aboutsummaryrefslogtreecommitdiffstats
path: root/python/fatcat_tools/importers/datacite.py
blob: aaf1af2c44e2b9669a1a41da09192488c55611f9 (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
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
"""
Prototype importer for datacite.org data.

Example input document at: https://gist.github.com/miku/5610a2d64e3fee82d16f5d3f3a295fc8.

Datacite being an aggregator, the data is varied and exposes a couple of
problems in content and structure. A few fields habe their own parsing
functions (parse_datacite_...), which can be tested more easily.
"""

import collections
import datetime
import hashlib
import json
import sqlite3
import sys

import dateparser
import fatcat_openapi_client
import langdetect
import pycountry

from fatcat_tools.normal import clean_doi
from fatcat_tools.transforms import entity_to_dict

from .common import EntityImporter, clean

# Cutoff length for abstracts.
MAX_ABSTRACT_LENGTH = 2048

# https://guide.fatcat.wiki/entity_container.html#container_type-vocabulary
CONTAINER_TYPE_MAP = {
    'Journal': 'journal',
    'Series': 'journal',
    'Book Series': 'book-series',
}

# The docs/guide should be the cannonical home for these mappings; update there
# first.  Map various datacite type types to CSL-ish types. None means TODO or
# remove.
DATACITE_TYPE_MAP = {
    'ris': {
        'THES': 'thesis',
        'SOUND': 'song', # 99.9% maps to citeproc song, so use that (exception: report)
        'CHAP': 'chapter',
        'FIGURE': 'figure',
        'RPRT': 'report',
        'JOUR': 'article-journal',
        'MPCT': 'motion_picture',
        'GEN': 'article-journal', # GEN consist of 99% article and report, post-weblog, misc - and one dataset
        'BOOK': 'book',
        'DATA': 'dataset',
        'COMP': 'software',
    },
    'schemaOrg': {
        'Dataset': 'dataset',
        'Book': 'book',
        'ScholarlyArticle': 'article-journal',
        'ImageObject': 'graphic',
        'Collection': None,
        'MediaObject': None,
        'Event': None,
        'SoftwareSourceCode': 'software',
        'Chapter': 'chapter',
        'CreativeWork': None, # Seems to be a catch-all resourceType, from PGRFA Material, Pamphlet, to music score.
        'PublicationIssue': 'article',
        'AudioObject': None,
        'Thesis': 'thesis',
    },
    'citeproc': {
        'article': 'article',
        'article-journal': 'article-journal',
        'article-magazine': 'article-magazine',
        'article-newspaper': 'article-newspaper',
        'bill': 'bill',
        'book': 'book',
        'broadcast': 'broadcast',
        'chapter': 'chapter',
        'dataset': 'dataset',
        'entry-dictionary': 'entry-dictionary',
        'entry-encyclopedia': 'entry-encyclopedia',
        'entry': 'entry',
        'figure': 'figure',
        'graphic': 'graphic',
        'interview': 'interview',
        'legal_case': 'legal_case',
        'legislation': 'legislation',
        'manuscript': 'manuscript',
        'map': 'map',
        'motion_picture': 'motion_picture',
        'musical_score': 'musical_score',
        'pamphlet': 'pamphlet',
        'paper-conference': 'paper-conference',
        'patent': 'patent',
        'personal_communication': 'personal_communication',
        'post': 'post',
        'post-weblog': 'post-weblog',
        'report': 'report',
        'review-book': 'review-book',
        'review': 'review',
        'song': 'song',
        'speech': 'speech',
        'thesis': 'thesis',
        'treaty': 'treaty',
        'webpage': 'webpage',
    },  # https://docs.citationstyles.org/en/master/specification.html#appendix-iii-types
    'bibtex': {
        'phdthesis': 'thesis',
        'inbook': 'chapter',
        'misc': None,
        'article': 'article-journal',
        'book': 'book',
    },
    'resourceTypeGeneral': {
        'Image': 'graphic',
        'Dataset': 'dataset',
        'PhysicalObject': None,
        'Collection': None,
        'Text': None, # "Greyliterature, labnotes, accompanyingmaterials"
        'Sound': None,
        'InteractiveResource': None,
        'Event': None,
        'Software': 'software',
        'Other': None,
        'Workflow': None,
        'Audiovisual': None,
    } # https://schema.datacite.org/meta/kernel-4.0/doc/DataCite-MetadataKernel_v4.0.pdf#page=32
}

