How to get every PLOS article DOI using PLOS's search API

plos_dois.py

""" This small script will return a list of every PLOS article DOI.
It uses regular expressions to return only DOIs to full articles.
For more information on text and data-mining PLOS content, see http://api.plos.org/text-and-data-mining/
"""
import requests

def get_all_solr_dois():
    """
    Get every article published by PLOS, up to 500,000, as indexed by Solr on api.plos.org.
    URL includes regex to exclude sub-DOIs and image DOIs.
    :return: list of DOIs for all PLOS articles
    """
    solr_magic_url = ('http://api.plos.org/terms?terms.fl=id&terms.limit=500000&wt=json&indent=true&terms.regex='
                      '10%5C.1371%5C/(journal%5C.p%5Ba-zA-Z%5D%7B3%7D%5C.%5B%5Cd%5D%7B7%7D$%7Cannotation%5C/'
                      '%5Ba-zA-Z0-9%5D%7B8%7D-%5Ba-zA-Z0-9%5D%7B4%7D-%5Ba-zA-Z0-9%5D%7B4%7D-%5Ba-zA-Z0-9%5D'
                      '%7B4%7D-%5Ba-zA-Z0-9%5D%7B12%7D$)')
    results = requests.get(solr_magic_url).json()
    solr_dois = [id for id in results['terms']['id'] if isinstance(id, str)]

    return solr_dois

if __name__ == "__main__":
    plos_dois = get_all_solr_dois()