gistfile1.txt

# -*- coding: utf-8 -*-

import collections
import json
import orcid
import requests
import sys
import time


def retrieveFromORCID(auth_orcid):
    def findDOI(paper):
        for i in paper['external-ids']['external-id']:
            if i['external-id-type'] == 'doi':
                return i['external-id-value']
        return None

    api = orcid.PublicAPI(orcid_api_id, orcid_api_secret)
    token = api.get_search_token_from_orcid()
    try:
        record = api.read_record_public(auth_orcid, 'record', token)
        name = record['person']['name']
        works = api.read_record_public(auth_orcid, 'works', token)['group']
    except:
        return None, None, None

    # Store DOIs as keys in an ordered dict, with uppercase, to avoid duplicates
    doilist = collections.OrderedDict()
    for paper in works:
        doi = findDOI(paper)
        if doi:
            doilist[doi.upper()] = (doi, paper['work-summary'][0]['title']['title']['value'])

    return name["given-names"]["value"], name["family-name"]["value"], doilist


def HALfromDOI(doi):
    p = {'wt': 'json',
         'q': 'doiId_s:"' + doi + '"',
         'fl': 'openAccess_bool,docid'}
    r = requests.get('http://api.archives-ouvertes.fr/search/', params=p)
    r = r.json()

    if 'response' in r and r['response']['numFound'] > 0:
        return r
    else:
        # Search should be case insensitive, but it is not :(
        # Work around this bug by performing a second search on lowercase
        # if there were no results
        p = {'wt': 'json',
            'q': 'doiId_s:"' + doi.lower() + '"',
            'fl': 'openAccess_bool,docid'}
        r = requests.get('http://api.archives-ouvertes.fr/search/', params=p)
        r = r.json()

        if 'response' in r and r['response']['numFound'] > 0:
            return r
        else:
            p = {'wt': 'json',
                'q': 'doiId_s:"' + doi.upper() + '"',
                'fl': 'openAccess_bool,docid'}
            r = requests.get('http://api.archives-ouvertes.fr/search/', params=p)
            r = r.json()
            return r


def papersAvailable(doilist):
    s = '<table>'
    s += '<tr><td>Title</td><td>DOI</td><td>In HAL?</td><td>Full text?</td></tr>'
    for doi, title in doilist.values():
        r = HALfromDOI(doi)

        hal = False
        oa = False
        if 'response' in r and r['response']['numFound'] > 0:
            hal = True
            oa = r['response']['docs'][0]['openAccess_bool']

        if hal:
            color = '#40d040' if oa else '#d0d040'
        else:
            color = '#d06060'

        hal = '+' if hal else '–'
        oa = '+' if oa else '–'
        s += f'<tr style="background-color: {color}"><td>{title}</td><td>{doi}</td>'
        s += f'<td style="text-align: center">{hal}</td><td style="text-align: center">{oa}</td></tr>'

    s += '</table>'
    return s


def auditORCID(auth_orcid):
    first_name, last_name, doilist = retrieveFromORCID(auth_orcid)
    if doilist:
        s = f'<h1>Papers for {first_name} {last_name} (ORCID: {auth_orcid})</h1>'
        s += papersAvailable(doilist)
        return s
    else:
        return f'ERROR: Could not find ORCID record for: {auth_orcid}'


def landingPage():
    s = '<h1>Quickly check if your papers are on HAL</h1>'
    s += 'Enter your ORCID here:'
    s += '<form><input type="text" name="orcid" /><input type="submit" value="Submit"></form>'
    s += 'This can take up to 2 minutes, please wait after you click...'
    return s