Merge pull request #30 from DataSketches/strings-to-char-array
convert strings to char[] before passing to sketch to avoid costly UTF-8
diff --git a/src/main/java/com/yahoo/sketches/hive/hll/SketchState.java b/src/main/java/com/yahoo/sketches/hive/hll/SketchState.java
index 9c190b4..27a5e33 100644
--- a/src/main/java/com/yahoo/sketches/hive/hll/SketchState.java
+++ b/src/main/java/com/yahoo/sketches/hive/hll/SketchState.java
@@ -46,7 +46,8 @@
sketch_.update(PrimitiveObjectInspectorUtils.getLong(data, keyObjectInspector));
return;
case STRING:
- sketch_.update(PrimitiveObjectInspectorUtils.getString(data, keyObjectInspector));
+ // conversion to char[] avoids costly UTF-8 encoding
+ sketch_.update(PrimitiveObjectInspectorUtils.getString(data, keyObjectInspector).toCharArray());
return;
default:
throw new IllegalArgumentException(
