dreamcatcher.rb

#! /usr/bin/ruby

## Elliott Kember's Delete MM_ Script.
# I bet you thought I wouldn't write this!

## Usage: 
# $ ruby dreamcatcher.rb emailaddress password

## dreamcatcher.rb

# What it does:
# It checks all your new emails for links which contain "MM_" in their HTML.
# That means they use Dreamweaver.
# That means you don't want their emails, so we delete them.

# We need some things from the grocery store
require 'net/imap'
require 'net/http'
require 'rubygems'
require 'mms2r'
require 'activesupport' # For TMail.

# And some settings
config = {}
config['username'] = ARGV[0] # => 'elliott.kember@gmail.com'
config['password'] = ARGV[1] # => 'secret'
config['timeout'] = ARGV[2].nil? ? 60 : ARGV[2].to_i  # seconds
config['host'] = ARGV[3] || 'imap.gmail.com'
config['port'] = ARGV[4] || '993'

latest_uid = nil

# Starting time to check from
t = (Time.new - 120)

# Keep doing this forever.
while true do
  
  # Let em know what we're doing.
  puts "Begin MM_ scanning..."
  
  # Connect and log in to imap
  imap = Net::IMAP.new(config['host'], config['port'], true)
  imap.login(config['username'], config['password'])
  imap.select('Inbox')
  
  # Search for not deleted
  
  if !latest_uid.nil?
    puts "Using latest uid"
    uids = imap.uid_search("#{latest_uid}:*, NOT SEEN")
    uids.delete latest_uid
  else
    uids = imap.uid_search(["NOT", "SEEN", "SINCE", t])
  end
  
  latest_uid = uids.max if uids.length > 0
  puts "Latest uid: #{latest_uid}"
  
  t = Time.new # Next time, we'll go from now.
  
  puts "Checking emails: #{uids.join(',')}"
  
  if uids.length > 0
    uids.each do |uid|
      # Has to be an integer
      uid = uid.to_i
      # Find the email
      source   = imap.uid_fetch(uid, 'RFC822')
      if source.nil?
        next # It's gone, forget it
      else
        # Get it
        source = source.first.attr['RFC822']
        # Parse it
        mail            = TMail::Mail.parse(source)                               # TMail object for some stuff
        # This is where MMS2R comes in handy - plain text body content.
        mms             = MMS2R::Media.new(mail)                                  # MMS object for plain-text content
        plain_body_text = mms.body                                                # See?
        # Regex for finding URLs
        url_regexp = /http[s]?:\/\/\w/
        # Find 'em
        urls = plain_body_text.split.grep(url_regexp)
        # Check each one
        delete_this_email = false
        
        # Check each url for MM_
        urls.each do |url|
          url = url.gsub(/<|>/, '')
          # Downloadify it
          r = Net::HTTP.get_response(URI.parse(url).host, URI.parse(url).path)
          # Check it for MM_
          if r.body.include? 'MM_'
            puts "Found #{uid} - #{mail.subject} from #{mail.from} - deleting that sucker."
            delete_this_email = true
            # I've seen enough. GTFO.
            break
          else
            # Good, clean email.
            puts "#{url} is clean."
          end
        end
        
        # If we're deleting this email, delete it! What a worthless comment.
        if delete_this_email
          imap.uid_store(uid, "+FLAGS", [:Deleted])
          puts "Deleting that email. Spammers!"
        else
          puts "Email was clean. Next!"
        end
        
      end
    end
  else
    puts "Nothing found. For now. Trying again in #{config['timeout']} seconds."
  end
  
  puts "Done for now."
  
  # Log out so we can reconnect, otherwise this gets cached.
  # This seems really awful, but keeping the connection open got messy.
  # GMail seems to cache things between searches which makes things hard.
  imap.logout
  
  # Sleep for x seconds so GMail doesn't hate us.
  sleep config['timeout']
end