I'm developing a simple Ajax backend that presently only consists of one URLconf and a single view. I'm trying to do test-driven development and have a few initial tests present, but 'manage.py test' is crapping out on me.
Initially, I got ImproperlyConfiguredError for not having DATABASE_* configured in the settings file. Googling this brought me to http://docs.djangoproject.com/en/dev/topics/testing/#using-different-testing-frameworks and Russell Keith-Magee's advice on the issue (e.g. http://groups.google.com/group/django-users/browse_thread/thread/fe90a261293ea995 and http://groups.google.com/group/django-users/browse_thread/thread/d3cac9794c5e59f1 ), suggesting creating a custom test runner by copying the code of the default runner w/o the test database creation, and then defining TEST_RUNNER in the settings file. That's what I did, but a different set of errors appeared. First, here's the code of my custom run_tests, with the database bits commented out. (Based off trunk r9800) - - - - - import unittest from django.conf import settings from django.db.models import get_app, get_apps from django.test.utils import setup_test_environment, teardown_test_environment from django.test.simple import build_suite, reorder_suite def run_tests(test_labels, verbosity=1, interactive=True, extra_tests= []): """A copy of django.test.simple.run_tests with the test DB creation commented out.""" setup_test_environment() settings.DEBUG = False suite = unittest.TestSuite() if test_labels: for label in test_labels: if '.' in label: suite.addTest(build_test(label)) else: app = get_app(label) suite.addTest(build_suite(app)) else: for app in get_apps(): suite.addTest(build_suite(app)) for test in extra_tests: suite.addTest(test) suite = reorder_suite(suite, (TestCase,)) # old_name = settings.DATABASE_NAME # from django.db import connection # connection.creation.create_test_db(verbosity, autoclobber=not interactive) result = unittest.TextTestRunner(verbosity=verbosity).run(suite) # connection.creation.destroy_test_db(old_name, verbosity) teardown_test_environment() return len(result.failures) + len(result.errors) - - - - - Given that, the correct TEST_RUNNER setting and nine tests to be run, this is the output of ./manage.py test: - - - - - EEEEEEEEE ====================================================================== ERROR: Views should return HTTP 400 when receiving syntactically invalid JSON as POST. ---------------------------------------------------------------------- Traceback (most recent call last): File "/Library/Python/2.5/site-packages/django/test/testcases.py", line 238, in __call__ self._pre_setup() File "/Library/Python/2.5/site-packages/django/test/testcases.py", line 213, in _pre_setup self._fixture_setup() File "/Library/Python/2.5/site-packages/django/test/testcases.py", line 410, in _fixture_setup if not settings.DATABASE_SUPPORTS_TRANSACTIONS: File "/Library/Python/2.5/site-packages/django/conf/__init__.py", line 32, in __getattr__ return getattr(self._target, name) AttributeError: 'Settings' object has no attribute 'DATABASE_SUPPORTS_TRANSACTIONS' [...same traceback reported eight more times, once per test...] ---------------------------------------------------------------------- Ran 0 tests in 0.002s FAILED (errors=9) - - - - - Googling the issue, I stumbled across changeset 9756 ( http://code.djangoproject.com/changeset/9756 ) which appears to introduce the DATABASE_SUPPORTS_TRANSACTIONS setting. Thinking it might have introduced a bug, I tried running the tests against trunk r9755. This is what happened then with ./manage.py test: [...] ImportError: cannot import name reorder_suite reorder_suite was apparently introduced between r9755 and r9800, so I took out the import and replaced my custom run_tests to be based off r9755 (copy verbatim, comment out the DB bits just like before) and ran ./manage.py test once more. This is the result: - - - - - - - EEEEEEEEE ====================================================================== ERROR: Views should return HTTP 400 when receiving syntactically invalid JSON as POST. ---------------------------------------------------------------------- Traceback (most recent call last): File "/Library/Python/2.5/site-packages/django/test/testcases.py", line 207, in __call__ self._pre_setup() File "/Library/Python/2.5/site-packages/django/test/testcases.py", line 188, in _pre_setup call_command('flush', verbosity=0, interactive=False) File "/Library/Python/2.5/site-packages/django/core/management/ __init__.py", line 158, in call_command return klass.execute(*args, **options) File "/Library/Python/2.5/site-packages/django/core/management/ base.py", line 222, in execute output = self.handle(*args, **options) File "/Library/Python/2.5/site-packages/django/core/management/ base.py", line 351, in handle return self.handle_noargs(**options) File "/Library/Python/2.5/site-packages/django/core/management/ commands/flush.py", line 30, in handle_noargs sql_list = sql_flush(self.style, only_django=True) File "/Library/Python/2.5/site-packages/django/core/management/ sql.py", line 131, in sql_flush statements = connection.ops.sql_flush(style, tables, connection.introspection.sequence_list()) File "/Library/Python/2.5/site-packages/django/db/backends/ __init__.py", line 292, in sql_flush raise NotImplementedError() NotImplementedError [...same traceback reported eight more times, once per test...] ---------------------------------------------------------------------- Ran 0 tests in 0.016s FAILED (errors=9) - - - - - - - ...and this is the current, r9755-based custom run_tests method, the same as before but without the code introduced between r9755 and now: - - - - - - - def run_tests(test_labels, verbosity=1, interactive=True, extra_tests= []): setup_test_environment() settings.DEBUG = False suite = unittest.TestSuite() if test_labels: for label in test_labels: if '.' in label: suite.addTest(build_test(label)) else: app = get_app(label) suite.addTest(build_suite(app)) else: for app in get_apps(): suite.addTest(build_suite(app)) for test in extra_tests: suite.addTest(test) # old_name = settings.DATABASE_NAME # from django.db import connection # connection.creation.create_test_db(verbosity, autoclobber=not interactive) result = unittest.TextTestRunner(verbosity=verbosity).run(suite) # connection.creation.destroy_test_db(old_name, verbosity) teardown_test_environment() return len(result.failures) + len(result.errors) - - - - - - - I went as far as commenting out all "django-specific" bits of the test runner, removing the imports from django.* to make sure that run_tests was just using the unittest functionality, and added a single test manually to the suite variable, but the same NotImplementedError came up. Looks like the empty DATABASE_* settings are causing something somewhere in the chain to go to the BaseDatabaseOperations class and call its methods, all of which of course return NotImplementedError. This is now well beyond my skills, so I turn to you, the experts. What is the correct method of creating a custom test runner for testing Django without a database, and is this different pre- and post-r9756, since at least the error message is completely different? TIA, Jarkko Laiho --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Django users" group. To post to this group, send email to django-users@googlegroups.com To unsubscribe from this group, send email to django-users+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/django-users?hl=en -~----------~----~----~----~------~----~------~--~---
