add a semantic command interface to "semantic_components"

[tpoignonec]

Hi,

Is there any interest in defining a SemanticComponentCommandInterface analogous to the SemanticComponentInterface, but for hardware command interfaces?

Applications could include, among others, commonly used simple write-only devices such as LED, buzzers, beepers, etc.
It could also be used for GPIO arrays the same way the state interface can be used to read them in a standarized fashion.

This PR includes such a semantic component, as well as a basic example with a 3-channel (R/G/B) LED with two scenarios:

• Default interface names: <name>/r, <name>/g, and <name>/b
• Custom interface names using the following syntax:

#include "semantic_components/led_rgb_device.hpp"

std::string interface_name_r = "led/custom_r";
std::string interface_name_g = "led/custom_g";
std::string interface_name_b = "led/custom_b";

// Create LED device with custom interface names
auto led_device = std::make_unique<semantic_components::LEDRgbDevice>(
   interface_name_r, interface_name_g, interface_name_b);

Additionally, basic tests are proposed.

[codecov]

Codecov Report

Attention: Patch coverage is 95.40230% with 4 lines in your changes missing coverage. Please review.

Project coverage is 89.40%. Comparing base (d4139b7) to head (db374fe).

Files with missing lines	Patch %	Lines
...test/test_semantic_component_command_interface.hpp	60.00%	2 Missing ⚠️
...ace/include/semantic_components/led_rgb_device.hpp	93.33%	1 Missing ⚠️
...omponents/semantic_component_command_interface.hpp	94.11%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1945      +/-   ##
==========================================
+ Coverage   89.38%   89.40%   +0.01%     
==========================================
  Files         132      138       +6     
  Lines       14663    14756      +93     
  Branches     1258     1262       +4     
==========================================
+ Hits        13106    13192      +86     
- Misses       1090     1096       +6     
- Partials      467      468       +1     

Flag	Coverage Δ
unittests	89.40% <95.40%> (+0.01%)	⬆️

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

Files with missing lines	Coverage Δ
controller_interface/test/test_led_rgb_device.cpp	100.00% <100.00%> (ø)
controller_interface/test/test_led_rgb_device.hpp	100.00% <100.00%> (ø)
...test/test_semantic_component_command_interface.cpp	100.00% <100.00%> (ø)
...ace/include/semantic_components/led_rgb_device.hpp	93.33% <93.33%> (ø)
...omponents/semantic_component_command_interface.hpp	94.11% <94.11%> (ø)
...test/test_semantic_component_command_interface.hpp	60.00% <60.00%> (ø)

... and 1 file with indirect coverage changes

[Wiktor-99]

I've thought about it a little bit, and I'm not sure if we should add such a component when the GPIO command controller, which is very generic, is already available

[christophfroehlich]

Thanks for the PR including nice tests .
We discussed this shortly in todays WG meeting, and approve this addition.
I can understand the input @Wiktor-99, but imho

• it is an obvious extension of an existing concept
• it could also use joints and not only gpios

Please add it to the release_notes and some minor comments.

[tpoignonec]

Great!
I do the release note and changes ASAP.

[christophfroehlich]

LGTM overall, I just added some comments to be in line with #1940

Maybe you have to format my suggestions once again to satisfy pre-commit.

[tpoignonec]

My bad, hadn't rerun the unit tests after accepting the change suggestions (minor issue with LED constructor).
Should be fine now.

[Raivias]

Other than dropping the library LGTM.

[saikishor]

@tpoignonec Can you please fix the pre-commit jobs?

[mergify]

This pull request is in conflict. Could you fix it @tpoignonec?

[tpoignonec]

@saikishor I removed the constructor as discussed.

Due to these changes, the name variable does not really serve a purpose anymore.
A typical component would do

MyCommandInterface(const std::string & name)
  : SemanticComponentCommandInterface(name, {name + "/my_io_1",  ... , name + "/my_io_N"})
  {
  }

However, SemanticComponentCommandInterface does not actually use the variable name_ anymore.

Regardless, LGTM at this point.

PS: feel free to revert the last commit if needed