New aggregation and handle empty aggregation selection

Author: PrimozGodec

Orange/data/aggregate.py

@@ -99,7 +99,12 @@ def _compute_aggregation(
 
     def _aggregations_to_table(self, aggregations: List[pd.Series]) -> Table:
         """Concatenate aggregation series and convert back to Table"""
-        df = pd.concat(aggregations, axis=1)
+        if aggregations:
+            df = pd.concat(aggregations, axis=1)
+        else:
+            # when no aggregation is computed return a table with gropby columns
+            df = self.group_by.first()
+            df = df.drop(columns=df.columns)
         gb_attributes = df.index.names
         df = df.reset_index()  # move group by var that are in index to columns
         table = table_from_frame(df)

Orange/widgets/data/owgroupby.py

@@ -75,12 +75,10 @@ def concatenate(x):
         {ContinuousVariable, TimeVariable},
     ),
     "First value": Aggregation(
-        "first",
-        {ContinuousVariable, DiscreteVariable, StringVariable, TimeVariable}
+        "first", {ContinuousVariable, DiscreteVariable, StringVariable, TimeVariable}
     ),
     "Last value": Aggregation(
-        "last",
-        {ContinuousVariable, DiscreteVariable, StringVariable, TimeVariable}
+        "last", {ContinuousVariable, DiscreteVariable, StringVariable, TimeVariable}
     ),
     "Random value": Aggregation(
         lambda x: x.sample(1, random_state=0),
@@ -92,16 +90,19 @@ def concatenate(x):
     "Count": Aggregation(
         "size", {ContinuousVariable, DiscreteVariable, StringVariable, TimeVariable}
     ),
+    "Proportion defined": Aggregation(
+        lambda x: x.count() / x.size,
+        {ContinuousVariable, DiscreteVariable, StringVariable, TimeVariable},
+    ),
 }
 # list of ordered aggregation names is required on several locations so we
 # prepare it in advance
 AGGREGATIONS_ORD = list(AGGREGATIONS)
 
+# use first aggregation suitable for each type as default
 DEFAULT_AGGREGATIONS = {
-    ContinuousVariable: {next(iter(AGGREGATIONS.keys()))},
-    TimeVariable: {next(iter(AGGREGATIONS.keys()))},
-    DiscreteVariable: set(),
-    StringVariable: set(),
+    var: {next(name for name, agg in AGGREGATIONS.items() if var in agg.types)}
+    for var in (ContinuousVariable, TimeVariable, DiscreteVariable, StringVariable)
 }
 
 
@@ -114,7 +115,7 @@ class Result:
 def _run(
     data: Table,
     group_by_attrs: List[Variable],
-    aggregations: Dict[Variable, List[str]],
+    aggregations: Dict[Variable, Set[str]],
     result: Result,
     state: TaskState,
 ) -> Result:
@@ -383,7 +384,9 @@ def __rows_selected(self) -> None:
                 # check if aggregation active for all selected attributes,
                 # partially check if active for some else uncheck
                 cb.setCheckState(
-                    Qt.Checked if activated == {True} else (Qt.Unchecked if activated == {False} else Qt.PartiallyChecked)
+                    Qt.Checked
+                    if activated == {True}
+                    else (Qt.Unchecked if activated == {False} else Qt.PartiallyChecked)
                 )
 
     def __gb_changed(self) -> None:

Orange/widgets/data/tests/test_owgroupby.py

@@ -159,15 +159,36 @@ def test_attr_table_row_selection(self):
         )
         self.select_table_rows(table, [3])  # discrete variable
         self.assert_enabled_cbs(
-            {"Count defined", "Count", "Concatenate", "First value", "Last value", "Random value"}
+            {
+                "Count defined",
+                "Count",
+                "Concatenate",
+                "First value",
+                "Last value",
+                "Random value",
+            }
         )
         self.select_table_rows(table, [4])  # string variable
         self.assert_enabled_cbs(
-            {"Count defined", "Count", "Concatenate", "First value", "Last value", "Random value"}
+            {
+                "Count defined",
+                "Count",
+                "Concatenate",
+                "First value",
+                "Last value",
+                "Random value",
+            }
         )
         self.select_table_rows(table, [3, 4])  # string variable
         self.assert_enabled_cbs(
-            {"Count defined", "Count", "Concatenate", "First value", "Last value", "Random value"}
+            {
+                "Count defined",
+                "Count",
+                "Concatenate",
+                "First value",
+                "Last value",
+                "Random value",
+            }
         )
         self.select_table_rows(table, [2, 3, 4])  # string variable
         self.assert_enabled_cbs(
@@ -399,7 +420,13 @@ def test_aggregations_change(self):
             Qt.PartiallyChecked, self.widget.agg_checkboxes["Mode"].checkState()
         )
         self.assert_aggregations_equal(
-            ["Mean, Mode, Count defined and 1 more", "Mean, Mode, Count defined", "Mean", "Count defined, Count", ""]
+            [
+                "Mean, Mode, Count defined and 1 more",
+                "Mean, Mode, Count defined",
+                "Mean",
+                "Count defined, Count",
+                "",
+            ]
         )
         self.assertDictEqual(
             {