rubygems-search

A simple Rails interface to search through RubyGems data. Proof of concept for RubyConf 2013

A running example is deployed here - it will be down after RubyConf. Be nice - it's a small instance running ES/Mongo/Rails - don't break it :)

Install

Download the repo && run bundle install.

You need to have MongoDB running on localhost (port 27017) and ElasticSearch (port 9200). The app will try to connect on MongoDB, db=rubygems and collection=gems

Data

The project comes with no data. You need to populate your MongoDB using the RubyGems Crawler running on your local machine available here or by downloading an existing MongoDB dump available here

You can test your data is properly loaded using the mongo console:

$ mongo localhost:27017/rubygems
MongoDB shell version: 2.2.0
connecting to: localhost:27017/rubygems
> db.gems.count()
64741
>

Add MongoDB Indexes

This application defines a few indexes to navigate through MongoDB results. You need to install a few indexes and you can do that by running the following command:

$ rake db:mongoid:create_indexes RAILS_ENV=development I, [2013-11-08T08:08:38.704069 #18230] INFO -- : MONGOID: Created indexes on RubyGem: I, [2013-11-08T08:08:38.704215 #18230] INFO -- : MONGOID: Index: {:name=>1}, Options: {:unique=>true, :background=>true, :dropDups=>true}

Run the application

$ rails s

and connect to http://localhost:3000/

Search from command line

You can execute searches using Rails command line ('rails c'). The RubyGem class support 2 search methods:

RubyGem.simple_search(search_terms, options={}) RubyGem.advanced_search(search_conditions, options={})

The available options are:

:page Defines the results page number (for pagination) :per Defines the results page size (number of results per page) :facets (True/False) Include facets in search results

To use the 'simple_search', just pass a search string to the 'search_terms'. For example:

RubyGem.simple_search('twitter AND bootstrap NOT rails', page: 2, per: 10).results

To use the 'advanced_search' pass an hash of search terms to the 'search_conditions'. The available keys are:

:name, :info, :owners, :authors, :licenses, :version, :runtime_dependencies, :dev_dependencies

For example:

RubyGem.advanced_search({runtime_dependencies: 'tor'}, facets:true).results