Data cleaning and validation functions for processing various types of text emanating and describing the business world. This applies to human and company names, language, territory and country codes, corporate and tax identifiers, etc.

The underlying idea is that handling these sorts of descriptors is easy on first glance, but reveals a dizzying set of complexity when carried into production. This is why rigour consolidates implementations that have already met some edge cases and are well-tested.

You can just grab the library from PyPI:

pip install -U rigour

See: https://opensanctions.github.io/rigour/