def test_bamboo_service(self):
# comment out when we can test or mock it differently
raise SkipTest
service_url = 'http://bamboo.io/'
service_name = 'bamboo'
xml_submission1 = os.path.join(self.this_directory, u'fixtures',
u'dhisform_submission1.xml')
xml_submission2 = os.path.join(self.this_directory, u'fixtures',
u'dhisform_submission2.xml')
xml_submission3 = os.path.join(self.this_directory, u'fixtures',
u'dhisform_submission3.xml')
# make sure xform doesnt have a bamboo dataset
self.xform.bamboo_dataset = ''
self.xform.save()
# make a first submission without the service
self._make_submission(xml_submission1)
self.assertEqual(self.response.status_code, 201)
# add rest service AFTER 1st submission
self._add_rest_service(service_url, service_name)
# submit another one.
self._make_submission(xml_submission2)
self.assertEqual(self.response.status_code, 201)
self.wait(5)
# it should have created the whole dataset
xform = XForm.objects.get(id=self.xform.id)
self.assertTrue(xform.bamboo_dataset != ''
and xform.bamboo_dataset is not None)
dataset = Dataset(connection=Connection(service_url),
dataset_id=xform.bamboo_dataset)
self.assertEqual(dataset.get_info()['num_rows'], 2)
# submit a third one. check that we have 3 records
self._make_submission(xml_submission3)
self.assertEqual(self.response.status_code, 201)
self.wait(5)
self.assertEqual(dataset.get_info()['num_rows'], 3)
# test regeneration
dsi = dataset.get_info()
regen_url = reverse(link_to_bamboo,
kwargs={
'username': self.user.username,
'id_string': self.xform.id_string
})
response = self.client.post(regen_url, {})
# deleting DS redirects to profile page
self.assertEqual(response.status_code, 302)
self.wait(5)
xform = XForm.objects.get(id=self.xform.id)
self.assertTrue(xform.bamboo_dataset)
dataset = Dataset(connection=Connection(service_url),
dataset_id=xform.bamboo_dataset)
new_dsi = dataset.get_info()
self.assertEqual(new_dsi['num_rows'], dsi['num_rows'])
self.assertNotEqual(new_dsi['id'], dsi['id'])
def test_bamboo_service(self):
service_url = 'http://bamboo.io/'
service_name = 'bamboo'
# self._add_rest_service(service_url, service_name)
self.wait(2)
xml_submission1 = os.path.join(self.this_directory,
u'fixtures',
u'dhisform_submission1.xml')
xml_submission2 = os.path.join(self.this_directory,
u'fixtures',
u'dhisform_submission2.xml')
xml_submission3 = os.path.join(self.this_directory,
u'fixtures',
u'dhisform_submission3.xml')
# make a first submission without the service
self._make_submission(xml_submission1)
self.assertEqual(self.response.status_code, 201)
# add rest service AFTER 1st submission
self._add_rest_service(service_url, service_name)
# submit another one.
self._make_submission(xml_submission2)
self.assertEqual(self.response.status_code, 201)
self.wait(3)
# it should have created the whole dataset
xform = XForm.objects.get(id=self.xform.id)
self.assertTrue(xform.bamboo_dataset)
dataset = Dataset(connection=Connection(service_url),
dataset_id=xform.bamboo_dataset)
self.assertEqual(dataset.get_info()['num_rows'], 2)
# submit a third one. check that we have 3 records
self._make_submission(xml_submission3)
self.assertEqual(self.response.status_code, 201)
self.wait(3)
self.assertEqual(dataset.get_info()['num_rows'], 3)
# test regeneration
dsi = dataset.get_info()
regen_url = reverse(link_to_bamboo, kwargs={
'username': self.user.username,
'id_string': self.xform.id_string
})
response = self.client.post(regen_url, {})
# deleting DS redirects to profile page
self.assertEqual(response.status_code, 302)
self.wait(3)
xform = XForm.objects.get(id=self.xform.id)
self.assertTrue(xform.bamboo_dataset)
dataset = Dataset(connection=Connection(service_url),
dataset_id=xform.bamboo_dataset)
new_dsi = dataset.get_info()
self.assertEqual(new_dsi['num_rows'], dsi['num_rows'])
self.assertNotEqual(new_dsi['id'], dsi['id'])
def run_test_suite(dataset_file_path_list):
print "running test suite for %s" % " ".join(dataset_file_path_list)
alldata = []
for dataset_name in dataset_file_path_list:
d = {}
d['hostname'] = os.uname()[1]
d['bamboo_url'] = URL
d['unix_time'] = time.time()
conn = Connection(url=URL)
d['commit'] = conn.version['commit']
d['branch'] = conn.version['branch']
dataset = Dataset(connection=conn, path=dataset_name)
d['import_time'] = time_till_import_is_finished(dataset)
info = dataset.get_info()
d['row'] = info['num_rows']
d['col'] = info['num_columns']
d['add_1_calculations_time'] = time_to_add_1_calculations(dataset)
d['add_5_calculations_1by1_time'] = time_to_add_5_calculations_1by1(dataset)
d['add_5_calculations_batch_time'] = time_to_add_5_calculations_batch(dataset)
d['update_1_time'] = time_to_add_1_update(dataset)
d['update_5_1by1_time'] = time_to_add_5_update_1by1(dataset)
d['update_5_batch_time'] = time_to_add_5_update_batch(dataset)
dataset.delete()
alldata.append(d)
return alldata
def run_test_suite(dataset_file_path_list):
print "running test suite for %s" % " ".join(dataset_file_path_list)
alldata = []
for dataset_name in dataset_file_path_list:
d = {}
d['hostname'] = os.uname()[1]
d['bamboo_url'] = URL
d['unix_time'] = time.time()
conn = Connection(url=URL)
d['commit'] = conn.version['commit']
d['branch'] = conn.version['branch']
dataset = Dataset(connection=conn, path=dataset_name)
d['import_time'] = time_till_import_is_finished(dataset)
info = dataset.get_info()
d['row'] = info['num_rows']
d['col'] = info['num_columns']
d['add_1_calculations_time'] = time_to_add_1_calculations(dataset)
d['add_5_calculations_1by1_time'] = time_to_add_5_calculations_1by1(
dataset)
d['add_5_calculations_batch_time'] = time_to_add_5_calculations_batch(
dataset)
d['update_1_time'] = time_to_add_1_update(dataset)
d['update_5_1by1_time'] = time_to_add_5_update_1by1(dataset)
d['update_5_batch_time'] = time_to_add_5_update_batch(dataset)
dataset.delete()
alldata.append(d)
return alldata
DATASET_ID = sys.argv[-1]
CSV_URL = u"https://github.com/modilabs/bamboo-examples/raw/master/data/water_points.csv"
WAIT_INTERVAL = 10
WAIT_MAX = 180
if len(sys.argv) > 1 and DATASET_ID:
print("Retrieving dataset from UUID %s" % DATASET_ID)
dataset = Dataset(dataset_id=DATASET_ID, connection=connect)
else:
print(u"Creating dataset from %s" % CSV_URL)
dataset = Dataset(url=CSV_URL, connection=connect)
waited = 0
while(dataset.get_info().get(u'state', 'ready') != 'ready'):
if waited >= WAIT_MAX:
print(u"Unable to get dataset ready in time. Exiting.")
