[SQL] Code Cleanup: Left Semi Hash Join

Some improvement for PR apache#837, add another case to white list and use `filter` to build result iterator. Author: Daoyuan <daoyuan.wang@intel.com> Closes apache#1049 from adrian-wang/clean-LeftSemiJoinHash and squashes the following commits: b314d5a [Daoyuan] change hashSet name 27579a9 [Daoyuan] add semijoin to white list and use filter to create new iterator in LeftSemiJoinBNL Signed-off-by: Michael Armbrust <michael@databricks.com>
pdeyhim · Jun 25, 2014 · e991148 · e991148
1 parent ce0886f
commit e991148
Show file tree

Hide file tree

Showing 52 changed files with 374 additions and 33 deletions.
diff --git a/sql/core/src/main/scala/org/apache/spark/sql/execution/joins.scala b/sql/core/src/main/scala/org/apache/spark/sql/execution/joins.scala
@@ -169,51 +169,25 @@ case class LeftSemiJoinHash(
   def execute() = {
 
     buildPlan.execute().zipPartitions(streamedPlan.execute()) { (buildIter, streamIter) =>
-      val hashTable = new java.util.HashSet[Row]()
+      val hashSet = new java.util.HashSet[Row]()
       var currentRow: Row = null
 
       // Create a Hash set of buildKeys
       while (buildIter.hasNext) {
         currentRow = buildIter.next()
         val rowKey = buildSideKeyGenerator(currentRow)
         if(!rowKey.anyNull) {
-          val keyExists = hashTable.contains(rowKey)
+          val keyExists = hashSet.contains(rowKey)
           if (!keyExists) {
-            hashTable.add(rowKey)
+            hashSet.add(rowKey)
           }
         }
       }
 
-      new Iterator[Row] {
-        private[this] var currentStreamedRow: Row = _
-        private[this] var currentHashMatched: Boolean = false
-
-        private[this] val joinKeys = streamSideKeyGenerator()
-
-        override final def hasNext: Boolean =
-          streamIter.hasNext && fetchNext()
-
-        override final def next() = {
-          currentStreamedRow
-        }
-
-        /**
-         * Searches the streamed iterator for the next row that has at least one match in hashtable.
-         *
-         * @return true if the search is successful, and false the streamed iterator runs out of
-         *         tuples.
-         */
-        private final def fetchNext(): Boolean = {
-          currentHashMatched = false
-          while (!currentHashMatched && streamIter.hasNext) {
-            currentStreamedRow = streamIter.next()
-            if (!joinKeys(currentStreamedRow).anyNull) {
-              currentHashMatched = hashTable.contains(joinKeys.currentValue)
-            }
-          }
-          currentHashMatched
-        }
-      }
+      val joinKeys = streamSideKeyGenerator()
+      streamIter.filter(current => {
+        !joinKeys(current).anyNull && hashSet.contains(joinKeys.currentValue)
+      })
     }
   }
 }

