def test_rename_preserves_data():
'''
assert that renaming a column preserves the data
'''
df = Relation("../country.csv")
renamed_df = df.rename("governmentform", "formofgov")
assert df["governmentform"].equals(renamed_df["formofgov"])
def test_cartesian_product2():
data_real1 = {'student': ['Abby', 'Billy', 'Carson']}
data_real2 = {
'grade': ['A', 'A', 'B'],
'course': ['Math', 'Science', 'Math']
}
df1 = pd.DataFrame(data=data_real1)
df2 = pd.DataFrame(data=data_real2)
r1 = Relation(df1)
r2 = Relation(df2)
r = r1.cartesian_product(r2)
data_expected = {
'student': [
'Abby', 'Abby', 'Abby', 'Billy', 'Billy', 'Billy', 'Carson',
'Carson', 'Carson'
],
'grade': ['A', 'A', 'B', 'A', 'A', 'B', 'A', 'A', 'B'],
'course': [
'Math', 'Science', 'Math', 'Math', 'Science', 'Math', 'Math',
'Science', 'Math'
]
}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
assert r.equals(r_expected)
def test_cartesian_product1():
data_real1 = {
'country': ['USA', 'Canada', 'France'],
'continent': ['North America', 'North America', 'Europe']
}
data_real2 = {'gnp': [1234, 5678], 'population': [100, 250]}
df1 = pd.DataFrame(data=data_real1)
df2 = pd.DataFrame(data=data_real2)
r1 = Relation(df1)
r2 = Relation(df2)
r = r1.cartesian_product(r2)
data_expected = {
'country': ['USA', 'USA', 'Canada', 'Canada', 'France', 'France'],
'continent': [
'North America', 'North America', 'North America', 'North America',
'Europe', 'Europe'
],
'gnp': [1234, 5678, 1234, 5678, 1234, 5678],
'population': [100, 250, 100, 250, 100, 250]
}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
assert r.equals(r_expected)
def test_product_1():
data_expected = {
'animal_x': ['zebra', 'zebra', 'x-ray fish', 'x-ray fish'],
'name': ['adam', 'adam', 'dina', 'dina'],
'color': ['red', 'red', 'purple', 'purple'],
'animal_y': ['zebra', 'x-ray fish', 'zebra', 'x-ray fish'],
'age': [7, 678, 7, 678]
}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
data_real_table1 = {
'animal': ['zebra', 'x-ray fish'],
'name': ['adam', 'dina'],
'color': ['red', 'purple']
}
df_table1 = pd.DataFrame(data=data_real_table1)
r_table1 = Relation(df_table1)
data_real_table2 = {'animal': ['zebra', 'x-ray fish'], 'age': [7, 678]}
df_table2 = pd.DataFrame(data=data_real_table2)
r_table2 = Relation(df_table2)
r = r_table1.cartesian_product(r_table2)
assert r.equals(r_expected)
def test_groupby_2():
data_expected = {
'student': ['amanda', 'sam', 'tony'],
'classes': ['science', 'chinese', 'math'],
'grade': ['A', 'A', 'A']
}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
print(r_expected)
data_real_table1 = {
'student': ['amanda', 'sam', 'tony'],
'classes': ['science', 'chinese', 'math']
}
df_table1 = pd.DataFrame(data=data_real_table1)
r_table1 = Relation(df_table1)
data_real_table2 = {'grade': ['A']}
df_table2 = pd.DataFrame(data=data_real_table2)
r_table2 = Relation(df_table2)
r = r_table1.cartesian_product(r_table2)
print(r)
assert r.equals(r_expected)
def test_rename_nonexistent_cols():
'''
assert that renaming a fake column returns the same dataframe as before and that a new empty column is not added
'''
df = Relation("../country.csv")
renamed_df = df.rename("a_fake_column", "something_nice")
assert "something_nice" not in renamed_df.columns
assert df.equals(renamed_df)
def test_union_remove_duplicates():
'''
assert union removes duplicate rows by comparing lengths of a dataframe against the length of one unioned with itself
'''
df = Relation("../country.csv")
republic_df = df.query("governmentform == 'Republic'")
union_on_itself_df = republic_df.union(republic_df)
assert len(republic_df) == len(union_on_itself_df)
def test_foo():
df1 = Relation(
pd.read_csv("tests/test_outer_join/test_outer_join_1.csv", sep="|"))
df2 = Relation(
pd.read_csv("tests/test_outer_join/test_outer_join_2.csv", sep="|"))
df = df1.outerjoin(df2)
df_expected = pd.read_csv("tests/test_outer_join/test_outer_join_3.csv")
assert df.equals(df_expected)
def test_union_on_continent():
'''
assert the length a union on two dataframes queried by continent is the same as the sum of the previous two lengths
'''
df = Relation("../country.csv")
africa_df = df.query("continent == 'Africa'")
europe_df = df.query("continent == 'Europe'")
union_df = africa_df.union(europe_df)
assert (len(africa_df) + len(europe_df)) == len(union_df)
def test_antijoin(r):
r = country.query('continent == "North America"').project(
['name', 'region']).antijoin(
country.query('region == "Caribbean"').project(
