Fetch from githubarchive.org and drop into MongoDB...

gh_ark.rb

require 'open-uri'
require 'zlib'
require 'yajl'
require 'mongo'
require 'date'
require 'iconv'

include Mongo

@client = MongoClient.new('localhost', 27017)
@db = @client['gh_ark']
@coll = @db['events']
@coll.remove

sd = Date.parse('2012-01-03')
ed = Date.parse('2012-01-04')

# Loop over the days
sd.upto(ed) do |date|
  url = "http://data.githubarchive.org/#{date}-23.json.gz"
  puts "Getting #{url}"
  gz = open(url)
  js = Zlib::GzipReader.new(gz).read
  ic = Iconv.new('UTF-8//IGNORE', 'UTF-8')
  js = ic.iconv(js + ' ')[0..-2]

  Yajl::Parser.parse(js) do |event|
    puts "inserting events for #{date}..."
    puts event
    @coll.insert(event)
  end
end
puts "There were #{@coll.count} events."