Extract Service

Shared service to extract data out of websites.

API

You MUST use POST /extract to request extracts. We do not serve GET requests because we want to discourage people hotlinking to extracts, and also to control bot traffic. If you try to GET /extract it returns HTTP 405 Method Not Allowed.

Headers

• X-Forwarded-For — The remote client address making the request. This allows the extract service to log the address.
• User-Agent — The remote user-agent of the client making the request. This value is overridden by the ua parameter, if present.

POST /extract

Default	Required	Type
null	yes	N/A

Required Parameters

url

String representing the URL you want to extract data from.

Default	Required	Type
null	yes¹	String

The URL must be valid. The protocol must be included. You may not include authentication in the URL (see user/pass parameters for HTTP Basic Auth).

element

String representing the element to extract.

Default	Required	Type
null	yes¹	String

attribute

String representing the attribute to extract.

Default	Required	Type
null	yes¹	String

Optional Parameters

selector

Specify a CSS selector. Send something very specific, such as an #html-id. If you send a generic selector that matches many elements on your page, then extract Service will only return the FIRST element that matches your selector.

Default	Required	Type
null	no	String