This is an automated email from the ASF dual-hosted git repository.
zhangbutao pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/hive.git
The following commit(s) were added to refs/heads/master by this push:
new 81b394104b0 HIVE-23558: Remove compute_stats UDAF (#4928)(Butao Zhang,
reviewed by Ayush Saxena)
81b394104b0 is described below
commit 81b394104b08efdc42d782d444462f4fdc12a8fd
Author: Butao Zhang <[email protected]>
AuthorDate: Sun Dec 24 21:23:03 2023 +0800
HIVE-23558: Remove compute_stats UDAF (#4928)(Butao Zhang, reviewed by
Ayush Saxena)
---
data/files/datasets/temp_udf/load.hive.sql | 1 +
.../hadoop/hive/ql/exec/FunctionRegistry.java | 1 -
.../queries/clientnegative/compute_stats_long.q | 7 -
ql/src/test/queries/clientpositive/char_udf1.q | 1 +
.../queries/clientpositive/compute_stats_binary.q | 9 --
.../queries/clientpositive/compute_stats_boolean.q | 9 --
.../queries/clientpositive/compute_stats_date.q | 28 ----
.../queries/clientpositive/compute_stats_decimal.q | 11 --
.../queries/clientpositive/compute_stats_double.q | 9 --
.../clientpositive/compute_stats_empty_table.q | 12 --
.../queries/clientpositive/compute_stats_long.q | 9 --
.../queries/clientpositive/compute_stats_string.q | 9 --
.../reduceSinkDeDuplication_pRS_key_empty.q | 1 +
ql/src/test/queries/clientpositive/varchar_udf1.q | 1 +
ql/src/test/queries/clientpositive/vector_udf1.q | 1 +
.../clientnegative/compute_stats_long.q.out | 57 -------
.../clientpositive/llap/compute_stats_binary.q.out | 34 ----
.../llap/compute_stats_boolean.q.out | 34 ----
.../clientpositive/llap/compute_stats_date.q.out | 176 ---------------------
.../llap/compute_stats_decimal.q.out | Bin 1512 -> 0 bytes
.../clientpositive/llap/compute_stats_double.q.out | Bin 1443 -> 0 bytes
.../llap/compute_stats_empty_table.q.out | 62 --------
.../clientpositive/llap/compute_stats_long.q.out | Bin 1376 -> 0 bytes
.../clientpositive/llap/compute_stats_string.q.out | Bin 1448 -> 0 bytes
.../clientpositive/llap/show_functions.q.out | 3 -
25 files changed, 5 insertions(+), 470 deletions(-)
diff --git a/data/files/datasets/temp_udf/load.hive.sql
b/data/files/datasets/temp_udf/load.hive.sql
new file mode 100644
index 00000000000..b8a21d37a23
--- /dev/null
+++ b/data/files/datasets/temp_udf/load.hive.sql
@@ -0,0 +1 @@
+CREATE FUNCTION compute_stats AS
'org.apache.hadoop.hive.ql.udf.generic.GenericUDAFComputeStats';
\ No newline at end of file
diff --git a/ql/src/java/org/apache/hadoop/hive/ql/exec/FunctionRegistry.java
b/ql/src/java/org/apache/hadoop/hive/ql/exec/FunctionRegistry.java
index c0b9519944d..6c1c5fb0113 100644
--- a/ql/src/java/org/apache/hadoop/hive/ql/exec/FunctionRegistry.java
+++ b/ql/src/java/org/apache/hadoop/hive/ql/exec/FunctionRegistry.java
@@ -563,7 +563,6 @@ public final class FunctionRegistry {
system.registerGenericUDAF("ngrams", new GenericUDAFnGrams());
system.registerGenericUDAF("context_ngrams", new
GenericUDAFContextNGrams());
- system.registerGenericUDAF("compute_stats", new GenericUDAFComputeStats());
system.registerGenericUDF("ndv_compute_bit_vector",
GenericUDFNDVComputeBitVector.class);
system.registerGenericUDAF("compute_bit_vector_hll", new
GenericUDAFComputeBitVectorHLL());
system.registerGenericUDAF("compute_bit_vector_fm", new
GenericUDAFComputeBitVectorFMSketch());
diff --git a/ql/src/test/queries/clientnegative/compute_stats_long.q
b/ql/src/test/queries/clientnegative/compute_stats_long.q
deleted file mode 100644
index bf87393abd0..00000000000
--- a/ql/src/test/queries/clientnegative/compute_stats_long.q
+++ /dev/null
@@ -1,7 +0,0 @@
-create table tab_int(a int);
-
--- insert some data
-LOAD DATA LOCAL INPATH "../../data/files/int.txt" INTO TABLE tab_int;
-
--- this should raise an error since the number of bit vectors has a hard limit
at 1024
-select compute_stats(a, 'fm', 10000) from tab_int;
