hi Singh

   quotechar in the csv should be in pairs. like 
name, description, salary, age, dob 
tammy,$my name$,$900000$,22,19/10/2019 
tammy1,$delhi$,$32345%*$*,22,19/10/2019
tammy2,$banglore$,$543$,$44$,19/10/2019
tammy3,$city$,$343$,$22$,12/10/2019 
tammy4,$punjab$,$23423$,$55$,19/10/2019 


Harmeet Singh wrote
> Hi Team,
> 
> I am trying to load the CSV fie, which contains invalid quote char. But in
> the results the row is inserted and values are mix with next row without
> any waring and error. Following are the details: 
*
> CSV File: 
*
> 
> name, description, salary, age, dob
> tammy,$my name$,$900000$,22,19/10/2019
*
> tammy1,$delhi$,$32345%,22,19/10/2019
*
> 
*
> tammy2,$banglore$,$543$,$44$,19/10/2019
*
> tammy3,$city$,$343$,$22$,12/10/2019
> tammy4,$punjab$,$23423$,$55$,19/10/2019
> 
> In CSV row 2 contains invalid quote char.
*
> Table :
*
> 
> create table one (name string, description string, salary double, age int,
> dob timestamp) stored by 'carbondata';
*
> Load data: 
*
> 
> load data local inpath
> 'hdfs://localhost:54310/home/harmeet/dollarquote3.csv' into table one
> OPTIONS('QUOTECHAR'="$");
*
> Actual Results: 
*
> 
> +---------+--------------+-------+-----------+------+--+
> |  name   | description  |  dob  |  salary   | age  |
> +---------+--------------+-------+-----------+------+--+
> | tammy   | my name      | NULL  | 900000.0  | 22   |
*
> | tammy1  | delhi        | NULL  | NULL      | 543  |
*
> | tammy3  | city         | NULL  | 343.0     | 22   |
> | tammy4  | punjab       | NULL  | 23423.0   | 55   |
> +---------+--------------+-------+-----------+------+--+
> 
> In the result the tammy1 value contains salary of tammy2 record.
*
> Expected Result: 
*
> 
> May be an error while reading invalid quote char  OR Just ignore the
> invalid quote char row.





--
View this message in context: 
http://apache-carbondata-mailing-list-archive.1130556.n5.nabble.com/In-load-data-CSV-row-contains-invalid-quote-char-and-results-are-invalid-tp2347p2463.html
Sent from the Apache CarbonData Mailing List archive mailing list archive at 
Nabble.com.

Reply via email to