# DATACITE_UNKNOWN_MARKERS via https://support.datacite.org/docs/schema-values-unknown-information-v43.
DATACITE_UNKNOWN_MARKERS = (
    '(:unac)',  # temporarily inaccessible
    '(:unal)',  # unallowed, suppressed intentionally
    '(:unap)',  # not applicable, makes no sense
    '(:unas)',  # value unassigned (e.g., Untitled)
    '(:unav)',  # value unavailable, possibly unknown
    '(:unkn)',  # known to be unknown (e.g., Anonymous, Inconnue)
    '(:none)',  # never had a value, never will
    '(:null)',  # explicitly and meaningfully empty
    '(:tba)',  # to be assigned or announced later
    '(:etal)',  # too numerous to list (et alia)
)

# UNKNOWN_MARKERS joins official datacite markers with a generic tokens marking
# unknown values.
UNKNOWN_MARKERS = set(DATACITE_UNKNOWN_MARKERS).union(set((
    'NA',
    'NN',
    'n.a.',
    '[s.n.]',
)))

# TODO(martin): merge this with other maps, maybe.
LICENSE_SLUG_MAP = {
    "//creativecommons.org/licenses/by/2.0/": "CC-BY",
    "//creativecommons.org/licenses/by/2.0/uk/legalcode": "CC-BY",
    "//creativecommons.org/licenses/by/3.0/": "CC-BY",
    "//creativecommons.org/licenses/by/3.0/us": "CC-BY",
    "//creativecommons.org/licenses/by/4.0/": "CC-BY",
    "//creativecommons.org/licenses/by/4.0/deed.de/": "CC-BY",
    "//creativecommons.org/licenses/by/4.0/deed.en_US/": "CC-BY",
    "//creativecommons.org/licenses/by/4.0/legalcode/": "CC-BY",
    "//creativecommons.org/licenses/by-nc/2.0/": "CC-BY-NC",
    "//creativecommons.org/licenses/by-nc/3.0/": "CC-BY-NC",
    "//creativecommons.org/licenses/by-nc/4.0/": "CC-BY-NC",
    "//creativecommons.org/licenses/by-nc/4.0/legalcode": "CC-BY-NC",
    "//creativecommons.org/licenses/by-nc-nd/3.0/": "CC-BY-NC-ND",
    "//creativecommons.org/licenses/by-nc-nd/3.0/gr": "CC-BY-NC-ND",
    "//creativecommons.org/licenses/by-nc-nd/4.0/": "CC-BY-ND",
    "//creativecommons.org/licenses/by-nc-nd/4.0/legalcode": "CC-BY-ND",
    "//creativecommons.org/licenses/by-nc-sa/4.0/": "CC-BY-NC-SA",
    "//creativecommons.org/licenses/by-nd/4.0/": "CC-BY-ND",
    "//creativecommons.org/licenses/by-sa/3.0/de": "CC-BY-SA",
    "//creativecommons.org/licenses/by-sa/3.0/gr": "CC-BY-SA",
    "//creativecommons.org/licenses/by-sa/4.0/": "CC-BY-SA",
    "//creativecommons.org/licenses/by-sa/4.0/legalcode": "CC-BY-SA",
    "//creativecommons.org/licenses/CC-BY/4.0/": "CC-BY",
    "//creativecommons.org/licenses/publicdomain/zero/1.0/": "CC-0",
    "//creativecommons.org/publicdomain/zero/1.0/": "CC-0",
    "//creativecommons.org/publicdomain/zero/1.0/legalcode": "CC-0",
    "//opensource.org/licenses/MIT": "MIT",
    "//www.elsevier.com/open-access/userlicense/1.0": "ELSEVIER-USER-1.0",
    "//www.gnu.org/licenses/gpl-3.0.en.html": "GPLv3",
    "//www.gnu.org/licenses/old-licenses/gpl-2.0.en.html": "GPLv2",
    "//www.karger.com/Services/SiteLicenses": "KARGER",
    "//www.opensource.org/licenses/Apache-2.0": "Apache-2.0",
    "//www.opensource.org/licenses/BSD-3-Clause": "BSD-3-Clause",
    "//www.opensource.org/licenses/EUPL-1.1":
    "EUPL-1.1",  # redirects to EUPL-1.2
    "//www.opensource.org/licenses/MIT": "MIT",
    # "http://royalsocietypublishing.org/licence": "", # OA and "normal", https://royalsociety.org/journals/authors/licence-to-publish/
    # "http://rsc.li/journals-terms-of-use": "RSC",
    # "http://www.fu-berlin.de/sites/refubium/rechtliches/Nutzungsbedingungen": "", # 53 UrhG.
    # "http://www.nrcresearchpress.com/page/about/CorporateTextAndDataMining": "",
    # "http://www.springer.com/tdm": "",
    # "https://cds.unistra.fr/vizier-org/licences_vizier.html": "", # Maybe try to "SPN" those: https://web.archive.org/web/*/https://cds.unistra.fr/vizier-org/licences_vizier.html
    # "https://link.aps.org/licenses/aps-default-accepted-manuscript-license": "",
    # "https://oparu.uni-ulm.de/xmlui/license_opod_v1": "",
    # "https://publikationen.bibliothek.kit.edu/kitopen-lizenz": "",
    # "https://rightsstatements.org/page/InC/1.0?language=en": "",
    # "https://services.ceda.ac.uk/cedasite/register/info": "",
    # "https://wdc.dlr.de/ndmc/userfiles/file/NDMC-Data_Sharing_Principles.pdf": "", # 404
    # "https://www.cambridge.org/core/terms": "",
    # "https://www.elsevier.com/tdm/userlicense/1.0",
    # "info:eu-repo/semantics/closedAccess": "", # https://wiki.surfnet.nl/display/standards/info-eu-repo/#info-eu-repo-AccessRights
    # "info:eu-repo/semantics/embargoedAccess": "",
    # "info:eu-repo/semantics/openAccess": "",
    # Note: Some URLs pointing to licensing terms are not in WB yet (but would be nice).
}

