def testGettersAndSetters(self):
"""Test behavior of concept, slice, and table setters and getters."""
self.dspl_dataset.AddImport(
dspl_model.Import(namespace_id='import1',
namespace_url='import1_url'))
self.dspl_dataset.AddTopic(
dspl_model.Topic(topic_id='topic1',
children=[dspl_model.Topic(topic_id='topic2')]))
self.dspl_dataset.AddTopic(dspl_model.Topic(topic_id='topic3'))
self.dspl_dataset.AddConcept(dspl_model.Concept(concept_id='concept1'))
self.dspl_dataset.AddConcept(dspl_model.Concept(concept_id='concept2'))
self.dspl_dataset.AddSlice(dspl_model.Slice(slice_id='slice1'))
self.dspl_dataset.AddTable(dspl_model.Table(table_id='table1'))
self.assertEqual(
self.dspl_dataset.GetImport('import1').namespace_url,
'import1_url')
self.assertEqual(
self.dspl_dataset.GetTopic('topic1').topic_id, 'topic1')
self.assertEqual(
self.dspl_dataset.GetTopic('topic2').topic_id, 'topic2')
self.assertEqual(
self.dspl_dataset.GetTopic('topic3').topic_id, 'topic3')
self.assertEqual(
self.dspl_dataset.GetConcept('concept2').concept_id, 'concept2')
self.assertEqual(
self.dspl_dataset.GetSlice('slice1').slice_id, 'slice1')
self.assertEqual(
self.dspl_dataset.GetTable('table1').table_id, 'table1')
self.assertEqual(self.dspl_dataset.GetConcept('concept3'), None)
self.assertEqual(self.dspl_dataset.GetSlice('slice3'), None)
self.assertEqual(self.dspl_dataset.GetTable('table3'), None)
def ElementToTable(table_element, csv_path, load_all_data):
"""Convert an ElementTree table element into a Table object.
Args:
table_element: ElementTree element having data from <table>...</table>
section in an XML file
csv_path: Path to directory where CSV file associated with this table can
be found
load_all_data: Boolean indicating whether all CSV data should be loaded
Returns:
dspl_model.Table object
"""
dspl_table = dspl_model.Table()
dspl_table.table_id = table_element.get('id')
column_elements = table_element.findall(_DSPL_SCHEMA_PREFIX + 'column')
dspl_columns = []
for column_element in column_elements:
column_value_element = column_element.find(_DSPL_SCHEMA_PREFIX +
'value')
if column_value_element is not None:
constant_value = column_value_element.text
else:
constant_value = ''
dspl_column = dspl_model.TableColumn(
column_id=column_element.get('id'),
data_type=column_element.get('type'),
data_format=column_element.get('format', default=''),
constant_value=constant_value)
dspl_columns.append(dspl_column)
dspl_table.columns = dspl_columns
data_element = table_element.find(_DSPL_SCHEMA_PREFIX + 'data')
if data_element is not None:
file_element = data_element.find(_DSPL_SCHEMA_PREFIX + 'file')
if file_element is not None:
dspl_table.file_name = file_element.text.strip()
if dspl_table.file_name:
csv_file_path = os.path.join(csv_path, dspl_table.file_name)
dspl_table.table_data = _ReadCSVData(csv_file_path, load_all_data)
return dspl_table
def testTableData(self):
"""Test that Table objects materialize their data to CSV correctly."""
table_data = [['col1', 'col2', 'col3'],
['1/1/2010', 'blue', 'california'],
['1/2/2010', 'red', 'maine']]
dspl_table = dspl_model.Table(table_id='table',
file_name=self.csv_file_path,
table_data=table_data,
verbose=False)
dspl_table.MaterializeData('')
output_csv_file = open(self.csv_file_path, 'r')
output_csv_reader = csv.reader(output_csv_file)
self.assertEqual(next(output_csv_reader), ['col1', 'col2', 'col3'])
self.assertEqual(next(output_csv_reader),
['1/1/2010', 'blue', 'california'])
self.assertEqual(next(output_csv_reader), ['1/2/2010', 'red', 'maine'])
output_csv_file.close()
def testDatasetXMLCreation(self):
"""Create dataset using models, then compare output to expected XML."""
