HIVE-14503: Remove explicit order by in qfiles for union tests (Prasanth Jayachandran reviewed by Siddharth Seth, Hari)
Project: http://git-wip-us.apache.org/repos/asf/hive/repo Commit: http://git-wip-us.apache.org/repos/asf/hive/commit/d14673a0 Tree: http://git-wip-us.apache.org/repos/asf/hive/tree/d14673a0 Diff: http://git-wip-us.apache.org/repos/asf/hive/diff/d14673a0 Branch: refs/heads/master Commit: d14673a087370d28383eedbe20d09f0375bea7e7 Parents: 59a9511 Author: Prasanth Jayachandran <[email protected]> Authored: Fri Aug 19 18:38:14 2016 -0700 Committer: Prasanth Jayachandran <[email protected]> Committed: Fri Aug 19 18:38:14 2016 -0700 ---------------------------------------------------------------------- ql/src/test/queries/clientpositive/union32.q | 3 +- ql/src/test/queries/clientpositive/union34.q | 9 +- ql/src/test/queries/clientpositive/union36.q | 5 +- .../queries/clientpositive/unionDistinct_1.q | 17 +- .../queries/clientpositive/union_remove_1.q | 3 +- .../queries/clientpositive/union_remove_10.q | 3 +- .../queries/clientpositive/union_remove_11.q | 3 +- .../queries/clientpositive/union_remove_12.q | 3 +- .../queries/clientpositive/union_remove_13.q | 3 +- .../queries/clientpositive/union_remove_14.q | 3 +- .../queries/clientpositive/union_remove_15.q | 5 +- .../queries/clientpositive/union_remove_16.q | 5 +- .../queries/clientpositive/union_remove_17.q | 5 +- .../queries/clientpositive/union_remove_18.q | 7 +- .../queries/clientpositive/union_remove_19.q | 3 +- .../queries/clientpositive/union_remove_2.q | 3 +- .../queries/clientpositive/union_remove_20.q | 3 +- .../queries/clientpositive/union_remove_21.q | 3 +- .../queries/clientpositive/union_remove_22.q | 2 +- .../queries/clientpositive/union_remove_23.q | 3 +- .../queries/clientpositive/union_remove_24.q | 3 +- .../queries/clientpositive/union_remove_25.q | 3 +- .../queries/clientpositive/union_remove_26.q | 2 +- .../queries/clientpositive/union_remove_3.q | 3 +- .../queries/clientpositive/union_remove_4.q | 3 +- .../queries/clientpositive/union_remove_5.q | 3 +- .../queries/clientpositive/union_remove_6.q | 5 +- .../clientpositive/union_remove_6_subq.q | 4 +- .../queries/clientpositive/union_remove_7.q | 3 +- .../queries/clientpositive/union_remove_8.q | 3 +- .../queries/clientpositive/union_remove_9.q | 3 +- .../test/queries/clientpositive/union_script.q | 5 +- .../queries/clientpositive/union_type_chk.q | 6 +- ql/src/test/queries/clientpositive/union_view.q | 5 +- .../results/clientpositive/spark/union32.q.out | 37 ++--- .../results/clientpositive/spark/union34.q.out | 115 ++++++-------- .../clientpositive/spark/union_remove_1.q.out | 10 +- .../clientpositive/spark/union_remove_10.q.out | 10 +- .../clientpositive/spark/union_remove_11.q.out | 10 +- .../clientpositive/spark/union_remove_12.q.out | 10 +- .../clientpositive/spark/union_remove_13.q.out | 12 +- .../clientpositive/spark/union_remove_14.q.out | 10 +- .../clientpositive/spark/union_remove_15.q.out | 14 +- .../clientpositive/spark/union_remove_16.q.out | 14 +- .../clientpositive/spark/union_remove_17.q.out | 14 +- .../clientpositive/spark/union_remove_18.q.out | 18 ++- .../clientpositive/spark/union_remove_19.q.out | 10 +- .../clientpositive/spark/union_remove_2.q.out | 10 +- .../clientpositive/spark/union_remove_20.q.out | 10 +- .../clientpositive/spark/union_remove_21.q.out | 10 +- .../clientpositive/spark/union_remove_22.q.out | 4 +- .../clientpositive/spark/union_remove_23.q.out | 10 +- .../clientpositive/spark/union_remove_24.q.out | 10 +- .../clientpositive/spark/union_remove_25.q.out | 10 +- .../clientpositive/spark/union_remove_3.q.out | 10 +- .../clientpositive/spark/union_remove_4.q.out | 10 +- .../clientpositive/spark/union_remove_5.q.out | 10 +- .../clientpositive/spark/union_remove_6.q.out | 14 +- .../spark/union_remove_6_subq.q.out | 8 +- .../clientpositive/spark/union_remove_7.q.out | 10 +- .../clientpositive/spark/union_remove_8.q.out | 10 +- .../clientpositive/spark/union_remove_9.q.out | 10 +- .../clientpositive/spark/union_script.q.out | 14 +- .../clientpositive/spark/union_view.q.out | 63 ++++---- .../clientpositive/tez/unionDistinct_1.q.out | 154 ++++++------------- .../clientpositive/tez/union_type_chk.q.out | 18 ++- .../test/results/clientpositive/union32.q.out | 32 ++-- .../test/results/clientpositive/union34.q.out | 106 ++++++------- .../test/results/clientpositive/union36.q.out | 18 ++- .../clientpositive/unionDistinct_1.q.out | 132 +++------------- .../results/clientpositive/union_remove_1.q.out | 10 +- .../clientpositive/union_remove_10.q.out | 10 +- .../clientpositive/union_remove_11.q.out | 10 +- .../clientpositive/union_remove_12.q.out | 10 +- .../clientpositive/union_remove_13.q.out | 12 +- .../clientpositive/union_remove_14.q.out | 10 +- .../clientpositive/union_remove_15.q.out | 14 +- .../clientpositive/union_remove_16.q.out | 14 +- .../clientpositive/union_remove_17.q.out | 14 +- .../clientpositive/union_remove_18.q.out | 18 ++- .../clientpositive/union_remove_19.q.out | 10 +- .../results/clientpositive/union_remove_2.q.out | 10 +- .../clientpositive/union_remove_20.q.out | 10 +- .../clientpositive/union_remove_21.q.out | 10 +- .../clientpositive/union_remove_22.q.out | 4 +- .../clientpositive/union_remove_23.q.out | 10 +- .../clientpositive/union_remove_24.q.out | 10 +- .../clientpositive/union_remove_25.q.out | 10 +- .../results/clientpositive/union_remove_3.q.out | 10 +- .../results/clientpositive/union_remove_4.q.out | 10 +- .../results/clientpositive/union_remove_5.q.out | 10 +- .../results/clientpositive/union_remove_6.q.out | 14 +- .../clientpositive/union_remove_6_subq.q.out | 8 +- .../results/clientpositive/union_remove_7.q.out | 10 +- .../results/clientpositive/union_remove_8.q.out | 10 +- .../results/clientpositive/union_remove_9.q.out | 10 +- .../results/clientpositive/union_script.q.out | 14 +- .../results/clientpositive/union_type_chk.q.out | 18 ++- .../results/clientpositive/union_view.q.out | 54 +++---- 99 files changed, 700 insertions(+), 773 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union32.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union32.q b/ql/src/test/queries/clientpositive/union32.q index f47f0af..b18d484 100644 --- a/ql/src/test/queries/clientpositive/union32.q +++ b/ql/src/test/queries/clientpositive/union32.q @@ -12,8 +12,7 @@ EXPLAIN SELECT * FROM (SELECT CAST(key AS DOUBLE) AS key FROM t1 UNION ALL -SELECT CAST(key AS BIGINT) AS key FROM t2) a -ORDER BY key; +SELECT CAST(key AS BIGINT) AS key FROM t2) a; SELECT * FROM (SELECT CAST(key AS DOUBLE) AS key FROM t1 http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union34.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union34.q b/ql/src/test/queries/clientpositive/union34.q index 2ab16de..dbefc40 100644 --- a/ql/src/test/queries/clientpositive/union34.q +++ b/ql/src/test/queries/clientpositive/union34.q @@ -1,4 +1,5 @@ set hive.mapred.mode=nonstrict; +-- SORT_QUERY_RESULTS create table src10_1 (key string, value string); create table src10_2 (key string, value string); create table src10_3 (key string, value string); @@ -18,13 +19,13 @@ SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION ALL SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION ALL SELECT * FROM src10_4 ) alias0 -) alias1 order by key; +) alias1; SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION ALL SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION ALL SELECT * FROM src10_4 ) alias0 -) alias1 order by key; +) alias1; set hive.auto.convert.join=false; -- When we do not convert the Join of sub1 and sub0 into a MapJoin, @@ -36,10 +37,10 @@ SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION ALL SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION ALL SELECT * FROM src10_4 ) alias0 -) alias1 order by key; +) alias1; SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION ALL SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION ALL SELECT * FROM src10_4 ) alias0 -) alias1 order by key; +) alias1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union36.