# TODO(martin): drop this after 3.7 upgrade
try:
    isascii = str.isascii # new in 3.7, https://docs.python.org/3/library/stdtypes.html#str.isascii
except AttributeError:
    isascii = lambda s: len(s) == len(s.encode())


class DataciteImporter(EntityImporter):
    """
    Importer for datacite records.
    """
    def __init__(self,
                 api,
                 issn_map_file,
                 debug=False,
                 insert_log_file=None,
                 **kwargs):

        eg_desc = kwargs.get(
            'editgroup_description',
            "Automated import of Datacite DOI metadata, harvested from REST API"
        )
        eg_extra = kwargs.get('editgroup_extra', dict())
        eg_extra['agent'] = eg_extra.get('agent',
                                         'fatcat_tools.DataciteImporter')
        super().__init__(api,
                         issn_map_file=issn_map_file,
                         editgroup_description=eg_desc,
                         editgroup_extra=eg_extra,
                         **kwargs)

        self.create_containers = kwargs.get('create_containers', True)
        extid_map_file = kwargs.get('extid_map_file')
        self.extid_map_db = None
        if extid_map_file:
            db_uri = "file:{}?mode=ro".format(extid_map_file)
            print("Using external ID map: {}".format(db_uri), file=sys.stderr)
            self.extid_map_db = sqlite3.connect(db_uri, uri=True)
        else:
            print("Not using external ID map", file=sys.stderr)

        self.read_issn_map_file(issn_map_file)
        self.debug = debug
        self.insert_log_file = insert_log_file

        print('datacite with debug={}'.format(self.debug), file=sys.stderr)

    def lookup_ext_ids(self, doi):
        """
        Return dictionary of identifiers refering to the same things as the given DOI.
        """
        if self.extid_map_db is None:
            return dict(core_id=None,
                        pmid=None,
                        pmcid=None,
                        wikidata_qid=None,
                        arxiv_id=None,
                        jstor_id=None)
        row = self.extid_map_db.execute(
            "SELECT core, pmid, pmcid, wikidata FROM ids WHERE doi=? LIMIT 1",
            [doi.lower()]).fetchone()
        if row is None:
            return dict(core_id=None,
                        pmid=None,
                        pmcid=None,
                        wikidata_qid=None,
                        arxiv_id=None,
                        jstor_id=None)
        row = [str(cell or '') or None for cell in row]
        return dict(
            core_id=row[0],
            pmid=row[1],
            pmcid=row[2],
            wikidata_qid=row[3],
            # TODO:
            arxiv_id=None,
            jstor_id=None,
        )

    def parse_record(self, obj):
        """
        Mapping datacite JSON to ReleaseEntity.
        """
        if not obj or not isinstance(obj, dict):
            return None
        if 'attributes' not in obj:
            return None

        attributes = obj['attributes']
        doi = clean_doi(attributes.get('doi', '').lower())

        if not isascii(doi):
            print('[{}] skipping non-ascii doi for now'.format(doi))
            return None

