09-05-2017 03:04 PM
I have created tables in hive, now i would like to download those tables in csv format, i have searched online, so i got these below solutions, but i dont understand how to use these commands on cloudera. 1. hive -e 'select books from table' | sed 's/[[:space:]]\+/,/g' > /home/lvermeer/temp.csv 2. set hive.io.output.fileformat = CSVTextFile; INSERT OVERWRITE LOCAL DIRECTORY 'dir_path' SELECT FIELD1, FIELD2, FIELD3 FROM TABLE1; If anyone knows, kindly help me out. Thanks in advance.
09-06-2017 01:08 AM
if you have hive shell or beeline - you can excute the same code nothing different
or you can try hue web ui to export the hive results as .csv
althought hue is not that good in downloading big tables .
09-09-2017 09:27 PM
You can try this as well in Unix.
beeline - u 'your jdbc connection string' - - outputformat=csv2 -e "your query here" > output.csv
Let me know if it helps.
04-05-2018 02:58 AM
I have finally found a way to do that
insert overwrite local directory 'Desktop/svk1.csv' > row format delimited > fields terminated by ',' > select * from emp;
then open the directory, just rename the file with .csv extension.
02-24-2019 05:59 AM
In case you are doing it from Windows you can use Python script hivehoney to extract table data to local CSV file.
Execute it like this:
set PROXY_HOST=your_bastion_host set SERVICE_USER=you_func_user set LINUX_USER=your_SOID set LINUX_PWD=your_pwd python hh.py --query_file=query.sql