admin.py 31.5 KB
Newer Older
Markus Scheidgen's avatar
Markus Scheidgen committed
1
2
3
4
#
# Copyright The NOMAD Authors.
#
# This file is part of NOMAD. See https://nomad-lab.eu for further info.
5
6
7
8
9
#
# 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
#
Markus Scheidgen's avatar
Markus Scheidgen committed
10
#     http://www.apache.org/licenses/LICENSE-2.0
11
12
#
# Unless required by applicable law or agreed to in writing, software
Markus Scheidgen's avatar
Markus Scheidgen committed
13
# distributed under the License is distributed on an "AS IS" BASIS,
14
15
16
# 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.
Markus Scheidgen's avatar
Markus Scheidgen committed
17
18
#

Markus Scheidgen's avatar
Markus Scheidgen committed
19
import typing
20
import click
21
import datetime
Markus Scheidgen's avatar
Markus Scheidgen committed
22
23
import elasticsearch_dsl
import elasticsearch
24
import sys
25
import threading
26

27
from nomad import processing as proc, search, datamodel, infrastructure, utils, config
28
29
30
from nomad.cli.cli import cli


Markus Scheidgen's avatar
Markus Scheidgen committed
31
32
def __run_parallel(
        uploads, parallel: int, callable, label: str):
33
34
35
36
37
38
39
40
    if isinstance(uploads, (tuple, list)):
        uploads_count = len(uploads)

    else:
        uploads_count = uploads.count()
        uploads = list(uploads)  # copy the whole mongo query set to avoid cursor timeouts

    cv = threading.Condition()
Markus Scheidgen's avatar
Markus Scheidgen committed
41
    threads: typing.List[threading.Thread] = []
42
43
44
45
46
47
48
49
50
51
52
53
54
55

    state = dict(
        completed_count=0,
        skipped_count=0,
        available_threads_count=parallel)

    logger = utils.get_logger(__name__)

    print('%d uploads selected, %s ...' % (uploads_count, label))

    def process_upload(upload: proc.Upload):
        logger.info('%s started' % label, upload_id=upload.upload_id)

        completed = False
Markus Scheidgen's avatar
Markus Scheidgen committed
56
        if callable(upload, logger):
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
            completed = True

        with cv:
            state['completed_count'] += 1 if completed else 0
            state['skipped_count'] += 1 if not completed else 0
            state['available_threads_count'] += 1

            print(
                '   %s %s and skipped %s of %s uploads' %
                (label, state['completed_count'], state['skipped_count'], uploads_count))

            cv.notify()

    for upload in uploads:
        with cv:
            cv.wait_for(lambda: state['available_threads_count'] > 0)
            state['available_threads_count'] -= 1
            thread = threading.Thread(target=lambda: process_upload(upload))
            threads.append(thread)
            thread.start()

    for thread in threads:
        thread.join()


Markus Scheidgen's avatar
Markus Scheidgen committed
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
def __run_processing(
        uploads, parallel: int, process, label: str, reprocess_running: bool = False):

    def run_process(upload, logger):
        if upload.process_running and not reprocess_running:
            logger.warn(
                'cannot trigger %s, since the upload is already/still processing' % label,
                current_process=upload.current_process,
                current_task=upload.current_task, upload_id=upload.upload_id)
            return False
        else:
            upload.reset()
            process(upload)
            upload.block_until_complete(interval=.5)

            if upload.tasks_status == proc.FAILURE:
                logger.info('%s with failure' % label, upload_id=upload.upload_id)

            logger.info('%s complete' % label, upload_id=upload.upload_id)
            return True

    __run_parallel(uploads, parallel=parallel, callable=run_process, label=label)


106
107
108
109
110
@cli.group(help='''The nomad admin commands to do nasty stuff directly on the databases.
                     Remember: With great power comes great responsibility!''')
@click.pass_context
def admin(ctx):
    pass
111
112


113
114
115
116
117
118
119
120
121
122
123
124
125
126
@admin.command(help='Reset/remove all databases.')
@click.option('--remove', is_flag=True, help='Do not just reset all dbs, but also remove them.')
@click.option('--i-am-really-sure', is_flag=True, help='Must be set for the command to to anything.')
def reset(remove, i_am_really_sure):
    if not i_am_really_sure:
        print('You do not seem to be really sure about what you are doing.')
        sys.exit(1)

    infrastructure.setup_mongo()
    infrastructure.setup_elastic()

    infrastructure.reset(remove)