sys.exit()
print(u"Dataset not ready. Waiting 10s...")
time.sleep(WAIT_INTERVAL)
waited += WAIT_INTERVAL
print(u"---")
print(dataset.get_info())
print(u"---")
print(u"Creating calculations")
print dataset.add_aggregation(formula='water_points=count()', groups=['district'])
print dataset.add_aggregation(formula='district_pop=sum(community_pop)', groups=['district'])
class TestDataset(TestBase):
def setUp(self):
TestBase.setUp(self)
self._create_dataset_from_file()
def _create_dataset_from_file(self):
self.dataset = Dataset(path=self.CSV_FILE, connection=self.connection)
self.wait()
def _create_aux_dataset_from_file(self):
self.aux_dataset = Dataset(path=self.AUX_CSV_FILE,
connection=self.connection)
self.wait()
def _wait_for_dataset_ready(self):
while self.dataset.state == 'pending':
self.wait()
def test_create_dataset_from_json(self):
dataset = Dataset(path=self.JSON_FILE,
data_format='json',
connection=self.connection)
self.assertTrue(dataset.id is not None)
self._cleanup(dataset)
def test_create_dataset_from_schema(self):
dataset = Dataset(schema_path=self.SCHEMA_FILE,
connection=self.connection)
self.assertTrue(dataset.id is not None)
self._cleanup(dataset)
# schema string
schema_str = open(self.SCHEMA_FILE).read()
dataset = Dataset(schema_content=schema_str,
connection=self.connection)
self.assertTrue(dataset.id is not None)
self._cleanup(dataset)
def test_create_dataset_from_schema_with_data(self):
# schema + JSON data
dataset = Dataset(path=self.JSON_FILE,
data_format='json',
schema_path=self.SCHEMA_FILE,
connection=self.connection)
self.assertTrue(dataset.id is not None)
self._cleanup(dataset)
# schema + CSV data
dataset = Dataset(path=self.CSV_FILE,
data_format='csv',
schema_path=self.SCHEMA_FILE,
connection=self.connection)
self.assertTrue(dataset.id is not None)
self._cleanup(dataset)
def test_create_dataset_default_connection(self):
dataset = Dataset(path=self.CSV_FILE,
connection=self.default_connection)
self._cleanup(dataset)
def test_create_dataset_no_info(self):
with self.assertRaises(PyBambooException):
Dataset()
def test_create_dataset_bad_data_format(self):
with self.assertRaises(PyBambooException):
Dataset(path=self.CSV_FILE,
data_format='BAD',
connection=self.connection)
def test_create_dataset_from_file(self):
# created in TestDataset.setUp()
self.assertTrue(self.dataset.id is not None)
def test_create_dataset_from_url(self):
dataset = Dataset(
url='http://formhub.org/mberg/forms/good_eats/data.csv',
connection=self.connection)
self.assertTrue(self.dataset.id is not None)
self._cleanup(dataset)
def test_reset_dataset(self):
dataset_id = self.dataset._id
self.dataset.reset(path=self.CSV_FILE, connection=self.connection)
self.assertEqual(self.dataset._id, dataset_id)
def test_reset_dataset_no_dataset_id(self):
self.dataset.delete()
with self.assertRaises(PyBambooException):
self.dataset.reset()
def test_na_values(self):
dataset = Dataset(path=self.CSV_FILE,
connection=self.connection,
na_values=['n/a'])
self.wait()
first_row = dataset.get_data(query={
'food_type': 'street_meat',
'amount': 2,
'rating': 'delectible',
'risk_factor': 'low_risk'
},
limit=1)[-1]
self.assertEqual(first_row.get('comments'), 'null')
self._cleanup(dataset)
def test_resample(self):
data = self.dataset.resample(date_column='submit_date',
interval='D',
how='mean')
self.assertTrue(data)
def test_resample_with_query(self):
data = self.dataset.resample(date_column='submit_date',
interval='D',
query={"food_type": "street_meat"},
how='sum')
self.assertTrue(data)
def test_rolling(self):
data = self.dataset.rolling(win_type='boxcar', window=3)
self.assertTrue(isinstance(data, list))
def test_set_info(self):
description = u"Meals rating worldwide"
attribution = u"mberg"
label = u"Good Eats"
license = u"Public Domain"
self.dataset.set_info(attribution=attribution,
description=description,
label=label,
license=license)
infos = self.dataset.get_info()
self.assertEqual(infos['description'], description)
self.assertEqual(infos['attribution'], attribution)
self.assertEqual(infos['label'], label)
self.assertEqual(infos['license'], license)
def test_index_present(self):
data = self.dataset.get_data(index=True)
self.assertTrue('index' in data[-1].keys())
def test_str(self):
self.assertEqual(str(self.dataset), self.dataset.id)
def test_version(self):
self.assert_keys_in_dict(self.VERSION_KEYS, self.dataset.version)
def test_columns(self):
self.wait() # have to wait, bamboo issue #284
cols = self.dataset.columns
keys = self.dataset.get_info()['schema'].keys()
for key in keys:
self.assertTrue(key in cols)
for col in cols:
self.assertTrue(col in keys)
def test_state(self):
self.assertEqual(self.dataset.state, 'ready')
def test_num_columns(self):
self.assertEqual(self.dataset.num_columns, 15)
def test_num_rows(self):
self.assertEqual(self.dataset.num_rows, 19)
def test_count(self):
self.wait()
count = self.dataset.count(field='food_type', method='count')
self.assertEqual(count, 19)
def test_data_count(self):
self._wait_for_dataset_ready() # TODO: is this necessary?
