data.py 21.3 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
20
21
22
23
24
25
26
# 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.

"""
This module comprises a set of persistent document classes that hold all user related
data. These are information about users, their uploads and datasets, the associated
calculations, and files


.. autoclass:: Calc
    :members:
.. autoclass:: Upload
    :members:
"""

27
from typing import List, Any, ContextManager, Tuple, Generator
28
from mongoengine import StringField, DateTimeField, DictField
Markus Scheidgen's avatar
Markus Scheidgen committed
29
import logging
30
from structlog import wrap_logger
31
from contextlib import contextmanager
Markus Scheidgen's avatar
Markus Scheidgen committed
32

Markus Scheidgen's avatar
Markus Scheidgen committed
33
34
from nomad import utils, coe_repo, config, infrastructure
from nomad.files import PathObject, UploadFiles, ExtractError, ArchiveBasedStagingUploadFiles
35
from nomad.processing.base import Proc, Chord, process, task, PENDING, SUCCESS, FAILURE
Markus Scheidgen's avatar
Markus Scheidgen committed
36
from nomad.parsing import parsers, parser_dict
Markus Scheidgen's avatar
Markus Scheidgen committed
37
from nomad.normalizing import normalizers
Markus Scheidgen's avatar
Markus Scheidgen committed
38
from nomad.datamodel import UploadWithMetadata, CalcWithMetadata
Markus Scheidgen's avatar
Markus Scheidgen committed
39
40


Markus Scheidgen's avatar
Markus Scheidgen committed
41
class Calc(Proc):
Markus Scheidgen's avatar
Markus Scheidgen committed
42
43
44
45
46
47
48
49
50
51
    """
    Instances of this class represent calculations. This class manages the elastic
    search index entry, files, and archive for the respective calculation.

    It also contains the calculations processing and its state.

    The attribute list, does not include the various repository properties generated
    while parsing, including ``program_name``, ``program_version``, etc.

    Attributes:
52
        calc_id: the calc_id of this calc
Markus Scheidgen's avatar
Markus Scheidgen committed
53
54
55
56
        parser: the name of the parser used to process this calc
        upload_id: the id of the upload used to create this calculation
        mainfile: the mainfile (including path in upload) that was used to create this calc
    """
57
    calc_id = StringField(primary_key=True)
Markus Scheidgen's avatar
Markus Scheidgen committed
58
59
60
61
62
63
    upload_id = StringField()
    mainfile = StringField()
    parser = StringField()

    meta: Any = {
        'indices': [
64
            'upload_id', 'mainfile', 'code', 'parser', 'tasks_status'
Markus Scheidgen's avatar
Markus Scheidgen committed
65
66
67
68
69
70
        ]
    }

    def __init__(self, *args, **kwargs):
        super().__init__(*args, **kwargs)
        self._parser_backend = None
71
72
        self._upload: Upload = None
        self._upload_files: ArchiveBasedStagingUploadFiles = None
73
        self._calc_proc_logwriter = None
74
        self._calc_proc_logwriter_ctx: ContextManager = None
Markus Scheidgen's avatar
Markus Scheidgen committed
75
76
77

    @classmethod
    def get(cls, id):
78
        return cls.get_by_id(id, 'calc_id')
Markus Scheidgen's avatar
Markus Scheidgen committed
79

Markus Scheidgen's avatar
Markus Scheidgen committed
80
    @property
81
82
    def mainfile_file(self) -> PathObject:
        return self.upload_files.raw_file_object(self.mainfile)
Markus Scheidgen's avatar
Markus Scheidgen committed
83

84
85
86
87
88
89
    @property
    def upload(self) -> 'Upload':
        if not self._upload:
            self._upload = Upload.get(self.upload_id)
        return self._upload

90
91
92
    @property
    def upload_files(self) -> ArchiveBasedStagingUploadFiles:
        if not self._upload_files:
93
            self._upload_files = ArchiveBasedStagingUploadFiles(self.upload_id, is_authorized=lambda: True, local_path=self.upload.local_path)
94
95
        return self._upload_files

