# Snowflake

[Snowflake](https://www.snowflake.com/) can either be used as a data
source or a data destination in Zuar Runner.

.. image:: assets/snowflake-1.png
   :alt: Zuar Runner and Snowflake

Source plugin example: [Query](/jobs/query/)

Destination plugin examples: [CSV](/jobs/csv/),
[Salesforce](/connectors/salesforce/),
[SQL](/jobs/sql/)

## Using Snowflake as a Data Destination

Note the following considerations when sending Zuar Runner job output to Snowflake.

Zuar Runnner will automatically:

- Create the Snowflake database schema if it doesn't exist
- Create the Snowflake database tables if they don't exist
- Determine data types for Snowflake columns
- Add new columns to Snowflake tables based on new fields in source data
- Adjust Snowflake tables based on changes in source data

## Snowflake Database URL Setup

The [database url](/databases/database-urls/) structure
for connecting to a Snowflake database uses this format:

`snowflake://<user_login_name>:<password>@<account_identifier>/<database>?warehouse=<warehouse_name>`

- `user_login_name`: The username Runner uses to authenticate to your Snowflake instance
- `password`: The password of the Snowflake user specified above
- `account_identifier`:
  - After logging into your Snowflake account console in a web browser,
  the account identifier should be found in the browser URL here:
  **https://`<account_identifier>`.snowflakecomputing.com**
  - Do *not* include the snowflakecomputing.com domain name as part of your account identifier.
  - More information on determining your [account identifier](https://docs.snowflake.com/user-guide/admin-account-identifier).
- `database`: Name of the database where Runner should write job output
- `warehouse_name`: Name of the warehouse that contains the above database

At a mimimum, we recommend including all of the components above. You can also include `schema` and `role`
if the defaults for the user account are not appropriate to your specific needs using this format:

`snowflake://<user_login_name>:<password>@<account_identifier>/<database_name>/<schema_name>?warehouse=<warehouse_name>&role=<role_name>`

`schema` is also an option in the Runner job configuration. Use this option if you need
to change schemas on a per-job basis.

.. NOTE::

   When outputting to a Snowflake database, leaving the `schema` blank
   will create a table in the **public** schema.

Reference: [Using Snowflake via the Python connector](https://docs.snowflake.com/developer-guide/python-connector/sqlalchemy#required-parameters).

Here's an example of using a Snowflake database as a destination in a
[CSV](/jobs/csv/) job:

.. image:: assets/snowflake__output.png
   :alt: CSV JOB

## SQL

Zuar Runner can send [SQL](/jobs/sql/) statements to a Snowflake
database. Use [Snowflake SQL syntax](https://docs.snowflake.com/en/sql-reference-commands) in these Zuar Runner SQL jobs.

## Cheat Sheet

The Zuar team has ample experience working with Snowflake. The team
has created this cheat sheet to help you get started: [Snowflake Cheat
Sheet](https://www.zuar.com/blog/snowflake-cheat-sheet/)