py-hiverunner

Library provides python API for Klarna's HiveRunner.

Install

Install py-hiverunner package with pip:

$ python -m pip install py-hiverunner

Usage

Before using py-hiverunner you need start JVM with facade service based on original Klarna's HiveRunner - java-hiverunner. This repo provides docker container for Py4J Java Gateway for HiveRunner.

You can pull this:

$ docker pull la9ran9e/py-hiverunner

and then run the container:

$ docker run -ti -p 25333:25333 -p 25334:25334 la9ran9e/py-hiverunner

After that you will have working server with Java HiveRunner.

Try this:

from py_hiverunner import hiverunner
from pprint import pprint


with hiverunner() as hive:
    hive.execute_query("create schema meh")
    hive.execute_query("create table meh.nonsub(a int, b string, c array<string>)")
    hive.execute_query("insert into meh.nonsub select 1, 'la9ran9e', array('1', 'a', 'b', '6')")
    hive.execute_query("insert into meh.nonsub select 2, 'la9ran9e', array('1', 'b', 'b', '6')")
    hive.execute_query("insert into meh.nonsub select 3, 'la9ran9e', array('1', 'c', 'b', '6')")
    hive.execute_query("insert into meh.nonsub select 4, '', array('1', 'd', 'b', '6')")

    hive.execute_query("create table meh.sub(a int, b string, c boolean)")
    hive.execute_query("insert into meh.sub select 1, 'la9ran9e', true")

    print("RESULT:")
    pprint(hive.execute_query("""
        select
            *
        from
            meh.sub as sub
        inner join
            meh.nonsub as nonsub
        on
            sub.b = nonsub.b
    """))