* Bump python to 3.7.10 to try to fix issues with typing module
* Remove requirement on typing again
* Remove bigquery and activedata references (#7051)
* Remove activedata and bigquery again
* Add importlib-metadata to dev requirements
* Remove remnants of selenium tests from setup.cfg
* Don't call taskcluster.aio.createSession outside of async functions (it's not allowed)
* Disable newrelic linters to fix issue with imp module
See https://discuss.newrelic.com/t/python-warnings-during-pytest/114897
* Fix linting errors
* Use pytest-xdist to speed up tests
* Properly ignore newrelic warnings
* Disable django.contrib.staticfiles in tests, because it breaks everything
* Set TREEHERDER_DEBUG=False for unit tests, because it breaks some tests
* Fix linting issues
*Create new Tox environment for linting checks
* Move production environment variables into Heroku's tox environment
* Call `tox -e docker` instead of custom command
* Upload code coverage file
* Run JS tests before trying to submit coverage results
* Upgrade very old pip package
Some changes in here:
* Making tox a bit more the centralized point for configurying job execution
* Move Heroku code paths into tox from .travis.yml
* Move to more modern Travis image (xenial 16.04 -> bionic 18.04)
* Make Python tests outside of Docker to run slow tests (DB required)
* Cache tox packages to speed up following Travis runs
* Remove coverage data before gathering new data (to prevent using old data)
This makes sure that serving the documentation via poetry does not regress
and we have a single way of generating them.
Setting the build backend in `pyproject.toml` makes `pip` to use `poetry` to install the dependencies. By doing so, readthedocs will be able to install the required dependencies to generate the docs (since they don't support poetry directly).
Ben Hearsum (he/him)
Armen Zambrano G
Cameron Dawson
Armen Zambrano
Kyle Lahnakoski
Shubhank Saxena
Suyash Salampuria