data.py 24.7 KB
Newer Older
Markus Scheidgen's avatar
Markus Scheidgen committed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
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, BooleanField
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
    @property
    def upload(self) -> 'Upload':
        if not self._upload:
            self._upload = Upload.get(self.upload_id)
91
            self._upload.worker_hostname = self.worker_hostname
92
93
        return self._upload

94
95
96
    @property
    def upload_files(self) -> ArchiveBasedStagingUploadFiles:
        if not self._upload_files:
97
98
            self._upload_files = ArchiveBasedStagingUploadFiles(
                self.upload_id, is_authorized=lambda: True, upload_path=self.upload.upload_path)
99
100
        return self._upload_files

Markus Scheidgen's avatar
Markus Scheidgen committed
101
    def get_logger(self, **kwargs):
102
103
104
105
        """
        Returns a wrapped logger that additionally saves all entries to the calculation
        processing log in the archive.
        """
106
107
108
        logger = super().get_logger()
        logger = logger.bind(
            upload_id=self.upload_id, mainfile=self.mainfile, calc_id=self.calc_id, **kwargs)
109

110
        if self._calc_proc_logwriter_ctx is None:
111
            self._calc_proc_logwriter_ctx = self.upload_files.archive_log_file(self.calc_id, 'wt')
112
            self._calc_proc_logwriter = self._calc_proc_logwriter_ctx.__enter__()  # pylint: disable=E1101
113

114
        def save_to_calc_log(logger, method_name, event_dict):
115
116
117
118
119
120
121
122
123
124
125
            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')

126
127
            return event_dict

128
        return wrap_logger(logger, processors=[save_to_calc_log])
129

Markus Scheidgen's avatar
Markus Scheidgen committed
130
    @process
131
    def process_calc(self):
132
        logger = self.get_logger()
133
        if self.upload is None:
134
            logger.error('calculation upload does not exist')
Markus Scheidgen's avatar
Markus Scheidgen committed
135
136
137
138
139
140

        try:
            self.parsing()
            self.normalizing()
            self.archiving()
        finally:
Markus Scheidgen's avatar
Markus Scheidgen committed
141
            # close loghandler that was not closed due to failures
142
            try:
143
144
145
                if self._calc_proc_logwriter is not None:
                    self._calc_proc_logwriter.close()
                    self._calc_proc_logwriter = None
146
147
148
            except Exception as e:
                logger.error('could not close calculation proc log', exc_info=e)

149
150
151
152
153
154
155
    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
156
157
158

    @task
    def parsing(self):
159
        context = dict(parser=self.parser, step=self.parser)
160
        logger = self.get_logger(**context)
161
        parser = parser_dict[self.parser]
Markus Scheidgen's avatar
Markus Scheidgen committed
162

163
        with utils.timer(logger, 'parser executed', input_size=self.mainfile_file.size):
164
165
166
167
168
169
170
            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)
171
                return
Markus Scheidgen's avatar
Markus Scheidgen committed
172

173
174
        self._parser_backend.openNonOverlappingSection('section_calculation_info')
        self._parser_backend.addValue('upload_id', self.upload_id)
175
176
        self._parser_backend.addValue('calc_id', self.calc_id)
        self._parser_backend.addValue('calc_hash', self.upload_files.calc_hash(self.mainfile))
177
178
179
        self._parser_backend.addValue('main_file', self.mainfile)
        self._parser_backend.addValue('parser_name', self.parser)

Markus Scheidgen's avatar
Markus Scheidgen committed
180
        if self._parser_backend.status[0] != 'ParseSuccess':
181
            logger.error(self._parser_backend.status[1])
Markus Scheidgen's avatar
Markus Scheidgen committed
182
            error = self._parser_backend.status[1]
183
            self._parser_backend.addValue('parse_status', 'ParseFailure')
184
            self.fail(error, level=logging.INFO, **context)
185
186
187
188
189
        else:
            self._parser_backend.addValue('parse_status', 'ParseSuccess')

        self._parser_backend.closeNonOverlappingSection('section_calculation_info')

190
191
192
193
194
195
        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')

196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
        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:
218
            errors = self._parser_backend.status[1]
219
220
221
222
            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
223
224
225
226
227

