Skip to main content
New Vectors only

​
Overview

Google Drive is Google’s cloud file storage. Datazone signs in with a Google Cloud service account and indexes the documents in the folders and shared drives you share with it into a Vector. Access is read-only (drive.readonly scope).
Google Drive sources hold documents, not tables. They are used by Vectors only — they have no extracts and no schema to infer.

​
Prerequisites

Create a service account, enable the Google Drive API, and share the content you want to index with the service account. See How to get the credentials below.

​
Connection Parameters

Set on the source when you create it. Names match the API and CLI payload.
​
string
required
Name — a unique identifier for your Google Drive source.
​
string
required
Service Account Key JSON — the full contents of the service account’s JSON key file. Stored encrypted.

​
How to get the credentials

1

Create or pick a Google Cloud project

In the Google Cloud console, select a project or create a new one.
2

Enable the Google Drive API

Go to APIs & Services → Library, search for Google Drive API, and click Enable.
3

Create a service account

Go to IAM & Admin → Service Accounts → Create service account. Give it a name such as datazone-drive. It needs no project roles.
4

Create a JSON key

Open the service account, go to Keys → Add key → Create new key → JSON. A .json file is downloaded. Paste its entire contents into Service Account Key JSON in Datazone.
5

Share your content with the service account

Copy the service account’s email (it looks like datazone-drive@my-project.iam.gserviceaccount.com) and share with it, as Viewer:
  • a folder or file in My Drive: right-click → Share, or
  • a shared drive: Manage members → add the email.
The key file grants access to everything shared with the service account. Treat it like a password, and don’t commit it to a repository. If your organization blocks service account key creation (the iam.disableServiceAccountKeyCreation policy), ask a Google Cloud admin to allow it for your project.

​
Using the source in a Vector

When you create a Vector, choose a connected source as the data source, then pick a path. The vector stores them as source and source_path, and both must be set together. The top level contains the shared drives the service account belongs to and the items shared with it. A path is the folder names joined by /: The path picker in the Datazone UI lets you browse the folders instead of typing the path.

​
Limitations

  • Only content shared with the service account is visible. A folder that is not shared will give not found or not shared with the service account.
  • Subfolders are indexed recursively.
  • Supported file types: PDF, CSV, DOCX, XLSX, XLS, MD, TXT. Other files are skipped.
  • Google Docs are exported to DOCX and Google Sheets to XLSX before indexing. Other Google-native files, such as Slides, are skipped.
  • Names must match exactly. If two folders at the same level share a name, the first one found is used.

​
Troubleshooting