aboutsummaryrefslogtreecommitdiffstats
path: root/python/fatcat_tools/importers/matched.py
blob: 3ef617d300fb2a3574a09c77f72c4f9663d69a1f (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

import sys
import json
import sqlite3
import itertools
import fatcat_client

from fatcat_tools.normal import *
from .common import EntityImporter, clean, make_rel_url, SANE_MAX_RELEASES, SANE_MAX_URLS


class MatchedImporter(EntityImporter):
    """
    Importer for "file to crossref DOI" matches.

    These matches are currently generated by Internet Archive hadoop jobs
    written in scala (part of the 'sandcrawler' repo/project), but could be
    generated by other parties as well.

    Input format is JSON with keys:
    - dois (list)
    - sha1 (hex)
    - md5 (hex)
    - sha256 (hex)
    - size (int)
    - cdx (list of objects)
        - dt (optional; if included creates wayback link)
        - url
    - mimetype
    - urls (list of strings... or objects?)

    Future handlings/extensions:
    - core_id, wikidata_id, pmcid, pmid: not as lists
    """

    def __init__(self, api, **kwargs):

        eg_desc = kwargs.pop('editgroup_description',
            "Import of large-scale file-to-release match results. Source of metadata varies.")
        eg_extra = kwargs.pop('editgroup_extra', dict())
        eg_extra['agent'] = eg_extra.get('agent', 'fatcat_tools.MatchedImporter')
        super().__init__(api,
            editgroup_description=eg_desc,
            editgroup_extra=eg_extra,
            **kwargs)
        self.default_link_rel = kwargs.get("default_link_rel", "web")
        self.default_mimetype = kwargs.get("default_mimetype", None)

    def want(self, raw_record):
        return True

    def parse_record(self, obj):
        dois = [d.lower() for d in obj.get('dois', [])]

        # lookup dois
        re_list = set()
        for doi in dois:
            doi = clean_doi(doi)
            if not doi:
                self.counts['skip-bad-doi'] += 1
                return None
            try:
                re = self.api.lookup_release(doi=doi)
            except fatcat_client.rest.ApiException as err:
                if err.status != 404:
                    raise err
                re = None
            if re is None:
                #print("DOI not found: {}".format(doi))
                pass
            else:
                re_list.add(re.ident)

        # look up other external ids
        for extid_type in ('arxiv', 'pmid', 'pmcid', 'jstor', 'wikidata_qid', 'core', 'isbn13', 'ark'):
            extid = obj.get(extid_type)
            if extid:
                try:
                    re = self.api.lookup_release(**{extid_type: extid})
                except fatcat_client.rest.ApiException as err:
                    if err.status != 404:
                        raise err
                    re = None
                if re is None:
                    pass
                else:
                    re_list.add(re.ident)

        release_ids = list(re_list)
        if len(release_ids) == 0:
            self.counts['skip-no-releases'] += 1
            return None
        if len(release_ids) > SANE_MAX_RELEASES:
            self.counts['skip-too-many-releases'] += 1
            return None

        # parse URLs and CDX
        urls = set()
        for url in obj.get('url', []):
            url = make_rel_url(url, default_link_rel=self.default_link_rel)
            if url != None:
                urls.add(url)
        for cdx in obj.get('cdx', []):
            original = cdx['url']
            if cdx.get('dt'):
                wayback = "https://web.archive.org/web/{}/{}".format(
                    cdx['dt'],
                    original)
                urls.add(("webarchive", wayback))
            url = make_rel_url(original, default_link_rel=self.default_link_rel)
            if url != None:
                urls.add(url)
        urls = [fatcat_client.FileUrl(rel=rel, url=url) for (rel, url) in urls]
        if len(urls) == 0:
            self.counts['skip-no-urls'] += 1
            return None
        if len(urls) > SANE_MAX_URLS:
            self.counts['skip-too-many-urls'] += 1
            return None

        size = obj.get('size')
        if size:
            size = int(size)

        mimetype = obj.get('mimetype', self.default_mimetype)
        if not mimetype and urls:
            if urls[0].url.endswith('.pdf'):
                mimetype = 'application/pdf'

        fe = fatcat_client.FileEntity(
            md5=obj.get('md5'),
            sha1=obj['sha1'],
            sha256=obj.get('sha256'),
            size=size,
            mimetype=mimetype,
            release_ids=release_ids,
            urls=urls,
        )
        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_client.rest.ApiException as err:
            if err.status != 404:
                raise err

        if not existing:
            return True

        fe.release_ids = list(set(fe.release_ids + existing.release_ids))
        if set(fe.release_ids) == set(existing.release_ids) and len(existing.urls) > 0:
            # no new release matches *and* there are already existing URLs
            self.counts['exists'] += 1
            return False

        # merge the existing into this one and update
        existing.urls = list(set([(u.rel, u.url) for u in fe.urls + existing.urls]))
        existing.urls = [fatcat_client.FileUrl(rel=rel, url=url) for (rel, url) in existing.urls]
        if len(existing.urls) > SANE_MAX_URLS:
            self.counts['skip-update-too-many-url'] += 1
            return None
        existing.release_ids = list(set(fe.release_ids + existing.release_ids))
        if len(existing.release_ids) > SANE_MAX_RELEASES:
            self.counts['skip-update-too-many-releases'] += 1
            return None
        existing.mimetype = existing.mimetype or fe.mimetype
        existing.size = existing.size or fe.size
        existing.md5 = existing.md5 or fe.md5
        existing.sha1 = existing.sha1 or fe.sha1
        existing.sha256 = existing.sha256 or fe.sha256
        self.api.update_file(self.get_editgroup_id(), existing.ident, existing)
        self.counts['update'] += 1
        return False

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