        # Contributors. Many nameIdentifierSchemes, we do not use (yet):
        # "attributes.creators[].nameIdentifiers[].nameIdentifierScheme":
        # ["LCNA", "GND", "email", "NAF", "OSF", "RRID", "ORCID",
        # "SCOPUS", "NRCPID", "schema.org", "GRID", "MGDS", "VIAF", "JACoW-ID"].
        contribs = []

        # Names, that should be ignored right away.
        name_blacklist = set(('Occdownload Gbif.Org',))

        for i, c in enumerate(attributes['creators']):
            nameType = c.get('nameType', '') or ''
            if nameType in ('', 'Personal'):
                creator_id = None
                for nid in c.get('nameIdentifiers', []):
                    name_scheme = nid.get('nameIdentifierScheme', '') or ''
                    if not name_scheme.lower() == "orcid":
                        continue
                    orcid = nid.get('nameIdentifier',
                                    '').replace('https://orcid.org/', '')
                    if not orcid:
                        continue
                    creator_id = self.lookup_orcid(orcid)
                    # TODO(martin): If creator_id is None, should we create creators?

                # If there are multiple affiliation strings, use the first one.
                affiliations = c.get('affiliation', []) or []
                raw_affiliation = None
                if len(affiliations) == 0:
                    raw_affiliation = None
                else:
                    raw_affiliation = clean(affiliations[0])

                name = c.get('name')
                given_name = c.get('givenName')
                surname = c.get('familyName')

                if name:
                    name = clean(name)

                if name in name_blacklist:
                    continue

                if given_name:
                    given_name = clean(given_name)

                if surname:
                    surname = clean(surname)

                if not name:
                    continue

                if raw_affiliation == '':
                    continue

                if name.lower() in UNKNOWN_MARKERS:
                    continue

                # Unpack name, if we have an index form (e.g. 'Razis, Panos A') into 'Panos A razis'.
                if name:
                    name = index_form_to_display_name(name)

                contribs.append(
                    fatcat_openapi_client.ReleaseContrib(
                        creator_id=creator_id,
                        index=i,
                        raw_name=name,
                        given_name=given_name,
                        surname=surname,
                        role='author',
                        raw_affiliation=raw_affiliation,
                    ))
            elif nameType == 'Organizational':
                name = c.get('name', '') or ''
                if name in UNKNOWN_MARKERS:
                    continue
                if len(name) < 3:
                    continue
                extra = {'organization': name}
                contribs.append(fatcat_openapi_client.ReleaseContrib(
                    index=i, extra=extra))
            else:
                print('[{}] unknown name type: {}'.format(doi, nameType), file=sys.stderr)

        # Title, may come with "attributes.titles[].titleType", like
        # "AlternativeTitle", "Other", "Subtitle", "TranslatedTitle"
        titles = attributes.get('titles', []) or []
        title, original_language_title, subtitle = parse_datacite_titles(
            titles)

        if title is None:
            print('[{}] skipping record w/o title: {}'.format(doi, obj), file=sys.stderr)
            return False

        title = clean(title)
        if not title:
            print('[{}] skipping record w/o title: {}'.format(doi, obj), file=sys.stderr)
            return False

        if not subtitle:
            subtitle = None
        else:
            subtitle = clean(subtitle)

        # Dates. A few internal dates (registered, created, updated) and
        # published (0..2554). We try to work with typed date list, in
        # "attributes.dates[].dateType", values: "Accepted", "Available"
        # "Collected", "Copyrighted", "Created", "Issued", "Submitted",
        # "Updated", "Valid".
        release_date, release_month, release_year = parse_datacite_dates(
            attributes.get('dates', []))

        # Start with clear stages, e.g. published. TODO(martin): we could
        # probably infer a bit more from the relations, e.g.
        # "IsPreviousVersionOf" or "IsNewVersionOf".
        release_stage = 'published'

        # TODO(martin): If 'state' is not 'findable' or 'isActive' is not true,
        # we might want something else than 'published'. See also:
        # https://support.datacite.org/docs/doi-states.

