aboutsummaryrefslogtreecommitdiffstats
path: root/python/fatcat_web/search.py
blob: 4a87c735c3df8738fa327331b2b997566c4c8362 (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

"""
Helpers for doing elasticsearch queries (used in the web interface; not part of
the formal API)

TODO: ELASTICSEARCH_*_INDEX should probably be factored out and just hard-coded
"""

import datetime
import requests
from flask import abort, flash
from fatcat_web import app

def do_search(index, request, limit=30, offset=0, deep_page_limit=2000):

    # Sanity checks
    if limit > 100:
        limit = 100
    if offset < 0:
        offset = 0
    if offset > deep_page_limit:
        # Avoid deep paging problem.
        offset = deep_page_limit

    request["size"] = int(limit)
    request["from"] = int(offset)
    # print(request)
    resp = requests.get("%s/%s/_search" %
            (app.config['ELASTICSEARCH_BACKEND'], index),
        json=request)

    if resp.status_code == 400:
        print("elasticsearch 400: " + str(resp.content))
        flash("Search query failed to parse; you might need to use quotes.<p><code>{}</code>".format(resp.content))
        abort(resp.status_code)
    elif resp.status_code != 200:
        print("elasticsearch non-200 status code: " + str(resp.status_code))
        print(resp.content)
        abort(resp.status_code)

    content = resp.json()
    results = [h['_source'] for h in content['hits']['hits']]
    for h in results:
        # Handle surrogate strings that elasticsearch returns sometimes,
        # probably due to mangled data processing in some pipeline.
        # "Crimes against Unicode"; production workaround
        for key in h:
            if type(h[key]) is str:
                h[key] = h[key].encode('utf8', 'ignore').decode('utf8')

    return {"count_returned": len(results),
            "count_found": content['hits']['total'],
            "results": results,
            "offset": offset,
            "deep_page_limit": deep_page_limit}


def do_release_search(q, limit=30, fulltext_only=True, offset=0):

    #print("Search hit: " + q)
    if limit > 100:
        # Sanity check
        limit = 100

    # Convert raw DOIs to DOI queries
    if len(q.split()) == 1 and q.startswith("10.") and q.count("/") >= 1:
        q = 'doi:"{}"'.format(q)

    if fulltext_only:
        q += " in_web:true"

    search_request = {
        "query": {
            "query_string": {
                "query": q,
                "default_operator": "AND",
                "analyze_wildcard": True,
                "lenient": True,
                "fields": ["biblio"],
            },
        },
    }

    resp = do_search(app.config['ELASTICSEARCH_RELEASE_INDEX'], search_request, offset=offset)
    for h in resp['results']:
        # Ensure 'contrib_names' is a list, not a single string
        if type(h['contrib_names']) is not list:
            h['contrib_names'] = [h['contrib_names'], ]
        h['contrib_names'] = [name.encode('utf8', 'ignore').decode('utf8') for name in h['contrib_names']]
    resp["query"] = { "q": q }
    resp["limit"] = limit
    return resp


def do_container_search(q, limit=30, offset=0):

    # Convert raw ISSN-L to ISSN-L query
    if len(q.split()) == 1 and len(q) == 9 and q[0:4].isdigit() and q[4] == '-':
        q = 'issnl:"{}"'.format(q)

    search_request = {
        "query": {
            "query_string": {
                "query": q,
                "default_operator": "AND",
                "analyze_wildcard": True,
                "lenient": True,
                "fields": ["biblio"],
            },
        },
    }

    resp = do_search(app.config['ELASTICSEARCH_CONTAINER_INDEX'], search_request, limit=limit, offset=offset)
    resp["query"] = { "q": q }
    resp["limit"] = limit
    return resp

def get_elastic_entity_stats():
    """
    TODO: files, filesets, webcaptures (no schema yet)

    Returns dict:
        changelog: {latest: {index, datetime}}
        release: {total, refs_total}
        papers: {total, in_web, in_oa, in_kbart, in_web_not_kbart}
    """

    stats = {}

    # 2. releases
    #  - total count
    #  - total citation records
    #  - total (paper, chapter, proceeding)
    #  - " with fulltext on web
    #  - " open access
    #  - " not in KBART, in IA
    #
    # Can do the above with two queries:
    #  - all releases, aggregate count and sum(ref_count)
    #  - in-scope works, aggregate count by (fulltext, OA, kbart/ia)

    # 2a. release totals
    query = {
        "size": 0,
        "aggs": {
            "release_ref_count": { "sum": { "field": "ref_count" } }
        }
    }
    resp = requests.get(
        "{}/fatcat_release/_search".format(app.config['ELASTICSEARCH_BACKEND']),
        json=query,
        params=dict(request_cache="true"))
    # TODO: abort()
    resp.raise_for_status()
    resp = resp.json()
    stats['release'] = {
        "total": resp['hits']['total'],
        "refs_total": int(resp['aggregations']['release_ref_count']['value']),
    }

