Developer notes

Pre-commit hook

Install our pre-commit hook as follows:

pip install pre-commit
pre-commit install

This will check your changes before each commit to ensure that they conform with our code style standards. We use black to reformat Python code, and PrettierJS for JavaScript. We also run ESLint to catch several common Redux usage bugs.

Testing

To execute the test suite:

  • Install geckodriver and Firefox

  • To run all tests: make test

  • To run a single test: ./baselayer/tools/test_frontend.py skyportal/tests/frontend/<test_file>.py::test_<specific_test>

This fires up the test server and runs that test. To make things faster you can also run the test server separately:

make run_testing   # fire up the server with `test_conf.yaml`

pytest skyportal/tests/api  # test api
pytest skyportal/tests/frontend/test_user.py  # run subset of frontend tests
pytest skyportal/tests/frontend/test_user.py::test_user_profile_fetching  # run a single test

On Linux, the tests can be run in “headless” mode (no browser display):

  • Install xvfb (sudo apt-get install xvfb)

  • make test_headless

Or, as described above, launch the test server with make run_testing, and then call

xvfb-run pytest skyportal/tests/frontend/test_user.py

It is not noticeably faster to run the test suite headlessly.

Test fixtures

The SkyPortal test suite leverages pytest fixtures to generate and isolate test data for each test. Fixtures are declared in the skyportal/tests/conftest.py file, and they utilize factories implemented in skyportal/tests/fixtures.py to generate new instances of fixtures as needed.

Each test will request any fixtures that it requires, and as part of the set-up for that test, new test instances of the fixtures are created. This means that each test is guaranteed to work with fresh, isolated test data in order to avoid any unintended dependencies on other tests. Fixtures should be used when possible in writing new tests to promote safe, repeatable test results.

SkyPortal test fixtures are implemented with teardown logic, such that any database records generated for that fixture are deleted from the test database upon completion of a test. This serves to avoid bloating of the test database as more tests are run. Note that each fixture defined in conftest.py uses the yield keyword to return its test object. Any code following yield in a pytest fixture is run as teardown logic when that fixture goes out of scope; you will find that in SkyPortal this involves calling the relevant teardown() functions on the fixture objects created. These teardown() functions are implemented for each SQLAlchemyModelFactory subclass defined in fixtures.py.

Any new fixtures and fixture factories being added should follow this same structure, taking care to delete all fixture data (including SubFactory-generated records, i.e. a new User created to be the author of a new Comment). Foreign key constraints and cascade behavior defined on the database models can be used to simplify this logic but can also lead to subtle errors in tearing down fixtures and should be taken into careful consideration.

Debugging

  • Run make log to watch log output

  • Run make stop to stop any running web services.

  • Run make attach to attach to output of webserver, e.g. for use with pdb.set_trace()

  • Run make check-js-updates to see which Javascript packages are eligible for an upgrade.

Database

All interactions with the database are performed by way of SQLAlchemy using the Pyscopg2 backend. Some standard but not necessarily obvious usage patterns we make use of include:

  • Logic for connecting to the DB, refreshing tables, etc. is found in baselayer/model_utils.py:

from baselayer.app.env import load_env
from skyportal.models import DBSession, init_db
env, cfg = load_env()
init_db(**cfg['database'])

  • The session object controls various DB state operations:

DBSession().add(obj)  # add a new object into the DB
DBSession().commit()  # commit modifications to objects
DBSession().rollback()  # recover after a DB error

For more information on how to use the session, see baselayer/doc/dev.md.

  • Generic logic applicable to any model is included in the base model class baselayer.app.models.Base (to_dict, __str__, etc.), but can be overridden within a specific model

  • Models can be queried directly (User.query.all()), or more specific queries can be constructed via the session object (DBSession().query(User.id).all())

  • A more modern way to interact with the models is using select statements: Obj.select(current_user_id).where(Obj.id == 'ZTF20aahxjps')

  • To get the actual objects create a session (either using self.Session() on the handler or using DBSession() which does not apply permissioning). Then use the session: session.execute(stmt).all() or session.scalars(stmt).all(). The scalars is preferred when selecting whole objects or single columns. The execute is better for getting multiple columns as tuples.

  • Best practice is to call the session as a context manager (i.e., with self.Session() as session:) but only if the session does not need to stay open.

  • Convenience functionality:

    • Join relationships: some multi-step relationships are defined through joins using the secondary parameter to eliminate queries from the intermediate table; e.g., User.acls instad of [r.acls for r in User.roles]

    • Association proxies: shortcut to some attribute of a related object; e.g., User.permissions instead of [a.id for a in User.acls]

    • Joined loads: this allows for a single query to also include child/related objects; often used in handlers when we know that information about related objects will also be needed.

    • to_json(): often from a handler we return an ORM object, which gets converted to JSON via json_util.to_json(obj.to_dict()). This also includes the attributes of any children that were loaded via joinedload or by accessing them directly. For example:

      • User.query.first().to_dict() will not contain information about the user’s permissions

      • u = User.query.first(); u.acls; u.to_dict() does include a list of the user’s ACLs

Docker images

Run make docker-images to build and push to Docker hub.

Debugging GitHub Actions

You can insert the following step to debug your workflow:

- name: Setup tmate session
  uses: mxschmitt/action-tmate@v2

It will print a command that you can use to SSH into the runner.