summaryrefslogtreecommitdiffstats
path: root/python/fatcat_tools/importers/doaj_article.py
blob: 30007857153ab9e7814a76902d3238922479acc8 (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
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
"""
Importer for DOAJ article-level metadata, schema v1.

DOAJ API schema and docs: https://doaj.org/api/v1/docs
"""

import warnings
import datetime
from typing import List, Optional

import fatcat_openapi_client
from fatcat_tools.normal import (clean_doi, clean_str, parse_month,
    clean_orcid, detect_text_lang, parse_lang_name, parse_country_name,
    clean_pmid, clean_pmcid)
from fatcat_tools.importers.common import EntityImporter

# Cutoff length for abstracts.
MAX_ABSTRACT_LENGTH = 2048


class DoajArticleImporter(EntityImporter):

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

        eg_desc = kwargs.get(
            'editgroup_description',
            "Automated import of DOAJ article metadata, harvested from REST API or bulk dumps"
        )
        eg_extra = kwargs.get('editgroup_extra', dict())
        eg_extra['agent'] = eg_extra.get('agent',
                                         'fatcat_tools.DoajArticleImporter')
        # ensure default is to not do updates with this worker (override super() default)
        kwargs['do_updates'] = kwargs.get("do_updates", False)
        super().__init__(api,
                         issn_map_file=issn_map_file,
                         editgroup_description=eg_desc,
                         editgroup_extra=eg_extra,
                         **kwargs)

        self.this_year = datetime.datetime.now().year
        self.read_issn_map_file(issn_map_file)

    def want(self, obj):
        return True

    def parse_record(self, obj):
        """
        bibjson {
            abstract (string, optional),
            author (Array[bibjson.author], optional),
            identifier (Array[bibjson.identifier]),
            journal (bibjson.journal, optional),
            keywords (Array[string], optional),
            link (Array[bibjson.link], optional),
            month (string, optional),
            subject (Array[bibjson.subject], optional),
            title (string),
            year (string, optional)
        }
        bibjson.journal {
            country (string, optional),
            end_page (string, optional),
            language (Array[string], optional),
            license (Array[bibjson.journal.license], optional),
            number (string, optional),
            publisher (string, optional),
            start_page (string, optional),
            title (string, optional),
            volume (string, optional)
        }
        """

        if not obj or not isinstance(obj, dict) or not 'bibjson' in obj:
            self.counts['skip-empty'] += 1
            return None

        bibjson = obj['bibjson']

        title = clean_str(bibjson.get('title'), force_xml=True)
        if not title:
            self.counts['skip-title'] += 1
            return False

        container_name = clean_str(bibjson['journal']['title'])
        container_id = None
        # NOTE: 'issns' not documented in API schema
        for issn in bibjson['journal']['issns']:
            issnl = self.issn2issnl(issn)
            if issnl:
                container_id = self.lookup_issnl(self.issn2issnl(issn))
            if container_id:
                # don't store container_name when we have an exact match
                container_name = None
                break

        volume = clean_str(bibjson['journal'].get('volume'))
        # NOTE: this schema seems to use "number" as "issue number"
        issue = clean_str(bibjson['journal'].get('number'))
        publisher = clean_str(bibjson['journal'].get('publisher'))

        try:
            release_year = int(bibjson.get('year'))
        except (TypeError, ValueError):
            release_year = None
        release_month = parse_month(clean_str(bibjson.get('month')))

        # block bogus far-future years/dates
        if release_year is not None and (release_year > (self.this_year + 5) or release_year < 1000):
            release_month = None
            release_year = None

        license_slug = self.doaj_license_slug(bibjson['journal'].get('license'))
        country = parse_country_name(bibjson['journal'].get('country'))
        language = None
        for raw in bibjson['journal'].get('language') or []:
            language = parse_lang_name(raw)
            if language:
                break

        # pages
        # NOTE: error in API docs? seems like start_page not under 'journal' object
        start_page = clean_str(bibjson['journal'].get('start_page')) or clean_str(bibjson.get('start_page'))
        end_page = clean_str(bibjson['journal'].get('end_page')) or clean_str(bibjson.get('end_page'))
        pages: Optional[str] = None
        if start_page and end_page:
            pages = f"{start_page}-{end_page}"
        elif start_page:
            pages = start_page

        doaj_article_id = obj['id'].lower()
        ext_ids = self.doaj_ext_ids(bibjson['identifier'], doaj_article_id)
        abstracts = self.doaj_abstracts(bibjson)
        contribs = self.doaj_contribs(bibjson.get('author') or [])

        # DOAJ-specific extra
        doaj_extra = dict()
        if bibjson.get('subject'):
            doaj_extra['subject'] = bibjson.get('subject')
        if bibjson.get('keywords'):
            doaj_extra['keywords'] = [k for k in [clean_str(s) for s in bibjson.get('keywords')] if k]

        # generic extra
        extra = dict()
        if country:
            extra['country'] = country
        if not container_id and container_name:
            extra['container_name'] = container_name
        if release_year and release_month:
            # TODO: schema migration
            extra['release_month'] = release_month

        if doaj_extra:
            extra['doaj'] = doaj_extra
        if not extra:
            extra = None

        re = fatcat_openapi_client.ReleaseEntity(
            work_id=None,
            container_id=container_id,
            release_type='article-journal',
            release_stage='published',
            title=title,
            release_year=release_year,
            #release_date,
            publisher=publisher,
            ext_ids=ext_ids,
            contribs=contribs,
            volume=volume,
            issue=issue,
            pages=pages,
            language=language,
            abstracts=abstracts,
            extra=extra,
            license_slug=license_slug,
        )
        re = self.biblio_hacks(re)
        return re

