def test_neg(self):
# - This column
# ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 1.88 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 1.83 | 130 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Flash | Male | 1.83 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 1.70 | 133 | Green | Black | | True | True |
# ---------------------------------------------------------------------------------
# | Superman | Male | 1.905 | 235 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Joker | Male | 1.83 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
dcf = BatFrame(DatasetTest.dataset_url)
s = (-dcf["Height"]).toPandas()
self.assertAlmostEqual(s["Batman"], -1.88)
self.assertAlmostEqual(s["Wonder Woman"], -1.83)
self.assertAlmostEqual(s["Flash"], -1.83)
self.assertAlmostEqual(s["Catwoman"], -1.70)
self.assertAlmostEqual(s["Superman"], -1.905)
self.assertAlmostEqual(s["Joker"], -1.83)
def test_rows(self):
bf = BatFrame(DatasetTest.dataset_url)
rows = bf.rows
self.assertEqual(len(rows), 6, rows)
self.assertTrue("Batman" in rows)
self.assertTrue("Wonder Woman" in rows)
self.assertTrue("Flash" in rows)
self.assertTrue("Catwoman" in rows)
self.assertTrue("Superman" in rows)
self.assertTrue("Joker" in rows)
def test_columns(self):
bf = BatFrame(DatasetTest.dataset_url)
columns = bf.columns
self.assertEqual(len(columns), 8, columns)
self.assertTrue("Gender" in columns)
self.assertTrue("Height" in columns)
self.assertTrue("Weight" in columns)
self.assertTrue("Eyes" in columns)
self.assertTrue("Hair" in columns)
self.assertTrue("Skin" in columns)
self.assertTrue("Good" in columns)
self.assertTrue("Bad" in columns)
def test_alignment(self):
bf = BatFrame(DatasetTest.dataset_url)
height = bf['Height'].values
weight = bf['Weight'].values
msg = "Got heights: {} \nand weights: {}\n Length of {} and {}"
msg = msg.format(height, weight, len(height), len(weight))
self.assertTrue(len(height) == len(weight), msg=msg)
expected = [("Batman", 188, 210), ("Wonder Woman", 183, 130),
("Flash", 183, 190), ("Catwoman", 170, 133),
("Superman", 190.5, 235), ("Joker", 183, 183)]
for t in zip(bf.rows, height, weight):
msg = "{} is not in the expected values {}".format(t, expected)
self.assertTrue(t in expected, msg=msg)
def test_unique(self):
bf = BatFrame(DatasetTest.dataset_url)
height = bf['Height'].unique()
e_height = [1.88, 1.83, 1.70, 1.905]
msg = "Got {} different values ({}) but expected 4".format(
len(height), height)
self.assertEqual(len(height), 4, msg=msg)
for h in height:
msg = "{} not in {}".format(h, e_height)
self.assertTrue(h in e_height, msg=msg)
weight = bf['Weight'].unique()
e_weight = [210, 130, 190, 133, 235, 183]
msg = "Got {} different values ({}) but expected 6".format(
len(weight), weight)
self.assertEqual(len(weight), 6, msg=msg)
for w in weight:
msg = "{} not in {}".format(w, e_weight)
self.assertTrue(w in e_weight, msg=msg)
def test_min(self):
# min(This column)
# ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 1.88 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 1.83 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 1.83 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 1.70 | 133 | Green | Black | | True | True |
# ---------------------------------------------------------------------------------
# | Superman | Male | 1.905 | 235 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Joker | Male | 1.83 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
bf = BatFrame(DatasetTest.dataset_url)
self.assertEqual(bf["Weight"].min(), 130)
self.assertEqual(min(bf["Weight"]), 130)
def test_head(self):
bf = BatFrame(DatasetTest.dataset_url)
head = bf['Gender'].head()
self.assertEqual(len(head), 5)
def test_ix(self):
#
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True |
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
bf = BatFrame(DatasetTest.dataset_url)
Wonder_Woman = bf.ix["Wonder Woman"]
self.assertEqual(Wonder_Woman['Gender'].values[0], "Female")
self.assertEqual(Wonder_Woman['Height'].values[0], 183)
self.assertEqual(Wonder_Woman['Weight'].values[0], 130)
self.assertEqual(Wonder_Woman['Eyes'].values[0], "Blue")
self.assertEqual(Wonder_Woman['Hair'].values[0], "Black")
self.assertTrue(Wonder_Woman['Good'].values[0])
self.assertFalse(Wonder_Woman['Bad'].values[0])
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True |
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
