[1/2] spark git commit: [SPARK-16621][SQL] Generate stable SQLs in SQLBuilder

2016-07-27 Thread rxin 
Repository: spark
Updated Branches:
  refs/heads/branch-2.0 44234b1c4 -> be9965b07


http://git-wip-us.apache.org/repos/asf/spark/blob/be9965b0/sql/hive/src/test/resources/sqlgen/rollup_cube_4_2.sql
--
diff --git a/sql/hive/src/test/resources/sqlgen/rollup_cube_4_2.sql 
b/sql/hive/src/test/resources/sqlgen/rollup_cube_4_2.sql
index eebef6a..8bf1645 100644
--- a/sql/hive/src/test/resources/sqlgen/rollup_cube_4_2.sql
+++ b/sql/hive/src/test/resources/sqlgen/rollup_cube_4_2.sql
@@ -2,4 +2,4 @@
 SELECT count(*) as cnt, key % 5 as k1, key - 5 as k2, grouping_id() FROM 
parquet_t1
 GROUP BY key % 5, key - 5 WITH CUBE
 

-SELECT `gen_attr` AS `cnt`, `gen_attr` AS `k1`, `gen_attr` AS `k2`, `gen_attr` 
AS `grouping_id()` FROM (SELECT count(1) AS `gen_attr`, (`gen_attr` % CAST(5 AS 
BIGINT)) AS `gen_attr`, (`gen_attr` - CAST(5 AS BIGINT)) AS `gen_attr`, 
grouping_id() AS `gen_attr` FROM (SELECT `key` AS `gen_attr`, `value` AS 
`gen_attr` FROM `default`.`parquet_t1`) AS gen_subquery_0 GROUP BY (`gen_attr` 
% CAST(5 AS BIGINT)), (`gen_attr` - CAST(5 AS BIGINT)) GROUPING 
SETS(((`gen_attr` % CAST(5 AS BIGINT)), (`gen_attr` - CAST(5 AS BIGINT))), 
((`gen_attr` % CAST(5 AS BIGINT))), ((`gen_attr` - CAST(5 AS BIGINT))), ())) AS 
gen_subquery_1
+SELECT `gen_attr_3` AS `cnt`, `gen_attr_4` AS `k1`, `gen_attr_5` AS `k2`, 
`gen_attr_6` AS `grouping_id()` FROM (SELECT count(1) AS `gen_attr_3`, 
(`gen_attr_7` % CAST(5 AS BIGINT)) AS `gen_attr_4`, (`gen_attr_7` - CAST(5 AS 
BIGINT)) AS `gen_attr_5`, grouping_id() AS `gen_attr_6` FROM (SELECT `key` AS 
`gen_attr_7`, `value` AS `gen_attr_8` FROM `default`.`parquet_t1`) AS 
gen_subquery_0 GROUP BY (`gen_attr_7` % CAST(5 AS BIGINT)), (`gen_attr_7` - 
CAST(5 AS BIGINT)) GROUPING SETS(((`gen_attr_7` % CAST(5 AS BIGINT)), 
(`gen_attr_7` - CAST(5 AS BIGINT))), ((`gen_attr_7` % CAST(5 AS BIGINT))), 
((`gen_attr_7` - CAST(5 AS BIGINT))), ())) AS gen_subquery_1

http://git-wip-us.apache.org/repos/asf/spark/blob/be9965b0/sql/hive/src/test/resources/sqlgen/rollup_cube_5_1.sql
--
diff --git a/sql/hive/src/test/resources/sqlgen/rollup_cube_5_1.sql 
b/sql/hive/src/test/resources/sqlgen/rollup_cube_5_1.sql
index 9474233..17e78a0 100644
--- a/sql/hive/src/test/resources/sqlgen/rollup_cube_5_1.sql
+++ b/sql/hive/src/test/resources/sqlgen/rollup_cube_5_1.sql
@@ -3,4 +3,4 @@ SELECT count(*) AS cnt, key % 5 AS k1, key - 5 AS k2, 
grouping_id(key % 5, key -
 FROM (SELECT key, key%2, key - 5 FROM parquet_t1) t GROUP BY key%5, key-5
 WITH ROLLUP
 

