data.py 22.6 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, Dict
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
32
import os.path
Markus Scheidgen's avatar
Markus Scheidgen committed
33

34
from nomad import utils, coe_repo, config, infrastructure, search
Markus Scheidgen's avatar
Markus Scheidgen committed
35
from nomad.files import PathObject, UploadFiles, ExtractError, ArchiveBasedStagingUploadFiles
36
from nomad.processing.base import Proc, process, task, PENDING, SUCCESS, FAILURE
37
from nomad.parsing import parser_dict, match_parser
Markus Scheidgen's avatar
Markus Scheidgen committed
38
from nomad.normalizing import normalizers
Markus Scheidgen's avatar
Markus Scheidgen committed
39
from nomad.datamodel import UploadWithMetadata, CalcWithMetadata
Markus Scheidgen's avatar
Markus Scheidgen committed
40
41


Markus Scheidgen's avatar
Markus Scheidgen committed
42
class Calc(Proc):
Markus Scheidgen's avatar
Markus Scheidgen committed
43
44
45
46
47
48
49
50
51
52
    """
    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:
53
        calc_id: the calc_id of this calc
Markus Scheidgen's avatar
Markus Scheidgen committed
54
55
56
57
        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
    """
58
    calc_id = StringField(primary_key=True)
Markus Scheidgen's avatar
Markus Scheidgen committed
59
60
61
62
63
64
    upload_id = StringField()
    mainfile = StringField()
    parser = StringField()

    meta: Any = {
        'indices': [
65
            'upload_id', 'mainfile', 'code', 'parser', 'tasks_status'
Markus Scheidgen's avatar
Markus Scheidgen committed
66
67
68
69
70
71
        ]
    }

    def __init__(self, *args, **kwargs):
        super().__init__(*args, **kwargs)
        self._parser_backend = None
72
73
        self._upload: Upload = None
        self._upload_files: ArchiveBasedStagingUploadFiles = None
74
        self._calc_proc_logwriter = None
75
        self._calc_proc_logwriter_ctx: ContextManager = None
Markus Scheidgen's avatar
Markus Scheidgen committed
76
77
78

    @classmethod
    def get(cls, id):
79
        return cls.get_by_id(id, 'calc_id')
Markus Scheidgen's avatar
Markus Scheidgen committed
80

Markus Scheidgen's avatar
Markus Scheidgen committed
81
    @property
82
83
    def mainfile_file(self) -> PathObject:
        return self.upload_files.raw_file_object(self.mainfile)
Markus Scheidgen's avatar
Markus Scheidgen committed
84

85
86
87
88
89
90
    @property
    def upload(self) -> 'Upload':
        if not self._upload:
            self._upload = Upload.get(self.upload_id)
        return self._upload

91
92
93
    @property
    def upload_files(self) -> ArchiveBasedStagingUploadFiles:
        if not self._upload_files:
94
            self._upload_files = ArchiveBasedStagingUploadFiles(self.upload_id, is_authorized=lambda: True, local_path=self.upload.local_path)
95
96
        return self._upload_files

Markus Scheidgen's avatar
Markus Scheidgen committed
97
    def get_logger(self, **kwargs):
98
99
100
101
        """
        Returns a wrapped logger that additionally saves all entries to the calculation
        processing log in the archive.
        """
102
103
104
        logger = super().get_logger()
        logger = logger.bind(
            upload_id=self.upload_id, mainfile=self.mainfile, calc_id=self.calc_id, **kwargs)
105

106
        if self._calc_proc_logwriter_ctx is None:
107
            self._calc_proc_logwriter_ctx = self.upload_files.archive_log_file(self.calc_id, 'wt')
108
            self._calc_proc_logwriter = self._calc_proc_logwriter_ctx.__enter__()  # pylint: disable=E1101
109

110
        def save_to_calc_log(logger, method_name, event_dict):
111
112
113
114
115
116
117
118
119
120
121
            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')

