fxcoudert/OA_audit.py

## OA_audit.py
# -*- coding: utf-8 -*-

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


orcid_api_id = 'xxxxxx'
orcid_api_secret = 'xxxxxx'
base_api_ua = 'xxxxxx'
orcid_api_email = 'xxxxxx'


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_id:' + doi,
         'fl': 'openAccess_bool,docid,halId_s'}
    r = requests.get('http://api.archives-ouvertes.fr/search/', params=p)
    return r.json()


def unpaywall(doi):
    r = requests.get(f'https://api.unpaywall.org/v2/{doi}',
                     params={"email": "fxcoudert@gmail.com"})
    try:
        r = r.json()
    except:
        return None

    if 'error' in r:
        return None
    if r['is_oa']:
        return r['best_oa_location']['url_for_landing_page']
    return None


def base_search(doi):
    r = requests.get('https://api.base-search.net/cgi-bin/BaseHttpSearchInterface.fcgi',
                     params={'func': 'PerformSearch', 'query': f'dcdoi:{doi}', 'boost': 'oa', 'format': 'json'},
                     headers={'user-agent': 'fx.coudert@chimieparistech.psl.eu'})
    docs = r.json()['response']['docs']
    for doc in docs:
        if doc['dcoa'] == 1:
            return doc['dcidentifier'][0]
    return None


def isOA(doi):
    x = unpaywall(doi)
    if x:
        return x
    x = base_search(doi)
    if x:
        return x
    return None


def papersAvailable(doilist, auth_orcid):
    res = []
    for doi, title in doilist.values():
        res.append((doi, title, isOA(doi)))

    nOA = sum(1 for x in res if x[2])
    score = 100 * nOA / len(res)
    s = f'<p style="font-size: 140%">Your OA score is: <b>{score:.1f}%</b></p>'

    s += '<p style="margin-top: 30px">Here is the detail of your papers,'
    s += f' as recorded on <a href="https://orcid.org/{auth_orcid}">ORCID {auth_orcid}</a>:</p>'
    s += '<table>'
    s += '<tr><td>Title</td><td>DOI</td><td>Open access?</td></tr>'
    for doi, title, oa in res:
        color = '#80d080' if oa else '#d0d080'
        href = 'https://doi.org/' + doi
        s += f'<tr style="background-color: {color}"><td><a href="{href}">{title}</a></td><td><a href="{href}">{doi}</a></td>'

        if oa:
            s += f'<td style="text-align: center"><a href="{oa}">Yes</a></td></tr>'
        else:
            s += f'<td style="text-align: center">Not found</td></tr>'

    s += '</table>'

    s += '<p style="margin-top: 30px">Open access data is taken from <a href="https://unpaywall.org">Unpaywall</a>'
    s += ' and <a href="https://www.base-search.net">BASE</a>.</p>'

    return s


def auditORCID(auth_orcid):
    first_name, last_name, doilist = retrieveFromORCID(auth_orcid)
    if doilist:
        s = header(f'Papers for {first_name} {last_name}')
        s += papersAvailable(doilist, auth_orcid)
        return s
    else:
        s = header('Error')
        s += f'<p>Could not find ORCID record for: <a href="https://orcid.org/{auth_orcid}">{auth_orcid}</a>.</p>'
        s += '<p>Either you do not have a DOI, or that DOI does not have any papers listed.</p>'
        return s


def header(title):
    return f'''
<!DOCTYPE html>
<html lang="en">

<head>
  <title>{title}</title>

  <meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
  <link rel="stylesheet" type="text/css" href="default.css" />
  <meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
  <style>
  a {{
    color: black;
  }}
  table a {{
    text-decoration: none;
  }}
  table, th, td {{
    padding: 2px;
    font-size: 95%;
  }}
  </style>
</head>

<body>
<h1>{title}</h1>
    '''

def landingPage():
    s = header("Check your Open Access score!")
    s += 'Enter your ORCID here:'
    s += '<form><input type="text" name="orcid" /><input type="submit" value="Submit"></form>'
    s += 'This can take up to 3 minutes, please wait after you click...'
    return s
	# -- coding: utf-8 --