trinity = bf.ix[["Wonder Woman", "Batman", "Superman"]]
self.assertTupleEqual(trinity.toPandas().shape, (3, 7))
self.assertItemsEqual(trinity['Gender'].values, ["Female", "Male", "Male"])
self.assertItemsEqual(trinity['Height'].values, [183, 188, 190.5])
self.assertItemsEqual(trinity['Weight'].values, [130, 210, 235])
self.assertItemsEqual(trinity['Eyes'].values, ["Blue"]*3)
self.assertItemsEqual(trinity['Hair'].values, ["Black"]*3)
self.assertItemsEqual(trinity['Good'].values, [True]*3)
self.assertItemsEqual(trinity['Bad'].values, [False]*3)
# These columns
# ⇓ ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True |
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
trinity = bf.ix[["Batman", "Superman"], ["Gender", "Height"]].toPandas()
self.assertTupleEqual(trinity.shape, (2, 2))
self.assertItemsEqual(trinity['Gender'].values, ["Male", "Male"])
self.assertItemsEqual(trinity['Height'].values, [188, 190.5])
def test_le(self):
bf = BatFrame(DatasetTest.dataset_url)
# Testing INTEGER lesser than or equal
# This column <= 183
# ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True | ⇐ This row
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True | ⇐ This row
# ---------------------------------------------------------------------------------
def helper(rowName):
df.ix[rowName]
df = bf[bf["Height"] <= 183].toPandas()
self.assertEqual(len(df), 4)
# Only Catwoman is smaller than 183 cm
self.assertTupleEqual(df.ix["Wonder Woman"].shape, (8, ))
self.assertTupleEqual(df.ix["Flash"].shape, (8, ))
self.assertTupleEqual(df.ix["Catwoman"].shape, (8, ))
self.assertTupleEqual(df.ix["Joker"].shape, (8, ))
# Batman and Superman are the biggest
self.assertRaises(KeyError, helper, "Batman")
self.assertRaises(KeyError, helper, "Superman")
# Testing FLOAT lesser than or equal
# This column <= 190.5
# ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True | ⇐ This row
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True | ⇐ This row
# ---------------------------------------------------------------------------------
df = bf[bf["Height"] <= 190.5].toPandas()
self.assertEqual(len(df), 6)
# All but Superman are smaller than 190.5 cm
self.assertTupleEqual(df.ix["Batman"].shape, (8, ))
self.assertTupleEqual(df.ix["Catwoman"].shape, (8, ))
self.assertTupleEqual(df.ix["Wonder Woman"].shape, (8, ))
self.assertTupleEqual(df.ix["Flash"].shape, (8, ))
self.assertTupleEqual(df.ix["Joker"].shape, (8, ))
self.assertTupleEqual(df.ix["Superman"].shape, (8, ))
# No one is taller than 190.5
# Testing Column le
# This le This
# ⇓ <= ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True |
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True | ⇐ This row
# ---------------------------------------------------------------------------------
df = bf[bf["Height"] <= bf["Weight"]].toPandas()
self.assertEqual(len(df), 4)
# The bulky ones (Height > Weight)
self.assertTupleEqual(df.ix["Batman"].shape, (8, ))
self.assertTupleEqual(df.ix["Flash"].shape, (8, ))
self.assertTupleEqual(df.ix["Superman"].shape, (8, ))
self.assertTupleEqual(df.ix["Joker"].shape, (8, ))
# The thin and tall
self.assertRaises(KeyError, helper, "Wonder Woman")
self.assertRaises(KeyError, helper, "Catwoman")
def test_ge(self):
bf = BatFrame(DatasetTest.dataset_url)
# Testing INTEGER greater than or equal
# This column >= 183
# ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True |
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True | ⇐ This row
# ---------------------------------------------------------------------------------
def helper(rowName):
df.ix[rowName]
df = bf[bf["Height"] >= 183].toPandas()
self.assertEqual(len(df), 5)
# Batman and Superman are taller or equal to 183 cm
self.assertTupleEqual(df.ix["Batman"].shape, (8, ))
self.assertTupleEqual(df.ix["Wonder Woman"].shape, (8, ))
self.assertTupleEqual(df.ix["Flash"].shape, (8, ))
self.assertTupleEqual(df.ix["Joker"].shape, (8, ))
self.assertTupleEqual(df.ix["Superman"].shape, (8, ))
# Only Catwoman is stricly shorter than 183 cm
self.assertRaises(KeyError, helper, "Catwoman")
# Testing FLOAT greater than or equal
# This column >= 190.5
# ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True |
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
df = bf[bf["Height"] >= 190.5].toPandas()
self.assertEqual(len(df), 1)
# Only Superman is taller or equal to 190.5 cm
self.assertTupleEqual(df.ix["Superman"].shape, (7, ))
