This repo shows how to work with Micropython on an ESP32 device. Features: * it uses Microsoft VSCode. It makes use of the VSCode remote container extension. * The extension connects to a Docker container in which all the ESP32 tooling has been installed. The tooling has been compiled from source. * Various utilities have been installed (mpfshell, micropython-cli, esptool) * An example of a custom C++ module is added to Micropython.
In the Docker image the following tools have been installed:
1. xtensa-esp32 toolchain (compiled from source!) in /esp/crosstool-NG
2. esp-idf (freetos ESP32 library) in /esp/esp-idf
3. micropython in /root/micropython
1. Micropython-wrap in /root/micropython/extmode/micropython-wrap
4. esptool (allows to erase flash)
5. mpfshell (allows to put files on the ESP32 flash, start the REPL)
6. micropy-cli (allows the setup micropython projects including intellisense and pylinting)
The ex_bash script in the .devcontainer folder can be used to start the container. The ex_bash script starts the container with 'privileged' permissions in order to have access the USB port. Furthermore it mounts the users home dir in the container, so you will have access to you files. Lastly it will correct the Docker user permissions which will ensure that file permissions are handled correctly.
An example custom c++ module is setup in /root/micropython/extmod/custommodule. The C++ code uses the micropython-wrap library to reduce the amount of boilerplate code. The custommodule is registered in the micropython main code (see micropython/ports/esp32/main.c)
When modifying the code in custommodule.cpp you can run the following to recompile the Micropython firmware:
cd /root/micropython/ports/esp32 && ESPIDF=$IDF_PATH make -j4 -f lolin32_makefile
1. Install VSCode and Docker
2. Install the remote docker extension in VSCode
3. Go into the .devcontainer folder of this repo
There are two version available. One based on ESP IDF v3.3 and one based on ESP IDF v4.0beta1.
3. Run "make pull_hub" to get the EPS32 tooling image from Docker hub, or run "make image" to locally build the docker image containing the tooling (this will take quite a long time!)
4. First plug in the ESP32 device (otherwise /dev/ttyUSB won't be available in the container).
5. Go into the .devcontainer folder of this repo and run ./ex_bash (this creates the container and mounts your home dir inside the container)
3. Run "make pull_hub_v4" to get the EPS32 tooling image from Docker hub, or run "make image_v4" to locally build the docker image containing the tooling (this will take quite a long time!)
4. First plug in the ESP32 device (otherwise /dev/ttyUSB won't be available in the container).
5. Go into the .devcontainer folder of this repo and run
<pre>SOURCE_IMAGE="planetbosman/esp32_vscode_v4" ./ex_bash</pre>
This creates the container and mounts your home dir inside the container.
6. Add the following to: ~/.config/Code/User/globalStorage/ms-vscode-remote.remote-containers/imageConfigs/planetbosman%2fesp32_vscode.json <BR>
Note replace workspacefolder path to the path of the example project in this repo.
{ "remoteUser": "vscode", "extensions": [ "ms-python.python", "junhuanchen.mpfshell" ], "workspaceFolder": "/home/dinne/vscode_esp32/example_project" }
7. In vscode choose ctrl-shift+P: Attach to running container
8. Ctrl+Shit+P: Run the command "Python: Select interpreter (python 3), select python Linter
9. open a terminal. Run micropython (this generates the import stubs)
- Use esptool to erase flash
esptool.py --chip esp32 --port /dev/ttyUSB0 erase_flash
10. To flash the micropython firmware open a VScode terminal and: i
cd /root/micropython/ports/esp32 && sudo make deploy
11. open folder "example_project". Open a terminal and upload various files
mpfs -n --nohelp -c "open ttyUSB0; put ssd1306.py; put main.py;"
12. Reset the device
esptool.py --chip esp32 --port /dev/ttyUSB0 --after hard_reset chip_id
13. Start a REPL (the REPL can be quit with CTRL+])
mpfs -n --nohelp -c "open ttyUSB0; r;"