repo.py 31.7 KB
Newer Older
Markus Scheidgen's avatar
Markus Scheidgen committed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
# Copyright 2018 Markus Scheidgen
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an"AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

"""
The repository API of the nomad@FAIRDI APIs. Currently allows to resolve repository
meta-data.
"""

20
from typing import List, Dict, Any
Markus Scheidgen's avatar
Markus Scheidgen committed
21
from flask_restplus import Resource, abort, fields
22
from flask import request, g
23
from elasticsearch_dsl import Q
24
from elasticsearch.exceptions import NotFoundError
25
import elasticsearch.helpers
Markus Scheidgen's avatar
Markus Scheidgen committed
26

27
28
from nomad import search, utils, datamodel, processing as proc, infrastructure
from nomad.app.utils import rfc3339DateTime, RFC3339DateTime, with_logger
29
from nomad.app.optimade import filterparser
30
from nomad.datamodel import UserMetadata, Dataset, User
Markus Scheidgen's avatar
Markus Scheidgen committed
31

Markus Scheidgen's avatar
Markus Scheidgen committed
32
from .api import api
33
from .auth import authenticate
34
from .common import pagination_model, pagination_request_parser, calc_route
Markus Scheidgen's avatar
Markus Scheidgen committed
35

36
ns = api.namespace('repo', description='Access repository metadata.')
Markus Scheidgen's avatar
Markus Scheidgen committed
37
38
39
40
41


@calc_route(ns)
class RepoCalcResource(Resource):
    @api.response(404, 'The upload or calculation does not exist')
42
    @api.response(401, 'Not authorized to access the calculation')
43
    @api.response(200, 'Metadata send', fields.Raw)
44
    @api.doc('get_repo_calc')
45
    @authenticate()
46
    def get(self, upload_id, calc_id):
Markus Scheidgen's avatar
Markus Scheidgen committed
47
48
49
        """
        Get calculation metadata in repository form.

50
        Repository metadata only entails the quantities shown in the repository.
51
        Calcs are references via *upload_id*, *calc_id* pairs.
Markus Scheidgen's avatar
Markus Scheidgen committed
52
53
        """
        try:
54
55
56
57
58
59
60
61
            calc = search.Entry.get(calc_id)
        except NotFoundError:
            abort(404, message='There is no calculation %s/%s' % (upload_id, calc_id))

        if calc.with_embargo or not calc.published:
            if g.user is None:
                abort(401, message='Not logged in to access %s/%s.' % (upload_id, calc_id))

62
            if not (any(g.user.user_id == user.user_id for user in calc.owners) or g.user.is_admin):
63
64
65
                abort(401, message='Not authorized to access %s/%s.' % (upload_id, calc_id))

        return calc.to_dict(), 200
Markus Scheidgen's avatar
Markus Scheidgen committed
66
67


Markus Scheidgen's avatar
Markus Scheidgen committed
68
repo_calcs_model_fields = {
Markus Scheidgen's avatar
Markus Scheidgen committed
69
    'pagination': fields.Nested(pagination_model, skip_none=True),
70
71
72
73
    'scroll': fields.Nested(allow_null=True, skip_none=True, model=api.model('Scroll', {
        'total': fields.Integer(description='The total amount of hits for the search.'),
        'scroll_id': fields.String(allow_null=True, description='The scroll_id that can be used to retrieve the next page.'),
        'size': fields.Integer(help='The size of the returned scroll page.')})),
74
75
76
    'results': fields.List(fields.Raw, description=(
        'A list of search results. Each result is a dict with quantitie names as key and '
        'values as values')),
77
78
79
80
    'statistics': fields.Raw(description=(
        'A dict with all statistics. Each statistic is dictionary with a metrics dict as '
        'value and quantity value as key. The possible metrics are code runs(calcs), %s. '
        'There is a pseudo quantity "total" with a single value "all" that contains the '
81
        ' metrics over all results. ' % ', '.join(datamodel.Domain.instance.metrics_names))),
Markus Scheidgen's avatar
Markus Scheidgen committed
82
83
84
85
86
}
for group_name, (group_quantity, _) in search.groups.items():
    repo_calcs_model_fields[group_name] = fields.Nested(api.model('RepoDatasets', {
        'after': fields.String(description='The after value that can be used to retrieve the next %s.' % group_name),
        'values': fields.Raw(description='A dict with %s as key. The values are dicts with "total" and "examples" keys.' % group_quantity)
Markus Scheidgen's avatar
Markus Scheidgen committed
87
    }), skip_none=True)
Markus Scheidgen's avatar
Markus Scheidgen committed
88
repo_calcs_model = api.model('RepoCalculations', repo_calcs_model_fields)
Markus Scheidgen's avatar
Markus Scheidgen committed
89

