mirror of
https://github.com/apache/impala.git
synced 2026-02-01 21:00:29 -05:00
a93f2c2675
This change compute the real number of true and false statistics information for boolean columns. Before this, impala used to set numTrues and numFalses to hardcoded -1 to indicate that its statistics is missing. Test Done: Append the numTrue and numFalse test for all the statistics-related test cases including the non-incremental, incremental and other test cases. Change-Id: I991bee8e7fdc644d908289f5fe2ee8032cc2c431 Reviewed-on: http://gerrit.cloudera.org:8080/14666 Reviewed-by: Impala Public Jenkins <impala-public-jenkins@cloudera.com> Tested-by: Impala Public Jenkins <impala-public-jenkins@cloudera.com>
378 lines
12 KiB
Plaintext
378 lines
12 KiB
Plaintext
====
|
|
---- QUERY
|
|
# IMPALA-867: Test computing stats on Avro tables created by Hive with
|
|
# matching/mismatched column definitions and Avro schema.
|
|
# Clone the used tables here.
|
|
create table avro_hive_alltypes
|
|
like functional_avro_snap.alltypes;
|
|
create table avro_hive_alltypes_extra_coldef
|
|
like functional_avro_snap.alltypes_extra_coldef;
|
|
create table avro_hive_alltypes_missing_coldef
|
|
like functional_avro_snap.alltypes_missing_coldef;
|
|
create table avro_hive_alltypes_type_mismatch
|
|
like functional_avro_snap.alltypes_type_mismatch;
|
|
create table avro_hive_no_avro_schema
|
|
like functional_avro_snap.no_avro_schema;
|
|
====
|
|
---- QUERY
|
|
# Avro table with matching column definitions and Avro schema
|
|
compute stats avro_hive_alltypes
|
|
---- RESULTS
|
|
'Updated 0 partition(s) and 11 column(s).'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
show table stats avro_hive_alltypes
|
|
---- LABELS
|
|
YEAR, MONTH, #ROWS, #FILES, SIZE, BYTES CACHED, CACHE REPLICATION, FORMAT, INCREMENTAL STATS, LOCATION
|
|
---- RESULTS
|
|
'Total','',0,0,'0B','0B','','','',''
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, STRING, STRING, STRING, STRING, STRING, STRING
|
|
====
|
|
---- QUERY
|
|
show column stats avro_hive_alltypes
|
|
---- LABELS
|
|
COLUMN, TYPE, #DISTINCT VALUES, #NULLS, MAX SIZE, AVG SIZE, #TRUES, #FALSES
|
|
---- RESULTS
|
|
'id','INT',0,0,4,4,-1,-1
|
|
'bool_col','BOOLEAN',2,0,1,1,0,0
|
|
'tinyint_col','INT',0,0,4,4,-1,-1
|
|
'smallint_col','INT',0,0,4,4,-1,-1
|
|
'int_col','INT',0,0,4,4,-1,-1
|
|
'bigint_col','BIGINT',0,0,8,8,-1,-1
|
|
'float_col','FLOAT',0,0,4,4,-1,-1
|
|
'double_col','DOUBLE',0,0,8,8,-1,-1
|
|
'date_string_col','STRING',0,0,0,0,-1,-1
|
|
'string_col','STRING',0,0,0,0,-1,-1
|
|
'timestamp_col','STRING',0,0,0,0,-1,-1
|
|
'year','INT',0,0,4,4,-1,-1
|
|
'month','INT',0,0,4,4,-1,-1
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, BIGINT, DOUBLE, BIGINT, BIGINT
