def test_dataset_subscription_info(self, mock_fetch):
# Given
mock_fetch.return_value = test_subscription_info
credentials = Credentials('fake_user', '1234')
dataset = Dataset(db_dataset1)
# When
info = dataset.subscription_info(credentials)
# Then
mock_fetch.assert_called_once_with(db_dataset1['id'], 'dataset',
credentials)
assert isinstance(info, SubscriptionInfo)
assert info.id == test_subscription_info['id']
assert info.estimated_delivery_days == test_subscription_info[
'estimated_delivery_days']
assert info.subscription_list_price == test_subscription_info[
'subscription_list_price']
assert info.tos == test_subscription_info['tos']
assert info.tos_link == test_subscription_info['tos_link']
assert info.licenses == test_subscription_info['licenses']
assert info.licenses_link == test_subscription_info['licenses_link']
assert info.rights == test_subscription_info['rights']
assert str(info) == 'Properties: id, estimated_delivery_days, ' + \
'subscription_list_price, tos, tos_link, ' + \
'licenses, licenses_link, rights'
def test_dataset_is_available_in_with_empty_field(self):
# Given
db_dataset = dict(db_dataset1)
db_dataset['available_in'] = None
dataset_null = Dataset(db_dataset)
# Then
assert not dataset_null._is_available_in('bq')
def test_summary_tail(self):
# Given
dataset = Dataset(db_dataset2)
# When
summary = dataset.tail()
# Then
assert isinstance(summary, pd.DataFrame)
def test_summary_counts(self):
# Given
dataset = Dataset(db_dataset2)
# When
summary = dataset.counts()
# Then
assert isinstance(summary, pd.Series)
def test_summary_fields_by_type(self):
# Given
dataset = Dataset(db_dataset2)
# When
summary = dataset.fields_by_type()
# Then
assert isinstance(summary, pd.Series)
def test_dataset_subscription_info_default_credentials(self, mocked_credentials, mock_fetch):
# Given
expected_credentials = Credentials('fake_user', '1234')
mocked_credentials.return_value = expected_credentials
dataset = Dataset(db_dataset1)
# When
dataset.subscription_info()
# Then
mock_fetch.assert_called_once_with(db_dataset1['id'], 'dataset', expected_credentials)
def test_dataset_subscribe_default_credentials(self, mocked_credentials, mock_display_form, mock_subscription_ids):
# Given
expected_credentials = Credentials('fake_user', '1234')
mocked_credentials.return_value = expected_credentials
dataset = Dataset(db_dataset1)
# When
dataset.subscribe()
# Then
mock_subscription_ids.assert_called_once_with(expected_credentials)
mock_display_form.assert_called_once_with(db_dataset1['id'], 'dataset', expected_credentials)
def test_dataset_is_exported_as_dict(self):
# Given
dataset = Dataset(db_dataset1)
excluded_fields = ['summary_json', 'available_in']
expected_dict = {key: value for key, value in db_dataset1.items() if key not in excluded_fields}
# When
dataset_dict = dataset.to_dict()
# Then
assert isinstance(dataset_dict, dict)
assert dataset_dict == expected_dict
def test_dataset_subscription_info_wrong_credentials(self):
# Given
wrong_credentials = 1234
dataset = Dataset(db_dataset1)
# When
with pytest.raises(ValueError) as e:
dataset.subscription_info(wrong_credentials)
# Then
assert str(e.value) == ('Credentials attribute is required. '
'Please pass a `Credentials` instance '
'or use the `set_default_credentials` function.')
def test_dataset_subscribe_existing(self, mock_display_message, mock_display_form, mock_subscription_ids):
# Given
expected_id = db_dataset1['id']
expected_subscribed_ids = [expected_id]
mock_subscription_ids.return_value = expected_subscribed_ids
credentials = Credentials('fake_user', '1234')
dataset = Dataset(db_dataset1)
# When
dataset.subscribe(credentials)
# Then
mock_subscription_ids.assert_called_once_with(credentials)
mock_display_message.assert_called_once_with(expected_id, 'dataset')
assert not mock_display_form.called
def test_get_all_datasets_credentials_without_do_enabled(self, mocked_repo):
# Given
def raise_exception(a, b):
raise ServerErrorException(['The user does not have Data Observatory enabled'])
mocked_repo.side_effect = raise_exception
credentials = Credentials('fake_user', '1234')
# When
with pytest.raises(Exception) as e:
Dataset.get_all(credentials=credentials)
# Then
assert str(e.value) == (
'We are sorry, the Data Observatory is not enabled for your account yet. '
'Please contact your customer success manager or send an email to '
'[email protected] to request access to it.')