90

91
92
repo_calc_id_model = api.model('RepoCalculationId', {
    'upload_id': fields.String(), 'calc_id': fields.String()
Markus Scheidgen's avatar
Markus Scheidgen committed
93
94
})

95
96
97
98
99
100
101
102
103
104
105
106

def add_common_parameters(request_parser):
    request_parser.add_argument(
        'owner', type=str,
        help='Specify which calcs to return: ``all``, ``public``, ``user``, ``staging``, default is ``all``')
    request_parser.add_argument(
        'from_time', type=lambda x: rfc3339DateTime.parse(x),
        help='A yyyy-MM-ddTHH:mm:ss (RFC3339) minimum entry time (e.g. upload time)')
    request_parser.add_argument(
        'until_time', type=lambda x: rfc3339DateTime.parse(x),
        help='A yyyy-MM-ddTHH:mm:ss (RFC3339) maximum entry time (e.g. upload time)')

107
    for quantity in search.quantities.values():
108
        request_parser.add_argument(
109
            quantity.name, help=quantity.description,
110
            action=quantity.argparse_action if quantity.multi else None)
111
112


Markus Scheidgen's avatar
Markus Scheidgen committed
113
repo_request_parser = pagination_request_parser.copy()
114
add_common_parameters(repo_request_parser)
115
116
117
118
repo_request_parser.add_argument(
    'scroll', type=bool, help='Enable scrolling')
repo_request_parser.add_argument(
    'scroll_id', type=str, help='The id of the current scrolling window to use.')
119
120
repo_request_parser.add_argument(
    'date_histogram', type=bool, help='Add an additional aggregation over the upload time')
Markus Scheidgen's avatar
Markus Scheidgen committed
121
repo_request_parser.add_argument(
122
    'metrics', type=str, action='append', help=(
123
        'Metrics to aggregate over all quantities and their values as comma separated list. '
124
        'Possible values are %s.' % ', '.join(datamodel.Domain.instance.metrics_names)))
Markus Scheidgen's avatar
Markus Scheidgen committed
125
126
repo_request_parser.add_argument(
    'statistics', type=bool, help=('Return statistics.'))
Markus Scheidgen's avatar
Markus Scheidgen committed
127

Markus Scheidgen's avatar
Markus Scheidgen committed
128
129
130
131
132
133
134
for group_name in search.groups:
    repo_request_parser.add_argument(
        group_name, type=bool, help=('Return %s group data.' % group_name))
    repo_request_parser.add_argument(
        '%s_after' % group_name, type=str,
        help='The last %s id of the last scroll window for the %s group' % (group_name, group_name))

135

136
137
138
139
search_request_parser = api.parser()
add_common_parameters(search_request_parser)


140
def add_query(search_request: search.SearchRequest, args: Dict[str, Any]):
141
    """
142
    Help that adds query relevant request args to the given SearchRequest.
143
    """
144
    args = {key: value for key, value in args.items() if value is not None}
145

146
    # owner
147
    owner = args.get('owner', 'all')
148
149
    try:
        search_request.owner(
150
            owner,
151
152
            g.user.user_id if g.user is not None else None)
    except ValueError as e:
153
        abort(401, getattr(e, 'message', 'Invalid owner parameter: %s' % owner))
154
155
156
157
    except Exception as e:
        abort(400, getattr(e, 'message', 'Invalid owner parameter'))