-SELECT `gen_attr` AS `cnt`, `gen_attr` AS `k1`, `gen_attr` AS `k2`, `gen_attr` 
AS `k3` FROM (SELECT count(1) AS `gen_attr`, (`gen_attr` % CAST(5 AS BIGINT)) 
AS `gen_attr`, (`gen_attr` - CAST(5 AS BIGINT)) AS `gen_attr`, grouping_id() AS 
`gen_attr` FROM (SELECT `gen_attr`, (`gen_attr` % CAST(2 AS BIGINT)) AS 
`gen_attr`, (`gen_attr` - CAST(5 AS BIGINT)) AS `gen_attr` FROM (SELECT `key` 
AS `gen_attr`, `value` AS `gen_attr` FROM `default`.`parquet_t1`) AS 
gen_subquery_0) AS t GROUP BY (`gen_attr` % CAST(5 AS BIGINT)), (`gen_attr` - 
CAST(5 AS BIGINT)) GROUPING SETS(((`gen_attr` % CAST(5 AS BIGINT)), (`gen_attr` 
- CAST(5 AS BIGINT))), ((`gen_attr` % CAST(5 AS BIGINT))), ())) AS 
gen_subquery_1
+SELECT `gen_attr_3` AS `cnt`, `gen_attr_4` AS `k1`, `gen_attr_5` AS `k2`, 
`gen_attr_6` AS `k3` FROM (SELECT count(1) AS `gen_attr_3`, (`gen_attr_7` % 
CAST(5 AS BIGINT)) AS `gen_attr_4`, (`gen_attr_7` - CAST(5 AS BIGINT)) AS 
`gen_attr_5`, grouping_id() AS `gen_attr_6` FROM (SELECT `gen_attr_7`, 
(`gen_attr_7` % CAST(2 AS BIGINT)) AS `gen_attr_8`, (`gen_attr_7` - CAST(5 AS 
BIGINT)) AS `gen_attr_9` FROM (SELECT `key` AS `gen_attr_7`, `value` AS 
`gen_attr_12` FROM `default`.`parquet_t1`) AS gen_subquery_0) AS t GROUP BY 
(`gen_attr_7` % CAST(5 AS BIGINT)), (`gen_attr_7` - CAST(5 AS BIGINT)) GROUPING 
SETS(((`gen_attr_7` % CAST(5 AS BIGINT)), (`gen_attr_7` - CAST(5 AS BIGINT))), 
((`gen_attr_7` % CAST(5 AS BIGINT))), ())) AS gen_subquery_1

http://git-wip-us.apache.org/repos/asf/spark/blob/be9965b0/sql/hive/src/test/resources/sqlgen/rollup_cube_5_2.sql
--
diff --git a/sql/hive/src/test/resources/sqlgen/rollup_cube_5_2.sql 
b/sql/hive/src/test/resources/sqlgen/rollup_cube_5_2.sql
index d36f43d..72506ef 100644
--- a/sql/hive/src/test/resources/sqlgen/rollup_cube_5_2.sql
+++ b/sql/hive/src/test/resources/sqlgen/rollup_cube_5_2.sql
@@ -3,4 +3,4 @@ SELECT count(*) AS cnt, key % 5 AS k1, key - 5 AS k2, 
grouping_id(key % 5, key -
 FROM (SELECT key, key % 2, key - 5 FROM parquet_t1) t GROUP BY key % 5, key - 5
 WITH CUBE
 

-SELECT

[1/2] spark git commit: [SPARK-16621][SQL] Generate stable SQLs in SQLBuilder

2016-07-26 Thread lian 
Repository: spark
Updated Branches:
  refs/heads/master 738b4cc54 -> 5b8e848bb


http://git-wip-us.apache.org/repos/asf/spark/blob/5b8e848b/sql/hive/src/test/resources/sqlgen/rollup_cube_4_2.sql
--
diff --git a/sql/hive/src/test/resources/sqlgen/rollup_cube_4_2.sql 
b/sql/hive/src/test/resources/sqlgen/rollup_cube_4_2.sql
index eebef6a..8bf1645 100644
--- a/sql/hive/src/test/resources/sqlgen/rollup_cube_4_2.sql
+++ b/sql/hive/src/test/resources/sqlgen/rollup_cube_4_2.sql
@@ -2,4 +2,4 @@
 SELECT count(*) as cnt, key % 5 as k1, key - 5 as k2, grouping_id() FROM 
parquet_t1
 GROUP BY key % 5, key - 5 WITH CUBE
 

-SELECT `gen_attr` AS `cnt`, `gen_attr` AS `k1`, `gen_attr` AS `k2`, `gen_attr` 
AS `grouping_id()` FROM (SELECT count(1) AS `gen_attr`, (`gen_attr` % CAST(5 AS 
BIGINT)) AS `gen_attr`, (`gen_attr` - CAST(5 AS BIGINT)) AS `gen_attr`, 
grouping_id() AS `gen_attr` FROM (SELECT `key` AS `gen_attr`, `value` AS 
`gen_attr` FROM `default`.`parquet_t1`) AS gen_subquery_0 GROUP BY (`gen_attr` 
% CAST(5 AS BIGINT)), (`gen_attr` - CAST(5 AS BIGINT)) GROUPING 
SETS(((`gen_attr` % CAST(5 AS BIGINT)), (`gen_attr` - CAST(5 AS BIGINT))), 
((`gen_attr` % CAST(5 AS BIGINT))), ((`gen_attr` - CAST(5 AS BIGINT))), ())) AS 
gen_subquery_1
+SELECT `gen_attr_3` AS `cnt`, `gen_attr_4` AS `k1`, `gen_attr_5` AS `k2`, 
`gen_attr_6` AS `grouping_id()` FROM (SELECT count(1) AS `gen_attr_3`, 
(`gen_attr_7` % CAST(5 AS BIGINT)) AS `gen_attr_4`, (`gen_attr_7` - CAST(5 AS 
BIGINT)) AS `gen_attr_5`, grouping_id() AS `gen_attr_6` FROM (SELECT `key` AS 
`gen_attr_7`, `value` AS `gen_attr_8` FROM `default`.`parquet_t1`) AS 
gen_subquery_0 GROUP BY (`gen_attr_7` % CAST(5 AS BIGINT)), (`gen_attr_7` - 
CAST(5 AS BIGINT)) GROUPING SETS(((`gen_attr_7` % CAST(5 AS BIGINT)), 
(`gen_attr_7` - CAST(5 AS BIGINT))), ((`gen_attr_7` % CAST(5 AS BIGINT))), 
((`gen_attr_7` - CAST(5 AS BIGINT))), ())) AS gen_subquery_1