self.dspl_dataset.name = 'My Dataset'
self.dspl_dataset.description = 'My Dataset Description'
self.dspl_dataset.url = 'url1'
self.dspl_dataset.provider_name = 'Googler'
self.dspl_dataset.provider_url = 'url2'
self.dspl_dataset.AddImport(
dspl_model.Import(namespace_id='imported_namespace1',
namespace_url='http://imported_namespace1_url'))
self.dspl_dataset.AddImport(
dspl_model.Import(namespace_id='imported_namespace2',
namespace_url='http://imported_namespace2_url'))
topic1 = dspl_model.Topic(topic_id='topic1', topic_name='topic1_name')
topic2 = dspl_model.Topic(topic_id='topic2', topic_name='topic2_name')
topic3 = dspl_model.Topic(topic_id='topic3', topic_name='topic3_name')
topic4 = dspl_model.Topic(topic_id='topic4', topic_name='topic4_name')
topic1.children = [topic2, topic3]
self.dspl_dataset.AddTopic(topic1)
self.dspl_dataset.AddTopic(topic4)
self.dspl_dataset.AddConcept(
dspl_model.Concept(
concept_id='concept1',
concept_name='Concept 1',
concept_description='Concept 1 Description',
table_ref='table2',
concept_extension_reference='entity:entity',
data_type='string',
attributes=[
dspl_model.Attribute(concept_ref='attribute_concept',
value='attribute_value')
],
properties=[
dspl_model.Property(concept_ref='property_concept'),
dspl_model.Property(concept_ref='another_property_concept',
is_parent=True)
]))
self.dspl_dataset.AddConcept(
dspl_model.Concept(concept_id='concept2',
concept_name='Concept 2',
concept_description='Concept 2 Description',
data_type='integer',
topic_references=['topic1', 'topic2']))
self.dspl_dataset.AddConcept(
dspl_model.Concept(concept_id='geo:country',
concept_reference='geo:country',
data_type='string'))
self.dspl_dataset.AddSlice(
dspl_model.Slice(slice_id='data_slice',
dimension_refs=['concept1', 'geo:country'],
metric_refs=['concept2'],
dimension_map={
'concept1': 'concept_column1',
'geo:country': 'concept_column3'
},
metric_map={'concept2': 'concept_column2'},
table_ref='table3'))
self.dspl_dataset.AddTable(
dspl_model.Table(
table_id='table',
columns=[
dspl_model.TableColumn('col1', 'string', '', ''),
dspl_model.TableColumn('col2', 'integer', '', '1234')
],
file_name='mydata.csv',
verbose=False))
xml_output = self.dspl_dataset.ToXMLElement()
for element_tuple in itertools.izip_longest(
xml_output.getiterator(),
xml.etree.ElementTree.fromstring(TEST_DSPL_XML).getiterator()):
constructed_element = element_tuple[0]
expected_element = element_tuple[1]
# Compare tag names
constructed_tag_name = constructed_element.tag
# Remove namespace prefixes from expected tag
expected_tag_name = re.search('^(?:\{.*\})?(.*)$',
expected_element.tag).group(1)
self.assertEqual(constructed_tag_name, expected_tag_name)
# Compare tag attributes, ignoring these for dspl and value tags
if (constructed_element.tag != 'dspl'
and constructed_element.tag != 'value'):
self.assertEqual(sorted(constructed_element.items()),
sorted(expected_element.items()))
# Compare tag text
constructed_text = constructed_element.text
expected_text = expected_element.text
# Handle differences in how test and expected results handle text
if expected_text:
expected_text = expected_text.strip()
if expected_text == '':
expected_text = None
self.assertEqual(constructed_text, expected_text)