def testDataContextNode(self):
# This happens to test the "value_of" tag as well.
lang_model = language_model.LanguageModel('|')
foo_def_dict = {
'className': 'Foo',
'type': 'string',
}
prototype = data_types.DataType(foo_def_dict,
None,
language_model=lang_model)
dv = data_value.DataValue('four', prototype)
source = '{% value_of data %}'
template = django_template.Template(source)
context = self._GetContext({'data': dv})
self.assertEquals('"four"', template.render(context))
context = self._GetContext({'data': 'foo'})
self.assertRaises(django_template.TemplateSyntaxError, template.render,
context)
bar_def_dict = {
'className': 'Foo',
'type': 'parrot',
}
prototype = data_types.DataType(bar_def_dict,
None,
language_model=lang_model)
dv = data_value.DataValue('fred', prototype)
context = self._GetContext({'data': dv})
self.assertRaises(django_template.TemplateSyntaxError, template.render,
context)
def _CreateDataValue(self, value, val_type):
def_dict = {
'className': 'Foo',
'type': val_type,
}
prototype = data_types.DataType(
def_dict, None, language_model=self.language_model)
dv = data_value.DataValue(value, prototype)
return dv
def testDataValue(self):
foo_def_dict = {
'className': 'Foo',
'type': 'string',
}
prototype = data_types.DataType(foo_def_dict,
None,
language_model=self.language_model)
dv = data_value.DataValue(3, prototype)
# Basic Checks
self.assertEqual(3, dv.value)
self.assertEqual(prototype, dv.data_type)
self.assertEqual({}, dv.metadata)
self.assertEqual('Foo', dv.code_type)
dv.metadata['foo'] = 'bar'
self.assertEqual({'foo': 'bar'}, dv.metadata)
dv.SetValue('four')
self.assertEqual(dv.value, 'four')
self.assertEqual(self.language_model, dv.GetLanguageModel())
other_language_model = language_model.LanguageModel(
class_name_delimiter='+')
dv.SetLanguageModel(other_language_model)
self.assertEqual(other_language_model, dv.GetLanguageModel())
# Now that we've set a local language model... make sure the codepath
# for setting the data_type's language model gets exercised.
self.assertEqual('Foo', dv.code_type)
# Check that the constructor doesn't freak if an odd object is passed in
dv = data_value.DataValue(object, prototype)
self.assertEqual(dv.value, object)
# A standard case is the prototype being a Property object. It is not
# uncommon that the Property's data_type is a SchemaReference. To verify
# this case is handled correctly we must fake an API.
bar_def_dict = {
'className': 'Foo',
'type': 'string',
}
class MockApi(object):
def __init__(self):
self.model_module = None
def SetSchema(self, s):
self.schema = s
def SetSchemaRef(self, schema_ref):
self.schema_ref = schema_ref
# pylint: disable=unused-argument
def ToClassName(self, name, element, element_type):
return name
# pylint: disable=unused-argument
def SchemaByName(self, schema_name):
return self.schema
# pylint: disable=unused-argument
def DataTypeFromJson(self,
unused_def_dict,
tentative_class_name,
parent=None,
wire_name=None):
return self.schema_ref
def NestedClassNameForProperty(self, name, owning_schema):
return '%s%s' % (owning_schema.class_name, name)
mock_api = MockApi()
bar_schema = schema.Schema(mock_api, 'Bar', bar_def_dict)
mock_api.SetSchema(bar_schema)
schema_ref = data_types.SchemaReference('Bar', mock_api)
mock_api.SetSchemaRef(schema_ref)
prototype = schema.Property(mock_api, schema_ref, 'baz', foo_def_dict)
dv = data_value.DataValue('3', prototype)
# Assure all the unwrapping gymnastics in the DataValue constructor did
# their job correctly.
self.assertEqual(mock_api.schema, dv.data_type)