        # Publisher. A few NA values. A few bogus values.
        publisher = attributes.get('publisher')

        if publisher in UNKNOWN_MARKERS | set(('Unpublished', 'Unknown')):
            publisher = None
            release_stage = None
        if publisher is not None and len(publisher) > 80:
            # Arbitrary magic value max length. TODO(martin): better heuristic,
            # but factored out; first we have to log misses. Example:
            # "ETH-Bibliothek Zürich, Bildarchiv / Fotograf: Feller,
            # Elisabeth, Empfänger, Unbekannt, Fotograf / Fel_041033-RE /
            # Unbekannt, Nutzungsrechte müssen durch den Nutzer abgeklärt
            # werden"
            publisher = None

        if publisher:
            publisher = clean(publisher)

        # Container. For the moment, only ISSN as container.
        container_id = None
        container_name = None

        container = attributes.get('container', {}) or {}
        if container.get('type') in CONTAINER_TYPE_MAP.keys():
            container_type = CONTAINER_TYPE_MAP.get(container['type'])
            if container.get('identifier') and container.get(
                    'identifierType') == 'ISSN':
                issn = container.get('identifier')
                if len(issn) == 8:
                    issn = issn[:4] + "-" + issn[4:]
                issnl = self.issn2issnl(issn)
                if issnl is not None:
                    container_id = self.lookup_issnl(issnl)

                    if container_id is None and container.get('title'):
                        container_name = container.get('title')
                        if isinstance(container_name, list):
                            if len(container_name) > 0:
                                print('[{}] too many container titles: {}'.format(doi,
                                    len(container_name)))
                                container_name = container_name[0]
                        assert isinstance(container_name, str)
                        ce = fatcat_openapi_client.ContainerEntity(
                            issnl=issnl,
                            container_type=container_type,
                            name=container_name,
                        )
                        ce_edit = self.create_container(ce)
                        container_id = ce_edit.ident
                        self._issnl_id_map[issnl] = container_id
                else:
                    # TODO(martin): factor this out into a testable function.
                    # TODO(martin): "container_name": "№1(1) (2018)" / 10.26087/inasan.2018.1.1.013
                    container_name = container.get('title')
                    if isinstance(container_name, list):
                        if len(container_name) > 0:
                            print('[{}] too many container titles: {}'.format(doi,
                                len(container_name)))
                            container_name = container_name[0]

        # Volume and issue.
        volume = container.get('volume')
        issue = container.get('issue')

        if volume:
            volume = clean(volume)

        if issue:
            issue = clean(issue)

        # Pages.
        pages = None

        first_page = container.get('firstPage')
        last_page = container.get('lastPage')

        if first_page and last_page:
            try:
                _ = int(first_page) < int(last_page)
                pages = '{}-{}'.format(first_page, last_page)
            except ValueError as err:
                # TODO(martin): This is more debug than info.
                # print('[{}] {}'.format(doi, err), file=sys.stderr)
                pass

        if not pages and first_page:
            pages = first_page

        # License.
        license_slug = None
        license_extra = []

        for l in attributes.get('rightsList', []):
            slug = lookup_license_slug(l.get('rightsUri'))
            if slug:
                license_slug = slug
            license_extra.append(l)

        # Release type. Try to determine the release type from a variety of
        # types supplied in datacite. The "attributes.types.resourceType" is
        # uncontrolled (170000+ unique values, from "null", "Dataset" to
        # "Jupyter Notebook" and "Macroseismic Data Points" or "2 days of IP
        # flows in 2009") citeproc may be the closest, but not always supplied.
        # Order lookup roughly by completeness of mapping.
        for typeType in ('citeproc', 'ris', 'schemaOrg', 'bibtex', 'resourceTypeGeneral'):
            value = attributes.get('types', {}).get(typeType)
            release_type = DATACITE_TYPE_MAP.get(typeType, {}).get(value)
            if release_type is not None:
                break

        if release_type is None:
            print("[{}] no mapped type: {}".format(doi, value), file=sys.stderr)

        # release_type exception: Global Biodiversity Information Facility
        # publishes highly interesting datasets, but titles are mostly the same
        # ("GBIF Occurrence Download" or "Occurrence Download"); set
        # release_type to "stub" (CSL/FC).
        if publisher == 'The Global Biodiversity Information Facility':
            release_type = 'stub'

        # release_type exception: lots of "Experimental Crystal Structure Determination"
        if publisher == 'Cambridge Crystallographic Data Centre':
            release_type = 'entry'

        # Supplement files, e.g. "Additional file 1: ASE constructs in questionnaire."
        if title.lower().startswith('additional file'):
            release_type = 'stub'

        # Language values are varied ("ger", "es", "English", "ENG", "en-us",
        # "other", ...). Try to crush it with langcodes: "It may sound to you
        # like langcodes solves a pretty boring problem. At one level, that's
        # right. Sometimes you have a boring problem, and it's great when a
        # library solves it for you." -- TODO(martin): We need more of these.
        language = None

        value = attributes.get('language', '') or ''
        try:
            language = pycountry.languages.lookup(value).alpha_2
        except (LookupError, AttributeError) as err:
            pass
            # TODO(martin): Print this on debug level, only.
            # print('[{}] language lookup miss for {}: {}'.format(doi, value, err), file=sys.stderr)