count = self.dataset.get_data(count=True)
self.assertEqual(count, 19)
def test_delete_dataset(self):
self.dataset.delete()
self.assertTrue(self.dataset._id is None)
def test_invalid_dataset(self):
self.dataset.delete()
with self.assertRaises(PyBambooException):
self.dataset.delete()
def test_add_calculation(self):
result = self.dataset.add_calculation(name='double_amount',
formula='amount * 2')
self.assertTrue(result)
def test_add_calculations(self):
formulae = [
{
'name': 'double_amount',
'formula': 'amount * 2'
},
{
'name': 'triple_amount',
'formula': 'amount * 3'
},
]
result = self.dataset.add_calculations(json=formulae)
self.assertTrue(result)
def test_add_invalid_calculation_a_priori(self):
bad_calcs = [
{
'name': None,
'formula': 'ok'
},
{
'name': 'number',
'formula': 3
},
{
'name': 'number',
'formula': 'ok',
'groups': 3
},
]
for calc in bad_calcs:
with self.assertRaises(PyBambooException):
self.dataset.add_calculation(**calc)
with self.assertRaises(PyBambooException):
self.dataset.add_calculations()
def test_add_invalid_calculation_a_posteriori(self):
result = self.dataset.add_calculation(name='double_amount',
formula='BAD')
self.assertEqual(result, False)
def test_add_aggregation(self):
result = self.dataset.add_calculation(name='sum_amount',
formula='sum(amount)')
self.assertTrue(result)
self.dataset.has_aggs_to_remove = True
def test_add_aggregation_with_groups(self):
result = self.dataset.add_calculation(name='sum_amount',
formula='sum(amount)',
groups=['food_type'])
self.assertTrue(result)
result = self.dataset.add_calculation(name='sum_amount',
formula='sum(amount)',
groups=['food_type', 'rating'])
self.assertTrue(result)
self.dataset.has_aggs_to_remove = True
def test_add_aggregation_invalid_groups(self):
with self.assertRaises(PyBambooException):
self.dataset.add_calculation(name='sum_amount',
formula='sum(amount)',
groups='BAD')
def test_remove_calculation(self):
name = 'double_amount'
self.dataset.add_calculation(name=name, formula='amount * 2')
result = self.dataset.remove_calculation(name)
self.assertTrue(result)
def test_remove_aggregation(self):
name = 'sum_amount'
result = self.dataset.add_calculation(name=name, formula='sum(amount)')
self.assertTrue(result)
result = self.dataset.remove_calculation(name)
self.assertTrue(result)
self.dataset.has_aggs_to_remove = True
def test_remove_calculation_fail(self):
result = self.dataset.remove_calculation('bad')
self.assertFalse(result)
def test_get_calculations(self):
calc_keys = ['state', 'formula', 'group', 'name']
result = self.dataset.add_calculation(name='double_amount',
formula='amount * 2')
self.assertEqual(result, True)
result = self.dataset.get_calculations()
self.assertTrue(isinstance(result, list))
for calc in result:
self.assertTrue(isinstance(calc, dict))
keys = calc.keys()
for key in calc_keys:
self.assertTrue(key in keys)
self.assertEqual(result[0]['state'], 'pending')
self.wait()
self.wait()
result = self.dataset.get_calculations()
self.assertEqual(result[0]['state'], 'ready')
def test_get_aggregate_datasets(self):
result = self.dataset.get_aggregate_datasets()
self.assertTrue(isinstance(result, dict))
self.assertEqual(len(result), 0)
self.dataset.add_calculation(name='sum_amount', formula='sum(amount)')
self.wait()
self.wait()
result = self.dataset.get_aggregate_datasets()
self.assertTrue(isinstance(result, dict))
self.assertEqual(len(result), 1)
self.assertTrue('' in result.keys())
self.assertTrue(isinstance(result[''], Dataset))
self.dataset.add_calculation(name='sum_amount',
formula='sum(amount)',
groups=['food_type'])
self.wait()
self.wait()
result = self.dataset.get_aggregate_datasets()
self.assertTrue(isinstance(result, dict))
self.assertEqual(len(result), 2)
self.assertTrue('food_type' in result.keys())
self.assertTrue(isinstance(result['food_type'], Dataset))
self.dataset.has_aggs_to_remove = True
def test_get_aggregate_datasets_no_aggregations(self):
result = self.dataset.get_aggregate_datasets()
self.assertTrue(isinstance(result, dict))
self.assertEqual(len(result), 0)
def test_get_summary(self):
self.wait() # TODO: remove (bamboo issue #276)
result = self.dataset.get_summary()
self.assertTrue(isinstance(result, dict))
cols = self.dataset.columns
keys = result.keys()
for col in cols:
self.assertTrue(col in keys)
def test_get_summary_with_select(self):
self.wait() # TODO: remove (bamboo issue #276)
result = self.dataset.get_summary(select=['food_type'])
self.assertEqual(len(result), 1)
self.assertTrue('food_type' in result.keys())
result = self.dataset.get_summary(select=['food_type', 'rating'])
self.assertEqual(len(result), 2)
result_keys = result.keys()
self.assertTrue('food_type' in result_keys)