def test_missing_fields_are_mapped_as_None(self, mocked_repo):
# Given
mocked_repo.return_value = [{'id': 'dataset1'}]
repo = DatasetRepository()
expected_datasets = CatalogList([
Dataset({
'id': 'dataset1',
'slug': None,
'name': None,
'description': None,
'provider_id': None,
'provider_name': None,
'category_id': None,
'category_name': None,
'data_source_id': None,
'country_id': None,
'lang': None,
'geography_id': None,
'geography_name': None,
'geography_description': None,
'temporal_aggregation': None,
'time_coverage': None,
'update_frequency': None,
'version': None,
'is_public_data': None,
'summary_json': None
})
])
# When
datasets = repo.get_all()
# Then
assert datasets == expected_datasets
def test_dataset_download_without_do_enabled(self, download_stream_mock,
get_by_id_mock, get_all_mock):
# Given
get_by_id_mock.return_value = test_dataset1
dataset = Dataset.get(test_dataset1.id)
get_all_mock.return_value = []
def raise_exception(limit=None,
order_by=None,
sql_query=None,
add_geom=None,
is_geography=None):
raise ServerErrorException(
['The user does not have Data Observatory enabled'])
download_stream_mock.side_effect = raise_exception
credentials = Credentials('fake_user', '1234')
# When
with pytest.raises(Exception) as e:
dataset.to_csv('fake_path', credentials)
# Then
assert str(e.value) == (
'We are sorry, the Data Observatory is not enabled for your account yet. '
'Please contact your customer success manager or send an email to '
'[email protected] to request access to it.')
def test_dataset_subscription_info_without_do_enabled(self, mock_fetch):
# Given
def raise_exception(a, b, c):
raise ServerErrorException(['The user does not have Data Observatory enabled'])
mock_fetch.side_effect = raise_exception
dataset = Dataset(db_dataset1)
credentials = Credentials('fake_user', '1234')
# When
with pytest.raises(Exception) as e:
dataset.subscription_info(credentials)
# Then
assert str(e.value) == (
'We are sorry, the Data Observatory is not enabled for your account yet. '
'Please contact your customer success manager or send an email to '
'[email protected] to request access to it.')
def test_dataset_is_represented_with_classname_and_slug(self):
# Given
dataset = Dataset(db_dataset1)
# When
dataset_repr = repr(dataset)
# Then
assert dataset_repr == "<Dataset.get('{id}')>".format(id=db_dataset1['slug'])
def test_dataset_is_printed_with_classname(self):
# Given
dataset = Dataset(db_dataset1)
# When
dataset_str = str(dataset)
# Then
assert dataset_str == "<Dataset.get('{id}')>".format(id=db_dataset1['slug'])
def test_dataset_download_not_subscribed_but_public(self, mocked_bq_client, get_by_id_mock, get_all_mock):
# Given
get_by_id_mock.return_value = test_dataset1 # is public
dataset = Dataset.get(test_dataset1.id)
get_all_mock.return_value = []
mocked_bq_client.return_value = BigQueryClientMock()
credentials = Credentials('fake_user', '1234')
dataset.to_csv('fake_path', credentials)
def test_dataset_download_not_subscribed_but_public(
self, download_stream_mock, get_by_id_mock, get_all_mock):
# Given
get_by_id_mock.return_value = test_dataset1 # is public
dataset = Dataset.get(test_dataset1.id)
get_all_mock.return_value = []
download_stream_mock.return_value = []
credentials = Credentials('fake_user', '1234')
dataset.to_csv('fake_path', credentials)
def test_summary_values(self):
# Given
dataset = Dataset(db_dataset2)
# When
summary = dataset.summary
# Then
assert summary == dataset.data['summary_json']
def test_dataset_download_not_subscribed_but_public(
self, mock_download_stream, mock_get_by_id):
# Given
mock_get_by_id.return_value = test_dataset1 # is public
dataset = Dataset.get(test_dataset1.id)
mock_download_stream.return_value = []
credentials = Credentials('fake_user', '1234')
dataset.to_csv('fake_path', credentials)
os.remove('fake_path')
def test_dataset_download(self, mocked_bq_client, get_by_id_mock, get_all_mock):
# Given
get_by_id_mock.return_value = test_dataset1
dataset = Dataset.get(test_dataset1.id)
get_all_mock.return_value = [dataset]
mocked_bq_client.return_value = BigQueryClientMock()
credentials = Credentials('fake_user', '1234')
# Then
dataset.to_csv('fake_path', credentials)
def test_get_dataset_by_id(self, mocked_repo):
# Given
mocked_repo.return_value = test_dataset1