|
|
====
|
|
---- QUERY
|
|
# Avro table with an extra column definition.
|
|
compute stats avro_hive_alltypes_extra_coldef
|
|
---- RESULTS
|
|
'Updated 0 partition(s) and 12 column(s).'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
show table stats avro_hive_alltypes_extra_coldef
|
|
---- LABELS
|
|
YEAR, MONTH, #ROWS, #FILES, SIZE, BYTES CACHED, CACHE REPLICATION, FORMAT, INCREMENTAL STATS, LOCATION
|
|
---- RESULTS
|
|
'Total','',0,0,'0B','0B','','','',''
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, STRING, STRING, STRING, STRING, STRING, STRING
|
|
====
|
|
---- QUERY
|
|
show column stats avro_hive_alltypes_extra_coldef
|
|
---- LABELS
|
|
COLUMN, TYPE, #DISTINCT VALUES, #NULLS, MAX SIZE, AVG SIZE, #TRUES, #FALSES
|
|
---- RESULTS
|
|
'id','INT',0,0,4,4,-1,-1
|
|
'bool_col','BOOLEAN',2,0,1,1,0,0
|
|
'tinyint_col','TINYINT',0,0,1,1,-1,-1
|
|
'smallint_col','SMALLINT',0,0,2,2,-1,-1
|
|
'int_col','INT',0,0,4,4,-1,-1
|
|
'bigint_col','BIGINT',0,0,8,8,-1,-1
|
|
'float_col','FLOAT',0,0,4,4,-1,-1
|
|
'double_col','DOUBLE',0,0,8,8,-1,-1
|
|
'date_string_col','STRING',0,0,0,0,-1,-1
|
|
'string_col','STRING',0,0,0,0,-1,-1
|
|
'timestamp_col','TIMESTAMP',0,0,16,16,-1,-1
|
|
'extra_col','STRING',0,0,0,0,-1,-1
|
|
'year','INT',0,0,4,4,-1,-1
|
|
'month','INT',0,0,4,4,-1,-1
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, BIGINT, DOUBLE, BIGINT, BIGINT
|
|
====
|
|
---- QUERY
|
|
# Avro table with missing two column definitions.
|
|
compute stats avro_hive_alltypes_missing_coldef
|
|
---- RESULTS
|
|
'Updated 0 partition(s) and 9 column(s).'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
show table stats avro_hive_alltypes_missing_coldef
|
|
---- LABELS
|
|
YEAR, MONTH, #ROWS, #FILES, SIZE, BYTES CACHED, CACHE REPLICATION, FORMAT, INCREMENTAL STATS, LOCATION
|
|
---- RESULTS
|
|
'Total','',0,0,'0B','0B','','','',''
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, STRING, STRING, STRING, STRING, STRING, STRING
|
|
====
|
|
---- QUERY
|
|
show column stats avro_hive_alltypes_missing_coldef
|
|
---- LABELS
|
|
COLUMN, TYPE, #DISTINCT VALUES, #NULLS, MAX SIZE, AVG SIZE, #TRUES, #FALSES
|
|
---- RESULTS
|
|
'id','INT',0,0,4,4,-1,-1
|
|
'bool_col','BOOLEAN',2,0,1,1,0,0
|
|
'smallint_col','SMALLINT',0,0,2,2,-1,-1
|
|
'int_col','INT',0,0,4,4,-1,-1
|
|
'bigint_col','BIGINT',0,0,8,8,-1,-1
|
|
'float_col','FLOAT',0,0,4,4,-1,-1
|
|
'double_col','DOUBLE',0,0,8,8,-1,-1
|
|
'date_string_col','STRING',0,0,0,0,-1,-1
|
|
'string_col','STRING',0,0,0,0,-1,-1
|
|
'year','INT',0,0,4,4,-1,-1
|
|
'month','INT',0,0,4,4,-1,-1
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, BIGINT, DOUBLE, BIGINT, BIGINT
