[CARBONDATA-586] Create table with 'Char' data type but it workes as 'String' data type

• Type: Improvement
• Status: Resolved
• Priority: Minor
• Resolution: Won't Fix
• Affects Version/s: 1.0.0-incubating
• Component/s: data-load
• Environment: spark 1.6.2; spark 2.1
• Assignee: David Cai
• Reporter: Anurag Srivastava
• Created: 03/Jan/17 07:58
• Updated: 10/May/17 07:53
• Resolved: 10/May/17 07:53

Description

I am trying to use Char data type with Carbon Data latest version and it created successfully. When I started loading data in this that time I found that it is taking data more then its size.

I have checked it with hive and there it is working fine.

EX :-

1. Carbon Data :

1.1 create table test_carbon (name char(10)) stored by 'org.apache.carbondata.format';

1.2 desc test_carbon;

Output :
---------------------------------+

col_name

data_type

comment

-------------------------------------

name

string

-------------------------------------

1.3 LOAD DATA INPATH 'hdfs://localhost:54310/test.csv' into table test_carbon OPTIONS ('FILEHEADER'='name');

1.4 select * from test_carbon;

Output :
------------------------

name

------------------------

Anurag Srivasrata

Robert

james james

------------------------

2. Hive :

2.1 create table test_hive (name char(10));

2.2 desc test_hive;

Output :
------------------------------------

col_name

data_type

comment

------------------------------------

name

char(10)

NULL

------------------------------------

2.3 LOAD DATA INPATH 'hdfs://localhost:54310/test.csv' into table test_hive;

2.4 select * from test_hive;

Output :
----------------

name

----------------

james jame

Anurag Sri

Robert

----------------

So as hive truncate remaining string with Char data type in carbon data it should work like hive.