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
|
#!/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 sys
import json
import argparse
import datetime
from grobid2json import teixml2json
from sandcrawler import *
def run_extract_json(args):
wayback_client = WaybackClient()
if args.jobs > 1:
worker = PdfExtractWorker(wayback_client, sink=None, thumbnail_sink=args.thumbnail_sink)
multi_worker = MultiprocessWrapper(worker, args.sink)
pusher = JsonLinePusher(multi_worker, args.json_file, batch_size=args.jobs)
else:
worker = PdfExtractWorker(wayback_client, sink=args.sink, thumbnail_sink=args.thumbnail_sink)
pusher = JsonLinePusher(worker, args.json_file)
pusher.run()
def run_extract_cdx(args):
wayback_client = WaybackClient()
if args.jobs > 1:
worker = PdfExtractWorker(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 = PdfExtractWorker(wayback_client, sink=args.sink, thumbnail_sink=args.thumbnail_sink)
pusher = CdxLinePusher(
worker,
args.cdx_file,
filter_http_statuses=[200, 226],
filter_mimetypes=['application/pdf'],
)
pusher.run()
def run_extract_zipfile(args):
if args.jobs > 1:
print("multi-processing: {}".format(args.jobs), file=sys.stderr)
worker = PdfExtractBlobWorker(sink=None, thumbnail_sink=None)
multi_worker = MultiprocessWrapper(worker, args.sink, jobs=args.jobs)
pusher = ZipfilePusher(multi_worker, args.zip_file, batch_size=args.jobs)
else:
worker = PdfExtractBlobWorker(sink=args.sink, thumbnail_sink=None)
pusher = ZipfilePusher(worker, args.zip_file)
pusher.run()
def run_single(args):
worker = PdfExtractBlobWorker(sink=None, thumbnail_sink=None)
with open(args.pdf_file, 'rb') as pdf_file:
pdf_bytes = pdf_file.read()
result = worker.process(pdf_bytes)
print(json.dumps(result.to_pdftext_dict(), sort_keys=True))
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")
subparsers = parser.add_subparsers()
sub_extract_json = subparsers.add_parser('extract-json',
help="for each JSON line with CDX info, fetches PDF and does PDF 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 PDF 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 PDF 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_single = subparsers.add_parser('single',
help="opens single PDF and extracts it")
sub_single.set_defaults(func=run_single)
sub_single.add_argument('pdf_file',
help="single PDF to extract",
type=str)
args = parser.parse_args()
if not args.__dict__.get("func"):
print("tell me what to do!", file=sys.stderr)
sys.exit(-1)
args.text_sink = None
args.thumbnail_sink = None
if args.kafka_mode:
text_topic = "sandcrawler-{}.pdftext".format(args.kafka_env)
thumbnail_topic = "sandcrawler-{}.thumbnail-180px-jpeg".format(args.kafka_env)
args.sink = KafkaCompressSink(kafka_hosts=args.kafka_hosts,
produce_topic=text_topic)
args.thumbnail_sink = KafkaCompressSink(kafka_hosts=args.kafka_hosts,
produce_topic=thumbnail_topic)
print("Running in kafka output mode, publishing to {} and {}\n".format(
text_topic, thumbnail_topic), file=sys.stderr)
else:
args.sink = None
args.func(args)
if __name__ == '__main__':
main()
|
