[CARBONDATA-1446] Alter query throws invalid exception while executing on range partitioned table

• Type: Bug
• Status: Closed
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: 1.2.0
• Fix Version/s: 1.2.0
• Component/s: data-query
• Environment: spark 2.1
• Assignee: Unassigned
• Reporter: Vandana Yadav
• Created: 04/Sep/17 06:24
• Updated: 01/Nov/17 06:14
• Resolved: 12/Sep/17 14:13

Description

Alter query throws invalid exception while executing on range partitioned table with invalid partition id

Steps to reproduce:
1) Create table
CREATE TABLE uniqdata_part1 (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) PARTITIONED BY (CUST_ID int) STORED BY 'org.apache.carbondata.format' TBLPROPERTIES ('PARTITION_TYPE'='RANGE','RANGE_INFO'='9090,9500,9800',"TABLE_BLOCKSIZE"= "256 MB")

2)Load data into the table:
LOAD DATA INPATH 'hdfs://localhost:54310/uniqdata/2000_UniqData.csv' into table uniqdata_part1 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')

3)Execute Query:
ALTER TABLE uniqdata_part1 SPLIT PARTITION(6) INTO ('9800','9900')

4) Result On beeline:
Error: java.lang.RuntimeException: Range info must be in ascending order, please check again! (state=,code=0)

5) Expected Result:
As we are giving invalid partition id then error/exception should be like invalid partition id (related to partition id rather than range info).

Issue Links

• links to: GitHub Pull Request #1320