    @task
    def normalizing(self):
        for normalizer in normalizers:
            normalizer_name = normalizer.__name__
228
            context = dict(normalizer=normalizer_name, step=normalizer_name)
229
            logger = self.get_logger(**context)
Markus Scheidgen's avatar
Markus Scheidgen committed
230
231

            with utils.timer(
232
                    logger, 'normalizer executed', input_size=self.mainfile_file.size):
233
                with self.use_parser_backend(normalizer_name) as backend:
234
235
236
237
238
239
240
                    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
241

242
243
            failed = self._parser_backend.status[0] != 'ParseSuccess'
            if failed:
244
                logger.error(self._parser_backend.status[1])
Markus Scheidgen's avatar
Markus Scheidgen committed
245
                error = self._parser_backend.status[1]
246
                self.fail(error, level=logging.WARNING, error=error, **context)
247
248
249
250
                break
            else:
                logger.debug(
                    'completed normalizer successfully', normalizer=normalizer_name)
Markus Scheidgen's avatar
Markus Scheidgen committed
251
252
253

    @task
    def archiving(self):
254
255
        logger = self.get_logger()

256
257
        calc_with_metadata = self._parser_backend.to_calc_with_metadata()

258
        # persist the repository metadata
259
        with utils.timer(logger, 'saved repo metadata', step='metadata'):
260
261
262
            self.upload_files.metadata.insert(calc_with_metadata.to_dict())

        # index in search
263
        with utils.timer(logger, 'indexed', step='index'):
264
265
            calc_with_metadata.update(published=False, uploader=self.upload.uploader.to_popo())
            search.Entry.from_calc_with_metadata(calc_with_metadata).save()
266

267
        # persist the archive
Markus Scheidgen's avatar
Markus Scheidgen committed
268
        with utils.timer(
269
                logger, 'archived', step='archive',
Markus Scheidgen's avatar
Markus Scheidgen committed
270
                input_size=self.mainfile_file.size) as log_data:
271
            with self.upload_files.archive_file(self.calc_id, 'wt') as out:
272
273
                self._parser_backend.write_json(out, pretty=True)

274
            log_data.update(archive_size=self.upload_files.archive_file_object(self.calc_id).size)
Markus Scheidgen's avatar
Markus Scheidgen committed
275
276
277
278

        # close loghandler
        if self._calc_proc_logwriter is not None:
            with utils.timer(
279
                    logger, 'archived log', step='logs',
Markus Scheidgen's avatar
Markus Scheidgen committed
280
                    input_size=self.mainfile_file.size) as log_data:
281
                self._calc_proc_logwriter_ctx.__exit__(None, None, None)  # pylint: disable=E1101
282
                self._calc_proc_logwriter = None
Markus Scheidgen's avatar
Markus Scheidgen committed
283

284
                log_data.update(log_size=self.upload_files.archive_log_file_object(self.calc_id).size)
Markus Scheidgen's avatar
Markus Scheidgen committed
285

286

287
class Upload(Proc):
Markus Scheidgen's avatar
Markus Scheidgen committed
288
289
290
291
292
293
    """
    Represents uploads in the databases. Provides persistence access to the files storage,
    and processing state.

    Attributes:
        name: optional user provided upload name
294
295
        upload_path: the path were the uploaded files was stored
        temporary: True if the uploaded file should be removed after extraction
296
        metadata: optional user provided additional meta data
Markus Scheidgen's avatar
Markus Scheidgen committed
297
298
        upload_id: the upload id generated by the database
        upload_time: the timestamp when the system realised the upload
299
        user_id: the id of the user that created this upload
Markus Scheidgen's avatar
Markus Scheidgen committed
300
301
302
303
    """
    id_field = 'upload_id'

    upload_id = StringField(primary_key=True)
304
305
    upload_path = StringField(default=None)
    temporary = BooleanField(default=False)
Markus Scheidgen's avatar
Markus Scheidgen committed
306
307

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

312
313
    queue = 'uploads'

Markus Scheidgen's avatar
Markus Scheidgen committed
314
315
    meta: Any = {
        'indexes': [
316
            'user_id', 'tasks_status'
Markus Scheidgen's avatar
Markus Scheidgen committed
317
318
319
320
321
        ]
    }

    def __init__(self, **kwargs):
        super().__init__(**kwargs)
322
        self._upload_files: ArchiveBasedStagingUploadFiles = None
Markus Scheidgen's avatar
Markus Scheidgen committed
323
324
325
326
327
328

