retrieve_catalog_metadata.rb

#!/usr/bin/env ruby
#

require 'httparty'
require 'json'

results = HTTParty.get("http://data.cityofchicago.org/api/search/views.json?limitTo=TABLES")

# Category -> [ Dataset, Dataset ]
by_categories = Hash.new { |h,k| h[k] = Array.new } 
results["results"].each do |dataset|
  dataset = dataset["view"]

  size = dataset["columns"].collect { |c| c["cachedContents"]["null"] + c["cachedContents"]["non_null"] }.max
  by_categories[dataset["category"]] << {
    "url" => "https://data.cityofchicago.org/d/#{dataset["id"]}",
    "name" => dataset["name"],
    "size" => size,
    "value" => 42 # Not sure how this is calculated
  }
end

# Formulate the final hash
output = {
  "name" => "Chicago Data Portal",
  "children" => by_categories.collect { |cat,ds| { "name" => cat, "children" => ds } }
}

puts JSON.pretty_generate(output)