IPCC AR6 2022 QIDs

dois.py

"""
If query.sparql and the input data file are downloaded, 
the script can simply be called as:

python iq-to-qid.py
"""

import argparse
import os
import time
import pandas as pd

import requests
from requests.exceptions import HTTPError
from tqdm import tqdm, trange

HEADERS = {"User-Agent": "ID-to-QID"}


def main(
    test: bool = False,
    inputFile: str = "deduplicated-dois.csv",
    batchSize: int = 100,
):
    """From a list of IDs, go through and sequentially output a list of their properties and values."""

    if not os.path.isfile(inputFile):
        print(inputFile + " does not exist")
        exit()

    with open("query.sparql") as r:
        query = r.readlines()
        query = "".join(query)

    with open(inputFile) as f:
        inputList = [line.strip() for line in f]

    data = []

    for i in trange(0, len(inputList), batchSize):
        IDs = inputList[i:i+batchSize]

        IDstring = " ".join(["'" + q + "'" for q in IDs])

        data += getData(query, IDstring)

        if test:
            break

    df = pd.DataFrame(data, columns =['qid', 'doi'])
    df.to_csv("qids-dois.csv", index=False)


def getData(query, IDstring):
    data = runQuery(
        query.format(
            values=IDstring,
        )
    )

    output = []

    for item in data["results"]["bindings"]:
        QID = item["item"]["value"][31:]
        id = item["id"]["value"]
        output.append([QID, id])

    return output


def runQuery(query):
    url = "https://query.wikidata.org/sparql"
    params = {"query": query, "format": "json"}
    try:
        response = requests.get(url, params=params, headers=HEADERS)
        return response.json()
    except HTTPError as e:
        print(response.text)
        print(e.response.text)
        exit()


def timer(tick, msg=""):
    print("--- %s %.3f seconds ---" % (msg, time.time() - tick))
    return time.time()


def defineArgParser():
    """Creates parser for command line arguments"""
    parser = argparse.ArgumentParser(
        description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter
    )

    parser.add_argument(
        "-t",
        "--test",
        help="",
        action="store_true",
    )

    return parser


if __name__ == "__main__":

    argParser = defineArgParser()
    clArgs = argParser.parse_args()

    tick = time.time()
    main(
        test=clArgs.test,
    )
    timer(tick)

dois.sparql

SELECT ?item ?doi {{
  VALUES ?doi {{ {values} }}
  
  ?item wdt:P356 ?doi .
  
  SERVICE wikibase:label {{ bd:serviceParam wikibase:language "en" }}
}}

qid-doi.csv