|
|
====
|
|
---- QUERY
|
|
# Avro table with one column definition having a different
|
|
# type than the Avro schema (bigint_col is a string).
|
|
compute stats avro_hive_alltypes_type_mismatch
|
|
---- RESULTS
|
|
'Updated 0 partition(s) and 11 column(s).'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
show table stats avro_hive_alltypes_type_mismatch
|
|
---- LABELS
|
|
YEAR, MONTH, #ROWS, #FILES, SIZE, BYTES CACHED, CACHE REPLICATION, FORMAT, INCREMENTAL STATS, LOCATION
|
|
---- RESULTS
|
|
'Total','',0,0,'0B','0B','','','',''
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, STRING, STRING, STRING, STRING, STRING, STRING
|
|
====
|
|
---- QUERY
|
|
show column stats avro_hive_alltypes_type_mismatch
|
|
---- LABELS
|
|
COLUMN, TYPE, #DISTINCT VALUES, #NULLS, MAX SIZE, AVG SIZE, #TRUES, #FALSES
|
|
---- RESULTS
|
|
'id','INT',0,0,4,4,-1,-1
|
|
'bool_col','BOOLEAN',2,0,1,1,0,0
|
|
'tinyint_col','TINYINT',0,0,1,1,-1,-1
|
|
'smallint_col','SMALLINT',0,0,2,2,-1,-1
|
|
'int_col','INT',0,0,4,4,-1,-1
|
|
'bigint_col','STRING',0,0,0,0,-1,-1
|
|
'float_col','FLOAT',0,0,4,4,-1,-1
|
|
'double_col','DOUBLE',0,0,8,8,-1,-1
|
|
'date_string_col','STRING',0,0,0,0,-1,-1
|
|
'string_col','STRING',0,0,0,0,-1,-1
|
|
'timestamp_col','TIMESTAMP',0,0,16,16,-1,-1
|
|
'year','INT',0,0,4,4,-1,-1
|
|
'month','INT',0,0,4,4,-1,-1
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, BIGINT, DOUBLE, BIGINT, BIGINT
|
|
====
|
|
---- QUERY
|
|
# Avro table without an Avro schema created by Hive.
|
|
# The Avro schema is inferred from the column definitions,
|
|
compute stats avro_hive_no_avro_schema
|
|
---- RESULTS
|
|
'Updated 0 partition(s) and 11 column(s).'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
show table stats avro_hive_no_avro_schema
|
|
---- LABELS
|
|
YEAR, MONTH, #ROWS, #FILES, SIZE, BYTES CACHED, CACHE REPLICATION, FORMAT, INCREMENTAL STATS, LOCATION
|
|
---- RESULTS
|
|
'Total','',0,0,'0B','0B','','','',''
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, STRING, STRING, STRING, STRING, STRING, STRING
|
|
====
|
|
---- QUERY
|
|
show column stats avro_hive_no_avro_schema
|
|
---- LABELS
|
|
COLUMN, TYPE, #DISTINCT VALUES, #NULLS, MAX SIZE, AVG SIZE, #TRUES, #FALSES
|
|
---- RESULTS
|
|
'id','INT',0,0,4,4,-1,-1
|
|
'bool_col','BOOLEAN',2,0,1,1,0,0
|
|
'tinyint_col','INT',0,0,4,4,-1,-1
|
|
'smallint_col','INT',0,0,4,4,-1,-1
|
|
'int_col','INT',0,0,4,4,-1,-1
|
|
'bigint_col','BIGINT',0,0,8,8,-1,-1
|
|
'float_col','FLOAT',0,0,4,4,-1,-1
|
|
'double_col','DOUBLE',0,0,8,8,-1,-1
|
|
'date_string_col','STRING',0,0,0,0,-1,-1
|
|
'string_col','STRING',0,0,0,0,-1,-1
|
|
'timestamp_col','STRING',0,0,0,0,-1,-1
|
|
'year','INT',0,0,4,4,-1,-1
|
|
'month','INT',0,0,4,4,-1,-1
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, BIGINT, DOUBLE, BIGINT, BIGINT
