aboutsummaryrefslogtreecommitdiffstats
path: root/python/fatcat_tools/importers/ingest.py
blob: 7dad13ce5d5deac943ad9087e7d10509a4d475f1 (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

import sys
import json
import base64
import itertools
import fatcat_openapi_client
from .common import EntityImporter, clean, make_rel_url, SANE_MAX_RELEASES, SANE_MAX_URLS, b32_hex


class IngestFileResultImporter(EntityImporter):

    def __init__(self, api, require_grobid=True, **kwargs):

        eg_desc = kwargs.pop('editgroup_description',
            "Files crawled from web using sandcrawler ingest tool")
        eg_extra = kwargs.pop('editgroup_extra', dict())
        eg_extra['agent'] = eg_extra.get('agent', 'fatcat_tools.IngestFileResultImporter')
        super().__init__(api,
            editgroup_description=eg_desc,
            editgroup_extra=eg_extra,
            **kwargs)
        self.default_link_rel = kwargs.get("default_link_rel", "web")
        assert self.default_link_rel
        self.default_mimetype = kwargs.get("default_mimetype", None)
        self.do_updates = kwargs.get("do_updates", False)
        self.require_grobid = require_grobid
        if self.require_grobid:
            print("Requiring GROBID status == 200")
        else:
            print("NOT checking GROBID success")
        self.ingest_request_source_whitelist = ['fatcat-changelog']
        if kwargs.get('skip_source_whitelist', False):
            self.ingest_request_source_whitelist = []

    def want(self, row):
        """
        Logic here probably needs work:

        - Direct ingests via DOI from fatcat-changelog should probably go
          through regardless of GROBID status
        - We should filter/block things like single-page PDFs here
        - public/anonymous submissions could require successful biblio-glutton
          match, or some other sanity check on the fatcat side (eg, fuzzy title
          match)
        - handle the case of release_stage not being 'published'; if pre-print,
          potentially create a new release.

        The current logic is intentionally conservative as a first step.
        """
        if row.get('hit') != True:
            self.counts['skip-hit'] += 1
            return False
        if self.ingest_request_source_whitelist and row['request'].get('ingest_request_source') not in self.ingest_request_source_whitelist:
            self.counts['skip-ingest_request_source'] += 1
            return False
        if not row.get('file_meta'):
            self.counts['skip-file-meta'] += 1
            return False
        if self.require_grobid and row.get('grobid', {}).get('status_code') != 200:
            self.counts['skip-grobid'] += 1
            return False

        return True

    def parse_record(self, row):

        request = row['request']
        fatcat = request.get('fatcat')
        file_meta = row['file_meta']

        # identify release by fatcat ident, or extid lookup, or biblio-glutton match
        release_ident = None
        if fatcat and fatcat.get('release_ident'):
            release_ident = fatcat.get('release_ident')
        elif request.get('ext_ids'):
            # if no fatcat ident, try extids
            for extid_type in ('doi', 'pmid', 'pmcid', 'arxiv'):
                extid = request['ext_ids'].get(extid_type)
                if not extid:
                    continue
                try:
                    release = self.api.lookup_release(**{extid_type: extid})
                except fatcat_openapi_client.rest.ApiException as err:
                    if err.status == 404:
                        continue
                    elif err.status == 400:
                        self.counts['warn-extid-invalid'] += 1
                        continue
                release_ident = release.ident
                break
        if not release_ident and row.get('grobid'):
            # try biblio-glutton extracted hit
            if row['grobid'].get('fatcat_ident'):
                release = row['grobid']['fatcat_ident'].split('_')[-1]

        if not release_ident:
            self.counts['skip-release-not-found'] += 1
            return None

        cdx = row.get('cdx')
        if not cdx:
            # TODO: support archive.org hits?
            self.counts['skip-no-cdx'] += 1
            return None

        url = make_rel_url(cdx['url'], self.default_link_rel)

        if not url:
            self.counts['skip-url'] += 1
            return None
        wayback = "https://web.archive.org/web/{}/{}".format(
            cdx['datetime'],
            cdx['url'])
        urls = [url, ("webarchive", wayback)]

        urls = [fatcat_openapi_client.FileUrl(rel=rel, url=url) for (rel, url) in urls]

        fe = fatcat_openapi_client.FileEntity(
            md5=file_meta['md5hex'],
            sha1=file_meta['sha1hex'],
            sha256=file_meta['sha256hex'],
            size=file_meta['size_bytes'],
            mimetype=file_meta['mimetype'] or self.default_mimetype,
            release_ids=[release_ident],
            urls=urls,
        )
        if fatcat and fatcat.get('edit_extra'):
            fe.edit_extra = fatcat['edit_extra']
        if request.get('ingest_request_source'):
            if not fe.edit_extra:
                fe.edit_extra = dict()
            fe.edit_extra['ingest_request_source'] = request['ingest_request_source']
        return fe

    def try_update(self, fe):
        # lookup sha1, or create new entity
        existing = None
        try:
            existing = self.api.lookup_file(sha1=fe.sha1)
        except fatcat_openapi_client.rest.ApiException as err:
            if err.status != 404:
                raise err

        if not existing:
            return True

        if (fe.release_ids[0] in existing.release_ids) and existing.urls:
            # TODO: could still, in theory update with the new URL?
            self.counts['exists'] += 1
            return False

        if not self.do_updates:
            self.counts['skip-update-disabled'] += 1
            return False

        # TODO: for now, never update
        self.counts['skip-update-disabled'] += 1
        return False

    def insert_batch(self, batch):
        self.api.create_file_auto_batch(fatcat_openapi_client.FileAutoBatch(
            editgroup=fatcat_openapi_client.Editgroup(
                description=self.editgroup_description,
                extra=self.editgroup_extra),
            entity_list=batch))