122
123
            return event_dict

124
        return wrap_logger(logger, processors=[save_to_calc_log])
125

Markus Scheidgen's avatar
Markus Scheidgen committed
126
    @process
127
    def process_calc(self):
128
        logger = self.get_logger()
129
        if self.upload is None:
130
            logger.error('calculation upload does not exist')
Markus Scheidgen's avatar
Markus Scheidgen committed
131
132
133
134
135
136

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

145
146
147
148
149
150
151
    def on_process_complete(self, process_name):
        # the save might be necessary to correctly read the join condition from the db
        self.save()
        # in case of error, the process_name might be unknown
        if process_name == 'process_calc' or process_name is None:
            self.upload.reload()
            self.upload.check_join()
Markus Scheidgen's avatar
Markus Scheidgen committed
152
153
154

    @task
    def parsing(self):
155
        context = dict(parser=self.parser, step=self.parser)
156
        logger = self.get_logger(**context)
157
        parser = parser_dict[self.parser]
Markus Scheidgen's avatar
Markus Scheidgen committed
158

159
        with utils.timer(logger, 'parser executed', input_size=self.mainfile_file.size):
160
161
            self._parser_backend = parser.run(
                self.upload_files.raw_file_object(self.mainfile).os_path, logger=logger)
Markus Scheidgen's avatar
Markus Scheidgen committed
162

163
164
        self._parser_backend.openNonOverlappingSection('section_calculation_info')
        self._parser_backend.addValue('upload_id', self.upload_id)
165
166
        self._parser_backend.addValue('calc_id', self.calc_id)
        self._parser_backend.addValue('calc_hash', self.upload_files.calc_hash(self.mainfile))
167
168
169
        self._parser_backend.addValue('main_file', self.mainfile)
        self._parser_backend.addValue('parser_name', self.parser)

Markus Scheidgen's avatar
Markus Scheidgen committed
170
        if self._parser_backend.status[0] != 'ParseSuccess':
171
            logger.error(self._parser_backend.status[1])
Markus Scheidgen's avatar
Markus Scheidgen committed
172
            error = self._parser_backend.status[1]
173
            self._parser_backend.addValue('parse_status', 'ParseFailure')
174
            self.fail(error, level=logging.DEBUG, **context)
175
176
177
178
179
        else:
            self._parser_backend.addValue('parse_status', 'ParseSuccess')

        self._parser_backend.closeNonOverlappingSection('section_calculation_info')

180
181
182
183
184
185
        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')

186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
        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:
208
            errors = self._parser_backend.tasks_status[1]
209
210
211
212
            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
213
214
215
216
217

    @task
    def normalizing(self):
        for normalizer in normalizers:
            normalizer_name = normalizer.__name__
218
            context = dict(normalizer=normalizer_name, step=normalizer_name)
219
            logger = self.get_logger(**context)
Markus Scheidgen's avatar
Markus Scheidgen committed
220
221

            with utils.timer(
222
                    logger, 'normalizer executed', input_size=self.mainfile_file.size):
223
224
                with self.use_parser_backend(normalizer_name) as backend:
                    normalizer(backend).normalize(logger=logger)
Markus Scheidgen's avatar
Markus Scheidgen committed
225

226
227
            failed = self._parser_backend.status[0] != 'ParseSuccess'
            if failed:
228
                logger.error(self._parser_backend.status[1])
Markus Scheidgen's avatar
Markus Scheidgen committed
229
                error = self._parser_backend.status[1]
230
                self.fail(error, level=logging.WARNING, **context)
231
232
233
234
                break
            else:
                logger.debug(
                    'completed normalizer successfully', normalizer=normalizer_name)
Markus Scheidgen's avatar
Markus Scheidgen committed
235
236
237

    @task
    def archiving(self):
238
239
        logger = self.get_logger()

240
241
        calc_with_metadata = self._parser_backend.to_calc_with_metadata()