diff --git a/ql/src/test/queries/clientpositive/char_udf1.q
b/ql/src/test/queries/clientpositive/char_udf1.q
index 20440436150..4a6ac4020c3 100644
--- a/ql/src/test/queries/clientpositive/char_udf1.q
+++ b/ql/src/test/queries/clientpositive/char_udf1.q
@@ -1,4 +1,5 @@
--! qt:dataset:src
+--! qt:dataset:temp_udf
drop table char_udf_1;
create table char_udf_1 (c1 string, c2 string, c3 char(10), c4 char(20));
diff --git a/ql/src/test/queries/clientpositive/compute_stats_binary.q
b/ql/src/test/queries/clientpositive/compute_stats_binary.q
deleted file mode 100644
index fd15634f202..00000000000
--- a/ql/src/test/queries/clientpositive/compute_stats_binary.q
+++ /dev/null
@@ -1,9 +0,0 @@
-create table tab_binary(a binary);
-
--- insert some data
-LOAD DATA LOCAL INPATH "../../data/files/binary.txt" INTO TABLE tab_binary;
-
-select count(*) from tab_binary;
-
--- compute statistical summary of data
-select compute_stats(a, 16) from tab_binary;
diff --git a/ql/src/test/queries/clientpositive/compute_stats_boolean.q
b/ql/src/test/queries/clientpositive/compute_stats_boolean.q
deleted file mode 100644
index cddb53f8f64..00000000000
--- a/ql/src/test/queries/clientpositive/compute_stats_boolean.q
+++ /dev/null
@@ -1,9 +0,0 @@
-create table tab_bool(a boolean);
-
--- insert some data
-LOAD DATA LOCAL INPATH "../../data/files/bool.txt" INTO TABLE tab_bool;
-
-select count(*) from tab_bool;
-
--- compute statistical summary of data
-select compute_stats(a, 16) from tab_bool;
diff --git a/ql/src/test/queries/clientpositive/compute_stats_date.q
b/ql/src/test/queries/clientpositive/compute_stats_date.q
deleted file mode 100644
index bf478526ba8..00000000000
--- a/ql/src/test/queries/clientpositive/compute_stats_date.q
+++ /dev/null
@@ -1,28 +0,0 @@
-
-create table tab_date (
- origin_city_name string,
- dest_city_name string,
- fl_date date,
- arr_delay float,
- fl_num int
-);
-
--- insert some data
-load data local inpath '../../data/files/flights_join.txt' overwrite into
table tab_date;
-
-select count(*) from tab_date;
-
--- compute statistical summary of data
-select compute_stats(fl_date, 'hll') from tab_date;
-
-explain
-analyze table tab_date compute statistics for columns fl_date;
-
-analyze table tab_date compute statistics for columns fl_date;
-
-describe formatted tab_date fl_date;
-
--- Update stats manually. Try both yyyy-mm-dd and integer value for high/low
value
-alter table tab_date update statistics for column fl_date set ('numDVs'='19',
'highValue'='2015-01-01', 'lowValue'='0');
-
-describe formatted tab_date fl_date;
diff --git a/ql/src/test/queries/clientpositive/compute_stats_decimal.q
b/ql/src/test/queries/clientpositive/compute_stats_decimal.q
deleted file mode 100644
index 2beafaf219a..00000000000
--- a/ql/src/test/queries/clientpositive/compute_stats_decimal.q
+++ /dev/null
@@ -1,11 +0,0 @@
-set hive.stats.autogather=true;
-
-create table tab_decimal(a decimal(35,3));
-
--- insert some data
-LOAD DATA LOCAL INPATH "../../data/files/decimal.txt" INTO TABLE tab_decimal;
-
-select count(*) from tab_decimal;
-
--- compute statistical summary of data
-select compute_stats(a, 'fm', 18) from tab_decimal;
diff --git a/ql/src/test/queries/clientpositive/compute_stats_double.q
b/ql/src/test/queries/clientpositive/compute_stats_double.q
deleted file mode 100644
index 6bae0643a89..00000000000
--- a/ql/src/test/queries/clientpositive/compute_stats_double.q
+++ /dev/null
@@ -1,9 +0,0 @@
-create table tab_double(a double);
-
--- insert some data
-LOAD DATA LOCAL INPATH "../../data/files/double.txt" INTO TABLE tab_double;
-
-select count(*) from tab_double;
-
--- compute statistical summary of data
-select compute_stats(a, 'fm', 16) from tab_double;
diff --git a/ql/src/test/queries/clientpositive/compute_stats_empty_table.q