        # Abstracts appear in "attributes.descriptions[].descriptionType", some
        # of the observed values: "Methods", "TechnicalInfo",
        # "SeriesInformation", "Other", "TableOfContents", "Abstract". The
        # "Other" fields might contain references or related articles (with
        # DOI). TODO(martin): maybe try to parse out some of those refs.
        abstracts = []
        descs = attributes.get('descriptions', []) or []
        for desc in descs:
            if not desc.get('descriptionType') == 'Abstract':
                continue
            if len(desc.get('description', '') or '') < 10:
                continue
            text = desc.get('description', '')
            if len(text) > MAX_ABSTRACT_LENGTH:
                text = text[:MAX_ABSTRACT_LENGTH] + " [...]"
            lang = None
            try:
                lang = langdetect.detect(text)
            except langdetect.lang_detect_exception.LangDetectException as err:
                print('[{}] language detection failed: {}'.format(doi, err),
                      file=sys.stderr)
            abstracts.append(
                fatcat_openapi_client.ReleaseAbstract(
                    mimetype="text/plain",
                    content=clean(text),
                    lang=lang,
                ))

        # References and relations. Datacite include many relation types in
        # "attributes.relatedIdentifiers[].relationType", e.g.
        # "IsPartOf", "IsPreviousVersionOf", "Continues", "IsVariantFormOf",
        # "IsSupplementTo", "Cites", "IsSupplementedBy", "IsDocumentedBy", "HasVersion",
        # "IsCitedBy", "IsMetadataFor", "IsNewVersionOf", "IsIdenticalTo", "HasPart",
        # "References", "Reviews", "HasMetadata", "IsContinuedBy", "IsVersionOf",
        # "IsDerivedFrom", "IsSourceOf".
        #
        # For the moment, we only care about References.
        refs, ref_index = [], 0

        relIds = attributes.get('relatedIdentifiers', []) or []
        for rel in relIds:
            if not rel.get('relationType', '') in ('References', 'Cites'):
                continue
            ref_extra = dict()
            if rel.get('relatedIdentifierType', '') == 'DOI':
                ref_extra['doi'] = rel.get('relatedIdentifier')
            if not ref_extra:
                ref_extra = None
            refs.append(
                fatcat_openapi_client.ReleaseRef(
                    index=ref_index,
                    extra=ref_extra,
                ))
            ref_index += 1

        # More specific release_type via 'Reviews' relationsship.
        for rel in relIds:
            if rel.get('relatedIdentifierType', '') != 'Reviews':
                continue
            release_type = 'review'

        # Extra information.
        extra_datacite = dict()

        if license_extra:
            extra_datacite['license'] = license_extra
        if attributes.get('subjects'):
            extra_datacite['subjects'] = attributes['subjects']

        # Include version information.
        metadata_version = attributes.get('metadataVersion') or ''
        schema_version = attributes.get('schemaVersion') or ''

        if metadata_version:
            extra_datacite['metadataVersion'] = metadata_version
        if schema_version:
            extra_datacite['schemaVersion'] = schema_version

        # Include resource types.
        types = attributes.get('types', {}) or {}
        resource_type = types.get('resourceType', '') or ''
        resource_type_general = types.get('resourceTypeGeneral', '') or ''

        if resource_type:
            extra_datacite['resourceType'] = resource_type
        if resource_type_general:
            extra_datacite['resourceTypeGeneral'] = resource_type_general

        # Include certain relations from relatedIdentifiers. Keeping the
        # original structure of data here, which is a list of dicts, with
        # relation type, identifer and identifier type (mostly).
        relations = []
        for rel in relIds:
            if rel.get('relationType') in ('IsPartOf', 'Reviews', 'Continues',
                                           'IsVariantFormOf', 'IsSupplementTo',
                                           'HasVersion', 'IsMetadataFor',
                                           'IsNewVersionOf', 'IsIdenticalTo',
                                           'IsVersionOf', 'IsDerivedFrom',
                                           'IsSourceOf'):
                relations.append(rel)

        if relations:
            extra_datacite['relations'] = relations

        extra = dict()

