class BigQueryTest(unittest.TestCase):
def setUp(self):
self.project_id = os.getenv('PROJECT_ID')
self.dataset_id = os.getenv('DATASET_ID', 'test_dataset')
self.table_id = os.getenv('TABLE_ID', 'test_table') + '_' + str(int(time.time()))
self.view_id = os.getenv('VIEW_ID', 'test_view') + '_' + str(int(time.time()))
if self.project_id is None:
print('PROJECT_ID is not defined.')
sys.exit(1)
self.bq = BigQuery(self.project_id)
if self.bq.exists_dataset(self.dataset_id):
self.bq.drop_dataset(self.dataset_id, delete_contents=True)
self.bq.create_dataset(self.dataset_id)
self.bq.dataset_id = self.dataset_id # Set default datasetId
schema = [
{ 'name': 'id', 'type': 'INTEGER', 'mode': 'REQUIRED' },
{ 'name': 'name', 'type': 'STRING', 'mode': 'REQUIRED' },
{ 'name': 'birth', 'type': 'RECORD', 'mode': 'NULLABLE', 'fields': [
{ 'name': 'year', 'type': 'INTEGER', 'mode': 'REQUIRED' },
{ 'name': 'month', 'type': 'INTEGER', 'mode': 'REQUIRED' },
{ 'name': 'day', 'type': 'INTEGER', 'mode': 'REQUIRED' },
]},
{ 'name': 'url', 'type': 'STRING', 'mode': 'REPEATED' },
]
self.bq.create_table(self.table_id, schema=schema)
def TearDown(self):
self.bq.drop_table(self.table_id)
self.bq.drop_dataset(self.dataset_id, delete_contents=True)
def test_error_invalid_rows(self):
rows = [
{ 'id': 1, 'name': 'foo' },
{ 'id': 2 },
{ 'id': 'three', 'name': 'baz' },
]
with self.assertRaises(BigQueryError):
self.bq.load(self.table_id, rows)
res = self.bq.dump_table(self.table_id)
self.assertEqual(0, len(res))
pprint(res)
def test_error_unknown_values(self):
rows = [
{ 'id': 1, 'name': 'foo', 'unknown_field': 'unknown_value' },
]
with self.assertRaises(BigQueryError):
self.bq.load(self.table_id, rows)
def test_normal_unknown_values(self):
rows = [
{ 'id': 1, 'name': 'foo', 'unknown_field': 'unknown_value' },
]
self.bq.load(self.table_id, rows, ignore_unknown_values=True)
res = self.bq.dump_table(self.table_id)
self.assertEqual(1, len(res))
pprint(res)
def test_normal_async(self):
filepath = os.path.dirname(os.path.abspath(__file__)) + '/data.json'
job_id = self.bq.load(self.table_id, filepath, async=True)
self.assertTrue(re.match(r'job_', job_id))
print(job_id)
while True:
res = self.bq.info_job(job_id)
state = res['status']['state']
print(state)
if state == 'DONE': break
time.sleep(2)
res = self.bq.dump_table(self.table_id)
self.assertEqual(3, len(res))
pprint(res)
def test_normal_from_csv(self):
filepath = os.path.dirname(os.path.abspath(__file__)) + '/data.csv'
schema = [
{ 'name': 'id', 'type': 'INTEGER', 'mode': 'REQUIRED' },
{ 'name': 'name', 'type': 'STRING', 'mode': 'REQUIRED' },
]
res = self.bq.load(self.table_id, filepath, schema=schema, skip_leading_rows=1)
self.assertTrue(bool(res))
res = self.bq.dump_table(self.table_id)
self.assertEqual(3, len(res))
pprint(res)
def test_normal_from_json(self):
filepath = os.path.dirname(os.path.abspath(__file__)) + '/data.json'
res = self.bq.load(self.table_id, filepath)
self.assertTrue(bool(res))
res = self.bq.dump_table(self.table_id)
self.assertEqual(3, len(res))
pprint(res)
def test_normal_from_obj(self):
rows = [
{ 'id': 1, 'name': 'foo' },
{ 'id': 2, 'name': 'bar', 'birth': { 'year': 2015, 'month': 10, 'day': 28 } },
{ 'id': 3, 'name': 'baz', 'url': [
'http://www.yahoo.co.jp/',
'http://www.google.co.jp/',
]}
]
res = self.bq.load(self.table_id, rows)
self.assertTrue(bool(res))
res = self.bq.dump_table(self.table_id)
self.assertEqual(3, len(res))
pprint(res)
class BigQueryTest(unittest.TestCase):
def setUp(self):
self.project_id = os.getenv('PROJECT_ID')
self.dataset_id = os.getenv('DATASET_ID', 'test_dataset')
self.table_id = os.getenv('TABLE_ID', 'test_table') + '_' + str(int(time.time()))
self.view_id = os.getenv('VIEW_ID', 'test_view') + '_' + str(int(time.time()))
if self.project_id is None:
print('PROJECT_ID is not defined.')
