This is an old revision of the document!


Hive

Migrate CSV to hive table

DROP TABLE IF EXISTS schema.tbl_external_csv;
DROP TABLE IF EXISTS schema.tbl_internal;
 
--
-- Create external csv reader
--
CREATE EXTERNAL TABLE IF NOT EXISTS schema.tbl_external_csv(
    name STRING,
    guid STRING,
    src_date String,
    KEY INT)
COMMENT 'tbl_internal external table'
ROW FORMAT DELIMITED
FIELDS TERMINATED BY ','
STORED AS TEXTFILE
LOCATION '/user/raw_user/raw_files'
tblproperties("skip.header.line.count"="1");
 
--
-- Create tbl_internal
--
CREATE TABLE IF NOT EXISTS schema.tbl_internal (
    name STRING,
    guid STRING,
    `date` TIMESTAMP,
    KEY INT
) STORED AS parquet
"""
 
 
 
hiveContext.sql("TRUNCATE TABLE `schema`.`tbl_internal`")
hiveContext.sql("""
    INSERT OVERWRITE TABLE `schema`.`tbl_internal`
    SELECT
        `client_name`,
        `topguid`,
        from_unixtime(unix_timestamp(`src_date`, 'dd/MM/yyyy'),'yyyy-MM-dd') AS `date`,
        `key`
    FROM `schema`.`tbl_external_csv`
kb/hive.1511173415.txt.gz · Last modified: (external edit)
Back to top
Driven by DokuWiki Recent changes RSS feed Valid CSS Valid XHTML 1.0