A pytest plugin for verifying alembic migrations.
A pytest plugin to test alembic migrations (with default tests) and which enables you to write tests specific to your migrations.
$ pip install pytest-alembic $ pytest --test-alembic ... ::pytest_alembic/tests/model_definitions_match_ddl <- . PASSED [ 25%] ::pytest_alembic/tests/single_head_revision <- . PASSED [ 50%] ::pytest_alembic/tests/up_down_consistency <- . PASSED [ 75%] ::pytest_alembic/tests/upgrade <- . PASSED [100%] ============================== 4 passed in 2.32s ===============================
Have you ever merged a change to your models and you forgot to generate a migration?
Have you ever written a migration only to realize that it fails when there’s data in the table?
Have you ever written a perfect migration only to merge it and later find out that someone else merged also merged a migration and your CD is now broken!?
pytest-alembic is meant to (with a little help) solve all these
problems and more. Note, due to a few different factors, there may
be some minimal required
however most of it is boilerplate akin to the setup required for alembic
Assert that there only exists one head revision.
We’re not sure what realistic scenario involves a diverging history to be desirable. We have only seen it be the result of uncaught merge conflicts resulting in a diverged history, which lazily breaks during deployment.
Assert that the revision history can be run through from base to head.
Assert that the state of the migrations matches the state of the models describing the DDL.
In general, the set of migrations in the history should coalesce into DDL which is described by the current set of models. Therefore, a call to
revision --autogenerateshould always generate an empty migration (e.g. find no difference between your database (i.e. migrations history) and your models).
Assert that all downgrades succeed.
While downgrading may not be lossless operation data-wise, there’s a theory of database migrations that says that the revisions in existence for a database should be able to go from an entirely blank schema to the finished product, and back again.
Let us know if you have any ideas for more built-in tests which would be generally useful for most alembic histories!
Sometimes when writing a particularly knarly data migration, it helps to be able to practice a little timely TDD, since there’s always the potential you’ll trash your actual production data.
pytest-alembic, you can write tests directly, in the same way
that you would normally, through the use of the
def test_knarly_migration_xyz123(alembic_engine, alembic_runner): # Migrate up to, but not including this new migration alembic_runner.migrate_up_before('xyz123') # Perform some very specific data setup, because this migration is sooooo complex. # ... alembic_engine.execute(table.insert(id=1, name='foo')) alembic_runner.migrate_up_one()
alembic_runner has a number of methods designed to make it convenient
to change the state of your database up, down, and all around.
pip install "pytest-alembic"
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
|Filename, size||File type||Python version||Upload date||Hashes|
|Filename, size pytest_alembic-0.3.1-py3-none-any.whl (17.7 kB)||File type Wheel||Python version py3||Upload date||Hashes View|
|Filename, size pytest-alembic-0.3.1.tar.gz (14.6 kB)||File type Source||Python version None||Upload date||Hashes View|
Hashes for pytest_alembic-0.3.1-py3-none-any.whl