Impselect is simple module for data selection from Impala. It supports caching, batch select and creating temporary tables.
pip install git+https://github.com/genichyar/impselect
For convenience you can create settings file .impselect.txt in user home directory. This file has JSON format.
{
"connection": {"host": "localhost", "database": "db_name"},
"tmpdir": "C:\\impselect",
"verbose": 1,
"try_except": {"timeout": 10, "count": 10}
}# First of all, we import library and create Impala object.
from impselect import Impala
i = Impala('your project name')
# Simple select.
df = i.select('SELECT * FROM users')
# Select with caching result to file.
df = i.select('SELECT * FROM users', 'task_name_1')
# Select with creating temporary table.
df = i.select('SELECT * FROM users', table_name='table_name') # Also you can use caching to file.
# Batch select.
df = i.select_batch('SELECT * FROM users WHERE day = unix_timestamp("{itervar}")',
['2017-06-04', '2017-06-05'], 'task_name_2')