http://git-wip-us.apache.org/repos/asf/spark/blob/5b8e848b/sql/hive/src/test/resources/sqlgen/rollup_cube_5_1.sql
--
diff --git a/sql/hive/src/test/resources/sqlgen/rollup_cube_5_1.sql 
b/sql/hive/src/test/resources/sqlgen/rollup_cube_5_1.sql
index 9474233..17e78a0 100644
--- a/sql/hive/src/test/resources/sqlgen/rollup_cube_5_1.sql
+++ b/sql/hive/src/test/resources/sqlgen/rollup_cube_5_1.sql
@@ -3,4 +3,4 @@ SELECT count(*) AS cnt, key % 5 AS k1, key - 5 AS k2, 
grouping_id(key % 5, key -
 FROM (SELECT key, key%2, key - 5 FROM parquet_t1) t GROUP BY key%5, key-5
 WITH ROLLUP
 

-SELECT `gen_attr` AS `cnt`, `gen_attr` AS `k1`, `gen_attr` AS `k2`, `gen_attr` 
AS `k3` FROM (SELECT count(1) AS `gen_attr`, (`gen_attr` % CAST(5 AS BIGINT)) 
AS `gen_attr`, (`gen_attr` - CAST(5 AS BIGINT)) AS `gen_attr`, grouping_id() AS 
`gen_attr` FROM (SELECT `gen_attr`, (`gen_attr` % CAST(2 AS BIGINT)) AS 
`gen_attr`, (`gen_attr` - CAST(5 AS BIGINT)) AS `gen_attr` FROM (SELECT `key` 
AS `gen_attr`, `value` AS `gen_attr` FROM `default`.`parquet_t1`) AS 
gen_subquery_0) AS t GROUP BY (`gen_attr` % CAST(5 AS BIGINT)), (`gen_attr` - 
CAST(5 AS BIGINT)) GROUPING SETS(((`gen_attr` % CAST(5 AS BIGINT)), (`gen_attr` 
- CAST(5 AS BIGINT))), ((`gen_attr` % CAST(5 AS BIGINT))), ())) AS 
gen_subquery_1
+SELECT `gen_attr_3` AS `cnt`, `gen_attr_4` AS `k1`, `gen_attr_5` AS `k2`, 
`gen_attr_6` AS `k3` FROM (SELECT count(1) AS `gen_attr_3`, (`gen_attr_7` % 
CAST(5 AS BIGINT)) AS `gen_attr_4`, (`gen_attr_7` - CAST(5 AS BIGINT)) AS 
`gen_attr_5`, grouping_id() AS `gen_attr_6` FROM (SELECT `gen_attr_7`, 
(`gen_attr_7` % CAST(2 AS BIGINT)) AS `gen_attr_8`, (`gen_attr_7` - CAST(5 AS 
BIGINT)) AS `gen_attr_9` FROM (SELECT `key` AS `gen_attr_7`, `value` AS 
`gen_attr_12` FROM `default`.`parquet_t1`) AS gen_subquery_0) AS t GROUP BY 
(`gen_attr_7` % CAST(5 AS BIGINT)), (`gen_attr_7` - CAST(5 AS BIGINT)) GROUPING 
SETS(((`gen_attr_7` % CAST(5 AS BIGINT)), (`gen_attr_7` - CAST(5 AS BIGINT))), 
((`gen_attr_7` % CAST(5 AS BIGINT))), ())) AS gen_subquery_1

http://git-wip-us.apache.org/repos/asf/spark/blob/5b8e848b/sql/hive/src/test/resources/sqlgen/rollup_cube_5_2.sql
--
diff --git a/sql/hive/src/test/resources/sqlgen/rollup_cube_5_2.sql 
b/sql/hive/src/test/resources/sqlgen/rollup_cube_5_2.sql
index d36f43d..72506ef 100644
--- a/sql/hive/src/test/resources/sqlgen/rollup_cube_5_2.sql
+++ b/sql/hive/src/test/resources/sqlgen/rollup_cube_5_2.sql
@@ -3,4 +3,4 @@ SELECT count(*) AS cnt, key % 5 AS k1, key - 5 AS k2, 
grouping_id(key % 5, key -
 FROM (SELECT key, key % 2, key - 5 FROM parquet_t1) t GROUP BY key % 5, key - 5
 WITH CUBE
 

-SELECT