def setUp(self):
self.first_name_col = DataColumn(["Stefie", "Piotr"])
self.last_name_col = DataColumn(["Tellex", "Mitros"])
self.language_col = DataColumn(["Ruby", "C"])
self.coding_friends = ColumnGroup([("FirstName", self.first_name_col),
("LastName" , self.last_name_col),
("Language" , self.language_col)])
def test_select(self):
"""Test creating a ColumnGroup by selecting columns"""
cg = self.coding_friends.select("Language", "FirstName")
self.assertEqual(cg, ColumnGroup([("Language", self.language_col),
("FirstName", self.first_name_col)]) )
cg2 = self.coding_friends.select("Language", ("FirstName", "first"))
self.assertEqual(cg2, ColumnGroup([("Language", self.language_col),
("first", self.first_name_col)]) )
def test_adding_col_groups(self):
"""Test that we can create new ColumnGroups by adding old ones"""
cg1 = ColumnGroup( zip(["firstName", "lastName"],
[self.first_name_col, self.last_name_col]))
cg2 = ColumnGroup( [("language", self.language_col)] )
cg3 = cg1 + cg2
self.assertEqual(cg3.column_names, ("firstName", "lastName", "language"))
self.assertEqual(cg3.columns, (self.first_name_col, self.last_name_col,
self.language_col))
def test_non_strings(self):
"""Test ColumnGroups with non-string values"""
num_col = DataColumn([100, 200, 300, 400])
bool_col = DataColumn([True, False, True, True])
none_col = DataColumn(["", None, None, ""])
names = ["numbers", "booleans", "nones" ]
cols = [ num_col , bool_col , none_col]
cg = ColumnGroup( zip(names, cols) )
print cg.to_csv()
def test_duplicate_col_name(self):
"""Don't allow duplicate column names"""
c1 = DataColumn(["one", "two"])
# This should be fine...
cg = ColumnGroup([("name1",c1), ("name 2", c1)])
# This should blow up because the names are not unique
self.assertRaises( ValueError,
lambda: ColumnGroup([("name1",c1), ("name1", c1)]) )
def test_sorted(self):
"""Test that we can create a ColumnGroup with columns sorted by column name."""
cg = self.coding_friends.sorted()
self.assertEqual(cg, ColumnGroup([("FirstName", self.first_name_col),
("Language" , self.language_col),
("LastName" , self.last_name_col)]))
# But it shouldn't have changed the original
self.assertEqual(self.coding_friends,
ColumnGroup([("FirstName", self.first_name_col),
("LastName" , self.last_name_col),
("Language" , self.language_col)]) )
def test_parse_csv(self):
"""Check normal CSV parsing with spaces"""
csv_text = """FirstName, LastName, FavoriteColor
Regina, Eum,Purple
David, Ormsbee,Blue """
cg = ColumnGroup.from_csv(csv_text)
self.assertEqual(cg.FirstName, cg[0])
self.assertEqual(cg.LastName, cg[1])
self.assertEqual(cg.FavoriteColor, cg[2])
self.assertEqual(cg.column_names,
('FirstName', 'LastName', 'FavoriteColor'))
self.assertEqual(cg.FirstName,
DataColumn(["Regina", "David"]) )
self.assertEqual(cg.LastName,
DataColumn(["Eum", "Ormsbee"]) )
self.assertEqual(cg.FavoriteColor,
DataColumn(["Purple", "Blue"]) )
def test_reject(self):
"""Test creating a ColumnGroup by rejecting columns"""
cg = self.coding_friends.reject("Language", "LastName", "Asparagus")
self.assertEqual(cg, ColumnGroup([("FirstName", self.first_name_col)]) )
def test_map_col_names(self):
"""Test that creating a new CG and mapping the column names works."""
