aboutsummaryrefslogtreecommitdiffstats
path: root/fatcat_scholar/transform.py
blob: 5bdd0c7a625f9a14d47e957acf9495d52343ed2f (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
import sys
import argparse
from typing import List, Dict, Optional, Any, Sequence

from dynaconf import settings
from fatcat_openapi_client import ReleaseEntity, FileEntity

from fatcat_scholar.api_entities import *
from fatcat_scholar.schema import *
from fatcat_scholar.grobid2json import teixml2json


def es_fulltext_from_sim(sim: Dict[str, Any]) -> Optional[ScholarFulltext]:
    if not sim["page_texts"]:
        return None
    first_page = sim["page_texts"][0]["page_num"]
    issue_item = sim["issue_item"]
    return ScholarFulltext(
        lang_code=None,  # TODO: pub/issue metadata? or langdetect?
        body="\n".join([p["raw_text"] for p in sim["page_texts"]]),
        # acknowledgement=None,
        # annex=None,
        release_ident=sim.get("release_ident"),
        # file_ident=None,
        # file_sha1=None,
        # file_mimetype=None,
        thumbnail_url=f"https://archive.org/serve/{issue_item}/__ia_thumb.jpg",
        access_url=f"https://archive.org/details/{issue_item}/page/{first_page}",
        access_type=AccessType.ia_sim,
    )


def es_sim_from_sim(sim: Dict[str, Any]) -> ScholarSim:
    first_page = None
    if sim["page_texts"]:
        first_page = sim["page_texts"][0]["page_num"]
    return ScholarSim(
        issue_item=sim["issue_item"],
        pub_collection=sim["pub_item_metadata"]["metadata"]["identifier"],
        sim_pubid=sim["issue_item_metadata"]["metadata"]["sim_pubid"],
        first_page=first_page,
    )


SIM_RELEASE_TYPE_MAP = {
    "Scholarly Journals": "article-journal",
    # TODO:
}
SIM_LANG_MAP = {
    "English": "en",
    # TODO:
}
SIM_COUNTRY_MAP = {
    "Netherlands": "nl",
    # TODO:
}


def es_biblio_from_sim(sim: Dict[str, Any]) -> ScholarBiblio:

    issue_meta = sim["issue_item_metadata"]["metadata"]
    pub_meta = sim["pub_item_metadata"]["metadata"]

    first_page = None
    if sim["page_texts"]:
        first_page = sim["page_texts"][0]["page_num"]
    container_name = sim["pub_item_metadata"]["metadata"]["title"]
    last_word = container_name.split()[-1]
    if len(last_word) == 9 and last_word[4] == "-":
        container_name = container_name[:-10]

    issns = []
    raw_issn = issue_meta.get("issn")
    if raw_issn and len(raw_issn) == 9:
        issns.append(raw_issn)

    volume = issue_meta.get("volume")
    volume_int = None
    if volume and volume.isdigit():
        volume_int = int(volume)
    issue = issue_meta.get("issue")
    issue_int = None
    if issue and issue.isdigit():
        issue_int = int(issue)

    date = issue_meta.get("date")
    release_year = None
    if date and len(date) > 4 and date[:4].isdigit():
        release_year = int(date[:4])

    release_date = None
    if len(date) == len("2000-01-01"):
        release_date = date

    return ScholarBiblio(
        # release_ident=release.ident,
        title=None,
        # subtitle=None,
        # original_title=release.original_title,
        release_date=release_date,
        release_year=release_year,
        release_type=SIM_RELEASE_TYPE_MAP.get(pub_meta.get("pub_type")),
        release_stage="published",  # as a default
        # withdrawn_status=release.withdrawn_status,
        lang_code=SIM_LANG_MAP.get(pub_meta.get("language")),
        country_code=SIM_COUNTRY_MAP.get(pub_meta.get("country")),
        volume=volume,
        volume_int=volume_int,
        issue=issue,
        issue_int=issue_int,
        pages=sim.get("pages"),
        first_page=first_page,
        first_page_int=None,
        # number=None,
        # no external identifiers
        # license_slug=release.license_slug,
        publisher=issue_meta.get("publisher"),
        container_name=container_name,
        container_original_name=None,  # TODO pass-through
        container_ident=None,  # TODO: pass-through
        container_type=None,  # TODO
        container_issnl=None,  # TODO: pass-through
        issns=issns,
        # no contrib/affiliation info
        contrib_names=[],
        affiliations=[],
    )