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union36.q b/ql/src/test/queries/clientpositive/union36.q index c38e7b1..b79ff0f 100644 --- a/ql/src/test/queries/clientpositive/union36.q +++ b/ql/src/test/queries/clientpositive/union36.q @@ -1,9 +1,10 @@ set hive.mapred.mode=nonstrict; set hive.cbo.enable=false; -select (x/sum(x) over()) as y from(select cast(1 as decimal(10,0)) as x from (select * from src limit 2)s1 union all select cast(1 as decimal(10,0)) x from (select * from src limit 2) s2 union all select '100000000' x from (select * from src limit 2) s3)u order by y; +-- SORT_QUERY_RESULTS +select (x/sum(x) over()) as y from(select cast(1 as decimal(10,0)) as x from (select * from src limit 2)s1 union all select cast(1 as decimal(10,0)) x from (select * from src limit 2) s2 union all select '100000000' x from (select * from src limit 2) s3)u; -select (x/sum(x) over()) as y from(select cast(1 as decimal(10,0)) as x from (select * from src limit 2)s1 union all select cast(1 as decimal(10,0)) x from (select * from src limit 2) s2 union all select cast (null as string) x from (select * from src limit 2) s3)u order by y; +select (x/sum(x) over()) as y from(select cast(1 as decimal(10,0)) as x from (select * from src limit 2)s1 union all select cast(1 as decimal(10,0)) x from (select * from src limit 2) s2 union all select cast (null as string) x from (select * from src limit 2) s3)u; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/unionDistinct_1.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/unionDistinct_1.q b/ql/src/test/queries/clientpositive/unionDistinct_1.q index 0d53a96..5c52d9b 100644 --- a/ql/src/test/queries/clientpositive/unionDistinct_1.q +++ b/ql/src/test/queries/clientpositive/unionDistinct_1.q @@ -386,16 +386,14 @@ from ( select transform(key, value) using 'cat' as (key2, value2) from src UNION DISTINCT - select key as key2, value as value2 from src) s -order by s.key2, s.value2; + select key as key2, value as value2 from src) s; select s.key2, s.value2 from ( select transform(key, value) using 'cat' as (key2, value2) from src UNION DISTINCT - select key as key2, value as value2 from src) s -order by s.key2, s.value2; + select key as key2, value as value2 from src) s; -- union24.q @@ -874,8 +872,7 @@ EXPLAIN SELECT * FROM (SELECT CAST(key AS DOUBLE) AS key FROM t1 UNION DISTINCT -SELECT CAST(key AS BIGINT) AS key FROM t2) a -ORDER BY key; +SELECT CAST(key AS BIGINT) AS key FROM t2) a; SELECT * FROM (SELECT CAST(key AS DOUBLE) AS key FROM t1 @@ -1013,13 +1010,13 @@ SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION DISTINCT SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION DISTINCT SELECT * FROM src10_4 ) alias0 -) alias1 order by key; +) alias1; SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION DISTINCT SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION DISTINCT SELECT * FROM src10_4 ) alias0 -) alias1 order by key; +) alias1; set hive.auto.convert.join=false; -- When we do not convert the Join of sub1 and sub0 into a MapJoin, @@ -1031,13 +1028,13 @@ SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION DISTINCT SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION DISTINCT SELECT * FROM src10_4 ) alias0 -) alias1 order by key; +) alias1; SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION DISTINCT SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION DISTINCT SELECT * FROM src10_4 ) alias0 -) alias1 order by key; +) alias1; -- union4.q http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_1.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_1.q b/ql/src/test/queries/clientpositive/union_remove_1.q index 702f71a..0a69068 100644 --- a/ql/src/test/queries/clientpositive/union_remove_1.q +++ b/ql/src/test/queries/clientpositive/union_remove_1.q @@ -7,6 +7,7 @@ set hive.merge.mapfiles=false; set hive.merge.mapredfiles=false; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- There is no need to write the temporary results of the sub-queries, and then read them @@ -42,4 +43,4 @@ FROM ( desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_10.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_10.q b/ql/src/test/queries/clientpositive/union_remove_10.q index 508f354..7892e15 100644 --- a/ql/src/test/queries/clientpositive/union_remove_10.q +++ b/ql/src/test/queries/clientpositive/union_remove_10.q @@ -8,6 +8,7 @@ set hive.merge.mapredfiles=true; set hive.merge.smallfiles.avgsize=1; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a map-only query, and the -- other one contains a nested union where one of the sub-queries requires a map-reduce @@ -55,4 +56,4 @@ select * FROM ( desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_11.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_11.q b/ql/src/test/queries/clientpositive/union_remove_11.q index bdfbf66..fd41648 100644 --- a/ql/src/test/queries/clientpositive/union_remove_11.q +++ b/ql/src/test/queries/clientpositive/union_remove_11.q @@ -8,6 +8,7 @@ set hive.merge.mapredfiles=true; set hive.merge.smallfiles.avgsize=1; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a map-only query, and the -- other one contains a nested union where also contains map only sub-queries), @@ -55,4 +56,4 @@ select * FROM ( desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_12.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_12.q b/ql/src/test/queries/clientpositive/union_remove_12.q index f6436f5..b665666 100644 --- a/ql/src/test/queries/clientpositive/union_remove_12.q +++ b/ql/src/test/queries/clientpositive/union_remove_12.q @@ -9,6 +9,7 @@ set hive.merge.mapredfiles=true; set hive.merge.smallfiles.avgsize=1; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a map-only query, and the -- other one is a map-join query), followed by select star and a file sink. @@ -49,4 +50,4 @@ FROM inputTbl1 a join inputTbl1 b on a.key=b.key desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_13.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_13.q b/ql/src/test/queries/clientpositive/union_remove_13.q index b02451b..11077fd 100644 --- a/ql/src/test/queries/clientpositive/union_remove_13.q +++ b/ql/src/test/queries/clientpositive/union_remove_13.q @@ -9,6 +9,7 @@ set hive.merge.mapredfiles=true; set hive.merge.smallfiles.avgsize=1; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a mapred query, and the -- other one is a map-join query), followed by select star and a file sink. @@ -49,4 +50,4 @@ FROM inputTbl1 a join inputTbl1 b on a.key=b.key desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_14.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_14.q b/ql/src/test/queries/clientpositive/union_remove_14.q index bec6226..b559b35 100644 --- a/ql/src/test/queries/clientpositive/union_remove_14.q +++ b/ql/src/test/queries/clientpositive/union_remove_14.q @@ -9,6 +9,7 @@ set hive.auto.convert.join=true; set hive.merge.smallfiles.avgsize=1; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a map-only query, and the -- other one contains a join, which should be performed as a map-join query at runtime), @@ -50,4 +51,4 @@ FROM inputTbl1 a join inputTbl1 b on a.key=b.key desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_15.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_15.q b/ql/src/test/queries/clientpositive/union_remove_15.q index e384739..43c7834 100644 --- a/ql/src/test/queries/clientpositive/union_remove_15.q +++ b/ql/src/test/queries/clientpositive/union_remove_15.q @@ -10,6 +10,7 @@ set hive.exec.dynamic.partition.mode=nonstrict; set hive.exec.dynamic.partition=true; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- and the results are written to a table using dynamic partitions. @@ -50,5 +51,5 @@ desc formatted outputTbl1; show partitions outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 where ds = '1' order by key, `values`; -select * from outputTbl1 where ds = '2' order by key, `values`; +select * from outputTbl1 where ds = '1'; +select * from outputTbl1 where ds = '2'; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_16.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_16.q b/ql/src/test/queries/clientpositive/union_remove_16.q index 41305bf..eca8aeb 100644 --- a/ql/src/test/queries/clientpositive/union_remove_16.q +++ b/ql/src/test/queries/clientpositive/union_remove_16.q @@ -11,6 +11,7 @@ set mapred.input.dir.recursive=true; set hive.exec.dynamic.partition.mode=nonstrict; set hive.exec.dynamic.partition=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- and the results are written to a table using dynamic partitions. @@ -49,5 +50,5 @@ desc formatted outputTbl1; show partitions outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 where ds = '1' order by key, `values`; -select * from outputTbl1 where ds = '2' order by key, `values`; +select * from outputTbl1 where ds = '1'; +select * from outputTbl1 where ds = '2'; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_17.