242
        # persist the repository metadata
243
244
245
246
247
        with utils.timer(logger, 'saved repo metadata', step='persist'):
            self.upload_files.metadata.insert(calc_with_metadata.to_dict())

        # index in search
        with utils.timer(logger, 'indexed', step='persist'):
248
249
            calc_with_metadata.update(published=False, uploader=self.upload.uploader.to_popo())
            search.Entry.from_calc_with_metadata(calc_with_metadata).save()
250

251
        # persist the archive
Markus Scheidgen's avatar
Markus Scheidgen committed
252
        with utils.timer(
253
                logger, 'archived', step='persist',
Markus Scheidgen's avatar
Markus Scheidgen committed
254
                input_size=self.mainfile_file.size) as log_data:
255
            with self.upload_files.archive_file(self.calc_id, 'wt') as out:
256
257
                self._parser_backend.write_json(out, pretty=True)

258
            log_data.update(archive_size=self.upload_files.archive_file_object(self.calc_id).size)
Markus Scheidgen's avatar
Markus Scheidgen committed
259
260
261
262

        # close loghandler
        if self._calc_proc_logwriter is not None:
            with utils.timer(
263
                    logger, 'archived log', step='persist',
Markus Scheidgen's avatar
Markus Scheidgen committed
264
                    input_size=self.mainfile_file.size) as log_data:
265
                self._calc_proc_logwriter_ctx.__exit__(None, None, None)  # pylint: disable=E1101
266
                self._calc_proc_logwriter = None
Markus Scheidgen's avatar
Markus Scheidgen committed
267

268
                log_data.update(log_size=self.upload_files.archive_log_file_object(self.calc_id).size)
Markus Scheidgen's avatar
Markus Scheidgen committed
269

Markus Scheidgen's avatar
Markus Scheidgen committed
270
271
272
273
    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
274

Markus Scheidgen's avatar
Markus Scheidgen committed
275
276
        _data = upload_files.metadata.get(self.calc_id)
        return CalcWithMetadata(**_data)
277
278


279
class Upload(Proc):
Markus Scheidgen's avatar
Markus Scheidgen committed
280
281
282
283
284
285
    """
    Represents uploads in the databases. Provides persistence access to the files storage,
    and processing state.

    Attributes:
        name: optional user provided upload name
286
        local_path: optional local path, e.g. for files that are already somewhere on the server
287
        metadata: optional user provided additional meta data
Markus Scheidgen's avatar
Markus Scheidgen committed
288
289
        upload_id: the upload id generated by the database
        upload_time: the timestamp when the system realised the upload
290
        user_id: the id of the user that created this upload
Markus Scheidgen's avatar
Markus Scheidgen committed
291
292
293
294
295
296
    """
    id_field = 'upload_id'

    upload_id = StringField(primary_key=True)

    name = StringField(default=None)
297
    local_path = StringField(default=None)
298
    metadata = DictField(default=None)
Markus Scheidgen's avatar
Markus Scheidgen committed
299
    upload_time = DateTimeField()
300
    user_id = StringField(required=True)
Markus Scheidgen's avatar
Markus Scheidgen committed
301
302
303

    meta: Any = {
        'indexes': [
304
            'user_id', 'tasks_status'
Markus Scheidgen's avatar
Markus Scheidgen committed
305
306
307
308
309
        ]
    }

    def __init__(self, **kwargs):
        super().__init__(**kwargs)
310
        self._upload_files: ArchiveBasedStagingUploadFiles = None
Markus Scheidgen's avatar
Markus Scheidgen committed
311
312
313
314
315
316

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

    @classmethod
317
    def user_uploads(cls, user: coe_repo.User) -> List['Upload']:
Markus Scheidgen's avatar
Markus Scheidgen committed
318
        """ Returns all uploads for the given user. Currently returns all uploads. """
319
        return cls.objects(user_id=str(user.user_id))
Markus Scheidgen's avatar
Markus Scheidgen committed
320