    # time range
158
159
    from_time_str = args.get('from_time', None)
    until_time_str = args.get('until_time', None)
160
161

    try:
162
163
164
        from_time = rfc3339DateTime.parse(from_time_str) if from_time_str is not None else None
        until_time = rfc3339DateTime.parse(until_time_str) if until_time_str is not None else None
        search_request.time_range(start=from_time, end=until_time)
165
166
167
    except Exception:
        abort(400, message='bad datetime format')

168
169
    # optimade
    try:
170
        optimade = args.get('optimade', None)
171
172
173
174
175
176
        if optimade is not None:
            q = filterparser.parse_filter(optimade)
            search_request.query(q)
    except filterparser.FilterException:
        abort(400, message='could not parse optimade query')

177
178
    # search parameter
    search_request.search_parameters(**{
179
        key: value for key, value in args.items()
180
        if key not in ['optimade'] and key in search.quantities})
181
182


Markus Scheidgen's avatar
Markus Scheidgen committed
183
184
@ns.route('/')
class RepoCalcsResource(Resource):
185
    @api.doc('search')
186
    @api.response(400, 'Invalid requests, e.g. wrong owner type or bad search parameters')
Markus Scheidgen's avatar
Markus Scheidgen committed
187
    @api.expect(repo_request_parser, validate=True)
188
    @api.marshal_with(repo_calcs_model, skip_none=True, code=200, description='Search results send')
189
    @authenticate()
Markus Scheidgen's avatar
Markus Scheidgen committed
190
191
    def get(self):
        """
192
        Search for calculations in the repository form, paginated.
193
194

        The ``owner`` parameter determines the overall entries to search through.
195
196
197
198
        Possible values are: ``all`` (show all entries visible to the current user), ``public``
        (show all publically visible entries), ``user`` (show all user entries, requires login),
        ``staging`` (show all user entries in staging area, requires login).

199
200
201
202
203
        You can use the various quantities to search/filter for. For some of the
        indexed quantities this endpoint returns aggregation information. This means
        you will be given a list of all possible values and the number of entries
        that have the certain value. You can also use these aggregations on an empty
        search to determine the possible values.
204
205
206

        The pagination parameters allows determine which page to return via the
        ``page`` and ``per_page`` parameters. Pagination however, is limited to the first
Markus Scheidgen's avatar
Markus Scheidgen committed
207
208
209
210
211
212
213
214
        100k (depending on ES configuration) hits.

        An alternative to pagination is to use ``scroll`` and ``scroll_id``. With ``scroll``
        you will get a ``scroll_id`` on the first request. Each call with ``scroll`` and
        the respective ``scroll_id`` will return the next ``per_page`` (here the default is 1000)
        results. Scroll however, ignores ordering and does not return aggregations.
        The scroll view used in the background will stay alive for 1 minute between requests.
        If the given ``scroll_id`` is not available anymore, a HTTP 400 is raised.
215
216
217
218
219

        The search will return aggregations on a predefined set of quantities. Aggregations
        will tell you what quantity values exist and how many entries match those values.

        Ordering is determined by ``order_by`` and ``order`` parameters.
220
        """
221
222

        try:
223
224
225
226
227
228
229
230
231
232
233
234
            args = {
                key: value for key, value in repo_request_parser.parse_args().items()
                if value is not None}

            scroll = args.get('scroll', False)
            scroll_id = args.get('scroll_id', None)
            page = args.get('page', 1)
            per_page = args.get('per_page', 10 if not scroll else 1000)
            order = args.get('order', -1)
            order_by = args.get('order_by', 'formula')

            date_histogram = args.get('date_histogram', False)
235
            metrics: List[str] = request.args.getlist('metrics')
Markus Scheidgen's avatar
Markus Scheidgen committed
236

