You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@hive.apache.org by li...@apache.org on 2016/12/23 03:28:04 UTC
[4/4] hive git commit: HIVE-15357: Fix and re-enable the spark-only
tests (Rui reviewed by Chao)
HIVE-15357: Fix and re-enable the spark-only tests (Rui reviewed by Chao)
Project: http://git-wip-us.apache.org/repos/asf/hive/repo
Commit: http://git-wip-us.apache.org/repos/asf/hive/commit/858ce8c2
Tree: http://git-wip-us.apache.org/repos/asf/hive/tree/858ce8c2
Diff: http://git-wip-us.apache.org/repos/asf/hive/diff/858ce8c2
Branch: refs/heads/master
Commit: 858ce8c22b65057e8cb3d9d87cb9b9dfb0f8666c
Parents: ee35ccb
Author: Rui Li <li...@apache.org>
Authored: Fri Dec 23 11:27:47 2016 +0800
Committer: Rui Li <li...@apache.org>
Committed: Fri Dec 23 11:27:47 2016 +0800
----------------------------------------------------------------------
.../hadoop/hive/cli/control/CliConfigs.java | 1 +
.../hive/ql/optimizer/physical/Vectorizer.java | 1 +
.../spark_dynamic_partition_pruning.q | 100 +-
...spark_vectorized_dynamic_partition_pruning.q | 1 +
.../spark/spark_combine_equivalent_work.q.out | 2 -
.../spark/spark_dynamic_partition_pruning.q.out | 1650 ++++++++----------
.../spark_dynamic_partition_pruning_2.q.out | 506 +++---
...k_vectorized_dynamic_partition_pruning.q.out | 1507 +++++++---------
8 files changed, 1726 insertions(+), 2042 deletions(-)
----------------------------------------------------------------------
http://git-wip-us.apache.org/repos/asf/hive/blob/858ce8c2/itests/util/src/main/java/org/apache/hadoop/hive/cli/control/CliConfigs.java
----------------------------------------------------------------------
diff --git a/itests/util/src/main/java/org/apache/hadoop/hive/cli/control/CliConfigs.java b/itests/util/src/main/java/org/apache/hadoop/hive/cli/control/CliConfigs.java
index c5e027b..af8ec67 100644
--- a/itests/util/src/main/java/org/apache/hadoop/hive/cli/control/CliConfigs.java
+++ b/itests/util/src/main/java/org/apache/hadoop/hive/cli/control/CliConfigs.java
@@ -479,6 +479,7 @@ public class CliConfigs {
setQueryDir("ql/src/test/queries/clientpositive");
includesFrom(testConfigProps, "miniSparkOnYarn.query.files");
+ includesFrom(testConfigProps, "spark.only.query.files");
setResultsDir("ql/src/test/results/clientpositive/spark");
setLogDir("itests/qtest-spark/target/qfile-results/clientpositive/spark");
http://git-wip-us.apache.org/repos/asf/hive/blob/858ce8c2/ql/src/java/org/apache/hadoop/hive/ql/optimizer/physical/Vectorizer.java
----------------------------------------------------------------------
diff --git a/ql/src/java/org/apache/hadoop/hive/ql/optimizer/physical/Vectorizer.java b/ql/src/java/org/apache/hadoop/hive/ql/optimizer/physical/Vectorizer.java
index 468ccaf..2a99274 100644
--- a/ql/src/java/org/apache/hadoop/hive/ql/optimizer/physical/Vectorizer.java
+++ b/ql/src/java/org/apache/hadoop/hive/ql/optimizer/physical/Vectorizer.java
@@ -2501,6 +2501,7 @@ public class Vectorizer implements PhysicalPlanResolver {
case EXTRACT:
case EVENT:
case HASHTABLESINK:
+ case SPARKPRUNINGSINK:
vectorOp = OperatorFactory.getVectorOperator(
op.getCompilationOpContext(), op.getConf(), vContext);
break;
http://git-wip-us.apache.org/repos/asf/hive/blob/858ce8c2/ql/src/test/queries/clientpositive/spark_dynamic_partition_pruning.q
----------------------------------------------------------------------
diff --git a/ql/src/test/queries/clientpositive/spark_dynamic_partition_pruning.q b/ql/src/test/queries/clientpositive/spark_dynamic_partition_pruning.q
index 8b83ef6..3b6810b 100644
--- a/ql/src/test/queries/clientpositive/spark_dynamic_partition_pruning.q
+++ b/ql/src/test/queries/clientpositive/spark_dynamic_partition_pruning.q
@@ -1,58 +1,58 @@
-set hive.support.sql11.reserved.keywords=false;
set hive.optimize.ppd=true;
set hive.ppd.remove.duplicatefilters=true;
set hive.spark.dynamic.partition.pruning=true;
set hive.optimize.metadataonly=false;
set hive.optimize.index.filter=true;
+set hive.strict.checks.cartesian.product=false;
-- SORT_QUERY_RESULTS
select distinct ds from srcpart;
select distinct hr from srcpart;
-EXPLAIN create table srcpart_date as select ds as ds, ds as date from srcpart group by ds;
-create table srcpart_date as select ds as ds, ds as date from srcpart group by ds;
