def test_groupby_1():
data_expected = {'animal': ['x-ray fish', 'yak', 'zebra'],
'count_name': [1,2,1]}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
data_real = {'animal': ['zebra', 'yak', 'yak', 'x-ray fish'],
'name': ['adam', 'bob', 'charlie', 'dina']}
df = pd.DataFrame(data=data_real)
r = Relation(df)
r = r.groupby(['animal']).count('name')
assert r.equals(r_expected)
def test_groupby3():
data_real = {
'course': ['DBMS', 'OSA', 'ML', 'OSA', 'OSA'],
'student': ['Abby', 'Abby', 'Abby', 'Bob', 'Carson']
}
df = pd.DataFrame(data=data_real)
r = Relation(df)
r = r.groupby('student').count('course')
data_expected = {
'student': ['Abby', 'Bob', 'Carson'],
'count_course': [3, 1, 1]
}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
assert r.equals(r_expected)
def test_groupby1():
data_real = {
'country': ['USA', 'Canada', 'France'],
'continent': ['North America', 'North America', 'Europe']
}
df = pd.DataFrame(data=data_real)
r = Relation(df)
r = r.groupby('continent').count('country')
data_expected = {
'continent': ['Europe', 'North America'],
'count_country': [1, 2]
}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
assert r.equals(r_expected)
def test_groupby2():
#This query does not necessarily represent a useful query, but is only intended to test the functionality
data_real = {
'major': ['CS', 'CS', 'Math', 'Math', 'DS'],
'gradyear': [2016, 2019, 2015, 2020, 2020]
}
df = pd.DataFrame(data=data_real)
r = Relation(df)
r = r.groupby('major').max('gradyear')
data_expected = {
'major': ['CS', 'DS', 'Math'],
'max_gradyear': [2019, 2020, 2020]
}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
assert r.equals(r_expected)
def test_union_compatibility():
'''
assert that union only accepts compatable dataframes by attempting an illegal union
'''
df = Relation("../country.csv")
monarchy_df = df.query("governmentform == 'Monarchy'")
groupby_gov_df = df.groupby(['governmentform']).count('name')
try:
monarchy_df.union(groupby_gov_df)
except ValueError as e:
assert str(e) == "Relations must be Union compatible"
except:
'''
all other errors result in test failure
'''
pytest.fail(
f"Illegal union returned an unknown exception: {sys.exc_info()[0]} : {sys.exc_info()[1]}"
)
def test_groupby_2():
data_expected = {'student': ['amanda', 'sam', 'sam', 'tony'],
'grade': ['A', 'A', 'C', 'B'],
'count_classes': [2,2,1,1]}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
print(r_expected)
data_real = {'student': ['amanda', 'amanda', 'sam', 'sam', 'sam', 'tony'],
'classes': ['math', 'science', 'math', 'science', 'chinese', 'math'],
'grade': ['A', 'A', 'C', 'A', 'A', 'B']}
df = pd.DataFrame(data=data_real)
r = Relation(df)
r = r.groupby(['student', 'grade']).count('classes')
print(r)
assert r.equals(r_expected)