Python RecordIndexer.process_bulk_queue方法代码示例

# 需要导入模块: from invenio_indexer.api import RecordIndexer [as 别名]
# 或者: from invenio_indexer.api.RecordIndexer import process_bulk_queue [as 别名]
def update_expired_embargoes():
    """Release expired embargoes every midnight."""
    logger = current_app.logger
    base_url = urlunsplit((
        current_app.config.get('PREFERRED_URL_SCHEME', 'http'),
        current_app.config['JSONSCHEMAS_HOST'],
        current_app.config.get('APPLICATION_ROOT') or '', '', ''
    ))
    # The task needs to run in a request context as JSON Schema validation
    # will use url_for.
    with current_app.test_request_context('/', base_url=base_url):
        s = B2ShareRecordsSearch(
            using=current_search_client,
            index='records'
        ).query(
            'query_string',
            query='open_access:false AND embargo_date:{{* TO {0}}}'.format(
                datetime.now(timezone.utc).isoformat()
            ),
            allow_leading_wildcard=False
        ).fields([])
        record_ids = [hit.meta.id for hit in s.scan()]
        if record_ids:
            logger.info('Changing access of {} embargoed publications'
                        ' to public.'.format(len(record_ids)))
        for record in Record.get_records(record_ids):
            logger.debug('Making embargoed publication {} public'.format(
                record.id))
            record['open_access'] = True
            record.commit()
        db.session.commit()

        indexer = RecordIndexer()
        indexer.bulk_index(record_ids)
        indexer.process_bulk_queue()

# 需要导入模块: from invenio_indexer.api import RecordIndexer [as 别名]
# 或者: from invenio_indexer.api.RecordIndexer import process_bulk_queue [as 别名]
def update_expired_embargos():
    """Release expired embargoes every midnight."""
    record_ids = AccessRight.get_expired_embargos()
    for record in Record.get_records(record_ids):
        record['access_right'] = AccessRight.OPEN
        record.commit()
    db.session.commit()

    indexer = RecordIndexer()
    indexer.bulk_index(record_ids)
    indexer.process_bulk_queue()

# 需要导入模块: from invenio_indexer.api import RecordIndexer [as 别名]
# 或者: from invenio_indexer.api.RecordIndexer import process_bulk_queue [as 别名]
def test_record_indexing(app, queue):
    """Run record autoindexer."""
    @before_record_index.connect_via(app)
    def remove_schema(sender, json=None, record=None):
        if '$schema' in json:
            del json['$schema']

    models_committed.connect(process_models_committed_signal, sender=app)

    with app.app_context():

        current_search_client.indices.delete_alias('_all', '_all',
                                                   ignore=[400, 404])
        current_search_client.indices.delete('*')
        aliases = current_search_client.indices.get_aliases()
        assert 0 == len(aliases)

    runner = CliRunner()
    script_info = ScriptInfo(create_app=lambda info: app)

    with runner.isolated_filesystem():
        result = runner.invoke(cmd, ['destroy', '--yes-i-know'],
                               obj=script_info)
        result = runner.invoke(cmd, ['init'],
                               obj=script_info)
        assert 0 == result.exit_code

    with app.app_context():
        from invenio_records.models import RecordMetadata
        with db.session.begin_nested():
            record1 = RecordMetadata(json={
                '$schema': ('http://example.com/schemas/'  # external site
                            'records/default-v1.0.0.json'),
                'title': 'Test1',
            })
            db.session.add(record1)
            record2 = RecordMetadata(json={
                '$schema': {
                    '$ref': ('http://example.com/schemas/'  # external site
                             'records/default-v1.0.0.json')
                },
                'title': 'Test2',
            })
            db.session.add(record2)
        db.session.commit()

        record_indexer = RecordIndexer(queue=queue)
        result = record_indexer.process_bulk_queue()
        assert 2 == len(list(result))

        response = current_search_client.get(
            index='records-default-v1.0.0',
            id=record1.id,
        )
        assert str(record1.id) == response['_id']

        response = current_search_client.get(
            index='records-default-v1.0.0',
            id=record2.id,
        )
        assert str(record2.id) == response['_id']

        db.session.delete(record1)
        db.session.commit()

        record_indexer.process_bulk_queue()

        response = current_search_client.get(
            index='records-default-v1.0.0',
            id=record1.id,
            ignore=404,
        )
        assert not response['found']

    # Clean-up:
    with app.app_context():
        result = runner.invoke(cmd, ['destroy', '--yes-i-know'],
                               obj=script_info)
        assert 0 == result.exit_code