def join(res1, res2):
conn = joins.pop(0)
tb1 = res1[0]
col_order_1 = res1[1]
tb2 = res2[0]
col_order_2 = res2[1]
if conn == ' UNETELO CN ':
return man.unite_tables(tb1, tb2, col_order_1, col_order_2, False)
elif conn == ' UNETELO TODO CN ':
return man.unite_tables(tb1, tb2, col_order_1, col_order_2, True)
elif conn == ' COMUN CN ':
return man.intersect_tables(tb1, tb2, col_order_1, col_order_2)
elif conn == ' SACALE ':
return man.left_outer_intersect(tb1, tb2, col_order_1, col_order_2)
else:
raise ValueError('Do not recognize connective')
def test_unite_tables_with_repeat(self):
expected = [('a', 1, 0.1), ('b', 2, 0.2), ('c', 3, 0.3), ('d', 4, 0.4),
('e', 5, 0.5), ('a', 1, 0.1), ('b', 2, 0.2), ('c', 3, 0.3),
('i', 9, 0.9), ('j', 10, 1.0)]
test = man.unite_tables(self.tb51, self.tb52, self.tb51_order,
self.tb52_order, True)
test_tb = test[0]
# Check column entries, row order may be different, but change is
# uniform across all columns
str_col = test_tb['str']
int_col = test_tb['int']
float_col = test_tb['float']
i = str_col.index('a')
j = str_col.index('b')
k = str_col.index('c')
l = str_col.index('d')
m = str_col.index('e')
n = str_col.index('i')
o = str_col.index('j')
# Same row as a
self.assertEqual(int_col[i], 1)
self.assertEqual(float_col[i], 0.1)
# Same row as b
self.assertEqual(int_col[j], 2)
self.assertEqual(float_col[j], 0.2)
# Same row as c
self.assertEqual(int_col[k], 3)
self.assertEqual(float_col[k], 0.3)
# Same row as d
self.assertEqual(int_col[l], 4)
self.assertEqual(float_col[l], 0.4)
# Same row as e
self.assertEqual(int_col[m], 5)
self.assertEqual(float_col[m], 0.5)
# Same row as n
self.assertEqual(int_col[n], 9)
self.assertEqual(float_col[n], 0.9)
# Same row as o
self.assertEqual(int_col[o], 10)
self.assertEqual(float_col[o], 1.0)
# Count that there are repeats
counter_str = col.Counter(str_col)
self.assertEqual(counter_str['a'], 2)
self.assertEqual(counter_str['b'], 2)
self.assertEqual(counter_str['c'], 2)
counter_int = col.Counter(int_col)
self.assertEqual(counter_int[1], 2)
self.assertEqual(counter_int[2], 2)
self.assertEqual(counter_int[3], 2)
counter_float = col.Counter(float_col)
self.assertEqual(counter_float[0.1], 2)
self.assertEqual(counter_float[0.2], 2)
self.assertEqual(counter_float[0.3], 2)
def test_unite_tables_with_repeat(self):
expected = [('a', 1, 0.1), ('b', 2, 0.2), ('c', 3, 0.3), ('d', 4, 0.4),
('e', 5, 0.5), ('a', 1, 0.1), ('b', 2, 0.2), ('c', 3, 0.3),
('i', 9, 0.9), ('j', 10, 1.0)]
test = man.unite_tables(self.tb51, self.tb52,
self.tb51_order, self.tb52_order, True)
test_tb = test[0]
# Check column entries, row order may be different, but change is
# uniform across all columns
str_col = test_tb['str']
int_col = test_tb['int']
float_col = test_tb['float']
i = str_col.index('a')
j = str_col.index('b')
k = str_col.index('c')
l = str_col.index('d')
m = str_col.index('e')
n = str_col.index('i')
o = str_col.index('j')
# Same row as a
self.assertEqual(int_col[i], 1)
self.assertEqual(float_col[i], 0.1)
# Same row as b
self.assertEqual(int_col[j], 2)
self.assertEqual(float_col[j], 0.2)
# Same row as c
self.assertEqual(int_col[k], 3)
self.assertEqual(float_col[k], 0.3)
# Same row as d
self.assertEqual(int_col[l], 4)
self.assertEqual(float_col[l], 0.4)
# Same row as e
self.assertEqual(int_col[m], 5)
self.assertEqual(float_col[m], 0.5)
# Same row as n
self.assertEqual(int_col[n], 9)
self.assertEqual(float_col[n], 0.9)
# Same row as o
self.assertEqual(int_col[o], 10)
self.assertEqual(float_col[o], 1.0)
# Count that there are repeats
counter_str = col.Counter(str_col)
self.assertEqual(counter_str['a'], 2)
self.assertEqual(counter_str['b'], 2)
self.assertEqual(counter_str['c'], 2)
counter_int = col.Counter(int_col)
self.assertEqual(counter_int[1], 2)
self.assertEqual(counter_int[2], 2)
self.assertEqual(counter_int[3], 2)
counter_float = col.Counter(float_col)
self.assertEqual(counter_float[0.1], 2)
self.assertEqual(counter_float[0.2], 2)
self.assertEqual(counter_float[0.3], 2)