# Everyone else are smaller
self.assertRaises(KeyError, helper, "Batman")
self.assertRaises(KeyError, helper, "Catwoman")
self.assertRaises(KeyError, helper, "Wonder Woman")
self.assertRaises(KeyError, helper, "Flash")
self.assertRaises(KeyError, helper, "Joker")
# Testing Column ge
# This ge This
# ⇓ >= ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True | ⇐ This row
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True | ⇐ This row
# ---------------------------------------------------------------------------------
df = bf[bf["Height"] >= bf["Weight"]].toPandas()
self.assertEqual(len(df), 3)
# The tall and thin (Height > Weight)
self.assertTupleEqual(df.ix["Wonder Woman"].shape, (8, ))
self.assertTupleEqual(df.ix["Catwoman"].shape, (8, ))
self.assertTupleEqual(df.ix["Joker"].shape, (8, ))
# the bulky ones
self.assertRaises(KeyError, helper, "Batman")
self.assertRaises(KeyError, helper, "Flash")
self.assertRaises(KeyError, helper, "Superman")
def test_ne(self):
# This column != Male
# ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True | ⇐ This row
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
# Testing string equality
bf = BatFrame(DatasetTest.dataset_url)
df = bf[bf["Gender"] != "Male"].toPandas()
self.assertEqual(len(df), 2)
self.assertTupleEqual(df.ix["Wonder Woman"].shape, (7, ))
self.assertTupleEqual(df.ix["Catwoman"].shape, (7, ))
def helper(rowName):
df.ix[rowName]
self.assertRaises(KeyError, helper, "Batman")
self.assertRaises(KeyError, helper, "Flash")
self.assertRaises(KeyError, helper, "Joker")
self.assertRaises(KeyError, helper, "Superman")
# Testing INTEGER equality
# This column != 183
# ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True | ⇐ This row
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
df = bf[bf["Height"] != 183].toPandas()
self.assertEqual(len(df), 3)
# Wonder Woman, Flash and Joker are all 183cm tall
self.assertTupleEqual(df.ix["Batman"].shape, (7, ))
self.assertTupleEqual(df.ix["Catwoman"].shape, (7, ))
self.assertTupleEqual(df.ix["Superman"].shape, (7, ))
# Batman, Catwoman and Superman are either taller or smaller
self.assertRaises(KeyError, helper, "Wonder Woman")
self.assertRaises(KeyError, helper, "Flash")
self.assertRaises(KeyError, helper, "Joker")
# Testing FLOAT equality
# This column != 190.5
# ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True | ⇐ This row
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True | ⇐ This row
# ---------------------------------------------------------------------------------
df = bf[bf["Height"] != 190.5].toPandas()
self.assertEqual(len(df), 5)
# Only Superman is 190.5 cm tall
self.assertTupleEqual(df.ix["Batman"].shape, (8, ))
self.assertTupleEqual(df.ix["Catwoman"].shape, (8, ))
self.assertTupleEqual(df.ix["Wonder Woman"].shape, (8, ))
self.assertTupleEqual(df.ix["Flash"].shape, (8, ))
self.assertTupleEqual(df.ix["Joker"].shape, (8, ))
# Everyone else are either taller or smaller
self.assertRaises(KeyError, helper, "Superman")
# Testing Column inequality
# Those column not equal
# ⇓ != ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True | ⇐ This row
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
df = bf[bf["Eyes"] != bf["Hair"]].toPandas()
self.assertEqual(len(df), 5)
# Everyone but the Joker don't have matching hair and eye color
self.assertTupleEqual(df.ix["Batman"].shape, (7, ))
self.assertTupleEqual(df.ix["Wonder Woman"].shape, (7, ))
self.assertTupleEqual(df.ix["Flash"].shape, (7, ))
self.assertTupleEqual(df.ix["Catwoman"].shape, (7, ))
self.assertTupleEqual(df.ix["Superman"].shape, (7, ))
# Joker, all alone in his corner
self.assertRaises(KeyError, helper, "Joker")
def test_complex_slicing(self):
# True
# ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True | ⇐ This row
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True | ⇐ This row
# ---------------------------------------------------------------------------------
bf = BatFrame(DatasetTest.dataset_url)
filtered = bf[bf["Bad"]]
print(filtered)
self.assertTupleEqual(filtered.toPandas().shape, (2, 8))
rowNames = filtered.rows
expected = ["Catwoman", "Joker"]
for hero in expected:
msg = "{} expected to be in the results".format(hero)
self.assertTrue(hero in rowNames, msg=msg)
# Not True
# ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True |
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
