data.py 23.3 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, 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
162
163
164
165
166
            try:
                self._parser_backend = parser.run(
                    self.upload_files.raw_file_object(self.mainfile).os_path, logger=logger)
            except Exception as e:
                self.fail(
                    'parser failed with exception', level=logging.ERROR,
                    exc_info=e, error=str(e), **context)
Markus Scheidgen's avatar
Markus Scheidgen committed
167

168
169
        self._parser_backend.openNonOverlappingSection('section_calculation_info')
        self._parser_backend.addValue('upload_id', self.upload_id)
170
171
        self._parser_backend.addValue('calc_id', self.calc_id)
        self._parser_backend.addValue('calc_hash', self.upload_files.calc_hash(self.mainfile))
172
173
174
        self._parser_backend.addValue('main_file', self.mainfile)
        self._parser_backend.addValue('parser_name', self.parser)

Markus Scheidgen's avatar
Markus Scheidgen committed
175
        if self._parser_backend.status[0] != 'ParseSuccess':
176
            logger.error(self._parser_backend.status[1])
Markus Scheidgen's avatar
Markus Scheidgen committed
177
            error = self._parser_backend.status[1]
178
            self._parser_backend.addValue('parse_status', 'ParseFailure')
179
            self.fail(error, level=logging.INFO, **context)
180
181
182
183
184
        else:
            self._parser_backend.addValue('parse_status', 'ParseSuccess')

        self._parser_backend.closeNonOverlappingSection('section_calculation_info')

185
186
187
188
189
190
        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')

191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
        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:
213
            errors = self._parser_backend.tasks_status[1]
214
215
216
217
            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
218
219
220
221
222

    @task
    def normalizing(self):
        for normalizer in normalizers:
            normalizer_name = normalizer.__name__
223
            context = dict(normalizer=normalizer_name, step=normalizer_name)
224
            logger = self.get_logger(**context)
Markus Scheidgen's avatar
Markus Scheidgen committed
225
226

            with utils.timer(
227
                    logger, 'normalizer executed', input_size=self.mainfile_file.size):
228
                with self.use_parser_backend(normalizer_name) as backend:
229
230
231
232
233
234
235
                    try:
                        normalizer(backend).normalize(logger=logger)
                    except Exception as e:
                        self.fail(
                            'normalizer failed with exception', level=logging.ERROR,
                            exc_info=e, error=str(e), **context)
                        self._parser_backend.status = ['ParseFailure', str(e)]
Markus Scheidgen's avatar
Markus Scheidgen committed
236

237
238
            failed = self._parser_backend.status[0] != 'ParseSuccess'
            if failed:
239
                logger.error(self._parser_backend.status[1])
Markus Scheidgen's avatar
Markus Scheidgen committed
240
                error = self._parser_backend.status[1]
241
                self.fail(error, level=logging.WARNING, error=error, **context)
242
243
244
245
                break
            else:
                logger.debug(
                    'completed normalizer successfully', normalizer=normalizer_name)
Markus Scheidgen's avatar
Markus Scheidgen committed
246
247
248

    @task
    def archiving(self):
249
250
        logger = self.get_logger()

251
252
        calc_with_metadata = self._parser_backend.to_calc_with_metadata()

253
        # persist the repository metadata
254
255
256
257
258
        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'):
259
260
            calc_with_metadata.update(published=False, uploader=self.upload.uploader.to_popo())
            search.Entry.from_calc_with_metadata(calc_with_metadata).save()
261

262
        # persist the archive
Markus Scheidgen's avatar
Markus Scheidgen committed
263
        with utils.timer(
264
                logger, 'archived', step='persist',
Markus Scheidgen's avatar
Markus Scheidgen committed
265
                input_size=self.mainfile_file.size) as log_data:
266
            with self.upload_files.archive_file(self.calc_id, 'wt') as out:
267
268
                self._parser_backend.write_json(out, pretty=True)

269
            log_data.update(archive_size=self.upload_files.archive_file_object(self.calc_id).size)
Markus Scheidgen's avatar
Markus Scheidgen committed
270
271
272
273

        # close loghandler
        if self._calc_proc_logwriter is not None:
            with utils.timer(
274
                    logger, 'archived log', step='persist',
Markus Scheidgen's avatar
Markus Scheidgen committed
275
                    input_size=self.mainfile_file.size) as log_data:
276
                self._calc_proc_logwriter_ctx.__exit__(None, None, None)  # pylint: disable=E1101
277
                self._calc_proc_logwriter = None
Markus Scheidgen's avatar
Markus Scheidgen committed
278

279
                log_data.update(log_size=self.upload_files.archive_log_file_object(self.calc_id).size)
Markus Scheidgen's avatar
Markus Scheidgen committed
280

Markus Scheidgen's avatar
Markus Scheidgen committed
281
282
283
284
    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
285

Markus Scheidgen's avatar
Markus Scheidgen committed
286
287
        _data = upload_files.metadata.get(self.calc_id)
        return CalcWithMetadata(**_data)
288
289


290
class Upload(Proc):
Markus Scheidgen's avatar
Markus Scheidgen committed
291
292
293
294
295
296
    """
    Represents uploads in the databases. Provides persistence access to the files storage,
    and processing state.