b/ql/src/test/queries/clientpositive/compute_stats_empty_table.q
deleted file mode 100644
index d4ed93fa639..00000000000
--- a/ql/src/test/queries/clientpositive/compute_stats_empty_table.q
+++ /dev/null
@@ -1,12 +0,0 @@
-create table tab_empty(a boolean, b int, c double, d string, e binary);
-
-select count(*) from tab_empty;
-
--- compute statistical summary of data
-select compute_stats(a, 16) from tab_empty;
-select compute_stats(b, 16) from tab_empty;
-select compute_stats(c, 16) from tab_empty;
-select compute_stats(d, 16) from tab_empty;
-select compute_stats(e, 16) from tab_empty;
-
-
diff --git a/ql/src/test/queries/clientpositive/compute_stats_long.q
b/ql/src/test/queries/clientpositive/compute_stats_long.q
deleted file mode 100644
index 48f4ebb9791..00000000000
--- a/ql/src/test/queries/clientpositive/compute_stats_long.q
+++ /dev/null
@@ -1,9 +0,0 @@
-create table tab_int(a int);
-
--- insert some data
-LOAD DATA LOCAL INPATH "../../data/files/int.txt" INTO TABLE tab_int;
-
-select count(*) from tab_int;
-
--- compute statistical summary of data
-select compute_stats(a, 'fm', 16) from tab_int;
diff --git a/ql/src/test/queries/clientpositive/compute_stats_string.q
b/ql/src/test/queries/clientpositive/compute_stats_string.q
deleted file mode 100644
index 79a531e8ec4..00000000000
--- a/ql/src/test/queries/clientpositive/compute_stats_string.q
+++ /dev/null
@@ -1,9 +0,0 @@
-create table tab_string(a string);
-
--- insert some data
-LOAD DATA LOCAL INPATH "../../data/files/string.txt" INTO TABLE tab_string;
-
-select count(*) from tab_string;
-
--- compute statistical summary of data
-select compute_stats(a, 'fm', 16) from tab_string;
diff --git
a/ql/src/test/queries/clientpositive/reduceSinkDeDuplication_pRS_key_empty.q
b/ql/src/test/queries/clientpositive/reduceSinkDeDuplication_pRS_key_empty.q
index f92229560ae..8af3ff49385 100644
--- a/ql/src/test/queries/clientpositive/reduceSinkDeDuplication_pRS_key_empty.q
+++ b/ql/src/test/queries/clientpositive/reduceSinkDeDuplication_pRS_key_empty.q
@@ -1,4 +1,5 @@
--! qt:dataset:src
+--! qt:dataset:temp_udf
set hive.mapred.mode=nonstrict;
set hive.cbo.enable=false;
diff --git a/ql/src/test/queries/clientpositive/varchar_udf1.q
b/ql/src/test/queries/clientpositive/varchar_udf1.q
index 7aee1ce2251..80e184a8c19 100644
--- a/ql/src/test/queries/clientpositive/varchar_udf1.q
+++ b/ql/src/test/queries/clientpositive/varchar_udf1.q
@@ -1,4 +1,5 @@
--! qt:dataset:src
+--! qt:dataset:temp_udf
drop table varchar_udf_1;
create table varchar_udf_1 (c1 string, c2 string, c3 varchar(10), c4
varchar(20));
diff --git a/ql/src/test/queries/clientpositive/vector_udf1.q
b/ql/src/test/queries/clientpositive/vector_udf1.q
index a258025dc10..5a642dafb28 100644
--- a/ql/src/test/queries/clientpositive/vector_udf1.q
+++ b/ql/src/test/queries/clientpositive/vector_udf1.q
@@ -1,4 +1,5 @@
--! qt:dataset:src
+--! qt:dataset:temp_udf
SET hive.vectorized.execution.enabled=true;
set hive.fetch.task.conversion=none;
diff --git a/ql/src/test/results/clientnegative/compute_stats_long.q.out
b/ql/src/test/results/clientnegative/compute_stats_long.q.out
deleted file mode 100644
index 990a9417907..00000000000
--- a/ql/src/test/results/clientnegative/compute_stats_long.q.out
+++ /dev/null
@@ -1,57 +0,0 @@
-PREHOOK: query: create table tab_int(a int)
-PREHOOK: type: CREATETABLE
-PREHOOK: Output: database:default
-PREHOOK: Output: default@tab_int
-POSTHOOK: query: create table tab_int(a int)
-POSTHOOK: type: CREATETABLE
-POSTHOOK: Output: database:default
-POSTHOOK: Output: default@tab_int
-PREHOOK: query: LOAD DATA LOCAL INPATH "../../data/files/int.txt" INTO TABLE
tab_int
-PREHOOK: type: LOAD
-#### A masked pattern was here ####
-PREHOOK: Output: default@tab_int
-POSTHOOK: query: LOAD DATA LOCAL INPATH "../../data/files/int.txt" INTO TABLE
tab_int