127
128
129
130
131
132
@admin.command(help='Reset all "stuck" in processing uploads and calc in low level mongodb operations.')
@click.option('--zero-complete-time', is_flag=True, help='Sets the complete time to epoch zero.')
def reset_processing(zero_complete_time):
    infrastructure.setup_mongo()

    def reset_collection(cls):
133
        in_processing = cls.objects(process_status__in=[proc.PROCESS_RUNNING, proc.base.PROCESS_CALLED])
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
        print('%d %s processes need to be reset due to incomplete process' % (in_processing.count(), cls.__name__))
        in_processing.update(
            process_status=None,
            current_process=None,
            worker_hostname=None,
            celery_task_id=None,
            errors=[], warnings=[],
            complete_time=datetime.datetime.fromtimestamp(0) if zero_complete_time else datetime.datetime.now(),
            current_task=None,
            tasks_status=proc.base.CREATED)

        in_tasks = cls.objects(tasks_status__in=[proc.PENDING, proc.RUNNING])
        print('%d %s processes need to be reset due to incomplete tasks' % (in_tasks.count(), cls.__name__))
        in_tasks.update(
            current_task=None,
            tasks_status=proc.base.CREATED,
            errors=[], warnings=[],
            complete_time=datetime.datetime.fromtimestamp(0) if zero_complete_time else datetime.datetime.now())

    reset_collection(proc.Calc)
    reset_collection(proc.Upload)


157
158
159
160
161
162
163
164
@admin.command(help='Check and lift embargo of data with expired embargo period.')
@click.option('--dry', is_flag=True, help='Do not lift the embargo, just show what needs to be done.')
@click.option('--parallel', default=1, type=int, help='Use the given amount of parallel processes. Default is 1.')
def lift_embargo(dry, parallel):
    infrastructure.setup_mongo()
    infrastructure.setup_elastic()

    request = search.SearchRequest()
Markus Scheidgen's avatar
Markus Scheidgen committed
165
    request.q = elasticsearch_dsl.Q('term', with_embargo=True) & elasticsearch_dsl.Q('term', published=True)
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
    request.quantity('upload_id', 1000)
    result = request.execute()

    uploads_to_repack = []
    for upload_id in result['quantities']['upload_id']['values']:
        upload = proc.Upload.get(upload_id)
        embargo_length = upload.embargo_length
        if embargo_length is None:
            embargo_length = 36
            upload.embargo_length = 36

        if upload.upload_time + datetime.timedelta(days=int(embargo_length * 365 / 12)) < datetime.datetime.now():
            print('need to lift the embargo of %s (upload_time=%s, embargo=%d)' % (
                upload.upload_id, upload.upload_time, embargo_length))

            if not dry:
                proc.Calc._get_collection().update_many(
                    {'upload_id': upload_id},
                    {'$set': {'metadata.with_embargo': False}})
                uploads_to_repack.append(upload)
                upload.save()

188
189
                with upload.entries_metadata() as entries:
                    search.index_all(entries)
190
191
192
193
194

    if not dry:
        __run_processing(uploads_to_repack, parallel, lambda upload: upload.re_pack(), 're-packing')


195
@admin.command(help='(Re-)index all calcs.')
196
@click.option('--threads', type=int, default=1, help='Number of threads to use.')
197
198
@click.option('--dry', is_flag=True, help='Do not index, just compute entries.')
def index(threads, dry):
199
200
201
202
203
204
    infrastructure.setup_mongo()
    infrastructure.setup_elastic()

    all_calcs = proc.Calc.objects().count()
    print('indexing %d ...' % all_calcs)

205
    def elastic_updates():
206
207
208
        with utils.ETA(all_calcs, '   index %10d or %10d calcs, ETA %s') as eta:
            for calc in proc.Calc.objects():
                eta.add()
209
210
                entry_metadata = datamodel.EntryMetadata.m_from_dict(calc.metadata)
                entry = entry_metadata.a_elastic.create_index_entry().to_dict(include_meta=True)
211
212
                entry['_op_type'] = 'index'
                yield entry
Markus Scheidgen's avatar
Markus Scheidgen committed
213

