EuropeanParliament

/

eurovoc_en

like 2

You can also utilize a SPARQL endpoint for this purpose.

import pandas as pd
import time
from SPARQLWrapper import SPARQLWrapper, JSON
from tqdm import tqdm

def get_eurovoc_code(label, max_retries=5, sleep_time=5):
"""This function returns the Eurovoc code for a given label."""

# Define the SPARQL endpoint and set the return format
sparql = SPARQLWrapper("https://publications.europa.eu/webapi/rdf/sparql")
sparql.setReturnFormat(JSON)

# Set the query with updated conditions
query = f"""
PREFIX cdm: <http://publications.europa.eu/ontology/cdm#>
SELECT * WHERE {{
    ?eurovoc <http://www.w3.org/2004/02/skos/core#prefLabel> "{label}"@en.
    FILTER(regex(str(?eurovoc), "http://eurovoc.europa.eu/" ))
}}
"""

sparql.setQuery(query)

retries = 0
while retries < max_retries:
    try:
        # Execute the query and return the Eurovoc code
        results = sparql.query().convert()
        for result in results["results"]["bindings"]:
            r = result["eurovoc"]["value"].split("/")[-1]
            return r  # assuming the Eurovoc code is the last part of the URI
        return None  # if no code found
    except Exception as e:
        if "HTTP Error 503" in str(e):
            retries += 1
            time.sleep(sleep_time)
        else:
            raise e
raise Exception(f"Failed after {max_retries} attempts for label: {label}")