cg = self.coding_friends.map_names({ "FirstName" : "first", "LastName" : "last"})
self.assertEqual(cg, ColumnGroup([("first", self.first_name_col),
("last", self.last_name_col),
("Language" , self.language_col)]) )
def test_empty(self):
"""Test that empty ColumnGroups are ok"""
cg = ColumnGroup(()) # empty column group
self.assertFalse(cg)
print cg
class TestColumnGroup(unittest.TestCase):
def setUp(self):
self.first_name_col = DataColumn(["Stefie", "Piotr"])
self.last_name_col = DataColumn(["Tellex", "Mitros"])
self.language_col = DataColumn(["Ruby", "C"])
self.coding_friends = ColumnGroup([("FirstName", self.first_name_col),
("LastName" , self.last_name_col),
("Language" , self.language_col)])
def test_duplicate_col_name(self):
"""Don't allow duplicate column names"""
c1 = DataColumn(["one", "two"])
# This should be fine...
cg = ColumnGroup([("name1",c1), ("name 2", c1)])
# This should blow up because the names are not unique
self.assertRaises( ValueError,
lambda: ColumnGroup([("name1",c1), ("name1", c1)]) )
def test_parse_csv(self):
"""Check normal CSV parsing with spaces"""
csv_text = """FirstName, LastName, FavoriteColor
Regina, Eum,Purple
David, Ormsbee,Blue """
cg = ColumnGroup.from_csv(csv_text)
self.assertEqual(cg.FirstName, cg[0])
self.assertEqual(cg.LastName, cg[1])
self.assertEqual(cg.FavoriteColor, cg[2])
self.assertEqual(cg.column_names,
('FirstName', 'LastName', 'FavoriteColor'))
self.assertEqual(cg.FirstName,
DataColumn(["Regina", "David"]) )
self.assertEqual(cg.LastName,
DataColumn(["Eum", "Ormsbee"]) )
self.assertEqual(cg.FavoriteColor,
DataColumn(["Purple", "Blue"]) )
def test_select(self):
"""Test creating a ColumnGroup by selecting columns"""
cg = self.coding_friends.select("Language", "FirstName")
self.assertEqual(cg, ColumnGroup([("Language", self.language_col),
("FirstName", self.first_name_col)]) )
cg2 = self.coding_friends.select("Language", ("FirstName", "first"))
self.assertEqual(cg2, ColumnGroup([("Language", self.language_col),
("first", self.first_name_col)]) )
# def test_selectf(self):
# """Test creating a ColumnGroup by rejecting a columns based on a
# function."""
# cg = self.coding_friends.select(lambda c: c.startswith('L'))
# self.assertEqual(cg, ColumnGroup([("LastName" , self.last_name_col),
# ("Language" , self.language_col)]))
def test_sorted(self):
"""Test that we can create a ColumnGroup with columns sorted by column name."""
cg = self.coding_friends.sorted()
self.assertEqual(cg, ColumnGroup([("FirstName", self.first_name_col),
("Language" , self.language_col),
("LastName" , self.last_name_col)]))
# But it shouldn't have changed the original
self.assertEqual(self.coding_friends,
ColumnGroup([("FirstName", self.first_name_col),
("LastName" , self.last_name_col),
("Language" , self.language_col)]) )
def test_map_col_names(self):
"""Test that creating a new CG and mapping the column names works."""
cg = self.coding_friends.map_names({ "FirstName" : "first", "LastName" : "last"})
self.assertEqual(cg, ColumnGroup([("first", self.first_name_col),
("last", self.last_name_col),
("Language" , self.language_col)]) )
def test_reject(self):
"""Test creating a ColumnGroup by rejecting columns"""
cg = self.coding_friends.reject("Language", "LastName", "Asparagus")
self.assertEqual(cg, ColumnGroup([("FirstName", self.first_name_col)]) )
# def test_rejectf(self):
# """Test creating a ColumnGroup by rejecting a columns based on a
# function."""
# pass
def test_non_strings(self):
"""Test ColumnGroups with non-string values"""
num_col = DataColumn([100, 200, 300, 400])
bool_col = DataColumn([True, False, True, True])
none_col = DataColumn(["", None, None, ""])
names = ["numbers", "booleans", "nones" ]
cols = [ num_col , bool_col , none_col]
cg = ColumnGroup( zip(names, cols) )
print cg.to_csv()
def test_empty(self):
"""Test that empty ColumnGroups are ok"""
cg = ColumnGroup(()) # empty column group
self.assertFalse(cg)
print cg
def test_adding_col_groups(self):
"""Test that we can create new ColumnGroups by adding old ones"""
cg1 = ColumnGroup( zip(["firstName", "lastName"],
[self.first_name_col, self.last_name_col]))
cg2 = ColumnGroup( [("language", self.language_col)] )
cg3 = cg1 + cg2
self.assertEqual(cg3.column_names, ("firstName", "lastName", "language"))
self.assertEqual(cg3.columns, (self.first_name_col, self.last_name_col,
self.language_col))