Markus Scheidgen's avatar
Markus Scheidgen committed
96
    def get_logger(self, **kwargs):
97
98
99
100
        """
        Returns a wrapped logger that additionally saves all entries to the calculation
        processing log in the archive.
        """
101
102
103
        logger = super().get_logger()
        logger = logger.bind(
            upload_id=self.upload_id, mainfile=self.mainfile, calc_id=self.calc_id, **kwargs)
104

105
        if self._calc_proc_logwriter_ctx is None:
106
            self._calc_proc_logwriter_ctx = self.upload_files.archive_log_file(self.calc_id, 'wt')
107
            self._calc_proc_logwriter = self._calc_proc_logwriter_ctx.__enter__()  # pylint: disable=E1101
108

109
        def save_to_calc_log(logger, method_name, event_dict):
110
111
112
113
114
115
116
117
118
119
120
            if self._calc_proc_logwriter is not None:
                program = event_dict.get('normalizer', 'parser')
                event = event_dict.get('event', '')
                entry = '[%s] %s: %s' % (method_name, program, event)
                if len(entry) > 120:
                    self._calc_proc_logwriter.write(entry[:120])
                    self._calc_proc_logwriter.write('...')
                else:
                    self._calc_proc_logwriter.write(entry)
                self._calc_proc_logwriter.write('\n')

121
122
            return event_dict

123
        return wrap_logger(logger, processors=[save_to_calc_log])
124

Markus Scheidgen's avatar
Markus Scheidgen committed
125
    @process
126
    def process_calc(self):
127
        logger = self.get_logger()
128
        if self.upload is None:
129
            logger.error('calculation upload does not exist')
Markus Scheidgen's avatar
Markus Scheidgen committed
130
131
132
133
134
135

        try:
            self.parsing()
            self.normalizing()
            self.archiving()
        finally:
Markus Scheidgen's avatar
Markus Scheidgen committed
136
            # close loghandler that was not closed due to failures
137
            try:
138
139
140
                if self._calc_proc_logwriter is not None:
                    self._calc_proc_logwriter.close()
                    self._calc_proc_logwriter = None
141
142
143
144
            except Exception as e:
                logger.error('could not close calculation proc log', exc_info=e)

            # inform parent proc about completion
145
            self.upload.completed_child()
Markus Scheidgen's avatar
Markus Scheidgen committed
146
147
148

    @task
    def parsing(self):
149
        context = dict(parser=self.parser, step=self.parser)
150
        logger = self.get_logger(**context)
151
        parser = parser_dict[self.parser]
Markus Scheidgen's avatar
Markus Scheidgen committed
152

153
        with utils.timer(logger, 'parser executed', input_size=self.mainfile_file.size):
154
155
            self._parser_backend = parser.run(
                self.upload_files.raw_file_object(self.mainfile).os_path, logger=logger)
Markus Scheidgen's avatar
Markus Scheidgen committed
156

157
158
        self._parser_backend.openNonOverlappingSection('section_calculation_info')
        self._parser_backend.addValue('upload_id', self.upload_id)
159
160
        self._parser_backend.addValue('calc_id', self.calc_id)
        self._parser_backend.addValue('calc_hash', self.upload_files.calc_hash(self.mainfile))
161
162
163
        self._parser_backend.addValue('main_file', self.mainfile)
        self._parser_backend.addValue('parser_name', self.parser)

Markus Scheidgen's avatar
Markus Scheidgen committed
164
        if self._parser_backend.status[0] != 'ParseSuccess':
165
            logger.error(self._parser_backend.status[1])
Markus Scheidgen's avatar
Markus Scheidgen committed
166
            error = self._parser_backend.status[1]
167
            self._parser_backend.addValue('parse_status', 'ParseFailure')
168
            self.fail(error, level=logging.DEBUG, **context)
169
170
171
172
173
        else:
            self._parser_backend.addValue('parse_status', 'ParseSuccess')

        self._parser_backend.closeNonOverlappingSection('section_calculation_info')

174
175
176
177
178
179
        self._parser_backend.openNonOverlappingSection('section_repository_info')
        self._parser_backend.addValue('repository_archive_gid', '%s/%s' % (self.upload_id, self.calc_id))
        self._parser_backend.addValue(
            'repository_filepaths', self.upload_files.calc_files(self.mainfile))
        self._parser_backend.closeNonOverlappingSection('section_repository_info')

