data.py 24.1 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
    upload_id = StringField()
    mainfile = StringField()
    parser = StringField()

63
64
    queue = 'calcs'

Markus Scheidgen's avatar
Markus Scheidgen committed
65
    meta: Any = {
Markus Scheidgen's avatar
Markus Scheidgen committed
66
67
        'indexes': [
            'upload_id', 'mainfile', 'parser', 'tasks_status'
Markus Scheidgen's avatar
Markus Scheidgen committed
68
69
70
71
72
73
        ]
    }

    def __init__(self, *args, **kwargs):
        super().__init__(*args, **kwargs)
        self._parser_backend = None
74
75
        self._upload: Upload = None
        self._upload_files: ArchiveBasedStagingUploadFiles = None
76
        self._calc_proc_logwriter = None
77
        self._calc_proc_logwriter_ctx: ContextManager = None
Markus Scheidgen's avatar
Markus Scheidgen committed
78
79
80

    @classmethod
    def get(cls, id):
81
        return cls.get_by_id(id, 'calc_id')
Markus Scheidgen's avatar
Markus Scheidgen committed
82

Markus Scheidgen's avatar
Markus Scheidgen committed
83
    @property
84
85
    def mainfile_file(self) -> PathObject:
        return self.upload_files.raw_file_object(self.mainfile)
Markus Scheidgen's avatar
Markus Scheidgen committed
86

87
88
89
90
91
92
    @property
    def upload(self) -> 'Upload':
        if not self._upload:
            self._upload = Upload.get(self.upload_id)
        return self._upload

93
94
95
    @property
    def upload_files(self) -> ArchiveBasedStagingUploadFiles:
        if not self._upload_files:
96
            self._upload_files = ArchiveBasedStagingUploadFiles(self.upload_id, is_authorized=lambda: True, local_path=self.upload.local_path)
97
98
        return self._upload_files

Markus Scheidgen's avatar
Markus Scheidgen committed
99
    def get_logger(self, **kwargs):
100
101
102
103
        """
        Returns a wrapped logger that additionally saves all entries to the calculation
        processing log in the archive.
        """
104
105
106
        logger = super().get_logger()
        logger = logger.bind(
            upload_id=self.upload_id, mainfile=self.mainfile, calc_id=self.calc_id, **kwargs)
107

108
        if self._calc_proc_logwriter_ctx is None:
109
            self._calc_proc_logwriter_ctx = self.upload_files.archive_log_file(self.calc_id, 'wt')
110
            self._calc_proc_logwriter = self._calc_proc_logwriter_ctx.__enter__()  # pylint: disable=E1101
111

112
        def save_to_calc_log(logger, method_name, event_dict):
113
114
115
116
117
118
119
120
121
122
123
            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')

124
125
            return event_dict

126
        return wrap_logger(logger, processors=[save_to_calc_log])
127

Markus Scheidgen's avatar
Markus Scheidgen committed
128
    @process
129
    def process_calc(self):
130
        logger = self.get_logger()
131
        if self.upload is None:
132
            logger.error('calculation upload does not exist')
Markus Scheidgen's avatar
Markus Scheidgen committed
133
134
135
136
137
138

        try:
            self.parsing()
            self.normalizing()
            self.archiving()
        finally:
Markus Scheidgen's avatar
Markus Scheidgen committed
139
            # close loghandler that was not closed due to failures
140
            try:
141
142
143
                if self._calc_proc_logwriter is not None:
                    self._calc_proc_logwriter.close()
                    self._calc_proc_logwriter = None
144
145
146
            except Exception as e:
                logger.error('could not close calculation proc log', exc_info=e)

147
148
149
150
151
152
153
    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
154
155
156

    @task
    def parsing(self):
157
        context = dict(parser=self.parser, step=self.parser)
158
        logger = self.get_logger(**context)
159
        parser = parser_dict[self.parser]
Markus Scheidgen's avatar
Markus Scheidgen committed
160

