def test_RN_Handles_BlankLines(self):
# make sure the name lookup initialisers deal well with empty lines
self.name_generator = \
WeightedChoice(test_data_input_file("lookupwithgaps.csv"), name_field="Forename")
self.assertEqual(
len(self.name_generator.items),
len(list(csv.DictReader(open(test_data_input_file("lookupwithgaps.csv")))))
)
def test_RN_Handles_BlankLines(self):
# make sure the name lookup initialisers deal well with empty lines
self.name_generator = \
WeightedChoice(test_data_input_file("lookupwithgaps.csv"), name_field="Forename")
self.assertEqual(
len(self.name_generator.items),
len(
list(
csv.DictReader(
open(test_data_input_file("lookupwithgaps.csv"))))))
def test_RN_Uses_All_Names(self):
"""
test that RandomNames uses all available options.
Method: probabilistic test to make sure no names are excluded from results using
small lookup file with just a few items and enough iterations to make all items
extremely likely to be chosen at least once
"""
test_fname = test_data_input_file("minimallookup.csv")
# get checklist
checklist = list(csv.DictReader(open(test_fname)))
# how many samples make it *extremely* unlikely that any item was not selected
# at least once?
# set up name generator
name_generator = WeightedChoice(test_fname, name_field="Forename")
# check all names are represented in output
for name_check in checklist:
for i in range(self.binary_check_sample_size):
new_name = name_generator.name()
if name_check["Forename"] == new_name: break
else:
self.fail("Name %s (1 of %d) was not generated even after %d iterations"\
% (name_check["Forename"],
len(name_generator.items),
self.binary_check_sample_size))
def test_RN_Uses_All_Names(self):
"""
test that RandomNames uses all available options.
Method: probabilistic test to make sure no names are excluded from results using
small lookup file with just a few items and enough iterations to make all items
extremely likely to be chosen at least once
"""
test_fname = test_data_input_file("minimallookup.csv")
# get checklist
checklist = list(csv.DictReader(open(test_fname)))
# how many samples make it *extremely* unlikely that any item was not selected
# at least once?
# set up name generator
name_generator = WeightedChoice(test_fname, name_field="Forename")
# check all names are represented in output
for name_check in checklist:
for i in range(self.binary_check_sample_size):
new_name = name_generator.name()
if name_check["Forename"] == new_name: break
else:
self.fail("Name %s (1 of %d) was not generated even after %d iterations"\
% (name_check["Forename"],
len(name_generator.items),
self.binary_check_sample_size))
def test_RN_Handles_NonNumPopularity(self):
# make sure the name popularity mechanisms deal with empty lines
with self.assertRaises(MissingPopularityException):
name_generator = WeightedChoice(
test_data_input_file("lookupMissingPopularity.csv"),
name_field="Forename")
def test_RN_Handles_NonNumPopularity(self):
# make sure the name popularity mechanisms deal with empty lines
with self.assertRaises(MissingPopularityException):
name_generator = WeightedChoice(
test_data_input_file("lookupMissingPopularity.csv"),
name_field="Forename")