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


	orcid_api_id = 'xxxxxx'
	orcid_api_secret = 'xxxxxx'
	base_api_ua = 'xxxxxx'
	orcid_api_email = 'xxxxxx'


	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_id:' + doi,
	'fl': 'openAccess_bool,docid,halId_s'}
	r = requests.get('http://api.archives-ouvertes.fr/search/', params=p)
	return r.json()


	def unpaywall(doi):
	r = requests.get(f'https://api.unpaywall.org/v2/{doi}',
	params={"email": "fxcoudert@gmail.com"})
	try:
	r = r.json()
	except:
	return None

	if 'error' in r:
	return None
	if r['is_oa']:
	return r['best_oa_location']['url_for_landing_page']
	return None


	def base_search(doi):
	r = requests.get('https://api.base-search.net/cgi-bin/BaseHttpSearchInterface.fcgi',
	params={'func': 'PerformSearch', 'query': f'dcdoi:{doi}', 'boost': 'oa', 'format': 'json'},
	headers={'user-agent': 'fx.coudert@chimieparistech.psl.eu'})
	docs = r.json()['response']['docs']
	for doc in docs:
	if doc['dcoa'] == 1:
	return doc['dcidentifier'][0]
	return None


	def isOA(doi):
	x = unpaywall(doi)
	if x:
	return x
	x = base_search(doi)
	if x:
	return x
	return None


	def papersAvailable(doilist, auth_orcid):
	res = []
	for doi, title in doilist.values():
	res.append((doi, title, isOA(doi)))

	nOA = sum(1 for x in res if x[2])
	score = 100 * nOA / len(res)
	s = f'<p style="font-size: 140%">Your OA score is: <b>{score:.1f}%</b></p>'

	s += '<p style="margin-top: 30px">Here is the detail of your papers,'
	s += f' as recorded on <a href="https://orcid.org/{auth_orcid}">ORCID {auth_orcid}</a>:</p>'
	s += '<table>'
	s += '<tr><td>Title</td><td>DOI</td><td>Open access?</td></tr>'
	for doi, title, oa in res:
	color = '#80d080' if oa else '#d0d080'
	href = 'https://doi.org/' + doi
	s += f'<tr style="background-color: {color}"><td><a href="{href}">{title}</a></td><td><a href="{href}">{doi}</a></td>'

	if oa:
	s += f'<td style="text-align: center"><a href="{oa}">Yes</a></td></tr>'
	else:
	s += f'<td style="text-align: center">Not found</td></tr>'

	s += '</table>'

	s += '<p style="margin-top: 30px">Open access data is taken from <a href="https://unpaywall.org">Unpaywall</a>'
	s += ' and <a href="https://www.base-search.net">BASE</a>.</p>'

	return s


	def auditORCID(auth_orcid):
	first_name, last_name, doilist = retrieveFromORCID(auth_orcid)
	if doilist:
	s = header(f'Papers for {first_name} {last_name}')
	s += papersAvailable(doilist, auth_orcid)
	return s
	else:
	s = header('Error')
	s += f'<p>Could not find ORCID record for: <a href="https://orcid.org/{auth_orcid}">{auth_orcid}</a>.</p>'
	s += '<p>Either you do not have a DOI, or that DOI does not have any papers listed.</p>'
	return s


	def header(title):
	return f'''
	<!DOCTYPE html>
	<html lang="en">

	<head>
	<title>{title}</title>

	<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
	<link rel="stylesheet" type="text/css" href="default.css" />
	<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
	<style>
	a {{
	color: black;
	}}
	table a {{
	text-decoration: none;
	}}
	table, th, td {{
	padding: 2px;
	font-size: 95%;
	}}
	</style>
	</head>

	<body>
	<h1>{title}</h1>
	'''

	def landingPage():
	s = header("Check your Open Access score!")
	s += 'Enter your ORCID here:'
	s += '<form><input type="text" name="orcid" /><input type="submit" value="Submit"></form>'
	s += 'This can take up to 3 minutes, please wait after you click...'
	return s