161
        with utils.timer(logger, 'parser executed', input_size=self.mainfile_file.size):
162
163
164
165
166
167
168
            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)
169
                return
Markus Scheidgen's avatar
Markus Scheidgen committed
170

171
172
        self._parser_backend.openNonOverlappingSection('section_calculation_info')
        self._parser_backend.addValue('upload_id', self.upload_id)
173
174
        self._parser_backend.addValue('calc_id', self.calc_id)
        self._parser_backend.addValue('calc_hash', self.upload_files.calc_hash(self.mainfile))
175
176
177
        self._parser_backend.addValue('main_file', self.mainfile)
        self._parser_backend.addValue('parser_name', self.parser)

Markus Scheidgen's avatar
Markus Scheidgen committed
178
        if self._parser_backend.status[0] != 'ParseSuccess':
179
            logger.error(self._parser_backend.status[1])
Markus Scheidgen's avatar
Markus Scheidgen committed
180
            error = self._parser_backend.status[1]
181
            self._parser_backend.addValue('parse_status', 'ParseFailure')
182
            self.fail(error, level=logging.INFO, **context)
183
184
185
186
187
        else:
            self._parser_backend.addValue('parse_status', 'ParseSuccess')

        self._parser_backend.closeNonOverlappingSection('section_calculation_info')

188
189
190
191
192
193
        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')

194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
        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:
216
            errors = self._parser_backend.tasks_status[1]
217
218
219
220
            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
221
222
223
224
225

    @task
    def normalizing(self):
        for normalizer in normalizers:
            normalizer_name = normalizer.__name__
226
            context = dict(normalizer=normalizer_name, step=normalizer_name)
227
            logger = self.get_logger(**context)
Markus Scheidgen's avatar
Markus Scheidgen committed
228
229

            with utils.timer(
230
                    logger, 'normalizer executed', input_size=self.mainfile_file.size):
231
                with self.use_parser_backend(normalizer_name) as backend:
232
233
234
235
236
237
238
                    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
239

240
241
            failed = self._parser_backend.status[0] != 'ParseSuccess'
            if failed:
242
                logger.error(self._parser_backend.status[1])
Markus Scheidgen's avatar
Markus Scheidgen committed
243
                error = self._parser_backend.status[1]
244
                self.fail(error, level=logging.WARNING, error=error, **context)
245
246
247
248
                break
            else:
                logger.debug(
                    'completed normalizer successfully', normalizer=normalizer_name)
Markus Scheidgen's avatar
Markus Scheidgen committed
249
250
251

    @task
    def archiving(self):
252
253
        logger = self.get_logger()

254
255
        calc_with_metadata = self._parser_backend.to_calc_with_metadata()

256
        # persist the repository metadata
257
258
259
260
261
        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'):
262
263
            calc_with_metadata.update(published=False, uploader=self.upload.uploader.to_popo())
            search.Entry.from_calc_with_metadata(calc_with_metadata).save()
264

265
        # persist the archive
Markus Scheidgen's avatar
Markus Scheidgen committed
266
        with utils.timer(
267
                logger, 'archived', step='persist',
Markus Scheidgen's avatar
Markus Scheidgen committed
268
                input_size=self.mainfile_file.size) as log_data:
269
            with self.upload_files.archive_file(self.calc_id, 'wt') as out:
270
271
                self._parser_backend.write_json(out, pretty=True)

272
            log_data.update(archive_size=self.upload_files.archive_file_object(self.calc_id).size)
Markus Scheidgen's avatar
Markus Scheidgen committed
273
274
275
276

        # close loghandler
        if self._calc_proc_logwriter is not None:
            with utils.timer(
277
                    logger, 'archived log', step='persist',
Markus Scheidgen's avatar
Markus Scheidgen committed
278
                    input_size=self.mainfile_file.size) as log_data:
279
                self._calc_proc_logwriter_ctx.__exit__(None, None, None)  # pylint: disable=E1101
280
                self._calc_proc_logwriter = None
Markus Scheidgen's avatar
Markus Scheidgen committed
281

282
                log_data.update(log_size=self.upload_files.archive_log_file_object(self.calc_id).size)
Markus Scheidgen's avatar
Markus Scheidgen committed
283

284

285
class Upload(Proc):
Markus Scheidgen's avatar
Markus Scheidgen committed
286
287
288
289
290
291
    """
    Represents uploads in the databases. Provides persistence access to the files storage,
    and processing state.

