A script to grab ADS private libraries and send them to ORCID

adsjson_to_orcid.py

from __future__ import division, print_function

"""
This script takes in the URL to the JSON version of an ADS private library, and outputs a bibtex file suitable for importing into ORCiD
"""

import json
import urllib2


def extract_bibcodes(urloflibrary):
    u = urllib2.urlopen(urloflibrary)
    try:
        jsitems = json.loads(u.read())['items']
    finally:
        u.close()

    adsnames = []
    for i in jsitems:
        if i['itemtype'] == 'ads/itemtype:pub':
            adsnames.append(i['basic']['name'])
        else:
            print('The item type is "{0}" not pub, so I don\'t know what to do '
                  'with it'.format(i['itemtype']))
    return adsnames


BIBCODE_TEMPLATE = 'http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode={bibcode}&data_type=BIBTEX&db_key=AST&nocookieset=1'
def bibcodes_to_bibtex(adsnames):
    bibtexs = []
    for name in adsnames:
        url = BIBCODE_TEMPLATE.format(bibcode=name)
        u = urllib2.urlopen(url)
        try:
            response = u.read()
        finally:
            u.close()

        bibtexs.append('@' + '@'.join(response.split('@')[1:]))
    return _hack_to_make_orcid_parse_the_bibtex_damnit(bibtexs)


def _hack_to_make_orcid_parse_the_bibtex_damnit(bibtexs):
    newbibtexs = []
    for b in bibtexs:
        bi = b.replace('month = jan', 'month = Jan')
        lines = []
        for l in bi.split('\n'):
            if l.lstrip().startswith('@ARTICLE'):
                lines.append(l.replace('&', ''))
            else:
                lines.append(l)
        newbibtexs.append('\n'.join(lines))
    return newbibtexs


if __name__ == '__main__':
    import argparse

    p = argparse.ArgumentParser()
    p.add_argument('jsonlibraryurl')
    p.add_argument('-o', '--outfile', default=None)

    ops = p.parse_args()

    adsnames = extract_bibcodes(ops.jsonlibraryurl)
    bibtexs = bibcodes_to_bibtex(adsnames)

    if ops.outfile is None:
        for l in bibtexs:
            print(l)
    else:
        with open(ops.outfile, 'w') as f:
            for l in bibtexs:
                f.write(l)
                f.write('\n')