pytest-bq 0.0.3

BigQuery fixtures and fixture factories for Pytest.

Pytest BQ

This is a pytest plugin in similar vein to pytest-postgres and pytest-kafka. It runs a local version of bigquery that you can use throughout your test suite.

This would have been much more painful without Mirakuru and bigquery-emulator; this is a simple wrapper around those tools.

Installation

This tool requires you to have a copy of the bigquery-emulator binary somewhere on your path. Depending upon your architecture you'll need a different version of the tool.

wget https://github.com/goccy/bigquery-emulator/releases/download/v0.6.1/bigquery-emulator-linux-amd64
mv bigquery-emulator-linux-amd64 /usr/local/bin/bigquery-emulator

To install this library:

pip install pytest-bq

Demo

# conftest.py
from pytest_bq.factories import client as bq_client
from pytest_bq.factories import proc as bq_process

# Create a process and a local client that targets that process.
bq_proc = bq_process.bq_proc(executable='/usr/local/bin/bigquery-emulator', project_id='test')
bqlocal = bq_client.bqlocal("bq_proc")

# tests/test_bq.py
from google.cloud import bigquery
from pytest_bq.executor.process import BQExecutor


def test_can_list_datasets(bq_proc: BQExecutor, bqlocal: bigquery.Client) -> None:
    """MVP to ensure everything works."""
    test_dataset = "test_base"
    bqlocal.create_dataset(test_dataset)
    datasets = [x.full_dataset_id for x in bqlocal.list_datasets()]

    assert test_dataset in datasets

Contributing

PRs are accepted.

# Install the dependencies with:
pip install .[test]
# Install pre-commit hooks.
pre-commit install
# Validate everything passes.
pre-commit run --all
# Run the tests.
pytest tests/

TODOs

• Validate data_from_yaml files.
• Tools for dummy data generation.