Markus Scheidgen's avatar
Markus Scheidgen committed
237
238
            with_statistics = args.get('statistics', False) or \
                any(args.get(group_name, False) for group_name in search.groups)
239
240
241
242
243
244
245
        except Exception as e:
            abort(400, message='bad parameters: %s' % str(e))

        search_request = search.SearchRequest()
        add_query(search_request, args)
        if date_histogram:
            search_request.date_histogram()
246

247
        try:
248
            assert page >= 1
249
            assert per_page >= 0
250
251
252
        except AssertionError:
            abort(400, message='invalid pagination')

253
254
255
        if order not in [-1, 1]:
            abort(400, message='invalid pagination')

256
257
        for metric in metrics:
            if metric not in search.metrics_names:
258
259
                abort(400, message='there is no metric %s' % metric)

Markus Scheidgen's avatar
Markus Scheidgen committed
260
261
        if with_statistics:
            search_request.default_statistics(metrics_to_use=metrics)
262

Markus Scheidgen's avatar
Markus Scheidgen committed
263
264
265
266
            additional_metrics = [
                metric
                for group_name, (_, metric) in search.groups.items()
                if args.get(group_name, False)]
267
268
269

            total_metrics = metrics + additional_metrics

Markus Scheidgen's avatar
Markus Scheidgen committed
270
271
            search_request.totals(metrics_to_use=total_metrics)
            search_request.statistic('authors', 1000)
272

273
        try:
274
            if scroll:
275
                results = search_request.execute_scrolled(scroll_id=scroll_id, size=per_page)
276

277
            else:
Markus Scheidgen's avatar
Markus Scheidgen committed
278
279
280
281
282
                for group_name, (group_quantity, _) in search.groups.items():
                    if args.get(group_name, False):
                        search_request.quantity(
                            group_quantity, size=per_page, examples=1,
                            after=request.args.get('%s_after' % group_name, None))
283

284
285
                results = search_request.execute_paginated(
                    per_page=per_page, page=page, order=order, order_by=order_by)
286
287

                # TODO just a work around to make things prettier
Markus Scheidgen's avatar
Markus Scheidgen committed
288
289
290
291
292
                if with_statistics:
                    statistics = results['statistics']
                    if 'code_name' in statistics and 'currupted mainfile' in statistics['code_name']:
                        del(statistics['code_name']['currupted mainfile'])

293
294
295
                if 'quantities' in results:
                    quantities = results.pop('quantities')

Markus Scheidgen's avatar
Markus Scheidgen committed
296
297
298
                for group_name, (group_quantity, _) in search.groups.items():
                    if args.get(group_name, False):
                        results[group_name] = quantities[group_quantity]
299

300
            return results, 200
Markus Scheidgen's avatar
Markus Scheidgen committed
301
302
        except search.ScrollIdNotFound:
            abort(400, 'The given scroll_id does not exist.')
303
        except KeyError as e:
304
305
            import traceback
            traceback.print_exc()
306
            abort(400, str(e))
307

308
309
310
311
312
313
314
315

query_model_parameters = {
    'owner': fields.String(description='Specify which calcs to return: ``all``, ``public``, ``user``, ``staging``, default is ``all``'),
    'from_time': RFC3339DateTime(description='A yyyy-MM-ddTHH:mm:ss (RFC3339) minimum entry time (e.g. upload time)'),
    'until_time': RFC3339DateTime(description='A yyyy-MM-ddTHH:mm:ss (RFC3339) maximum entry time (e.g. upload time)')
}

for quantity in search.quantities.values():
316
    if quantity.multi and quantity.argparse_action is None:
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
        def field(**kwargs):
            return fields.List(fields.String(**kwargs))
    else:
        field = fields.String
    query_model_parameters[quantity.name] = field(description=quantity.description)

repo_query_model = api.model('RepoQuery', query_model_parameters, skip_none=True)


def repo_edit_action_field(quantity):
    if quantity.is_scalar:
        return fields.Nested(repo_edit_action_model, description=quantity.description, skip_none=True)
    else:
        return fields.List(
            fields.Nested(repo_edit_action_model, skip_none=True), description=quantity.description)