# When
dataset = Dataset.get(test_dataset1.id)
# Then
assert isinstance(dataset, object)
assert isinstance(dataset, Dataset)
assert dataset == test_dataset1
def test_get_all_datasets(self, mocked_repo):
# Given
mocked_repo.return_value = test_datasets
# When
datasets = Dataset.get_all()
# Then
assert isinstance(datasets, list)
assert isinstance(datasets, CatalogList)
assert datasets == test_datasets
def test_dataset_download(self, download_stream_mock, get_by_id_mock,
get_all_mock):
# Given
get_by_id_mock.return_value = test_dataset1
dataset = Dataset.get(test_dataset1.id)
get_all_mock.return_value = [dataset]
download_stream_mock.return_value = []
credentials = Credentials('fake_user', '1234')
# Then
dataset.to_csv('fake_path', credentials)
def test_dataset_download(self, mock_download_stream, mock_get_by_id,
mock_subscription_ids):
# Given
mock_get_by_id.return_value = test_dataset1
dataset = Dataset.get(test_dataset1.id)
mock_download_stream.return_value = []
mock_subscription_ids.return_value = [test_dataset1.id]
credentials = Credentials('fake_user', '1234')
# Then
dataset.to_csv('fake_path', credentials)
os.remove('fake_path')
def test_get_all_datasets_credentials(self, mocked_repo):
# Given
mocked_repo.return_value = test_datasets
credentials = Credentials('fake_user', '1234')
# When
datasets = Dataset.get_all(credentials=credentials)
# Then
mocked_repo.assert_called_once_with(None, credentials)
assert isinstance(datasets, list)
assert isinstance(datasets, CatalogList)
assert datasets == test_datasets
def test_dataset_is_available_in(self):
# Given
dataset_in_bq = Dataset(db_dataset1)
dataset_not_in_bq = Dataset(db_dataset2)
# Then
assert dataset_in_bq._is_available_in('bq')
assert not dataset_not_in_bq._is_available_in('bq')
def test_dataset_not_subscribed_download_fails(self, get_by_id_mock,
get_all_mock):
# mock dataset
get_by_id_mock.return_value = test_dataset2 # is private
dataset = Dataset.get(test_dataset2.id)
get_all_mock.return_value = []
credentials = Credentials('fake_user', '1234')
# When
with pytest.raises(Exception) as e:
dataset.to_csv('fake_path', credentials)
# Then
assert str(e.value) == ('You are not subscribed to this Dataset yet. '
'Please, use the subscribe method first.')
def test_dataset_download_without_do_enabled(self, mocked_bq_client, get_by_id_mock, get_all_mock):
# Given
get_by_id_mock.return_value = test_dataset1
dataset = Dataset.get(test_dataset1.id)
get_all_mock.return_value = []
mocked_bq_client.return_value = BigQueryClientMock(
ServerErrorException(['The user does not have Data Observatory enabled'])
)
credentials = Credentials('fake_user', '1234')
# When
with pytest.raises(Exception) as e:
dataset.to_csv('fake_path', credentials)
# Then
assert str(e.value) == (
'We are sorry, the Data Observatory is not enabled for your account yet. '
'Please contact your customer success manager or send an email to '
'[email protected] to request access to it.')
def test_dataset_properties(self):
# Given
dataset = Dataset(db_dataset1)
# When
dataset_id = dataset.id
slug = dataset.slug
name = dataset.name
description = dataset.description
provider = dataset.provider
category = dataset.category
data_source = dataset.data_source
country = dataset.country
language = dataset.language
geography = dataset.geography
temporal_aggregation = dataset.temporal_aggregation
time_coverage = dataset.time_coverage
update_frequency = dataset.update_frequency
version = dataset.version
is_public_data = dataset.is_public_data
summary = dataset.summary
# Then
assert dataset_id == db_dataset1['id']
assert slug == db_dataset1['slug']
assert name == db_dataset1['name']
assert description == db_dataset1['description']
assert provider == db_dataset1['provider_id']
assert category == db_dataset1['category_id']
assert data_source == db_dataset1['data_source_id']
assert country == db_dataset1['country_id']
assert language == db_dataset1['lang']
assert geography == db_dataset1['geography_id']
assert temporal_aggregation == db_dataset1['temporal_aggregation']
assert time_coverage == db_dataset1['time_coverage']
assert update_frequency == db_dataset1['update_frequency']
assert version == db_dataset1['version']
assert is_public_data == db_dataset1['is_public_data']
assert summary == db_dataset1['summary_json']
