aboutsummaryrefslogtreecommitdiffstats
path: root/python/fatcat_ingest.py
blob: 21597fae655b2d826cd2ce4ecbd2200a71737045 (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
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
#!/usr/bin/env python3

"""
Intended to be a command line interface to "Save Paper Now" and ingest
request/response.
"""

import argparse
import json
import sys
from collections import Counter

import elasticsearch
import raven
from elasticsearch_dsl import Q, Search

from fatcat_tools import kafka_fail_fast, public_api, simple_kafka_producer
from fatcat_tools.transforms import release_ingest_request

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


def _init_search(args):

    # ensure API connection works
    args.api.get_changelog()

    client = elasticsearch.Elasticsearch(args.elasticsearch_endpoint)
    search = Search(using=client, index=args.elasticsearch_index)
    return search


def _run_search_dump(args, search):

    if args.dry_run:
        print("=== THIS IS A DRY RUN ===")

    kafka_producer = None
    if args.kafka_request_topic:
        ingest_file_request_topic = args.kafka_request_topic
    else:
        ingest_file_request_topic = "sandcrawler-{}.ingest-file-requests-daily".format(args.env)
    if args.enqueue_kafka:
        print(
            "Will send ingest requests to kafka topic: {}".format(ingest_file_request_topic),
            file=sys.stderr,
        )
        kafka_producer = simple_kafka_producer(args.kafka_hosts)

    if args.limit is not None:
        search = search[: args.limit]

    if args.before_year:
        search = search.filter("exists", field="release_year").filter(
            "range", release_date=dict(lt=args.before_year)
        )
    if args.after_year:
        search = search.filter("exists", field="release_year").filter(
            "range", release_date=dict(gte=args.after_year)
        )

    if not args.allow_non_oa:
        search = search.filter("term", is_oa=True)

    if args.release_types:
        release_types = args.release_types.split(",")
        search = search.filter("terms", release_type=release_types)
    else:
        search = search.filter(
            "bool", must_not=[Q("terms", release_type=["stub", "component"])]
        )

    counts = Counter({"ingest_request": 0, "elasticsearch_release": 0, "estimate": 0})
    search = search.params()
    counts["estimate"] = search.count()
    print(
        "Expecting {} release objects in search queries".format(counts["estimate"]),
        file=sys.stderr,
    )

    # don't try to clean up scroll if we are connected to public server (behind
    # nginx proxy that disallows DELETE)
    if args.elasticsearch_endpoint in (
        "https://search.fatcat.wiki",
        "https://search.qa.fatcat.wiki",
    ):
        search = search.params(clear_scroll=False)

    results = search.scan()
    for esr in results:
        if args.limit and counts["ingest_request"] >= args.limit:
            break
        counts["elasticsearch_release"] += 1
        release = args.api.get_release(esr.ident)
        ingest_request = release_ingest_request(
            release,
            ingest_request_source="fatcat-ingest",
            ingest_type=args.ingest_type,
        )
        if not ingest_request:
            continue
        if args.force_recrawl:
            ingest_request["force_recrawl"] = True
        counts["ingest_request"] += 1
        if args.dry_run:
            continue
        if kafka_producer is not None:
            kafka_producer.produce(
                ingest_file_request_topic,
                json.dumps(ingest_request).encode("utf-8"),
                # key=None,
                on_delivery=kafka_fail_fast,
            )
            counts["kafka"] += 1
        else:
            print(json.dumps(ingest_request))
    if kafka_producer is not None:
        kafka_producer.flush()
    print(counts, file=sys.stderr)
    if args.dry_run:
        print("=== THIS WAS A DRY RUN ===")


def run_ingest_container(args):
    """
    This command queries elasticsearch for releases from a given container (eg,
    journal), and prepares ingest requests for them.

