Running Test Suite

Running the Automatic Tests

XPlane2Blender comes with several hundred automated tests to cover a wide variety of export capabilities. The test suite runs the exporter inside Blender in headless mode and checks output results.

The outer script that runs the tests is the Python3 script tests.py - it can be run with -h to see all command line arguments. The test has a few requirements to run properly:

1. It needs a version of Blender 2.8x to run against. If the command line tool 'blender' in the current search path isn't a 2.8x Blender, you can use --blender to specify an install path. For example, on OS X you might do

python3 tests.py --blender /Applications/Blender.app/Contents/MacOS/blender -q -p

on Linux, the simplest thing can be to untar a tarball of the app into your home directory, e.g.

python3 tests.py --blender ../blender-2.83.19-linux-x64/blender -c -q

2. The sim-link addons/modules in the git repo needs to point back to the root of the git repo - see addons/addons.README for the gory details. If the simlink isn't working (E.g. because your GIT client didn't set up the unix flags properly), the script won't work.

Useful Flags:

• -q quiets output to something reasonable.
• -c continue after a test failure, useful for getting a summary of all failures.

Open Issues:

All of the above assumes a unix environment and a functioning python3. TODO: running the tests on Windows.

It would be useful to be able to run the add-on without needing an addons/modules directory, but I haven't found a way to do that with Blender.