321
322
323
324
    @property
    def uploader(self):
        return coe_repo.User.from_user_id(self.user_id)

Markus Scheidgen's avatar
Markus Scheidgen committed
325
326
327
328
329
330
331
332
333
334
335
    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.
336
337

        Arguments:
338
            user (coe_repo.User): The user that created the upload.
Markus Scheidgen's avatar
Markus Scheidgen committed
339
        """
340
        user: coe_repo.User = kwargs['user']
341
342
343
        del(kwargs['user'])
        if 'upload_id' not in kwargs:
            kwargs.update(upload_id=utils.create_uuid())
344
        kwargs.update(user_id=str(user.user_id))
Markus Scheidgen's avatar
Markus Scheidgen committed
345
        self = super().create(**kwargs)
346

Markus Scheidgen's avatar
Markus Scheidgen committed
347
        self._continue_with('uploading')
348

Markus Scheidgen's avatar
Markus Scheidgen committed
349
350
        return self

351
352
353
354
355
356
357
358
359
360
361
362
363
364
    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'):
365
366
367
            with utils.timer(
                    logger, 'upload deleted from index', step='delete',
                    upload_size=self.upload_files.size):
368
                search.delete_upload(self.upload_id)
369

370
371
372
373
374
375
376
            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
377

378
    @process
379
    def publish_upload(self):
380
381
382
383
384
385
386
        """
        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()
387

388
        with utils.lnr(logger, 'publish failed'):
389
390
            upload_with_metadata = self.to_upload_with_metadata()

391
            with utils.timer(
392
                    logger, 'upload added to repository', step='publish',
393
                    upload_size=self.upload_files.size):
394
                coe_repo.Upload.add(upload_with_metadata)
395
396

            with utils.timer(
397
                    logger, 'staged upload files packed', step='publish',
398
                    upload_size=self.upload_files.size):
399
400
401
402
                for calc_metadata in upload_with_metadata.calcs:
                    calc_metadata.published = True
                    self.upload_files.metadata.update(
                        calc_id=calc_metadata.calc_id, updates=calc_metadata.to_dict())
403
404
                self.upload_files.pack()

405
406
407
            with utils.timer(
                    logger, 'index updated', step='publish',
                    upload_size=self.upload_files.size):
408
409
                coe_upload = coe_repo.Upload.from_upload_id(upload_with_metadata.upload_id)
                if coe_upload is not None:
410
411
                    search.publish(
                        [coe_calc.to_calc_with_metadata() for coe_calc in coe_upload.calcs])
412

413
            with utils.timer(
414
                    logger, 'staged upload deleted', step='publish',
415
416
417
                    upload_size=self.upload_files.size):
                self.upload_files.delete()
                self.delete()
Markus Scheidgen's avatar
Markus Scheidgen committed
418

419
        return True  # do not save the process status on the delete upload
420

Markus Scheidgen's avatar
Markus Scheidgen committed
421
    @process
422
    def process_upload(self):
Markus Scheidgen's avatar
Markus Scheidgen committed
423
424
425
426
427
428
429
        self.extracting()
        self.parse_all()

    @task
    def uploading(self):
        pass

430
    @property
431
432
    def upload_files(self) -> ArchiveBasedStagingUploadFiles:
        if not self._upload_files:
433
            self._upload_files = ArchiveBasedStagingUploadFiles(self.upload_id, is_authorized=lambda: True, local_path=self.local_path)
434
        return self._upload_files
435

Markus Scheidgen's avatar
Markus Scheidgen committed
436
437
    @task
    def extracting(self):
438
439
440
441
442
443
        """
        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
444
445
        logger = self.get_logger()
        try:
Markus Scheidgen's avatar
Markus Scheidgen committed
446
447
            with utils.timer(
                    logger, 'upload extracted', step='extracting',
448
449
                    upload_size=self.upload_files.size):
                self.upload_files.extract()
450
        except KeyError:
451
            self.fail('processing requested for non existing upload', log_level=logging.ERROR)
452
453
454
            return
        except ExtractError:
            self.fail('bad .zip/.tar file', log_level=logging.INFO)
Markus Scheidgen's avatar
Markus Scheidgen committed
455
456
            return