        # "1.0.0", "v1.305.2019", "Final", "v1.0.0", "v0.3.0", "1", "0.19.0",
        # "3.1", "v1.1", "{version}", "4.0", "10329", "11672", "11555",
        # "v1.4.5", "2", "V1", "v3.0", "v0", "v0.6", "11124", "v1.0-beta", "1st
        # Edition", "20191024", "v2.0.0", "v0.9.3", "10149", "2.0", null,
        # "v0.1.1", "3.0", "1.0", "3", "v1.12.2", "20191018", "v0.3.1", "v1.0",
        # "10161", "10010691", "10780", # "Presentación"
        version = attributes.get('version')

        # top-level extra keys
        if not container_id and container_name:
            extra['container_name'] = container_name

        # Always include datacite key, even if value is empty (dict).
        extra['datacite'] = extra_datacite

        # Preparation for a schema update.
        if release_month:
            extra['release_month'] = release_month

        extids = self.lookup_ext_ids(doi=doi)

        # Assemble release.
        re = fatcat_openapi_client.ReleaseEntity(
            work_id=None,
            container_id=container_id,
            release_type=release_type,
            release_stage=release_stage,
            title=title,
            subtitle=subtitle,
            original_title=original_language_title,
            release_year=release_year,
            release_date=release_date,
            publisher=publisher,
            ext_ids=fatcat_openapi_client.ReleaseExtIds(
                doi=doi,
                pmid=extids['pmid'],
                pmcid=extids['pmcid'],
                wikidata_qid=extids['wikidata_qid'],
                core=extids['core_id'],
                arxiv=extids['arxiv_id'],
                jstor=extids['jstor_id'],
            ),
            contribs=contribs,
            volume=volume,
            issue=issue,
            pages=pages,
            language=language,
            abstracts=abstracts,
            refs=refs,
            extra=extra,
            license_slug=license_slug,
            version=version,
        )
        return re

    def try_update(self, re):
        """
        When debug is true, write the RE to stdout, not to the database. Might
        hide schema mismatch bugs.
        """
        if self.debug is True:
            print(json.dumps(entity_to_dict(re, api_client=None)))
            return False

        # lookup existing DOI (don't need to try other ext idents for crossref)
        existing = None
        try:
            existing = self.api.lookup_release(doi=re.ext_ids.doi)
        except fatcat_openapi_client.rest.ApiException as err:
            if err.status != 404:
                raise err
            # doesn't exist, need to update
            return True

        # eventually we'll want to support "updates", but for now just skip if
        # entity already exists
        if existing:
            self.counts['exists'] += 1
            return False

        return True

    def insert_batch(self, batch):
        print('inserting batch ({})'.format(len(batch)), file=sys.stderr)
        if self.insert_log_file:
            with open(self.insert_log_file, 'a') as f:
                for doc in batch:
                    json.dump(entity_to_dict(doc, api_client=None), f)
                    f.write('\n')
        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 lookup_license_slug(raw):
    """
    TODO(martin): reuse from or combine with crossref, maybe.
    """
    if not raw:
        return None
    raw = raw.strip().replace('http://', '//').replace('https://', '//')
    if 'creativecommons.org' in raw.lower():
        raw = raw.lower()
        raw = raw.replace('/legalcode', '/').replace('/uk', '')
        if not raw.endswith('/'):
            raw = raw + '/'
    return LICENSE_SLUG_MAP.get(raw)


def find_original_language_title(item, min_length=4, max_questionmarks=3):
    """
    Perform a few checks before returning a potential original language title.

    Example input: {'title': 'Some title', 'original_language_title': 'Some title'}
    """
    if not 'original_language_title' in item:
        return None
    title = item.get('title')
    if not title:
        return None
    original_language_title = item.get('original_language_title')
    if isinstance(original_language_title,
                  str) and title != original_language_title:
        if len(original_language_title) < min_length:
            return None
        if original_language_title.count('?') > max_questionmarks:
            return None
        return original_language_title
    if isinstance(original_language_title, dict):
        content = original_language_title.get('__content__', '') or ''
        if content and content != title and not content.count(
                '?') > max_questionmarks:
            return content
    return None


def parse_datacite_titles(titles):
    """
    Given a list of title items from datacite, return 3-tuple (title,
    original_language_title, subtitle).