    Attributes:
        name: optional user provided upload name
297
        local_path: optional local path, e.g. for files that are already somewhere on the server
298
        metadata: optional user provided additional meta data
Markus Scheidgen's avatar
Markus Scheidgen committed
299
300
        upload_id: the upload id generated by the database
        upload_time: the timestamp when the system realised the upload
301
        user_id: the id of the user that created this upload
Markus Scheidgen's avatar
Markus Scheidgen committed
302
303
304
305
306
307
    """
    id_field = 'upload_id'

    upload_id = StringField(primary_key=True)

    name = StringField(default=None)
308
    local_path = StringField(default=None)
309
    metadata = DictField(default=None)
Markus Scheidgen's avatar
Markus Scheidgen committed
310
    upload_time = DateTimeField()
311
    user_id = StringField(required=True)
Markus Scheidgen's avatar
Markus Scheidgen committed
312
313
314

    meta: Any = {
        'indexes': [
315
            'user_id', 'tasks_status'
Markus Scheidgen's avatar
Markus Scheidgen committed
316
317
318
319
320
        ]
    }

    def __init__(self, **kwargs):
        super().__init__(**kwargs)
321
        self._upload_files: ArchiveBasedStagingUploadFiles = None
Markus Scheidgen's avatar
Markus Scheidgen committed
322
323
324
325
326
327

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

    @classmethod
328
    def user_uploads(cls, user: coe_repo.User) -> List['Upload']:
Markus Scheidgen's avatar
Markus Scheidgen committed
329
        """ Returns all uploads for the given user. Currently returns all uploads. """
330
        return cls.objects(user_id=str(user.user_id))
Markus Scheidgen's avatar
Markus Scheidgen committed
331

332
333
334
335
    @property
    def uploader(self):
        return coe_repo.User.from_user_id(self.user_id)

Markus Scheidgen's avatar
Markus Scheidgen committed
336
337
338
339
340
341
342
343
344
345
346
    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.
347
348

        Arguments:
349
            user (coe_repo.User): The user that created the upload.
Markus Scheidgen's avatar
Markus Scheidgen committed
350
        """
351
        user: coe_repo.User = kwargs['user']
352
353
354
        del(kwargs['user'])
        if 'upload_id' not in kwargs:
            kwargs.update(upload_id=utils.create_uuid())
355
        kwargs.update(user_id=str(user.user_id))
Markus Scheidgen's avatar
Markus Scheidgen committed
356
        self = super().create(**kwargs)
357

Markus Scheidgen's avatar
Markus Scheidgen committed
358
        self._continue_with('uploading')
359

Markus Scheidgen's avatar
Markus Scheidgen committed
360
361
        return self

362
363
364
365
366
367
368
369
370
371
372
373
374
375
    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'):
376
377
378
            with utils.timer(
                    logger, 'upload deleted from index', step='delete',
                    upload_size=self.upload_files.size):
379
                search.delete_upload(self.upload_id)
380

381
382
383
384
385
386
387
            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
388

389
    @process
390
    def publish_upload(self):
391
392
393
394
395
396
397
        """
        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()
398

399
        with utils.lnr(logger, 'publish failed'):
400
401
            upload_with_metadata = self.to_upload_with_metadata()

402
            with utils.timer(
403
                    logger, 'upload added to repository', step='publish',
404
                    upload_size=self.upload_files.size):
405
                coe_repo.Upload.add(upload_with_metadata)
406
407

            with utils.timer(
408
                    logger, 'staged upload files packed', step='publish',
409
                    upload_size=self.upload_files.size):
410
411
412
413
                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())
414
415
                self.upload_files.pack()

416
417
418
            with utils.timer(
                    logger, 'index updated', step='publish',
                    upload_size=self.upload_files.size):
419
420
                coe_upload = coe_repo.Upload.from_upload_id(upload_with_metadata.upload_id)
                if coe_upload is not None:
421
422
                    search.publish(
                        [coe_calc.to_calc_with_metadata() for coe_calc in coe_upload.calcs])
423

424
            with utils.timer(
425
                    logger, 'staged upload deleted', step='publish',
426
427
428
                    upload_size=self.upload_files.size):
                self.upload_files.delete()
                self.delete()
Markus Scheidgen's avatar
Markus Scheidgen committed
429

