HTTP File

The HTTP File connector lets Weld fetch a single file from any HTTPS URL and load it into your destination schema — no server, storage credentials, or file-transfer protocol required. Point Weld at a URL that returns a CSV or NDJSON file, such as a data export endpoint or a report generated by another system, and Weld will download it on each sync and load it as a table.

Unlike file-transfer connectors such as FTP and SFTP, the HTTP File connector does not browse folders or match multiple files with a regex pattern — it always downloads the exact URL you configure, and only supports Full sync.

🔧 Setup Guide

Step 1 - Begin Configuration

  1. After selecting Data Sources in the side menu, click + New on the top right.
  2. Find the HTTP File connector from the list and click on it.
  3. In the connector setup form, enter the connection name of your choice.
  4. Fill in the required field:
  • File URL - the URL Weld will fetch the file from on each sync. The URL must use https://.
  1. Optionally, expand Add authentication header if the URL is gated behind an API key or token, and fill in:
  • Header name - the name of the HTTP header Weld will send with the request (e.g. Authorization).
  • Header value - the value of that header (e.g. Bearer <token>).

Both the header name and header value must be provided together, or left empty together.

  1. Click Connect. Weld will make a request to the URL (and abort it as soon as the response headers are received) to confirm it's reachable and, if configured, that the authentication header is accepted.

Step 2 - Data to sync

For the stream created from this connection, fill in:

  • Stream name - the name Weld will use for this stream/table.
  • File type - the format of the file at the configured URL:
    • CSV - parsed automatically based on the file's delimiter.
    • NDJSON - newline-delimited JSON, one JSON object per line.

Weld reads the first row of the file to build a preview of the resulting table, so you can review the schema and remove columns before confirming.

Full sync only:

The HTTP File connector always downloads and reloads the entire file on every run — incremental sync and history tables are not supported, since there is no reliable way to detect only new or changed rows from a flat file at a static URL.

Step 3 - Configure sync

  1. Select how often you would like the data to sync.
  2. Provide a unique destination schema name.

Weld will take over from here and commence syncing data from your configured URL.

Was this page helpful?