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_17.q b/ql/src/test/queries/clientpositive/union_remove_17.q index 8f91d03..59a3a9c 100644 --- a/ql/src/test/queries/clientpositive/union_remove_17.q +++ b/ql/src/test/queries/clientpositive/union_remove_17.q @@ -10,6 +10,7 @@ set hive.exec.dynamic.partition.mode=nonstrict; set hive.exec.dynamic.partition=true; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- and the results are written to a table using dynamic partitions. @@ -46,5 +47,5 @@ desc formatted outputTbl1; show partitions outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 where ds = '1' order by key, `values`; -select * from outputTbl1 where ds = '2' order by key, `values`; +select * from outputTbl1 where ds = '1'; +select * from outputTbl1 where ds = '2'; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_18.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_18.q b/ql/src/test/queries/clientpositive/union_remove_18.q index ad38742..98ee7d0 100644 --- a/ql/src/test/queries/clientpositive/union_remove_18.q +++ b/ql/src/test/queries/clientpositive/union_remove_18.q @@ -10,6 +10,7 @@ set hive.exec.dynamic.partition.mode=nonstrict; set hive.exec.dynamic.partition=true; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- There is no need to write the temporary results of the sub-queries, and then read them @@ -49,6 +50,6 @@ desc formatted outputTbl1; show partitions outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 where ds = '11' order by key, `values`; -select * from outputTbl1 where ds = '18' order by key, `values`; -select * from outputTbl1 where ds is not null order by key, `values`, ds; +select * from outputTbl1 where ds = '11'; +select * from outputTbl1 where ds = '18'; +select * from outputTbl1 where ds is not null; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_19.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_19.q b/ql/src/test/queries/clientpositive/union_remove_19.q index 34f8b5a..9e47254 100644 --- a/ql/src/test/queries/clientpositive/union_remove_19.q +++ b/ql/src/test/queries/clientpositive/union_remove_19.q @@ -7,6 +7,7 @@ set hive.merge.mapfiles=false; set hive.merge.mapredfiles=false; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- There is no need to write the temporary results of the sub-queries, and then read them @@ -90,4 +91,4 @@ FROM ( ) b where b.key >= 7; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_2.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_2.q b/ql/src/test/queries/clientpositive/union_remove_2.q index 274608c..2ebc1ad 100644 --- a/ql/src/test/queries/clientpositive/union_remove_2.q +++ b/ql/src/test/queries/clientpositive/union_remove_2.q @@ -7,6 +7,7 @@ set hive.merge.mapfiles=false; set hive.merge.mapredfiles=false; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 3 subqueries is performed (exactly one of which requires a map-reduce job) -- followed by select star and a file sink. @@ -47,5 +48,5 @@ FROM ( desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_20.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_20.q b/ql/src/test/queries/clientpositive/union_remove_20.q index 93a49f3..700ee4d 100644 --- a/ql/src/test/queries/clientpositive/union_remove_20.q +++ b/ql/src/test/queries/clientpositive/union_remove_20.q @@ -7,6 +7,7 @@ set hive.merge.mapfiles=false; set hive.merge.mapredfiles=false; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select and a file sink -- However, the order of the columns in the select list is different. So, union cannot @@ -43,4 +44,4 @@ FROM ( desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_21.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_21.q b/ql/src/test/queries/clientpositive/union_remove_21.q index fa8ff27..1b1472e 100644 --- a/ql/src/test/queries/clientpositive/union_remove_21.q +++ b/ql/src/test/queries/clientpositive/union_remove_21.q @@ -7,6 +7,7 @@ set hive.merge.mapfiles=false; set hive.merge.mapredfiles=false; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select and a file sink -- However, all the columns are not selected. So, union cannot @@ -43,4 +44,4 @@ FROM ( desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_22.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_22.q b/ql/src/test/queries/clientpositive/union_remove_22.q index 0e97cae..d4d3cbc 100644 --- a/ql/src/test/queries/clientpositive/union_remove_22.q +++ b/ql/src/test/queries/clientpositive/union_remove_22.q @@ -63,4 +63,4 @@ FROM ( ) a; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_23.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_23.q b/ql/src/test/queries/clientpositive/union_remove_23.q index 436719d..98ae503 100644 --- a/ql/src/test/queries/clientpositive/union_remove_23.q +++ b/ql/src/test/queries/clientpositive/union_remove_23.q @@ -7,6 +7,7 @@ set hive.merge.mapfiles=false; set hive.merge.mapredfiles=false; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- There is no need to write the temporary results of the sub-queries, and then read them @@ -45,4 +46,4 @@ FROM ( desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_24.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_24.q b/ql/src/test/queries/clientpositive/union_remove_24.q index 3a1e225..7ed80d1 100644 --- a/ql/src/test/queries/clientpositive/union_remove_24.q +++ b/ql/src/test/queries/clientpositive/union_remove_24.q @@ -7,6 +7,7 @@ set hive.merge.mapfiles=false; set hive.merge.mapredfiles=false; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- There is no need to write the temporary results of the sub-queries, and then read them @@ -41,4 +42,4 @@ SELECT * FROM desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_25.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_25.q b/ql/src/test/queries/clientpositive/union_remove_25.q index d70adb9..78fa9aa 100644 --- a/ql/src/test/queries/clientpositive/union_remove_25.q +++ b/ql/src/test/queries/clientpositive/union_remove_25.q @@ -8,6 +8,7 @@ set hive.merge.mapfiles=false; set hive.merge.mapredfiles=false; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- There is no need to write the temporary results of the sub-queries, and then read them @@ -45,7 +46,7 @@ FROM ( desc formatted outputTbl1 partition(ds='2004'); set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; explain insert overwrite table outputTbl2 partition(ds) http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_26.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_26.q b/ql/src/test/queries/clientpositive/union_remove_26.q index d35d4e2..4e77d73 100644 --- a/ql/src/test/queries/clientpositive/union_remove_26.q +++ b/ql/src/test/queries/clientpositive/union_remove_26.q @@ -108,4 +108,4 @@ select count(*) from ( UNION ALL SELECT count(1) as rowcnt, min(val) as ms, max(val) as mx from inputTbl2 UNION ALL - SELECT count(1) as rowcnt, min(val) as ms, max(val) as mx from inputTbl3) t; \ No newline at end of file + SELECT count(1) as rowcnt, min(val) as ms, max(val) as mx from inputTbl3) t; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_3.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_3.q b/ql/src/test/queries/clientpositive/union_remove_3.q index 2e6e3e6..b0f63fc 100644 --- a/ql/src/test/queries/clientpositive/union_remove_3.q +++ b/ql/src/test/queries/clientpositive/union_remove_3.q @@ -7,6 +7,7 @@ set hive.merge.mapfiles=false; set hive.merge.mapredfiles=false; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->remove->filesink optimization -- Union of 3 subqueries is performed (all of which are map-only queries) -- followed by select star and a file sink. @@ -47,5 +48,5 @@ FROM ( desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_4.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_4.q b/ql/src/test/queries/clientpositive/union_remove_4.q index a08bcc1..9ee6282 100644 --- a/ql/src/test/queries/clientpositive/union_remove_4.q +++ b/ql/src/test/queries/clientpositive/union_remove_4.q @@ -8,6 +8,7 @@ set hive.merge.mapredfiles=true; set mapred.input.dir.recursive=true; set hive.merge.smallfiles.avgsize=1; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- There is no need to write the temporary results of the sub-queries, and then read them @@ -43,4 +44,4 @@ FROM ( desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_5.