214
215
216
    if dry:
        for _ in elastic_updates():
            pass
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
    else:
        if threads > 1:
            print('  use %d threads' % threads)
            for _ in elasticsearch.helpers.parallel_bulk(
                    infrastructure.elastic_client, elastic_updates(), chunk_size=500,
                    thread_count=threads):
                pass
        else:
            elasticsearch.helpers.bulk(
                infrastructure.elastic_client, elastic_updates())
        search.refresh()

    print('')
    print('indexing completed')


@admin.command()
@click.option('--threads', type=int, default=1, help='Number of threads to use.')
@click.option('--code', multiple=True, type=str, help='Index only calculcations of given codes.')
@click.option('--dry', is_flag=True, help='Do not index, just compute entries.')
@click.option('--in-place', is_flag=True, default=False, help='Perform indexing in the current elastic search index. Meant only for small reindex operations.')
@click.option('-n', type=int, default=None, help='Number of calculations to process. Leave undefined to process all calculations.')
@click.option('--source',
              type=click.Choice(['mongo', 'es'], case_sensitive=True))
def index_materials(threads, code, dry, in_place, n, source):
    """(Re-)index all materials.

244
245
246
247
248
249
250
251
252
    This command will is used to completely rebuild the materials index or
    parts of it. You can choose which source is used for building the index: In
    general ElasticSearch is preferred, but if it is not available also MongoDB
    can be used.

    By default this command will start to rebuild the index from scratch.  This
    can be done on a "live" system because a new temporary index is used. If
    you use the --in-place option, the indexing will be run on the same index
    that is currently in use.
253
    """
254

Markus Scheidgen's avatar
Markus Scheidgen committed
255
256
257
258
259
    from nomad.datamodel.material import Material, Calculation
    from nomad.datamodel.encyclopedia import EncyclopediaMetadata
    from nomad.search import material_document
    from nomad.datamodel.material import Material, Calculation, Method, Properties, IdealizedStructure, Energies, Workflow, Bulk

260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
    def create_entry(material, calc, in_place):
        """Creates an ES update operation that inserts the full material info
        if entry does not exists, otherwise only adds the calculation into the
        nested subdocument, possibly replacing the old one if in_place is True.
        """
        entry = {}
        entry['_op_type'] = 'update'
        entry['_index'] = target_index_name
        entry['_id'] = material.material_id
        entry['_type'] = 'doc'
        entry['_source'] = {
            "upsert": material.m_to_dict(include_defaults=False, partial="es"),
            "doc_as_upsert": False,
            "script": {
                "params": {
                    "calc": calc.m_to_dict(include_defaults=False, partial="es")
                },
            }
        }
        if in_place:
            entry['_source']["script"]["source"] = "ctx._source.calculations.removeIf(x -> x.calc_id == params.calc.calc_id); ctx._source.calculations.add(params.calc)"
        else:
            entry['_source']["script"]["source"] = "ctx._source.calculations.add(params.calc)"
        return entry

285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
    chunk_size = 500
    infrastructure.setup_mongo()
    client = infrastructure.setup_elastic()

    # In order to do the reindexing with zero downtime, two different indices
    # are rotated and an alias is used
    old_index_name = list(client.indices.get(config.elastic.materials_index_name).keys())[0]
    if in_place:
        target_index_name = old_index_name
    else:
        if old_index_name == config.elastic.materials_index_name + "_a":
            target_index_name = config.elastic.materials_index_name + "_b"
        elif old_index_name == config.elastic.materials_index_name + "_b":
            target_index_name = config.elastic.materials_index_name + "_a"
        else:
            raise ValueError(
                "Unrecognized index name accociated with the alias {}"
                .format(config.elastic.materials_index_name)
            )

    if source == "mongo":
306
307
308
309
310
311
312
        mongo_db = infrastructure.mongo_client[config.mongo.db_name]
        mongo_collection = mongo_db['archive']
        if code:
            collection = mongo_collection.find({"section_metadata.dft.code_name": {"$in": code}})
        else:
            collection = mongo_collection.find()
        all_calcs = collection.count()
313
314
315
316
317
318
        print('indexing materials from %d calculations ...' % all_calcs)

