85.70%           
=======================================
  Files         123      123           
  Lines       11897    11912   +15     
  Branches     1015     1015           
=======================================
+ Hits        10195    10209   +14     
- Misses       1379     1380    +1     
  Partials      323      323           

Flag	Coverage Δ
unittests	85.70% <97.46%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
...ller/include/mecanum_drive_controller/odometry.hpp	100.00% <ø> (ø)
..._controller/test/test_mecanum_drive_controller.cpp	100.00% <100.00%> (ø)
..._drive_controller/src/mecanum_drive_controller.cpp	90.23% <94.28%> (-0.46%)	⬇️

... and 1 file with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[saikishor]

Suggested change

	input_ref_.try_set(emtpy_msg);
	input_ref_.set(emtpy_msg);

Shouldn't it be a set in this case? As we are in the activate callback

[christophfroehlich]

on_activate is in the rt thread, isn't it?

[saikishor]

Yes, you are right. In this particular case, it is wiser to use a set, as you want the controller to start with a zero state. Moreover, it is hard to have it blocking, unless you keep spamming the topic even before the controller started

[christophfroehlich]

but if the lock fails here, wouldn't it be very likely that the recently received command can be read with the try_get at the beginning of the update method?

[christophfroehlich]

then this blocking set here has no use