> ## Documentation Index
> Fetch the complete documentation index at: https://knowledge.cloudquant.com/llms.txt
> Use this file to discover all available pages before exploring further.

# File Upload Management

> Upload, replace, download, and delete files behind file-upload connections without backend access

# File upload management

The **File Management** panel on a file-upload connection lets Admins manage the physical file behind a dataset: upload, replace, download, or delete — without access to underlying storage.

Supported formats for upload connections include `.csv`, `.tsv`, `.psv`, `.txt`, and `.xlsx`. See [Supported Data Formats](/datasource-config/supported-formats) for the full list.

<Note>
  Admin privileges or higher on the connection are required. The panel appears only on **File Upload** connections, not S3, SharePoint, database, or API sources.
</Note>

## Open File Management

1. Sign in to the admin portal.
2. Go to **Connections** and open the connection.
3. Scroll to **File Management**.

| Panel state            | Meaning                 | Action                                |
| ---------------------- | ----------------------- | ------------------------------------- |
| Empty (no file listed) | No file associated yet  | **Upload**                            |
| File name shown        | File already associated | **Replace**, **Download**, **Delete** |

## Upload a new file

Use when the connection has no file yet, or after a delete.

<Steps>
  <Step title="Confirm empty state">
    The panel shows no file name and an **Upload** control.
  </Step>

  <Step title="Upload">
    Click **Upload**, select the file, and wait for the progress indicator to complete.
  </Step>

  <Step title="Verify">
    Confirm the file name appears and run a small test query on the linked dataset.
  </Step>
</Steps>

If the dataset was newly created, finish column, date, and key configuration before notifying users.

## Replace an existing file

Use when data was refreshed or corrected but dataset name, columns, entitlements, and mappings should stay the same.

<Warning>
  The replacement file must use the **same format** and **same column structure** as the original. Added, removed, or renamed columns require re-onboarding instead of replace.
</Warning>

<Steps>
  <Step title="Optional download">
    Use **Download** to inspect the current file first.
  </Step>

  <Step title="Replace">
    Click **Replace**, select the new file, and wait for completion.
  </Step>

  <Step title="Verify">
    Query a value you know exists only in the new file to confirm the swap.
  </Step>
</Steps>

If **pre-generated cache** is enabled, trigger **Trigger Cache Pre-Generation** or wait for the next scheduled run so queries do not serve stale cache from the old file.

## After upload or replace

| Aspect                  | Behavior                                   |
| ----------------------- | ------------------------------------------ |
| Dataset name            | Unchanged                                  |
| Column configuration    | Unchanged — verify headers still align     |
| Entitlements            | Unchanged                                  |
| Translation mappings    | Resolve against the new file on next query |
| Previous file (replace) | Permanently removed                        |

## Troubleshooting

**Upload/Replace not visible** — Confirm Admin+ role and File Upload connection type.

**Validation error** — Check delimiter, encoding, and that format matches (`.csv` for `.csv`).

**No data after upload** — Column headers or date/key fields may not match the file; re-onboard if structure changed.

**Stale data after replace** — Refresh pre-generated cache if enabled.