Contributions are very welcome, I will gladly review and discuss any merge requests. If you have questions about the code and architecture, feel free to open an issue. This file should give an overview about some internals of input-remapper.
All pull requests will at some point require unittests (see below for more info). The code coverage may only be improved, not decreased. It also has to be mostly compliant with pylint.
To quickly restart input-remapper without pkexec prompts, you can use
sudo pkill -f input-remapper && sudo input-remapper-reader-service -d & sudo input-remapper-service -d & input-remapper-gtk -dmypy inputremapper # find typing issues
black . # auto-format all code in-place
pip install pylint-pydantic --user # https://github.com/fcfangcc/pylint-pydantic
pylint inputremapper # get a code quality rating from pylintPylint gives lots of great advice on how to write better python code and even detects errors. Mypy checks for typing errors. Use black to format it.
You should be able to use your IDEs built in python unittest features to run tests. But you can also run them from your console:
pip install psutil # https://github.com/giampaolo/psutil
pip install -e .
sudo pkill -f input-remapper
python3 -m unittest discover -s ./tests/This assumes you are using your system's pip. If you are in a virtual env,
a sudo pip install is not recommended. See Scripts for alternatives.
python -m unittest tests/unit/test_daemon.py
python -m unittest tests.unit.test_ipc.TestPipe -k "test_pipe" -f
# See `python -m unittest -h` for more.
Don't use your computer during integration tests to avoid interacting with the gui, which might make tests fail.
To read events for manual testing, evtest is very helpful.
Add -d to input-remapper-gtk to get debug output.
Tests are in https://github.com/sezanzeb/input-remapper/tree/main/tests
https://github.com/sezanzeb/input-remapper/blob/main/tests/test.py patches some modules
and runs tests. The tests need patches because every environment that runs them will be
different. By using patches they all look the same to the individual tests. Some
patches also allow to make some handy assertions, like the write_history of UInput.
Test files are usually named after the module they are in.
In the tearDown functions, usually one of quick_cleanup or cleanup should be called.
This avoids making a test fail that comes after your new test, because some state
variables might still be modified by yours.
To automate some of the development tasks, you can use the
setup.sh script. The script avoids using pip for installation.
Instead, it uses either your local python3 in your virtual env, or using
/usr/bin/python3 explicitly. For more information run
scripts/setup.sh help
Do not use GTKs foreach methods, because when the function fails it just freezes up
completely. Use get_children() and iterate over it with regular python for loops.
Use gtk_iteration() in tests when interacting with GTK methods to trigger events to
be emitted.
Do not do from evdev import list_devices; list_devices(), and instead do
import evdev; evdev.list_devices(). The first variant cannot be easily patched in
tests (there are ways, but as far as I can tell it has to be configured individually
for each source-file/module). The second option allows for patches to be defiend in
one central places. Importing KEY_*, BTN_*, etc. constants via from evdev is
fine.
ssh/login into a debian/ubuntu environment
scripts/build.shThis will generate input-remapper/deb/input-remapper-2.0.1.deb
# https://github.com/nedbat/coveragepy https://github.com/giampaolo/psutil
pip install coverage anybadge pylint psutil
sudo pkill -f input-remapper
# Make sure input-remapper is uninstalled, then install it editable (without sudo
# should be fine), so that the path for the coverage collection is correct.
# Use `find /usr/ -iname "*inputremapper*"` to check if it is uninstalled.
pip install -e .
./scripts/badges.shNew badges, if needed, will be created in readme/ and they just need to be commited.
To regenerate the po/input-remapper.pot file, run
xgettext -k --keyword=translatable --sort-output -o po/input-remapper.pot data/input-remapper.glade
xgettext --keyword=_ -L Python --sort-output -jo po/input-remapper.pot inputremapper/configs/mapping.py inputremapper/gui/*.py inputremapper/gui/components/*.pyThis is the template file that you can copy to fill in the translations. Also create a
corresponding symlink, like ln -s it_IT.po it.po, because some environments expect
different names, apparently. See https://github.com/sezanzeb/input-remapper/tree/main/po
for examples.
There is a miro board describing input-remappers architecture:
https://miro.com/app/board/uXjVPLa8ilM=/?share_link_id=272180986764