        # Bulk update
        def elastic_updates():
            with utils.ETA(all_calcs, '   index %10d of %10d calcs, ETA %s') as eta:
                i_calc = 0
319
                for mongo_archive in collection:
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
                    i_calc += 1
                    if n is not None:
                        if i_calc > n:
                            return
                    eta.add()

                    # Do not process entries that do not have the material
                    # information
                    try:
                        status = mongo_archive["section_metadata"]["encyclopedia"]["status"]
                        if status != EncyclopediaMetadata.status.type.success:
                            raise AttributeError
                    except (KeyError, AttributeError, IndexError):
                        continue

                    # Create material information
                    metadata = mongo_archive["section_metadata"]
                    encyclopedia = EncyclopediaMetadata.m_from_dict(metadata["encyclopedia"])
                    dft = metadata["dft"]
                    material: Material = Material()
                    material.material_id = encyclopedia.material.material_id
                    material.material_type = encyclopedia.material.material_type
                    material.material_name = encyclopedia.material.material_name
                    material.material_classification = encyclopedia.material.material_classification
                    material.formula = encyclopedia.material.formula
                    material.formula_reduced = encyclopedia.material.formula_reduced
                    material.species_and_counts = encyclopedia.material.species_and_counts
                    material.species = encyclopedia.material.species
                    enc_bulk = encyclopedia.material.bulk
                    if enc_bulk:
                        bulk = Bulk.m_from_dict(enc_bulk.m_to_dict())
                        material.m_add_sub_section(Material.bulk, bulk)

                    # Create calculation info for this entry
                    calc = Calculation()
                    calc.calc_id = metadata["calc_id"]
                    calc.upload_id = metadata["upload_id"]
                    mongo_calc = proc.Calc.get(calc.calc_id)
                    calc.published = mongo_calc["metadata"]["published"]
                    calc.with_embargo = mongo_calc["metadata"]["with_embargo"]
                    calc.owners = [mongo_calc["metadata"]["uploader"]] + mongo_calc["metadata"]["shared_with"]
                    enc_idealized_structure = encyclopedia.material.idealized_structure
                    idealized_structure = IdealizedStructure()
                    cell_volume = enc_idealized_structure.cell_volume
                    if cell_volume is not None:
                        idealized_structure.cell_volume = cell_volume
                    idealized_structure.lattice_parameters = enc_idealized_structure.lattice_parameters
                    calc.m_add_sub_section(Calculation.idealized_structure, idealized_structure)
                    enc_method = encyclopedia.method
                    method = Method.m_from_dict(enc_method.m_to_dict())
                    method.program_name = dft["code_name"]
                    method.program_version = dft["code_version"]
                    method.basis_set = dft["basis_set"]
                    calc.m_add_sub_section(Calculation.method, method)
                    enc_props = encyclopedia.properties

                    # Properties may not exist at all
                    if enc_props is not None:
                        properties = Properties()

                        # Energies may not be present in all calculations
                        try:
                            energies = Energies.m_from_dict(enc_props.energies.m_to_dict())
                            properties.m_add_sub_section(Properties.energies, energies)
                        except AttributeError:
                            pass

                        properties.has_electronic_dos = enc_props.electronic_dos is not None
                        properties.has_electronic_band_structure = enc_props.electronic_band_structure is not None
                        properties.has_thermodynamical_properties = enc_props.thermodynamical_properties is not None
                        atomic_density = enc_props.atomic_density
                        if atomic_density is not None:
                            properties.atomic_density = atomic_density
                        mass_density = enc_props.mass_density
                        if mass_density is not None:
                            properties.mass_density = mass_density
                        band_gap = enc_props.band_gap
                        if band_gap is not None:
                            properties.band_gap = band_gap
                        band_gap_direct = enc_props.band_gap_direct
                        if band_gap_direct is not None:
                            properties.band_gap_direct = band_gap_direct
                        calc.m_add_sub_section(Calculation.properties, properties)

                    workflow = Workflow()
                    workflow.workflow_type = encyclopedia.calculation.calculation_type
                    calc.m_add_sub_section(Calculation.workflow, workflow)
                    material.m_add_sub_section(Material.calculations, calc)

                    # Update entry that inserts the full material info if entry
410
411
412
413
                    # does not exists, otherwise only adds the calculation into
                    # the nested subdocument
                    yield create_entry(material, calc, in_place)

