opml_to_li.rb

# Takes an OPML file from google reader and spits out a set of <li>s for
# each unique blog in the list, sorted by name, with the tags in ()s.
require 'rubygems'
require 'nokogiri'
require 'open-uri'

def get_tag(opml_node)
  if opml_node.nil?
    nil
  elsif opml_node['htmlUrl'].nil?
    opml_node['title']
  else
    get_tag(opml_node.parent)
  end
end

opml_file = ARGV[0]
relevant_tags = ARGV[1..-1]
blogs = {}

doc = Nokogiri::XML(File.read(opml_file))
doc.xpath('//outline').each do |entry|
  name = entry['title']
  url = entry['htmlUrl']
  if url
    blogs[url] ||= {:name => name, :tags => []}
    blogs[url][:tags] << get_tag(entry)
  end
end

unless relevant_tags.empty?
  blogs.each_key do |k|
    blogs[k][:tags].reject! { |tag| ! relevant_tags.include?(tag) }
  end
  blogs.delete_if { |key, value| value[:tags].empty? }
end

puts "<ul>"
blogs.keys.sort_by { |k| blogs[k][:name].downcase.sub(/^the /,'').gsub(/[^\w\d]/, '') }.each do |url|
  blog = blogs[url]
  s = " <li><a href=\"#{url}\">#{blog[:name]}</a>"
  s << " (#{blog[:tags].join(', ')})" unless blog[:tags].empty?
  s << "</li>"
  puts s
end
puts "</ul>"
puts <<SUMMARY
<div>Count: #{blogs.size}</div>
<div>This list was produced by downloading <a href="http://www.google.com/reader/subscriptions/export">google reader's OPML listing of my subscriptions</a> and running this:<br/>
<code>ruby <a href="http://gist.github.com/135314">ompl_to_li.rb</a> #{ARGV.join(' ')}</code>
</div>
SUMMARY