Constraint: Fix __str__ for non-string column keys
Pandas DataFrame columns may be named with any hashable object.
diff --git a/tests/test_constraint.py b/tests/test_constraint.py
index 438fa97..f75b61a 100644
--- a/tests/test_constraint.py
+++ b/tests/test_constraint.py
@@ -18,7 +18,7 @@
import unittest
from trappy.plotter import AttrConf
-from trappy.plotter.Constraint import ConstraintManager
+from trappy.plotter.Constraint import Constraint, ConstraintManager
class TestConstraintManager(unittest.TestCase):
"""Test trappy.plotter.ConstraintManager"""
@@ -184,3 +184,13 @@
constraint = iter(c_mgr).next()
series = constraint.result[AttrConf.PIVOT_VAL]
self.assertEquals(len(series), 3)
+
+class TestConstraint(unittest.TestCase):
+ def test_str_constraint(self):
+ """str(constraint) doesn't fail when the column is not a string"""
+ dfr = pd.DataFrame({12: [1, 2, 3], 13: [3, 4, 5]})
+
+ constraint = Constraint(dfr, AttrConf.PIVOT, 12, template=None,
+ trace_index=0, filters={}, window=None)
+
+ self.assertEqual(str(constraint), "DataFrame 0:12")