    @staticmethod
    def biblio_hacks(re):
        """
        This function handles known special cases. For example,
        publisher-specific or platform-specific workarounds.
        """
        return re

    def try_update(self, re):

        # lookup existing release by DOAJ article id
        existing = None
        try:
            existing = self.api.lookup_release(doaj=re.ext_ids.doaj)
        except fatcat_openapi_client.rest.ApiException as err:
            if err.status != 404:
                raise err
            # doesn't exist, need to update
            return True

        # then try other ext_id lookups
        if not existing:
            for extid_type in ('doi', 'pmid', 'pmcid'):
                extid_val = re.ext_ids.__dict__[extid_type]
                if not extid_val:
                    continue
                try:
                    existing = self.api.lookup_release(**{extid_type: extid_val})
                except fatcat_openapi_client.rest.ApiException as err:
                    if err.status != 404:
                        raise err
                if existing:
                    if existing.ext_ids.doaj:
                        warn_str = f"unexpected DOAJ ext_id match after lookup failed doaj={re.ext_ids.doaj} ident={existing.ident}"
                        warnings.warn(warn_str)
                        self.counts["skip-doaj-id-mismatch"] += 1
                        return None
                    break

        # TODO: in the future could do fuzzy match here, eg using elasticsearch

        # create entity
        if not existing:
            return True

        # other logic could go here about skipping updates
        if not self.do_updates or existing.ext_ids.doaj:
            self.counts['exists'] += 1
            return False

        # fields to copy over for update
        existing.ext_ids.doaj = existing.ext_ids.doaj or re.ext_ids.doaj
        existing.release_type = existing.release_type or re.release_type
        existing.release_stage = existing.release_stage or re.release_stage
        existing.container_id = existing.container_id or re.container_id
        existing.abstracts = existing.abstracts or re.abstracts
        existing.extra['doaj'] = re.extra['doaj']
        existing.volume = existing.volume or re.volume
        existing.issue = existing.issue or re.issue
        existing.pages = existing.pages or re.pages
        existing.language = existing.language or re.language

        try:
            self.api.update_release(self.get_editgroup_id(), existing.ident, existing)
            self.counts['update'] += 1
        except fatcat_openapi_client.rest.ApiException as err:
            # there is a code path where we try to update the same release
            # twice in a row; if that happens, just skip
            # NOTE: API behavior might change in the future?
            if "release_edit_editgroup_id_ident_id_key" in err.body:
                self.counts['skip-update-conflict'] += 1
                return False
            else:
                raise err

        return False

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

    def doaj_abstracts(self, bibjson: dict) -> List[fatcat_openapi_client.ReleaseAbstract]:
        text = clean_str(bibjson.get('abstract'))
        if not text or len(text) < 10:
            return []
        if len(text) > MAX_ABSTRACT_LENGTH:
            text = text[:MAX_ABSTRACT_LENGTH] + " [...]"

        lang = detect_text_lang(text)

        abstract = fatcat_openapi_client.ReleaseAbstract(
            mimetype="text/plain",
            content=text,
            lang=lang,
        )

        return [abstract,]

    def doaj_contribs(self, authors: List[dict]) -> List[fatcat_openapi_client.ReleaseContrib]:
        """
        bibjson.author {
            affiliation (string, optional),
            name (string),
            orcid_id (string, optional)
        }
        """
        contribs = []
        index = 0
        for author in authors:
            if not author.get('name'):
                continue
            creator_id = None
            orcid = clean_orcid(author.get('orcid_id'))
            if orcid:
                creator_id = self.lookup_orcid(orcid)
            contribs.append(fatcat_openapi_client.ReleaseContrib(
                raw_name=author.get('name'),
                role='author',
                index=index,
                creator_id=creator_id,
                raw_affiliation=clean_str(author.get('affiliation')),
            ))
            index += 1
        return contribs

    def doaj_ext_ids(self, identifiers: List[dict], doaj_article_id: str) -> fatcat_openapi_client.ReleaseExtIds:
        """
        bibjson.identifier {
            id (string),
            type (string)
        }
        """

        assert doaj_article_id.isalnum() and len(doaj_article_id) == 32

        doi: Optional[str] = None
        pmid: Optional[str] = None
        pmcid: Optional[str] = None
        for id_obj in identifiers:
            if id_obj['type'].lower() == 'doi':
                doi = clean_doi(id_obj['id'])
            elif id_obj['type'].lower() == 'pmid':
                pmid = clean_pmid(id_obj['id'])
            elif id_obj['type'].lower() == 'pmcid':
                pmcid = clean_pmcid(id_obj['id'])

        return fatcat_openapi_client.ReleaseExtIds(
            doaj=doaj_article_id,
            doi=doi,
            pmid=pmid,
            pmcid=pmcid,
        )

    def doaj_license_slug(self, license_list: List[dict]) -> Optional[str]:
        """
        bibjson.journal.license {
            open_access (boolean, optional),
            title (string, optional),
            type (string, optional),
            url (string, optional),
            version (string, optional)
        }
        """
        if not license_list:
            return None
        for license in license_list:
            if not license.get('open_access'):
                continue
            slug = license.get('type')
            if slug.startswith('CC '):
                slug = slug.replace('CC ', 'cc-').lower()
                return slug
        return None