def _add_file_release_meta(
    fulltext: ScholarFulltext, pdf_meta: Optional[dict], re: ReleaseEntity, fe: FileEntity
) -> ScholarFulltext:
    best_url = None
    best_url_type = None
    for url in fe.urls:
        best_url = url.url
        best_url_type = AccessType.web
        if "//archive.org/" in url.url:
            best_url_type = AccessType.ia_file
            break
        elif "//web.archive.org/" in url.url:
            best_url_type = AccessType.wayback
            break
        if url.rel == "repository":
            best_url_type = AccessType.repository
        # TODO: more file-to-access logic

    fulltext.release_ident = re.ident
    fulltext.file_ident = fe.ident
    fulltext.file_sha1 = fe.sha1
    fulltext.file_mimetype = fe.mimetype
    fulltext.access_url = best_url
    fulltext.access_type = best_url_type
    if pdf_meta is not None and pdf_meta["pdf_meta"].get("has_page0_thumbnail"):
        # eg: https://blobs.fatcat.wiki/thumbnail/pdf/32/29/322909fe57cef73b10a166996a4528d337026d16.180px.jpg
        fulltext.thumbnail_url = f"{ settings.THUMBNAIL_URL_PREFIX }{ fe.sha1[0:2] }/{ fe.sha1[2:4] }/{ fe.sha1 }.180px.jpg"
    return fulltext


def es_fulltext_from_grobid(
    tei_xml: str, pdf_meta: Optional[dict], re: ReleaseEntity, fe: FileEntity
) -> Optional[ScholarFulltext]:
    obj = teixml2json(tei_xml)
    if not obj.get("body"):
        return None
    ret = ScholarFulltext(
        lang_code=obj.get("lang"),
        body=obj.get("body"),
        acknowledgement=obj.get("acknowledgement"),
        annex=obj.get("annex"),
    )
    return _add_file_release_meta(ret, pdf_meta, re, fe)


def es_fulltext_from_pdftotext(
    raw_text: str, pdf_meta: Optional[dict], re: ReleaseEntity, fe: FileEntity
) -> Optional[ScholarFulltext]:

    ret = ScholarFulltext(
        lang_code=re.language,
        body=raw_text,
        acknowledgement=None,
        annex=None,
    )
    return _add_file_release_meta(ret, pdf_meta, re, fe)


def transform_heavy(heavy: IntermediateBundle) -> Optional[ScholarDoc]:

    tags: List[str] = []
    work_ident: Optional[str] = None
    sim_issue: Optional[str] = None
    abstracts: List[ScholarAbstract] = []
    fulltext: Optional[ScholarFulltext] = None
    primary_release: Optional[ReleaseEntity] = None

    ia_sim: Optional[ScholarSim] = None
    if heavy.sim_fulltext is not None:
        ia_sim = es_sim_from_sim(heavy.sim_fulltext)
        fulltext = es_fulltext_from_sim(heavy.sim_fulltext)

    if heavy.doc_type == DocType.sim_page:
        assert ia_sim is not None
        assert heavy.sim_fulltext is not None
        key = f"page_{ia_sim.issue_item}_{ia_sim.first_page}"
        sim_issue = ia_sim.issue_item
        biblio = es_biblio_from_sim(heavy.sim_fulltext)
        # fulltext extracted from heavy.sim_fulltext above
    elif heavy.doc_type == DocType.work:
        work_ident = heavy.releases[0].work_id
        key = f"work_{work_ident}"
        assert heavy.biblio_release_ident
        primary_release = [
            r for r in heavy.releases if r.ident == heavy.biblio_release_ident
        ][0]
        biblio = es_biblio_from_release(primary_release)

        # TODO: abstracts from releases also; abstracts_dict; abstracts from GROBID parse
        abstracts = es_abstracts_from_release(primary_release)
    else:
        raise NotImplementedError(f"doc_type: {heavy.doc_type}")

    if heavy.grobid_fulltext:
        fulltext_release = [
            r
            for r in heavy.releases
            if r.ident == heavy.grobid_fulltext["release_ident"]
        ][0]
        fulltext_file = [
            f
            for f in fulltext_release.files
            if f.ident == heavy.grobid_fulltext["file_ident"]
        ][0]
        fulltext = es_fulltext_from_grobid(
            heavy.grobid_fulltext["tei_xml"], heavy.pdf_meta, fulltext_release, fulltext_file
        )

    if not fulltext and heavy.pdftotext_fulltext:
        fulltext_release = [
            r
            for r in heavy.releases
            if r.ident == heavy.pdftotext_fulltext["release_ident"]
        ][0]
        fulltext_file = [
            f
            for f in fulltext_release.files
            if f.ident == heavy.pdftotext_fulltext["file_ident"]
        ][0]
        fulltext = es_fulltext_from_pdftotext(
            heavy.pdftotext_fulltext["raw_text"], heavy.pdf_meta, fulltext_release, fulltext_file
        )

