MNT: add a warning if a categroical string array is all convertible to numbers or dates

jklymak · jklymak · commit 5797522f0e77 · 2019-01-18T14:05:31.000-08:00
diff --git a/lib/matplotlib/category.py b/lib/matplotlib/category.py
@@ -12,10 +12,12 @@
 """
 
 from collections import OrderedDict
+import dateutil.parser
 import itertools
 
 import numpy as np
 
+import matplotlib.cbook as cbook
 import matplotlib.units as units
 import matplotlib.ticker as ticker
 
@@ -174,6 +176,20 @@ def __init__(self, data=None):
         if data is not None:
             self.update(data)
 
+    @staticmethod
+    def _str_is_convertable(val):
+        """
+        Helper method to see if string can be cast to float or parsed as date.
+        """
+        try:
+            float(val)
+        except ValueError:
+            try:
+                dateutil.parser.parse(val)
+            except ValueError:
+                return False
+        return True
+
     def update(self, data):
         """Maps new values to integer identifiers.
 
@@ -189,11 +205,20 @@ def update(self, data):
         """
         data = np.atleast_1d(np.array(data, dtype=object))
 
+        convertable = True
         for val in OrderedDict.fromkeys(data):
+            # OrderedDict just iterates over unique values in data.
             if not isinstance(val, (str, bytes)):
                 raise TypeError("{val!r} is not a string".format(val=val))
+            # check if we can convert string to number or date...
+            convertable = (convertable and self._str_is_convertable(val))
             if val not in self._mapping:
                 self._mapping[val] = next(self._counter)
+        if convertable:
+            cbook._warn_external('using category units to plot a list of '
+                          'strings that is a;; floats or parsable as dates. '
+                          'If you do not mean these to be categories, cast '
+                          'to the approriate data type before plotting.')
 
 
 # Register the converter with Matplotlib's unit framework
diff --git a/lib/matplotlib/tests/test_category.py b/lib/matplotlib/tests/test_category.py
@@ -203,7 +203,8 @@ class TestPlotNumlike(object):
     @pytest.mark.parametrize("ndata", numlike_data, ids=numlike_ids)
     def test_plot_numlike(self, ax, plotter, ndata):
         counts = np.array([4, 6, 5])
-        plotter(ax, ndata, counts)
+        with pytest.warns(UserWarning, match='using category units to plot'):
+            plotter(ax, ndata, counts)
         axis_test(ax.xaxis, ndata)
 
 
@@ -261,12 +262,14 @@ def test_update_plot(self, ax, plotter):
 
     PLOT_BROKEN_IDS = ["scatter", "plot", "bar"]
 
+    @pytest.mark.filterwarnings('ignore::UserWarning')
     @pytest.mark.parametrize("plotter", PLOT_BROKEN_LIST, ids=PLOT_BROKEN_IDS)
     @pytest.mark.parametrize("xdata", fvalues, ids=fids)
     def test_mixed_type_exception(self, ax, plotter, xdata):
         with pytest.raises(TypeError):
             plotter(ax, xdata, [1, 2])
 
+    @pytest.mark.filterwarnings('ignore::UserWarning')
     @pytest.mark.parametrize("plotter", PLOT_BROKEN_LIST, ids=PLOT_BROKEN_IDS)
     @pytest.mark.parametrize("xdata", fvalues, ids=fids)
     def test_mixed_type_update_exception(self, ax, plotter, xdata):
