fix(layer): cast polars decimal columns to float

plotnine/layer.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import decimal
 import typing
 from copy import copy, deepcopy
 from typing import Iterable, List, cast, overload
@@ -31,6 +32,23 @@
     )
 
 
+def _convert_decimal_columns(data: pd.DataFrame) -> pd.DataFrame:
+    """
+    Cast columns of decimal.Decimal values to float
+
+    polars' `to_pandas()` maps a Decimal column onto a pandas object column
+    holding `decimal.Decimal` values. plotnine treats object columns as
+    discrete, so such a column cannot be used with a continuous scale. Casting
+    it to float lets it behave like any other numeric column.
+    """
+    for col in data.columns:
+        if data[col].dtype == object:
+            non_null = data[col].dropna()
+            if len(non_null) and isinstance(non_null.iloc[0], decimal.Decimal):
+                data[col] = data[col].astype("float64")
+    return data
+
+
 class layer:
     """
     Layer
@@ -222,7 +240,9 @@ def _make_layer_data(self, plot_data: DataLike | None):
         if plot_data is None:
             data = pd.DataFrame()
         elif hasattr(plot_data, "to_pandas"):
-            data = cast("DataFrameConvertible", plot_data).to_pandas()
+            data = _convert_decimal_columns(
+                cast("DataFrameConvertible", plot_data).to_pandas()
+            )
         else:
             data = cast("pd.DataFrame", plot_data)
 
@@ -249,9 +269,9 @@ def _make_layer_data(self, plot_data: DataLike | None):
         else:
             # Recognise polars dataframes
             if hasattr(self._data, "to_pandas"):
-                self.data = cast(
-                    "DataFrameConvertible", self._data
-                ).to_pandas()
+                self.data = _convert_decimal_columns(
+                    cast("DataFrameConvertible", self._data).to_pandas()
+                )
             elif isinstance(self._data, pd.DataFrame):
                 self.data = self._data.copy()
             else:

tests/test_ggplot_internals.py

@@ -12,6 +12,7 @@
     coord_trans,
     facet_null,
     geom_bar,
+    geom_col,
     geom_histogram,
     geom_line,
     geom_point,
@@ -21,6 +22,7 @@
     labs,
     lims,
     scale_x_continuous,
+    scale_y_continuous,
     stage,
     stat_identity,
     theme,
@@ -365,6 +367,29 @@ def to_pandas(self):
     assert p2 == "to_pandas"
 
 
+def test_to_pandas_decimal_columns():
+    # polars' to_pandas() returns a Decimal column as an object column of
+    # decimal.Decimal values, which would otherwise be treated as discrete
+    # and rejected by a continuous scale. See GH #1033.
+    from decimal import Decimal
+
+    class DecimalData:
+        def to_pandas(self):
+            return pd.DataFrame(
+                {
+                    "x": ["a", "b", "c"],
+                    "y": [Decimal(1), Decimal(2), Decimal(3)],
+                }
+            )
+
+    p = (
+        ggplot(DecimalData(), aes("x", "y"))
+        + geom_col()
+        + scale_y_continuous(limits=(0, 5))
+    )
+    p._build()
+
+
 def test_callable_as_data():
     def _fn(data):
         return data.rename(columns={"xx": "x", "yy": "y"})