filtered = bf[~bf["Bad"]]
print(filtered)
self.assertTupleEqual(filtered.toPandas().shape, (4, 7))
rowNames = filtered.rows
expected = ["Batman", "Wonder Woman", "Flash", "Superman"]
for hero in expected:
msg = "{} expected to be in the results".format(hero)
self.assertTrue(hero in rowNames, msg=msg)
# True & True
# ⇓ ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True | ⇐ This row
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
filtered = bf[bf["Bad"] & bf["Good"]]
print(filtered)
self.assertTupleEqual(filtered.toPandas().shape, (1, 7))
rowNames = filtered.rows
expected = ["Catwoman"]
for hero in expected:
msg = "{} expected to be in the results".format(hero)
self.assertTrue(hero in rowNames, msg=msg)
# True & Not True
# ⇓ ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True | ⇐ This row
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True | ⇐ This row
# ---------------------------------------------------------------------------------
filtered = bf[bf["Bad"] & ~bf["Good"]]
print(filtered)
self.assertTupleEqual(filtered.toPandas().shape, (1, 8))
rowNames = filtered.rows
expected = ["Joker"]
for hero in expected:
msg = "{} expected to be in the results".format(hero)
self.assertTrue(hero in rowNames, msg=msg)
# True | True
# ⇓ ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True | ⇐ This row
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True | ⇐ This row
# ---------------------------------------------------------------------------------
filtered = bf[bf["Bad"] | bf["Good"]]
print(filtered)
self.assertTupleEqual(filtered.toPandas().shape, (6, 8))
rowNames = filtered.rows
expected = [
"Catwoman", "Batman", "Wonder Woman", "Flash", "Superman", "Joker"
]
for hero in expected:
msg = "{} expected to be in the results".format(hero)
self.assertTrue(hero in rowNames, msg=msg)
# True | Not True
# ⇓ ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True | ⇐ This row
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True | ⇐ This row
# ---------------------------------------------------------------------------------
filtered = bf[bf["Bad"] | ~bf["Good"]]
print(filtered)
self.assertTupleEqual(filtered.toPandas().shape, (2, 8))
rowNames = filtered.rows
expected = ["Catwoman", "Joker"]
for hero in expected:
msg = "{} expected to be in the results".format(hero)
self.assertTrue(hero in rowNames, msg=msg)
def test_sort(self):
# Sort on this column
# ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True |
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
bf = BatFrame(DatasetTest.dataset_url)
sort = bf.sort("Weight") # ascending=True is default
rows = sort.rows
expected_rows = [
"Wonder Woman", "Catwoman", "Joker", "Flash", "Batman", "Superman"
]
for row, e_row in zip(rows, expected_rows):
self.assertEqual(row, e_row)
values = sort["Weight"].values
expected_values = [130, 133, 183, 190, 210, 235]
for value, e_value in zip(values, expected_values):
self.assertEqual(value, e_value)
# Reverse sort aka descending
sort = bf.sort("Weight", ascending=False)
rows = sort.rows
for row, e_row in zip(rows, reversed(expected_rows)):
self.assertEqual(row, e_row)
values = sort["Weight"].values
for value, e_value in zip(values, reversed(expected_values)):
self.assertEqual(value, e_value)
# Sort on those column
# ⇓ ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True |
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
bf = BatFrame(DatasetTest.dataset_url)
sort = bf.sort(["Height", "Weight"]) # ascending=True is default
rows = sort.rows
expected_rows = [
"Catwoman", "Wonder Woman", "Joker", "Flash", "Batman", "Superman"
]
for row, e_row in zip(rows, expected_rows):
self.assertEqual(row, e_row)
heights = sort["Height"].values
expected_height = [170, 183, 183, 183, 188, 190.5]
for height, e_height in zip(heights, expected_height):
self.assertEqual(height, e_height)
weights = sort["Weight"].values
expected_weight = [133, 130, 183, 190, 210, 235]
for weight, e_weight in zip(weights, expected_weight):
self.assertEqual(weight, e_weight)
# Reverse sort aka descending. Both are descending
sort = bf.sort(["Height", "Weight"], ascending=False)
rows = sort.rows
for row, e_row in zip(rows, reversed(expected_rows)):
self.assertEqual(row, e_row)
heights = sort["Height"].values
for height, e_height in zip(heights, reversed(expected_height)):
self.assertEqual(height, e_height)