+EXPLAIN create table srcpart_date as select ds as ds, ds as `date` from srcpart group by ds;
+create table srcpart_date as select ds as ds, ds as `date` from srcpart group by ds;
create table srcpart_hour as select hr as hr, hr as hour from srcpart group by hr;
-create table srcpart_date_hour as select ds as ds, ds as date, hr as hr, hr as hour from srcpart group by ds, hr;
+create table srcpart_date_hour as select ds as ds, ds as `date`, hr as hr, hr as hour from srcpart group by ds, hr;
create table srcpart_double_hour as select (hr*2) as hr, hr as hour from srcpart group by hr;
-- single column, single key
-EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = '2008-04-08';
-select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = '2008-04-08';
+EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = '2008-04-08';
+select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = '2008-04-08';
set hive.spark.dynamic.partition.pruning=false;
-EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = '2008-04-08';
-select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = '2008-04-08';
+EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = '2008-04-08';
+select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = '2008-04-08';
set hive.spark.dynamic.partition.pruning=true;
select count(*) from srcpart where ds = '2008-04-08';
-- multiple sources, single key
EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) join srcpart_hour on (srcpart.hr = srcpart_hour.hr)
-where srcpart_date.date = '2008-04-08' and srcpart_hour.hour = 11;
+where srcpart_date.`date` = '2008-04-08' and srcpart_hour.hour = 11;
select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) join srcpart_hour on (srcpart.hr = srcpart_hour.hr)
-where srcpart_date.date = '2008-04-08' and srcpart_hour.hour = 11;
+where srcpart_date.`date` = '2008-04-08' and srcpart_hour.hour = 11;
set hive.spark.dynamic.partition.pruning=false;
EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) join srcpart_hour on (srcpart.hr = srcpart_hour.hr)
-where srcpart_date.date = '2008-04-08' and srcpart_hour.hour = 11;
+where srcpart_date.`date` = '2008-04-08' and srcpart_hour.hour = 11;
select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) join srcpart_hour on (srcpart.hr = srcpart_hour.hr)
-where srcpart_date.date = '2008-04-08' and srcpart_hour.hour = 11;
+where srcpart_date.`date` = '2008-04-08' and srcpart_hour.hour = 11;
set hive.spark.dynamic.partition.pruning=true;
select count(*) from srcpart where hr = 11 and ds = '2008-04-08';
-- multiple columns single source
-EXPLAIN select count(*) from srcpart join srcpart_date_hour on (srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr) where srcpart_date_hour.date = '2008-04-08' and srcpart_date_hour.hour = 11;
-select count(*) from srcpart join srcpart_date_hour on (srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr) where srcpart_date_hour.date = '2008-04-08' and srcpart_date_hour.hour = 11;
+EXPLAIN select count(*) from srcpart join srcpart_date_hour on (srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr) where srcpart_date_hour.`date` = '2008-04-08' and srcpart_date_hour.hour = 11;
+select count(*) from srcpart join srcpart_date_hour on (srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr) where srcpart_date_hour.`date` = '2008-04-08' and srcpart_date_hour.hour = 11;
set hive.spark.dynamic.partition.pruning=false;
-EXPLAIN select count(*) from srcpart join srcpart_date_hour on (srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr) where srcpart_date_hour.date = '2008-04-08' and srcpart_date_hour.hour = 11;
-select count(*) from srcpart join srcpart_date_hour on (srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr) where srcpart_date_hour.date = '2008-04-08' and srcpart_date_hour.hour = 11;
+EXPLAIN select count(*) from srcpart join srcpart_date_hour on (srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr) where srcpart_date_hour.`date` = '2008-04-08' and srcpart_date_hour.hour = 11;
+select count(*) from srcpart join srcpart_date_hour on (srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr) where srcpart_date_hour.`date` = '2008-04-08' and srcpart_date_hour.hour = 11;
set hive.spark.dynamic.partition.pruning=true;
select count(*) from srcpart where ds = '2008-04-08' and hr = 11;
-- empty set
-EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = 'I DONT EXIST';
-select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = 'I DONT EXIST';
+EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = 'I DONT EXIST';
+select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = 'I DONT EXIST';
set hive.spark.dynamic.partition.pruning=false;
-EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = 'I DONT EXIST';
-select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = 'I DONT EXIST';
+EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = 'I DONT EXIST';
+select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = 'I DONT EXIST';
set hive.spark.dynamic.partition.pruning=true;
select count(*) from srcpart where ds = 'I DONT EXIST';
@@ -75,34 +75,34 @@ select count(*) from srcpart where cast(hr as string) = 11;
-- parent is reduce tasks
-EXPLAIN select count(*) from srcpart join (select ds as ds, ds as date from srcpart group by ds) s on (srcpart.ds = s.ds) where s.date = '2008-04-08';
-select count(*) from srcpart join (select ds as ds, ds as date from srcpart group by ds) s on (srcpart.ds = s.ds) where s.date = '2008-04-08';
+EXPLAIN select count(*) from srcpart join (select ds as ds, ds as `date` from srcpart group by ds) s on (srcpart.ds = s.ds) where s.`date` = '2008-04-08';
+select count(*) from srcpart join (select ds as ds, ds as `date` from srcpart group by ds) s on (srcpart.ds = s.ds) where s.`date` = '2008-04-08';
select count(*) from srcpart where ds = '2008-04-08';
-- non-equi join
-EXPLAIN select count(*) from srcpart, srcpart_date_hour where (srcpart_date_hour.date = '2008-04-08' and srcpart_date_hour.hour = 11) and (srcpart.ds = srcpart_date_hour.ds or srcpart.hr = srcpart_date_hour.hr);
-select count(*) from srcpart, srcpart_date_hour where (srcpart_date_hour.date = '2008-04-08' and srcpart_date_hour.hour = 11) and (srcpart.ds = srcpart_date_hour.ds or srcpart.hr = srcpart_date_hour.hr);
+EXPLAIN select count(*) from srcpart, srcpart_date_hour where (srcpart_date_hour.`date` = '2008-04-08' and srcpart_date_hour.hour = 11) and (srcpart.ds = srcpart_date_hour.ds or srcpart.hr = srcpart_date_hour.hr);
+select count(*) from srcpart, srcpart_date_hour where (srcpart_date_hour.`date` = '2008-04-08' and srcpart_date_hour.hour = 11) and (srcpart.ds = srcpart_date_hour.ds or srcpart.hr = srcpart_date_hour.hr);
-- old style join syntax
-EXPLAIN select count(*) from srcpart, srcpart_date_hour where srcpart_date_hour.date = '2008-04-08' and srcpart_date_hour.hour = 11 and srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr;
-select count(*) from srcpart, srcpart_date_hour where srcpart_date_hour.date = '2008-04-08' and srcpart_date_hour.hour = 11 and srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr;
+EXPLAIN select count(*) from srcpart, srcpart_date_hour where srcpart_date_hour.`date` = '2008-04-08' and srcpart_date_hour.hour = 11 and srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr;
+select count(*) from srcpart, srcpart_date_hour where srcpart_date_hour.`date` = '2008-04-08' and srcpart_date_hour.hour = 11 and srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr;
-- left join
-EXPLAIN select count(*) from srcpart left join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = '2008-04-08';
-EXPLAIN select count(*) from srcpart_date left join srcpart on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = '2008-04-08';
+EXPLAIN select count(*) from srcpart left join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = '2008-04-08';
+EXPLAIN select count(*) from srcpart_date left join srcpart on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = '2008-04-08';
-- full outer
-EXPLAIN select count(*) from srcpart full outer join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = '2008-04-08';
+EXPLAIN select count(*) from srcpart full outer join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = '2008-04-08';
-- with static pruning
EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) join srcpart_hour on (srcpart.hr = srcpart_hour.hr)
-where srcpart_date.date = '2008-04-08' and srcpart_hour.hour = 11 and srcpart.hr = 11;
+where srcpart_date.`date` = '2008-04-08' and srcpart_hour.hour = 11 and srcpart.hr = 11;
select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) join srcpart_hour on (srcpart.hr = srcpart_hour.hr)