    Attributes:
        name: optional user provided upload name
292
        local_path: optional local path, e.g. for files that are already somewhere on the server
293
        metadata: optional user provided additional meta data
Markus Scheidgen's avatar
Markus Scheidgen committed
294
295
        upload_id: the upload id generated by the database
        upload_time: the timestamp when the system realised the upload
296
        user_id: the id of the user that created this upload
Markus Scheidgen's avatar
Markus Scheidgen committed
297
298
299
300
301
302
    """
    id_field = 'upload_id'

    upload_id = StringField(primary_key=True)

    name = StringField(default=None)
303
    local_path = StringField(default=None)
304
    metadata = DictField(default=None)
Markus Scheidgen's avatar
Markus Scheidgen committed
305
    upload_time = DateTimeField()
306
    user_id = StringField(required=True)
Markus Scheidgen's avatar
Markus Scheidgen committed
307

308
309
    queue = 'uploads'

Markus Scheidgen's avatar
Markus Scheidgen committed
310
311
    meta: Any = {
        'indexes': [
312
            'user_id', 'tasks_status'
Markus Scheidgen's avatar
Markus Scheidgen committed
313
314
315
316
317
        ]
    }

    def __init__(self, **kwargs):
        super().__init__(**kwargs)
318
        self._upload_files: ArchiveBasedStagingUploadFiles = None
Markus Scheidgen's avatar
Markus Scheidgen committed
319
320
321
322
323
324

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

    @classmethod
325
    def user_uploads(cls, user: coe_repo.User) -> List['Upload']:
Markus Scheidgen's avatar
Markus Scheidgen committed
326
        """ Returns all uploads for the given user. Currently returns all uploads. """
327
        return cls.objects(user_id=str(user.user_id))
Markus Scheidgen's avatar
Markus Scheidgen committed
328

329
330
331
332
    @property
    def uploader(self):
        return coe_repo.User.from_user_id(self.user_id)

Markus Scheidgen's avatar
Markus Scheidgen committed
333
334
    def get_logger(self, **kwargs):
        logger = super().get_logger()
335
        logger = logger.bind(upload_id=self.upload_id, upload_name=self.name, **kwargs)
Markus Scheidgen's avatar
Markus Scheidgen committed
336
337
338
339
340
341
342
343
        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.
344
345

        Arguments:
346
            user (coe_repo.User): The user that created the upload.
Markus Scheidgen's avatar
Markus Scheidgen committed
347
        """
348
        user: coe_repo.User = kwargs['user']
349
350
351
        del(kwargs['user'])
        if 'upload_id' not in kwargs:
            kwargs.update(upload_id=utils.create_uuid())
352
        kwargs.update(user_id=str(user.user_id))
Markus Scheidgen's avatar
Markus Scheidgen committed
353
        self = super().create(**kwargs)
354

Markus Scheidgen's avatar
Markus Scheidgen committed
355
        self._continue_with('uploading')
356

Markus Scheidgen's avatar
Markus Scheidgen committed
357
358
        return self

359
360
361
362
363
364
365
366
367
368
369
370
371
372
    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'):
373
374
375
            with utils.timer(
                    logger, 'upload deleted from index', step='delete',
                    upload_size=self.upload_files.size):
376
                search.delete_upload(self.upload_id)
377

378
379
380
381
382
383
384
            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
385

386
    @process
387
    def publish_upload(self):
388
389
390
391
392
393
394
        """
        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()
Markus Scheidgen's avatar
Markus Scheidgen committed
395
        logger.info('started to publish', step='publish')
396

