# first load data as csv

create table sales_data_v2

> (

> p_type string,

> total_sales int

> )

> row format delimited

> fields terminated by ',';

load data local inpath 'file:///tmp/hive_class/sales_data_raw.csv' into table sales_data_v2;

# command to create identical table

create table sales_data_v2_bkup as select * from sales_data_v2;

# describe command for a table

describe extended sales_data_v2;

# create a table which will store data in parquet

create table sales_data_pq_final

> (

> product_type string,

> total_sales int

> )

> stored as parquet;

# load data in parquet file

from sales_data_v2 insert overwrite table sales_data_pq_final select *;