180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
        self.add_processor_info(self.parser)

    @contextmanager
    def use_parser_backend(self, processor_name):
        self._parser_backend.reset_status()
        yield self._parser_backend
        self.add_processor_info(processor_name)

    def add_processor_info(self, processor_name: str) -> None:
        self._parser_backend.openContext('/section_calculation_info/0')
        self._parser_backend.openNonOverlappingSection('section_archive_processing_info')
        self._parser_backend.addValue('archive_processor_name', processor_name)

        if self._parser_backend.status[0] == 'ParseSuccess':
            warnings = getattr(self._parser_backend, '_warnings', [])
            if len(warnings) > 0:
                self._parser_backend.addValue('archive_processor_status', 'WithWarnings')
                self._parser_backend.addValue('archive_processor_warning_number', len(warnings))
                self._parser_backend.addArrayValues('archive_processor_warnings', [str(warning) for warning in warnings])
            else:
                self._parser_backend.addValue('archive_processor_status', 'Success')
        else:
202
            errors = self._parser_backend.tasks_status[1]
203
204
205
206
            self._parser_backend.addValue('archive_processor_error', str(errors))

        self._parser_backend.closeNonOverlappingSection('section_archive_processing_info')
        self._parser_backend.closeContext('/section_calculation_info/0')
Markus Scheidgen's avatar
Markus Scheidgen committed
207
208
209
210
211

    @task
    def normalizing(self):
        for normalizer in normalizers:
            normalizer_name = normalizer.__name__
212
            context = dict(normalizer=normalizer_name, step=normalizer_name)
213
            logger = self.get_logger(**context)
Markus Scheidgen's avatar
Markus Scheidgen committed
214
215

            with utils.timer(
216
                    logger, 'normalizer executed', input_size=self.mainfile_file.size):
217
218
                with self.use_parser_backend(normalizer_name) as backend:
                    normalizer(backend).normalize(logger=logger)
Markus Scheidgen's avatar
Markus Scheidgen committed
219

220
221
            failed = self._parser_backend.status[0] != 'ParseSuccess'
            if failed:
222
                logger.error(self._parser_backend.status[1])
Markus Scheidgen's avatar
Markus Scheidgen committed
223
                error = self._parser_backend.status[1]
224
                self.fail(error, level=logging.WARNING, **context)
225
226
227
228
                break
            else:
                logger.debug(
                    'completed normalizer successfully', normalizer=normalizer_name)
Markus Scheidgen's avatar
Markus Scheidgen committed
229
230
231

    @task
    def archiving(self):
232
233
        logger = self.get_logger()

234
        # persist the repository metadata
235
        with utils.timer(logger, 'indexed', step='index'):
Markus Scheidgen's avatar
Markus Scheidgen committed
236
237
            self.upload_files.metadata.insert(
                self._parser_backend.to_calc_with_metadata().to_dict())
238

239
        # persist the archive
Markus Scheidgen's avatar
Markus Scheidgen committed
240
241
242
        with utils.timer(
                logger, 'archived', step='archive',
                input_size=self.mainfile_file.size) as log_data:
243
            with self.upload_files.archive_file(self.calc_id, 'wt') as out:
244
245
                self._parser_backend.write_json(out, pretty=True)

246
            log_data.update(archive_size=self.upload_files.archive_file_object(self.calc_id).size)
Markus Scheidgen's avatar
Markus Scheidgen committed
247
248
249
250
251
252

        # close loghandler
        if self._calc_proc_logwriter is not None:
            with utils.timer(
                    logger, 'archived log', step='archive_log',
                    input_size=self.mainfile_file.size) as log_data:
253
                self._calc_proc_logwriter_ctx.__exit__(None, None, None)  # pylint: disable=E1101
254
                self._calc_proc_logwriter = None
Markus Scheidgen's avatar
Markus Scheidgen committed
255

256
                log_data.update(log_size=self.upload_files.archive_log_file_object(self.calc_id).size)
Markus Scheidgen's avatar
Markus Scheidgen committed
257

Markus Scheidgen's avatar
Markus Scheidgen committed
258
259
260
261
    def to_calc_with_metadata(self) -> CalcWithMetadata:
        upload_files = UploadFiles.get(self.upload_id, is_authorized=lambda: True)
        if self.upload_files is None:
            raise KeyError
262

Markus Scheidgen's avatar
Markus Scheidgen committed
263
264
        _data = upload_files.metadata.get(self.calc_id)
        return CalcWithMetadata(**_data)
265
266


Markus Scheidgen's avatar
Markus Scheidgen committed
267
class Upload(Chord):
Markus Scheidgen's avatar
Markus Scheidgen committed
268
269
270
271
272
273
    """
    Represents uploads in the databases. Provides persistence access to the files storage,
    and processing state.