457
    def match_mainfiles(self) -> Generator[Tuple[str, object], None, None]:
458
459
460
461
462
463
464
        """
        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
        """
465
        directories_with_match: Dict[str, str] = dict()
466
        for filename in self.upload_files.raw_file_manifest():
467
            try:
468
                parser = match_parser(filename, self.upload_files)
469
                if parser is not None:
470
471
472
473
474
475
476
477
                    directory = os.path.dirname(filename)
                    if directory in directories_with_match:
                        self.warnings.append(
                            'The directory %s contains data from multiple code runs. '
                            'Nomad only processed %s.' % (directory, os.path.basename(filename)))
                    else:
                        directories_with_match[directory] = filename
                        yield filename, parser
478
479
480
481
            except Exception as e:
                self.get_logger().error(
                    'exception while matching pot. mainfile',
                    mainfile=filename, exc_info=e)
482

Markus Scheidgen's avatar
Markus Scheidgen committed
483
484
    @task
    def parse_all(self):
485
        """
486
        Identified mainfile/parser combinations among the upload's files, creates
487
488
        respective :class:`Calc` instances, and triggers their processing.
        """
489
490
        logger = self.get_logger()

Markus Scheidgen's avatar
Markus Scheidgen committed
491
492
        with utils.timer(
                logger, 'upload extracted', step='matching',
493
494
                upload_size=self.upload_files.size):
            for filename, parser in self.match_mainfiles():
495
                calc = Calc.create(
496
                    calc_id=self.upload_files.calc_id(filename),
497
498
499
                    mainfile=filename, parser=parser.name,
                    upload_id=self.upload_id)

500
                calc.process_calc()
Markus Scheidgen's avatar
Markus Scheidgen committed
501

502
503
504
505
506
507
508
509
510
511
512
513
    def on_process_complete(self, process_name):
        if process_name == 'process_upload':
            self.check_join()

    def check_join(self):
        total_calcs = self.total_calcs
        processed_calcs = self.processed_calcs

        self.get_logger().debug('check join', processed_calcs=processed_calcs, total_calcs=total_calcs)
        if not self.process_running and processed_calcs >= total_calcs:
            self.get_logger().debug('join')
            self.join()
514
515
516

    def join(self):
        self.cleanup()
Markus Scheidgen's avatar
Markus Scheidgen committed
517
518
519

    @task
    def cleanup(self):
520
521
522
523
524
525
526
527
528
529
        # 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
        ])
530
531
532
533
534
535
536
        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
537
538

    @property
539
    def processed_calcs(self):
540
        return Calc.objects(upload_id=self.upload_id, tasks_status__in=[SUCCESS, FAILURE]).count()
541
542
543
544
545
546
547

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

    @property
    def failed_calcs(self):
548
        return Calc.objects(upload_id=self.upload_id, tasks_status=FAILURE).count()
549

550
551
    @property
    def pending_calcs(self):
552
        return Calc.objects(upload_id=self.upload_id, tasks_status=PENDING).count()
553

554
555
    def all_calcs(self, start, end, order_by='mainfile'):
        return Calc.objects(upload_id=self.upload_id)[start:end].order_by(order_by)
556
557
558

    @property
    def calcs(self):
Markus Scheidgen's avatar
Markus Scheidgen committed
559
        return Calc.objects(upload_id=self.upload_id, tasks_status=SUCCESS)
Markus Scheidgen's avatar
Markus Scheidgen committed
560
561
562
563
564
565
566
567
568
569
570

    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)
571
572
            if metadata is not None:
                calc.apply_user_metadata(metadata)
Markus Scheidgen's avatar
Markus Scheidgen committed
573
574
575
576
577
578
579
580
581
582
583
            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