def test_set_time_index(sample_df):
# Test setting time index with set_time_index()
dt = DataTable(sample_df)
new_dt = dt.set_time_index('signup_date')
assert new_dt is not dt
assert dt.time_index is None
assert new_dt.time_index == 'signup_date'
assert 'time_index' in new_dt.columns['signup_date'].semantic_tags
non_index_cols = [
col for col in new_dt.columns.values() if col.name != 'signup_date'
]
assert all(
['time_index' not in col.semantic_tags for col in non_index_cols])
# Test changing time index with set_time_index()
sample_df['transaction_date'] = pd.to_datetime('2015-09-02')
dt = DataTable(sample_df)
new_dt = dt.set_time_index('signup_date')
assert new_dt.time_index == 'signup_date'
new_dt2 = new_dt.set_time_index('transaction_date')
assert 'time_index' in new_dt2.columns['transaction_date'].semantic_tags
non_index_cols = [
col for col in new_dt2.columns.values()
if col.name != 'transaction_date'
]
assert all(
['time_index' not in col.semantic_tags for col in non_index_cols])
# Test setting index using setter
dt = DataTable(sample_df)
assert dt.time_index is None
dt.time_index = 'signup_date'
assert dt.time_index == 'signup_date'
assert 'time_index' in dt.columns['signup_date'].semantic_tags
non_index_cols = [
col for col in dt.columns.values() if col.name != 'signup_date'
]
assert all(
['time_index' not in col.semantic_tags for col in non_index_cols])
# Test changing time index with setter
sample_df['transaction_date'] = pd.to_datetime('2015-09-02')
dt = DataTable(sample_df)
dt.time_index = 'signup_date'
assert dt.time_index == 'signup_date'
dt.time_index = 'transaction_date'
assert 'time_index' in dt.columns['transaction_date'].semantic_tags
non_index_cols = [
col for col in dt.columns.values() if col.name != 'transaction_date'
]
assert all(
['time_index' not in col.semantic_tags for col in non_index_cols])
def test_set_index_twice(sample_df):
dt = DataTable(sample_df, index='id', time_index='signup_date')
original_df = dt.df.copy()
dt_index_twice = dt.set_index('id')
assert 'index' in dt_index_twice['id'].semantic_tags
assert dt_index_twice.index == 'id'
assert dt_index_twice == dt
pd.testing.assert_frame_equal(to_pandas(original_df),
to_pandas(dt_index_twice.df))
dt_time_index_twice = dt.set_time_index('signup_date')
assert 'time_index' in dt_time_index_twice['signup_date'].semantic_tags
assert dt_time_index_twice.time_index == 'signup_date'
assert dt_time_index_twice == dt
pd.testing.assert_frame_equal(to_pandas(original_df),
to_pandas(dt_time_index_twice.df))
dt.index = 'id'
assert 'index' in dt['id'].semantic_tags
assert dt.index == 'id'
pd.testing.assert_frame_equal(to_pandas(original_df), to_pandas(dt.df))
dt.time_index = 'signup_date'
assert 'time_index' in dt['signup_date'].semantic_tags
assert dt.time_index == 'signup_date'
pd.testing.assert_frame_equal(to_pandas(original_df), to_pandas(dt.df))
def test_numeric_index_strings(time_index_df):
error_msg = 'Time index column must contain datetime or numeric values'
with pytest.raises(TypeError, match=error_msg):
DataTable(time_index_df, time_index='strs')
error_msg = 'Time index column must contain datetime or numeric values'
with pytest.raises(TypeError, match=error_msg):
DataTable(time_index_df,
time_index='ints',
logical_types={'ints': 'Categorical'})
error_msg = 'Time index column must contain datetime or numeric values'
with pytest.raises(TypeError, match=error_msg):
DataTable(time_index_df,
time_index='letters',
logical_types={'strs': 'Integer'})
dt = DataTable(time_index_df,
time_index='strs',
logical_types={'strs': 'Double'})
date_col = dt['strs']
assert dt.time_index == 'strs'
assert date_col.logical_type == Double
assert date_col.semantic_tags == {'time_index', 'numeric'}
dt = DataTable(time_index_df, logical_types={'strs': 'Double'})
dt = dt.set_time_index('strs')
date_col = dt['strs']
assert dt.time_index == 'strs'
assert date_col.logical_type == Double
assert date_col.semantic_tags == {'time_index', 'numeric'}
def test_numeric_time_index_dtypes(numeric_time_index_df):
dt = DataTable(numeric_time_index_df, time_index='ints')
date_col = dt['ints']
assert dt.time_index == 'ints'
assert date_col.logical_type == Integer
assert date_col.semantic_tags == {'time_index', 'numeric'}
dt = dt.set_time_index('floats')
date_col = dt['floats']
assert dt.time_index == 'floats'
assert date_col.logical_type == Double
assert date_col.semantic_tags == {'time_index', 'numeric'}
dt = dt.set_time_index('with_null')
date_col = dt['with_null']
assert dt.time_index == 'with_null'
if ks and isinstance(numeric_time_index_df, ks.DataFrame):
ltype = Double
else:
ltype = Integer
assert date_col.logical_type == ltype
assert date_col.semantic_tags == {'time_index', 'numeric'}