self.assertTrue('food_type' in result_keys)
def test_get_summary_bad_select(self):
with self.assertRaises(PyBambooException):
self.dataset.get_summary(select='BAD')
def test_get_summary_with_query(self):
self.wait() # TODO: remove (bamboo issue #276)
self.dataset.get_summary(query={'food_type': 'lunch'})
def test_get_summary_bad_query(self):
with self.assertRaises(PyBambooException):
self.dataset.get_summary(query='BAD')
def test_get_summary_with_groups(self):
self.wait() # TODO: remove (bamboo issue #276)
result = self.dataset.get_summary(groups=['food_type'])
self.assertEqual(len(result), 1)
values = self.dataset.get_summary(
select=['food_type'])['food_type']['summary'].keys()
self.assertTrue('food_type' in result.keys())
self.assertTrue(isinstance(result['food_type'], dict))
keys = result['food_type'].keys()
for val in values:
self.assertTrue(val in keys)
def test_get_summary_bad_groups(self):
with self.assertRaises(PyBambooException):
self.dataset.get_summary(groups='BAD')
def test_get_info(self):
info_keys = [
'attribution',
'description',
'license',
'created_at',
'updated_at',
'label',
'num_columns',
'num_rows',
'id',
'schema',
]
schema_keys = [
'simpletype',
'olap_type',
'label',
]
self.wait() # have to wait, bamboo issue #284
result = self.dataset.get_info()
self.assertTrue(isinstance(result, dict))
for key in info_keys:
self.assertTrue(key in result.keys())
self.assertEqual(result['num_columns'], 15)
self.assertEqual(result['num_rows'], 19)
schema = result['schema']
self.assertTrue(isinstance(schema, dict))
self.assertEqual(len(schema.keys()), 15)
for col_name, col_info in schema.iteritems():
for key in schema_keys:
self.assertTrue(key in col_info.keys())
def test_get_data(self):
self.wait()
result = self.dataset.get_data()
self.assertTrue(isinstance(result, list))
self.assertEqual(len(result), 19)
def test_get_data_with_select(self):
self.wait()
result = self.dataset.get_data(select=['food_type', 'amount'])
self.assertEqual(len(result), 19)
for row in result:
self.assertEqual(len(row), 2)
cols = row.keys()
self.assertTrue('food_type' in cols)
self.assertTrue('amount' in cols)
def test_get_data_with_query(self):
self.wait() # TODO: remove (bamboo issue #285)
result = self.dataset.get_data(query={'food_type': 'lunch'})
self.assertEqual(len(result), 7)
def test_get_data_with_select_and_query(self):
self.wait() # TODO: remove (bamboo issue #285)
result = self.dataset.get_data(select=['food_type', 'amount'],
query={'food_type': 'lunch'})
self.assertEqual(len(result), 7)
for row in result:
self.assertEqual(len(row), 2)
cols = row.keys()
self.assertTrue('food_type' in cols)
self.assertTrue('amount' in cols)
def test_get_data_with_format(self):
self.wait() # TODO: remove (bamboo issue #285)
result = self.dataset.get_data(format='csv')
self.assertTrue(isinstance(result, basestring))
def test_get_data_invalid_select(self):
with self.assertRaises(PyBambooException):
self.dataset.get_data(select='BAD')
def test_get_data_invalid_query(self):
with self.assertRaises(PyBambooException):
self.dataset.get_data(query='BAD')
def test_get_data_with_invalid_format(self):
with self.assertRaises(PyBambooException):
self.dataset.get_data(format='BAD')
def test_get_data_bad_query(self):
self.wait() # TODO: remove (bamboo issue #285)
result = self.dataset.get_data(query={'BAD': 'BAD'})
self.assertFalse(result)
def test_update_data(self):
row = {
'food_type': 'morning_food',
'amount': 10.0,
'risk_factor': 'high_risk',
'rating': 'delectible',
}
result = self.dataset.update_data([row])
self.wait(15)
result = self.dataset.get_data()
self.assertTrue(isinstance(result, list))
self.assertEqual(len(result), 20)
def test_update_data_no_data(self):
with self.assertRaises(PyBambooException):
self.dataset.update_data([])
def test_update_data_bad_data(self):
bad_rows = [{}, [[]], [{'exception': Exception()}]]
for rows in bad_rows:
with self.assertRaises(PyBambooException):
self.dataset.update_data(rows)
def test_merge(self):
# already have one dataset in self.dataset
dataset = Dataset(path=self.CSV_FILE, connection=self.connection)
result = Dataset.merge([self.dataset, dataset],
connection=self.connection)
self.assertTrue(isinstance(result, Dataset))
self._cleanup(dataset)
self._cleanup(result)
def test_merge_default_connection(self):
dataset = Dataset(path=self.CSV_FILE,
connection=self.default_connection)
other_dataset = Dataset(path=self.CSV_FILE,
connection=self.default_connection)
result = Dataset.merge([dataset, other_dataset])
self.assertTrue(isinstance(result, Dataset))
self._cleanup(dataset)
self._cleanup(other_dataset)
self._cleanup(result)
def test_merge_bad_datasets(self):
dataset = {}