    Attributes:
        name: optional user provided upload name
274
        local_path: optional local path, e.g. for files that are already somewhere on the server
275
        metadata: optional user provided additional meta data
Markus Scheidgen's avatar
Markus Scheidgen committed
276
277
        upload_id: the upload id generated by the database
        upload_time: the timestamp when the system realised the upload
278
        user_id: the id of the user that created this upload
Markus Scheidgen's avatar
Markus Scheidgen committed
279
280
281
282
283
284
    """
    id_field = 'upload_id'

    upload_id = StringField(primary_key=True)

    name = StringField(default=None)
285
    local_path = StringField(default=None)
286
    metadata = DictField(default=None)
Markus Scheidgen's avatar
Markus Scheidgen committed
287
    upload_time = DateTimeField()
288
    user_id = StringField(required=True)
Markus Scheidgen's avatar
Markus Scheidgen committed
289
290
291

    meta: Any = {
        'indexes': [
292
            'user_id', 'tasks_status'
Markus Scheidgen's avatar
Markus Scheidgen committed
293
294
295
296
297
        ]
    }

    def __init__(self, **kwargs):
        super().__init__(**kwargs)
298
        self._upload_files: ArchiveBasedStagingUploadFiles = None
Markus Scheidgen's avatar
Markus Scheidgen committed
299
300
301
302
303
304

    @classmethod
    def get(cls, id):
        return cls.get_by_id(id, 'upload_id')

    @classmethod
305
    def user_uploads(cls, user: coe_repo.User) -> List['Upload']:
Markus Scheidgen's avatar
Markus Scheidgen committed
306
        """ Returns all uploads for the given user. Currently returns all uploads. """
307
        return cls.objects(user_id=str(user.user_id))
Markus Scheidgen's avatar
Markus Scheidgen committed
308

309
310
311
312
    @property
    def uploader(self):
        return coe_repo.User.from_user_id(self.user_id)

Markus Scheidgen's avatar
Markus Scheidgen committed
313
314
315
316
317
318
319
320
321
322
323
    def get_logger(self, **kwargs):
        logger = super().get_logger()
        logger = logger.bind(upload_id=self.upload_id, **kwargs)
        return logger

    @classmethod
    def create(cls, **kwargs) -> 'Upload':
        """
        Creates a new upload for the given user, a user given name is optional.
        It will populate the record with a signed url and pending :class:`UploadProc`.
        The upload will be already saved to the database.
324
325

        Arguments:
326
            user (coe_repo.User): The user that created the upload.
Markus Scheidgen's avatar
Markus Scheidgen committed
327
        """
328
        user: coe_repo.User = kwargs['user']
329
330
331
        del(kwargs['user'])
        if 'upload_id' not in kwargs:
            kwargs.update(upload_id=utils.create_uuid())
332
        kwargs.update(user_id=str(user.user_id))
Markus Scheidgen's avatar
Markus Scheidgen committed
333
        self = super().create(**kwargs)
334

Markus Scheidgen's avatar
Markus Scheidgen committed
335
        self._continue_with('uploading')
336

Markus Scheidgen's avatar
Markus Scheidgen committed
337
338
        return self

339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
    def delete(self):
        """ Deletes this upload process state entry and its calcs. """
        Calc.objects(upload_id=self.upload_id).delete()
        super().delete()