-POSTHOOK: type: LOAD
-#### A masked pattern was here ####
-POSTHOOK: Output: default@tab_int
-PREHOOK: query: select compute_stats(a, 'fm', 10000) from tab_int
-PREHOOK: type: QUERY
-PREHOOK: Input: default@tab_int
-#### A masked pattern was here ####
-Status: Failed
-Vertex failed, vertexName=Map 1, vertexId=vertex_#ID#, diagnostics=[Task
failed, taskId=task_#ID#, diagnostics=[TaskAttempt 0 failed, info=[Error: Error
while running task ( failure ) : attempt_#ID#:java.lang.RuntimeException:
java.lang.RuntimeException: org.apache.hadoop.hive.ql.metadata.HiveException:
Hive Runtime Error while processing row
-#### A masked pattern was here ####
-Caused by: java.lang.RuntimeException:
org.apache.hadoop.hive.ql.metadata.HiveException: Hive Runtime Error while
processing row
-#### A masked pattern was here ####
-Caused by: org.apache.hadoop.hive.ql.metadata.HiveException: Hive Runtime
Error while processing row
-#### A masked pattern was here ####
-Caused by: org.apache.hadoop.hive.ql.metadata.HiveException: The maximum
allowed value for number of bit vectors is 1024, but was passed 10000 bit
vectors
-#### A masked pattern was here ####
-], TaskAttempt 1 failed, info=[Error: Error while running task ( failure ) :
attempt_#ID#:java.lang.RuntimeException: java.lang.RuntimeException:
org.apache.hadoop.hive.ql.metadata.HiveException: Hive Runtime Error while
processing row
-#### A masked pattern was here ####
-Caused by: java.lang.RuntimeException:
org.apache.hadoop.hive.ql.metadata.HiveException: Hive Runtime Error while
processing row
-#### A masked pattern was here ####
-Caused by: org.apache.hadoop.hive.ql.metadata.HiveException: Hive Runtime
Error while processing row
-#### A masked pattern was here ####
-Caused by: org.apache.hadoop.hive.ql.metadata.HiveException: The maximum
allowed value for number of bit vectors is 1024, but was passed 10000 bit
vectors
-#### A masked pattern was here ####
-]], Vertex did not succeed due to OWN_TASK_FAILURE, failedTasks:1
killedTasks:0, Vertex vertex_#ID# [Map 1] killed/failed due to:OWN_TASK_FAILURE]
-[Masked Vertex killed due to OTHER_VERTEX_FAILURE]
-DAG did not succeed due to VERTEX_FAILURE. failedVertices:1 killedVertices:1
-FAILED: Execution Error, return code 2 from
org.apache.hadoop.hive.ql.exec.tez.TezTask. Vertex failed, vertexName=Map 1,
vertexId=vertex_#ID#, diagnostics=[Task failed, taskId=task_#ID#,
diagnostics=[TaskAttempt 0 failed, info=[Error: Error while running task (
failure ) : attempt_#ID#:java.lang.RuntimeException:
java.lang.RuntimeException: org.apache.hadoop.hive.ql.metadata.HiveException:
Hive Runtime Error while processing row
-#### A masked pattern was here ####
-Caused by: java.lang.RuntimeException:
org.apache.hadoop.hive.ql.metadata.HiveException: Hive Runtime Error while
processing row
-#### A masked pattern was here ####
-Caused by: org.apache.hadoop.hive.ql.metadata.HiveException: Hive Runtime
Error while processing row
-#### A masked pattern was here ####
-Caused by: org.apache.hadoop.hive.ql.metadata.HiveException: The maximum
allowed value for number of bit vectors is 1024, but was passed 10000 bit
vectors
-#### A masked pattern was here ####
-], TaskAttempt 1 failed, info=[Error: Error while running task ( failure ) :
attempt_#ID#:java.lang.RuntimeException: java.lang.RuntimeException:
org.apache.hadoop.hive.ql.metadata.HiveException: Hive Runtime Error while
processing row
-#### A masked pattern was here ####
-Caused by: java.lang.RuntimeException:
org.apache.hadoop.hive.ql.metadata.HiveException: Hive Runtime Error while
processing row
-#### A masked pattern was here ####
-Caused by: org.apache.hadoop.hive.ql.metadata.HiveException: Hive Runtime
Error while processing row
-#### A masked pattern was here ####
-Caused by: org.apache.hadoop.hive.ql.metadata.HiveException: The maximum
