fix histogram plot failure on float('inf') (skrub-data#1151)

* fix histogram plot failure on float('inf')

* changelog

* _

* _

CHANGES.rst

@@ -96,7 +96,7 @@ Bug fixes
 
 * The :class:`TableReport` would raise an exception when a column contained
   infinite values. This has been fixed in :pr:`1150` by :user:`Jérôme Dockès
-  <jeromedockes>`.
+  <jeromedockes>` and :pr:`1151` by Jérôme Dockès.
 
 Release 0.3.1
 =============

skrub/_reporting/_plotting.py

@@ -9,6 +9,7 @@
 import warnings
 
 import matplotlib
+import numpy as np
 from matplotlib import pyplot as plt
 
 from skrub import _dataframe as sbd
@@ -118,7 +119,14 @@ def _adjust_fig_size(fig, ax, target_w, target_h):
 def histogram(col, color=COLOR_0):
     """Histogram for a numeric column."""
     col = sbd.drop_nulls(col)
+    if sbd.is_float(col):
+        # avoid any issues with pandas nullable dtypes
+        # (to_numpy can yield a numpy array with object dtype in old pandas
+        # version if there are inf or nan)
+        col = sbd.to_float32(col)
     values = sbd.to_numpy(col)
+    if np.issubdtype(values.dtype, np.floating):
+        values = values[np.isfinite(values)]
     fig, ax = plt.subplots()
     _despine(ax)
     ax.hist(values, color=color)

skrub/_reporting/tests/test_table_report.py

@@ -1,5 +1,6 @@
 import json
 import re
+import warnings
 
 from skrub import TableReport, ToDatetime
 from skrub import _dataframe as sbd
@@ -99,3 +100,16 @@ def test_duplicate_columns(pd_module):
     df = pd_module.make_dataframe({"a": [1, 2], "b": [3, 4]})
     df.columns = ["a", "a"]
     TableReport(df).html()
+
+
+def test_infinite_values(df_module):
+    # Non-regression for https://github.com/skrub-data/skrub/issues/1134
+    # (histogram plot failing with infinite values)
+    with warnings.catch_warnings():
+        # convert_dtypes() emits spurious warning while deciding whether to cast to int
+        warnings.filterwarnings("ignore", message="invalid value encountered in cast")
+        df = df_module.make_dataframe(
+            dict(a=[float("inf"), 1.0, 2.0], b=[0.0, 1.0, 2.0])
+        )
+
+    TableReport(df).html()