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_5.q b/ql/src/test/queries/clientpositive/union_remove_5.q index 4ac3edf..5d3c482 100644 --- a/ql/src/test/queries/clientpositive/union_remove_5.q +++ b/ql/src/test/queries/clientpositive/union_remove_5.q @@ -8,6 +8,7 @@ set hive.merge.mapredfiles=true; set hive.merge.smallfiles.avgsize=1; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 3 subqueries is performed (exactly one of which requires a map-reduce job) -- followed by select star and a file sink. @@ -49,4 +50,4 @@ FROM ( desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_6.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_6.q b/ql/src/test/queries/clientpositive/union_remove_6.q index f37f409..23eb760 100644 --- a/ql/src/test/queries/clientpositive/union_remove_6.q +++ b/ql/src/test/queries/clientpositive/union_remove_6.q @@ -7,6 +7,7 @@ set hive.merge.mapfiles=false; set hive.merge.mapredfiles=false; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (all of which are mapred queries) -- followed by select star and a file sink in 2 output tables. @@ -38,5 +39,5 @@ insert overwrite table outputTbl1 select * insert overwrite table outputTbl2 select *; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; -select * from outputTbl2 order by key, `values`;; +select * from outputTbl1; +select * from outputTbl2; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_6_subq.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_6_subq.q b/ql/src/test/queries/clientpositive/union_remove_6_subq.q index 6853728..3607a52 100644 --- a/ql/src/test/queries/clientpositive/union_remove_6_subq.q +++ b/ql/src/test/queries/clientpositive/union_remove_6_subq.q @@ -43,8 +43,8 @@ insert overwrite table outputTbl1 select * insert overwrite table outputTbl2 select *; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; -select * from outputTbl2 order by key, `values`; +select * from outputTbl1; +select * from outputTbl2; -- The following queries guarantee the correctness. explain http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_7.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_7.q b/ql/src/test/queries/clientpositive/union_remove_7.q index cd6a75a..5e0d14b 100644 --- a/ql/src/test/queries/clientpositive/union_remove_7.q +++ b/ql/src/test/queries/clientpositive/union_remove_7.q @@ -7,6 +7,7 @@ set hive.merge.mapfiles=false; set hive.merge.mapredfiles=false; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- There is no need to write the temporary results of the sub-queries, and then read them @@ -44,4 +45,4 @@ FROM ( desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_8.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_8.q b/ql/src/test/queries/clientpositive/union_remove_8.q index 93aabe6..1e049ba 100644 --- a/ql/src/test/queries/clientpositive/union_remove_8.q +++ b/ql/src/test/queries/clientpositive/union_remove_8.q @@ -7,6 +7,7 @@ set hive.merge.mapfiles=false; set hive.merge.mapredfiles=false; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 3 subqueries is performed (exactly one of which requires a map-reduce job) -- followed by select star and a file sink. @@ -49,4 +50,4 @@ FROM ( desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_remove_9.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_remove_9.q b/ql/src/test/queries/clientpositive/union_remove_9.q index 1bb8d6d..b819560 100644 --- a/ql/src/test/queries/clientpositive/union_remove_9.q +++ b/ql/src/test/queries/clientpositive/union_remove_9.q @@ -8,6 +8,7 @@ set hive.merge.mapredfiles=true; set hive.merge.smallfiles.avgsize=1; set mapred.input.dir.recursive=true; +-- SORT_QUERY_RESULTS -- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which contains a union and is map-only), -- and the other one is a map-reduce query followed by select star and a file sink. @@ -53,4 +54,4 @@ select * FROM ( desc formatted outputTbl1; set hive.input.format=org.apache.hadoop.hive.ql.io.HiveInputFormat; -select * from outputTbl1 order by key, `values`; +select * from outputTbl1; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_script.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_script.q b/ql/src/test/queries/clientpositive/union_script.q index 7687104..e5b160f 100644 --- a/ql/src/test/queries/clientpositive/union_script.q +++ b/ql/src/test/queries/clientpositive/union_script.q @@ -1,8 +1,9 @@ set hive.mapred.mode=nonstrict; +-- SORT_QUERY_RESULTS select * from ( - select transform(key) using 'cat' as cola from src)s order by cola; + select transform(key) using 'cat' as cola from src)s; select * from ( select transform(key) using 'cat' as cola from src union all - select transform(key) using 'cat' as cola from src) s order by cola; + select transform(key) using 'cat' as cola from src) s; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_type_chk.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_type_chk.q b/ql/src/test/queries/clientpositive/union_type_chk.q index a25aeda..ff2e7cf 100644 --- a/ql/src/test/queries/clientpositive/union_type_chk.q +++ b/ql/src/test/queries/clientpositive/union_type_chk.q @@ -1,7 +1,7 @@ set hive.mapred.mode=nonstrict; set hive.cbo.enable=false; -set hive.mapred.mode=nonstrict; -select (x/sum(x) over()) as y from(select cast(1 as decimal(10,0)) as x from (select * from src limit 2)s1 union all select cast(1 as decimal(10,0)) x from (select * from src limit 2) s2 union all select '100000000' x from (select * from src limit 2) s3)u order by y; +-- SORT_QUERY_RESULTS +select (x/sum(x) over()) as y from(select cast(1 as decimal(10,0)) as x from (select * from src limit 2)s1 union all select cast(1 as decimal(10,0)) x from (select * from src limit 2) s2 union all select '100000000' x from (select * from src limit 2) s3)u; -select (x/sum(x) over()) as y from(select cast(1 as decimal(10,0)) as x from (select * from src limit 2)s1 union all select cast(1 as decimal(10,0)) x from (select * from src limit 2) s2 union all select cast (null as string) x from (select * from src limit 2) s3)u order by y; +select (x/sum(x) over()) as y from(select cast(1 as decimal(10,0)) as x from (select * from src limit 2)s1 union all select cast(1 as decimal(10,0)) x from (select * from src limit 2) s2 union all select cast (null as string) x from (select * from src limit 2) s3)u; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/queries/clientpositive/union_view.q ---------------------------------------------------------------------- diff --git a/ql/src/test/queries/clientpositive/union_view.q b/ql/src/test/queries/clientpositive/union_view.q index fa39450..c4f63e6 100644 --- a/ql/src/test/queries/clientpositive/union_view.q +++ b/ql/src/test/queries/clientpositive/union_view.q @@ -63,12 +63,13 @@ SELECT key, value, ds FROM src_union_3 EXPLAIN SELECT key, value, ds FROM src_union_view WHERE key=86 AND ds ='1'; EXPLAIN SELECT key, value, ds FROM src_union_view WHERE key=86 AND ds ='2'; EXPLAIN SELECT key, value, ds FROM src_union_view WHERE key=86 AND ds ='3'; -EXPLAIN SELECT key, value, ds FROM src_union_view WHERE key=86 AND ds IS NOT NULL order by ds; +EXPLAIN SELECT key, value, ds FROM src_union_view WHERE key=86 AND ds IS NOT NULL; SELECT key, value, ds FROM src_union_view WHERE key=86 AND ds ='1'; SELECT key, value, ds FROM src_union_view WHERE key=86 AND ds ='2'; SELECT key, value, ds FROM src_union_view WHERE key=86 AND ds ='3'; -SELECT key, value, ds FROM src_union_view WHERE key=86 AND ds IS NOT NULL order by ds; +-- SORT_BEFORE_DIFF +SELECT key, value, ds FROM src_union_view WHERE key=86 AND ds IS NOT NULL; EXPLAIN SELECT count(1) from src_union_view WHERE ds ='1'; EXPLAIN SELECT count(1) from src_union_view WHERE ds ='2'; http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/results/clientpositive/spark/union32.