diff --git a/sql/hive/src/test/resources/golden/semijoin-0-1631b71327abf75b96116036b977b26c b/sql/hive/src/test/resources/golden/semijoin-0-1631b71327abf75b96116036b977b26c
diff --git a/sql/hive/src/test/resources/golden/semijoin-1-e7c99e1f46d502edbb0925d75aab5f0c b/sql/hive/src/test/resources/golden/semijoin-1-e7c99e1f46d502edbb0925d75aab5f0c
@@ -0,0 +1,11 @@
+0	val_0
+0	val_0
+0	val_0
+2	val_2
+4	val_4
+5	val_5
+5	val_5
+5	val_5
+8	val_8
+9	val_9
+10	val_10
diff --git a/sql/hive/src/test/resources/golden/semijoin-10-ffd4fb3a903a6725ccb97d5451a3fec6 b/sql/hive/src/test/resources/golden/semijoin-10-ffd4fb3a903a6725ccb97d5451a3fec6
diff --git a/sql/hive/src/test/resources/golden/semijoin-11-246a40dcafe077f02397e30d227c330 b/sql/hive/src/test/resources/golden/semijoin-11-246a40dcafe077f02397e30d227c330
@@ -0,0 +1,8 @@
+0	val_0
+0	val_0
+0	val_0
+4	val_2
+8	val_4
+10	val_5
+10	val_5
+10	val_5
diff --git a/sql/hive/src/test/resources/golden/semijoin-12-6d93a9d332ba490835b17f261a5467df b/sql/hive/src/test/resources/golden/semijoin-12-6d93a9d332ba490835b17f261a5467df
diff --git a/sql/hive/src/test/resources/golden/semijoin-13-18282d38b6efc0017089ab89b661764f b/sql/hive/src/test/resources/golden/semijoin-13-18282d38b6efc0017089ab89b661764f
diff --git a/sql/hive/src/test/resources/golden/semijoin-14-19cfcefb10e1972bec0ffd421cd79de7 b/sql/hive/src/test/resources/golden/semijoin-14-19cfcefb10e1972bec0ffd421cd79de7
diff --git a/sql/hive/src/test/resources/golden/semijoin-15-1de6eb3f357bd1c4d02ab4d19d43589 b/sql/hive/src/test/resources/golden/semijoin-15-1de6eb3f357bd1c4d02ab4d19d43589
@@ -0,0 +1,11 @@
+val_0
+val_0
+val_0
+val_10
+val_2
+val_4
+val_5
+val_5
+val_5
+val_8
+val_9
diff --git a/sql/hive/src/test/resources/golden/semijoin-16-d3a72a90515ac4a8d8e9ac923bcda3d b/sql/hive/src/test/resources/golden/semijoin-16-d3a72a90515ac4a8d8e9ac923bcda3d
diff --git a/sql/hive/src/test/resources/golden/semijoin-17-f0f8720cfd11fd71af17b310dc2d1019 b/sql/hive/src/test/resources/golden/semijoin-17-f0f8720cfd11fd71af17b310dc2d1019
@@ -0,0 +1,3 @@
+0	val_0
+0	val_0
+0	val_0
diff --git a/sql/hive/src/test/resources/golden/semijoin-18-f7b2ce472443982e32d954cbb5c96765 b/sql/hive/src/test/resources/golden/semijoin-18-f7b2ce472443982e32d954cbb5c96765
diff --git a/sql/hive/src/test/resources/golden/semijoin-19-b1f1c7f701abe81c01e72fb98f0bd13f b/sql/hive/src/test/resources/golden/semijoin-19-b1f1c7f701abe81c01e72fb98f0bd13f
@@ -0,0 +1,3 @@
+val_10
+val_8
+val_9
diff --git a/sql/hive/src/test/resources/golden/semijoin-2-deb9c3286ae8e851b1fdb270085b16bc b/sql/hive/src/test/resources/golden/semijoin-2-deb9c3286ae8e851b1fdb270085b16bc
diff --git a/sql/hive/src/test/resources/golden/semijoin-20-b7a8ebaeb42b2eaba7d97cadc3fd96c1 b/sql/hive/src/test/resources/golden/semijoin-20-b7a8ebaeb42b2eaba7d97cadc3fd96c1
diff --git a/sql/hive/src/test/resources/golden/semijoin-21-480418a0646cf7260b494b9eb4821bb6 b/sql/hive/src/test/resources/golden/semijoin-21-480418a0646cf7260b494b9eb4821bb6
diff --git a/sql/hive/src/test/resources/golden/semijoin-22-b6aebd98f7636cda7b24e0bf84d7ba41 b/sql/hive/src/test/resources/golden/semijoin-22-b6aebd98f7636cda7b24e0bf84d7ba41
