Expand LuxGroupby Tests and add bug fixes

Makefile

@@ -3,4 +3,5 @@ init:
 test:
 	black --check .
 	python -m pytest tests/
+
 .PHONY: init test

lux/core/groupby.py

@@ -60,6 +60,13 @@ def filter(self, *args, **kwargs):
         ret_val.pre_aggregated = False  # Returned LuxDataFrame isn't pre_aggregated
         return ret_val
 
+    def apply(self, *args, **kwargs):
+        ret_val = super(LuxDataFrameGroupBy, self).apply(*args, **kwargs)
+        for attr in self._metadata:
+            ret_val.__dict__[attr] = getattr(self, attr, None)
+        ret_val.pre_aggregated = False  # Returned LuxDataFrame isn't pre_aggregated
+        return ret_val
+
     def size(self, *args, **kwargs):
         ret_val = super(LuxDataFrameGroupBy, self).size(*args, **kwargs)
         for attr in self._metadata:

lux/core/series.py

@@ -49,6 +49,7 @@ class LuxSeries(pd.Series):
         "_pandas_only",
         "pre_aggregated",
         "_type_override",
+        "name",
     ]
 
     _default_metadata = {
@@ -111,10 +112,12 @@ def __repr__(self):
         from lux.core.frame import LuxDataFrame
 
         series_repr = super(LuxSeries, self).__repr__()
+
+        ldf = LuxDataFrame(self)
+
         # Default column name 0 causes errors
         if self.name is None:
-            self.name = " "
-        ldf = LuxDataFrame(self)
+            ldf = ldf.rename(columns={0: " "})
         self._ldf = ldf
 
         try:

tests/conftest.py

@@ -29,4 +29,5 @@ def global_var():
         "_pandas_only",
         "pre_aggregated",
         "_type_override",
+        "name",
     ]

tests/test_columns.py

@@ -131,3 +131,17 @@ def test_int_columns(global_var):
     assert list(df.recommendation.keys()) == ["Enhance", "Filter", "Generalize"]
     df.intent = [0]
     assert list(df.recommendation.keys()) == ["Enhance", "Filter"]
+
+
+def test_name_column(global_var):
+    df = pd.read_csv("lux/data/car.csv")
+    new_df = df.rename(columns={"Name": "name"})
+    assert list(new_df.recommendation.keys()) == [
+        "Correlation",
+        "Distribution",
+        "Occurrence",
+        "Temporal",
+    ]
+    assert len(new_df.recommendation["Correlation"])
+    assert new_df["name"][0] != None
+    assert (new_df["name"].unique() != None)[0]

tests/test_groupby.py

@@ -23,6 +23,7 @@ def test_agg(global_var):
     new_df = df[["Horsepower", "Brand"]].groupby("Brand").agg(sum)
     new_df._repr_html_()
     assert new_df.history[0].name == "groupby"
+    assert new_df.pre_aggregated
 
 
 def test_shortcut_agg(global_var):
@@ -31,6 +32,7 @@ def test_shortcut_agg(global_var):
     new_df = df[["MilesPerGal", "Brand"]].groupby("Brand").sum()
     new_df._repr_html_()
     assert new_df.history[0].name == "groupby"
+    assert new_df.pre_aggregated
 
 
 def test_agg_mean(global_var):
@@ -39,6 +41,7 @@ def test_agg_mean(global_var):
     new_df = df.groupby("Origin").mean()
     new_df._repr_html_()
     assert new_df.history[0].name == "groupby"
+    assert new_df.pre_aggregated
 
 
 def test_agg_size(global_var):
@@ -47,6 +50,7 @@ def test_agg_size(global_var):
     new_df = df.groupby("Brand").size().to_frame()
     new_df._repr_html_()
     assert new_df.history[0].name == "groupby"
+    assert new_df.pre_aggregated
 
 
 def test_filter(global_var):
@@ -55,3 +59,13 @@ def test_filter(global_var):
     new_df = df.groupby("Origin").filter(lambda x: x["Weight"].mean() > 3000)
     new_df._repr_html_()
     assert new_df.history[0].name == "groupby"
+    assert not new_df.pre_aggregated
+
+
+def test_get_group(global_var):
+    df = pytest.car_df
+    df._repr_html_()
+    new_df = df.groupby("Origin").get_group("Japan")
+    new_df._repr_html_()
+    assert new_df.history[0].name == "groupby"
+    assert not new_df.pre_aggregated

tests/test_series.py

@@ -46,6 +46,7 @@ def test_df_to_series():
         "_pandas_only",
         "pre_aggregated",
         "_type_override",
+        "name",
     ], "Metadata is lost when going from Dataframe to Series."
     assert df.cardinality is not None, "Metadata is lost when going from Dataframe to Series."
     assert series.name == "Weight", "Pandas Series original `name` property not retained."