    @process
    def delete_upload(self):
        """
        Deletes of the upload, including its processing state and
        staging files.
        """
        logger = self.get_logger()

        with utils.lnr(logger, 'staged upload delete failed'):
            with utils.timer(
                    logger, 'staged upload deleted', step='delete',
                    upload_size=self.upload_files.size):
                self.upload_files.delete()
                self.delete()

        return True  # do not save the process status on the delete upload
360

361
    @process
362
    def publish_upload(self):
363
364
365
366
367
368
369
        """
        Moves the upload out of staging to add it to the coe repository. It will
        pack the staging upload files in to public upload files, add entries to the
        coe repository db and remove this instance and its calculation from the
        processing state db.
        """
        logger = self.get_logger()
370

371
        with utils.lnr(logger, 'publish failed'):
372
            with utils.timer(
373
                    logger, 'upload added to repository', step='publish',
374
                    upload_size=self.upload_files.size):
Markus Scheidgen's avatar
Markus Scheidgen committed
375
                coe_repo.Upload.add(self.to_upload_with_metadata())
376
377

            with utils.timer(
378
                    logger, 'staged upload files packed', step='publish',
379
380
381
382
                    upload_size=self.upload_files.size):
                self.upload_files.pack()

            with utils.timer(
383
                    logger, 'staged upload deleted', step='publish',
384
385
386
                    upload_size=self.upload_files.size):
                self.upload_files.delete()
                self.delete()
Markus Scheidgen's avatar
Markus Scheidgen committed
387

388
        return True  # do not save the process status on the delete upload
389

Markus Scheidgen's avatar
Markus Scheidgen committed
390
    @process
391
    def process_upload(self):
Markus Scheidgen's avatar
Markus Scheidgen committed
392
393
394
395
396
397
398
        self.extracting()
        self.parse_all()

    @task
    def uploading(self):
        pass

399
    @property
400
401
    def upload_files(self) -> ArchiveBasedStagingUploadFiles:
        if not self._upload_files:
402
            self._upload_files = ArchiveBasedStagingUploadFiles(self.upload_id, is_authorized=lambda: True, local_path=self.local_path)
403
        return self._upload_files
404

Markus Scheidgen's avatar
Markus Scheidgen committed
405
406
    @task
    def extracting(self):
407
408
409
410
411
412
        """
        Task performed before the actual parsing/normalizing. Extracting and bagging
        the uploaded files, computing all keys, create an *upload* entry in the NOMAD-coe
        repository db, etc.
        """
        # extract the uploaded file, this will also create a bagit bag.
Markus Scheidgen's avatar
Markus Scheidgen committed
413
414
        logger = self.get_logger()
        try:
Markus Scheidgen's avatar
Markus Scheidgen committed
415
416
            with utils.timer(
                    logger, 'upload extracted', step='extracting',
417
418
                    upload_size=self.upload_files.size):
                self.upload_files.extract()
419
        except KeyError:
420
            self.fail('processing requested for non existing upload', log_level=logging.ERROR)
421
422
423
            return
        except ExtractError:
            self.fail('bad .zip/.tar file', log_level=logging.INFO)
Markus Scheidgen's avatar
Markus Scheidgen committed
424
425
            return

426
    def match_mainfiles(self) -> Generator[Tuple[str, object], None, None]:
427
428
429
430
431
432
433
        """
        Generator function that matches all files in the upload to all parsers to
        determine the upload's mainfiles.

        Returns:
            Tuples of mainfile, filename, and parsers
        """
434
        for filename in self.upload_files.raw_file_manifest():
435
436
            for parser in parsers:
                try:
437
                    with self.upload_files.raw_file(filename) as mainfile_f:
438
                        if parser.is_mainfile(filename, lambda fn: mainfile_f):
439
                            yield filename, parser
440
                except Exception as e:
441
                    self.get_logger().error(
442
443
444
                        'exception while matching pot. mainfile',
                        mainfile=filename, exc_info=e)

Markus Scheidgen's avatar
Markus Scheidgen committed
445
446
    @task
    def parse_all(self):
447
        """
448
        Identified mainfile/parser combinations among the upload's files, creates
449
450
        respective :class:`Calc` instances, and triggers their processing.
        """
451
452
        logger = self.get_logger()

