Add support for positional target frame arguments for transform wrench node

[vedh1234]

This PR adds support for passing target frames as positional CLI arguments to wrench_transformer_node, similar to tf2_echo. ( Issue #2027 ) Positional arguments override the existing target_frames parameter when both are provided.

Includes:

• CLI parsing for non-ROS arguments
• Parameter override logic
• Updated documentation
• Test for positional arguments

[thedevmystic]

Phenomenal job on adding this features. Appreciate it, @vedh1234

[thedevmystic]

I would suggest why don't we keep the standalone executable comment too, and introduce tf2_echo frames option as an "option". So that previous usage of the controller can be still used. And the previous documentation is still valid.

[vedh1234]

The node cannot run without specifying target_frames (it throws an initialization error). I saw the tf2_echo doc which also mentions frame usage, that's why I thought of adding it and removing the older comment. So do you think it's still valid to keep the standalone executable comment or we are supposed to add what successfully runs ?

[thedevmystic]

If the older implementation of force_torque_sensor_broadcaster can run without specifying the tf_echo argument.

Then, I think we shouldn't throw an error if it is missing.

We should introduce tf_echo as an "option", not a requirement. To keep the backward compatibility, when tf_echo wasn't introduced.

We should make it optional, by providing a default value when none is provided, but currently it is a requirement. This will break the usage of old existing users.

Am I missing something, if so, feel free to inform me.

[vedh1234]

Thanks for the clarification!

Just to confirm — the previous implementation of wrench_transformer_node also required target_frames to be non-empty. The parameter had a not_empty<> validator, so launching the node without specifying target_frames already resulted in a parameter validation error.

So the node was never runnable without frames — only the documentation implied it.

This PR doesn’t change that requirement, it just adds an additional way (positional args) to provide the same frames. The parameter-based approach is still fully supported and backward compatible.

[thedevmystic]

Thanks for clarification.

But this means the not_empty<> is incorrect, as the original documentation explicitly said, it can be used as an standalone executable, without specifying the target frames. As also target frames is initialized as empty list (default value). I think it is a long run inconsistency.

Because, something that is forbidden to be empty, can't have a default value.

---

Orginal documentation:

The package provides a standalone ROS 2 node wrench_transformer_node that transforms wrench messages published by the ForceTorqueSensorBroadcaster controller to different target frames using TF2. This is useful when applications need force/torque data in coordinate frames other than the sensor frame.

The node subscribes to wrench messages from the broadcaster (either raw or filtered) and publishes transformed versions to separate topics for each target frame.

Usage
The wrench transformer node can be launched as a standalone executable:

ros2 run force_torque_sensor_broadcaster wrench_transformer_node

---

As we can see, we don't need to specify target frames, do we?

So, I think wrench_transformer_node was intended to be used without explicit target_frames. And tf_echo is a option, not an requirement.

So, i think we should remove the validation, as default value is provided, if it is empty.

force_torque_wrench_transformer:
  target_frames: {
    type: string_array,
    default_value: [],
    description: "...",
    # validation: {
    #   not_empty<>: null
    # }
  }

What do you say, @vedh1234?

[christophfroehlich]

what would be the usage of the wrench_transformer_node without target frames? that was not the original intention. The current state of the documentation fixes this, thanks!

[codecov]

Codecov Report

❌ Patch coverage is 95.65217% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 84.93%. Comparing base (36cfca8) to head (07857d5).

Files with missing lines	Patch %	Lines
...ensor_broadcaster/test/test_wrench_transformer.cpp	94.11%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2040      +/-   ##
==========================================
+ Coverage   84.91%   84.93%   +0.01%     
==========================================
  Files         148      148              
  Lines       14367    14389      +22     
  Branches     1230     1233       +3     
==========================================
+ Hits        12200    12221      +21     
  Misses       1740     1740              
- Partials      427      428       +1     

Flag	Coverage Δ
unittests	84.93% <95.65%> (+0.01%)	⬆️

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

Files with missing lines	Coverage Δ
...rque_sensor_broadcaster/src/wrench_transformer.cpp	80.00% <100.00%> (+1.33%)	⬆️
...ensor_broadcaster/test/test_wrench_transformer.cpp	91.43% <94.11%> (+0.16%)	⬆️

🚀 New features to boost your workflow:

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

[christophfroehlich]

Thanks a lot, LGTM!

[christophfroehlich]

what would be the usage of the wrench_transformer_node without target frames? that was not the original intention. The current state of the documentation fixes this, thanks!