ArchiveBot spike.

Gemfile

@@ -0,0 +1,9 @@
+source "https://rubygems.org"
+
+gem 'capybara'
+gem 'cinch', :git => 'https://github.com/cinchrb/cinch.git'
+gem 'poltergeist'
+gem 'redis-namespace'
+gem 'sidekiq'
+gem 'trollop'
+gem 'uuidtools'

Gemfile.lock

@@ -0,0 +1,60 @@
+GIT
+  remote: https://github.com/cinchrb/cinch.git
+  revision: a78bfad4adea7498fcfefc2f1acf84d5e70bcb9c
+  specs:
+    cinch (2.0.9)
+
+GEM
+  remote: https://rubygems.org/
+  specs:
+    capybara (2.1.0)
+      mime-types (>= 1.16)
+      nokogiri (>= 1.3.3)
+      rack (>= 1.0.0)
+      rack-test (>= 0.5.4)
+      xpath (~> 2.0)
+    celluloid (0.15.0)
+      timers (~> 1.1.0)
+    cliver (0.2.1)
+    connection_pool (1.1.0)
+    json (1.8.0)
+    mime-types (1.25)
+    mini_portile (0.5.1)
+    multi_json (1.7.9)
+    nokogiri (1.6.0)
+      mini_portile (~> 0.5.0)
+    poltergeist (1.4.1)
+      capybara (~> 2.1.0)
+      cliver (~> 0.2.1)
+      multi_json (~> 1.0)
+      websocket-driver (>= 0.2.0)
+    rack (1.5.2)
+    rack-test (0.6.2)
+      rack (>= 1.0)
+    redis (3.0.4)
+    redis-namespace (1.3.1)
+      redis (~> 3.0.0)
+    sidekiq (2.14.0)
+      celluloid (>= 0.14.1)
+      connection_pool (>= 1.0.0)
+      json
+      redis (>= 3.0.4)
+      redis-namespace
+    timers (1.1.0)
+    trollop (2.0)
+    uuidtools (2.1.4)
+    websocket-driver (0.2.3)
+    xpath (2.0.0)
+      nokogiri (~> 1.3)
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  capybara
+  cinch!
+  poltergeist
+  redis-namespace
+  sidekiq
+  trollop
+  uuidtools

archive.rb

@@ -0,0 +1,9 @@
+require 'sidekiq'
+
+class Archive
+  include Sidekiq::Worker
+
+  def perform(uri, ident)
+    # We do nothing for now!
+  end
+end

bot.rb

@@ -0,0 +1,42 @@
+require 'cinch'
+require 'redis-namespace'
+require 'trollop'
+require 'uri'
+
+require File.expand_path('../brain', __FILE__)
+
+opts = Trollop.options do
+  opt :server, 'IRC server, expressed as a URI (irc://SERVER:PORT or //SERVER:PORT)', :type => String
+  opt :nick, 'Nick to use', :default => 'ArchiveBot'
+  opt :channels, 'Comma-separated list of channels', :type => String
+  opt :schemes, 'Comma-separated list of acceptable URI schemes', :default => 'http,https'
+  opt :redis, 'Redis server for job tracking', :default => 'redis://localhost:6379'
+  opt :redis_namespace, 'Redis namespace for job tracking', :default => 'archivebot'
+end
+
+%w(server nick channels).each do |opt|
+  Trollop.die "#{opt} is required" unless opts[opt.to_sym]
+end
+
+schemes = opts[:schemes].split(',').map(&:strip)
+channels = opts[:channels].split(',').map(&:strip)
+uri = URI.parse(opts[:server])
+rconn = Redis.new(:url => opts[:redis])
+redis = Redis::Namespace.new(opts[:redis_namespace], :redis => rconn)
+
+bot = Cinch::Bot.new do
+  configure do |c|
+    c.server = uri.host
+    c.port = uri.port
+    c.nick = opts[:nick]
+    c.channels = channels
+  end
+
+  brain = Brain.new(redis, schemes)
+
+  on :message, /\A\!archive (.+)\Z/ do |m, param|
+    brain.request_archive(m, param)
+  end
+end
+
+bot.start

brain.rb

@@ -0,0 +1,50 @@
+require 'uri'
+
+require File.expand_path('../archive', __FILE__)
+require File.expand_path('../job_tracking', __FILE__)
+
+class Brain
+  include JobTracking
+
+  attr_reader :redis
+  attr_reader :schemes
+
+  def initialize(redis, schemes)
+    @redis = redis
+    @schemes = schemes
+  end
+
+  def request_archive(m, param)
+    # Do we have a valid URI?
+    begin
+      uri = URI.parse(param).normalize
+    rescue URI::InvalidURIError => e
+      reply m, "Sorry, that doesn't look like a URL to me."
+      return
+    end
+
+    # Is the URI in our list of recognized schemes?
+    if !schemes.include?(uri.scheme)
+      reply m, "Sorry, I can only handle #{schemes.join(', ')}."
+      return
+    end
+
+    # Is the job already known?
+    ident = job_ident(uri)
+    if has_job?(ident, redis)
+      reply m, "That URL is already being processed.  Use !status #{ident} for updates."
+      return
+    end
+
+    # OK, add the job and queue it up.
+    add_job(ident, redis)
+    Archive.perform_async(uri, ident)
+    reply m, "Archiving #{uri.to_s}; use !status #{ident} for updates."
+  end
+
+  private
+
+  def reply(m, *args)
+    m.reply "#{m.user.nick}: #{args.join(' ')}"
+  end
+end

job_tracking.rb

@@ -0,0 +1,20 @@
+require 'uuidtools'
+
+module JobTracking
+  ARCHIVEBOT_V0_NAMESPACE = UUIDTools::UUID.parse('82244de1-c354-4c89-bf2b-f153ce23af43')
+
+  def job_ident(uri)
+    uuid = UUIDTools::UUID.sha1_create(ARCHIVEBOT_V0_NAMESPACE, uri.to_s)
+
+    # shorten it up a bit
+    uuid.to_i.to_s(36)
+  end
+
+  def has_job?(ident, redis)
+    redis.sismember('jobs', ident)
+  end
+
+  def add_job(ident, redis)
+    redis.sadd('jobs', ident)
+  end
+end