aboutsummaryrefslogtreecommitdiffstats
path: root/python/grobid_tool.py
blob: 782bc13be8764f114aabbece56c5b3139ea86994 (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
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
#!/usr/bin/env python3
"""
These are generally for running one-off tasks from the command line. Output
might go to stdout, or might go to Kafka topic.

Example of large parallel run, locally:

    cat /srv/sandcrawler/tasks/ungrobided.2019-09-23.json         | pv -l | parallel -j30 --pipe         ./grobid_tool.py --kafka-env prod --kafka-hosts wbgrp-svc263.us.archive.org:9092,wbgrp-svc284.us.archive.org:9092,wbgrp-svc285.us.archive.org:9092 --kafka-mode --grobid-host http://localhost:8070 -j0 extract-json -
"""

import argparse
import json
import sys

from grobid_tei_xml import parse_document_xml

from sandcrawler import *
from sandcrawler.grobid import CrossrefRefsWorker


def run_extract_json(args):
    grobid_client = GrobidClient(host_url=args.grobid_host)
    wayback_client = WaybackClient()
    if args.jobs > 1:
        worker = GrobidWorker(grobid_client, wayback_client, sink=None)
        multi_worker = MultiprocessWrapper(worker, args.sink)
        pusher = JsonLinePusher(multi_worker, args.json_file, batch_size=args.jobs)
    else:
        worker = GrobidWorker(grobid_client, wayback_client, sink=args.sink)
        pusher = JsonLinePusher(worker, args.json_file)
    pusher.run()


def run_extract_cdx(args):
    grobid_client = GrobidClient(host_url=args.grobid_host)
    wayback_client = WaybackClient()
    if args.jobs > 1:
        worker = GrobidWorker(grobid_client, wayback_client, sink=None)
        multi_worker = MultiprocessWrapper(worker, args.sink)
        pusher = CdxLinePusher(
            multi_worker,
            args.cdx_file,
            filter_http_statuses=[200, 226],
            filter_mimetypes=["application/pdf"],
            batch_size=args.jobs,
        )
    else:
        worker = GrobidWorker(grobid_client, wayback_client, sink=args.sink)
        pusher = CdxLinePusher(
            worker,
            args.cdx_file,
            filter_http_statuses=[200, 226],
            filter_mimetypes=["application/pdf"],
        )
    pusher.run()


def run_extract_zipfile(args):
    grobid_client = GrobidClient(host_url=args.grobid_host)
    if args.jobs > 1:
        print("multi-processing: {}".format(args.jobs), file=sys.stderr)
        worker = GrobidBlobWorker(grobid_client, sink=None)
        multi_worker = MultiprocessWrapper(worker, args.sink, jobs=args.jobs)
        pusher = ZipfilePusher(multi_worker, args.zip_file, batch_size=args.jobs)
    else:
        worker = GrobidBlobWorker(grobid_client, sink=args.sink)
        pusher = ZipfilePusher(worker, args.zip_file)
    pusher.run()


def run_transform(args):
    grobid_client = GrobidClient()
    for line in args.json_file:
        if not line.strip():
            continue
        line = json.loads(line)
        if args.metadata_only:
            out = grobid_client.metadata(line)
        else:
            tei_doc = parse_document_xml(line["tei_xml"])
            out = tei_doc.to_legacy_dict()
        if out:
            if "source" in line:
                out["source"] = line["source"]
            print(json.dumps(out))


def run_parse_crossref_refs(args):
    grobid_client = GrobidClient(host_url=args.grobid_host)
    worker = CrossrefRefsWorker(grobid_client, sink=args.sink)
    pusher = JsonLinePusher(worker, args.json_file)
    pusher.run()


def main():
    parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter)
    parser.add_argument(
        "--kafka-mode", action="store_true", help="send output to Kafka (not stdout)"
    )
    parser.add_argument(
        "--kafka-hosts",
        default="localhost:9092",
        help="list of Kafka brokers (host/port) to use",
    )
    parser.add_argument(
        "--kafka-env", default="dev", help="Kafka topic namespace to use (eg, prod, qa, dev)"
    )
    parser.add_argument(
        "-j", "--jobs", default=8, type=int, help="parallelism for batch CPU jobs"
    )
    parser.add_argument(
        "--grobid-host", default="https://grobid.qa.fatcat.wiki", help="GROBID API host/port"
    )
    subparsers = parser.add_subparsers()

    sub_extract_json = subparsers.add_parser(
        "extract-json",
        help="for each JSON line with CDX info, fetches PDF and does GROBID extraction",
    )
    sub_extract_json.set_defaults(func=run_extract_json)
    sub_extract_json.add_argument(
        "json_file",
        help="JSON file to import from (or '-' for stdin)",
        type=argparse.FileType("r"),
    )

    sub_extract_cdx = subparsers.add_parser(
        "extract-cdx", help="for each CDX line, fetches PDF and does GROBID extraction"
    )
    sub_extract_cdx.set_defaults(func=run_extract_cdx)
    sub_extract_cdx.add_argument(
        "cdx_file",
        help="CDX file to import from (or '-' for stdin)",
        type=argparse.FileType("r"),
    )

    sub_extract_zipfile = subparsers.add_parser(
        "extract-zipfile",
        help="opens zipfile, iterates over PDF files inside and does GROBID extract for each",
    )
    sub_extract_zipfile.set_defaults(func=run_extract_zipfile)
    sub_extract_zipfile.add_argument("zip_file", help="zipfile with PDFs to extract", type=str)

    sub_parse_crossref_refs = subparsers.add_parser(
        "parse-crossref-refs",
        help="reads Crossref metadata records, parses any unstructured refs with GROBID",
    )
    sub_parse_crossref_refs.set_defaults(func=run_parse_crossref_refs)
    sub_parse_crossref_refs.add_argument(
        "json_file",
        help="JSON-L file to process (or '-' for stdin)",
        type=argparse.FileType("r"),
    )

    sub_transform = subparsers.add_parser("transform")
    sub_transform.set_defaults(func=run_transform)
    sub_transform.add_argument(
        "--metadata-only",
        action="store_true",
        help="Only pass through bibliographic metadata, not fulltext",
    )
    sub_transform.add_argument(
        "json_file",
        help="convert TEI-XML to JSON. Input is JSON lines with tei_xml field",
        type=argparse.FileType("r"),
    )

    args = parser.parse_args()
    if not args.__dict__.get("func"):
        parser.print_help(file=sys.stderr)
        sys.exit(-1)

    args.sink = None
    if args.kafka_mode:
        produce_topic = "sandcrawler-{}.grobid-output-pg".format(args.kafka_env)
        print("Running in kafka output mode, publishing to {}\n".format(produce_topic))
        args.sink = KafkaCompressSink(kafka_hosts=args.kafka_hosts, produce_topic=produce_topic)

    args.func(args)


if __name__ == "__main__":
    main()