Convert a list of DOIs to BibTex strings

doi2bib.py

#!/usr/bin/env python3

def get_json(doi:str) -> dict:
    # get citation dict from doi string
    import re
    import urllib.request
    from urllib.error import HTTPError
    import json
    from pprint import pprint

    for match in re.findall('doi.org/(.*)', doi):
        match = match
        break
    else:
        match = doi

    req = urllib.request.Request('https://dx.doi.org/' + match)
    # see https://citation.crosscite.org/docs.html
    req.add_header('Accept', 'application/vnd.citationstyles.csl+json') # JSON
    # req.add_header('Accept', 'application/x-bibtex') # BIBTEX
    try:
        with urllib.request.urlopen(req) as f:
            d = f.read().decode()
    except HTTPError as e:
        print(f"HTTP Error for {doi}: {e.code}")
        exit(1)
    except Error as e:
        raise

    return json.loads(d)



def bibify(d:dict, to_str=True, abbreviate_journal=True, titlecase=True) -> dict:
    # filter the output of get_json() to bibtex format
    bibtex = {}
    bibtex['ENTRYTYPE'] = 'article' if 'article' in d['type'] else d['title']
    bibtex['title']  = d['title']
    if titlecase:
        bibtex['title'] = bibtex['title'].title()
    bibtex['author'] = ' and '.join( f"{dd['family']}, {dd['given']}" for dd in d['author'] )
    if abbreviate_journal and 'container-title-short' in d:
        bibtex['journal'] = ' '.join(i+'.' if not i.endswith('.') else i for i in d['container-title-short'].split())
    else:
        bibtex['journal'] = d['container-title']

    if 'page' in d:
        bibtex['pages'] = '--'.join(d['page'].split('-'))
    if 'number' in d:
        bibtex['issue'] = d['number']

    y, *md = d['published']['date-parts'][0]
    bibtex['year'] = y
    if md:
        bibtex['month'] = md[0]

    for k in 'doi', 'url', 'volume', 'publisher':
        if k in d:
            bibtex[k] = d[k]

    bibtex['ID'] = d['author'][0]['family'] + f"{y}"

    if to_str:
        for k,v in bibtex.items():
            bibtex[k] = str(v)

    return bibtex



def dump(d:dict) -> str:
    # dump the output of `bibify()` to string
    out = f"@{d['ENTRYTYPE']}" + '{' + f"{d['ID']},\n"
    for k,v in d.items():
        if k not in {'ENTRYTYPE', 'ID'}:
            out += f"  {k} = " + '{' + str(v) + '},\n'
    out += '}\n\n'
    return out






if __name__ == '__main__':
    import argparse
    parser = argparse.ArgumentParser(description='DOI to BibTex converter')
    parser.add_argument('DOIs', nargs='+', help='DOI strings to be converted to BibTex')
    parser.add_argument('-t', '--titlecase',  action='store_true', help='Use title case for titles')
    parser.add_argument('-a', '--abbreviate', action='store_true', help='Abbreviate journal names')
    args = parser.parse_args()

    try:
        from bibtexparser.bibdatabase import BibDatabase
        from bibtexparser import dumps as _dumps
        bibparser = True
        _db  = BibDatabase()
        db   = _db.entries
        dumps = lambda db: _dumps(_db)
        # p = bibtexparser.bparser.BibTexParser(common_strings=True)  # For month names.
    except ImportError:
        bibparser = False
        dumps = lambda db: ''.join(dump(i) for i in db)
        db = []

    for doi in args.DOIs:
        d = get_json(doi)
        d = bibify(d, abbreviate_journal=args.abbreviate, titlecase=args.titlecase)
        db.append(d)

    print()
    print(dumps(db))