Google Sheets Connector
Treats a Google Sheets spreadsheet as a database — each spreadsheet is a "schema" and
each sheet inside it is a "table".
What it lets you do
- Test that DagFlux can reach Google Sheets with your authorised account.
- List the sheets inside any spreadsheet you have access to.
- Inspect the columns of a sheet (DagFlux infers them from the first rows of the sheet).
- Run a row-fetch query against any sheet.
- Stream large sheets in batches.
- Upload data back to a new sheet.
Setup
Google Sheets uses OAuth. When you create the connection, DagFlux opens your browser at Google's
authorisation page, runs a small local listener for the redirect, exchanges the returned code for
tokens, and saves the tokens encrypted. The required scopes cover spreadsheet read/write plus the
limited Drive access needed for picking files.
Tokens auto-refresh in the background — DagFlux refreshes within a five-minute buffer of the expiry
time so a workflow run never lands on a stale token.
Behaviour notes
- A spreadsheet is identified by its spreadsheet ID; a sheet by its sheet name.
- Schema is inferred from the first few rows of the sheet (the first row is treated as the header).
- When you point a Data Source node at a Google Sheet, DagFlux imports the data into the local SQL workspace at run time so downstream nodes can query it like any other table. Column names are sanitised; duplicates are disambiguated; types are inferred from the values.
- Bucket / sheet name is used as a prefix when DagFlux creates the local table, so importing the same sheet from different spreadsheets doesn't collide.
