Affiliation Regex Parser

A lightweight, regex-based Python library for parsing academic author affiliation strings into structured fields (departments, institutes, universities, organizations, cities, countries, emails, postcodes, and unknown segments).

Examples and usage patterns are documented in cookbook.ipynb.

Install

Core install:

pip install affiliation-regex-parser

Optional Excel support (for worldcities.xlsx-based city→country inference):

pip install affiliation-regex-parser[excel]

Features

• Regex-first, deterministic parsing (no ML dependencies).
• Pluggable architecture via providers (e.g., custom city lists, custom inference).
• Optional city→country inference using a world cities dataset.
• Designed for batch parsing (reuse one parser instance).

Data attribution (worldcities.xlsx)

This project can use a worldcities.xlsx dataset sourced from SimpleMaps World Cities (free version), licensed under Creative Commons Attribution 4.0 (CC BY 4.0).

Source:

https://simplemaps.com/data/world-cities

If you redistribute the dataset with this package, ensure attribution is preserved per CC BY 4.0.

Documentation

• See cookbook.ipynb for practical examples and common configurations.
• See the docstrings in AffiliationRegexParser and provider classes for API details.

License

MIT License. See LICENSE.

Project status

This package is under active development; the public API aims to remain stable, but outputs may improve over time as patterns and fixtures evolve.