You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@drill.apache.org by ja...@apache.org on 2015/11/05 06:56:42 UTC
[4/9] drill git commit: DRILL-3793: New MergeJoin and add
RecordIterator interface
http://git-wip-us.apache.org/repos/asf/drill/blob/2bc16a90/exec/java-exec/src/test/java/org/apache/drill/exec/physical/impl/join/TestMergeJoinAdvanced.java
----------------------------------------------------------------------
diff --git a/exec/java-exec/src/test/java/org/apache/drill/exec/physical/impl/join/TestMergeJoinAdvanced.java b/exec/java-exec/src/test/java/org/apache/drill/exec/physical/impl/join/TestMergeJoinAdvanced.java
index 54539fd..ec7411d 100644
--- a/exec/java-exec/src/test/java/org/apache/drill/exec/physical/impl/join/TestMergeJoinAdvanced.java
+++ b/exec/java-exec/src/test/java/org/apache/drill/exec/physical/impl/join/TestMergeJoinAdvanced.java
@@ -23,8 +23,15 @@ import org.apache.drill.exec.ExecConstants;
import org.apache.drill.exec.planner.physical.PlannerSettings;
import org.junit.AfterClass;
import org.junit.BeforeClass;
+import org.junit.Ignore;
import org.junit.Test;
+import java.io.BufferedWriter;
+import java.io.File;
+import java.io.FileWriter;
+import java.io.IOException;
+import java.util.Random;
+
public class TestMergeJoinAdvanced extends BaseTestQuery {
// Have to disable hash join to test merge join in this class
@@ -96,4 +103,99 @@ public class TestMergeJoinAdvanced extends BaseTestQuery {
setSessionOption(ExecConstants.MAX_WIDTH_PER_NODE_KEY, String.valueOf(ExecConstants.MAX_WIDTH_PER_NODE.getDefault().num_val));
}
}
+
+ private static void generateData(final BufferedWriter leftWriter, final BufferedWriter rightWriter,
+ final long left, final long right) throws IOException {
+ for (int i=0; i < left; ++i) {
+ leftWriter.write(String.format("{ \"k\" : %d , \"v\": %d }", 10000, i));
+ }
+ leftWriter.write(String.format("{ \"k\" : %d , \"v\": %d }", 10001, 10001));
+ leftWriter.write(String.format("{ \"k\" : %d , \"v\": %d }", 10002, 10002));
+
+ for (int i=0; i < right; ++i) {
+ rightWriter.write(String.format("{ \"k1\" : %d , \"v1\": %d }", 10000, i));
+ }
+ rightWriter.write(String.format("{ \"k1\" : %d , \"v1\": %d }", 10004, 10004));
+ rightWriter.write(String.format("{ \"k1\" : %d , \"v1\": %d }", 10005, 10005));
+ rightWriter.write(String.format("{ \"k1\" : %d , \"v1\": %d }", 10006, 10006));
+
+ leftWriter.close();
+ rightWriter.close();
+ }
+
+ private static void testMultipleBatchJoin(final long right, final long left,
+ final String joinType, final long expected) throws Exception {
+ final String leftSide = BaseTestQuery.getTempDir("merge-join-left.json");
+ final String rightSide = BaseTestQuery.getTempDir("merge-join-right.json");
+ final BufferedWriter leftWriter = new BufferedWriter(new FileWriter(new File(leftSide)));
+ final BufferedWriter rightWriter = new BufferedWriter(new FileWriter(new File(rightSide)));
+ generateData(leftWriter, rightWriter, left, right);
+ final String query1 = String.format("select count(*) c1 from dfs_test.`%s` L %s join dfs_test.`%s` R on L.k=R.k1",
+ leftSide, joinType, rightSide);
+ testBuilder()
+ .sqlQuery(query1)
+ .optionSettingQueriesForTestQuery("alter session set `planner.enable_hashjoin` = false")
+ .unOrdered()
+ .baselineColumns("c1")
+ .baselineValues(expected)
+ .go();
+ }
+
+ @Test
+ public void testMergeInnerJoinLargeRight() throws Exception {
+ testMultipleBatchJoin(1000l, 5000l, "inner", 5000l * 1000l);
+ }
+
+ @Test
+ public void testMergeLeftJoinLargeRight() throws Exception {
+ testMultipleBatchJoin(1000l, 5000l, "left", 5000l * 1000l +2l);
+ }
+
+ @Test
+ public void testMergeRightJoinLargeRight() throws Exception {
+ testMultipleBatchJoin(1000l, 5000l, "right", 5000l*1000l +3l);
+ }
+
+ @Test
+ public void testMergeInnerJoinLargeLeft() throws Exception {
+ testMultipleBatchJoin(5000l, 1000l, "inner", 5000l*1000l);
+ }
+
+ @Test
+ public void testMergeLeftJoinLargeLeft() throws Exception {
+ testMultipleBatchJoin(5000l, 1000l, "left", 5000l*1000l + 2l);
+ }
+
+ @Test
+ public void testMergeRightJoinLargeLeft() throws Exception {
+ testMultipleBatchJoin(5000l, 1000l, "right", 5000l*1000l + 3l);
+ }
+
+ // Following tests can take some time.
+ @Test
+ @Ignore
+ public void testMergeInnerJoinRandomized() throws Exception {
+ final Random r = new Random();
+ final long right = r.nextInt(10001) + 1l;
+ final long left = r.nextInt(10001) + 1l;
+ testMultipleBatchJoin(left, right, "inner", left*right);
+ }
+
+ @Test
+ @Ignore
+ public void testMergeLeftJoinRandomized() throws Exception {
+ final Random r = new Random();
+ final long right = r.nextInt(10001) + 1l;
+ final long left = r.nextInt(10001) + 1l;
+ testMultipleBatchJoin(left, right, "left", left*right + 2l);
+ }
+
+ @Test
+ @Ignore
+ public void testMergeRightJoinRandomized() throws Exception {
+ final Random r = new Random();
+ final long right = r.nextInt(10001) + 1l;
+ final long left = r.nextInt(10001) + 1l;
+ testMultipleBatchJoin(left, right, "right", left * right + 3l);
+ }
}