diff --git a/sql/hive/src/test/resources/golden/semijoin-23-ed730ccdf552c07e7a82ba6b7fd3fbda b/sql/hive/src/test/resources/golden/semijoin-23-ed730ccdf552c07e7a82ba6b7fd3fbda
@@ -0,0 +1,5 @@
+4	val_2
+8	val_4
+10	val_5
+10	val_5
+10	val_5
diff --git a/sql/hive/src/test/resources/golden/semijoin-24-d16b37134de78980b2bf96029e8265c3 b/sql/hive/src/test/resources/golden/semijoin-24-d16b37134de78980b2bf96029e8265c3
diff --git a/sql/hive/src/test/resources/golden/semijoin-25-be2bd011cc80b480b271a08dbf381e9b b/sql/hive/src/test/resources/golden/semijoin-25-be2bd011cc80b480b271a08dbf381e9b
@@ -0,0 +1,19 @@
+0
+0
+0
+0
+0
+0
+2
+4
+4
+5
+5
+5
+8
+8
+9
+10
+10
+10
+10
diff --git a/sql/hive/src/test/resources/golden/semijoin-26-f1d3bab29f1ebafa148dbe3816e1da25 b/sql/hive/src/test/resources/golden/semijoin-26-f1d3bab29f1ebafa148dbe3816e1da25
diff --git a/sql/hive/src/test/resources/golden/semijoin-27-391c256254d171973f02e7f33672ce1d b/sql/hive/src/test/resources/golden/semijoin-27-391c256254d171973f02e7f33672ce1d
@@ -0,0 +1,4 @@
+0	val_0
+0	val_0
+0	val_0
+8	val_8
diff --git a/sql/hive/src/test/resources/golden/semijoin-28-b56400f6d9372f353cf7292a2182e963 b/sql/hive/src/test/resources/golden/semijoin-28-b56400f6d9372f353cf7292a2182e963
diff --git a/sql/hive/src/test/resources/golden/semijoin-29-9efeef3d3c38e22a74d379978178c4f5 b/sql/hive/src/test/resources/golden/semijoin-29-9efeef3d3c38e22a74d379978178c4f5
@@ -0,0 +1,14 @@
+0	val_0	0	val_0
+0	val_0	0	val_0
+0	val_0	0	val_0
+0	val_0	0	val_0
+0	val_0	0	val_0
+0	val_0	0	val_0
+0	val_0	0	val_0
+0	val_0	0	val_0
+0	val_0	0	val_0
+4	val_4	4	val_2
+8	val_8	8	val_4
+10	val_10	10	val_5
+10	val_10	10	val_5
+10	val_10	10	val_5
diff --git a/sql/hive/src/test/resources/golden/semijoin-3-b4d4317dd3a10e18502f20f5c5250389 b/sql/hive/src/test/resources/golden/semijoin-3-b4d4317dd3a10e18502f20f5c5250389
@@ -0,0 +1,11 @@
+0	val_0
+0	val_0
+0	val_0
+4	val_2
+8	val_4
+10	val_5
+10	val_5
+10	val_5
+16	val_8
+18	val_9
+20	val_10
diff --git a/sql/hive/src/test/resources/golden/semijoin-30-dd901d00fce5898b03a57cbc3028a70a b/sql/hive/src/test/resources/golden/semijoin-30-dd901d00fce5898b03a57cbc3028a70a
diff --git a/sql/hive/src/test/resources/golden/semijoin-31-e5dc4d8185e63e984aa4e3a2e08bd67 b/sql/hive/src/test/resources/golden/semijoin-31-e5dc4d8185e63e984aa4e3a2e08bd67
@@ -0,0 +1,14 @@
+0	val_0
+0	val_0
+0	val_0
+0	val_0
+0	val_0
+0	val_0
+2	val_2
+4	val_4
+5	val_5
+5	val_5
+5	val_5
+8	val_8
+9	val_9
+10	val_10
diff --git a/sql/hive/src/test/resources/golden/semijoin-32-23017c7663f2710265a7e2a4a1606d39 b/sql/hive/src/test/resources/golden/semijoin-32-23017c7663f2710265a7e2a4a1606d39
diff --git a/sql/hive/src/test/resources/golden/semijoin-33-ed499f94c6e6ac847ef5187b3b43bbc5 b/sql/hive/src/test/resources/golden/semijoin-33-ed499f94c6e6ac847ef5187b3b43bbc5
@@ -0,0 +1,14 @@
+0
+0
+0
+0
+0
+0
+4
+4
+8
+8
+10
+10
+10
+10
diff --git a/sql/hive/src/test/resources/golden/semijoin-34-5e1b832090ab73c141c1167d5b25a490 b/sql/hive/src/test/resources/golden/semijoin-34-5e1b832090ab73c141c1167d5b25a490
