Vandana Yadav created CARBONDATA-2225:
-----------------------------------------
Summary: Inaccurate result displays while using histogram_numeric
Key: CARBONDATA-2225
URL: https://issues.apache.org/jira/browse/CARBONDATA-2225
Project: CarbonData
Issue Type: Bug
Components: data-query
Affects Versions: 1.4.0
Environment: spark 2.2.1
Reporter: Vandana Yadav
Attachments: 2000_UniqData.csv, 3000_1_UniqData.csv,
4000_UniqData.csv, 6000_UniqData.csv, 7000_UniqData.csv
Inaccurate result displays while using histogram_numeric
Steps to reproduce:
1: Create carbon table:
a) CREATE TABLE uniqdata (CUST_ID int,CUST_NAME String,ACTIVE_EMUI_VERSION
string, DOB timestamp, DOJ timestamp, BIGINT_COLUMN1 bigint,BIGINT_COLUMN2
bigint,DECIMAL_COLUMN1 decimal(30,10), DECIMAL_COLUMN2
decimal(36,10),Double_COLUMN1 double, Double_COLUMN2 double,INTEGER_COLUMN1
int) STORED BY 'org.apache.carbondata.format' TBLPROPERTIES ("TABLE_BLOCKSIZE"=
"256 MB");
b) LOAD DATA INPATH 'hdfs://localhost:54310/Data/uniqdata/2000_UniqData.csv'
into table uniqdata OPTIONS('DELIMITER'=',' ,
'QUOTECHAR'='"','BAD_RECORDS_ACTION'='FORCE','FILEHEADER'='CUST_ID,CUST_NAME,ACTIVE_EMUI_VERSION,DOB,DOJ,BIGINT_COLUMN1,BIGINT_COLUMN2,DECIMAL_COLUMN1,DECIMAL_COLUMN2,Double_COLUMN1,Double_COLUMN2,INTEGER_COLUMN1');
c) LOAD DATA INPATH 'hdfs://localhost:54310/Data/uniqdata/4000_UniqData.csv'
into table uniqdata OPTIONS('DELIMITER'=',' ,
'QUOTECHAR'='"','BAD_RECORDS_ACTION'='FORCE','FILEHEADER'='CUST_ID,CUST_NAME,ACTIVE_EMUI_VERSION,DOB,DOJ,BIGINT_COLUMN1,BIGINT_COLUMN2,DECIMAL_COLUMN1,DECIMAL_COLUMN2,Double_COLUMN1,Double_COLUMN2,INTEGER_COLUMN1');
d) LOAD DATA INPATH 'hdfs://localhost:54310/Data/uniqdata/6000_UniqData.csv'
into table uniqdata OPTIONS('DELIMITER'=',' ,
'QUOTECHAR'='"','BAD_RECORDS_ACTION'='FORCE','FILEHEADER'='CUST_ID,CUST_NAME,ACTIVE_EMUI_VERSION,DOB,DOJ,BIGINT_COLUMN1,BIGINT_COLUMN2,DECIMAL_COLUMN1,DECIMAL_COLUMN2,Double_COLUMN1,Double_COLUMN2,INTEGER_COLUMN1');
e) LOAD DATA INPATH 'hdfs://localhost:54310/Data/uniqdata/7000_UniqData.csv'
into table uniqdata OPTIONS('DELIMITER'=',' ,
'QUOTECHAR'='"','BAD_RECORDS_ACTION'='FORCE','FILEHEADER'='CUST_ID,CUST_NAME,ACTIVE_EMUI_VERSION,DOB,DOJ,BIGINT_COLUMN1,BIGINT_COLUMN2,DECIMAL_COLUMN1,DECIMAL_COLUMN2,Double_COLUMN1,Double_COLUMN2,INTEGER_COLUMN1');
f) LOAD DATA INPATH 'hdfs://localhost:54310/Data/uniqdata/3000_1_UniqData.csv'
into table uniqdata OPTIONS('DELIMITER'=',' ,
'QUOTECHAR'='"','BAD_RECORDS_ACTION'='FORCE','FILEHEADER'='CUST_ID,CUST_NAME,ACTIVE_EMUI_VERSION,DOB,DOJ,BIGINT_COLUMN1,BIGINT_COLUMN2,DECIMAL_COLUMN1,DECIMAL_COLUMN2,Double_COLUMN1,Double_COLUMN2,INTEGER_COLUMN1');
2) Create Hive Table:
a) CREATE TABLE uniqdata_h (CUST_ID int,CUST_NAME String,ACTIVE_EMUI_VERSION
string, DOB timestamp, DOJ timestamp, BIGINT_COLUMN1 bigint,BIGINT_COLUMN2
bigint,DECIMAL_COLUMN1 decimal(30,10), DECIMAL_COLUMN2
decimal(36,10),Double_COLUMN1 double, Double_COLUMN2 double,INTEGER_COLUMN1
int) ROW FORMAT DELIMITED FIELDS TERMINATED BY ',';
b) load data local inpath
'/home/knoldus/Desktop/csv/TestData/Data/uniqdata/2000_UniqData.csv' into table
uniqdata_h;
c) load data local inpath
'/home/knoldus/Desktop/csv/TestData/Data/uniqdata/4000_UniqData.csv' into table
uniqdata_h;
d) load data local inpath
'/home/knoldus/Desktop/csv/TestData/Data/uniqdata/6000_UniqData.csv' into table
uniqdata_h;
e) load data local inpath
'/home/knoldus/Desktop/csv/TestData/Data/uniqdata/7000_UniqData.csv' into table
uniqdata_h;
f)load data local inpath
'/home/knoldus/Desktop/csv/TestData/Data/uniqdata/3000_1_UniqData.csv' into
table uniqdata_h;
3) Execute Query:
a) hive:
select histogram_numeric(1,2) from uniqdata_h where CUST_ID IS NULL or DOB IS
NOT NULL or BIGINT_COLUMN1 =1233720368578 or DECIMAL_COLUMN1 =
12345678901.1234000058 or Double_COLUMN1 = 1.12345674897976E10 or
INTEGER_COLUMN1 IS NULL ;
output:
+------------------------------------------------+--+
| histogram_numeric( 1, 2) |
+------------------------------------------------+--+
| [\{"x":1.0,"y":12040.0},\{"x":1.0,"y":10026.0}] |
+------------------------------------------------+--+
1 row selected (3.173 seconds)
b) Carbon:
select histogram_numeric(1,2) from uniqdata where CUST_ID IS NULL or DOB IS NOT
NULL or BIGINT_COLUMN1 =1233720368578 or DECIMAL_COLUMN1 =
12345678901.1234000058 or Double_COLUMN1 = 1.12345674897976E10 or
INTEGER_COLUMN1 IS NULL ;
output:
+-----------------------------------------------+--+
| histogram_numeric( 1, 2) |
+-----------------------------------------------+--+
| [\{"x":1.0,"y":13039.0},\{"x":1.0,"y":9027.0}] |
+-----------------------------------------------+--+
1 row selected (0.95 seconds)
4) Expected Result: Result of spark hive and carbon should be same for the same
function with same data.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