Markus Scheidgen's avatar
Markus Scheidgen committed
453
        # TODO: deal with multiple possible parser specs
Markus Scheidgen's avatar
Markus Scheidgen committed
454
455
        with utils.timer(
                logger, 'upload extracted', step='matching',
456
                upload_size=self.upload_files.size):
457
            total_calcs = 0
458
            for filename, parser in self.match_mainfiles():
459
                calc = Calc.create(
460
                    calc_id=self.upload_files.calc_id(filename),
461
462
463
                    mainfile=filename, parser=parser.name,
                    upload_id=self.upload_id)

464
                calc.process_calc()
465
                total_calcs += 1
Markus Scheidgen's avatar
Markus Scheidgen committed
466

467
468
469
470
471
        # have to save the total_calcs information for chord management
        self.spwaned_childred(total_calcs)

    def join(self):
        self.cleanup()
Markus Scheidgen's avatar
Markus Scheidgen committed
472
473
474

    @task
    def cleanup(self):
475
476
477
478
479
480
481
482
483
484
        # send email about process finish
        user = self.uploader
        name = '%s %s' % (user.first_name, user.last_name)
        message = '\n'.join([
            'Dear %s,' % name,
            '',
            'your data %suploaded %s has completed processing.' % (
                self.name if self.name else '', self.upload_time.isoformat()),
            'You can review your data on your upload page: %s' % config.services.upload_url
        ])
485
486
487
488
489
490
491
        try:
            infrastructure.send_mail(
                name=name, email=user.email, message=message, subject='Processing completed')
        except Exception as e:
            # probably due to email configuration problems
            # don't fail or present this error to clients
            self.logger.error('could not send after processing email', exc_info=e)
Markus Scheidgen's avatar
Markus Scheidgen committed
492
493

    @property
494
    def processed_calcs(self):
495
        return Calc.objects(upload_id=self.upload_id, tasks_status__in=[SUCCESS, FAILURE]).count()
496
497
498
499
500
501
502

    @property
    def total_calcs(self):
        return Calc.objects(upload_id=self.upload_id).count()

    @property
    def failed_calcs(self):
503
        return Calc.objects(upload_id=self.upload_id, tasks_status=FAILURE).count()
504

505
506
    @property
    def pending_calcs(self):
507
        return Calc.objects(upload_id=self.upload_id, tasks_status=PENDING).count()
508

509
510
    def all_calcs(self, start, end, order_by='mainfile'):
        return Calc.objects(upload_id=self.upload_id)[start:end].order_by(order_by)
511
512
513

    @property
    def calcs(self):
Markus Scheidgen's avatar
Markus Scheidgen committed
514
        return Calc.objects(upload_id=self.upload_id, tasks_status=SUCCESS)
Markus Scheidgen's avatar
Markus Scheidgen committed
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

    def to_upload_with_metadata(self) -> UploadWithMetadata:
        calc_metadata = dict()
        user_upload_time = None
        if self.metadata is not None:
            user_upload_time = self.metadata.get('_upload_time')
            for calc in self.metadata.get('calculations', []):
                calc_metadata[calc['mainfile']] = calc

        def apply_metadata(calc):
            metadata = calc_metadata.get(calc.mainfile, self.metadata)
            if metadata is None:
                return calc

            calc.pid = metadata.get('_pid')
            calc.comment = metadata.get('comment')
            calc.upload_time = metadata.get('_upload_time')
            uploader_id = metadata.get('_uploader')
            if uploader_id is not None:
                calc.uploader = utils.POPO(id=uploader_id)
            calc.references = [utils.POPO(value=ref) for ref in metadata.get('references', [])]
            calc.with_embargo = metadata.get('with_embargo', False)
            calc.coauthors = [
                utils.POPO(id=user) for user in metadata.get('coauthors', [])]
            calc.shared_with = [
                utils.POPO(id=user) for user in metadata.get('shared_with', [])]
            calc.datasets = [
                utils.POPO(id=ds['id'], doi=utils.POPO(value=ds.get('_doi')), name=ds.get('_name'))
                for ds in metadata.get('datasets', [])]

            return calc

        result = UploadWithMetadata(
            upload_id=self.upload_id,
            uploader=utils.POPO(id=int(self.user_id)),
            upload_time=self.upload_time if user_upload_time is None else user_upload_time)

        result.calcs = [
            apply_metadata(calc.to_calc_with_metadata()) for calc in self.calcs]

        return result