Implement TokenizeOperator in addition to the changes made by Zachary Heilbron.
Now, the compiler builds the bulk-load plan.
For other details, please refer to the corresponding Hyracks changes.

Change-Id: I646539d88ea2bdc6da4fbf2b6e9460a6189125ff
Reviewed-on: http://fulliautomatix.ics.uci.edu:8443/79
Tested-by: Jenkins <jenkins@fulliautomatix.ics.uci.edu>
Reviewed-by: Young-Seok Kim <kisskys@gmail.com>
diff --git a/asterix-app/src/test/resources/optimizerts/results/rtree-secondary-index.plan b/asterix-app/src/test/resources/optimizerts/results/rtree-secondary-index.plan
index c288790..d86e4f5 100644
--- a/asterix-app/src/test/resources/optimizerts/results/rtree-secondary-index.plan
+++ b/asterix-app/src/test/resources/optimizerts/results/rtree-secondary-index.plan
@@ -1,3 +1,12 @@
+-- SINK  |PARTITIONED|
+  -- ONE_TO_ONE_EXCHANGE  |PARTITIONED|
+    -- STREAM_PROJECT  |PARTITIONED|
+      -- ONE_TO_ONE_EXCHANGE  |PARTITIONED|
+        -- BULKLOAD  |PARTITIONED|
+          -- HASH_PARTITION_MERGE_EXCHANGE MERGE:[$$1(ASC)] HASH:[$$1]  |PARTITIONED|
+            -- DATASOURCE_SCAN  |PARTITIONED|
+              -- ONE_TO_ONE_EXCHANGE  |PARTITIONED|
+                -- EMPTY_TUPLE_SOURCE  |PARTITIONED|
 -- DISTRIBUTE_RESULT  |PARTITIONED|
   -- ONE_TO_ONE_EXCHANGE  |PARTITIONED|
     -- STREAM_PROJECT  |PARTITIONED|