DOI python regex

doi_regex.py

import re


def main():
    tests = ["10.1021/cb3006787",
             "10.1021/acschembio.5b00308",
             "10.7164/antibiotics.45.1853",
             "10.1016/S0953-7562(09)80401-2"
             ]

    regexp = re.compile('^(10.\d{4,9})\/([-._;()/:A-Za-z0-9]+)$')

    for t in tests:
        res = re.match(regexp, t)
        print('Overall match: %s' % res.group(0))
        print('Group 1 match: %s' % res.group(1))
        print('Group 2 match: %s\n' % res.group(2))

if __name__ == '__main__':
    main()

Adapted from CrossRef.
Should capture ~99% of DOIs.
If in a pinch, '^(10.\d{4,9})\/)([^s]+)$' should be fairly robust, but may allow for false positives.