    # TODO: additional access list
    access_dict = dict()
    if fulltext and fulltext.access_type:
        access_dict[fulltext.access_type] = ScholarAccess(
            access_type=fulltext.access_type,
            access_url=fulltext.access_url,
            mimetype=fulltext.file_mimetype,
            file_ident=fulltext.file_ident,
            release_ident=fulltext.release_ident,
        )
    if ia_sim and not AccessType.ia_sim in access_dict:
        access_dict[AccessType.ia_sim] = ScholarAccess(
            access_type=AccessType.ia_sim,
            access_url=f"https://archive.org/details/{ia_sim.issue_item}/page/{ia_sim.first_page}",
        )

    # TODO: additional abstracts

    # tags
    if biblio.license_slug and biblio.license_slug.lower().startswith("cc-"):
        tags.append("oa")
    if primary_release and primary_release.container:
        container = primary_release.container
        if container.extra:
            if container.extra.get("doaj"):
                tags.append("doaj")
                tags.append("oa")
            if container.extra.get("road"):
                tags.append("road")
                tags.append("oa")
            if container.extra.get("szczepanski"):
                tags.append("szczepanski")
                tags.append("oa")
            if container.extra.get("ia", {}).get("longtail_oa"):
                tags.append("longtail")
                tags.append("oa")
            if container.extra.get("sherpa_romeo", {}).get("color") == "white":
                tags.append("oa")
            if container.extra.get("default_license", "").lower().startswith("cc-"):
                tags.append("oa")
            if container.extra.get("platform"):
                # scielo, ojs, wordpress, etc
                tags.append(container.extra["platform"].lower())
    if biblio.doi_prefix == "10.2307":
        tags.append("jstor")

    # biorxiv/medrxiv hacks
    if not biblio.container_name and biblio.release_stage != "published":
        for _, acc in access_dict.items():
            if "://www.medrxiv.org/" in acc.access_url:
                biblio.container_name = "medRxiv"
                if biblio.release_stage == None:
                    biblio.release_stage = "submitted"
            elif "://www.biorxiv.org/" in acc.access_url:
                biblio.container_name = "bioRxiv"
                if biblio.release_stage == None:
                    biblio.release_stage = "submitted"
    tags = list(set(tags))

    return ScholarDoc(
        key=key,
        collapse_key=sim_issue or work_ident,
        doc_type=heavy.doc_type.value,
        doc_index_ts=datetime.datetime.utcnow(),
        work_ident=work_ident,
        tags=tags,
        biblio=biblio,
        fulltext=fulltext,
        ia_sim=ia_sim,
        abstracts=abstracts,
        releases=[es_release_from_release(r) for r in heavy.releases],
        access=list(access_dict.values()),
    )


def run_transform(infile: Sequence) -> None:
    for line in infile:
        obj = json.loads(line)

        heavy = IntermediateBundle(
            doc_type=DocType(obj["doc_type"]),
            releases=[
                entity_from_json(json.dumps(re), ReleaseEntity)
                for re in obj["releases"]
            ],
            biblio_release_ident=obj.get("biblio_release_ident"),
            grobid_fulltext=obj.get("grobid_fulltext"),
            pdftotext_fulltext=obj.get("pdftotext_fulltext"),
            pdf_meta=obj.get("pdf_meta"),
            sim_fulltext=obj.get("sim_fulltext"),
        )
        es_doc = transform_heavy(heavy)
        if not es_doc:
            continue
        print(es_doc.json())


def main() -> None:
    """
    Run this command like:

        python -m fatcat_scholar.transform
    """

    parser = argparse.ArgumentParser(
        formatter_class=argparse.ArgumentDefaultsHelpFormatter
    )
    subparsers = parser.add_subparsers()

    sub = subparsers.add_parser(
        "run_transform", help="iterates through 'heavy' intermediate"
    )
    sub.set_defaults(func="run_transform")
    sub.add_argument(
        "json_file",
        help="intermediate globs as JSON-lines",
        nargs="?",
        default=sys.stdin,
        type=argparse.FileType("r"),
    )

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

    if args.func == "run_transform":
        run_transform(infile=args.json_file)
    else:
        raise NotImplementedError(args.func)


if __name__ == "__main__":
    main()