repo_edit_action_model = api.model('RepoEditAction', {
335
    'value': fields.String(description='The value/values that is set as a string.'),
336
337
338
339
340
341
342
343
344
345
346
347
    'success': fields.Boolean(description='If this can/could be done. Only in API response.'),
    'message': fields.String(descriptin='A message that details the action result. Only in API response.')
})

repo_edit_model = api.model('RepoEdit', {
    'verify': fields.Boolean(description='If true, no action is performed.'),
    'query': fields.Nested(repo_query_model, skip_none=True, description='New metadata will be applied to query results.'),
    'actions': fields.Nested(
        api.model('RepoEditActions', {
            quantity.name: repo_edit_action_field(quantity)
            for quantity in UserMetadata.m_def.all_quantities.values()
        }), skip_none=True,
348
349
350
        description='Each action specifies a single value (even for multi valued quantities).'),
    'success': fields.Boolean(description='If the overall edit can/could be done. Only in API response.'),
    'message': fields.String(description='A message that details the overall edit result. Only in API response.')
351
352
353
})


354
def edit(parsed_query: Dict[str, Any], logger, mongo_update: Dict[str, Any] = None, re_index=True) -> List[str]:
355
    # get all calculations that have to change
356
357
358
359
360
361
362
363
    with utils.timer(logger, 'edit query executed'):
        search_request = search.SearchRequest()
        add_query(search_request, parsed_query)
        upload_ids = set()
        calc_ids = []
        for hit in search_request.execute_scan():
            calc_ids.append(hit['calc_id'])
            upload_ids.add(hit['upload_id'])
364
365

    # perform the update on the mongo db
366
367
368
369
370
    with utils.timer(logger, 'edit mongo update executed', size=len(calc_ids)):
        if mongo_update is not None:
            n_updated = proc.Calc.objects(calc_id__in=calc_ids).update(multi=True, **mongo_update)
            if n_updated != len(calc_ids):
                logger.error('edit repo did not update all entries', payload=mongo_update)
371
372

    # re-index the affected entries in elastic search
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
    with utils.timer(logger, 'edit elastic update executed', size=len(calc_ids)):
        if re_index:
            def elastic_updates():
                for calc in proc.Calc.objects(calc_id__in=calc_ids):
                    entry = search.Entry.from_calc_with_metadata(
                        datamodel.CalcWithMetadata(**calc['metadata']))
                    entry = entry.to_dict(include_meta=True)
                    entry['_op_type'] = 'index'
                    yield entry

            _, failed = elasticsearch.helpers.bulk(
                infrastructure.elastic_client, elastic_updates(), stats_only=True)
            search.refresh()
            if failed > 0:
                logger.error(
                    'edit repo with failed elastic updates',
                    payload=mongo_update, nfailed=len(failed))
390

391
392
    return list(upload_ids)

393

394
395
396
397
398
399
400
401
def get_uploader_ids(query):
    """ Get all the uploader from the query, to check coauthers and shared_with for uploaders. """
    search_request = search.SearchRequest()
    add_query(search_request, query)
    search_request.quantity(name='uploader_id')
    return search_request.execute()['quantities']['uploader_id']['values']


402
403
@ns.route('/edit')
class EditRepoCalcsResource(Resource):
404
405
406
    @api.doc('edit_repo')
    @api.response(400, 'Invalid requests, e.g. wrong owner type or bad search parameters')
    @api.expect(repo_edit_model)
407
    @api.marshal_with(repo_edit_model, skip_none=True, code=200, description='Edit verified/performed')
408
409
410
411
    @authenticate()
    @with_logger
    def post(self, logger):
        """ Edit repository metadata. """
412
413