    # 2b. paper counts
    query = {
        "size": 0,
        "query": {
            "terms": { "release_type": [
                # "chapter", "thesis",
                "article-journal", "paper-conference",
            ] } },
        "aggs": { "paper_like": { "filters": { "filters": {
                "in_web": { "term": { "in_web": "true" } },
                "is_oa": { "term": { "is_oa": "true" } },
                "in_kbart": { "term": { "in_kbart": "true" } },
                "in_web_not_kbart": { "bool": { "filter": [
                        { "term": { "in_web": "true" } },
                        { "term": { "in_kbart": "false" } }
                ]}}
        }}}}
    }
    resp = requests.get(
        "{}/fatcat_release/_search".format(app.config['ELASTICSEARCH_BACKEND']),
        json=query,
        params=dict(request_cache="true"))
    # TODO: abort()
    resp.raise_for_status()
    resp = resp.json()
    buckets = resp['aggregations']['paper_like']['buckets']
    stats['papers'] = {
        'total': resp['hits']['total'],
        'in_web': buckets['in_web']['doc_count'],
        'is_oa': buckets['is_oa']['doc_count'],
        'in_kbart': buckets['in_kbart']['doc_count'],
        'in_web_not_kbart': buckets['in_web_not_kbart']['doc_count'],
    }

    # 3. containers
    #   => total count
    query = {
        "size": 0,
    }
    resp = requests.get(
        "{}/fatcat_container/_search".format(app.config['ELASTICSEARCH_BACKEND']),
        json=query,
        params=dict(request_cache="true"))
    # TODO: abort()
    resp.raise_for_status()
    resp = resp.json()
    stats['container'] = {
        "total": resp['hits']['total'],
    }

    return stats

def get_elastic_container_stats(ident, issnl=None):
    """
    Returns dict:
        ident
        issnl (optional)
        total
        in_web
        in_kbart
        preserved
    """

    query = {
        "size": 0,
        "query": {
            "term": { "container_id": ident }
        },
        "aggs": { "container_stats": { "filters": { "filters": {
                "in_web": { "term": { "in_web": "true" } },
                "in_kbart": { "term": { "in_kbart": "true" } },
                "is_preserved": { "term": { "is_preserved": "true" } },
        }}}}
    }
    resp = requests.get(
        "{}/fatcat_release/_search".format(app.config['ELASTICSEARCH_BACKEND']),
        json=query,
        params=dict(request_cache="true"))
    # TODO: abort()
    #print(resp.json())
    resp.raise_for_status()
    resp = resp.json()
    buckets = resp['aggregations']['container_stats']['buckets']
    stats = {
        'ident': ident,
        'issnl': issnl,
        'total': resp['hits']['total'],
        'in_web': buckets['in_web']['doc_count'],
        'in_kbart': buckets['in_kbart']['doc_count'],
        'is_preserved': buckets['is_preserved']['doc_count'],
    }

    return stats

def get_elastic_container_random_releases(ident, limit=5):
    """
    Returns a list of releases from the container.
    """

    assert limit > 0 and limit <= 100

    query = {
        "size": int(limit),
        "sort": [
            { "in_web": {"order": "desc"} },
            { "release_date": {"order": "desc"} },
        ],
        "query": {
            "bool": {
                "must": [
                    { "term": { "container_id": ident } },
                    { "range": { "release_year": { "lte": datetime.datetime.today().year } } },
                ],
            },
        },
    }
    resp = requests.get(
        "{}/fatcat_release/_search".format(app.config['ELASTICSEARCH_BACKEND']),
        json=query,
        params=dict(request_cache="true"))
    # TODO: abort()
    #print(resp.json())
    resp.raise_for_status()
    resp = resp.json()
    #print(resp)
    hits = [h['_source'] for h in resp['hits']['hits']]
    for h in hits:
        # Handle surrogate strings that elasticsearch returns sometimes,
        # probably due to mangled data processing in some pipeline.
        # "Crimes against Unicode"; production workaround
        for key in h:
            if type(h[key]) is str:
                h[key] = h[key].encode('utf8', 'ignore').decode('utf8')

    return hits

def get_elastic_container_histogram(ident):
    """
    Fetches a stacked histogram

    Filters to the past 500 years (at most), or about 1000 values.

    Returns a list of tuples:
        (year, in_ia, count)
    """

    query = {
        "aggs": {
            "year_in_ia": {
                "composite": {
                    "size": 1000,
                    "sources": [
                        {"year": {
                            "histogram": {
                                "field": "release_year",
                                "interval": 1,
                        }}},
                        {"in_ia": {
                            "terms": {
                                "field": "in_ia",
                        }}},
                    ],
                },
            },
        },
        "size": 0,
        "query": {
            "bool": {
                "must": [{
                    "range": {
                        "release_year": {
                            "gte": datetime.datetime.today().year - 499,
                            "lte": datetime.datetime.today().year,
                        }
                    }
                }],
                "filter": [{
                    "bool": {
                        "should": [{
                            "match": {
                                "container_id": ident
                            }
                        }],
                        "minimum_should_match": 1,
                    },
                }],
            }
        }
    }
    resp = requests.get(
        "{}/fatcat_release/_search".format(app.config['ELASTICSEARCH_BACKEND']),
        json=query,
        params=dict(request_cache="true"))
    resp.raise_for_status()
    # TODO: abort()
    resp = resp.json()
    #print(resp)
    vals = [(h['key']['year'], h['key']['in_ia'], h['doc_count'])
            for h in resp['aggregations']['year_in_ia']['buckets']]
    vals = sorted(vals)
    return vals