-where srcpart_date.date = '2008-04-08' and srcpart_hour.hour = 11 and srcpart.hr = 11;
+where srcpart_date.`date` = '2008-04-08' and srcpart_hour.hour = 11 and srcpart.hr = 11;
EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) join srcpart_hour on (srcpart.hr = srcpart_hour.hr)
-where srcpart_date.date = '2008-04-08' and srcpart.hr = 13;
+where srcpart_date.`date` = '2008-04-08' and srcpart.hr = 13;
select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) join srcpart_hour on (srcpart.hr = srcpart_hour.hr)
-where srcpart_date.date = '2008-04-08' and srcpart.hr = 13;
+where srcpart_date.`date` = '2008-04-08' and srcpart.hr = 13;
-- union + subquery
EXPLAIN select count(*) from srcpart where srcpart.ds in (select max(srcpart.ds) from srcpart union all select min(srcpart.ds) from srcpart);
@@ -117,26 +117,26 @@ set hive.auto.convert.join.noconditionaltask = true;
set hive.auto.convert.join.noconditionaltask.size = 10000000;
-- single column, single key
-EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = '2008-04-08';
-select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = '2008-04-08';
+EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = '2008-04-08';
+select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = '2008-04-08';
select count(*) from srcpart where ds = '2008-04-08';
-- multiple sources, single key
EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) join srcpart_hour on (srcpart.hr = srcpart_hour.hr)
-where srcpart_date.date = '2008-04-08' and srcpart_hour.hour = 11;
+where srcpart_date.`date` = '2008-04-08' and srcpart_hour.hour = 11;
select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) join srcpart_hour on (srcpart.hr = srcpart_hour.hr)
-where srcpart_date.date = '2008-04-08' and srcpart_hour.hour = 11;
+where srcpart_date.`date` = '2008-04-08' and srcpart_hour.hour = 11;
select count(*) from srcpart where hr = 11 and ds = '2008-04-08';
-- multiple columns single source
-EXPLAIN select count(*) from srcpart join srcpart_date_hour on (srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr) where srcpart_date_hour.date = '2008-04-08' and srcpart_date_hour.hour = 11;
-select count(*) from srcpart join srcpart_date_hour on (srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr) where srcpart_date_hour.date = '2008-04-08' and srcpart_date_hour.hour = 11;
+EXPLAIN select count(*) from srcpart join srcpart_date_hour on (srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr) where srcpart_date_hour.`date` = '2008-04-08' and srcpart_date_hour.hour = 11;
+select count(*) from srcpart join srcpart_date_hour on (srcpart.ds = srcpart_date_hour.ds and srcpart.hr = srcpart_date_hour.hr) where srcpart_date_hour.`date` = '2008-04-08' and srcpart_date_hour.hour = 11;
select count(*) from srcpart where ds = '2008-04-08' and hr = 11;
-- empty set
-EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = 'I DONT EXIST';
+EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = 'I DONT EXIST';
-- Disabled until TEZ-1486 is fixed
--- select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = 'I DONT EXIST';
+-- select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = 'I DONT EXIST';
-- expressions
EXPLAIN select count(*) from srcpart join srcpart_double_hour on (srcpart.hr = cast(srcpart_double_hour.hr/2 as int)) where srcpart_double_hour.hour = 11;
@@ -146,27 +146,27 @@ select count(*) from srcpart join srcpart_double_hour on (srcpart.hr*2 = srcpart
select count(*) from srcpart where hr = 11;
-- parent is reduce tasks
-EXPLAIN select count(*) from srcpart join (select ds as ds, ds as date from srcpart group by ds) s on (srcpart.ds = s.ds) where s.date = '2008-04-08';
-select count(*) from srcpart join (select ds as ds, ds as date from srcpart group by ds) s on (srcpart.ds = s.ds) where s.date = '2008-04-08';
+EXPLAIN select count(*) from srcpart join (select ds as ds, ds as `date` from srcpart group by ds) s on (srcpart.ds = s.ds) where s.`date` = '2008-04-08';
+select count(*) from srcpart join (select ds as ds, ds as `date` from srcpart group by ds) s on (srcpart.ds = s.ds) where s.`date` = '2008-04-08';
select count(*) from srcpart where ds = '2008-04-08';
-- left join
-EXPLAIN select count(*) from srcpart left join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = '2008-04-08';