        # basic body parsing and some semantic checks
414
415
416
417
418
419
420
421
422
423
        json_data = request.get_json()
        if json_data is None:
            json_data = {}
        query = json_data.get('query', {})

        owner = query.get('owner', 'user')
        if owner not in ['user', 'staging']:
            abort(400, 'Not a valid owner for edit %s. Edit can only be performed in user or staging' % owner)
        query['owner'] = owner

424
425
426
427
        if 'actions' not in json_data:
            abort(400, 'Missing key actions in edit data')
        actions = json_data['actions']
        verify = json_data.get('verify', False)
428

429
430
431
432
433
434
435
436
437
438
        # preparing the query of entries that are edited
        parsed_query = {}
        for quantity_name, quantity in search.quantities.items():
            if quantity_name in query:
                value = query[quantity_name]
                if quantity.multi and quantity.argparse_action == 'split' and not isinstance(value, list):
                    value = value.split(',')
                parsed_query[quantity_name] = value
        parsed_query['owner'] = owner

439
        # checking the edit actions and preparing a mongo update on the fly
440
        json_data['success'] = True
441
        mongo_update = {}
442
        uploader_ids = None
443
        lift_embargo = False
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
        removed_datasets = None

        with utils.timer(logger, 'edit verified'):
            for action_quantity_name, quantity_actions in actions.items():
                quantity = UserMetadata.m_def.all_quantities.get(action_quantity_name)
                if quantity is None:
                    abort(400, 'Unknown quantity %s' % action_quantity_name)

                quantity_flask = quantity.m_x('flask', {})
                if quantity_flask.get('admin_only', False):
                    if not g.user.is_admin():
                        abort(404, 'Only the admin user can set %s' % quantity.name)

                if isinstance(quantity_actions, list) == quantity.is_scalar:
                    abort(400, 'Wrong shape for quantity %s' % action_quantity_name)

                if not isinstance(quantity_actions, list):
                    quantity_actions = [quantity_actions]

                flask_verify = quantity_flask.get('verify', None)
                mongo_key = 'metadata__%s' % quantity.name
                has_error = False
                for action in quantity_actions:
                    action['success'] = True
                    action['message'] = None
                    action_value = action.get('value')
                    action_value = action_value if action_value is None else action_value.strip()

                    if action_value is None:
                        mongo_value = None
474

475
                    elif action_value == '':
476
                        mongo_value = None
477
478
479
480
481

                    elif flask_verify == datamodel.User:
                        try:
                            mongo_value = User.get(user_id=action_value).user_id
                        except KeyError:
482
483
                            action['success'] = False
                            has_error = True
484
                            action['message'] = 'User does not exist'
485
                            continue
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
                        if uploader_ids is None:
                            uploader_ids = get_uploader_ids(parsed_query)
                        if action_value in uploader_ids:
                            action['success'] = False
                            has_error = True
                            action['message'] = 'This user is already an uploader of one entry in the query'
                            continue

                    elif flask_verify == datamodel.Dataset:
                        try:
                            mongo_value = Dataset.m_def.m_x('me').get(
                                user_id=g.user.user_id, name=action_value).dataset_id
                        except KeyError:
                            action['message'] = 'Dataset does not exist and will be created'
                            mongo_value = None
                            if not verify:
                                dataset = Dataset(
                                    dataset_id=utils.create_uuid(), user_id=g.user.user_id,
                                    name=action_value)
                                dataset.m_x('me').create()
                                mongo_value = dataset.dataset_id

                    elif action_quantity_name == 'with_embargo':
                        # ignore the actual value ... just lift the embargo
                        mongo_value = False
                        lift_embargo = True