414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
    elif source == "es":
        s = elasticsearch_dsl.Search(index=config.elastic.index_name)
        filters = [elasticsearch_dsl.Q("term", encyclopedia__status="success")]
        if code:
            filters.append(elasticsearch_dsl.Q("terms", dft__code_name=code))
        query = elasticsearch_dsl.Q(
            "bool",
            filter=filters,
        )
        s = s.query(query)
        s = s.extra(**{
            "size": 0,
        })
        all_calcs = s.execute().hits.total
        print('indexing materials from %d calculations ...' % all_calcs)

        def elastic_updates():
            with utils.ETA(all_calcs, '   index %10d of %10d calcs, ETA %s', chunk_size) as eta:

                s = elasticsearch_dsl.Search(index=config.elastic.index_name)
                filters = [elasticsearch_dsl.Q("term", encyclopedia__status="success")]
                if code:
                    filters.append(elasticsearch_dsl.Q("terms", dft__code_name=code))
                query = elasticsearch_dsl.Q(
                    "bool",
                    filter=filters,
                )
                s = s.query(query)
                s = s.extra(**{
                    "size": chunk_size,
                })
                i_calc = 0
                for hit in s.scan():
                    i_calc += 1
                    if n is not None:
                        if i_calc > n:
                            return
                    eta.add()

                    material: Material = Material()
                    calc = Calculation()

                    # Check that all required information exists. If not, the
                    # calculation is skipped.
                    try:
                        material.material_id = hit.encyclopedia.material.material_id
                        material.material_type = hit.encyclopedia.material.material_type
                        material.formula = hit.encyclopedia.material.formula
                        material.formula_reduced = hit.encyclopedia.material.formula_reduced
                        material.species_and_counts = hit.encyclopedia.material.species_and_counts
                        material.species = hit.encyclopedia.material.species
                        calc.calc_id = hit.calc_id
                        calc.upload_id = hit.upload_id
                        calc.published = hit.published
                        calc.with_embargo = hit.with_embargo
                        calc.owners = [x.user_id for x in hit.owners]
                        idealized_structure = IdealizedStructure.m_from_dict(hit.encyclopedia.material.idealized_structure.to_dict())
                        calc.m_add_sub_section(Calculation.idealized_structure, idealized_structure)

                        method = Method.m_from_dict(hit.encyclopedia.method.to_dict())
                        method.program_name = hit.dft.code_name
                        method.program_version = hit.dft.code_version
                        method.basis_set = hit.dft.basis_set
                        calc.m_add_sub_section(Calculation.method, method)

                        workflow = Workflow()
                        workflow.workflow_type = hit.encyclopedia.calculation.calculation_type
                        calc.m_add_sub_section(Calculation.workflow, workflow)
                    except AttributeError:
                        continue

                    # Not all materials have a name
                    try:
                        material.material_name = hit.encyclopedia.material.material_name
                    except AttributeError:
                        pass

                    # Not all materials have a bulk section
                    try:
                        bulk = Bulk.m_from_dict(hit.encyclopedia.material.bulk)
                        material.m_add_sub_section(Material.bulk, bulk)
                    except AttributeError:
                        pass

                    # Properties may not exist at all
                    try:
                        enc_properties = hit.encyclopedia.properties
                    except AttributeError:
                        pass
                    else:
                        properties = Properties()

                        # Energies may not be present in all calculations
                        try:
                            energies = Energies.m_from_dict(enc_properties.energies.to_dict())
                            properties.m_add_sub_section(Properties.energies, energies)
                        except AttributeError:
                            pass

                        # Gather the boolean flags that indicate the presence of
                        # certain properties
                        try:
                            properties.has_electronic_dos = enc_properties.electronic_dos is not None
                        except AttributeError:
                            properties.has_electronic_dos = False
                        try:
                            properties.has_electronic_band_structure = enc_properties.electronic_band_structure is not None
                        except AttributeError:
                            properties.has_electronic_band_structure = False
                        try:
                            properties.has_thermodynamical_properties = enc_properties.thermodynamical_properties is not None
                        except AttributeError:
                            properties.has_thermodynamical_properties = False

                        # Not all materials have an atomic density
                        try:
                            properties.atomic_density = enc_properties.atomic_density
                        except AttributeError:
                            pass

