REF: Pre-empt ValueError in _aggregate_series_fast (pandas-dev#29500)

Author: jbrockmendel

pandas/_libs/reduction.pyx

@@ -323,6 +323,10 @@ cdef class SeriesGrouper(_BaseGrouper):
         #  safer obj._get_values(slice(None, 0))
         assert dummy is not None
 
+        if len(series) == 0:
+            # get_result would never assign `result`
+            raise ValueError("SeriesGrouper requires non-empty `series`")
+
         self.labels = labels
         self.f = f
 
@@ -408,8 +412,9 @@ cdef class SeriesGrouper(_BaseGrouper):
             islider.reset()
             vslider.reset()
 
-        if result is None:
-            raise ValueError("No result.")
+        # We check for empty series in the constructor, so should always
+        #  have result initialized by this point.
+        assert result is not None, "`result` has not been assigned."
 
         if result.dtype == np.object_:
             result = maybe_convert_objects(result)

pandas/core/groupby/ops.py

@@ -604,7 +604,11 @@ def agg_series(self, obj: Series, func):
         # Caller is responsible for checking ngroups != 0
         assert self.ngroups != 0
 
-        if is_extension_array_dtype(obj.dtype) and obj.dtype.kind != "M":
+        if len(obj) == 0:
+            # SeriesGrouper would raise if we were to call _aggregate_series_fast
+            return self._aggregate_series_pure_python(obj, func)
+
+        elif is_extension_array_dtype(obj.dtype) and obj.dtype.kind != "M":
             # _aggregate_series_fast would raise TypeError when
             #  calling libreduction.Slider
             # TODO: can we get a performant workaround for EAs backed by ndarray?
@@ -618,10 +622,7 @@ def agg_series(self, obj: Series, func):
         try:
             return self._aggregate_series_fast(obj, func)
         except ValueError as err:
-            if "No result." in str(err):
-                # raised in libreduction
-                pass
-            elif "Function does not reduce" in str(err):
+            if "Function does not reduce" in str(err):
                 # raised in libreduction
                 pass
             else:
@@ -632,6 +633,7 @@ def _aggregate_series_fast(self, obj, func):
         # At this point we have already checked that
         #  - obj.index is not a MultiIndex
         #  - obj is backed by an ndarray, not ExtensionArray
+        #  - len(obj) > 0
         #  - ngroups != 0
         func = self._is_builtin_func(func)
 

pandas/tests/groupby/test_bin_groupby.py

@@ -25,6 +25,16 @@ def test_series_grouper():
     tm.assert_almost_equal(counts, exp_counts)
 
 
+def test_series_grouper_requires_nonempty_raises():
+    # GH#29500
+    obj = Series(np.random.randn(10))
+    dummy = obj[:0]
+    labels = np.array([-1, -1, -1, 0, 0, 0, 1, 1, 1, 1], dtype=np.int64)
+
+    with pytest.raises(ValueError, match="SeriesGrouper requires non-empty `series`"):
+        libreduction.SeriesGrouper(dummy, np.mean, labels, 2, dummy)
+
+
 def test_series_bin_grouper():
     obj = Series(np.random.randn(10))
     dummy = obj[:0]