                        # check if necessary
                        search_request = search.SearchRequest()
                        add_query(search_request, parsed_query)
                        search_request.q = search_request.q & Q('term', with_embargo=True)
                        if search_request.execute()['total'] == 0:
                            action['success'] = False
                            has_error = True
                            action['message'] = 'There is no embargo to lift'
                            continue
                    else:
                        mongo_value = action_value

                    if len(quantity.shape) == 0:
                        mongo_update[mongo_key] = mongo_value
                    else:
                        mongo_values = mongo_update.setdefault(mongo_key, [])
                        if mongo_value is not None:
                            if mongo_value in mongo_values:
                                action['success'] = False
                                has_error = True
                                action['message'] = 'Duplicate values are not allowed'
                                continue
                            mongo_values.append(mongo_value)

                if len(quantity_actions) == 0 and len(quantity.shape) > 0:
                    mongo_update[mongo_key] = []

                if action_quantity_name == 'datasets':
                    # check if datasets edit is allowed and if datasets have to be removed
                    search_request = search.SearchRequest()
                    add_query(search_request, parsed_query)
                    search_request.quantity(name='dataset_id')
                    old_datasets = list(
                        search_request.execute()['quantities']['dataset_id']['values'].keys())

                    removed_datasets = []
                    for dataset_id in old_datasets:
                        if dataset_id not in mongo_update.get(mongo_key, []):
                            removed_datasets.append(dataset_id)

                    doi_ds = Dataset.m_def.m_x('me').objects(
                        dataset_id__in=removed_datasets, doi__ne=None).first()
                    if doi_ds is not None:
                        json_data['success'] = False
                        json_data['message'] = json_data.get('message', '') + \
                            'Edit would remove entries from a dataset with DOI (%s) ' % doi_ds.name
                        has_error = True
561

562
563
564
565
566
567
568
569
        # stop here, if client just wants to verify its actions
        if verify:
            return json_data, 200

        # stop if the action were not ok
        if has_error:
            return json_data, 400

570
        # perform the change
571
572
573
574
575
576
577
        upload_ids = edit(parsed_query, logger, mongo_update, True)

        # lift embargo
        if lift_embargo:
            for upload_id in upload_ids:
                upload = proc.Upload.get(upload_id)
                upload.re_pack()
578

579
580
581
582
        # remove old datasets
        if removed_datasets is not None:
            Dataset.m_def.m_x('me').objects(dataset_id__in=removed_datasets).delete()

583
        return json_data, 200
584

585

586
587
588
589
590
repo_quantity_model = api.model('RepoQuantity', {
    'after': fields.String(description='The after value that can be used to retrieve the next set of values.'),
    'values': fields.Raw(description='A dict with values as key. Values are dicts with "total" and "examples" keys.')
})

591
repo_quantity_values_model = api.model('RepoQuantityValues', {
592
593
594
595
596
    'quantity': fields.Nested(repo_quantity_model, allow_null=True)
})

repo_quantities_model = api.model('RepoQuantities', {
    'quantities': fields.List(fields.Nested(repo_quantity_model))
597
598
599
600
601
602
})

repo_quantity_search_request_parser = api.parser()
add_common_parameters(repo_quantity_search_request_parser)
repo_quantity_search_request_parser.add_argument(
    'after', type=str, help='The after value to use for "scrolling".')
603
repo_quantity_search_request_parser.add_argument(
604
605
606
    'size', type=int, help='The max size of the returned values.')


607
@ns.route('/quantity/<string:quantity>')
608
609
610
611
612
class RepoQuantityResource(Resource):
    @api.doc('quantity_search')
    @api.response(400, 'Invalid requests, e.g. wrong owner type, bad quantity, bad search parameters')
    @api.expect(repo_quantity_search_request_parser, validate=True)
    @api.marshal_with(repo_quantity_values_model, skip_none=True, code=200, description='Search results send')
613
    @authenticate()
614
615
616
617
618
619
620
621
622
623
624
625
626
627
    def get(self, quantity: str):
        """
        Retrieve quantity values from entries matching the search.

        You can use the various quantities to search/filter for. For some of the
        indexed quantities this endpoint returns aggregation information. This means
        you will be given a list of all possible values and the number of entries
        that have the certain value. You can also use these aggregations on an empty
        search to determine the possible values.

        There is no ordering and no pagination. Instead there is an 'after' key based
        scrolling. The result will contain an 'after' value, that can be specified
        for the next request. You can use the 'size' and 'after' parameters accordingly.

