gistfile1.rb

#!/usr/bin/env ruby
require 'rexml/document'
require 'net/http'
require 'time'

def get_tweets(user,pass,page)
  Net::HTTP.start('www.twitter.com') {|http|
      req = Net::HTTP::Get.new("/statuses/user_timeline.xml?count=200?page=#{page}")
      req.basic_auth user, pass
      response = http.request(req)
    }
end

if ARGV.size == 0
  puts "Usage: savetweets.rb <twitter username> <twitter password>"
  exit
end

doc = REXML::Document.new 
statuses = REXML::Element.new("statuses")
statuses.attributes["type"] = "array"
doc.elements << statuses

#http://apiwiki.twitter.com/Twitter-REST-API-Method%3A-statuses-user_timeline
#count.  Optional.  Specifies the number of statuses to retrieve. May not be greater than 200
#Clients may request up to 3,200 statuses via the page and count parameters for timeline REST API methods. 
page = 0
loop do
  page += 1
  warn "loading page #{page}"
  page_response = get_tweets(ARGV[0], ARGV[1], page)
  begin
    page_doc = REXML::Document.new page_response.body
    error = page_doc.root.elements["hash/error"]
    if error
      raise error.text
    end
    statuses = page_doc.root.elements["statuses"]
    statuses.elements.each do |status|
      statuses << status
    end
    warn "got #{statuses.size} tweets, #{doc.root.elements["statuses"].size} total"
  rescue REXML::ParseException
    puts "error parsing #{page_response.body}"
    break
  rescue RuntimeError=> e
    puts "#{e}"
    break
  end
end

statuses = doc.root.elements["statuses"]
puts "#{statuses.size} tweets total"
if statuses.size > 0
  start_date = Time.parse(statuses.elements[1].elements["created_at"].text)
  end_date = Time.parse(statuses.elements[statuses.elements.size].elements["created_at"].text)
  filename = "#{start_date.strftime("%Y-%m-%d")}-#{end_date.strftime("%Y-%m-%d")}.twitter.xml"

  warn "Saving tweets to #{filename}"
  File.open(filename, "w") do |f|
    f.write "#{doc}"
  end
else
  warn "No tweets retrieved."
end