Ebot is a scalable and distribuited Web crawler. The URLs are saved to a NOSQL database (which supports map/reduce queries) that you can query via RESTful HTTP requests or using your preferred programming languages. The URLs that need to be analyzed are sent to AMQP queues. In this way, it is possible to run several crawlers in parallel and stop and start them without losing URLs.
|Tags||crawler Web bot Erlang couchdb mochiweb rabbitmq amqp NoSQL riak webmachine|
|Operating Systems||OS Independent|
|Implementation||Erlang mochiweb rabbitmq apache couchdb webmachine riak|
Release Notes: A better plugin architecture. New plugins for saving image URLs and titles of HTML pages.
Release Notes: This release was updated to the latest releases of rabbitmq (2.2.0) and couchbeam. Some bugs were fixed.
Release Notes: Compatibility with Erlang R14A. Tested with Debian Testing.
Release Notes: Compatibility with the latest (development) releases of required libraries and software (rabbitmq, couchdb, webmachine, riak, etc.)
Release Notes: For better scalability, new amqp queues are used (ebot.new.* ebot.fetched.* ebot.completed.* ebot.refused.*) and the old core of the crawler (ebot_web) is now split in two different and parallel modules/processes (ebot_html and ebot_web).