-EXPLAIN select count(*) from srcpart_date left join srcpart on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = '2008-04-08';
+EXPLAIN select count(*) from srcpart left join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = '2008-04-08';
+EXPLAIN select count(*) from srcpart_date left join srcpart on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = '2008-04-08';
-- full outer
-EXPLAIN select count(*) from srcpart full outer join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.date = '2008-04-08';
+EXPLAIN select count(*) from srcpart full outer join srcpart_date on (srcpart.ds = srcpart_date.ds) where srcpart_date.`date` = '2008-04-08';
-- with static pruning
EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) join srcpart_hour on (srcpart.hr = srcpart_hour.hr)
-where srcpart_date.date = '2008-04-08' and srcpart_hour.hour = 11 and srcpart.hr = 11;
+where srcpart_date.`date` = '2008-04-08' and srcpart_hour.hour = 11 and srcpart.hr = 11;
select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) join srcpart_hour on (srcpart.hr = srcpart_hour.hr)
-where srcpart_date.date = '2008-04-08' and srcpart_hour.hour = 11 and srcpart.hr = 11;
+where srcpart_date.`date` = '2008-04-08' and srcpart_hour.hour = 11 and srcpart.hr = 11;
EXPLAIN select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) join srcpart_hour on (srcpart.hr = srcpart_hour.hr)
-where srcpart_date.date = '2008-04-08' and srcpart.hr = 13;
+where srcpart_date.`date` = '2008-04-08' and srcpart.hr = 13;
-- Disabled until TEZ-1486 is fixed
-- select count(*) from srcpart join srcpart_date on (srcpart.ds = srcpart_date.ds) join srcpart_hour on (srcpart.hr = srcpart_hour.hr)
--- where srcpart_date.date = '2008-04-08' and srcpart.hr = 13;
+-- where srcpart_date.`date` = '2008-04-08' and srcpart.hr = 13;
-- union + subquery
EXPLAIN select distinct(ds) from srcpart where srcpart.ds in (select max(srcpart.ds) from srcpart union all select min(srcpart.ds) from srcpart);
http://git-wip-us.apache.org/repos/asf/hive/blob/858ce8c2/ql/src/test/queries/clientpositive/spark_vectorized_dynamic_partition_pruning.q
----------------------------------------------------------------------
diff --git a/ql/src/test/queries/clientpositive/spark_vectorized_dynamic_partition_pruning.q b/ql/src/test/queries/clientpositive/spark_vectorized_dynamic_partition_pruning.q
index 293fcfc..791c3a9 100644
--- a/ql/src/test/queries/clientpositive/spark_vectorized_dynamic_partition_pruning.q
+++ b/ql/src/test/queries/clientpositive/spark_vectorized_dynamic_partition_pruning.q
@@ -4,6 +4,7 @@ set hive.spark.dynamic.partition.pruning=true;
set hive.optimize.metadataonly=false;
set hive.optimize.index.filter=true;
set hive.vectorized.execution.enabled=true;
+set hive.strict.checks.cartesian.product=false;
select distinct ds from srcpart;
http://git-wip-us.apache.org/repos/asf/hive/blob/858ce8c2/ql/src/test/results/clientpositive/spark/spark_combine_equivalent_work.q.out
----------------------------------------------------------------------
diff --git a/ql/src/test/results/clientpositive/spark/spark_combine_equivalent_work.q.out b/ql/src/test/results/clientpositive/spark/spark_combine_equivalent_work.q.out
index 93d07d6..b7116c1 100644
--- a/ql/src/test/results/clientpositive/spark/spark_combine_equivalent_work.q.out
+++ b/ql/src/test/results/clientpositive/spark/spark_combine_equivalent_work.q.out
@@ -30,11 +30,9 @@ POSTHOOK: Output: default@a1
POSTHOOK: Output: default@a1@end_dt=20161021
PREHOOK: query: insert into table a1 partition(END_DT='20161020') values('2000721360','20161001')
PREHOOK: type: QUERY
-PREHOOK: Input: default@values__tmp__table__1
PREHOOK: Output: default@a1@end_dt=20161020
POSTHOOK: query: insert into table a1 partition(END_DT='20161020') values('2000721360','20161001')
POSTHOOK: type: QUERY
-POSTHOOK: Input: default@values__tmp__table__1
POSTHOOK: Output: default@a1@end_dt=20161020
POSTHOOK: Lineage: a1 PARTITION(end_dt=20161020).kehhao SIMPLE [(values__tmp__table__1)values__tmp__table__1.FieldSchema(name:tmp_values_col1, type:string, comment:), ]
POSTHOOK: Lineage: a1 PARTITION(end_dt=20161020).start_dt SIMPLE [(values__tmp__table__1)values__tmp__table__1.FieldSchema(name:tmp_values_col2, type:string, comment:), ]