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

    @classmethod
329
    def user_uploads(cls, user: coe_repo.User) -> List['Upload']:
Markus Scheidgen's avatar
Markus Scheidgen committed
330
        """ Returns all uploads for the given user. Currently returns all uploads. """
331
        return cls.objects(user_id=str(user.user_id))
Markus Scheidgen's avatar
Markus Scheidgen committed
332

333
334
335
336
    @property
    def uploader(self):
        return coe_repo.User.from_user_id(self.user_id)

Markus Scheidgen's avatar
Markus Scheidgen committed
337
338
    def get_logger(self, **kwargs):
        logger = super().get_logger()
339
        logger = logger.bind(upload_id=self.upload_id, upload_name=self.name, **kwargs)
Markus Scheidgen's avatar
Markus Scheidgen committed
340
341
342
343
344
345
346
347
        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.
348
349

        Arguments:
350
            user (coe_repo.User): The user that created the upload.
Markus Scheidgen's avatar
Markus Scheidgen committed
351
        """
352
        user: coe_repo.User = kwargs['user']
353
354
355
        del(kwargs['user'])
        if 'upload_id' not in kwargs:
            kwargs.update(upload_id=utils.create_uuid())
356
        kwargs.update(user_id=str(user.user_id))
Markus Scheidgen's avatar
Markus Scheidgen committed
357
        self = super().create(**kwargs)
358

Markus Scheidgen's avatar
Markus Scheidgen committed
359
        self._continue_with('uploading')
360

Markus Scheidgen's avatar
Markus Scheidgen committed
361
362
        return self

363
364
365
366
367
368
369
370
371
372
373
374
375
376
    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'):
377
            with utils.timer(
378
                    logger, 'upload deleted from index', step='index',
379
                    upload_size=self.upload_files.size):
380
                search.delete_upload(self.upload_id)
381

382
            with utils.timer(
383
                    logger, 'staged upload deleted', step='files',
384
385
386
387
388
                    upload_size=self.upload_files.size):
                self.upload_files.delete()
                self.delete()

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

390
    @process
391
    def publish_upload(self):
392
393
394
395
396
397
398
        """
        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
399
        logger.info('started to publish', step='publish')
400

401
        with utils.lnr(logger, 'publish failed'):
402
403
            upload_with_metadata = self.to_upload_with_metadata()

404
            with utils.timer(
405
                    logger, 'upload added to repository', step='repo',
406
                    upload_size=self.upload_files.size):
407
                coe_repo.Upload.publish(upload_with_metadata)
408

409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
            with utils.timer(
                    logger, 'staged upload files packed', step='pack',
                    upload_size=self.upload_files.size):
                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')

                self.upload_files.pack()

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

            with utils.timer(
                    logger, 'staged upload deleted', step='delete staged',
                    upload_size=self.upload_files.size):
                self.upload_files.delete()
                self.delete()
Markus Scheidgen's avatar
Markus Scheidgen committed
436

437
        return True  # do not save the process status on the delete upload
438

Markus Scheidgen's avatar
Markus Scheidgen committed
439
    @process
440
    def process_upload(self):
Markus Scheidgen's avatar
Markus Scheidgen committed
441
442
443
444
445
446
447
        self.extracting()
        self.parse_all()

    @task
    def uploading(self):
        pass

448
    @property
449
450
    def upload_files(self) -> ArchiveBasedStagingUploadFiles:
        if not self._upload_files:
451
452
            self._upload_files = ArchiveBasedStagingUploadFiles(
                self.upload_id, is_authorized=lambda: True, upload_path=self.upload_path)
453
        return self._upload_files
454

Markus Scheidgen's avatar
Markus Scheidgen committed
455
456
    @task
    def extracting(self):
457
458
459
460
461
        """
        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.
        """
462
463
464
465
466
        # extract the uploaded file
        self._upload_files = ArchiveBasedStagingUploadFiles(
            upload_id=self.upload_id, is_authorized=lambda: True, create=True,
            upload_path=self.upload_path)

Markus Scheidgen's avatar
Markus Scheidgen committed
467
468
        logger = self.get_logger()
        try:
Markus Scheidgen's avatar
Markus Scheidgen committed
469
470
            with utils.timer(
                    logger, 'upload extracted', step='extracting',
471
472
                    upload_size=self.upload_files.size):
                self.upload_files.extract()
473
474
475
476

            if self.temporary:
                os.remove(self.upload_path)
                self.upload_path = None
477
        except KeyError:
478
            self.fail('processing requested for non existing upload', log_level=logging.ERROR)
479
480
481
            return
        except ExtractError:
            self.fail('bad .zip/.tar file', log_level=logging.INFO)
Markus Scheidgen's avatar
Markus Scheidgen committed
482
483
            return