other_dataset = []
with self.assertRaises(PyBambooException):
Dataset.merge([dataset, other_dataset], connection=self.connection)
def test_merge_fail(self):
other_dataset = Dataset('12345', connection=self.connection)
result = Dataset.merge([self.dataset, other_dataset],
connection=self.connection)
self.assertFalse(result)
def test_join(self):
self._create_aux_dataset_from_file()
self.wait()
result = Dataset.join(self.dataset,
self.aux_dataset,
'food_type',
connection=self.connection)
self.assertTrue(isinstance(result, Dataset))
self._cleanup(result)
def test_join_default_connection(self):
dataset = Dataset(path=self.CSV_FILE,
connection=self.default_connection)
aux_dataset = Dataset(path=self.AUX_CSV_FILE,
connection=self.default_connection)
self.wait()
result = Dataset.join(dataset, aux_dataset, 'food_type')
self.wait()
self.assertTrue(isinstance(result, Dataset))
self._cleanup(dataset)
self._cleanup(aux_dataset)
self._cleanup(result)
def test_join_bad_other_dataset(self):
with self.assertRaises(PyBambooException):
Dataset.join(self.dataset,
Exception(),
'food_type',
connection=self.connection)
def test_join_bad_on(self):
self._create_aux_dataset_from_file()
self.wait()
result = Dataset.join(self.dataset,
self.aux_dataset,
'BAD',
connection=self.connection)
self.assertFalse(result)
# /row/INDEX tests.
def test_get_row(self):
self.assertEqual(
self.dataset.get_row(0)['comments'], u"Try the yogurt drink")
def test_update_row(self):
index = 2
comment = 'test'
self.dataset.update_row(index, {'comments': comment})
self.assertEqual(self.dataset.get_row(index)['comments'], comment)
def test_delete_row(self):
self._wait_for_dataset_ready() # TODO: is this necessary?
index = 10
self.dataset.delete_row(index=index)
result = self.dataset.get_row(index)
self.assertTrue('error' in result)
with open(bamboo_id_file) as f:
bamboo_ids = json.loads(f.read())
if not bamboo_ids:
print '"%s" not found: exiting' % bamboo_id_file
sys.exit(0)
print 'current dataset status:'
print json.dumps(bamboo_ids, indent=4, sort_keys=True)
# upload originals
for sector in bamboo_ids.keys():
for name, id in bamboo_ids[sector]['originals'].iteritems():
if not id:
print 'dataset: %s not uploaded, uploading %s.csv' % (name, name)
dataset = Dataset(connection=connection,
path='csvs/originals/%s.csv' % name)
state = dataset.get_info()['state']
while state != 'ready':
time.sleep(1)
state = dataset.get_info()['state']
print state
bamboo_ids[sector]['originals'][name] = dataset.id
with open(bamboo_id_file, 'wb') as f:
f.write(json.dumps(bamboo_ids))
# merge originals
for sector in bamboo_ids.keys():
if not bamboo_ids[sector]['merged']:
print 'no merged dataset for sector: %s' % sector
datasets = [
Dataset(connection=connection, dataset_id=id)
for name, id in bamboo_ids[sector]['originals'].iteritems()
# get state of current datasets
with open(bamboo_id_file) as f:
bamboo_ids = json.loads(f.read())
if not bamboo_ids:
print '"%s" not found: exiting' % bamboo_id_file
sys.exit(0)
print 'current dataset status:'
print json.dumps(bamboo_ids, indent=4, sort_keys=True)
# upload originals
for sector in bamboo_ids.keys():
for name, id in bamboo_ids[sector]['originals'].iteritems():
if not id:
print 'dataset: %s not uploaded, uploading %s.csv' % (name, name)
dataset = Dataset(connection=connection, path='csvs/originals/%s.csv' % name)
state = dataset.get_info()['state']
while state != 'ready':
time.sleep(1)
state = dataset.get_info()['state']
print state
bamboo_ids[sector]['originals'][name] = dataset.id
with open(bamboo_id_file, 'wb') as f:
f.write(json.dumps(bamboo_ids))
# merge originals
for sector in bamboo_ids.keys():
if not bamboo_ids[sector]['merged']:
print 'no merged dataset for sector: %s' % sector
datasets = [Dataset(connection=connection, dataset_id=id) for name, id
in bamboo_ids[sector]['originals'].iteritems()]
print 'merging datasets: %s' % [dataset.id for dataset in datasets]
#! /usr/bin/env python
import time
from pybamboo.dataset import Dataset
d = Dataset(path='data/water_points.csv')
print 'dataset id: %s' % d.id
time.sleep(2)
info = d.get_info()
print 'num_columns: %s' % info['num_columns']
print 'num_rows: %s' % info['num_rows']
print 'columns: %s' % info['schema'].keys()
water_points = d.get_data(select=['communities_villages','water_functioning'])
print 'return all water points where water is functioning %s' % water_points
broken_water_points =d.get_data(select=['communities_villages'],query={'water_functioning': False})
#print 'Villages with broken water points %s' % len(broken_water_points)
summary = d.get_summary()
print 'Water point lift mechanism type summary %s' % summary['water_lift_mechanism_type']
