How to Build an ETL App for Impala Data in Python with CData

The rich ecosystem of Python modules lets you get to work quickly and integrate your systems more effectively. With the CData Python Connector for Impala and the petl framework, you can build Impala-connected applications and pipelines for extracting, transforming, and loading Impala data. This article shows how to connect to Impala with the CData Python Connector and use petl and pandas to extract, transform, and load Impala data.

With built-in, optimized data processing, the CData Python Connector offers unmatched performance for interacting with live Impala data in Python. When you issue complex SQL queries from Impala, the driver pushes supported SQL operations, like filters and aggregations, directly to Impala and utilizes the embedded SQL engine to process unsupported operations client-side (often SQL functions and JOIN operations).

Connecting to Impala Data

Connecting to Impala data looks just like connecting to any relational data source. Create a connection string using the required connection properties. For this article, you will pass the connection string as a parameter to the create_engine function.

In order to connect to Apache Impala, set the Server, Port, and ProtocolVersion. You may optionally specify a default Database. To connect using alternative methods, such as NOSASL, LDAP, or Kerberos, refer to the online Help documentation.

After installing the CData Impala Connector, follow the procedure below to install the other required modules and start accessing Impala through Python objects.

Install Required Modules

Use the pip utility to install the required modules and frameworks:

pip install petl
pip install pandas

Build an ETL App for Impala Data in Python

Once the required modules and frameworks are installed, we are ready to build our ETL app. Code snippets follow, but the full source code is available at the end of the article.

First, be sure to import the modules (including the CData Connector) with the following:

import petl as etl
import pandas as pd
import cdata.apacheimpala as mod

You can now connect with a connection string. Use the connect function for the CData Impala Connector to create a connection for working with Impala data.

cnxn = mod.connect("Server=127.0.0.1;Port=21050;")

Create a SQL Statement to Query Impala

Use SQL to create a statement for querying Impala. In this article, we read data from the Customers entity.

sql = "SELECT City, CompanyName FROM Customers WHERE Country = 'US'"

Extract, Transform, and Load the Impala Data

With the query results stored in a DataFrame, we can use petl to extract, transform, and load the Impala data. In this example, we extract Impala data, sort the data by the CompanyName column, and load the data into a CSV file.

Loading Impala Data into a CSV File

table1 = etl.fromdb(cnxn,sql)

table2 = etl.sort(table1,'CompanyName')

etl.tocsv(table2,'customers_data.csv')

In the following example, we add new rows to the Customers table.

Adding New Rows to Impala

table1 = [ ['City','CompanyName'], ['NewCity1','NewCompanyName1'], ['NewCity2','NewCompanyName2'], ['NewCity3','NewCompanyName3'] ]

etl.appenddb(table1, cnxn, 'Customers')

With the CData Python Connector for Impala, you can work with Impala data just like you would with any database, including direct access to data in ETL packages like petl.

Free Trial & More Information

Download a free, 30-day trial of the CData Python Connector for Impala to start building Python apps and scripts with connectivity to Impala data. Reach out to our Support Team if you have any questions.

---

Full Source Code

import petl as etl
import pandas as pd
import cdata.apacheimpala as mod

cnxn = mod.connect("Server=127.0.0.1;Port=21050;")

sql = "SELECT City, CompanyName FROM Customers WHERE Country = 'US'"

table1 = etl.fromdb(cnxn,sql)

table2 = etl.sort(table1,'CompanyName')

etl.tocsv(table2,'customers_data.csv')

table3 = [ ['City','CompanyName'], ['NewCity1','NewCompanyName1'], ['NewCity2','NewCompanyName2'], ['NewCity3','NewCompanyName3'] ]

etl.appenddb(table3, cnxn, 'Customers')