weights = sort["Weight"].values
for weight, e_weight in zip(weights, reversed(expected_weight)):
self.assertEqual(weight, e_weight)
# Different sorting order
sort = bf.sort(["Height", "Weight"], [True, False])
rows = sort.rows
expected_rows = [
"Catwoman", "Flash", "Joker", "Wonder Woman", "Batman", "Superman"
]
for row, e_row in zip(rows, expected_rows):
self.assertEqual(row, e_row)
heights = sort["Height"].values
expected_height = [170, 183, 183, 183, 188, 190.5]
for height, e_height in zip(heights, expected_height):
self.assertEqual(height, e_height)
weights = sort["Weight"].values
expected_weight = [133, 190, 183, 130, 210, 235]
for weight, e_weight in zip(weights, expected_weight):
self.assertEqual(weight, e_weight)
# Sort on this column
# ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True |
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
sort = bf.sort("Skin")
rows = sort.rows
expected_rows = [
"Superman", "Wonder Woman", "Catwoman", "Batman", "Flash", "Joker"
]
for row, e_row in zip(rows, expected_rows):
self.assertEqual(row, e_row)
heights = sort["Height"].values
expected_height = [190.5, 183, 170, 188, 183, 183]
for height, e_height in zip(heights, expected_height):
self.assertEqual(height, e_height)
weights = sort["Weight"].values
expected_weight = [235, 130, 133, 210, 190, 183]
for weight, e_weight in zip(weights, expected_weight):
self.assertEqual(weight, e_weight)
def test_head(self):
bf = BatFrame(DatasetTest.dataset_url)
head = bf.head().toPandas()
self.assertTupleEqual(head.shape, (5, 8))
def test_slicing(self):
# VERY IMPORTANT
# Slicing calls the *limit* REST parameter which will limit the number
# of rows returned based on the **hash** of the row name. The insertion
# order **does not** matter
bf = BatFrame(DatasetTest.dataset_url)
# Testing slicing [:3]
# ALL THE COLUMNS
# ⇓ ⇓ ⇓ ⇓ ⇓ ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False |
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True | ⇐ This row
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
df = bf[:3].toPandas()
# Right now the column Skin does not come out because no result have
# that column
self.assertTupleEqual(df.shape, (3, 7))
rowNames = df.index.values.tolist()
expected = expected = ["Superman", "Wonder Woman", "Catwoman"]
for hero in expected:
msg = "{} expected to be in the results".format(hero)
self.assertTrue(hero in rowNames, msg=msg)
# for i, hero in enumerate(expected):
# msg = "\n-------------Expecting---------------\n"
# msg += " ".join(expected)
# msg += "\n\n--------------Instead got--------------\n"
# msg += str(" ".join(rowNames))
# self.assertEqual(rowNames[i], hero, msg=msg)
# Testing slicing [2:5]
# ALL THE COLUMNS
# ⇓ ⇓ ⇓ ⇓ ⇓ ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True |
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
df = bf[2:5].toPandas()
self.assertTupleEqual(df.shape, (4, 8))
rowNames = df.index.values.tolist()
expected = ["Flash", "Batman", "Catwoman", "Joker"]
for hero in expected:
msg = "{} expected to be in the results".format(hero)
self.assertTrue(hero in rowNames, msg=msg)
# Testing slicing [2:5]
# ALL THE COLUMNS
# ⇓ ⇓ ⇓ ⇓ ⇓ ⇓
# ---------------------------------------------------------------------------------
# | | Gender | Height | Weight | Eyes | Hair | Skin | Good | Bad |
# ---------------------------------------------------------------------------------
# | Batman | Male | 188 | 210 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Wonder Woman | Female | 183 | 130 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Flash | Male | 183 | 190 | Green | Red | | True | False |
# ---------------------------------------------------------------------------------
# | Catwoman | Female | 170 | 133 | Green | Black | | True | True |
# ---------------------------------------------------------------------------------
# | Superman | Male | 190.5 | 235 | Blue | Black | | True | False | ⇐ This row
# ---------------------------------------------------------------------------------
# | Joker | Male | 183 | 183 | Green | Green | White | False | True |
# ---------------------------------------------------------------------------------
df = bf[4:].toPandas()
self.assertTupleEqual(df.shape, (2, 8))
rowNames = df.index.values.tolist()
expected = ["Flash", "Joker"]
for hero in expected:
msg = "{} expected to be in the results".format(hero)
self.assertTrue(hero in rowNames, msg=msg)