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/union32.q.out b/ql/src/test/results/clientpositive/spark/union32.q.out index 09fb01e..16cb243 100644 --- a/ql/src/test/results/clientpositive/spark/union32.q.out +++ b/ql/src/test/results/clientpositive/spark/union32.q.out @@ -38,7 +38,6 @@ SELECT * FROM (SELECT CAST(key AS DOUBLE) AS key FROM t1 UNION ALL SELECT CAST(key AS BIGINT) AS key FROM t2) a -ORDER BY key PREHOOK: type: QUERY POSTHOOK: query: -- Test simple union with double EXPLAIN @@ -46,7 +45,6 @@ SELECT * FROM (SELECT CAST(key AS DOUBLE) AS key FROM t1 UNION ALL SELECT CAST(key AS BIGINT) AS key FROM t2) a -ORDER BY key POSTHOOK: type: QUERY STAGE DEPENDENCIES: Stage-1 is a root stage @@ -55,8 +53,6 @@ STAGE DEPENDENCIES: STAGE PLANS: Stage: Stage-1 Spark - Edges: - Reducer 2 <- Map 1 (SORT, 1), Map 3 (SORT, 1) #### A masked pattern was here #### Vertices: Map 1 @@ -68,11 +64,14 @@ STAGE PLANS: expressions: UDFToDouble(key) (type: double) outputColumnNames: _col0 Statistics: Num rows: 10 Data size: 70 Basic stats: COMPLETE Column stats: NONE - Reduce Output Operator - key expressions: _col0 (type: double) - sort order: + + File Output Operator + compressed: false Statistics: Num rows: 20 Data size: 140 Basic stats: COMPLETE Column stats: NONE - Map 3 + table: + input format: org.apache.hadoop.mapred.SequenceFileInputFormat + output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat + serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe + Map 2 Map Operator Tree: TableScan alias: t2 @@ -81,23 +80,13 @@ STAGE PLANS: expressions: UDFToDouble(UDFToLong(key)) (type: double) outputColumnNames: _col0 Statistics: Num rows: 10 Data size: 70 Basic stats: COMPLETE Column stats: NONE - Reduce Output Operator - key expressions: _col0 (type: double) - sort order: + + File Output Operator + compressed: false Statistics: Num rows: 20 Data size: 140 Basic stats: COMPLETE Column stats: NONE - Reducer 2 - Reduce Operator Tree: - Select Operator - expressions: KEY.reducesinkkey0 (type: double) - outputColumnNames: _col0 - Statistics: Num rows: 20 Data size: 140 Basic stats: COMPLETE Column stats: NONE - File Output Operator - compressed: false - Statistics: Num rows: 20 Data size: 140 Basic stats: COMPLETE Column stats: NONE - table: - input format: org.apache.hadoop.mapred.SequenceFileInputFormat - output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat - serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe + table: + input format: org.apache.hadoop.mapred.SequenceFileInputFormat + output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat + serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe Stage: Stage-0 Fetch Operator http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/results/clientpositive/spark/union34.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/union34.q.out b/ql/src/test/results/clientpositive/spark/union34.q.out index 1a50c3c..c49c436 100644 --- a/ql/src/test/results/clientpositive/spark/union34.q.out +++ b/ql/src/test/results/clientpositive/spark/union34.q.out @@ -1,8 +1,10 @@ -PREHOOK: query: create table src10_1 (key string, value string) +PREHOOK: query: -- SORT_QUERY_RESULTS +create table src10_1 (key string, value string) PREHOOK: type: CREATETABLE PREHOOK: Output: database:default PREHOOK: Output: default@src10_1 -POSTHOOK: query: create table src10_1 (key string, value string) +POSTHOOK: query: -- SORT_QUERY_RESULTS +create table src10_1 (key string, value string) POSTHOOK: type: CREATETABLE POSTHOOK: Output: database:default POSTHOOK: Output: default@src10_1 @@ -67,7 +69,7 @@ SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION ALL SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION ALL SELECT * FROM src10_4 ) alias0 -) alias1 order by key +) alias1 PREHOOK: type: QUERY POSTHOOK: query: -- When we convert the Join of sub1 and sub0 into a MapJoin, -- we can use a single MR job to evaluate this entire query. @@ -76,7 +78,7 @@ SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION ALL SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION ALL SELECT * FROM src10_4 ) alias0 -) alias1 order by key +) alias1 POSTHOOK: type: QUERY STAGE DEPENDENCIES: Stage-2 is a root stage @@ -88,7 +90,7 @@ STAGE PLANS: Spark #### A masked pattern was here #### Vertices: - Map 3 + Map 2 Map Operator Tree: TableScan alias: src10_2 @@ -109,8 +111,6 @@ STAGE PLANS: Stage: Stage-1 Spark - Edges: - Reducer 2 <- Map 1 (SORT, 1), Map 4 (SORT, 1), Map 5 (SORT, 1) #### A masked pattern was here #### Vertices: Map 1 @@ -133,16 +133,18 @@ STAGE PLANS: 1 _col0 (type: string) outputColumnNames: _col0, _col1 input vertices: - 1 Map 3 + 1 Map 2 Statistics: Num rows: 11 Data size: 114 Basic stats: COMPLETE Column stats: NONE - Reduce Output Operator - key expressions: _col0 (type: string) - sort order: + + File Output Operator + compressed: false Statistics: Num rows: 31 Data size: 322 Basic stats: COMPLETE Column stats: NONE - value expressions: _col1 (type: string) + table: + input format: org.apache.hadoop.mapred.SequenceFileInputFormat + output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat + serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe Local Work: Map Reduce Local Work - Map 4 + Map 3 Map Operator Tree: TableScan alias: src10_3 @@ -151,12 +153,14 @@ STAGE PLANS: expressions: key (type: string), value (type: string) outputColumnNames: _col0, _col1 Statistics: Num rows: 10 Data size: 104 Basic stats: COMPLETE Column stats: NONE - Reduce Output Operator - key expressions: _col0 (type: string) - sort order: + + File Output Operator + compressed: false Statistics: Num rows: 31 Data size: 322 Basic stats: COMPLETE Column stats: NONE - value expressions: _col1 (type: string) - Map 5 + table: + input format: org.apache.hadoop.mapred.SequenceFileInputFormat + output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat + serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe + Map 4 Map Operator Tree: TableScan alias: src10_4 @@ -165,24 +169,13 @@ STAGE PLANS: expressions: key (type: string), value (type: string) outputColumnNames: _col0, _col1 Statistics: Num rows: 10 Data size: 104 Basic stats: COMPLETE Column stats: NONE - Reduce Output Operator - key expressions: _col0 (type: string) - sort order: + + File Output Operator + compressed: false Statistics: Num rows: 31 Data size: 322 Basic stats: COMPLETE Column stats: NONE - value expressions: _col1 (type: string) - Reducer 2 - Reduce Operator Tree: - Select Operator - expressions: KEY.reducesinkkey0 (type: string), VALUE._col0 (type: string) - outputColumnNames: _col0, _col1 - Statistics: Num rows: 31 Data size: 322 Basic stats: COMPLETE Column stats: NONE - File Output Operator - compressed: false - Statistics: Num rows: 31 Data size: 322 Basic stats: COMPLETE Column stats: NONE - table: - input format: org.apache.hadoop.mapred.SequenceFileInputFormat - output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat - serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe + table: + input format: org.apache.hadoop.mapred.SequenceFileInputFormat + output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat + serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe Stage: Stage-0 Fetch Operator @@ -194,7 +187,7 @@ PREHOOK: query: SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION ALL SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION ALL SELECT * FROM src10_4 ) alias0 -) alias1 order by key +) alias1 PREHOOK: type: QUERY PREHOOK: Input: default@src10_1 PREHOOK: Input: default@src10_2 @@ -205,7 +198,7 @@ POSTHOOK: query: SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION ALL SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION ALL SELECT * FROM src10_4 ) alias0 -) alias1 order by key +) alias1 POSTHOOK: type: QUERY POSTHOOK: Input: default@src10_1 POSTHOOK: Input: default@src10_2 @@ -251,7 +244,7 @@ SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION ALL SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION ALL SELECT * FROM src10_4 ) alias0 -) alias1 order by key +) alias1 PREHOOK: type: QUERY POSTHOOK: query: -- When we do not convert the Join of sub1 and sub0 into a MapJoin, -- we need to use two MR jobs to evaluate this query. @@ -262,7 +255,7 @@ SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION ALL SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION ALL SELECT * FROM src10_4 ) alias0 -) alias1 order by key +) alias1 POSTHOOK: type: QUERY STAGE DEPENDENCIES: Stage-1 is a root stage @@ -272,8 +265,7 @@ STAGE PLANS: Stage: Stage-1 Spark Edges: - Reducer 2 <- Map 1 (PARTITION-LEVEL SORT, 2), Map 4 (PARTITION-LEVEL SORT, 2) - Reducer 3 <- Map 5 (SORT, 1), Map 6 (SORT, 1), Reducer 2 (SORT, 1) + Reducer 2 <- Map 1 (PARTITION-LEVEL SORT, 2), Map 3 (PARTITION-LEVEL SORT, 2) #### A masked pattern was here #### Vertices: Map 1 @@ -294,7 +286,7 @@ STAGE PLANS: Map-reduce partition columns: _col0 (type: string) Statistics: Num rows: 10 Data size: 