Development tools library for python interpreter used for the FIRST Robotics Competition
pyfrc - RobotPy development library helper
pyfrc is a python 3 library designed to make developing code for RobotPy (the Python interpreter for the FIRST Robotics Competition) easier.
This library contains a few primary parts:
A built-in uploader that will upload your robot code to the robot
An implementation of wpilib that will run on your computer
This is a library designed to emulate parts of WPILib so you can more easily do unit testing of your robot code on any platform that supports python3, without having to have a cRio around for testing.
NOTE: This is not a complete implementation of WPILib. Add more things as needed, and submit patches! :)
Integration with the py.test testing tool to allow you to easily write unit tests for your robot code.
A robot simulator tool which allows you to run your code in (vaguely) real time and get simple feedback via a tk-based UI
using pip to install
The easiest installation is by using pip. On a linux/OSX system that has pip installed, just run the following command:
$ pip-3.2 install pyfrc
If you have python 3.3 installed, you may need to use ‘pip-3.3’ instead.
On Windows, I recommend using pip-Win to install packages. Download it from:
Once you’ve downloaded it, run it to install pip, and run the following command in its window:
pip install pyfrc
You must have the following python packages installed. Make sure that you install them for your python3 interpreter, as pyfrc only supports python 3.
- py.test (http://pytest.org/)
Once you have those installed, you can just install pyfrc the same way you would install most other python programs:
$ python3 setup.py install
code coverage support
If you wish to run code coverage testing, then you must install the following package:
- coverage (https://pypi.python.org/pypi/coverage)
It requires a compiler to install from source, so if you’re on Windows you probably just want to download the binary from pypi and install that, instead of trying to install from pip.
Once you modify your robot code, you can directly run your robot.py file and the pyfrc features will be enabled. You must modify your code slightly to make this work correctly.
Robot Code Modifications
There are a few modifications that you need to make to your robot.py to take advantage of the features provided by pyfrc.
- Your import statement must catch the wpilib import error and import wpilib from pyfrc instead.
- Your run() function must return the Robot object you create
- You must add a block that calls wpilib.run() at the bottom of your program
- You must define all of your motors and sensors inside of your robot class, and they cannot be global variables. This allows them to be reset each time a new test is created, as a new instance of your robot is created each time a test is run.
- import wpilib
- except ImportError:
- from pyfrc import wpilib
- def run():
robot = MyRobot() robot.StartCompetition()
- if __name__ == ‘__main__’:
Robot ‘physics model’
pyfrc now supports a simplistic custom physics model implementations for simulation and testing support. It can be as simple or complex as you want to make it. Hopefully in the future we will be adding helper functions to make this a lot easier to do.
The idea here is you provide a simulation object that overrides specific pieces of WPILib, and modifies motors/sensors accordingly depending on the state of the simulation. An example of this would be measuring a motor moving for a set period of time, and then changing a limit switch to turn on after that period of time. This can help you do more complex simulations of your robot code without too much extra effort.
By default, pyfrc doesn’t modify any of your inputs/outputs without being told to do so by your code or the simulation GUI.
See samples/physics for more details.
py.test unit testing integration support
pyfrc supports testing robot code using the py.test python testing tool.
See ‘samples/simple’ for an example test program that starts the robot code and runs it through autonomous mode and operator mode.
To run the unit tests, just run your robot.py with the following arguments:
$ python3 robot.py test
For more information on how to write py.test tests, see the documentation at http://pytest.org , or refer to the samples directory for examples.
If your test functions have any of the following arguments, then that argument will be an object as listed below:
- control: the wpilib.internal module
- fake_time: the module that controls time for wpilib, use Get() to retrieve the current simulation time
- robot: An instance of your robot class
- robot_file: the filename your robot code is started from
- robot_path: the directory that your robot is located
- wpilib: the wpilib module
code coverage for tests
pyfrc supports testing for code coverage using the coverage.py module. This feature can be used with any of the pyfrc commands and provide coverage information.
For example, to run the ‘test’ command to run unit tests:
$ python3 robot.py coverage test
Or to run coverage over the simulator:
$ python3 robot.py coverage sim
Running code coverage while the simulator is running is nice, because you don’t have to write unit tests to make sure that you’ve completely covered your code. Of course, you should write unit tests anyways… but this is good for developing code that needs to be run on the robot quickly and you need to make sure that you tested everything first.
When using the code coverage feature, what actually happens is robot.py gets executed again, except this time it is executed using the coverage module. This allows coverage.py to completely track code coverage, otherwise any modules that are imported by robot.py (and much of robot.py itself) would not be reported as covered.
Note: There is a py.test module called pytest-cov that is supposed to allow you to run code coverage tests. However, I’ve found that it doesn’t work particularly well for me, and doesn’t appear to be maintained anymore.
Note II: For some reason, when running the simulation under the code coverage tool, the output is buffered until the process exits. This does not happen under py.test, however. It’s not clear why this occurs.
The pyfrc robot simulator allows very simplistic simulation of your code in real time and displays the results in a (ugly) user interface. To run the simulator, run your robot.py with the following arguments:
$ python3 robot.py sim
If you wish to run so that your simulator can connect to the SmartDashboard, if you have pynetworktables installed you can run the following:
$ python3 robot.py netsim
Or you can use this instead:
$ python3 robot.py sim –enable-pynetworktables
As there is interest, I will add more features to the simulator. Please feel free to improve it and submit pull requests!
Adding custom tooltips to motors/sensors
If you move the mouse over the motors/sensors in the simulator user interface, you will notice that tooltips are shown which show which type of object is using the slot. pyfrc will now read the ‘label’ attribute from each object, and if present it will display that as the tooltip instead. For example:
motor = wpilib.Jaguar(1) motor.label = ‘whatzit motor’
I haven’t tested it on the cRio yet, but I believe that this should not affect operation on the robot, as RobotPy will just ignore the extra attribute.
The implementation of wpilib contained with pyfrc has a ‘fake’ implementation of SmartDashboard/NetworkTables within it. The simulator functionality can also use pynetworktables as the NetworkTables base when instructed.
The lib/pyfrc/wpilib directory is the code for wpilib directly copied from the RobotPy implementation. This code tries to load a module called ‘_wpilib’, which is a binary python module on the robot. However, in the directory lib/pyfrc/wpilib/_wpilib there is a python package which emulates a lot of the functionality found in the binary package for wpilib.
Contributing new changes
- Fork this git repository
- Create your feature branch (git checkout -b my-new-feature)
- Commit your changes (git commit -am ‘Add some feature’)
- Push to the branch (git push origin my-new-feature)
- Create new Pull Request on github