def test12(self):
df=DataFrame()
df['DUM']=range(48) # Shouldn't complain
self.assertEqual(df.keys(),['DUM'])
df['DUM']=['A' for i in range(48)] # Shouldn't complain
self.assertEqual(df.keys(),['DUM'])
self.assertEqual(df._sqltypesdict['DUM'],'text')
def test_kn(self):
df = DataFrame()
df.read_tbl('data/example.csv')
y = [23]*len(df['X'])
df['X'] = y
self.assertEqual(df.keys(), ['CASE', 'TIME', 'CONDITION', 'X'])
def test4(self):
df = DataFrame()
df[1] = list(range(100))
df[2] = ['bob' for i in range(100)]
df[3] = [i * 1.234232 for i in range(100)]
df[4] = ['bob' for i in range(50)] + list(range(50))
with self.assertRaises(TypeError) as cm:
df._build_sqlite3_tbl(list(df.keys())[:2], 42)
self.assertEqual(str(cm.exception), "'int' object is not iterable")
def test4(self):
df = DataFrame()
df[1] = range(100)
df[2] = ["bob" for i in range(100)]
df[3] = [i * 1.234232 for i in range(100)]
df[4] = ["bob" for i in range(50)] + range(50)
with self.assertRaises(TypeError) as cm:
df._build_sqlite3_tbl(df.keys()[:2], 42)
self.assertEqual(str(cm.exception), "'int' object is not iterable")
def test3(self):
tupa = ('a1','a2','a3')
tupb = ('a1','b2','b3')
df=DataFrame()
df.insert([(('a1','a2','a3'),34), (('a1','b2','b3'),1)])
df.insert([(('a1','a2','a3'),34), (('a1','b2','b3'),2)])
df.insert([(('a1','a2','a3'),42), (('a1','b2','b3'),3)])
namea,nameb = df.keys()
self.assertEqual(namea, tupa)
self.assertEqual(nameb, tupb)
def test31(self):
df = DataFrame()
df[1] = range(100)
df[2] = ["bob" for i in range(100)]
df[3] = [i * 1.234232 for i in range(100)]
df[4] = ["bob" for i in range(50)] + range(50)
shuffle(df[1])
shuffle(df[2])
shuffle(df[3])
with self.assertRaises(KeyError) as cm:
df._build_sqlite3_tbl(df.keys()[:2], ['4 != "bob"'])
self.assertEqual(str(cm.exception), "'4'")
def test31(self):
df = DataFrame()
df[1] = list(range(100))
df[2] = ['bob' for i in range(100)]
df[3] = [i * 1.234232 for i in range(100)]
df[4] = ['bob' for i in range(50)] + list(range(50))
shuffle(df[1])
shuffle(df[2])
shuffle(df[3])
with self.assertRaises(KeyError) as cm:
df._build_sqlite3_tbl(list(df.keys())[:2], ['4 != "bob"'])
self.assertEqual(str(cm.exception), "'4'")
def test1(self):
R = {'DUM': 'integer',
'COURSE': 'text',
'ERROR': 'integer',
'MODEL': 'text',
'TIMEOFDAY': 'text',
'SUBJECT': 'integer'}
df=DataFrame()
df.read_tbl('data/error~subjectXtimeofdayXcourseXmodel_MISSING.csv')
df['DUM']=range(48) # Shouldn't complain
self.assertEqual(df.keys(),
['SUBJECT', 'TIMEOFDAY', 'COURSE', 'MODEL', 'ERROR', 'DUM'])
for k in R:
self.assertEqual(df._sqltypesdict[k],R[k])
class Test_del_item(unittest.TestCase):
def setUp(self):
self.df=DataFrame()
self.df.read_tbl('data/error~subjectXtimeofdayXcourseXmodel_MISSING.csv')
del self.df['COURSE']
def test0(self):
self.assertEqual(self.df.keys(),
['SUBJECT', 'TIMEOFDAY', 'MODEL', 'ERROR'])
def test1(self):
self.assertEqual(list(self.df.types()),
['integer', 'text', 'text', 'integer'])
def test3(self):
self.assertEqual(len(self.df), 4)
def test22(self):
"""test with string keys and where condition"""
df = DataFrame()
df['1'] = list(range(100))
df['2'] = ['bob' for i in range(100)]
df['3'] = [i * 1.234232 for i in range(100)]
df['4'] = ['bob' for i in range(50)] + list(range(50))
shuffle(df['1'])
shuffle(df['2'])
shuffle(df['3'])
df._build_sqlite3_tbl(list(df.keys())[:2], ['4 not in ("bob")'])
df._execute('select * from TBL')