class TestDataset(TestBase):
def setUp(self):
TestBase.setUp(self)
self._create_dataset_from_file()
def _create_dataset_from_file(self):
self.dataset = Dataset(path=self.CSV_FILE,
connection=self.connection)
self.wait()
def _create_aux_dataset_from_file(self):
self.aux_dataset = Dataset(path=self.AUX_CSV_FILE,
connection=self.connection)
self.wait()
def _wait_for_dataset_ready(self):
while self.dataset.state == 'pending':
self.wait()
def test_create_dataset_from_json(self):
dataset = Dataset(path=self.JSON_FILE, data_format='json',
connection=self.connection)
self.assertTrue(dataset.id is not None)
self._cleanup(dataset)
def test_create_dataset_from_schema(self):
dataset = Dataset(schema_path=self.SCHEMA_FILE,
connection=self.connection)
self.assertTrue(dataset.id is not None)
self._cleanup(dataset)
# schema string
schema_str = open(self.SCHEMA_FILE).read()
dataset = Dataset(schema_content=schema_str,
connection=self.connection)
self.assertTrue(dataset.id is not None)
self._cleanup(dataset)
def test_create_dataset_from_schema_with_data(self):
# schema + JSON data
dataset = Dataset(path=self.JSON_FILE, data_format='json',
schema_path=self.SCHEMA_FILE,
connection=self.connection)
self.assertTrue(dataset.id is not None)
self._cleanup(dataset)
# schema + CSV data
dataset = Dataset(path=self.CSV_FILE, data_format='csv',
schema_path=self.SCHEMA_FILE,
connection=self.connection)
self.assertTrue(dataset.id is not None)
self._cleanup(dataset)
def test_create_dataset_default_connection(self):
dataset = Dataset(path=self.CSV_FILE,
connection=self.default_connection)
self._cleanup(dataset)
def test_create_dataset_no_info(self):
with self.assertRaises(PyBambooException):
Dataset()
def test_create_dataset_bad_data_format(self):
with self.assertRaises(PyBambooException):
Dataset(path=self.CSV_FILE, data_format='BAD',
connection=self.connection)
def test_create_dataset_from_file(self):
# created in TestDataset.setUp()
self.assertTrue(self.dataset.id is not None)
def test_create_dataset_from_url(self):
dataset = Dataset(
url='http://formhub.org/mberg/forms/good_eats/data.csv',
connection=self.connection)
self.assertTrue(self.dataset.id is not None)
self._cleanup(dataset)
def test_reset_dataset(self):
dataset_id = self.dataset._id
self.dataset.reset(path=self.CSV_FILE,
connection=self.connection)
self.assertEqual(self.dataset._id, dataset_id)
def test_reset_dataset_no_dataset_id(self):
self.dataset.delete()
with self.assertRaises(PyBambooException):
self.dataset.reset()
def test_na_values(self):
dataset = Dataset(
path=self.CSV_FILE,
connection=self.connection,
na_values=['n/a'])
self.wait()
first_row = dataset.get_data(query={'food_type': 'street_meat',
'amount': 2,
'rating': 'delectible',
'risk_factor': 'low_risk'},
limit=1)[-1]
self.assertEqual(first_row.get('comments'), 'null')
self._cleanup(dataset)
def test_resample(self):
data = self.dataset.resample(date_column='submit_date',
interval='D',
how='mean')
self.assertTrue(data)
def test_resample_with_query(self):
data = self.dataset.resample(date_column='submit_date',
interval='D',
query={"food_type": "street_meat"},
how='sum')
self.assertTrue(data)
def test_rolling(self):
data = self.dataset.rolling(win_type='boxcar', window=3)
self.assertTrue(isinstance(data, list))
def test_set_info(self):
description = u"Meals rating worldwide"
attribution = u"mberg"
label = u"Good Eats"
license = u"Public Domain"
self.dataset.set_info(attribution=attribution,
description=description,
label=label,
license=license)
infos = self.dataset.get_info()
self.assertEqual(infos['description'], description)
self.assertEqual(infos['attribution'], attribution)
self.assertEqual(infos['label'], label)
self.assertEqual(infos['license'], license)
def test_index_present(self):
data = self.dataset.get_data(index=True)
self.assertTrue('index' in data[-1].keys())
def test_str(self):
self.assertEqual(str(self.dataset), self.dataset.id)
def test_version(self):
self.assert_keys_in_dict(self.VERSION_KEYS,
self.dataset.version)
def test_columns(self):
self.wait() # have to wait, bamboo issue #284
cols = self.dataset.columns
keys = self.dataset.get_info()['schema'].keys()
for key in keys:
self.assertTrue(key in cols)
for col in cols:
self.assertTrue(col in keys)
def test_state(self):
self.assertEqual(self.dataset.state, 'ready')
def test_num_columns(self):
self.assertEqual(self.dataset.num_columns, 15)
def test_num_rows(self):
self.assertEqual(self.dataset.num_rows, 19)
def test_count(self):
self.wait()
count = self.dataset.count(field='food_type', method='count')
self.assertEqual(count, 19)
def test_data_count(self):
self._wait_for_dataset_ready() # TODO: is this necessary?