104 Basic stats: COMPLETE Column stats: NONE value expressions: _col1 (type: string) - Map 4 + Map 3 Map Operator Tree: TableScan alias: src10_2 @@ -311,7 +303,7 @@ STAGE PLANS: sort order: + Map-reduce partition columns: _col0 (type: string) Statistics: Num rows: 10 Data size: 104 Basic stats: COMPLETE Column stats: NONE - Map 5 + Map 4 Map Operator Tree: TableScan alias: src10_3 @@ -320,12 +312,14 @@ STAGE PLANS: expressions: key (type: string), value (type: string) outputColumnNames: _col0, _col1 Statistics: Num rows: 10 Data size: 104 Basic stats: COMPLETE Column stats: NONE - Reduce Output Operator - key expressions: _col0 (type: string) - sort order: + + File Output Operator + compressed: false Statistics: Num rows: 31 Data size: 322 Basic stats: COMPLETE Column stats: NONE - value expressions: _col1 (type: string) - Map 6 + table: + input format: org.apache.hadoop.mapred.SequenceFileInputFormat + output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat + serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe + Map 5 Map Operator Tree: TableScan alias: src10_4 @@ -334,11 +328,13 @@ STAGE PLANS: expressions: key (type: string), value (type: string) outputColumnNames: _col0, _col1 Statistics: Num rows: 10 Data size: 104 Basic stats: COMPLETE Column stats: NONE - Reduce Output Operator - key expressions: _col0 (type: string) - sort order: + + File Output Operator + compressed: false Statistics: Num rows: 31 Data size: 322 Basic stats: COMPLETE Column stats: NONE - value expressions: _col1 (type: string) + table: + input format: org.apache.hadoop.mapred.SequenceFileInputFormat + output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat + serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe Reducer 2 Reduce Operator Tree: Join Operator @@ -346,17 +342,6 @@ STAGE PLANS: Inner Join 0 to 1 outputColumnNames: _col0, _col1 Statistics: Num rows: 11 Data size: 114 Basic stats: COMPLETE Column stats: NONE - Reduce Output Operator - key expressions: _col0 (type: string) - sort order: + - Statistics: Num rows: 31 Data size: 322 Basic stats: COMPLETE Column stats: NONE - value expressions: _col1 (type: string) - Reducer 3 - Reduce Operator Tree: - Select Operator - expressions: KEY.reducesinkkey0 (type: string), VALUE._col0 (type: string) - outputColumnNames: _col0, _col1 - Statistics: Num rows: 31 Data size: 322 Basic stats: COMPLETE Column stats: NONE File Output Operator compressed: false Statistics: Num rows: 31 Data size: 322 Basic stats: COMPLETE Column stats: NONE @@ -375,7 +360,7 @@ PREHOOK: query: SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION ALL SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION ALL SELECT * FROM src10_4 ) alias0 -) alias1 order by key +) alias1 PREHOOK: type: QUERY PREHOOK: Input: default@src10_1 PREHOOK: Input: default@src10_2 @@ -386,7 +371,7 @@ POSTHOOK: query: SELECT * FROM ( SELECT sub1.key,sub1.value FROM (SELECT * FROM src10_1) sub1 JOIN (SELECT * FROM src10_2) sub0 ON (sub0.key = sub1.key) UNION ALL SELECT key,value FROM (SELECT * FROM (SELECT * FROM src10_3) sub2 UNION ALL SELECT * FROM src10_4 ) alias0 -) alias1 order by key +) alias1 POSTHOOK: type: QUERY POSTHOOK: Input: default@src10_1 POSTHOOK: Input: default@src10_2 http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/results/clientpositive/spark/union_remove_1.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/union_remove_1.q.out b/ql/src/test/results/clientpositive/spark/union_remove_1.q.out index 6d88345..ff18e4a 100644 --- a/ql/src/test/results/clientpositive/spark/union_remove_1.q.out +++ b/ql/src/test/results/clientpositive/spark/union_remove_1.q.out @@ -1,4 +1,5 @@ -PREHOOK: query: -- This is to test the union->selectstar->filesink optimization +PREHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- There is no need to write the temporary results of the sub-queries, and then read them -- again to process the union. The union can be removed completely. @@ -12,7 +13,8 @@ create table inputTbl1(key string, val string) stored as textfile PREHOOK: type: CREATETABLE PREHOOK: Output: database:default PREHOOK: Output: default@inputTbl1 -POSTHOOK: query: -- This is to test the union->selectstar->filesink optimization +POSTHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- There is no need to write the temporary results of the sub-queries, and then read them -- again to process the union. The union can be removed completely. @@ -190,11 +192,11 @@ Bucket Columns: [] Sort Columns: [] Storage Desc Params: serialization.format 1 -PREHOOK: query: select * from outputTbl1 order by key, `values` +PREHOOK: query: select * from outputTbl1 PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 order by key, `values` +POSTHOOK: query: select * from outputTbl1 POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 #### A masked pattern was here #### http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/results/clientpositive/spark/union_remove_10.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/union_remove_10.q.out b/ql/src/test/results/clientpositive/spark/union_remove_10.q.out index 7b4b0c4..ded8b40 100644 --- a/ql/src/test/results/clientpositive/spark/union_remove_10.q.out +++ b/ql/src/test/results/clientpositive/spark/union_remove_10.q.out @@ -1,4 +1,5 @@ -PREHOOK: query: -- This is to test the union->selectstar->filesink optimization +PREHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a map-only query, and the -- other one contains a nested union where one of the sub-queries requires a map-reduce -- job), followed by select star and a file sink. @@ -16,7 +17,8 @@ create table inputTbl1(key string, val string) stored as textfile PREHOOK: type: CREATETABLE PREHOOK: Output: database:default PREHOOK: Output: default@inputTbl1 -POSTHOOK: query: -- This is to test the union->selectstar->filesink optimization +POSTHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a map-only query, and the -- other one contains a nested union where one of the sub-queries requires a map-reduce -- job), followed by select star and a file sink. @@ -273,11 +275,11 @@ Bucket Columns: [] Sort Columns: [] Storage Desc Params: serialization.format 1 -PREHOOK: query: select * from outputTbl1 order by key, `values` +PREHOOK: query: select * from outputTbl1 PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 order by key, `values` +POSTHOOK: query: select * from outputTbl1 POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 #### A masked pattern was here #### http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/results/clientpositive/spark/union_remove_11.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/union_remove_11.q.out b/ql/src/test/results/clientpositive/spark/union_remove_11.q.out index 5d77d06..0fe1340 100644 --- a/ql/src/test/results/clientpositive/spark/union_remove_11.q.out +++ b/ql/src/test/results/clientpositive/spark/union_remove_11.q.out @@ -1,4 +1,5 @@ -PREHOOK: query: -- This is to test the union->selectstar->filesink optimization +PREHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a map-only query, and the -- other one contains a nested union where also contains map only sub-queries), -- followed by select star and a file sink. @@ -16,7 +17,8 @@ create table inputTbl1(key string, val string) stored as textfile PREHOOK: type: CREATETABLE PREHOOK: Output: database:default PREHOOK: Output: default@inputTbl1 -POSTHOOK: query: -- This is to test the union->selectstar->filesink optimization +POSTHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a map-only query, and the -- other one contains a nested union where also contains map only sub-queries), -- followed by select star and a file sink. @@ -263,11 +265,11 @@ Bucket Columns: [] Sort Columns: [] Storage Desc Params: serialization.format 1 -PREHOOK: query: select * from outputTbl1 order by key, `values` +PREHOOK: query: select * from outputTbl1 PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 order by key, `values` +POSTHOOK: query: select * from outputTbl1 POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 #### A masked pattern was here #### http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/results/clientpositive/spark/union_remove_12.