for i, (a, b) in enumerate(df.cur):
self.assertEqual(a, df['1'][i + 50])
self.assertEqual(b, df['2'][i + 50])
def test22(self):
"""test with string keys and where condition"""
df = DataFrame()
df["1"] = range(100)
df["2"] = ["bob" for i in range(100)]
df["3"] = [i * 1.234232 for i in range(100)]
df["4"] = ["bob" for i in range(50)] + range(50)
shuffle(df["1"])
shuffle(df["2"])
shuffle(df["3"])
df._build_sqlite3_tbl(df.keys()[:2], ['4 not in ("bob")'])
df._execute("select * from TBL")
for i, (a, b) in enumerate(df.cur):
self.assertEqual(a, df["1"][i + 50])
self.assertEqual(b, df["2"][i + 50])
def test3(self):
"""test with string keys and tuple where condition"""
df = DataFrame()
df[1] = range(100)
df[2] = ["bob" for i in range(100)]
df[3] = [i * 1.234232 for i in range(100)]
df[4] = ["bob" for i in range(50)] + range(50)
shuffle(df[1])
shuffle(df[2])
shuffle(df[3])
df._build_sqlite3_tbl(df.keys()[:2], [(4, "!=", "bob")])
df._execute("select * from TBL")
for i, (a, b) in enumerate(df.cur):
self.assertEqual(a, df[1][i + 50])
self.assertEqual(b, df[2][i + 50])
def test3(self):
"""test with string keys and tuple where condition"""
df = DataFrame()
df[1] = list(range(100))
df[2] = ['bob' for i in range(100)]
df[3] = [i * 1.234232 for i in range(100)]
df[4] = ['bob' for i in range(50)] + list(range(50))
shuffle(df[1])
shuffle(df[2])
shuffle(df[3])
df._build_sqlite3_tbl(list(df.keys())[:2], [(4, '!=', 'bob')])
df._execute('select * from TBL')
for i, (a, b) in enumerate(df.cur):
self.assertEqual(a, df[1][i + 50])
self.assertEqual(b, df[2][i + 50])
def test1(self):
"""test with integer keys subset of table"""
df = DataFrame()
df[1] = range(100)
df[2] = ["bob" for i in range(100)]
df[3] = [i * 1.234232 for i in range(100)]
df[4] = ["bob" for i in range(50)] + range(50)
shuffle(df[1])
shuffle(df[2])
shuffle(df[3])
shuffle(df[4])
df._build_sqlite3_tbl(df.keys()[:2])
df._execute("select * from TBL")
for i, (a, b) in enumerate(df.cur):
self.assertEqual(a, df[1][i])
self.assertEqual(b, df[2][i])
def test1(self):
"""test with integer keys subset of table"""
df = DataFrame()
df[1] = list(range(100))
df[2] = ['bob' for i in range(100)]
df[3] = [i * 1.234232 for i in range(100)]
df[4] = ['bob' for i in range(50)] + list(range(50))
shuffle(df[1])
shuffle(df[2])
shuffle(df[3])
shuffle(df[4])
df._build_sqlite3_tbl(list(df.keys())[:2])
df._execute('select * from TBL')
for i, (a, b) in enumerate(df.cur):
self.assertEqual(a, df[1][i])
self.assertEqual(b, df[2][i])
def test2(self):
df1=DataFrame()
df2=DataFrame()
df1.read_tbl('data/words~ageXcondition.csv')
df2.read_tbl('data/words~ageXcondition.csv')
M=df1.shape()[1]
# this should work
df1.attach(df2)
# df1 should have twice as many rows now
self.assertEqual(df1.shape()[1]/2,df2.shape()[1])
# go through and check data
for i in range(M):
for n in df1.keys():
if _isfloat(df1[n][i]):
self.assertAlmostEqual(df1[n][i],df1[n][M+i])
else:
self.assertEqual(df1[n][i],df1[n][M+i])
def test01(self):
"""test with integer keys"""
df = DataFrame()
df[1] = range(100)
df[2] = ['bob' for i in range(100)]
df[3] = [i * 1.234232 for i in range(100)]
df[4] = ['bob' for i in range(50)] + range(50)
shuffle(df[1])
shuffle(df[2])
shuffle(df[3])
shuffle(df[4])
df._build_sqlite3_tbl(df.keys())
df._execute('select * from TBL')
for i, (a, b, c, d) in enumerate(df.cur):
self.assertEqual(a, df[1][i])
