Python spark_job_results示例

示例#1

文件： test_scheduling.py 项目： g-k/telemetry-analysis-service

def test_spark_job_results(mocker, public):
    identifier = 'job-identifier'

    def mocked_list_objects(**kwargs):
        is_public = kwargs['Bucket'] == settings.AWS_CONFIG['PUBLIC_DATA_BUCKET']
        pub_prefix = 'pub' if is_public else ''
        return {
            'Contents': [
                {'Key': '{}/logs/{}my-log.txt'.format(identifier, pub_prefix)},
                {'Key': '{}/data/{}my-notebook.ipynb'.format(identifier, pub_prefix)},
                {'Key': '{}/data/{}sub/artifact.txt'.format(identifier, pub_prefix)},
            ]
        }
    mocker.patch.object(s3, 'list_objects_v2', mocked_list_objects)

    prefix = 'pub' if public else ''

    results = scheduling.spark_job_results(identifier, public)
    assert results == {
        'data': [
            '{}/data/{}my-notebook.ipynb'.format(identifier, prefix),
            '{}/data/{}sub/artifact.txt'.format(identifier, prefix),
        ],
        'logs': [
            '{}/logs/{}my-log.txt'.format(identifier, prefix)
        ]
    }

示例#2

文件： test_scheduling.py 项目： g-k/telemetry-analysis-service

def test_spark_job_results_empty(mocker):
    mocker.patch.object(s3, 'list_objects_v2', return_value={})

    results = scheduling.spark_job_results('job-identifier', True)
    assert results == {}