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/union_remove_12.q.out b/ql/src/test/results/clientpositive/spark/union_remove_12.q.out index 6a7fe69..7487cd3 100644 --- a/ql/src/test/results/clientpositive/spark/union_remove_12.q.out +++ b/ql/src/test/results/clientpositive/spark/union_remove_12.q.out @@ -1,4 +1,5 @@ -PREHOOK: query: -- This is to test the union->selectstar->filesink optimization +PREHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a map-only query, and the -- other one is a map-join query), followed by select star and a file sink. -- The union optimization is applied, and the union is removed. @@ -15,7 +16,8 @@ create table inputTbl1(key string, val string) stored as textfile PREHOOK: type: CREATETABLE PREHOOK: Output: database:default PREHOOK: Output: default@inputTbl1 -POSTHOOK: query: -- This is to test the union->selectstar->filesink optimization +POSTHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a map-only query, and the -- other one is a map-join query), followed by select star and a file sink. -- The union optimization is applied, and the union is removed. @@ -271,11 +273,11 @@ Bucket Columns: [] Sort Columns: [] Storage Desc Params: serialization.format 1 -PREHOOK: query: select * from outputTbl1 order by key, `values` +PREHOOK: query: select * from outputTbl1 PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 order by key, `values` +POSTHOOK: query: select * from outputTbl1 POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 #### A masked pattern was here #### http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/results/clientpositive/spark/union_remove_13.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/union_remove_13.q.out b/ql/src/test/results/clientpositive/spark/union_remove_13.q.out index 2d031f9..550c24b 100644 --- a/ql/src/test/results/clientpositive/spark/union_remove_13.q.out +++ b/ql/src/test/results/clientpositive/spark/union_remove_13.q.out @@ -1,4 +1,5 @@ -PREHOOK: query: -- This is to test the union->selectstar->filesink optimization +PREHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a mapred query, and the -- other one is a map-join query), followed by select star and a file sink. -- The union selectstar optimization should be performed, and the union should be removed. @@ -15,7 +16,8 @@ create table inputTbl1(key string, val string) stored as textfile PREHOOK: type: CREATETABLE PREHOOK: Output: database:default PREHOOK: Output: default@inputTbl1 -POSTHOOK: query: -- This is to test the union->selectstar->filesink optimization +POSTHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a mapred query, and the -- other one is a map-join query), followed by select star and a file sink. -- The union selectstar optimization should be performed, and the union should be removed. @@ -297,11 +299,11 @@ Bucket Columns: [] Sort Columns: [] Storage Desc Params: serialization.format 1 -PREHOOK: query: select * from outputTbl1 order by key, `values` +PREHOOK: query: select * from outputTbl1 PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 order by key, `values` +POSTHOOK: query: select * from outputTbl1 POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 #### A masked pattern was here #### @@ -313,8 +315,8 @@ POSTHOOK: Input: default@outputtbl1 3 13 7 1 7 17 -8 2 8 18 8 18 +8 2 8 28 8 28 http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/results/clientpositive/spark/union_remove_14.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/union_remove_14.q.out b/ql/src/test/results/clientpositive/spark/union_remove_14.q.out index 0be3d86..9002223 100644 --- a/ql/src/test/results/clientpositive/spark/union_remove_14.q.out +++ b/ql/src/test/results/clientpositive/spark/union_remove_14.q.out @@ -1,4 +1,5 @@ -PREHOOK: query: -- This is to test the union->selectstar->filesink optimization +PREHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a map-only query, and the -- other one contains a join, which should be performed as a map-join query at runtime), -- followed by select star and a file sink. @@ -16,7 +17,8 @@ create table inputTbl1(key string, val string) stored as textfile PREHOOK: type: CREATETABLE PREHOOK: Output: database:default PREHOOK: Output: default@inputTbl1 -POSTHOOK: query: -- This is to test the union->selectstar->filesink optimization +POSTHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 subqueries is performed (one of which is a map-only query, and the -- other one contains a join, which should be performed as a map-join query at runtime), -- followed by select star and a file sink. @@ -273,11 +275,11 @@ Bucket Columns: [] Sort Columns: [] Storage Desc Params: serialization.format 1 -PREHOOK: query: select * from outputTbl1 order by key, `values` +PREHOOK: query: select * from outputTbl1 PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 order by key, `values` +POSTHOOK: query: select * from outputTbl1 POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 #### A masked pattern was here #### http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/results/clientpositive/spark/union_remove_15.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/union_remove_15.q.out b/ql/src/test/results/clientpositive/spark/union_remove_15.q.out index c0ad4e6..c4d1542 100644 --- a/ql/src/test/results/clientpositive/spark/union_remove_15.q.out +++ b/ql/src/test/results/clientpositive/spark/union_remove_15.q.out @@ -1,4 +1,5 @@ -PREHOOK: query: -- This is to test the union->selectstar->filesink optimization +PREHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- and the results are written to a table using dynamic partitions. -- There is no need to write the temporary results of the sub-queries, and then read them @@ -15,7 +16,8 @@ create table inputTbl1(key string, val string) stored as textfile PREHOOK: type: CREATETABLE PREHOOK: Output: database:default PREHOOK: Output: default@inputTbl1 -POSTHOOK: query: -- This is to test the union->selectstar->filesink optimization +POSTHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- and the results are written to a table using dynamic partitions. -- There is no need to write the temporary results of the sub-queries, and then read them @@ -220,12 +222,12 @@ POSTHOOK: type: SHOWPARTITIONS POSTHOOK: Input: default@outputtbl1 ds=1 ds=2 -PREHOOK: query: select * from outputTbl1 where ds = '1' order by key, `values` +PREHOOK: query: select * from outputTbl1 where ds = '1' PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 PREHOOK: Input: default@outputtbl1@ds=1 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 where ds = '1' order by key, `values` +POSTHOOK: query: select * from outputTbl1 where ds = '1' POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 POSTHOOK: Input: default@outputtbl1@ds=1 @@ -235,12 +237,12 @@ POSTHOOK: Input: default@outputtbl1@ds=1 3 1 1 7 1 1 8 2 1 -PREHOOK: query: select * from outputTbl1 where ds = '2' order by key, `values` +PREHOOK: query: select * from outputTbl1 where ds = '2' PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 PREHOOK: Input: default@outputtbl1@ds=2 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 where ds = '2' order by key, `values` +POSTHOOK: query: select * from outputTbl1 where ds = '2' POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 POSTHOOK: Input: default@outputtbl1@ds=2 http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/results/clientpositive/spark/union_remove_16.