pyrf24 0.6.0

A python package wrapping the nRF24 C++ libraries.

Introduction

This is the official home of the python wrappers for the RF24 stack. It is meant for Linux-based SoC boards like the Raspberry Pi. Documentation is hosted at http://pyrf24.rtfd.io/.

Pinout

The nRF24L01’s CE and IRQ pins can be connected to other GPIO pins on the SoC. The MISO, MOSI, SCK are limited to the corresponding counterparts on the SoC’s SPI bus. The CSN pin is limited to the chosen SPI bus’s “Chip Select” options (also labeled as “CE” pins on many Raspberry Pi pinout diagrams). The following table shows the default pins used in all the examples for this package.

nRF24L01	Raspberry Pi
GND	GND
VCC	3V
CE	GPIO22
CSN	GPIO8 (CE0)
SCK	GPIO11 (SCK)
MOSI	GPIO10 (MOSI)
MISO	GPIO9 (MISO)
IRQ	GPIO24

The IRQ pin is not typically connected, and it is only used in the interrupt_configure example.

Warning

If connecting a nRF24L01+PA+LNA module to the Raspberry Pi, you MUST use a external 3V power supply because the Raspberry Pi (all models) do not provide enough power for the nRF24L01+PA+LNA modules.

Important

It is highly recommended that the nRF24L01’s VCC and GND pins have a parallel capacitor to stabilize the power supply. Usually 100 microfarad is enough, but the capacitance ultimately depends on the nature of your power supply’s stability.

Note

Notice that gpiod is used to watch the radio’s IRQ pin (exclusively during the interrupt_configure.py example).

gpiod is not required for normal usage (when not using the radio’s IRQ pin).

sudo apt install python3-dev
pip install gpiod

Installing from PyPI

Simply use:

python -m pip install pyrf24

We have distributed binary wheels to pypi.org for easy installation and automated dependency. These wheels specifically target any Linux platform on aarch64, armv7l, and x86_64 architectures.

Note

If you’re installing from a Linux machine that is not using an architecture aarch64, armv7l, or x86_64, then pip may try to build the package from source code. In this case, you’ll likely need to install some extra build dependencies:

sudo apt install python3-dev cmake

Installing from Github

Installing from source will require CMake and CPython headers:

sudo apt install python3-dev cmake

To build this python package locally, you need to have cloned this library’s repository with its submodules.

git clone --recurse-submodules https://github.com/nRF24/pyRF24.git
cd pyRF24
python -m pip install . -v

Hint

For consecutive build attempts, it is recommended to delete any previous build artifacts before attempting to build again.

rm -r build/ dist/

Note

The -v is optional. Here, we use it to show that pip isn’t frozen during the build process.

Installing the package can take a long time, and you might think that pip is frozen on the step labeled “Building wheel for pyrf24 (pyproject.toml)”. Just wait for about 5 minutes (maybe longer on older/slower variants of Raspberry Pi).

Building a wheel

Building a somewhat portable binary distribution for python packages involves building a .whl file known as a wheel. This wheel can be used to install the pyrf24 package on systems using the same version of CPython, CPU architecture, and C standard lib.

1. If building wheels is not done in an isolated build environment, it is advised that some build-time dependencies be installed manually to ensure up-to-date stable releases are used. Execute the following from the root directory of this repo:

   python -m pip install -r requirements-build.txt

   Note

   This step only needs to be done once.

2. Using the same directory that you cloned the pyrf24 library into:

   python -m pip wheel -w dist .

   Important

   It is recommended to purge any previous build artifacts before re-building the package.

   rm -r build/ dist/

3. To install a built wheel, simply pass the wheel’s path and file name to pip install:

   python -m pip install dist/pyrf24-MAJOR.MINOR.PATCH-cp3X-cp3X-linux_ARCH.whl

   Where the following would be replaced accordingly:

   • MAJOR.MINOR.PATCH is the current version of the pyrf24 package.

     • If not building a tagged commit, then the version will describe the commit relative to the number of commits since the latest tag. For example, 0.1.1.post1.dev3 is the third commit (dev3) since the first “post release” (post1) after the tagged version 0.1.1. This adhere’s to PEP440.

   • cp3X is the version of python used to build the wheel (ie cp39 for CPython 3.9) The second occurrence of cp3X describes the CPython ABI compatibility.

   • ARCH is the architecture type of the CPU. This corresponds to the compiler used. On Raspberry Pi OS (32 bit), this will be armv7l.

