Databricks Google Sheets Connector

0.8.0 verified Mon Apr 27 auth: no python

A Python library for reading and writing Google Sheets data from Databricks notebooks. Version 0.8.0 supports Python >=3.6. It wraps the Google Sheets API and provides a simple DataFrame-centric interface. Releases are infrequent; no breaking changes recently announced.

pip install databricks-google-sheets

Common errors

error ModuleNotFoundError: No module named 'databricks_google_sheets' ↓

cause Package not installed or installed under incorrect name (e.g., 'databricks-google-sheets' in pip list).

fix

Install via pip: pip install databricks-google-sheets. Import with underscores.

error GoogleSheetsClient.get_credentials() missing required positional argument: 'service_account_key' ↓

cause Did not provide the service account key JSON file path.

fix

Pass a valid path to a service account key JSON file: GoogleSheetsClient.get_credentials(service_account_key='/path/to/key.json')

error AttributeError: 'ServiceAccountCredentials' object has no attribute 'to_json' ↓

cause Incompatible version of google-auth or gspread. The library expects an older interface.

fix

Pin google-auth to <=1.35.0 and gspread to <=3.7.0: pip install google-auth==1.35.0 gspread==3.7.0

Warnings

gotcha The library expects a service account key file path, not JSON string. Use absolute path or Databricks secret scope. ↓

fix Store the key file in DBFS or use Databricks secrets with dbutils.secrets.get() and write to a temp file.

gotcha Google Sheets API quotas are low for free tier. Frequent writes may exceed limits. ↓

fix Implement rate limiting or batch updates using gspread directly if needed.

deprecated The method `write_to_sheet` expects a DataFrame; passing a list may cause silent errors. ↓

fix Always convert data to pandas DataFrame before calling write_to_sheet.

Imports

GoogleSheetsClient

wrong

from databricks.google_sheets import GoogleSheetsClient

correct

from databricks_google_sheets import GoogleSheetsClient

The correct package uses underscores, not dots.

Quickstart

Initialize client with service account credentials and read a sheet into a DataFrame.

from databricks_google_sheets import GoogleSheetsClient
import os

# Set credentials via environment variables or Databricks secrets
scope = ['https://www.googleapis.com/auth/spreadsheets']
creds = GoogleSheetsClient.get_credentials(
    service_account_key=os.environ.get('GOOGLE_APPLICATION_CREDENTIALS', '')
)
client = GoogleSheetsClient(creds)

# Read a sheet
url = 'https://docs.google.com/spreadsheets/d/your-sheet-id/edit'
df = client.read_from_url(url, sheet_name='Sheet1')
print(df.head())