The Harvester is a web-based Feed-Aggregator
Running instances can be seen at: blog-harvester.de or rubynetz.de
The Harvester eats the feeds you want and produces a static html/feed page that aggregates all those.
Harvester 0.8 is alpha quality! There are still some unfixed bugs (e.g. database stuff)
Install the harvester with
gem install harvester --pre
You can now create a new harvester project with
harvester new <project_directory>
Then you will need to configure config.yaml, collections.yaml and the templates to suit your needs (see below). You also have to run harvester db create (in that directory) to create your initial database structure.
config.yaml contains general configuration, like paths, logs and timeouts. You can also configure your database there. The default database backend is SQLite3, but all the following ones are supported:
-
SQLite3
-
PostgreSQL
-
MySQL
The collections.yaml file contains the links to your desired feeds.
Then run
harvester run
in the project directory or run
harvester run -c path/to/config.yaml
This will create the output files. That’s it ;).
Usually, you want to run this command automatically every x seconds/minutes. You can use a program like cron or simply run
harvester clock <time>
to start a simple scheduler.
There some more harvester commands to explore. Run
harvester
to get a command list.
…
-
Still some things broken after update
-
Fix database issues
-
Fix wrong tweet escaping (sometimes: ä–>ä)
-
Improve/fix feed parsing (especially atom)
-
Fix chart generation
-
Tidy up templates / remove astro-specific links
-
Extract mrss.rb into extra gem
-
Security issues
-
Optimize performance
-
Fix jabber bot
-
Implement some kind of tag filters