484
    def match_mainfiles(self) -> Generator[Tuple[str, object], None, None]:
485
486
487
488
489
490
491
        """
        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
        """
492
        directories_with_match: Dict[str, str] = dict()
493
        for filename in self.upload_files.raw_file_manifest():
494
            try:
495
                parser = match_parser(filename, self.upload_files)
496
                if parser is not None:
497
498
                    directory = os.path.dirname(filename)
                    if directory in directories_with_match:
499
500
501
                        # TODO this might give us the chance to store directory based relationship
                        # between calcs for the future?
                        pass
502
503
                    else:
                        directories_with_match[directory] = filename
504
505

                    yield filename, parser
506
507
508
509
            except Exception as e:
                self.get_logger().error(
                    'exception while matching pot. mainfile',
                    mainfile=filename, exc_info=e)
510

Markus Scheidgen's avatar
Markus Scheidgen committed
511
512
    @task
    def parse_all(self):
513
        """
514
        Identified mainfile/parser combinations among the upload's files, creates
515
516
        respective :class:`Calc` instances, and triggers their processing.
        """
517
518
        logger = self.get_logger()

Markus Scheidgen's avatar
Markus Scheidgen committed
519
520
        with utils.timer(
                logger, 'upload extracted', step='matching',
521
522
                upload_size=self.upload_files.size):
            for filename, parser in self.match_mainfiles():
523
                calc = Calc.create(
524
                    calc_id=self.upload_files.calc_id(filename),
525
                    mainfile=filename, parser=parser.name,
526
                    worker_hostname=self.worker_hostname,
527
528
                    upload_id=self.upload_id)

529
                calc.process_calc()
Markus Scheidgen's avatar
Markus Scheidgen committed
530

531
532
533
534
535
536
537
538
539
540
541
542
    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()
543
544
545

    def join(self):
        self.cleanup()
Markus Scheidgen's avatar
Markus Scheidgen committed
546
547
548

    @task
    def cleanup(self):
549
550
        search.refresh()

551
552
553
554
555
556
557
558
559
560
        # 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
        ])
561
562
563
564
565
566
567
        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
568
569

    @property
570
    def processed_calcs(self):
571
        return Calc.objects(upload_id=self.upload_id, tasks_status__in=[SUCCESS, FAILURE]).count()
572
573
574
575
576
577
578

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

    @property
    def failed_calcs(self):
579
        return Calc.objects(upload_id=self.upload_id, tasks_status=FAILURE).count()
580

581
582
    @property
    def pending_calcs(self):
583
        return Calc.objects(upload_id=self.upload_id, tasks_status=PENDING).count()
584

585
586
    def all_calcs(self, start, end, order_by='mainfile'):
        return Calc.objects(upload_id=self.upload_id)[start:end].order_by(order_by)
587
588
589

    @property
    def calcs(self):
Markus Scheidgen's avatar
Markus Scheidgen committed
590
        return Calc.objects(upload_id=self.upload_id, tasks_status=SUCCESS)
Markus Scheidgen's avatar
Markus Scheidgen committed
591
592

    def to_upload_with_metadata(self) -> UploadWithMetadata:
593
594
595
596
597
598
599
600
601
        # 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
602
        if self.metadata is not None:
603
604
605
            upload_metadata.update(self.metadata)
            if 'calculations' in upload_metadata:
                del(upload_metadata['calculations'])
Markus Scheidgen's avatar
Markus Scheidgen committed
606

607
608
            for calc in self.metadata.get('calculations', []):
                calc_metadatas[calc['mainfile']] = calc
Markus Scheidgen's avatar
Markus Scheidgen committed
609

610
        user_upload_time = upload_metadata.get('_upload_time', None)
Markus Scheidgen's avatar
Markus Scheidgen committed
611
612
613
614
615
        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)

616
617
618
619
620
621
622
623
624
        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)

625
626
            calc_metadata = dict(upload_metadata)
            calc_metadata.update(calc_metadatas.get(calc.mainfile, {}))
627
628
            calc_with_metadata.apply_user_metadata(calc_metadata)

Markus Scheidgen's avatar
Markus Scheidgen committed
629
            logger.debug('prepared calc with metadata', calc_id=calc_with_metadata.calc_id)
630
631
632
633
634
            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
635
636

        return result