Using the CData ODBC Driver for Cohere in PyCharm

The CData ODBC Drivers can be used in any environment that supports loading an ODBC Driver. In this tutorial we will explore using the CData ODBC Driver for Cohere from within PyCharm. Included are steps for adding the CData ODBC Driver as a data source, as well as basic PyCharm code to query the data source and display results.

To begin, this tutorial will assume that you have already installed the CData ODBC Driver for Cohere as well as PyCharm.

Add Pyodbc to the Project

Follow the steps below to add the pyodbc module to your project.

1. Click File -> Settings to open the project settings window.
2. Click Project Interpreter from the Project: YourProjectName menu.
3. To add pyodbc, click the + button and enter pyodbc.
4. Click Install Package to install pyodbc.

Connect to Cohere

You can now connect with an ODBC connection string or a DSN. See the Getting Started section in the CData driver documentation for a guide to creating a DSN on your OS.

Start by setting the Profile connection property to the location of the Cohere Profile on disk (e.g. C:\profiles\Cohere.apip). Next, set the ProfileSettings connection property to the connection string for Cohere (see below).

Cohere API Profile Settings

Obtain your API key from your Cohere account at dashboard.cohere.com by navigating to the API Keys section.

Below is the syntax for a DSN:

[CData API Source]
Driver = CData ODBC Driver for Cohere
Description = My Description
Profile = C:\profiles\Cohere.apip
ProfileSettings = 'APIKey = your_api_key'

Execute SQL to Cohere

Instantiate a Cursor and use the execute method of the Cursor class to execute any SQL statement.

import pyodbc

cnxn = pyodbc.connect('DRIVER={CData ODBC Driver for API};Profile = C:\profiles\Cohere.apip;ProfileSettings = 'APIKey = your_api_key';')
cursor = cnxn.cursor()
cursor.execute("SELECT Id, ConversationId FROM ChatMessages WHERE ConversationId = 'abc123'")
rows = cursor.fetchall()
for row in rows:
  print(row.Id, row.ConversationId)

After connecting to Cohere in PyCharm using the CData ODBC Driver, you will be able to build Python apps with access to Cohere data as if it were a standard database. If you have any questions, comments, or feedback regarding this tutorial, please contact us at [email protected].