This tool is a command-line interface (CLI). It enables superior communication with any embedded device. It was designed for:
- fast prototyping and debugging. Set everything up in a few minutes and start debugging any embedded device efficiently. Forget about
printf
. Forever. - communication-based applications. Stop re-writing custom protocols for each new project.
- real-time update of embedded application parameters. Tune your application without loosing time compiling & flashing just for parameter tuning.
- plot data from the device in real-time. Standard linear data is supported, but also arrays, sparse arrays. In the future, also Matrices, XYZ, and RGB-type codes.
- Reusability. The protocol is highly flexible, loosely coupled to your application. It can be used in a wide number of application scenarios.
This CLI provides a set of commands to connect to a device, read, plot, write data on it, log any received and sent data.
In the future: export to Excel and CSV and replay command in the CLI for offline inspection.
The communication protocol that carry all exchanged information is implemented in Python and C:
pytelemetry
for scripting the communication from your PCtelemetry
: for enabling communication in the embedded device.
Officially supported embedded platforms are for now Arduino
and Mbed
.
See the central documentation for installation instructions, tutorials, description of the protocol, etc.
Aan example of listing serial ports ls -s
, connecting to a device through COM20 serial com20 --bauds 115200
, listing all received topics ls
and opening a plot on topic touch plot touch
The command line interface can be started like this
python3 -m pytelemetrycli.cli
If everything is installed properly, :>
should welcome you.
pytelemetry terminal started. (type help for a list of commands.)
:> _
Without arguments, you get a list of all available commands. Otherwise the full command
documentation.
Without options, prints a list of all received topics.
With the --serial flag, prints a list of all available COM ports
Usage: ls [options]
Options:
-s, --serial Use this flag to print a list of all available serial ports
Connects pytelemetry to the serial port.
Usage: serial <port> [options]
Options:
-b X, --bauds X Connection speed in bauds [default: 9600]
Prints X last received samples from <topic>.
Usage: print <topic> [options]
Options:
-a X, --amount X Amount of samples to display [default: 1]
Publishes a (value | string) on <topic>.
Usage: pub (--u8 | --u16 | --u32 | --i8 | --i16 | --i32 | --f32 | --s) <topic> <value>
Plots <topic> in a graph window.
Usage: plot <topic>
Disconnects from any open connection.
Usage: disconnect
Exits the terminal application.
Usage: quit
pytelemetrycli
requires python 3.3+, PyQt4 and numpy.
It is recommended to download numpy
and PyQt4
wheels python packages (courtesy of Christoph Gohlke).
In case you were wondering, no you don't have to install Qt. The binary wheel is enough.
Then install with pip the downloaded files
pip install numpy-x.xx.x+vanilla-cp3x-none-winxxx.whl
pip install PyQt4-x.xx.x-cp3x-none-winxxx.whl
Then, simply install pytelemetrycli
with pip as usual
pip install pytelemetrycli
The easiest way to install numpy and PyQt4 seem to be using homebrew
.
lease note that you should also have installed python 3.5 with homebrew for this to work correctly.
Also, avoid to have another python 3.5 distribution on your system otherwise you will face import issues as well.
brew install python3
brew install pyqt --with-python3
pip3 install pytelemetrycli
The setup used for testing relies on miniconda.
conda install numpy
conda install pyqt
conda install pip
pip install pytelemetrycli
However, if you have PyQt4 and numpy already installed in your directory, simply run
pip install pytelemetrycli