def test_bf_03_single_col(self):
sampling_method = 'non'
sampling_rate = None
self.columns = [1]
field_freq, truncated = mod.build_freq(self.files, self.dialect, self.columns, self.number, sampling_method, sampling_rate)
assert(not truncated)
assert(sum(field_freq.values()) == 1000)
assert(len(field_freq) == 4) # it's possible (but extremely unlikely) that there could be fewer entries
for key in field_freq.keys():
assert(key[0] in ['A1','A2','A3','A4'])
def test_bf_01_multicol(self):
sampling_method = 'non'
sampling_rate = None
field_freq, truncated = mod.build_freq(self.files, self.dialect, self.columns, self.number, sampling_method, sampling_rate)
assert(not truncated)
assert(sum(field_freq.values()) == 1000)
assert(len(field_freq) == 8)
for key in field_freq.keys():
assert(key[0] in ['A1','A2','A3','A4'])
assert(key[1] in ['B1','B2'])
def test_bf_02_multicol_and_truncation(self):
sampling_method = 'non'
sampling_rate = None
self.number = 4
field_freq, truncated = mod.build_freq(self.files, self.dialect, self.columns, self.number, sampling_method, sampling_rate)
assert(truncated)
assert(len(field_freq) == 4) # it's possible (but extremely unlikely) that there could be fewer entries
for key in field_freq.keys():
assert(key[0] in ['A1','A2','A3','A4'])
assert(key[1] in ['B1','B2'])
def test_bf_03_interval_sampling(self):
sampling_method = 'interval'
sampling_rate = 10
field_freq, truncated = mod.build_freq(self.files, self.dialect, self.columns, self.number, sampling_method, sampling_rate)
assert(not truncated)
assert(sum(field_freq.values()) == 100)
assert(len(field_freq) == 8) # it's possible (but unlikely) that there could be fewer entries
for key in field_freq.keys():
assert(key[0] in ['A1','A2','A3','A4'])
assert(key[1] in ['B1','B2'])