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/union_remove_16.q.out b/ql/src/test/results/clientpositive/spark/union_remove_16.q.out index d02109a..961291b 100644 --- a/ql/src/test/results/clientpositive/spark/union_remove_16.q.out +++ b/ql/src/test/results/clientpositive/spark/union_remove_16.q.out @@ -1,4 +1,5 @@ -PREHOOK: query: -- This is to test the union->selectstar->filesink optimization +PREHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- and the results are written to a table using dynamic partitions. -- There is no need to write the temporary results of the sub-queries, and then read them @@ -14,7 +15,8 @@ create table inputTbl1(key string, val string) stored as textfile PREHOOK: type: CREATETABLE PREHOOK: Output: database:default PREHOOK: Output: default@inputTbl1 -POSTHOOK: query: -- This is to test the union->selectstar->filesink optimization +POSTHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- and the results are written to a table using dynamic partitions. -- There is no need to write the temporary results of the sub-queries, and then read them @@ -260,12 +262,12 @@ POSTHOOK: type: SHOWPARTITIONS POSTHOOK: Input: default@outputtbl1 ds=1 ds=2 -PREHOOK: query: select * from outputTbl1 where ds = '1' order by key, `values` +PREHOOK: query: select * from outputTbl1 where ds = '1' PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 PREHOOK: Input: default@outputtbl1@ds=1 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 where ds = '1' order by key, `values` +POSTHOOK: query: select * from outputTbl1 where ds = '1' POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 POSTHOOK: Input: default@outputtbl1@ds=1 @@ -275,12 +277,12 @@ POSTHOOK: Input: default@outputtbl1@ds=1 3 1 1 7 1 1 8 2 1 -PREHOOK: query: select * from outputTbl1 where ds = '2' order by key, `values` +PREHOOK: query: select * from outputTbl1 where ds = '2' PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 PREHOOK: Input: default@outputtbl1@ds=2 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 where ds = '2' order by key, `values` +POSTHOOK: query: select * from outputTbl1 where ds = '2' POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 POSTHOOK: Input: default@outputtbl1@ds=2 http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/results/clientpositive/spark/union_remove_17.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/union_remove_17.q.out b/ql/src/test/results/clientpositive/spark/union_remove_17.q.out index c8047a9..baed263 100644 --- a/ql/src/test/results/clientpositive/spark/union_remove_17.q.out +++ b/ql/src/test/results/clientpositive/spark/union_remove_17.q.out @@ -1,4 +1,5 @@ -PREHOOK: query: -- This is to test the union->selectstar->filesink optimization +PREHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- and the results are written to a table using dynamic partitions. -- There is no need for this optimization, since the query is a map-only query. @@ -12,7 +13,8 @@ create table inputTbl1(key string, val string) stored as textfile PREHOOK: type: CREATETABLE PREHOOK: Output: database:default PREHOOK: Output: default@inputTbl1 -POSTHOOK: query: -- This is to test the union->selectstar->filesink optimization +POSTHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- and the results are written to a table using dynamic partitions. -- There is no need for this optimization, since the query is a map-only query. @@ -192,12 +194,12 @@ POSTHOOK: type: SHOWPARTITIONS POSTHOOK: Input: default@outputtbl1 ds=1 ds=2 -PREHOOK: query: select * from outputTbl1 where ds = '1' order by key, `values` +PREHOOK: query: select * from outputTbl1 where ds = '1' PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 PREHOOK: Input: default@outputtbl1@ds=1 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 where ds = '1' order by key, `values` +POSTHOOK: query: select * from outputTbl1 where ds = '1' POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 POSTHOOK: Input: default@outputtbl1@ds=1 @@ -208,12 +210,12 @@ POSTHOOK: Input: default@outputtbl1@ds=1 7 1 1 8 1 1 8 1 1 -PREHOOK: query: select * from outputTbl1 where ds = '2' order by key, `values` +PREHOOK: query: select * from outputTbl1 where ds = '2' PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 PREHOOK: Input: default@outputtbl1@ds=2 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 where ds = '2' order by key, `values` +POSTHOOK: query: select * from outputTbl1 where ds = '2' POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 POSTHOOK: Input: default@outputtbl1@ds=2 http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/results/clientpositive/spark/union_remove_18.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/union_remove_18.q.out b/ql/src/test/results/clientpositive/spark/union_remove_18.q.out index 6a6de92..535d085 100644 --- a/ql/src/test/results/clientpositive/spark/union_remove_18.q.out +++ b/ql/src/test/results/clientpositive/spark/union_remove_18.q.out @@ -1,4 +1,5 @@ -PREHOOK: query: -- This is to test the union->selectstar->filesink optimization +PREHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- There is no need to write the temporary results of the sub-queries, and then read them -- again to process the union. The union can be removed completely. @@ -14,7 +15,8 @@ create table inputTbl1(key string, ds string) stored as textfile PREHOOK: type: CREATETABLE PREHOOK: Output: database:default PREHOOK: Output: default@inputTbl1 -POSTHOOK: query: -- This is to test the union->selectstar->filesink optimization +POSTHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- There is no need to write the temporary results of the sub-queries, and then read them -- again to process the union. The union can be removed completely. @@ -234,31 +236,31 @@ ds=13 ds=17 ds=18 ds=28 -PREHOOK: query: select * from outputTbl1 where ds = '11' order by key, `values` +PREHOOK: query: select * from outputTbl1 where ds = '11' PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 PREHOOK: Input: default@outputtbl1@ds=11 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 where ds = '11' order by key, `values` +POSTHOOK: query: select * from outputTbl1 where ds = '11' POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 POSTHOOK: Input: default@outputtbl1@ds=11 #### A masked pattern was here #### 1 1 11 1 1 11 -PREHOOK: query: select * from outputTbl1 where ds = '18' order by key, `values` +PREHOOK: query: select * from outputTbl1 where ds = '18' PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 PREHOOK: Input: default@outputtbl1@ds=18 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 where ds = '18' order by key, `values` +POSTHOOK: query: select * from outputTbl1 where ds = '18' POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 POSTHOOK: Input: default@outputtbl1@ds=18 #### A masked pattern was here #### 8 1 18 8 1 18 -PREHOOK: query: select * from outputTbl1 where ds is not null order by key, `values`, ds +PREHOOK: query: select * from outputTbl1 where ds is not null PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 PREHOOK: Input: default@outputtbl1@ds=11 @@ -268,7 +270,7 @@ PREHOOK: Input: default@outputtbl1@ds=17 PREHOOK: Input: default@outputtbl1@ds=18 PREHOOK: Input: default@outputtbl1@ds=28 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 where ds is not null order by key, `values`, ds +POSTHOOK: query: select * from outputTbl1 where ds is not null POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 POSTHOOK: Input: default@outputtbl1@ds=11 http://git-wip-us.apache.org/repos/asf/hive/blob/d14673a0/ql/src/test/results/clientpositive/spark/union_remove_19.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/union_remove_19.q.out b/ql/src/test/results/clientpositive/spark/union_remove_19.q.out index 59063a7..8c06c64 100644 --- a/ql/src/test/results/clientpositive/spark/union_remove_19.q.out +++ b/ql/src/test/results/clientpositive/spark/union_remove_19.q.out @@ -1,4 +1,5 @@ -PREHOOK: query: -- This is to test the union->selectstar->filesink optimization +PREHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- There is no need to write the temporary results of the sub-queries, and then read them -- again to process the union. The union can be removed completely. @@ -14,7 +15,8 @@ create table inputTbl1(key string, val string) stored as textfile PREHOOK: type: CREATETABLE PREHOOK: Output: database:default PREHOOK: Output: default@inputTbl1 -POSTHOOK: query: -- This is to test the union->selectstar->filesink optimization +POSTHOOK: query: -- SORT_QUERY_RESULTS +-- This is to test the union->selectstar->filesink optimization -- Union of 2 map-reduce subqueries is performed followed by select star and a file sink -- There is no need to write the temporary results of the sub-queries, and then read them -- again to process the union. The union can be removed completely. @@ -484,11 +486,11 @@ POSTHOOK: Input: default@inputtbl1 POSTHOOK: Output: default@outputtbl1 POSTHOOK: Lineage: outputtbl1.key EXPRESSION [(inputtbl1)inputtbl1.FieldSchema(name:key, type:string, comment:null), ] POSTHOOK: Lineage: outputtbl1.values EXPRESSION [(inputtbl1)inputtbl1.null, ] -PREHOOK: query: select * from outputTbl1 order by key, `values` +PREHOOK: query: select * from outputTbl1 PREHOOK: type: QUERY PREHOOK: Input: default@outputtbl1 #### A masked pattern was here #### -POSTHOOK: query: select * from outputTbl1 order by key, `values` +POSTHOOK: query: select * from outputTbl1 POSTHOOK: type: QUERY POSTHOOK: Input: default@outputtbl1 #### A masked pattern was here ####