self.assertEqual(b, df[2][i])
self.assertEqual(c, df[3][i])
self.assertEqual(d, str(df[4][i]))
def test2(self):
df1 = DataFrame()
df2 = DataFrame()
df1.read_tbl('data/words~ageXcondition.csv')
df2.read_tbl('data/words~ageXcondition.csv')
M = df1.shape()[1]
# this should work
df1.attach(df2)
# df1 should have twice as many rows now
self.assertEqual(df1.shape()[1] / 2, df2.shape()[1])
# go through and check data
for i in range(M):
for n in list(df1.keys()):
if _isfloat(df1[n][i]):
self.assertAlmostEqual(df1[n][i], df1[n][M + i])
else:
self.assertEqual(df1[n][i], df1[n][M + i])
def test02(self):
"""test with tuple keys"""
df = DataFrame()
## df.PRINTQUERIES = True
df[(1,)] = range(100)
df[(2,)] = ["bob" for i in range(100)]
df[(3,)] = [i * 1.234232 for i in range(100)]
df[(4,)] = ["bob" for i in range(50)] + range(50)
shuffle(df[(1,)])
shuffle(df[(2,)])
shuffle(df[(3,)])
shuffle(df[(4,)])
df._build_sqlite3_tbl(df.keys())
df._execute("select * from TBL")
for i, (a, b, c, d) in enumerate(df.cur):
self.assertEqual(a, df[(1,)][i])
self.assertEqual(b, df[(2,)][i])
self.assertEqual(c, df[(3,)][i])
self.assertEqual(d, str(df[(4,)][i]))
def test02(self):
"""test with tuple keys"""
df = DataFrame()
## df.PRINTQUERIES = True
df[(1, )] = list(range(100))
df[(2, )] = ['bob' for i in range(100)]
df[(3, )] = [i * 1.234232 for i in range(100)]
df[(4, )] = ['bob' for i in range(50)] + list(range(50))
shuffle(df[(1, )])
shuffle(df[(2, )])
shuffle(df[(3, )])
shuffle(df[(4, )])
df._build_sqlite3_tbl(list(df.keys()))
df._execute('select * from TBL')
for i, (a, b, c, d) in enumerate(df.cur):
self.assertEqual(a, df[(1, )][i])
self.assertEqual(b, df[(2, )][i])
self.assertEqual(c, df[(3, )][i])
self.assertEqual(d, str(df[(4, )][i]))
def test00(self):
"""test with string keys"""
df = DataFrame()
## df.PRINTQUERIES=True
df["1"] = range(100)
df["2"] = ["bob" for i in range(100)]
df["3"] = [i * 1.234232 for i in range(100)]
df["4"] = ["bob" for i in range(50)] + range(50)
df["5"] = np.sqrt(df["3"] * 100.0)
## print(df)
shuffle(df["1"])
shuffle(df["2"])
shuffle(df["3"])
df._build_sqlite3_tbl(df.keys())
df._execute("select * from TBL")
for i, (a, b, c, d, e) in enumerate(df.cur):
self.assertEqual(a, df["1"][i])
self.assertEqual(b, df["2"][i])
self.assertEqual(c, df["3"][i])
self.assertEqual(d, str(df["4"][i]))
def test00(self):
"""test with string keys"""
df = DataFrame()
## df.PRINTQUERIES=True
df['1'] = list(range(100))
df['2'] = ['bob' for i in range(100)]
df['3'] = [i * 1.234232 for i in range(100)]
df['4'] = ['bob' for i in range(50)] + list(range(50))
df['5'] = np.sqrt(df['3'] * 100.)
## print(df)
shuffle(df['1'])
shuffle(df['2'])
shuffle(df['3'])
df._build_sqlite3_tbl(list(df.keys()))
df._execute('select * from TBL')
for i, (a, b, c, d, e) in enumerate(df.cur):
self.assertEqual(a, df['1'][i])
self.assertEqual(b, df['2'][i])
self.assertEqual(c, df['3'][i])
self.assertEqual(d, str(df['4'][i]))
def test2(self):
df = DataFrame()
df.read_tbl('data/iqbrainsize.txt', delimiter='\t')
cor = df.correlation(df.keys())
def test11(self):
df=DataFrame()
df['DUM']=range(48) # Shouldn't complain
self.assertEqual(df.keys(),['DUM'])
def test21(self):
df=DataFrame()
df[1]=range(48)
df[2]=['A' for i in range(48)]
self.assertEqual(df.keys(),[1,2])
def test2(self):
df = DataFrame()
df.read_tbl('data/iqbrainsize.txt', delimiter='\t')
cor = df.correlation(list(df.keys()))