allowed value for number of bit vectors is 1024, but was passed 10000 bit
vectors
-#### A masked pattern was here ####
-]], Vertex did not succeed due to OWN_TASK_FAILURE, failedTasks:1
killedTasks:0, Vertex vertex_#ID# [Map 1] killed/failed due
to:OWN_TASK_FAILURE][Masked Vertex killed due to OTHER_VERTEX_FAILURE]DAG did
not succeed due to VERTEX_FAILURE. failedVertices:1 killedVertices:1
diff --git a/ql/src/test/results/clientpositive/llap/compute_stats_binary.q.out
b/ql/src/test/results/clientpositive/llap/compute_stats_binary.q.out
deleted file mode 100644
index fc90c894cd9..00000000000
--- a/ql/src/test/results/clientpositive/llap/compute_stats_binary.q.out
+++ /dev/null
@@ -1,34 +0,0 @@
-PREHOOK: query: create table tab_binary(a binary)
-PREHOOK: type: CREATETABLE
-PREHOOK: Output: database:default
-PREHOOK: Output: default@tab_binary
-POSTHOOK: query: create table tab_binary(a binary)
-POSTHOOK: type: CREATETABLE
-POSTHOOK: Output: database:default
-POSTHOOK: Output: default@tab_binary
-PREHOOK: query: LOAD DATA LOCAL INPATH "../../data/files/binary.txt" INTO
TABLE tab_binary
-PREHOOK: type: LOAD
-#### A masked pattern was here ####
-PREHOOK: Output: default@tab_binary
-POSTHOOK: query: LOAD DATA LOCAL INPATH "../../data/files/binary.txt" INTO
TABLE tab_binary
-POSTHOOK: type: LOAD
-#### A masked pattern was here ####
-POSTHOOK: Output: default@tab_binary
-PREHOOK: query: select count(*) from tab_binary
-PREHOOK: type: QUERY
-PREHOOK: Input: default@tab_binary
-#### A masked pattern was here ####
-POSTHOOK: query: select count(*) from tab_binary
-POSTHOOK: type: QUERY
-POSTHOOK: Input: default@tab_binary
-#### A masked pattern was here ####
-10
-PREHOOK: query: select compute_stats(a, 16) from tab_binary
-PREHOOK: type: QUERY
-PREHOOK: Input: default@tab_binary
-#### A masked pattern was here ####
-POSTHOOK: query: select compute_stats(a, 16) from tab_binary
-POSTHOOK: type: QUERY
-POSTHOOK: Input: default@tab_binary
-#### A masked pattern was here ####
-{"columntype":"Binary","maxlength":58,"avglength":32.5,"countnulls":0}
diff --git
a/ql/src/test/results/clientpositive/llap/compute_stats_boolean.q.out
b/ql/src/test/results/clientpositive/llap/compute_stats_boolean.q.out
deleted file mode 100644
index be90ee22273..00000000000
--- a/ql/src/test/results/clientpositive/llap/compute_stats_boolean.q.out
+++ /dev/null
@@ -1,34 +0,0 @@
-PREHOOK: query: create table tab_bool(a boolean)
-PREHOOK: type: CREATETABLE
-PREHOOK: Output: database:default
-PREHOOK: Output: default@tab_bool
-POSTHOOK: query: create table tab_bool(a boolean)
-POSTHOOK: type: CREATETABLE
-POSTHOOK: Output: database:default
-POSTHOOK: Output: default@tab_bool
-PREHOOK: query: LOAD DATA LOCAL INPATH "../../data/files/bool.txt" INTO TABLE
tab_bool
-PREHOOK: type: LOAD
-#### A masked pattern was here ####
-PREHOOK: Output: default@tab_bool
-POSTHOOK: query: LOAD DATA LOCAL INPATH "../../data/files/bool.txt" INTO TABLE
tab_bool
-POSTHOOK: type: LOAD
-#### A masked pattern was here ####
-POSTHOOK: Output: default@tab_bool
-PREHOOK: query: select count(*) from tab_bool
-PREHOOK: type: QUERY
-PREHOOK: Input: default@tab_bool
-#### A masked pattern was here ####
-POSTHOOK: query: select count(*) from tab_bool
-POSTHOOK: type: QUERY
-POSTHOOK: Input: default@tab_bool
-#### A masked pattern was here ####
-33
-PREHOOK: query: select compute_stats(a, 16) from tab_bool
-PREHOOK: type: QUERY
-PREHOOK: Input: default@tab_bool
-#### A masked pattern was here ####
-POSTHOOK: query: select compute_stats(a, 16) from tab_bool
-POSTHOOK: type: QUERY
-POSTHOOK: Input: default@tab_bool
-#### A masked pattern was here ####
-{"columntype":"Boolean","counttrues":13,"countfalses":19,"countnulls":1}
diff --git a/ql/src/test/results/clientpositive/llap/compute_stats_date.q.out
