Avoid error from relplot faceting variable name collision

doc/releases/v0.12.0.txt

@@ -38,7 +38,9 @@ A paper describing seaborn was published in the `Journal of Open Source Software
 
 - |Fix| In :func:`lmplot`, fixed a bug where `sharey=False` did not always work as expected (:pr:`2576`).
 
-- |Fix| In :func:`scatterplot` and :func:`lineplot`, fixed a bug where legend entries for the `size` semantic were incorrect when `size_norm` extrapolated beyond the range of the data (:pr:`2580`).
+- |Fix| In the relational plots, fixed a bug where legend entries for the `size` semantic were incorrect when `size_norm` extrapolated beyond the range of the data (:pr:`2580`).
+
+- |Fix| In :func:`relplot`, fixed an error that would be raised when a column used to facet shared a name with one of the plot variables (:pr:`2581`).
 
 - |Fix| In :func:`histplot` and :func:`kdeplot`, fixed a bug where the `alpha` parameter was ignored when `fill=False` (:pr:`2460`).
 

seaborn/relational.py

@@ -883,7 +883,7 @@ def relplot(
     if "ax" in kwargs:
         msg = (
             "relplot is a figure-level function and does not accept "
-            "the ax= paramter. You may wish to try {}".format(kind + "plot")
+            "the `ax` parameter. You may wish to try {}".format(kind + "plot")
         )
         warnings.warn(msg, UserWarning)
         kwargs.pop("ax")
@@ -940,10 +940,6 @@ def relplot(
     if kind == "scatter":
         plot_kws.pop("dashes")
 
-    # Define the named variables for plotting on each facet
-    plot_variables = {key: key for key in p.variables}
-    plot_kws.update(plot_variables)
-
     # Add the grid semantics onto the plotter
     grid_semantics = "row", "col"
     p.semantics = plot_semantics + grid_semantics
@@ -956,16 +952,26 @@ def relplot(
         ),
     )
 
+    # Define the named variables for plotting on each facet
+    # Rename the variables with a leading underscore to avoid
+    # collisions with faceting variable names
+    plot_variables = {v: f"_{v}" for v in variables}
+    plot_kws.update(plot_variables)
+
     # Pass the row/col variables to FacetGrid with their original
     # names so that the axes titles render correctly
     grid_kws = {v: p.variables.get(v, None) for v in grid_semantics}
-    full_data = p.plot_data.rename(columns=grid_kws)
+
+    # Rename the columns of the plot_data structure appropriately
+    new_cols = plot_variables.copy()
+    new_cols.update(grid_kws)
+    full_data = p.plot_data.rename(columns=new_cols)
 
     # Set up the FacetGrid object
     facet_kws = {} if facet_kws is None else facet_kws.copy()
-    facet_kws.update(grid_kws)
     g = FacetGrid(
-        data=full_data,
+        data=full_data.dropna(axis=1, how="all"),
+        **grid_kws,
         col_wrap=col_wrap, row_order=row_order, col_order=col_order,
         height=height, aspect=aspect, dropna=False,
         **facet_kws
@@ -991,6 +997,13 @@ def relplot(
                          title=p.legend_title,
                          adjust_subtitles=True)
 
+    # Rename the columns of the FacetGrid's `data` attribute
+    # to match the original column names
+    orig_cols = {
+        f"_{k}": f"_{k}_" if v is None else v for k, v in variables.items()
+    }
+    g.data = g.data.rename(columns=orig_cols)
+
     return g
 
 

seaborn/tests/test_relational.py

@@ -617,6 +617,28 @@ def test_relplot_legend(self, long_df):
         for line, color in zip(lines, palette):
             assert line.get_color() == color
 
+    def test_relplot_data_columns(self, long_df):
+
+        long_df = long_df.assign(x_var=long_df["x"], y_var=long_df["y"])
+        g = relplot(
+            data=long_df,
+            x="x_var", y="y_var",
+            hue=long_df["a"].to_numpy(), col="c"
+        )
+        assert g.data.columns.to_list() == ["x_var", "y_var", "_hue_", "c"]
+
+    def test_facet_variable_collision(self, long_df):
+
+        # https://github.com/mwaskom/seaborn/issues/2488
+        col_data = long_df["c"]
+        long_df = long_df.assign(size=col_data)
+
+        g = relplot(
+            data=long_df,
+            x="x", y="y", col="size",
+        )
+        assert g.axes.shape == (1, len(col_data.unique()))
+
     def test_ax_kwarg_removal(self, long_df):
 
         f, ax = plt.subplots()