pubmed_parser

Why we included this project

PubMed's XML records are messy enough that most teams end up writing the same fragile parsing code over and over. This library does that work once: it reads the PubMed Open-Access subset, MEDLINE XML, and Entrez E-utils responses into plain Python dictionaries, with parsers for article metadata, citations, paragraphs, images, captions, and grant IDs. For anyone doing text mining, bibliometrics, or biomedical NLP, that means the analysis gets your attention instead of the record formats, which vary between journals and database versions. The codebase is small enough to read and adapt, so when you need a field the built-in parsers don't expose, you can extend it rather than start from scratch.