diff --git a/sql/hive/src/test/resources/golden/semijoin-35-8d5731f26232f6e26dd8012461b08d99 b/sql/hive/src/test/resources/golden/semijoin-35-8d5731f26232f6e26dd8012461b08d99
@@ -0,0 +1,26 @@
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+4
+4
+8
+8
+10
+10
+10
+10
diff --git a/sql/hive/src/test/resources/golden/semijoin-36-b1159823dca8025926407f8aa921238d b/sql/hive/src/test/resources/golden/semijoin-36-b1159823dca8025926407f8aa921238d
diff --git a/sql/hive/src/test/resources/golden/semijoin-37-a15b9074f999ce836be5329591b968d0 b/sql/hive/src/test/resources/golden/semijoin-37-a15b9074f999ce836be5329591b968d0
@@ -0,0 +1,29 @@
+NULL
+NULL
+NULL
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+4
+4
+8
+8
+10
+10
+10
+10
diff --git a/sql/hive/src/test/resources/golden/semijoin-38-f37547c73a48ce3ba089531b176e6ba b/sql/hive/src/test/resources/golden/semijoin-38-f37547c73a48ce3ba089531b176e6ba
diff --git a/sql/hive/src/test/resources/golden/semijoin-39-c22a6f11368affcb80a9c80e653a47a8 b/sql/hive/src/test/resources/golden/semijoin-39-c22a6f11368affcb80a9c80e653a47a8
@@ -0,0 +1,29 @@
+NULL
+NULL
+NULL
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+4
+4
+8
+8
+10
+10
+10
+10
diff --git a/sql/hive/src/test/resources/golden/semijoin-4-dfdad5a2742f93e8ea888191460809c0 b/sql/hive/src/test/resources/golden/semijoin-4-dfdad5a2742f93e8ea888191460809c0
diff --git a/sql/hive/src/test/resources/golden/semijoin-40-32071a51e2ba6e86b1c5e40de55aae63 b/sql/hive/src/test/resources/golden/semijoin-40-32071a51e2ba6e86b1c5e40de55aae63
diff --git a/sql/hive/src/test/resources/golden/semijoin-41-cf74f73a33b1af8902b7970a9350b092 b/sql/hive/src/test/resources/golden/semijoin-41-cf74f73a33b1af8902b7970a9350b092
@@ -0,0 +1,29 @@
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+4
+4
+8
+8
+10
+10
+10
+10
+16
+18
+20
diff --git a/sql/hive/src/test/resources/golden/semijoin-42-6b4257a74fca627785c967c99547f4c0 b/sql/hive/src/test/resources/golden/semijoin-42-6b4257a74fca627785c967c99547f4c0
diff --git a/sql/hive/src/test/resources/golden/semijoin-43-e8a166ac2e94bf8d1da0fe91b0db2c81 b/sql/hive/src/test/resources/golden/semijoin-43-e8a166ac2e94bf8d1da0fe91b0db2c81
@@ -0,0 +1,31 @@
+NULL
+NULL
+NULL
+NULL
+NULL
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+4
+4
+8
+8
+10
+10
+10
+10
diff --git a/sql/hive/src/test/resources/golden/semijoin-44-945aaa3a24359ef73acab1e99500d5ea b/sql/hive/src/test/resources/golden/semijoin-44-945aaa3a24359ef73acab1e99500d5ea
diff --git a/sql/hive/src/test/resources/golden/semijoin-45-3fd94ffd4f1eb6cf83dcc064599bf12b b/sql/hive/src/test/resources/golden/semijoin-45-3fd94ffd4f1eb6cf83dcc064599bf12b
@@ -0,0 +1,42 @@
+NULL
+NULL
+NULL
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+0
+2
+4
+4
+5
+5
+5
+8
+8
+9
+10
+10
+10
+10
+10
+10
+10
+10
+10
+10
+10
+10
diff --git a/sql/hive/src/test/resources/golden/semijoin-46-620e01f81f6e5254b4bbe8fab4043ec0 b/sql/hive/src/test/resources/golden/semijoin-46-620e01f81f6e5254b4bbe8fab4043ec0
-Original file line number
+Diff line change
@@ -0,0 +1,11 @@
+	val_0
+	val_0
+	val_0
+	val_2
+	val_4
+	val_5
+	val_5
+	val_5
+	val_8
+	val_9
+	val_10