* Writing proper documentation and unit tests is highly encouraged. PyQtGraph uses nose / pytest style testing, so tests should usually be included in a tests/ directory adjacent to the relevant code.
PyQtGraph developers are highly encouraged to (but not required) to use [`pre-commit`](https://pre-commit.com/). `pre-commit` does a number of checks when attempting to commit the code to ensure it conforms to various standards, such as `flake8`, utf-8 encoding pragma, line-ending fixers, and so on. If any of the checks fail, the commit will be rejected, and you will have the opportunity to make the necessary fixes before adding and committing a file again. This ensures that every commit made conforms to (most) of the styling standards that the library enforces; and you will most likely pass the code style checks by the CI.
To make use of `pre-commit`, have it available in your `$PATH` and run `pre-commit install` from the root directory of PyQtGraph.
As PyQtGraph supports a wide array of Qt-bindings, and python versions, we make use of `tox` to test against most of the configurations in our test matrix. As some of the qt-bindings are only installable via `conda`, `conda` needs to be in your `PATH`, and we utilize the `tox-conda` plugin.
* Tests for a module should ideally cover all code in that module, i.e., statement coverage should be at 100%.
* To measure the test coverage, un `pytest --cov -n 4` to run the test suite with coverage on 4 cores.
### Continous Integration
For our Continuous Integration, we utilize Azure Pipelines. On each OS, we test the following 6 configurations
* Python2.7 with PyQt4
* Python2.7 with PySide
* Python3.6 with PyQt5-5.9
* Python3.6 with PySide2-5.9
* Python3.7 with PyQt5-5.12
* Python3.7 with PySide2-5.12
More information on coverage and test failures can be found on the respective tabs of the [build results page](https://dev.azure.com/pyqtgraph/pyqtgraph/_build?definitionId=1)