matplotlib
diff --git a/‎lib/matplotlib/category.py‎
Lines changed: 30 additions & 0 deletions b/‎lib/matplotlib/category.py‎
Lines changed: 30 additions & 0 deletions
@@ -12,14 +12,20 @@
 """
 
 from collections import OrderedDict
+import dateutil.parser
 import itertools
+import logging
 
 import numpy as np
 
+import matplotlib.cbook as cbook
 import matplotlib.units as units
 import matplotlib.ticker as ticker
 
 
+_log = logging.getLogger(__name__)
+
+
 class StrCategoryConverter(units.ConversionInterface):
     @staticmethod
     def convert(value, unit, axis):
@@ -168,6 +174,23 @@ def __init__(self, data=None):
         if data is not None:
             self.update(data)
 
+    @staticmethod
+    def _strs_are_convertible(vals):
+        """
+        Helper method to see if list of strings can all be cast to float or
+        parsed as date.
+        """
+
+        for val in vals:
+            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.
 
@@ -184,10 +207,17 @@ def update(self, data):
         data = np.atleast_1d(np.array(data, dtype=object))
 
         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))
             if val not in self._mapping:
                 self._mapping[val] = next(self._counter)
+        # check if we can convert all strings to number or date...
+        if self._strs_are_convertible(data):
+            _log.info('using category units to plot a list of '
+                      'strings that is all floats or parsable as dates. '
+                      'If you do not mean these to be categories, cast '
+                      'to the approriate data type before plotting.')
 
 
 # Connects the convertor to matplotlib