|
|
====
|
|
---- QUERY
|
|
# Test Avro table created without any column definitions.
|
|
create table avro_impala_alltypes_no_coldefs
|
|
partitioned by (year int, month int)
|
|
with serdeproperties
|
|
('avro.schema.url'='$FILESYSTEM_PREFIX/test-warehouse/avro_schemas/functional/alltypes.json')
|
|
stored as avro;
|
|
====
|
|
---- QUERY
|
|
compute stats avro_impala_alltypes_no_coldefs
|
|
---- RESULTS
|
|
'Updated 0 partition(s) and 11 column(s).'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
show table stats avro_impala_alltypes_no_coldefs
|
|
---- LABELS
|
|
YEAR, MONTH, #ROWS, #FILES, SIZE, BYTES CACHED, CACHE REPLICATION, FORMAT, INCREMENTAL STATS, LOCATION
|
|
---- RESULTS
|
|
'Total','',0,0,'0B','0B','','','',''
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, STRING, STRING, STRING, STRING, STRING, STRING
|
|
====
|
|
---- QUERY
|
|
show column stats avro_impala_alltypes_no_coldefs
|
|
---- LABELS
|
|
COLUMN, TYPE, #DISTINCT VALUES, #NULLS, MAX SIZE, AVG SIZE, #TRUES, #FALSES
|
|
---- RESULTS
|
|
'id','INT',0,0,4,4,-1,-1
|
|
'bool_col','BOOLEAN',2,0,1,1,0,0
|
|
'tinyint_col','INT',0,0,4,4,-1,-1
|
|
'smallint_col','INT',0,0,4,4,-1,-1
|
|
'int_col','INT',0,0,4,4,-1,-1
|
|
'bigint_col','BIGINT',0,0,8,8,-1,-1
|
|
'float_col','FLOAT',0,0,4,4,-1,-1
|
|
'double_col','DOUBLE',0,0,8,8,-1,-1
|
|
'date_string_col','STRING',0,0,0,0,-1,-1
|
|
'string_col','STRING',0,0,0,0,-1,-1
|
|
'timestamp_col','STRING',0,0,0,0,-1,-1
|
|
'year','INT',0,0,4,4,-1,-1
|
|
'month','INT',0,0,4,4,-1,-1
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, BIGINT, DOUBLE, BIGINT, BIGINT
|
|
====
|
|
---- QUERY
|
|
# IMPALA-1104: Test computing stats on Avro tables created by Impala
|
|
# with mismatched column definitions and Avro schema. Mismatched column name.
|
|
create table avro_impala_alltypes_bad_colname
|
|
(id int, bool_col boolean, tinyint_col int, smallint_col int, bad_int_col int,
|
|
bigint_col bigint, float_col float, double_col double, date_string_col string,
|
|
string_col string, timestamp_col timestamp)
|
|
partitioned by (year int, month int)
|
|
with serdeproperties
|
|
('avro.schema.url'='$FILESYSTEM_PREFIX/test-warehouse/avro_schemas/functional/alltypes.json')
|
|
stored as avro;
|
|
====
|
|
---- QUERY
|
|
compute stats avro_impala_alltypes_bad_colname
|
|
---- RESULTS
|
|
'Updated 0 partition(s) and 11 column(s).'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
show table stats avro_impala_alltypes_bad_colname
|
|
---- LABELS
|
|
YEAR, MONTH, #ROWS, #FILES, SIZE, BYTES CACHED, CACHE REPLICATION, FORMAT, INCREMENTAL STATS, LOCATION
|
|
---- RESULTS
|
|
'Total','',0,0,'0B','0B','','','',''
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, STRING, STRING, STRING, STRING, STRING, STRING
|
|
====
|
|
---- QUERY
|
|
show column stats avro_impala_alltypes_bad_colname
|
|
---- LABELS
|
|
COLUMN, TYPE, #DISTINCT VALUES, #NULLS, MAX SIZE, AVG SIZE, #TRUES, #FALSES
|
|
---- RESULTS
|
|
'id','INT',0,0,4,4,-1,-1
|
|
'bool_col','BOOLEAN',2,0,1,1,0,0
|
|
'tinyint_col','INT',0,0,4,4,-1,-1
|
|
'smallint_col','INT',0,0,4,4,-1,-1
|
|
'int_col','INT',0,0,4,4,-1,-1
|
|
'bigint_col','BIGINT',0,0,8,8,-1,-1
|
|
'float_col','FLOAT',0,0,4,4,-1,-1
|
|
'double_col','DOUBLE',0,0,8,8,-1,-1
|
|
'date_string_col','STRING',0,0,0,0,-1,-1
|
|
'string_col','STRING',0,0,0,0,-1,-1
|
|
'timestamp_col','STRING',0,0,0,0,-1,-1
|
|
'year','INT',0,0,4,4,-1,-1
|
|
'month','INT',0,0,4,4,-1,-1
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, BIGINT, DOUBLE, BIGINT, BIGINT
