def test_indexer_generate_modname_nunja_script_altname(self):
entry_point = EntryPoint.parse('example.mold = example.package:mold')
module = ModuleType('example.package')
nunja_template, nunja_script, nunja_modpath = generate_modname_nunja(
entry_point, module, fext='.tmpl')
# the names provided will be the fragments based on actual file
# name, with the final filename extension stripped.
name = nunja_script(['example', 'package', 'table', 'index'])
self.assertEqual(name, 'example.mold/table/index')
def test_indexer_generate_modname_nunja_modpath_pkg_resources(self):
entry_point = EntryPoint.parse('nunja.testing = nunja.testing:mold')
nunja_template, nunja_script, nunja_modpath = generate_modname_nunja(
entry_point, testing, fext='.tmpl')
# the names provided will be the fragments based on actual file
# name, with the final filename extension stripped.
name = nunja_modpath(testing, entry_point)
self.assertEqual(len(name), 1)
# make it easier to see wrong than just not True
self.assertIn(join('nunja', 'testing', 'mold'), name[0])
self.assertTrue(name[0].endswith(join('nunja', 'testing', 'mold')))
def test_indexer_modpath_pkg_resources_entry_point_not_import(self):
entry_point = EntryPoint.parse('nunjatest = nunja.testing:mold')
nunja_template, nunja_script, nunja_modpath = generate_modname_nunja(
entry_point, testing, fext='.tmpl')
with pretty_logging(logger='nunja', stream=StringIO()) as stream:
path = nunja_modpath(testing, entry_point)
msg = stream.getvalue()
self.assertEqual(msg, '')
self.assertTrue(path[0].endswith(join('nunja', 'testing', 'mold')))
def test_indexer_modpath_pkg_resources_entry_point_fail_import(self):
entry_point = EntryPoint.parse('example.mold = example.package:mold')
module = ModuleType('example.package')
nunja_template, nunja_script, nunja_modpath = generate_modname_nunja(
entry_point, module, fext='.tmpl')
with pretty_logging(logger='nunja', stream=StringIO()) as stream:
nunja_modpath(None, None)
msg = stream.getvalue()
self.assertIn("does not appear to be a valid module", msg)
self.assertIn("got unexpected entrypoint", msg)
def test_indexer_modpath_pkg_resources_entry_point_mismatch_module(self):
entry_point = EntryPoint.parse('example.mold = example.package1:mold')
module1 = ModuleType('example.package1')
module2 = ModuleType('example.package2')
nunja_template, nunja_script, nunja_modpath = generate_modname_nunja(
entry_point, module1, fext='.tmpl')
with pretty_logging(logger='nunja', stream=StringIO()) as stream:
nunja_modpath(module2, entry_point)
msg = stream.getvalue()
self.assertIn(
"modpath function created for <module 'example.package1'", msg)
self.assertIn("got unexpected module <module 'example.package2'", msg)
self.assertIn("could not be located as a module", msg)
def test_indexer_generate_modname_nunja_template(self):
entry_point = EntryPoint.parse(
'example.package.mold = example.package:mold')
module = ModuleType('example.package')
nunja_template, nunja_script, nunja_modpath = generate_modname_nunja(
entry_point, module, fext='.tmpl')
# the names provided will be the fragments based on actual file
# name, with the final filename extension stripped. However a
# **HUGE** difference is that the customized modpath function
# (nunja_modpath_pkg_resources) ALSO provides the first attrs
# (the name) to the module_base_path, thus generating the
# mapping that effectively truncates out the attrs.
# Normally this is what might be expected
# ['example', 'package', 'mold', 'table', 'template'])
# However, this is what we actually get
name = nunja_template(['example', 'package', 'table', 'template'])
# Naturally, this applies for the rest of the functions.
self.assertEqual(name, 'text!example.package.mold/table/template.tmpl')