def setUp(self):
self.task = AnswerDistributionOneFilePerCourseTask(
mapreduce_engine='local',
src=None,
dest=None,
name=None,
include=None,
output_root=None,
)
def test_output_path_for_opaque_key(self):
course_id = str(CourseLocator(org='foo', course='bar', run='baz'))
hashed_course_id = hashlib.sha1(course_id).hexdigest()
task = AnswerDistributionOneFilePerCourseTask(
mapreduce_engine='local',
src=None,
dest=None,
name='name',
include=None,
output_root='/tmp',
)
output_path = task.output_path_for_key(course_id)
expected_output_path = '/tmp/{0}/foo_bar_baz_answer_distribution.csv'.format(hashed_course_id)
self.assertEquals(output_path, expected_output_path)
def test_delete_output_root(self):
# It's still possible to use the delete option
# to get rid of the output_root directory.
task = AnswerDistributionOneFilePerCourseTask(
mapreduce_engine='local',
src=None,
dest=None,
name='name',
include=None,
output_root=self.output_root,
delete_output_root="true",
)
self.assertFalse(task.complete())
self.assertFalse(os.path.exists(self.output_root))
def test_delete_output_root(self):
# It's still possible to use the delete option
# to get rid of the output_root directory.
task = AnswerDistributionOneFilePerCourseTask(
mapreduce_engine='local',
src=None,
dest=None,
name='name',
include=None,
output_root=self.output_root,
delete_output_root="true",
)
self.assertFalse(task.complete())
self.assertFalse(os.path.exists(self.output_root))
def test_output_path_for_opaque_key(self):
course_id = str(CourseLocator(org='foo', course='bar', run='baz'))
hashed_course_id = hashlib.sha1(course_id).hexdigest()
task = AnswerDistributionOneFilePerCourseTask(
mapreduce_engine='local',
src=None,
dest=None,
name='name',
include=None,
output_root='/tmp',
)
output_path = task.output_path_for_key(course_id)
expected_output_path = '/tmp/{0}/foo_bar_baz_answer_distribution.csv'.format(
hashed_course_id)
self.assertEquals(output_path, expected_output_path)
def setUp(self):
self.task = AnswerDistributionOneFilePerCourseTask(
mapreduce_engine='local',
src=None,
dest=None,
name=None,
include=None,
output_root=None,
)
def test_no_delete_output_root(self):
# Not using the delete_output_root option will
# not delete the output_root.
self.assertTrue(os.path.exists(self.output_root))
AnswerDistributionOneFilePerCourseTask(
mapreduce_engine='local',
src=None,
dest=None,
name='name',
include=None,
output_root=self.output_root,
)
self.assertTrue(os.path.exists(self.output_root))
class AnswerDistributionOneFilePerCourseTaskTest(unittest.TestCase):
"""Tests for AnswerDistributionOneFilePerCourseTask class."""
def setUp(self):
self.task = AnswerDistributionOneFilePerCourseTask(
mapreduce_engine='local',
src=None,
dest=None,
name=None,
include=None,
output_root=None,
)
def test_map_single_value(self):
key, value = next(self.task.mapper('foo\tbar'))
self.assertEquals(key, 'foo')
self.assertEquals(value, 'bar')
def test_reduce_multiple_values(self):
field_names = AnswerDistributionPerCourseMixin.get_column_order()
# To test sorting, the first sample is made to sort after the
# second sample.
column_values_2 = [(k, unicode(k) + u'\u2603') for k in field_names]
column_values_2[3] = (column_values_2[3][0], 10)
column_values_1 = list(column_values_2)
column_values_1[4] = (column_values_1[4][0], u'ZZZZZZZZZZZ')
sample_input_1 = json.dumps(dict(column_values_1))
sample_input_2 = json.dumps(dict(column_values_2))
mock_output_file = Mock()
self.task.multi_output_reducer('foo',
iter([sample_input_1, sample_input_2]),
mock_output_file)
expected_header_string = ','.join(field_names) + '\r\n'
self.assertEquals(mock_output_file.write.mock_calls[0],
call(expected_header_string))
