Muck Raker

Muck Raker contains the jar files and rake tasks for running deamons to aggregates and recommend data. See muck-services for details.

Installation

Add to Gemfile:

gem '

Usage

Muck raker aggregates data feeds and analyzes the results to make recommendations.

Bring in required files - migrations etc:

muck:raker:sync - Copy muck-raker migrations, solr config files and initializer file into your application.

Running the Muck Raker Daemon

muck:raker:start - Start muck-raker daemon running continuously. This task forks and then checks for stale feeds every
60 seconds. When a stale feed is detected, it harvests the feed. If resources are added, updated, or deleted, it
indexes the new resources, generates recommendations for the new resources, and generates tags if needed. Once a week
the daemon updates recommendations for all resources and generates new tag clouds.

muck:raker:stop - Stop the muck raker daemon process.

Running Individual Steps

muck:raker:start_redo - Redo everything and quit. This task harvests, updates recommendations for all resources,
generates tags if needed, and generates new tag clouds. It does not fork. It stops after the one pass through.

muck:raker:harvest - Harvest stale feeds. Each feed has a "harvest_interval" that specifies how often the daemon
should harvest it to see if it has changed.. After that time period has elapsed the feed is considered stale. By
default that time period is 1 day.

muck:raker:index - Index new resources, delete resources that were deleted from the database, and update the index
for resources that were updated in the database.

muck:raker:index_redo - Updates the indexes for all resources.

muck:raker:recommend - Incrementally update recommendations (create recommendations for newly harvested records).

muck:raker:recommend_redo - Redo recommendations for all resources.

muck:raker:subjects - Auto-generate tags for newly harvested resources that don't have at least 4.

muck:raker:tag_clouds - Regenerate tag clouds.