430
        return True  # do not save the process status on the delete upload
431

Markus Scheidgen's avatar
Markus Scheidgen committed
432
    @process
433
    def process_upload(self):
Markus Scheidgen's avatar
Markus Scheidgen committed
434
435
436
437
438
439
440
        self.extracting()
        self.parse_all()

    @task
    def uploading(self):
        pass

441
    @property
442
443
    def upload_files(self) -> ArchiveBasedStagingUploadFiles:
        if not self._upload_files:
444
            self._upload_files = ArchiveBasedStagingUploadFiles(self.upload_id, is_authorized=lambda: True, local_path=self.local_path)
445
        return self._upload_files
446

Markus Scheidgen's avatar
Markus Scheidgen committed
447
448
    @task
    def extracting(self):
449
450
451
452
453
454
        """
        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
455
456
        logger = self.get_logger()
        try:
Markus Scheidgen's avatar
Markus Scheidgen committed
457
458
            with utils.timer(
                    logger, 'upload extracted', step='extracting',
459
460
                    upload_size=self.upload_files.size):
                self.upload_files.extract()
461
        except KeyError:
462
            self.fail('processing requested for non existing upload', log_level=logging.ERROR)
463
464
465
            return
        except ExtractError:
            self.fail('bad .zip/.tar file', log_level=logging.INFO)
Markus Scheidgen's avatar
Markus Scheidgen committed
466
467
            return

468
    def match_mainfiles(self) -> Generator[Tuple[str, object], None, None]:
469
470
471
472
473
474
475
        """
        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
        """
476
        directories_with_match: Dict[str, str] = dict()
477
        for filename in self.upload_files.raw_file_manifest():
478
            try:
479
                parser = match_parser(filename, self.upload_files)
480
                if parser is not None:
481
                    directory = os.path.dirname(filename)
482
483
                    # TODO this might give us the chance to store directory based relationship
                    # between calcs for the future?
484
                    if directory in directories_with_match:
485
                        self.info.append(
486
487
488
489
                            'The directory %s contains data from multiple code runs. '
                            'Nomad only processed %s.' % (directory, os.path.basename(filename)))
                    else:
                        directories_with_match[directory] = filename
490
491

                    yield filename, parser
492
493
494
495
            except Exception as e:
                self.get_logger().error(
                    'exception while matching pot. mainfile',
                    mainfile=filename, exc_info=e)
496

Markus Scheidgen's avatar
Markus Scheidgen committed
497
498
    @task
    def parse_all(self):
499
        """
500
        Identified mainfile/parser combinations among the upload's files, creates
501
502
        respective :class:`Calc` instances, and triggers their processing.
        """
503
504
        logger = self.get_logger()

Markus Scheidgen's avatar
Markus Scheidgen committed
505
506
        with utils.timer(
                logger, 'upload extracted', step='matching',
507
508
                upload_size=self.upload_files.size):
            for filename, parser in self.match_mainfiles():
509
                calc = Calc.create(
510
                    calc_id=self.upload_files.calc_id(filename),
511
512
513
                    mainfile=filename, parser=parser.name,
                    upload_id=self.upload_id)

514
                calc.process_calc()
Markus Scheidgen's avatar
Markus Scheidgen committed
515

516
517
518
519
520
521
522
523
524
525
526
527
    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()
528
529
530

    def join(self):
        self.cleanup()
Markus Scheidgen's avatar
Markus Scheidgen committed
531
532
533

    @task
    def cleanup(self):
534
535
536
537
538
539
540
541
542
543
        # 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
        ])
544
545
546
547
548
549
550
        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
551
552

    @property
553
    def processed_calcs(self):
554
        return Calc.objects(upload_id=self.upload_id, tasks_status__in=[SUCCESS, FAILURE]).count()
555
556
557
558
559
560
561

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

    @property
    def failed_calcs(self):
562
        return Calc.objects(upload_id=self.upload_id, tasks_status=FAILURE).count()
563

564
565
    @property
    def pending_calcs(self):
566
        return Calc.objects(upload_id=self.upload_id, tasks_status=PENDING).count()
567

568
569
    def all_calcs(self, start, end, order_by='mainfile'):
        return Calc.objects(upload_id=self.upload_id)[start:end].order_by(order_by)
570
571
572

    @property
    def calcs(self):
Markus Scheidgen's avatar
Markus Scheidgen committed
573
        return Calc.objects(upload_id=self.upload_id, tasks_status=SUCCESS)
Markus Scheidgen's avatar
Markus Scheidgen committed
574
575
576
577
578
579
580
581
582
583
584

    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)
585
586
            if metadata is not None:
                calc.apply_user_metadata(metadata)
Markus Scheidgen's avatar
Markus Scheidgen committed
587
588
589
590
591
592
593
594
595
596
597
            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