count = self.dataset.get_data(count=True)
self.assertEqual(count, 19)
def test_delete_dataset(self):
self.dataset.delete()
self.assertTrue(self.dataset._id is None)
def test_invalid_dataset(self):
self.dataset.delete()
with self.assertRaises(PyBambooException):
self.dataset.delete()
def test_add_calculation(self):
result = self.dataset.add_calculation(name='double_amount',
formula='amount * 2')
self.assertTrue(result)
def test_add_calculations(self):
formulae = [
{'name': 'double_amount', 'formula': 'amount * 2'},
{'name': 'triple_amount', 'formula': 'amount * 3'},
]
result = self.dataset.add_calculations(json=formulae)
self.assertTrue(result)
def test_add_invalid_calculation_a_priori(self):
bad_calcs = [
{'name': None, 'formula': 'ok'},
{'name': 'number', 'formula': 3},
{'name': 'number', 'formula': 'ok', 'groups': 3},
]
for calc in bad_calcs:
with self.assertRaises(PyBambooException):
self.dataset.add_calculation(**calc)
with self.assertRaises(PyBambooException):
self.dataset.add_calculations()
def test_add_invalid_calculation_a_posteriori(self):
result = self.dataset.add_calculation(name='double_amount',
formula='BAD')
self.assertEqual(result, False)
def test_add_aggregation(self):
result = self.dataset.add_calculation(name='sum_amount',
formula='sum(amount)')
self.assertTrue(result)
self.dataset.has_aggs_to_remove = True
def test_add_aggregation_with_groups(self):
result = self.dataset.add_calculation(
name='sum_amount',
formula='sum(amount)',
groups=['food_type'])
self.assertTrue(result)
result = self.dataset.add_calculation(
name='sum_amount',
formula='sum(amount)',
groups=['food_type', 'rating'])
self.assertTrue(result)
self.dataset.has_aggs_to_remove = True
def test_add_aggregation_invalid_groups(self):
with self.assertRaises(PyBambooException):
self.dataset.add_calculation(
name='sum_amount',
formula='sum(amount)',
groups='BAD')
def test_remove_calculation(self):
name = 'double_amount'
self.dataset.add_calculation(name=name,
formula='amount * 2')
result = self.dataset.remove_calculation(name)
self.assertTrue(result)
def test_remove_aggregation(self):
name = 'sum_amount'
result = self.dataset.add_calculation(name=name,
formula='sum(amount)')
self.assertTrue(result)
result = self.dataset.remove_calculation(name)
self.assertTrue(result)
self.dataset.has_aggs_to_remove = True
def test_remove_calculation_fail(self):
result = self.dataset.remove_calculation('bad')
self.assertFalse(result)
def test_get_calculations(self):
calc_keys = ['state', 'formula', 'group', 'name']
result = self.dataset.add_calculation(name='double_amount',
formula='amount * 2')
self.assertEqual(result, True)
result = self.dataset.get_calculations()
self.assertTrue(isinstance(result, list))
for calc in result:
self.assertTrue(isinstance(calc, dict))
keys = calc.keys()
for key in calc_keys:
self.assertTrue(key in keys)
self.assertEqual(result[0]['state'], 'pending')
self.wait()
self.wait()
result = self.dataset.get_calculations()
self.assertEqual(result[0]['state'], 'ready')
def test_get_aggregate_datasets(self):
result = self.dataset.get_aggregate_datasets()
self.assertTrue(isinstance(result, dict))
self.assertEqual(len(result), 0)
self.dataset.add_calculation(name='sum_amount',
formula='sum(amount)')
self.wait()
self.wait()
result = self.dataset.get_aggregate_datasets()
self.assertTrue(isinstance(result, dict))
self.assertEqual(len(result), 1)
self.assertTrue('' in result.keys())
self.assertTrue(isinstance(result[''], Dataset))
self.dataset.add_calculation(
name='sum_amount', formula='sum(amount)', groups=['food_type'])
self.wait()
self.wait()
result = self.dataset.get_aggregate_datasets()
self.assertTrue(isinstance(result, dict))
self.assertEqual(len(result), 2)
self.assertTrue('food_type' in result.keys())
self.assertTrue(isinstance(result['food_type'], Dataset))
self.dataset.has_aggs_to_remove = True
def test_get_aggregate_datasets_no_aggregations(self):
result = self.dataset.get_aggregate_datasets()
self.assertTrue(isinstance(result, dict))
self.assertEqual(len(result), 0)
def test_get_summary(self):
self.wait() # TODO: remove (bamboo issue #276)
result = self.dataset.get_summary()
self.assertTrue(isinstance(result, dict))
cols = self.dataset.columns
keys = result.keys()
for col in cols:
self.assertTrue(col in keys)
def test_get_summary_with_select(self):
self.wait() # TODO: remove (bamboo issue #276)
result = self.dataset.get_summary(select=['food_type'])
self.assertEqual(len(result), 1)
self.assertTrue('food_type' in result.keys())
result = self.dataset.get_summary(select=['food_type', 'rating'])
self.assertEqual(len(result), 2)
result_keys = result.keys()
self.assertTrue('food_type' in result_keys)
self.assertTrue('food_type' in result_keys)
def test_get_summary_bad_select(self):
with self.assertRaises(PyBambooException):
self.dataset.get_summary(select='BAD')
def test_get_summary_with_query(self):
self.wait() # TODO: remove (bamboo issue #276)
self.dataset.get_summary(query={'food_type': 'lunch'})
def test_get_summary_bad_query(self):
with self.assertRaises(PyBambooException):
self.dataset.get_summary(query='BAD')
def test_get_summary_with_groups(self):
self.wait() # TODO: remove (bamboo issue #276)
result = self.dataset.get_summary(groups=['food_type'])
self.assertEqual(len(result), 1)
values = self.dataset.get_summary(
select=['food_type'])['food_type']['summary'].keys()
self.assertTrue('food_type' in result.keys())
self.assertTrue(isinstance(result['food_type'], dict))
keys = result['food_type'].keys()
for val in values:
self.assertTrue(val in keys)
def test_get_summary_bad_groups(self):