sys.exit(1)
self.bq = BigQuery(self.project_id)
if self.bq.exists_dataset(self.dataset_id):
self.bq.drop_dataset(self.dataset_id, delete_contents=True)
self.bq.create_dataset(self.dataset_id)
self.bq.dataset_id = self.dataset_id # Set default datasetId
#schema = [
# { 'name': 'title', 'type': 'STRING', 'mode': 'REQUIRED' },
# { 'name': 'count', 'type': 'INTEGER', 'mode': 'REQUIRED' }
#]
#self.bq.create_table(self.table_id, schema=schema)
def TearDown(self):
self.bq.drop_table(self.table_id)
self.bq.drop_dataset(self.dataset_id, delete_contents=True)
# def test_error_allow_large_results(self):
# schema = [
# { 'name': 'word', 'type': 'STRING', 'mode': 'REQUIRED' },
# { 'name': 'word_count', 'type': 'INTEGER', 'mode': 'REQUIRED' },
# ]
# self.bq.create_table(self.table_id, schema=schema)
#
# query = 'SELECT word,word_count FROM [publicdata:samples.shakespeare] LIMIT 10'
#
# # Cannnot append to required fields when allowLargeResults is True
# with self.assertRaises(BigQueryError):
# res = self.bq.insert_from_select(self.table_id, query, allow_large_results=True)
def test_error_no_required_field(self):
schema = [
{ 'name': 'title', 'type': 'STRING', 'mode': 'REQUIRED' },
{ 'name': 'unique_words', 'type': 'INTEGER', 'mode': 'REQUIRED' },
{ 'name': 'required_field', 'type': 'STRING', 'mode': 'REQUIRED' },
]
self.bq.create_table(self.table_id, schema=schema)
query = 'SELECT TOP(corpus, 10) as title, COUNT(*) as unique_words ' \
+ 'FROM [publicdata:samples.shakespeare]'
# No Required Field
with self.assertRaises(BigQueryError):
self.bq.insert_from_select(self.table_id, query)
def test_error_schema_mismatch(self):
schema = [
{ 'name': 'title', 'type': 'STRING', 'mode': 'REQUIRED' },
{ 'name': 'id', 'type': 'INTEGER', 'mode': 'REQUIRED' },
]
self.bq.create_table(self.table_id, schema=schema)
query = 'SELECT title,id FROM [publicdata:samples.wikipedia] LIMIT 10'
# Can't change type form NULLABLE to REQUIRED
with self.assertRaises(BigQueryError):
self.bq.insert_from_select(self.table_id, query)
def test_normal_allow_large_results(self):
schema = [
{ 'name': 'word', 'type': 'STRING', 'mode': 'NULLABLE' },
{ 'name': 'word_count', 'type': 'INTEGER', 'mode': 'NULLABLE' },
]
self.bq.create_table(self.table_id, schema=schema)
query = 'SELECT word,word_count FROM [publicdata:samples.shakespeare] LIMIT 10'
res = self.bq.insert_from_select(self.table_id, query, allow_large_results=True)
self.assertTrue(bool(res))
pprint(res)
res = self.bq.dump_table(self.table_id)
self.assertEqual(10, len(res))
pprint(res)
def test_normal_insert_into_exists_table(self):
schema = [
{ 'name': 'word', 'type': 'STRING', 'mode': 'REQUIRED' },
{ 'name': 'word_count', 'type': 'INTEGER', 'mode': 'REQUIRED' },
]
self.bq.create_table(self.table_id, schema=schema)
query = 'SELECT word,word_count FROM [publicdata:samples.shakespeare] LIMIT 10'
res = self.bq.insert_from_select(self.table_id, query)
self.assertTrue(bool(res))
pprint(res)
res = self.bq.dump_table(self.table_id)
self.assertEqual(10, len(res))
pprint(res)
def test_normal_with_args(self):
query = 'SELECT TOP(corpus, 10) as title, COUNT(*) as unique_words FROM shakespeare'
res = self.bq.insert_from_select(
dest_project_id=self.project_id, dest_dataset_id=self.dataset_id, dest_table_id=self.table_id, query=query,
src_project_id='publicdata', src_dataset_id='samples'
)
self.assertTrue(bool(res))
pprint(res)
res = self.bq.dump_table(self.table_id)
self.assertEqual(10, len(res))
pprint(res)
def test_normal_async(self):
query = 'SELECT TOP(corpus, 10) as title, COUNT(*) as unique_words ' \
+ 'FROM [publicdata:samples.shakespeare]'
res = self.bq.insert_from_select(self.table_id, query, async=True)
self.assertTrue(re.match(r'job_', res))
print(res)
self.bq.wait_job(res)
res = self.bq.dump_table(self.table_id)
self.assertEqual(10, len(res))
pprint(res)
def test_normal(self):
query = 'SELECT TOP(corpus, 10) as title, COUNT(*) as unique_words ' \
+ 'FROM [publicdata:samples.shakespeare]'
res = self.bq.insert_from_select(self.table_id, query)
self.assertTrue(bool(res))
pprint(res)
res = self.bq.dump_table(self.table_id)
self.assertEqual(10, len(res))
pprint(res)
class BigQueryTest(unittest.TestCase):
def setUp(self):
self.project_id = os.getenv('PROJECT_ID')
self.dataset_id = os.getenv('DATASET_ID', 'test_dataset')
self.table_id = os.getenv('TABLE_ID', 'test_table') + '_' + str(int(time.time()))
self.view_id = os.getenv('VIEW_ID', 'test_view') + '_' + str(int(time.time()))
if self.project_id is None:
print('PROJECT_ID is not defined.')
