Merge pull request #29 from mbaak/master

Bug fixes to categorize and sparselybin for use with sparksql

Author: jpivarski

histogrammar/primitives/categorize.py

@@ -148,9 +148,9 @@ def __iadd__(self, other):
             self.entries += other.entries
             for k in self.keySet.union(other.keySet):
                 if k in self.bins and k in other.bins:
-                    bins[k] += other.bins[k]
+                    self.bins[k] += other.bins[k]
                 elif k not in self.bins and k in other.bins:
-                    bins[k] = self.bins[k].copy()
+                    self.bins[k] = other.bins[k].copy()
             return self
         else:
             raise ContainerException("cannot add {0} and {1}".format(self.name, other.name))

histogrammar/primitives/sparselybin.py

@@ -398,7 +398,7 @@ def _numpy(self, data, weights, shape):
         self.entries += float(newentries)
 
     def _sparksql(self, jvm, converter):
-        return converter.SparselyBin(self.binWidth, quantity.asSparkSQL(), self.value._sparksql(jvm, converter), self.nanflow._sparksql(jvm, converter), self.origin)
+        return converter.SparselyBin(self.binWidth, self.quantity.asSparkSQL(), self.value._sparksql(jvm, converter), self.nanflow._sparksql(jvm, converter), self.origin)
 
     @property
     def children(self):