397
        with utils.lnr(logger, 'publish failed'):
398
399
            upload_with_metadata = self.to_upload_with_metadata()

400
            with utils.timer(
401
                    logger, 'upload added to repository', step='publish',
402
                    upload_size=self.upload_files.size):
403
                coe_repo.Upload.add(upload_with_metadata)
404
405

            with utils.timer(
406
                    logger, 'staged upload files packed', step='publish',
407
                    upload_size=self.upload_files.size):
408
409
410
411
412
413
414
415
416
                coe_upload = coe_repo.Upload.from_upload_id(upload_with_metadata.upload_id)
                if coe_upload is not None:
                    for coe_calc in coe_upload.calcs:
                        calc_metadata = coe_calc.to_calc_with_metadata()
                        calc_metadata.published = True
                        self.upload_files.metadata.update(
                            calc_id=calc_metadata.calc_id, updates=calc_metadata.to_dict())
                    logger.info('metadata updated after publish to coe repo', step='publish')

417
418
                self.upload_files.pack()

419
420
421
            with utils.timer(
                    logger, 'index updated', step='publish',
                    upload_size=self.upload_files.size):
422
423
                coe_upload = coe_repo.Upload.from_upload_id(upload_with_metadata.upload_id)
                if coe_upload is not None:
424
425
                    search.publish(
                        [coe_calc.to_calc_with_metadata() for coe_calc in coe_upload.calcs])
426

427
            with utils.timer(
428
                    logger, 'staged upload deleted', step='publish',
429
430
431
                    upload_size=self.upload_files.size):
                self.upload_files.delete()
                self.delete()
Markus Scheidgen's avatar
Markus Scheidgen committed
432

433
        return True  # do not save the process status on the delete upload
434

Markus Scheidgen's avatar
Markus Scheidgen committed
435
    @process
436
    def process_upload(self):
Markus Scheidgen's avatar
Markus Scheidgen committed
437
438
439
440
441
442
443
        self.extracting()
        self.parse_all()

    @task
    def uploading(self):
        pass

444
    @property
445
446
    def upload_files(self) -> ArchiveBasedStagingUploadFiles:
        if not self._upload_files:
447
            self._upload_files = ArchiveBasedStagingUploadFiles(self.upload_id, is_authorized=lambda: True, local_path=self.local_path)
448
        return self._upload_files
449

Markus Scheidgen's avatar
Markus Scheidgen committed
450
451
    @task
    def extracting(self):
452
453
454
455
456
457
        """
        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
458
459
        logger = self.get_logger()
        try:
Markus Scheidgen's avatar
Markus Scheidgen committed
460
461
            with utils.timer(
                    logger, 'upload extracted', step='extracting',
462
463
                    upload_size=self.upload_files.size):
                self.upload_files.extract()
464
        except KeyError:
465
            self.fail('processing requested for non existing upload', log_level=logging.ERROR)
466
467
468
            return
        except ExtractError:
            self.fail('bad .zip/.tar file', log_level=logging.INFO)
Markus Scheidgen's avatar
Markus Scheidgen committed
469
470
            return

471
    def match_mainfiles(self) -> Generator[Tuple[str, object], None, None]:
472
473
474
475
476
477
478
        """
        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
        """
479
        directories_with_match: Dict[str, str] = dict()
480
        for filename in self.upload_files.raw_file_manifest():
481
            try:
482
                parser = match_parser(filename, self.upload_files)
483
                if parser is not None:
484
485
                    directory = os.path.dirname(filename)
                    if directory in directories_with_match:
486
487
488
                        # TODO this might give us the chance to store directory based relationship
                        # between calcs for the future?
                        pass
489
490
                    else:
                        directories_with_match[directory] = filename
491
492

                    yield filename, parser
493
494
495
496
            except Exception as e:
                self.get_logger().error(
                    'exception while matching pot. mainfile',
                    mainfile=filename, exc_info=e)
497