with self.assertRaises(PyBambooException):
self.dataset.get_summary(groups='BAD')
def test_get_info(self):
info_keys = [
'attribution',
'description',
'license',
'created_at',
'updated_at',
'label',
'num_columns',
'num_rows',
'id',
'schema',
]
schema_keys = [
'simpletype',
'olap_type',
'label',
]
self.wait() # have to wait, bamboo issue #284
result = self.dataset.get_info()
self.assertTrue(isinstance(result, dict))
for key in info_keys:
self.assertTrue(key in result.keys())
self.assertEqual(result['num_columns'], 15)
self.assertEqual(result['num_rows'], 19)
schema = result['schema']
self.assertTrue(isinstance(schema, dict))
self.assertEqual(len(schema.keys()), 15)
for col_name, col_info in schema.iteritems():
for key in schema_keys:
self.assertTrue(key in col_info.keys())
def test_get_data(self):
self.wait()
result = self.dataset.get_data()
self.assertTrue(isinstance(result, list))
self.assertEqual(len(result), 19)
def test_get_data_with_select(self):
self.wait()
result = self.dataset.get_data(select=['food_type', 'amount'])
self.assertEqual(len(result), 19)
for row in result:
self.assertEqual(len(row), 2)
cols = row.keys()
self.assertTrue('food_type' in cols)
self.assertTrue('amount' in cols)
def test_get_data_with_query(self):
self.wait() # TODO: remove (bamboo issue #285)
result = self.dataset.get_data(query={'food_type': 'lunch'})
self.assertEqual(len(result), 7)
def test_get_data_with_select_and_query(self):
self.wait() # TODO: remove (bamboo issue #285)
result = self.dataset.get_data(
select=['food_type', 'amount'], query={'food_type': 'lunch'})
self.assertEqual(len(result), 7)
for row in result:
self.assertEqual(len(row), 2)
cols = row.keys()
self.assertTrue('food_type' in cols)
self.assertTrue('amount' in cols)
def test_get_data_with_format(self):
self.wait() # TODO: remove (bamboo issue #285)
result = self.dataset.get_data(format='csv')
self.assertTrue(isinstance(result, basestring))
def test_get_data_invalid_select(self):
with self.assertRaises(PyBambooException):
self.dataset.get_data(select='BAD')
def test_get_data_invalid_query(self):
with self.assertRaises(PyBambooException):
self.dataset.get_data(query='BAD')
def test_get_data_with_invalid_format(self):
with self.assertRaises(PyBambooException):
self.dataset.get_data(format='BAD')
def test_get_data_bad_query(self):
self.wait() # TODO: remove (bamboo issue #285)
result = self.dataset.get_data(query={'BAD': 'BAD'})
self.assertFalse(result)
def test_update_data(self):
row = {
'food_type': 'morning_food',
'amount': 10.0,
'risk_factor': 'high_risk',
'rating': 'delectible',
}
result = self.dataset.update_data([row])
self.wait(15)
result = self.dataset.get_data()
self.assertTrue(isinstance(result, list))
self.assertEqual(len(result), 20)
def test_update_data_no_data(self):
with self.assertRaises(PyBambooException):
self.dataset.update_data([])
def test_update_data_bad_data(self):
bad_rows = [
{},
[[]],
[{'exception': Exception()}]
]
for rows in bad_rows:
with self.assertRaises(PyBambooException):
self.dataset.update_data(rows)
def test_merge(self):
# already have one dataset in self.dataset
dataset = Dataset(path=self.CSV_FILE,
connection=self.connection)
result = Dataset.merge([self.dataset, dataset],
connection=self.connection)
self.assertTrue(isinstance(result, Dataset))
self._cleanup(dataset)
self._cleanup(result)
def test_merge_default_connection(self):
dataset = Dataset(path=self.CSV_FILE,
connection=self.default_connection)
other_dataset = Dataset(path=self.CSV_FILE,
connection=self.default_connection)
result = Dataset.merge([dataset, other_dataset])
self.assertTrue(isinstance(result, Dataset))
self._cleanup(dataset)
self._cleanup(other_dataset)
self._cleanup(result)
def test_merge_bad_datasets(self):
dataset = {}
other_dataset = []
with self.assertRaises(PyBambooException):
Dataset.merge([dataset, other_dataset],
connection=self.connection)
def test_merge_fail(self):
other_dataset = Dataset('12345', connection=self.connection)
result = Dataset.merge([self.dataset, other_dataset],
connection=self.connection)
self.assertFalse(result)
def test_join(self):
self._create_aux_dataset_from_file()
self.wait()
result = Dataset.join(self.dataset, self.aux_dataset,
'food_type', connection=self.connection)
self.assertTrue(isinstance(result, Dataset))
self._cleanup(result)
def test_join_default_connection(self):
dataset = Dataset(path=self.CSV_FILE,
connection=self.default_connection)
aux_dataset = Dataset(path=self.AUX_CSV_FILE,
connection=self.default_connection)
self.wait()
result = Dataset.join(dataset, aux_dataset, 'food_type')
self.wait()
self.assertTrue(isinstance(result, Dataset))
self._cleanup(dataset)
self._cleanup(aux_dataset)
self._cleanup(result)
def test_join_bad_other_dataset(self):
with self.assertRaises(PyBambooException):
Dataset.join(self.dataset, Exception(), 'food_type',
connection=self.connection)
def test_join_bad_on(self):
self._create_aux_dataset_from_file()
self.wait()
result = Dataset.join(self.dataset, self.aux_dataset,
'BAD', connection=self.connection)
self.assertFalse(result)
# /row/INDEX tests.
def test_get_row(self):
self.assertEqual(self.dataset.get_row(0)['comments'],
u"Try the yogurt drink")
def test_update_row(self):
index = 2
comment = 'test'
self.dataset.update_row(index, {'comments': comment})
self.assertEqual(self.dataset.get_row(index)['comments'], comment)
def test_delete_row(self):
self._wait_for_dataset_ready() # TODO: is this necessary?
index = 10
self.dataset.delete_row(index=index)
result = self.dataset.get_row(index)
self.assertTrue('error' in result)
