Skip to content

Quantco/datajudge

Repository files navigation

datajudge

CI Documentation Status Conda-forge PypiVersion codecov.io

Express and test specifications against data from database.

Documentation

Supported backends

PostgreSQL Microsoft SQL Server Snowflake Google BigQuery DuckDB

Usage

datajudge can either be installed via pypi with pip install datajudge or via conda-forge with conda install datajudge -c conda-forge.

Please refer to the Getting Started section of our documentation for details.

Expressing an expectations between different tables from a database may look as such:

from datajudge import BetweenRequirement

companies_between_req = BetweenRequirement.from_tables(
    db_name1="example",
    table_name1="companies",
    db_name2="example",
    table_name2="companies_archive",
)

companies_between_req.add_row_superset_constraint(
    columns1=["name"], columns2=["name"], constant_max_missing_fraction=0
)

About

Assessing whether data from database complies with reference information.

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Contributors 19