|
|
====
|
|
---- QUERY
|
|
# IMPALA-1104: Test computing stats on Avro tables created by Impala
|
|
# with mismatched column definitions and Avro schema. Mismatched column type.
|
|
create table avro_impala_alltypes_bad_coltype
|
|
(id int, bool_col boolean, tinyint_col int, smallint_col int, int_col int,
|
|
bigint_col bigint, float_col float, double_col bigint, date_string_col string,
|
|
string_col string, timestamp_col timestamp)
|
|
partitioned by (year int, month int)
|
|
with serdeproperties
|
|
('avro.schema.url'='$FILESYSTEM_PREFIX/test-warehouse/avro_schemas/functional/alltypes.json')
|
|
stored as avro;
|
|
====
|
|
---- QUERY
|
|
compute stats avro_impala_alltypes_bad_coltype
|
|
---- RESULTS
|
|
'Updated 0 partition(s) and 11 column(s).'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
show table stats avro_impala_alltypes_bad_coltype
|
|
---- LABELS
|
|
YEAR, MONTH, #ROWS, #FILES, SIZE, BYTES CACHED, CACHE REPLICATION, FORMAT, INCREMENTAL STATS, LOCATION
|
|
---- RESULTS
|
|
'Total','',0,0,'0B','0B','','','',''
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, STRING, STRING, STRING, STRING, STRING, STRING
|
|
====
|
|
---- QUERY
|
|
show column stats avro_impala_alltypes_bad_coltype
|
|
---- LABELS
|
|
COLUMN, TYPE, #DISTINCT VALUES, #NULLS, MAX SIZE, AVG SIZE, #TRUES, #FALSES
|
|
---- RESULTS
|
|
'id','INT',0,0,4,4,-1,-1
|
|
'bool_col','BOOLEAN',2,0,1,1,0,0
|
|
'tinyint_col','INT',0,0,4,4,-1,-1
|
|
'smallint_col','INT',0,0,4,4,-1,-1
|
|
'int_col','INT',0,0,4,4,-1,-1
|
|
'bigint_col','BIGINT',0,0,8,8,-1,-1
|
|
'float_col','FLOAT',0,0,4,4,-1,-1
|
|
'double_col','DOUBLE',0,0,8,8,-1,-1
|
|
'date_string_col','STRING',0,0,0,0,-1,-1
|
|
'string_col','STRING',0,0,0,0,-1,-1
|
|
'timestamp_col','STRING',0,0,0,0,-1,-1
|
|
'year','INT',0,0,4,4,-1,-1
|
|
'month','INT',0,0,4,4,-1,-1
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, BIGINT, DOUBLE, BIGINT, BIGINT
|
|
====
|
|
---- QUERY
|
|
# Test that compute stats on a Hive-created Avro table without column defs
|
|
# works (HIVE-6308, IMPALA-867).
|
|
create table alltypes_no_coldef like functional_avro_snap.alltypes_no_coldef;
|
|
compute stats alltypes_no_coldef
|
|
---- RESULTS
|
|
'Updated 1 partition(s) and 11 column(s).'
|
|
---- TYPES
|
|
STRING
|
|
====
|
|
---- QUERY
|
|
show column stats alltypes_no_coldef
|
|
---- RESULTS
|
|
'id','INT',0,0,4,4,-1,-1
|
|
'bool_col','BOOLEAN',2,0,1,1,0,0
|
|
'tinyint_col','INT',0,0,4,4,-1,-1
|
|
'smallint_col','INT',0,0,4,4,-1,-1
|
|
'int_col','INT',0,0,4,4,-1,-1
|
|
'bigint_col','BIGINT',0,0,8,8,-1,-1
|
|
'float_col','FLOAT',0,0,4,4,-1,-1
|
|
'double_col','DOUBLE',0,0,8,8,-1,-1
|
|
'date_string_col','STRING',0,0,0,0,-1,-1
|
|
'string_col','STRING',0,0,0,0,-1,-1
|
|
'timestamp_col','STRING',0,0,0,0,-1,-1
|
|
---- TYPES
|
|
STRING, STRING, BIGINT, BIGINT, BIGINT, DOUBLE, BIGINT, BIGINT
|
|
====
|