Micropy Cli is a project management/generation tool for writing Micropython code in modern IDEs such as VSCode. Its primary goal is to automate the process of creating a workspace complete with:
You can download and install the latest version of this software from the Python package index (PyPI) as follows:
pip install --upgrade micropy-cli
If applicable, you can test out a pre-release by executing:
pip install --upgrade --pre micropy-cli
Creating a new project folder is as simple as:
micropy init <PROJECT NAME>
When creating a project with micropy-cli
, two special items are added:
.micropy/
foldermicropy.json
fileThe .micropy/
contains symlinks from your project to your $HOME/.micropy/stubs
folder. By doing this, micropy can reference the required stub files for your project as relative to it, rather than using absolute paths to $HOME/.micropy
. How does this benefit you? Thanks to this feature, you can feel free to push common setting files such as settings.json
and .pylint.rc
to your remote git repository. This way, others who clone your repo can achieve a matching workspace in their local environment.
Note: The generated
.micropy/
folder should be IGNORED by your VCS. It is created locally for each environment via themicropy.json
file.
The micropy.json
file contains information micropy needs in order to resolve your projects required files when other clone your repo. Think of it as a package.json
for micropython.
To setup a Micropy environment locally, simply:
micropy-cli
micropy
Micropy will automatically configure and install any stubs required by a project thanks to its micropy.json
file.
While all modules that are included in your targeted micropython firmware are available with autocompletion, intellisense, and linting, most projects require external dependencies.
Currently, handling dependencies with micropython is a bit tricky. Maybe you can install a cpython version of your requirement? Maybe you could just copy and paste it? What if it needs to be frozen?
Micropy handles all these issues for you automatically. Not only does it track your project's dependencies, it keeps both requirements.txt
and dev-requirements.txt
updated, enables autocompletion/intellisense for each dep, and allows you to import them just as you would on your device.
This allows you to include your requirement however you want, whether that be as a frozen module in your custom built firmware, or simply in the /lib
folder on your device.
To add a package as a requirement for your project, run:
micropy install <PACKAGE_NAMES>
while in your project's root directory.
This will automatically execute the following:
PACKAGE_NAMES
from pypi, as a url, or a local path.micropy
folder.micropy.json
requirements.txt
To install dev packages that are not needed on your device, but are needed for local development, add the --dev
flag. This will do everything above except stub the requirement.
You can also install all requirements found in micropy.json
/requirements.txt
/dev-requirements.txt
by executing micropy install
without passing any packages. Micropy will automatically do this when setting up a local environment of an existing micropy project.
Lets say your new project will depend on picoweb and blynklib. Plus, you'd like to use rshell to communicate directly with your device. After creating your project via micropy init
, you can install your requirements as so:
Now you or anybody cloning your project can import those requirements normally, and have the benefits of all the features micropy brings:
Stub files are the magic behind how micropy allows features such as linting, Intellisense, and autocompletion to work. To achieve the best results with MicropyCli, its important that you first add the appropriate stubs for the device/firmware your project uses.
Note: When working in a micropy project, all stub related commands will also be executed on the active project. (i.e if in a project and you run
micropy stubs add <stub-name>
, then that stub retrieved AND added to the active project.)
Adding stubs to Micropy is a breeze. Simply run: micropy stubs add <STUB_NAME>
By sourcing micropy-stubs, MicroPy has several premade stub packages to choose from.
These packages generally use the following naming schema:
<device>-<firmware>-<version>
For example, running micropy stubs add esp32-micropython-1.11.0
will install the following:
You can search stubs that are made available to Micropy via micropy stubs search <QUERY>
Alternatively, using micropy stubs add <PATH>
, you can manually add stubs to Micropy.
For manual stub generation, please see Josvel/micropython-stubber.
Using micropy stubs create <PORT/IP_ADDRESS>
, MicropyCli can automatically generate and add stubs from any Micropython device you have on hand. This can be done over both USB and WiFi.
Note: For stub creation, micropy-cli has additional dependencies.
These can be installed by executing:
pip install micropy-cli[create_stubs]
To list stubs you have installed, simply run micropy stubs list
.
To search for stubs for your device, use micropy stubs search <QUERY>
.
VSCode IntelliSense, Autocompletion & Linting capabilities
micropy-cli
and pymakr-vsc. Great place to start if you're new to this!Developing for the Raspberry Pi Pico in VS Code
micropy-cli
along with several other great development tools under the Development category.Josverl's Repo is full of information regarding Micropython compatibility with VSCode and more. To find out more about how this process works, take a look at it.
micropy-cli and micropy-stubs depend on micropython-stubber for its ability to generate frozen modules, create stubs on a pyboard, and more.