                        # Not all materials have a mass density
                        try:
                            properties.mass_density = enc_properties.mass_density
                        except AttributeError:
                            pass

                        # Not all materials have band gaps
                        try:
                            properties.band_gap = enc_properties.band_gap
                        except AttributeError:
                            pass

                        # Not all materials have band gap type
                        try:
                            properties.band_gap_direct = enc_properties.band_gap_direct
                        except AttributeError:
                            pass

                        calc.m_add_sub_section(Calculation.properties, properties)

                    material.m_add_sub_section(Material.calculations, calc)

556
                    yield create_entry(material, calc, in_place)
557
558
559

    if dry:
        for _ in elastic_updates():
560
            pass
561
    else:
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
        # Create new index into which the data will be inserted. The old index will
        # keep working while the new index is being built
        material_document.init(index=target_index_name)

        if threads > 1:
            print('  use %d threads' % threads)
            for _ in elasticsearch.helpers.parallel_bulk(
                    infrastructure.elastic_client, elastic_updates(), chunk_size=chunk_size,
                    thread_count=threads):
                pass
        else:
            elasticsearch.helpers.bulk(
                infrastructure.elastic_client, elastic_updates())
            search.refresh()

        # Changes materials index alias to point to the new index and remove the
        # old index.
        if not in_place:
            new_index = elasticsearch_dsl.Index(target_index_name)
            new_index.put_alias(name=config.elastic.materials_index_name)
            old_index = elasticsearch_dsl.Index(old_index_name)
            old_index.delete()
584

585
    print('')
586
    print('indexing completed')
587
588


589
@admin.group(help='Generate scripts and commands for nomad operation.')
590
591
592
593
def ops():
    pass


594
595
596
597
598
@ops.group(help='Tools for managing the DOS similarity data.')
def similarity():
    pass


599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
@ops.command(help=('Dump the mongo (calculation metadata) db.'))
@click.option('--restore', is_flag=True, help='Do not dump, but restore.')
def dump(restore: bool):
    date_str = datetime.datetime.utcnow().strftime('%Y_%m_%d')
    print('mongodump --host {} --port {} --db {} -o /backup/fairdi/mongo/{}'.format(
        config.mongo.host, config.mongo.port, config.mongo.db_name, date_str))


@ops.command(help=('Restore the mongo (calculation metadata) db.'))
@click.argument('PATH_TO_DUMP', type=str, nargs=1)
def restore(path_to_dump):
    print('mongorestore --host {} --port {} --db {} {}'.format(
        config.mongo.host, config.mongo.port, config.mongo.db_name, path_to_dump))


614
615
616
617
618
619
620
621
622
623
@ops.command(help=('Generate an nginx.conf to serve the GUI and proxy pass to API container.'))
@click.option('--prefix', type=str, default='/example_nomad', help='Url path prefix. Default is /example_nomd, can be empty str.')
def nginx_conf(prefix):
    prefix = prefix.rstrip('/')
    prefix = '/%s' % prefix.lstrip('/')

    print('''\
server {{
    listen        80;
    server_name   www.example.com;
624
    proxy_set_header Host $host;
625

626
627
    location / {{
        proxy_pass http://app:8000;
628
629
    }}

630
631
    location ~ {1}\\/?(gui)?$ {{
        rewrite ^ {1}/gui/ permanent;
632
633
    }}

634
635
636
637
638
639
640
641
642
    location {1}/gui/ {{
        proxy_intercept_errors on;
        error_page 404 = @redirect_to_index;
        proxy_pass http://app:8000;
    }}

    location @redirect_to_index {{
        rewrite ^ {1}/gui/index.html break;
        proxy_pass http://app:8000;
643
644
    }}

645
    location ~ \\/gui\\/(service-worker\\.js|meta\\.json)$ {{
646
647
648
649
650
        add_header Last-Modified $date_gmt;
        add_header Cache-Control 'no-store, no-cache, must-revalidate, proxy-revalidate, max-age=0';
        if_modified_since off;
        expires off;
        etag off;
651
        proxy_pass http://app:8000;
652
653
    }}

654
    location ~ \\/api\\/uploads\\/?$ {{
655
656
        client_max_body_size 35g;
        proxy_request_buffering off;
657
        proxy_pass http://app:8000;
658
659
    }}