sys.exit(1)
self.bq = BigQuery(self.project_id)
if self.bq.exists_dataset(self.dataset_id):
self.bq.drop_dataset(self.dataset_id, delete_contents=True)
self.bq.create_dataset(self.dataset_id)
self.bq.dataset_id = self.dataset_id # Set default datasetId
schema = [
{ 'name': 'id', 'type': 'INTEGER', 'mode': 'REQUIRED' },
{ 'name': 'name', 'type': 'STRING', 'mode': 'REQUIRED' },
{ 'name': 'birth', 'type': 'RECORD', 'mode': 'NULLABLE', 'fields': [
{ 'name': 'year', 'type': 'INTEGER', 'mode': 'REQUIRED' },
{ 'name': 'month', 'type': 'INTEGER', 'mode': 'REQUIRED' },
{ 'name': 'day', 'type': 'INTEGER', 'mode': 'REQUIRED' },
]},
{ 'name': 'url', 'type': 'STRING', 'mode': 'REPEATED' },
]
self.bq.create_table(self.table_id, schema=schema)
def TearDown(self):
self.bq.drop_table(self.table_id)
self.bq.drop_dataset(self.dataset_id, delete_contents=True)
@staticmethod
def sigalrm_handler(x, y):
raise Exception("Timeout")
def wait_insert(self):
signal.signal(signal.SIGALRM, BigQueryTest.sigalrm_handler)
signal.alarm(120)
while True:
res = self.bq.dump_table(self.table_id)
if bool(res):
signal.alarm(0)
return res
print('sleep...')
time.sleep(2)
def test_error(self):
rows = [
{ 'id': 1, 'name': 'foo' }, # normal
{ 'id': 2 }, # missing required field
{ 'id': 'three', 'name': 'baz' }, # invalid data type
]
with self.assertRaises(BigQueryError):
self.bq.insert(self.table_id, rows, ignore_unknown_values=True)
rows = [
{ 'id': 1, 'name': 'foo' },
{ 'id': 2, 'name': 'bar', 'unknown_fields': 'unknown_value' },
]
with self.assertRaises(BigQueryError):
self.bq.insert(self.table_id, rows, skip_invalid_rows=True)
def test_normal_skip_invalid_rows(self):
rows = [
{ 'id': 1, 'name': 'foo' }, # normal
{ 'id': 2 }, # missing required field
{ 'id': 'three', 'name': 'baz' }, # invalid data type
]
res = self.bq.insert(self.table_id, rows, skip_invalid_rows=True)
self.assertTrue(bool(res))
res = self.wait_insert()
self.assertEqual(1, len(res))
pprint(res)
def test_normal_ignore_unknown_values(self):
rows = [
{ 'id': 1, 'name': 'foo' }, # normal
{ 'id': 2, 'name': 'bar', 'unknown_field': 'unknown_value' },
]
res = self.bq.insert(self.table_id, rows, ignore_unknown_values=True)
self.assertTrue(bool(res))
res = self.wait_insert()
self.assertEqual(2, len(res))
pprint(res)
def test_normal(self):
rows = [
{ 'id': 1, 'name': 'foo' },
{ 'id': 2, 'name': 'bar', 'birth': { 'year': 2015, 'month': 10, 'day': 28 } },
{ 'id': 3, 'name': 'baz', 'url': [
'http://www.yahoo.co.jp/',
'http://www.google.co.jp/',
]}
]
res = self.bq.insert(self.table_id, rows)
self.assertTrue(bool(res))
res = self.wait_insert()
self.assertEqual(3, len(res))
pprint(res)