    By default it filters to releases which don't have any fulltext files
    archived in IA, and dumps the ingest requests as JSON.
    """

    search = _init_search(args).filter("term", in_ia=False)

    # filter/query by container
    if args.container_id:
        search = search.filter("term", container_id=args.container_id)
    elif args.issnl:
        search = search.filter("term", container_issnl=args.issnl)
    elif args.publisher:
        search = search.query("match", publisher=args.publisher)
    elif args.name:
        search = search.query("match", container_name=args.name)
    else:
        print(
            "You must supply at least one query/filter parameter! Eg, ISSN-L", file=sys.stderr
        )
        sys.exit(-1)

    return _run_search_dump(args, search)


def run_ingest_query(args):
    """
    Accepts a free-form Lucene query language string. Intended to work the same
    way as searches in the fatcat web interface.
    """

    search = (
        _init_search(args)
        .filter("term", in_ia=False)
        .query(
            "query_string",
            query=args.query,
            default_operator="AND",
            analyze_wildcard=True,
            lenient=True,
            fields=["title^5", "contrib_names^2", "container_title"],
        )
    )

    return _run_search_dump(args, search)


def run_ingest_extid(args):
    """
    Selects release entities where the external identifier (extid) exists
    """

    search = _init_search(args).filter("term", in_ia=False).filter("exists", field=args.extid)

    return _run_search_dump(args, search)


def main():
    parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter)
    parser.add_argument(
        "--fatcat-api-url", default="http://localhost:9411/v0", help="connect to this host/port"
    )
    parser.add_argument(
        "--enqueue-kafka",
        action="store_true",
        help="send ingest requests directly to sandcrawler kafka topic for processing",
    )
    parser.add_argument(
        "--kafka-hosts",
        default="localhost:9092",
        help="list of Kafka brokers (host/port) to use",
    )
    parser.add_argument("--kafka-request-topic", help="exact Kafka ingest request topic to use")
    parser.add_argument(
        "--elasticsearch-endpoint",
        default="https://search.fatcat.wiki",
        help="elasticsearch API. internal endpoint preferred, but public is default",
    )
    parser.add_argument(
        "--elasticsearch-index", default="fatcat_release", help="elasticsearch index to query"
    )
    parser.add_argument(
        "--env", default="dev", help="Kafka topic namespace to use (eg, prod, qa, dev)"
    )
    parser.add_argument(
        "--limit", default=None, type=int, help="Max number of search hits to return"
    )
    parser.add_argument(
        "--dry-run",
        action="store_true",
        help="runs through creating all ingest requests, but doesn't actually output or enqueue",
    )
    parser.add_argument(
        "--before-year",
        type=str,
        help="filters results to only with release_year before this (not inclusive)",
    )
    parser.add_argument(
        "--after-year",
        type=str,
        help="filters results to only with release_year after this (inclusive)",
    )
    parser.add_argument(
        "--release-types",
        type=str,
        help="filters results to specified release-types, separated by commas. By default, 'stub' is filtered out.",
    )
    parser.add_argument(
        "--allow-non-oa",
        action="store_true",
        help="By default, we limit to OA releases. This removes that filter",
    )
    parser.add_argument(
        "--force-recrawl",
        action="store_true",
        help="Tell ingest worker to skip GWB history lookup and do SPNv2 crawl",
    )
    parser.add_argument(
        "--ingest-type", default="pdf", help="What medium to ingest (pdf, xml, html)"
    )
    subparsers = parser.add_subparsers()

    sub_container = subparsers.add_parser(
        "container", help="Create ingest requests for releases from a specific container"
    )
    sub_container.set_defaults(func=run_ingest_container)
    sub_container.add_argument("--container-id", help="fatcat container entity ident")
    sub_container.add_argument("--issnl", help="ISSN-L of container entity")
    sub_container.add_argument("--publisher", help="publisher name")
    sub_container.add_argument("--name", help="container name")

    sub_query = subparsers.add_parser(
        "query", help="Create ingest requests for releases from a specific query"
    )
    sub_query.set_defaults(func=run_ingest_query)
    sub_query.add_argument("query", help="search query (same DSL as web interface search)")

    sub_extid = subparsers.add_parser(
        "extid", help="Create ingest requests for releases that have given extid defined"
    )
    sub_extid.set_defaults(func=run_ingest_extid)
    sub_extid.add_argument("extid", help="extid short name (as included in ES release schema)")

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

    args.api = public_api(args.fatcat_api_url)
    args.func(args)


if __name__ == "__main__":
    main()