660
    location ~ \\/api\\/(raw|archive) {{
661
        proxy_buffering off;
662
        proxy_pass http://app:8000;
663
664
    }}

665
    location ~ \\/api\\/mirror {{
666
        proxy_buffering off;
667
668
        proxy_read_timeout 600;
        proxy_pass http://app:8000;
669
    }}
670
}}'''.format(prefix))
671
672
673


@ops.command(help=('Generate a proxy pass config for apache2 reverse proxy servers.'))
674
@click.option('--prefix', type=str, default='app', help='The path prefix under which everything is proxy passed.')
675
676
@click.option('--host', type=str, default='130.183.207.104', help='The host to proxy to.')
@click.option('--port', type=str, default='30001', help='The port to proxy to.')
677
678
def apache_conf(prefix, host, port):
    print('''\
679
680
681
682
ProxyPass "/{0}" "http://{1}:{2}/{0}"
ProxyPassReverse "/{0}" "http://{1}:{2}/{0}"
<Proxy http://{1}:{2}/{0}>
    ProxyPreserveHost On
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
    <IfModule !mod_access_compat.c>
         Require all granted
     </IfModule>
     <IfModule mod_access_compat.c>
         Order allow,deny
         Allow from all
     </IfModule>
</Proxy>

RequestHeader set "X-Forwarded-Proto" expr=%{{REQUEST_SCHEME}}
RequestHeader set "X-Forwarded-SSL" expr=%{{HTTPS}}

ProxyPass /fairdi/keycloak http://{1}:8002/fairdi/keycloak
ProxyPassReverse /fairdi/keycloak http://{1}:8002/fairdi/keycloak
<Proxy http://{1}:8002/app>
     ProxyPreserveHost On
     <IfModule !mod_access_compat.c>
         Require all granted
     </IfModule>
     <IfModule mod_access_compat.c>
         Order allow,deny
         Allow from all
     </IfModule>
</Proxy>

RewriteEngine on
RewriteCond %{QUERY_STRING} ^pid=([^&]+)$
RewriteRule ^/NomadRepository-1.1/views/calculation.zul$ /{0}/gui/entry/pid/%1? [R=301]

AllowEncodedSlashes On
713
'''.format(prefix, host, port))  # type: ignore
714
715
716
717
718
719
720


@ops.command(help='Updates the AFLOW prototype information using the latest online version and writes the results to a python module in the given FILEPATH.')
@click.argument('FILEPATH', nargs=1, type=str)
@click.option('--matches-only', is_flag=True, help='Only update the match information that depends on the symmetry analysis settings. Will not perform and online update.')
@click.pass_context
def prototypes_update(ctx, filepath, matches_only):
721
722
    from nomad.cli.admin import prototypes
    prototypes.update_prototypes(ctx, filepath, matches_only)
723

724
725

@ops.command(help='Updates the springer database in nomad.config.normalize.springer_db_path.')
726
727
728
@click.option('--max-n-query', default=10, type=int, help='Number of unsuccessful springer request before returning an error. Default is 10.')
@click.option('--retry-time', default=120, type=int, help='Time in seconds to retry after unsuccessful request. Default is 120.')
def springer_update(max_n_query, retry_time):
729
    from nomad.cli.admin import springer
730
    springer.update_springer(max_n_query, retry_time)
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748


@similarity.command(help='Updates the msgpack file containing the similarity information.')
@click.option('--dir', "-d", "input_dir", type=str, help='Path of the folder containing the raw similarity information files')
@click.option('--out', "-o", type=str, help='Path of the output msgpack file.')
@click.option('--verbose', is_flag=True, help='Enable verbose output.')
def update(input_dir, out, verbose):
    from nomad.cli.admin import similarity
    similarity.update(input_dir, out, verbose)


@similarity.command(help='Ingests the given similarity information from an msgpack file into MongoDB.')
@click.option('--in', "-i", "input_path", type=str, help='Path of the ingested msgpack file.')
@click.option('--batch_size', type=int, default=10000, help='Batch size for MongoDB bulk ingestion.')
@click.option('--verbose', is_flag=True, help='Enable verbose output.')
def ingest(input_path, batch_size, verbose):
    from nomad.cli.admin import similarity
    similarity.ingest(input_path, batch_size, verbose)