['name', 'region'])).reset_index().drop(columns=["index"])
data_expected = Relation("tests/antijoin.csv",
sep="|").reset_index().drop(columns=["index"])
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
assert r.equals(r_expected)
def test_rename_removes_col():
'''
assert that renaming a column removes the first column header and adds the second
'''
df = Relation("../country.csv")
renamed_df = df.rename("indepyear", "yearofindep")
assert "indepyear" in df.columns
assert "yearofindep" not in df.columns
assert "indepyear" not in renamed_df.columns
assert "yearofindep" in renamed_df.columns
def test_rename_1():
data_expected = {'color': ['green', 'blue']}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
data_real = {'hello': ['green', 'blue']}
df = pd.DataFrame(data=data_real)
r = Relation(df)
r = r.rename("hello", "color")
assert r.equals(r_expected)
def test_rename_2():
data_expected = {'color': ['green', 'blue'], 'food': ['banana', 'cookie']}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
data_real = {'color': ['green', 'blue'], 'hello': ['banana', 'cookie']}
df = pd.DataFrame(data=data_real)
r = Relation(df)
r = r.rename("hello", "food")
assert r.equals(r_expected)
def test_foo():
data_real = {"name": ["Carol", "Bob"], "age": [86, 4]}
data_expected = {"name": ["Carol", "Bob"]}
df = pd.DataFrame(data=data_real)
df_expected = pd.DataFrame(data=data_expected)
r = Relation(df)
r_expected = Relation(df_expected)
r_expected_2 = Relation('tests/test_project_expected_1.csv', sep='|')
r2 = r.project(["name"])
assert r2.equals(r_expected)
assert r2.equals(r_expected_2)
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 r5():
data_real = {
"Class": ["History", "Art", "Drama", "History", "Art"],
"Start Time": ["10:00", "9:00", "12:00", "13:00", "18:00"]
}
df05 = pd.DataFrame(data=data_real)
return Relation(df05)
def r3():
data_real = {
"name": ["Larry", "Bob", "Lucy"],
"Birthyear": [1999, 2015, 1994]
}
df03 = pd.DataFrame(data=data_real)
return Relation(df03)
def r6():
data_real = {
"Year": [1996, 1996, 1997, 1998, 1998, 1998],
"Revenue": [2000, 3400, 1200, 500, 650, 200]
}
df06 = pd.DataFrame(data=data_real)
return Relation(df06)
def test_select_2(r):
r = r.select("height > 180")
data_expected = {"name":["Changmin","Yunho"], "age":[31,33], "height": [186,184]}
df_expected_2 = pd.DataFrame(data = data_expected)
r_expected_2 = Relation("tests/test_select_expected_1.csv", sep="|")
assert r.equals(r_expected_2)
def test_foo(r):
r = r.extend("birth", 2019 - r.age).project(["name", "age", "birth"]).head(10)
data_expected = {"name": ["Carol", "Bob"], "age": [86, 4], "birth": [1933, 2015]}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
assert r.equals(r_expected)
def test_semi_join_right(r_1, r_2):
data_expected = {'color': ['green', 'blue']}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
r = r_2.semi_join(r_1)
assert r.equals(r_expected)
def test_groupby03(r6):
data_expected = {
"Year": [1996, 1997, 1998],
"sum_Revenue": [5400, 1200, 1350]
}
df_expected = pd.DataFrame(data=data_expected)
r_groupbyExpected = Relation(df_expected)
assert r6.groupby("Year").sum("Revenue").equals(r_groupbyExpected)
def makeContact():
data4 = {
"id": [0, 1],
"email": ["*****@*****.**", "*****@*****.**"],
"phone": ["555-107-1234", "868-402-7539"]
}
contact = pd.DataFrame(data=data4)
return Relation(contact)
def test_outerjoin01(r, r2):
data_expected = {
"name": ["Carol", "Bob", "Larry", "Lucy"],
"age": [86, 4, 20, 25]
}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
assert r.outerJoin(r2).equals(r_expected)
def test_semi_join_left(r_1, r_2):
data_expected = {'color': ['red', 'purple']}
df_expected = pd.DataFrame(data=data_expected)
r_expected = Relation(df_expected)
r = r_1.semi_join(r_2)
assert r.equals(r_expected)
def r():
data_real = {
"firstName": ["Alex", "John", "Clark", "Haley"],
"degree": ["Math", "Drama", "Computer Science", "Math"],
"age": [20, 21, 19, 19]
}
df = pd.DataFrame(data=data_real)
return Relation(df)
def r1():
data_real = {
"name": ["Jaejoong", "Junsu", "Yoochun"],
"age": [33, 32, 33],
"height": [179, 178, 180]
}
df = pd.DataFrame(data=data_real)
return Relation(df)
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_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_product(r1, r2):
r = r1.product(r2)
r_expected = Relation("tests/test_product_expected.csv", sep="|")
print(r_expected)
print(r)
assert r.equals(r_expected)