aboutsummaryrefslogtreecommitdiffstats
path: root/python/sandcrawler_worker.py
blob: 1f02eead71dcf7e4f525a3ef4068f904dee4e76f (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
#!/usr/bin/env python3

"""
These are generally for continuously running workers that consume from Kafka.
Outputs might either be pushed back into Kafka, or directly into sandcrawler-db
or minio.
"""

import sys
import argparse
import datetime
import raven

from sandcrawler import *

# Yep, a global. Gets DSN from `SENTRY_DSN` environment variable
sentry_client = raven.Client()


def run_grobid_extract(args):
    consume_topic = "sandcrawler-{}.ungrobided-pg".format(args.env)
    produce_topic = "sandcrawler-{}.grobid-output-pg".format(args.env)
    sink = KafkaSink(kafka_hosts=args.kafka_hosts, produce_topic=produce_topic)
    worker = GrobidWorker(host_url=args.grobid_host, sink=sink)
    pusher = KafkaJsonPusher(sink=worker, group="grobid-extract")
    pusher.run()

def run_grobid_persist(args):
    consume_topic = "sandcrawler-{}.grobid-output-pg".format(args.env)
    sink = GrobidPersist(consume_topic=consume_topic)
    pusher = KafkaJsonPusher(sink)
    pusher.run()

def run_ingest_file(args):
    consume_topic = "sandcrawler-{}.ingest-file-requests".format(args.env)
    produce_topic = "sandcrawler-{}.ingest-file-results".format(args.env)
    sink = KafkaSink(kafka_hosts=args.kafka_hosts, produce_topic=produce_topic)
    grobid_client = GrobidClient(host_url=args.grobid_host)
    worker = IngestFileWorker(grobid_client=grobid_client, sink=sink)
    pusher = KafkaJsonPusher(worker=worker, kafka_hosts=args.kafka_hosts,
        consume_topic=consume_topic, group="grobid-extract")
    pusher.run()

def main():
    parser = argparse.ArgumentParser()
    parser.add_argument('--kafka-hosts',
        default="localhost:9092",
        help="list of Kafka brokers (host/port) to use")
    parser.add_argument('--env',
        default="dev",
        help="Kafka topic namespace to use (eg, prod, qa, dev)")
    parser.add_argument('--grobid-host',
        default="http://grobid.qa.fatcat.wiki",
        help="GROBID API host/port")
    subparsers = parser.add_subparsers()

    sub_grobid_extract = subparsers.add_parser('grobid-extract')
    sub_grobid_extract.set_defaults(func=run_grobid_extract)

    sub_grobid_persist = subparsers.add_parser('grobid-persist')
    sub_grobid_persist.set_defaults(func=run_grobid_persist)

    sub_ingest_file = subparsers.add_parser('ingest-file')
    sub_ingest_file.set_defaults(func=run_ingest_file)

    args = parser.parse_args()
    if not args.__dict__.get("func"):
        print("tell me what to do!")
        sys.exit(-1)

    args.func(args)

if __name__ == '__main__':
    main()