A morally ambiguous qutebrowser plugin for accessing academic papers either through the unpaywall.org API if a free copy exists in PMC, BioArxiv, etc. else through Sci-Hub

unpaywall

#!/usr/bin/env python3
"""
Qutebrowser userscript for finding a free version of first doi on webpage by using REST API of unpaywall.org.
"""

import requests
import json
import os
import sys
import re
from collections import Counter
from qutescript import userscript

user_email = "[email protected]"


# find DOIs on current page page
source = os.getenv("QUTE_TEXT")
with open(source) as f:
    text = f.read()

@userscript
def show_error(request):
    request.send_command('message-error {}'.format(error_msg))


doi_regex = re.compile(r'(10\.(\d)+/([^(\s\>\"\<)])+)')

dois = doi_regex.findall(text)
dois = Counter(e[0] for e in dois)
try:
	doi = dois.most_common(1)[0][0]
except IndexError:
    error_msg="no doi's found"
    show_error()

url = "https://api.unpaywall.org/v2/" + doi + "?email=" + user_email

api_response = requests.get(url, verify=True)
dl_url = ""

if(api_response.ok):
    json_load = json.loads(api_response.content)
    if "best_oa_location" in json_load:
        best_location_dict = eval(str(json_load["best_oa_location"]))
        if best_location_dict != None:
            # if there is a PMC article in html go to that
            if "url_for_landing_page" in json_load:
                dl_url = best_location_dict["url_for_landing_page"]
            # otherwise load pdf
            else:
                dl_url = best_location_dict["url_for_pdf"]


# if legal way didnt work then be more flexible:
if dl_url == "":
    dl_url = "https://sci-hub.tw/" + str(doi)


@userscript
def unpaywall_doi(request):
    request.send_command('open {}'.format(dl_url))


# run the function
unpaywall_doi()