Documents
The Documents tab is used for file-upload type data sources. Instead of connecting to a live database or storage system, file-upload data sources are built entirely from structured data files that you upload directly into Lakehousecat.
This tab is only relevant if the data source was created with the File Upload type. For connection-based data sources (databases, lake formats, object storage), this tab is not used.
Supported File Types
Only structured, tabular data files are supported:
- CSV (comma-separated values)
- Excel (
.xlsx,.xls)
Text documents, PDFs, and unstructured content are not supported. Lakehousecat is focused exclusively on tabular data.
Adding a File
- Open the data source editor and click the Documents tab.
- Click the + button to open the upload form.
- Fill in the required fields:
| Field | Required | Description |
|---|---|---|
| File | Yes | The CSV or Excel file to upload. Click the upload area to select a file from your local machine. |
| Document Name | Yes | A unique technical identifier for this file within the data source (e.g., quarterly_sales_q3). Used internally. |
| Title | Yes | A human-readable display name for the file (e.g., Quarterly Sales Report Q3 2024). |
- Click Save to upload the file.
Managing Documents
The Documents list shows all files uploaded to this data source. From the list you can:
- Edit — Update the document name or title.
- Delete — Remove the file from the data source.
When to Re-Run Extraction
After adding, replacing, or removing files, re-run the semantic extraction from the Operations tab to update the semantic layer with the new data content.
Notes
- Each file is stored as a separate document entry within the data source.
- A data source can contain multiple files — for example, one file per reporting period.
- The file content becomes the data that the AI can query. Ensure the files are clean, well-structured, and use consistent column naming across uploads.