628
629
630
        The result will contain a 'quantity' key with quantity values and the "after"
        value. There will be upto 'size' many values. For the rest of the values use the
        "after" parameter in another request.
631
632
        """

633
        search_request = search.SearchRequest()
634
635
636
637
        args = {
            key: value
            for key, value in repo_quantity_search_request_parser.parse_args().items()
            if value is not None}
638

639
640
641
        add_query(search_request, args)
        after = args.get('after', None)
        size = args.get('size', 100)
642
643
644
645
646
647

        try:
            assert size >= 0
        except AssertionError:
            abort(400, message='invalid size')

648
        search_request.quantity(quantity, size=size, after=after)
649
650

        try:
651
652
653
            results = search_request.execute()
            quantities = results.pop('quantities')
            results['quantity'] = quantities[quantity]
654
655
656
657
658
659

            return results, 200
        except KeyError as e:
            import traceback
            traceback.print_exc()
            abort(400, 'Given quantity does not exist: %s' % str(e))
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
repo_quantities_search_request_parser = api.parser()
add_common_parameters(repo_quantities_search_request_parser)
repo_quantities_search_request_parser.add_argument(
    'quantities', type=str, action='append',
    help='The quantities to retrieve values from')
repo_quantities_search_request_parser.add_argument(
    'size', type=int, help='The max size of the returned values.')


@ns.route('/quantities')
class RepoQuantitiesResource(Resource):
    @api.doc('quantities_search')
    @api.response(400, 'Invalid requests, e.g. wrong owner type, bad quantity, bad search parameters')
    @api.expect(repo_quantities_search_request_parser, validate=True)
    @api.marshal_with(repo_quantities_model, skip_none=True, code=200, description='Search results send')
    @authenticate()
    def get(self):
        """
        Retrieve quantity values for multiple quantities at once.

        You can use the various quantities to search/filter for. For some of the
        indexed quantities this endpoint returns aggregation information. This means
        you will be given a list of all possible values and the number of entries
        that have the certain value. You can also use these aggregations on an empty
        search to determine the possible values.

        There is no ordering and no pagination and not after key based scrolling. Instead
        there is an 'after' key based scrolling.

        The result will contain a 'quantities' key with a dict of quantity names and the
        retrieved values as values.
        """

        search_request = search.SearchRequest()
        args = {
            key: value
            for key, value in repo_quantities_search_request_parser.parse_args().items()
            if value is not None}

        add_query(search_request, args)
        quantities = args.get('quantities', [])
        size = args.get('size', 5)

        try:
            assert size >= 0
        except AssertionError:
            abort(400, message='invalid size')

        for quantity in quantities:
            try:
                search_request.quantity(quantity, size=size)
            except KeyError as e:
                import traceback
                traceback.print_exc()
                abort(400, 'Given quantity does not exist: %s' % str(e))

        return search_request.execute(), 200


721
722
723
724
725
@ns.route('/pid/<int:pid>')
class RepoPidResource(Resource):
    @api.doc('resolve_pid')
    @api.response(404, 'Entry with PID does not exist')
    @api.marshal_with(repo_calc_id_model, skip_none=True, code=200, description='Entry resolved')
Markus Scheidgen's avatar
Markus Scheidgen committed
726
    @authenticate()
727
    def get(self, pid: int):
728
729
730
731
        search_request = search.SearchRequest()

        if g.user is not None:
            search_request.owner('all', user_id=g.user.user_id)
732
        else:
733
734
735
736
737
738
739
740
741
742
743
            search_request.owner('all')

        search_request.search_parameter('pid', pid)

        results = list(search_request.execute_scan())
        total = len(results)

        if total == 0:
            abort(404, 'Entry with PID %d does not exist' % pid)

        if total > 1:
744
            utils.get_logger(__name__).error('Two entries for the same pid', pid=pid)
745
746
747
748
749

        result = results[0]
        return dict(
            upload_id=result['upload_id'],
            calc_id=result['calc_id'])