b/ql/src/test/results/clientpositive/llap/compute_stats_date.q.out
deleted file mode 100644
index b33be22d858..00000000000
--- a/ql/src/test/results/clientpositive/llap/compute_stats_date.q.out
+++ /dev/null
@@ -1,176 +0,0 @@
-PREHOOK: query: create table tab_date (
- origin_city_name string,
- dest_city_name string,
- fl_date date,
- arr_delay float,
- fl_num int
-)
-PREHOOK: type: CREATETABLE
-PREHOOK: Output: database:default
-PREHOOK: Output: default@tab_date
-POSTHOOK: query: create table tab_date (
- origin_city_name string,
- dest_city_name string,
- fl_date date,
- arr_delay float,
- fl_num int
-)
-POSTHOOK: type: CREATETABLE
-POSTHOOK: Output: database:default
-POSTHOOK: Output: default@tab_date
-PREHOOK: query: load data local inpath '../../data/files/flights_join.txt'
overwrite into table tab_date
-PREHOOK: type: LOAD
-#### A masked pattern was here ####
-PREHOOK: Output: default@tab_date
-POSTHOOK: query: load data local inpath '../../data/files/flights_join.txt'
overwrite into table tab_date
-POSTHOOK: type: LOAD
-#### A masked pattern was here ####
-POSTHOOK: Output: default@tab_date
-PREHOOK: query: select count(*) from tab_date
-PREHOOK: type: QUERY
-PREHOOK: Input: default@tab_date
-#### A masked pattern was here ####
-POSTHOOK: query: select count(*) from tab_date
-POSTHOOK: type: QUERY
-POSTHOOK: Input: default@tab_date
-#### A masked pattern was here ####
-20
-PREHOOK: query: select compute_stats(fl_date, 'hll') from tab_date
-PREHOOK: type: QUERY
-PREHOOK: Input: default@tab_date
-#### A masked pattern was here ####
-POSTHOOK: query: select compute_stats(fl_date, 'hll') from tab_date
-POSTHOOK: type: QUERY
-POSTHOOK: Input: default@tab_date
-#### A masked pattern was here ####
-{"columntype":"Date","min":"2000-11-20","max":"2010-10-29","countnulls":0,"numdistinctvalues":19,"ndvbitvector":HLL������������t������������������������R��ո��������������Y������������������������������������������������犞}
-PREHOOK: query: explain
-analyze table tab_date compute statistics for columns fl_date
-PREHOOK: type: ANALYZE_TABLE
-PREHOOK: Input: default@tab_date
-PREHOOK: Output: default@tab_date
-#### A masked pattern was here ####
-POSTHOOK: query: explain
-analyze table tab_date compute statistics for columns fl_date
-POSTHOOK: type: ANALYZE_TABLE
-POSTHOOK: Input: default@tab_date
-POSTHOOK: Output: default@tab_date
-#### A masked pattern was here ####
-STAGE DEPENDENCIES:
- Stage-0 is a root stage
- Stage-2 depends on stages: Stage-0
-
-STAGE PLANS:
- Stage: Stage-0
- Tez
-#### A masked pattern was here ####
- Edges:
- Reducer 2 <- Map 1 (CUSTOM_SIMPLE_EDGE)
-#### A masked pattern was here ####
- Vertices:
- Map 1
- Map Operator Tree:
- TableScan
- alias: tab_date
- Statistics: Num rows: 1 Data size: 56 Basic stats: COMPLETE
Column stats: NONE
- Select Operator
- expressions: fl_date (type: date)
- outputColumnNames: fl_date
- Statistics: Num rows: 1 Data size: 56 Basic stats:
COMPLETE Column stats: NONE
- Group By Operator
- aggregations: min(fl_date), max(fl_date), count(1),
count(fl_date), compute_bit_vector_hll(fl_date)
- minReductionHashAggr: 0.99
- mode: hash
- outputColumnNames: _col0, _col1, _col2, _col3, _col4
- Statistics: Num rows: 1 Data size: 328 Basic stats:
COMPLETE Column stats: NONE
- Reduce Output Operator
- null sort order:
- sort order:
- Statistics: Num rows: 1 Data size: 328 Basic stats:
COMPLETE Column stats: NONE
- value expressions: _col0 (type: date), _col1 (type:
date), _col2 (type: bigint), _col3 (type: bigint), _col4 (type: binary)
- Execution mode: llap
- LLAP IO: all inputs
- Reducer 2
- Execution mode: vectorized, llap
- Reduce Operator Tree:
- Group By Operator
- aggregations: min(VALUE._col0), max(VALUE._col1),
count(VALUE._col2), count(VALUE._col3), compute_bit_vector_hll(VALUE._col4)
- mode: mergepartial