Using a specific RF24 driver

By default, this package is built using the RF24 driver SPIDEV. If you want to build the package using a different RF24 driver (like RPi, MRAA, wiringPi, or pigpio), then it is necessary to use an environment variable containing additional arguments for CMake:

export CMAKE_ARGS="-DRF24_DRIVER=RPi"

Hint

You can also use this environment variable to enable debug output from different layers of the RF24 stack. For a list of supported options, look at the script in this repository’s cmake/using_flags.cmake.

The following value will turn on debug output for the RF24Mesh and RF24Network classes (respectively).

export CMAKE_ARGS="-DRF24MESH_DEBUG=ON -DRF24NETWORK_DEBUG=ON"

For RPi clones that use a different GPIO chip number, the CMAKE_ARGS can be used to adjust this too. The default GPIO chip number is set to /dev/gpiochip0.

export CMAKE_ARGS="-DRF24_LINUX_GPIO_CHIP=/dev/gpiochip4"

Then just build and install the package from source as usual.

python -m pip install . -v

Differences in API

This package intentionally adheres to PEP8 standards as much as possible. This means that class members’ names use snake casing (eg. get_dynamic_payload_size()) instead of using the C++ conventional camel casing (eg. getDynamicPayloadSize()). However, the older python wrappers provided with each C++ library (RF24, RF24Network, & RF24Mesh) had used camel casing. So, the API provided by this package exposes both snake cased and camel cased versions of the API. The camel cased API is not documented to avoid duplicate and complicated documentation.

radio.print_details()  # documented
# can also be invoked as
radio.printDetails()  # not documented

Some of the C++ functions that do not accept arguments are wrapped as a class property. But, the C++ style functions are still exposed. For example:

radio.listen = False
# is equivalent to
radio.stopListening()  # not documented

radio.listen = True
# is equivalent to
radio.startListening()  # not documented

Migrating to pyrf24

If you have a project that uses code from the older individually installed wrappers, then you can use this package as a drop-in replacement. You only need to change the import statements in your project’s source. Everything from the old individual wrappers is exposed through the pyrf24 package.

Using the old individual wrappers	Using the pyrf24 package
from RF24 import RF24, RF24_PA_LOW	from pyrf24 import RF24, RF24_PA_LOW
from RF24 import RF24 from RF24Network import RF24Network, RF24NetworkHeader	from pyrf24 import RF24, RF24Network, RF24NetworkHeader
from RF24 import RF24 from RF24Network import RF24Network from RF24Mesh import RF24Mesh	from pyrf24 import RF24, RF24Network, RF24Mesh

Python Type Hints

This package is designed to only function on Linux devices. But, it is possible to install this package on non-Linux devices to get the stub files which help auto-completion and type checking in various development environments.

Documentation

Each release has corresponding documentation hosted at http://pyrf24.rtfd.io/.

Before submitting contributions, you should make sure that any documentation changes build successfully. This can be done locally but on Linux only. The documentation of API requires this package (& all its latest changes) be installed.

This package’s documentation is built with the python package Sphinx and the sphinx-immaterial theme. It also uses the dot tool provided by the graphviz software to generate graphs.

1. Install Graphviz

   sudo apt-get install graphviz

2. Installing Sphinx necessities

   Note

   If you installed sphinx using apt, then it is likely out-of-date and will override any virtual python environments installation of Sphinx. Simply uninstall sphinx (using apt) will remedy this problem.

   python -m pip install -r docs/requirements.txt

   Important

   If pip outputs a warning about your path/to/Python/Python3x/Scripts folder not added to your OS environment variable PATH, then you will likely get an error message like sphinx-build command not found when building the documentation. For more information on installing sphinx, see the official Sphinx install instructions.

   Warning

   This documentation’s theme requires Sphinx v4.0+. So, it is not recommended to install sphinx from apt on Linux because the version distributed with the OS’s PPA repository may not be the most recent version of Sphinx.

3. Building the Documentation

   To build the documentation locally, the pyrf24 package needs to be installed first. Then run:

   cd docs
   sphinx-build -E -W . _build

   The docs/_build folder should now contain the html files that would be hosted on deployment. Direct your internet browser to the html files in this folder to make sure your changes have been rendered correctly.

   Note

   The flags -E and -W will ensure the docs fail to build on any error or warning (just like it does when deploying the docs online).