BUG: Fix scatter plot colors in groupby context to match line plot behavior (pandas-dev#59846)

Author: myenugula

Diff for: doc/source/whatsnew/v3.0.0.rst

@@ -762,6 +762,7 @@ Plotting
 - Bug in :meth:`DataFrame.plot.bar` with ``stacked=True`` where labels on stacked bars with zero-height segments were incorrectly positioned at the base instead of the label position of the previous segment (:issue:`59429`)
 - Bug in :meth:`DataFrame.plot.line` raising ``ValueError`` when set both color and a ``dict`` style (:issue:`59461`)
 - Bug in :meth:`DataFrame.plot` that causes a shift to the right when the frequency multiplier is greater than one. (:issue:`57587`)
+- Bug in :meth:`DataFrameGroupBy.plot` with scatter colors (:issue:`59846`)
 - Bug in :meth:`Series.plot` with ``kind="pie"`` with :class:`ArrowDtype` (:issue:`59192`)
 
 Groupby/resample/rolling

Diff for: pandas/core/groupby/groupby.py

@@ -431,11 +431,35 @@ def __init__(self, groupby: GroupBy) -> None:
         self._groupby = groupby
 
     def __call__(self, *args, **kwargs):
-        def f(self):
-            return self.plot(*args, **kwargs)
+        # Special case for scatter plots to enable automatic colors, similar to line plots
+        if kwargs.get("kind") == "scatter":
+            import matplotlib.pyplot as plt
 
-        f.__name__ = "plot"
-        return self._groupby._python_apply_general(f, self._groupby._selected_obj)
+            # Get colors from matplotlib's color cycle (similar to what LinePlot uses)
+            colors = plt.rcParams["axes.prop_cycle"].by_key()["color"]
+
+            # Determine the axis to plot on
+            if "ax" in kwargs:
+                ax = kwargs["ax"]
+            else:
+                _, ax = plt.subplots()
+
+            # Plot each group with a different color
+            results = {}
+            for i, (name, group) in enumerate(self._groupby):
+                group_kwargs = kwargs.copy()
+                group_kwargs["ax"] = ax
+                group_kwargs["color"] = colors[i % len(colors)]
+                results[name] = group.plot(*args, **group_kwargs)
+
+            return results
+        else:
+            # Original implementation for non-scatter plots
+            def f(self):
+                return self.plot(*args, **kwargs)
+
+            f.__name__ = "plot"
+            return self._groupby._python_apply_general(f, self._groupby._selected_obj)
 
     def __getattr__(self, name: str):
         def attr(*args, **kwargs):
@@ -546,8 +570,7 @@ def groups(self) -> dict[Hashable, Index]:
         2023-02-15    4
         dtype: int64
         >>> ser.resample("MS").groups
-        {Timestamp('2023-01-01 00:00:00'): np.int64(2),
-         Timestamp('2023-02-01 00:00:00'): np.int64(4)}
+        {Timestamp('2023-01-01 00:00:00'): 2, Timestamp('2023-02-01 00:00:00'): 4}
         """
         if isinstance(self.keys, list) and len(self.keys) == 1:
             warnings.warn(
@@ -614,7 +637,7 @@ def indices(self) -> dict[Hashable, npt.NDArray[np.intp]]:
         toucan  1  5  6
         eagle   7  8  9
         >>> df.groupby(by=["a"]).indices
-        {np.int64(1): array([0, 1]), np.int64(7): array([2])}
+        {1: array([0, 1]), 7: array([2])}
 
         For Resampler:
 

Diff for: pandas/tests/plotting/test_groupby.py

@@ -152,3 +152,52 @@ def test_groupby_hist_series_with_legend_raises(self):
 
         with pytest.raises(ValueError, match="Cannot use both legend and label"):
             g.hist(legend=True, label="d")
+
+    def test_groupby_scatter_colors(self):
+        # GH 59846 - Test that scatter plots use different colors for different groups
+        # similar to how line plots do
+        from matplotlib.collections import PathCollection
+        import matplotlib.pyplot as plt
+
+        # Create test data with distinct groups
+        df = DataFrame(
+            {
+                "x": [1, 2, 3, 4, 5, 6, 7, 8, 9],
+                "y": [1, 2, 3, 4, 5, 6, 7, 8, 9],
+                "group": ["A", "A", "A", "B", "B", "B", "C", "C", "C"],
+            }
+        )
+
+        # Set up a figure with both line and scatter plots
+        fig, (ax1, ax2) = plt.subplots(1, 2)
+
+        # Plot line chart (known to use different colors for different groups)
+        df.groupby("group").plot(x="x", y="y", ax=ax1, kind="line")
+
+        # Plot scatter chart (should also use different colors for different groups)
+        df.groupby("group").plot(x="x", y="y", ax=ax2, kind="scatter")
+
+        # Get the colors used in the line plot and scatter plot
+        line_colors = [line.get_color() for line in ax1.get_lines()]
+
+        # Get scatter colors
+        scatter_colors = []
+        for collection in ax2.collections:
+            if isinstance(collection, PathCollection):  # This is a scatter plot
+                # Get the face colors (might be array of RGBA values)
+                face_colors = collection.get_facecolor()
+                # If multiple points with same color, we get the first one
+                if face_colors.ndim > 1:
+                    scatter_colors.append(tuple(face_colors[0]))
+                else:
+                    scatter_colors.append(tuple(face_colors))
+
+        # Assert that we have the right number of colors (one per group)
+        assert len(line_colors) == 3
+        assert len(scatter_colors) == 3
+
+        # Assert that the colors are all different (fixed behavior)
+        assert len(set(line_colors)) == 3
+        assert len(set(scatter_colors)) == 3
+
+        plt.close(fig)