# Confirm that the second sample appears before the first.
expected_row_1 = ','.join(
unicode(v[1]).encode('utf8') for v in column_values_2) + '\r\n'
self.assertEquals(mock_output_file.write.mock_calls[1],
call(expected_row_1))
expected_row_2 = ','.join(
unicode(v[1]).encode('utf8') for v in column_values_1) + '\r\n'
self.assertEquals(mock_output_file.write.mock_calls[2],
call(expected_row_2))
def test_output_path_for_legacy_key(self):
course_id = 'foo/bar/baz'
hashed_course_id = hashlib.sha1(course_id).hexdigest()
task = AnswerDistributionOneFilePerCourseTask(
mapreduce_engine='local',
src=None,
dest=None,
name='name',
include=None,
output_root='/tmp',
)
output_path = task.output_path_for_key(course_id)
expected_output_path = '/tmp/{0}/foo_bar_baz_answer_distribution.csv'.format(
hashed_course_id)
self.assertEquals(output_path, expected_output_path)
def test_output_path_for_opaque_key(self):
course_id = str(CourseLocator(org='foo', course='bar', run='baz'))
hashed_course_id = hashlib.sha1(course_id).hexdigest()
task = AnswerDistributionOneFilePerCourseTask(
mapreduce_engine='local',
src=None,
dest=None,
name='name',
include=None,
output_root='/tmp',
)
output_path = task.output_path_for_key(course_id)
expected_output_path = '/tmp/{0}/foo_bar_baz_answer_distribution.csv'.format(
hashed_course_id)
self.assertEquals(output_path, expected_output_path)
class AnswerDistributionOneFilePerCourseTaskTest(unittest.TestCase):
"""Tests for AnswerDistributionOneFilePerCourseTask class."""
def setUp(self):
self.task = AnswerDistributionOneFilePerCourseTask(
mapreduce_engine='local',
src=None,
dest=None,
name=None,
include=None,
output_root=None,
)
def test_map_single_value(self):
key, value = next(self.task.mapper('foo\tbar'))
self.assertEquals(key, 'foo')
self.assertEquals(value, 'bar')
def test_reduce_multiple_values(self):
field_names = AnswerDistributionPerCourseMixin.get_column_order()
# To test sorting, the first sample is made to sort after the
# second sample.
column_values_2 = [(k, unicode(k) + u'\u2603') for k in field_names]
column_values_2[3] = (column_values_2[3][0], 10)
column_values_1 = list(column_values_2)
column_values_1[4] = (column_values_1[4][0], u'ZZZZZZZZZZZ')
sample_input_1 = json.dumps(dict(column_values_1))
sample_input_2 = json.dumps(dict(column_values_2))
mock_output_file = Mock()
self.task.multi_output_reducer('foo', iter([sample_input_1, sample_input_2]), mock_output_file)
expected_header_string = ','.join(field_names) + '\r\n'
self.assertEquals(mock_output_file.write.mock_calls[0], call(expected_header_string))
# Confirm that the second sample appears before the first.
expected_row_1 = ','.join(unicode(v[1]).encode('utf8') for v in column_values_2) + '\r\n'
self.assertEquals(mock_output_file.write.mock_calls[1], call(expected_row_1))
expected_row_2 = ','.join(unicode(v[1]).encode('utf8') for v in column_values_1) + '\r\n'
self.assertEquals(mock_output_file.write.mock_calls[2], call(expected_row_2))
def test_output_path_for_legacy_key(self):
course_id = 'foo/bar/baz'
hashed_course_id = hashlib.sha1(course_id).hexdigest()
task = AnswerDistributionOneFilePerCourseTask(
mapreduce_engine='local',
src=None,
dest=None,
name='name',
include=None,
output_root='/tmp',
)
output_path = task.output_path_for_key(course_id)
expected_output_path = '/tmp/{0}/foo_bar_baz_answer_distribution.csv'.format(hashed_course_id)
self.assertEquals(output_path, expected_output_path)
def test_output_path_for_opaque_key(self):
course_id = str(CourseLocator(org='foo', course='bar', run='baz'))
hashed_course_id = hashlib.sha1(course_id).hexdigest()
task = AnswerDistributionOneFilePerCourseTask(
mapreduce_engine='local',
src=None,
dest=None,
name='name',
include=None,
output_root='/tmp',
)
output_path = task.output_path_for_key(course_id)
expected_output_path = '/tmp/{0}/foo_bar_baz_answer_distribution.csv'.format(hashed_course_id)
self.assertEquals(output_path, expected_output_path)