elasticsearch-ais-extract.py

#!/usr/bin/env python3
import csv
import requests
import json
import os
import sys

def elastic(name, director, year):
    payload = {
      "_source": "FILMID",
      "size": 1,
      "query": {
        "bool": {
          "should": [
            { "match": { "NAZEV-SKUT": name }},
            { "match": { "REZIE": director }},
            { "match": { "ROK-VYROBY": year }}
          ]
        }
      }
    }

    r = requests.post('https://{0}/{1}/_search'.format(os.environ['ELASTIC_HOST'], os.environ['ELASTIC_INDEX']), json=payload, auth=(os.environ['ELASTIC_USER'], os.environ['ELASTIC_PASS']))
    try:
        ais_id = r.json()['hits']['hits'][0]['_source']['FILMID']
        print(ais_id)
    except Exception as e:
        print('ERROR')
        print('---------', file=sys.stderr)
        print(e, file=sys.stderr)
        print(json.dumps(payload), file=sys.stderr)
        print('---------', file=sys.stderr)

with open('a.tsv') as tsvfile:
  reader = csv.reader(tsvfile, delimiter='\t')
  for row in reader:
      name = row[2]
      director = row[3]
      year = row[4]
      elastic(name, director, year)