    Example input: [{"title": "Meeting Heterogeneity in Consumer Demand"}]
    """
    title, original_language_title, subtitle = None, None, None

    if titles is None:
        return title, original_language_title, subtitle
    if len(titles) == 0:
        return title, original_language_title, subtitle
    elif len(titles) == 1:
        original_language_title = find_original_language_title(titles[0])
        title = titles[0].get('title', '') or ''
        title = title.strip()
        if not title:
            title = None
        return title, original_language_title, subtitle
    else:
        for entry in titles:
            if not title and ('titleType' not in entry
                              or not entry.get('titleType')):
                title = entry.get('title').strip()
            if not subtitle and entry.get('titleType') == 'Subtitle':
                subtitle = entry.get('title', '').strip()
            if not original_language_title:
                original_language_title = find_original_language_title(entry)

    return title, original_language_title, subtitle


def parse_datacite_dates(dates):
    """
    Given a list of date fields (under .dates), return tuple, (release_date,
    release_year).
    """
    release_date, release_month, release_year = None, None, None

    if not dates:
        return release_date, release_month, release_year

    if not isinstance(dates, list):
        raise ValueError('expected a list of date items')

    # Observed values: "Available", "Submitted", "Valid", "Issued", "Accepted",
    # "Collected", "Updated", "Copyrighted", "Created"
    # Ignored for now: "Collected", "Issued"
    date_type_prio = (
        'Valid',
        'Available',
        'Accepted',
        'Submitted',
        'Copyrighted',
        'Created',
        'Updated',
    )

    # We need to note the granularity, since a string like "2019" would be
    # parsed into "2019-01-01", even though the month is unknown. Use 3
    # granularity types: 'y', 'm', 'd'.
    Pattern = collections.namedtuple('Pattern', 'layout granularity')

    # Before using (expensive) dateparser, try a few common patterns.
    common_patterns = (
        Pattern('%Y-%m-%d', 'd'),
        Pattern('%Y-%m', 'm'),
        Pattern('%Y-%m-%dT%H:%M:%SZ', 'd'),
        Pattern('%Y-%m-%dT%H:%M:%S', 'd'),
        Pattern('%Y', 'y'),
    )

    def parse_item(item):
        result, value, year_only = None, item.get('date', ''), False
        release_date, release_month, release_year = None, None, None

        for layout, granularity in common_patterns:
            try:
                result = datetime.datetime.strptime(value, layout)
            except ValueError:
                continue
            else:
                if granularity == 'y':
                    year_only = True
                break

        if result is None:
            print('fallback for {}'.format(value), file=sys.stderr)
            parser = dateparser.DateDataParser()
            try:
                # Results in a dict with keys: date_obj, period, locale.
                parse_result = parser.get_date_data(value)

                # A datetime object, later we need a date, only.
                result = parse_result['date_obj']
                if result is not None:
                    if parse_result['period'] == 'year':
                        return None, None, result.year
                    elif parse_result['period'] == 'month':
                        return None, result.month, result.year
                    else:
                        return result.date(), result.month, result.year
            except TypeError as err:
                print("{} date parsing failed with: {}".format(value, err),
                      file=sys.stderr)

        if result is None:
            # Unparsable date.
            return release_date, release_month, release_year

        if granularity != 'y':
            release_date = result.date()
        release_year = result.year
        if granularity in ('m', 'd'):
            release_month = result.month

        return release_date, release_month, release_year

    today = datetime.date.today()

    for prio in date_type_prio:
        for item in dates:
            if not item.get('dateType') == prio:
                continue

            release_date, release_month, release_year = parse_item(item)
            if release_date is None and release_year is None:
                continue

            if release_year < 1000 or release_year > today.year + 5:
                # Skip possibly bogus dates.
                release_year = None
                continue
            break
        else:
            continue
        break

    if release_date is None and release_year is None:
        for item in dates:
            release_date, release_month, release_year = parse_item(item)
            if release_year or release_date:
                break

    return release_date, release_month, release_year

def index_form_to_display_name(s):
    """
    Try to convert an index form name, like 'Razis, Panos A' into display_name,
    e.g. 'Panos A Razis'.
    """
    if ',' not in s:
        return s
    skip_on_chars = ['(', ')', '*']
    for char in skip_on_chars:
        if char in s:
            return s
    if s.count(',') > 1:
        # "Dr. Hina, Dr. Muhammad Usman Shahid, Dr. Muhammad Zeeshan Khan"
        return s
    stopwords = [s.lower() for s in (
        'Archive',
        'Collection',
        'Coordinator',
        'Department',
        'Germany',
        'International',
        'National',
        'Netherlands',
        'Office',
        'Organisation',
        'Organization',
        'Service',
        'Services',
        'United States',
        'University',
        'Verein',
        'Volkshochschule',
    )]
    lower = s.lower()
    for stop in stopwords:
        if stop in lower:
            return s

    a, b = s.split(',')
    return '{} {}'.format(b.strip(), a.strip())