Skip to content

Contributing¤

Thank you for your interest in contributing to ts-shape! This page covers the licensing landscape of the project and its dependencies so that contributors can make informed decisions when adding or updating packages.

For development setup, building, testing, and publishing workflows, see the Development Guide.

Project License¤

ts-shape is released under the MIT License. See the full license text in LICENSE.txt.

All contributions to this repository are accepted under the same MIT License.

Third-Party Dependency Licenses¤

The table below lists every runtime dependency declared in pyproject.toml, its license, and what it is used for in ts-shape.

Dependency Version License Purpose
pandas >= 2.1.0 BSD-3-Clause Core dataframe operations
numpy >= 1.26.4 BSD-3-Clause Numerical computations
scipy >= 1.13.1 BSD-3-Clause Scientific and statistical functions
sqlalchemy >= 2.0.32 MIT Database connectivity (TimescaleDB loader)
azure-storage-blob >= 12.19.1 MIT Azure Blob Storage loader
s3fs >= 2024.10.0 BSD-3-Clause S3 file system access
requests >= 2.32.3 Apache-2.0 HTTP requests

Optional Dependency Licenses¤

The packages below are not installed by default. They are declared as optional extras in pyproject.toml and are imported lazily via guarded try/except ImportError blocks, so they do not affect the license footprint of a base pip install ts-shape.

Dependency Version License Extra Purpose
pint >= 0.24 BSD-3-Clause units / dev Engineering unit conversion (UnitConverter)
scikit-learn >= 1.3.0 BSD-3-Clause ml Optional ML-based detectors (e.g. Isolation Forest outliers)
psycopg2-binary >= 2.9.9 LGPL-3.0-or-later (with exceptions) postgres PostgreSQL / TimescaleDB driver

psycopg2-binary is LGPL-licensed, but it is a separately installed, dynamically linked database driver — ts-shape neither bundles nor redistributes it, so its LGPL terms do not extend to ts-shape itself.

License Compatibility¤

All dependencies are compatible with the MIT License and use permissive licenses:

  • BSD-3-Clause (pandas, numpy, scipy, s3fs, pint, scikit-learn) — permissive, no restrictions beyond attribution.
  • MIT (sqlalchemy, azure-storage-blob) — same terms as ts-shape itself.
  • Apache-2.0 (requests) — permissive, compatible with MIT distribution.

PostgreSQL connectivity is handled through SQLAlchemy, which supports any DB-API 2.0 compatible driver. Users can install the driver of their choice (e.g. pip install ts-shape[postgres] for psycopg2-binary, or install psycopg, pg8000, etc.).

Guidelines for Adding New Dependencies¤

When proposing a new dependency, please consider:

  1. License compatibility — MIT, BSD, and Apache-2.0 are preferred. Avoid GPL-licensed packages as they are incompatible with MIT distribution.
  2. Necessity — prefer the standard library or existing dependencies where possible.
  3. Maintenance — choose well-maintained packages with active communities.
  4. Update requirements.in — add direct dependencies there, then run python scripts/requirements.py compile to pin versions (see Development Guide).