def make_random_bin_table(num_rows, num_cols, key_col1, key_col2, delim,
dst_path):
restricted_chars = ['\r', '\n'] + [delim]
key_col = random.randint(0, num_cols - 1)
good_keys1 = ['alpha', 'beta', 'gamma', 'delta', 'epsilon', 'zeta']
good_keys2 = [str(v) for v in range(20)]
result_table = list()
for r in rbql.xrange6(num_rows):
result_table.append(list())
for c in rbql.xrange6(num_cols):
if c == key_col1:
result_table[-1].append(random.choice(good_keys1))
elif c == key_col2:
result_table[-1].append(random.choice(good_keys2))
else:
dice = random.randint(1, 20)
if dice == 1:
result_table[-1].append(random.choice(good_keys1))
elif dice == 2:
result_table[-1].append(random.choice(good_keys2))
else:
result_table[-1].append(
make_random_csv_entry(0, 20, restricted_chars))
with codecs.open(dst_path, 'w', encoding='latin-1') as f:
for row in result_table:
f.write(delim.join(row))
if random.randint(0, 2) == 0:
f.write('\r\n')
else:
f.write('\n')
def generate_random_scenario(max_num_rows, max_num_cols, delims):
num_rows = random.randint(1, max_num_rows)
num_cols = random.randint(1, max_num_cols)
delim = random.choice(delims)
restricted_chars = ['\r', '\n'] + [delim]
key_col = random.randint(0, num_cols - 1)
good_keys = [
'Hello', 'Avada Kedavra ', ' ??????', '128', '3q295 fa#(@*$*)',
' abcdefg ', 'NR', 'a1', 'a2'
]
input_table = list()
for r in rbql.xrange6(num_rows):
input_table.append(list())
for c in rbql.xrange6(num_cols):
if c != key_col:
input_table[-1].append(
make_random_csv_entry(0, 20, restricted_chars))
else:
input_table[-1].append(random.choice(good_keys))
output_table = list()
target_key = random.choice(good_keys)
if random.choice([True, False]):
sql_op = '!='
output_table = [
row for row in input_table if row[key_col] != target_key
]
else:
sql_op = '=='
output_table = [
row for row in input_table if row[key_col] == target_key
]
query = 'select * where a{} {} "{}"'.format(key_col + 1, sql_op,
target_key)
return (input_table, query, output_table, delim)
def make_random_csv_entry(min_len, max_len, restricted_chars):
strlen = random.randint(min_len, max_len)
char_set = list(range(256))
restricted_chars = [ord(c) for c in restricted_chars]
char_set = [c for c in char_set if c not in restricted_chars]
data = list()
for i in rbql.xrange6(strlen):
data.append(random.choice(char_set))
pseudo_latin = bytes(bytearray(data)).decode('latin-1')
return pseudo_latin
def test_random_bin_tables(self):
test_name = 'test_random_bin_tables'
for subtest in rbql.xrange6(50):
input_table, query, canonic_table, delim = generate_random_scenario(
12, 12, ['\t', ',', ';'])
test_table = run_conversion_test_py(query,
input_table,
test_name,
delim=delim)
self.compare_tables(canonic_table, test_table)
test_table = run_conversion_test_js(query,
input_table,
test_name,
delim=delim)
self.compare_tables(canonic_table, test_table)
def compare_tables(self, canonic_table, test_table):
self.assertEqual(len(canonic_table), len(test_table))
for i in rbql.xrange6(len(canonic_table)):
self.assertEqual(len(canonic_table[i]), len(test_table[i]))
self.assertEqual(canonic_table[i], test_table[i])
self.assertEqual(canonic_table, test_table)