- outputColumnNames: _col0, _col1, _col2, _col3, _col4
- Statistics: Num rows: 1 Data size: 328 Basic stats: COMPLETE
Column stats: NONE
- Select Operator
- expressions: 'DATE' (type: string), _col0 (type: date),
_col1 (type: date), (_col2 - _col3) (type: bigint),
COALESCE(ndv_compute_bit_vector(_col4),0) (type: bigint), _col4 (type: binary)
- outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5
- Statistics: Num rows: 1 Data size: 328 Basic stats: COMPLETE
Column stats: NONE
- File Output Operator
- compressed: false
- Statistics: Num rows: 1 Data size: 328 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
-
- Stage: Stage-2
- Stats Work
- Basic Stats Work:
- Column Stats Desc:
- Columns: fl_date
- Column Types: date
- Table: default.tab_date
-
-PREHOOK: query: analyze table tab_date compute statistics for columns fl_date
-PREHOOK: type: ANALYZE_TABLE
-PREHOOK: Input: default@tab_date
-PREHOOK: Output: default@tab_date
-#### A masked pattern was here ####
-POSTHOOK: query: analyze table tab_date compute statistics for columns fl_date
-POSTHOOK: type: ANALYZE_TABLE
-POSTHOOK: Input: default@tab_date
-POSTHOOK: Output: default@tab_date
-#### A masked pattern was here ####
-PREHOOK: query: describe formatted tab_date fl_date
-PREHOOK: type: DESCTABLE
-PREHOOK: Input: default@tab_date
-POSTHOOK: query: describe formatted tab_date fl_date
-POSTHOOK: type: DESCTABLE
-POSTHOOK: Input: default@tab_date
-col_name fl_date
-data_type date
-min 2000-11-20
-max 2010-10-29
-num_nulls 0
-distinct_count 19
-avg_col_len
-max_col_len
-num_trues
-num_falses
-bit_vector HL
-comment from deserializer
-COLUMN_STATS_ACCURATE
{\"BASIC_STATS\":\"true\",\"COLUMN_STATS\":{\"fl_date\":\"true\"}}
-PREHOOK: query: alter table tab_date update statistics for column fl_date set
('numDVs'='19', 'highValue'='2015-01-01', 'lowValue'='0')
-PREHOOK: type: ALTERTABLE_UPDATETABLESTATS
-PREHOOK: Input: default@tab_date
-PREHOOK: Output: default@tab_date
-POSTHOOK: query: alter table tab_date update statistics for column fl_date set
('numDVs'='19', 'highValue'='2015-01-01', 'lowValue'='0')
-POSTHOOK: type: ALTERTABLE_UPDATETABLESTATS
-POSTHOOK: Input: default@tab_date
-POSTHOOK: Output: default@tab_date
-PREHOOK: query: describe formatted tab_date fl_date
-PREHOOK: type: DESCTABLE
-PREHOOK: Input: default@tab_date
-POSTHOOK: query: describe formatted tab_date fl_date
-POSTHOOK: type: DESCTABLE
-POSTHOOK: Input: default@tab_date
-col_name fl_date
-data_type date
-min 1970-01-01
-max 2015-01-01
-num_nulls 0
-distinct_count 19
-avg_col_len
-max_col_len
-num_trues
-num_falses
-bit_vector HL
-comment from deserializer
-COLUMN_STATS_ACCURATE
{\"BASIC_STATS\":\"true\",\"COLUMN_STATS\":{\"fl_date\":\"true\"}}
diff --git
a/ql/src/test/results/clientpositive/llap/compute_stats_decimal.q.out
b/ql/src/test/results/clientpositive/llap/compute_stats_decimal.q.out
deleted file mode 100644
index 810a8b232af..00000000000
Binary files
a/ql/src/test/results/clientpositive/llap/compute_stats_decimal.q.out and
/dev/null differ
diff --git a/ql/src/test/results/clientpositive/llap/compute_stats_double.q.out
b/ql/src/test/results/clientpositive/llap/compute_stats_double.q.out
deleted file mode 100644
index c6a90202249..00000000000
Binary files
a/ql/src/test/results/clientpositive/llap/compute_stats_double.q.out and
/dev/null differ
diff --git
a/ql/src/test/results/clientpositive/llap/compute_stats_empty_table.q.out
b/ql/src/test/results/clientpositive/llap/compute_stats_empty_table.q.out
deleted file mode 100644
index 2ce83e14bf1..00000000000
--- a/ql/src/test/results/clientpositive/llap/compute_stats_empty_table.q.out
+++ /dev/null
@@ -1,62 +0,0 @@
-PREHOOK: query: create table tab_empty(a boolean, b int, c double, d string, e
binary)
-PREHOOK: type: CREATETABLE
-PREHOOK: Output: database:default
-PREHOOK: Output: default@tab_empty