Markus Scheidgen's avatar
Markus Scheidgen committed
498
499
    @task
    def parse_all(self):
500
        """
501
        Identified mainfile/parser combinations among the upload's files, creates
502
503
        respective :class:`Calc` instances, and triggers their processing.
        """
504
505
        logger = self.get_logger()

Markus Scheidgen's avatar
Markus Scheidgen committed
506
507
        with utils.timer(
                logger, 'upload extracted', step='matching',
508
509
                upload_size=self.upload_files.size):
            for filename, parser in self.match_mainfiles():
510
                calc = Calc.create(
511
                    calc_id=self.upload_files.calc_id(filename),
512
513
514
                    mainfile=filename, parser=parser.name,
                    upload_id=self.upload_id)

515
                calc.process_calc()
Markus Scheidgen's avatar
Markus Scheidgen committed
516

517
518
519
520
521
522
523
524
525
526
527
528
    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()
529
530
531

    def join(self):
        self.cleanup()
Markus Scheidgen's avatar
Markus Scheidgen committed
532
533
534

    @task
    def cleanup(self):
535
536
537
538
539
540
541
542
543
544
        # 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
        ])
545
546
547
548
549
550
551
        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
552
553

    @property
554
    def processed_calcs(self):
555
        return Calc.objects(upload_id=self.upload_id, tasks_status__in=[SUCCESS, FAILURE]).count()
556
557
558
559
560
561
562

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

    @property
    def failed_calcs(self):
563
        return Calc.objects(upload_id=self.upload_id, tasks_status=FAILURE).count()
564

565
566
    @property
    def pending_calcs(self):
567
        return Calc.objects(upload_id=self.upload_id, tasks_status=PENDING).count()
568

569
570
    def all_calcs(self, start, end, order_by='mainfile'):
        return Calc.objects(upload_id=self.upload_id)[start:end].order_by(order_by)
571
572
573

    @property
    def calcs(self):
Markus Scheidgen's avatar
Markus Scheidgen committed
574
        return Calc.objects(upload_id=self.upload_id, tasks_status=SUCCESS)
Markus Scheidgen's avatar
Markus Scheidgen committed
575
576

    def to_upload_with_metadata(self) -> UploadWithMetadata:
577
578
579
580
581
582
583
584
585
        # TODO remove the very verbose metadata after debugging and optimizing

        logger = self.get_logger(step='publish')

        # prepare user metadata per upload and per calc
        logger.info('prepare user metadata per upload and per calc')
        calc_metadatas: Dict[str, Any] = dict()
        upload_metadata: Dict[str, Any] = dict()

Markus Scheidgen's avatar
Markus Scheidgen committed
586
        if self.metadata is not None:
587
588
589
            upload_metadata.update(self.metadata)
            if 'calculations' in upload_metadata:
                del(upload_metadata['calculations'])
Markus Scheidgen's avatar
Markus Scheidgen committed
590

591
592
            for calc in self.metadata.get('calculations', []):
                calc_metadatas[calc['mainfile']] = calc
Markus Scheidgen's avatar
Markus Scheidgen committed
593

594
        user_upload_time = upload_metadata.get('_upload_time', None)
Markus Scheidgen's avatar
Markus Scheidgen committed
595
596
597
598
599
        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)

600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
        logger.info('read calc data from files and apply user metadata')
        upload_files = UploadFiles.get(self.upload_id, is_authorized=lambda: True)
        if self.upload_files is None:
            raise KeyError

        def apply_metadata(calc):
            calc_data = upload_files.metadata.get(calc.calc_id)
            calc_with_metadata = CalcWithMetadata(**calc_data)

            calc_metadata = dict(calc_metadatas.get(calc.mainfile, {}))
            calc_metadata.update(upload_metadata)
            calc_with_metadata.apply_user_metadata(calc_metadata)

            logger.info('prepared calc with metadata', calc_id=calc_with_metadata.calc_id)
            return calc_with_metadata

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

        logger.info('prepared user metadata')
Markus Scheidgen's avatar
Markus Scheidgen committed
619
620

        return result