-POSTHOOK: query: create table tab_empty(a boolean, b int, c double, d string,
e binary)
-POSTHOOK: type: CREATETABLE
-POSTHOOK: Output: database:default
-POSTHOOK: Output: default@tab_empty
-PREHOOK: query: select count(*) from tab_empty
-PREHOOK: type: QUERY
-PREHOOK: Input: default@tab_empty
-#### A masked pattern was here ####
-POSTHOOK: query: select count(*) from tab_empty
-POSTHOOK: type: QUERY
-POSTHOOK: Input: default@tab_empty
-#### A masked pattern was here ####
-0
-PREHOOK: query: select compute_stats(a, 16) from tab_empty
-PREHOOK: type: QUERY
-PREHOOK: Input: default@tab_empty
-#### A masked pattern was here ####
-POSTHOOK: query: select compute_stats(a, 16) from tab_empty
-POSTHOOK: type: QUERY
-POSTHOOK: Input: default@tab_empty
-#### A masked pattern was here ####
-{"columntype":"Boolean","counttrues":0,"countfalses":0,"countnulls":0}
-PREHOOK: query: select compute_stats(b, 16) from tab_empty
-PREHOOK: type: QUERY
-PREHOOK: Input: default@tab_empty
-#### A masked pattern was here ####
-POSTHOOK: query: select compute_stats(b, 16) from tab_empty
-POSTHOOK: type: QUERY
-POSTHOOK: Input: default@tab_empty
-#### A masked pattern was here ####
-{"columntype":"Long","min":null,"max":null,"countnulls":0,"numdistinctvalues":0,"ndvbitvector":}
-PREHOOK: query: select compute_stats(c, 16) from tab_empty
-PREHOOK: type: QUERY
-PREHOOK: Input: default@tab_empty
-#### A masked pattern was here ####
-POSTHOOK: query: select compute_stats(c, 16) from tab_empty
-POSTHOOK: type: QUERY
-POSTHOOK: Input: default@tab_empty
-#### A masked pattern was here ####
-{"columntype":"Double","min":null,"max":null,"countnulls":0,"numdistinctvalues":0,"ndvbitvector":}
-PREHOOK: query: select compute_stats(d, 16) from tab_empty
-PREHOOK: type: QUERY
-PREHOOK: Input: default@tab_empty
-#### A masked pattern was here ####
-POSTHOOK: query: select compute_stats(d, 16) from tab_empty
-POSTHOOK: type: QUERY
-POSTHOOK: Input: default@tab_empty
-#### A masked pattern was here ####
-{"columntype":"String","maxlength":0,"avglength":0.0,"countnulls":0,"numdistinctvalues":0,"ndvbitvector":}
-PREHOOK: query: select compute_stats(e, 16) from tab_empty
-PREHOOK: type: QUERY
-PREHOOK: Input: default@tab_empty
-#### A masked pattern was here ####
-POSTHOOK: query: select compute_stats(e, 16) from tab_empty
-POSTHOOK: type: QUERY
-POSTHOOK: Input: default@tab_empty
-#### A masked pattern was here ####
-{"columntype":"Binary","maxlength":0,"avglength":0.0,"countnulls":0}
diff --git a/ql/src/test/results/clientpositive/llap/compute_stats_long.q.out
b/ql/src/test/results/clientpositive/llap/compute_stats_long.q.out
deleted file mode 100644
index 138ee3f1047..00000000000
Binary files a/ql/src/test/results/clientpositive/llap/compute_stats_long.q.out
and /dev/null differ
diff --git a/ql/src/test/results/clientpositive/llap/compute_stats_string.q.out
b/ql/src/test/results/clientpositive/llap/compute_stats_string.q.out
deleted file mode 100644
index 135ac45e754..00000000000
Binary files
a/ql/src/test/results/clientpositive/llap/compute_stats_string.q.out and
/dev/null differ
diff --git a/ql/src/test/results/clientpositive/llap/show_functions.q.out
b/ql/src/test/results/clientpositive/llap/show_functions.q.out
index e5a24964711..9642bfbd8b0 100644
--- a/ql/src/test/results/clientpositive/llap/show_functions.q.out
+++ b/ql/src/test/results/clientpositive/llap/show_functions.q.out
@@ -83,7 +83,6 @@ collect_list
collect_set
compute_bit_vector_fm
compute_bit_vector_hll
-compute_stats
concat
concat_ws
context_ngrams
@@ -533,7 +532,6 @@ collect_list
collect_set
compute_bit_vector_fm
compute_bit_vector_hll
-compute_stats
concat
concat_ws
context_ngrams
@@ -713,7 +711,6 @@ collect_list
collect_set
compute_bit_vector_fm
compute_bit_vector_hll
-compute_stats
concat
concat_ws
context_ngrams
