Created
May 13, 2011 05:53
-
-
Save jamiew/970048 to your computer and use it in GitHub Desktop.
Noodling with different ways of expanding short URLs in bulk
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
require 'rubygems' | |
require 'benchmark' | |
require 'pp' | |
require 'mechanize' | |
require 'eventmachine' | |
require 'em-http-request' | |
def syncronous(urls) | |
agent = Mechanize.new | |
expanded_urls = urls.map{|url| | |
begin | |
resp = agent.head(url) | |
resp.uri.to_s | |
rescue | |
STDERR.puts "Error fetching #{url.inspect} => #{$!.inspect}" | |
end | |
} | |
return expanded_urls | |
end | |
def iterator(urls, concurrency = 10) | |
expanded_urls = nil | |
EventMachine.run do | |
responses = EM::Iterator.new(urls, concurrency).map(proc{|url,iter| | |
http = EventMachine::HttpRequest.new(url).head(:redirects => 1) | |
http.callback { iter.return(http.response_header['LOCATION']) } | |
}, proc{|responses| | |
expanded_urls = responses | |
EventMachine.stop | |
}) | |
end | |
return expanded_urls | |
end | |
def multi(urls) | |
expanded_urls = nil | |
EventMachine.run do | |
multi = EventMachine::MultiRequest.new | |
urls.each{|url| multi.add(EventMachine::HttpRequest.new(url).head(:redirects => 1)) } | |
multi.callback do | |
expanded_urls = multi.responses[:succeeded].map{|r| r.response_header['LOCATION'] } | |
EventMachine.stop | |
end | |
end | |
return expanded_urls | |
end | |
# urls = ["http://owl.li/4TAMu", "http://nyti.ms/lXobRU", "http://youtu.be/ydbOwOpyF-o", "http://joint.im/beta/r/khPZl9sn4gT", "http://www.aim.com/av/", "http://nyti.ms/kMalpD", "http://nyti.ms/kushxM", "http://4sq.com/ls4SMM", "http://bzfd.it/lCFVTs", "http://29.media.tumblr.com/tumblr_ll3cdksMCl1qz6f9yo1_500.jpg", "http://bit.ly/mdmt3N", "http://bit.ly/juekXw", "http://nyti.ms/j4FEdf", "http://kck.st/l1PnlT", "http://j.mp/5epoi", "http://j.mp/kmrpev", "http://ttk.me/t4Bm5", "http://bit.ly/lTPQFN", "http://4sq.com/lFdbD1", "http://bit.ly/l6MXft", "http://nyti.ms/lSWIPU", "http://www.psfk.com/2011/05/the-quarterlife-crisis-young-insecure-and-depressed.html?sms_ss=twitter&at_xt=4dcc927f32bf1691,0", "http://su.pr/A1eCd9", "http://instagr.am/p/EK-_5/", "http://twitpic.com/4wv5oc", "http://bit.ly/lyeHQf", "http://bit.ly/jkOi8o", "http://4sq.com/mzo07c", "http://nyti.ms/iFlKgo", "http://wp.me/p4-Hw", "http://bzfd.it/llbwwg", "http://bit.ly/mv4h50", "http://bit.ly/j7RkNE", "http://mikufes.com", "http://mikufes.com/", "http://www.anime-expo.org/?p=6818", "http://bit.ly/k6hbm4", "http://nyti.ms/iLtXjX", "http://bit.ly/kHTq9e", "http://soupsoup.net/lTB4ym", "http://ttk.me/t4Bm1", "http://knowledge.wharton.upenn.edu/article.cfm?articleid=2755", "http://nyti.ms/jjTgpd", "http://bit.ly/kEW119", "http://su.pr/A6nk05", "http://instagr.am/p/ELE55/", "http://nyti.ms/kOseS2", "http://is.gd/fn8Vyz", "http://tcrn.ch/kx2nx0", "http://bit.ly/lR4bIZ", "http://bzfd.it/ka6KT7", "http://soupsoup.net/miMftM", "http://bit.ly/kHBc9P", "http://is.gd/ztnK1C", "http://youtu.be/qwvdxV26q8I", "http://ping.fm/yDzgX", "http://bit.ly/jvmnNp", "http://chzb.gr/rollercoasterchess", "http://bit.ly/hb3UbD", "http://bit.ly/lQ6WHv", "http://www.storyful.com/stories/gjdjks", "http://vhx.tv/73589", "http://econ.st/mJjaDX", "http://strawberrymoth.blogspot.com/2011/01/inspiring-style-files-presents.html", "http://twitpic.com/4wu6ll", "http://nyr.kr/lZDT7n", "http://nyti.ms/m1CfjD", "http://amarpai.com/bikemap/bikemap.html", "http://bit.ly/bjyRwD", "http://nyti.ms/io6orz", "http://nyti.ms/mtZNEn", "http://bit.ly/k9KYwL", "http://yfrog.com/h0xq7vqj", "http://hb.ly/mqh5Y9", "http://drbl.in/bkDf", "http://fncy.it/lEy1EK", "http://4sq.com/iRoVnp", "http://www.justin.tv/hatperson", "http://www.htmlfivewow.com/", "http://aim.com/av", "http://instagr.am/p/EK8mh/", "http://4sq.com/lLQRrp", "http://storify.com/hrheingold/students-selforganize-their-own-syllabus", "http://sfy.co/8Wm", "http://on.fb.me/kayJ2u", "http://frc.vc/3eN", "http://nyti.ms/ljFiEw", "http://bit.ly/kCjzKp", "http://twitpic.com/4wtowc", "http://nyti.ms/jViJ3s", "http://nyti.ms/iB1o3b", "http://bzfd.it/kTfJdD", "http://lockerz.com/s/101093774", "http://thepulsenetwork.com/technology/stevegarfield-tv-on-tpn/05-12-11-this-week-in-steve/", "http://thepulsenetwork.com/technology/stevegarfield-tv-on-tpn/05-12-11-video-soup/", "http://thepulsenetwork.com/technology/stevegarfield-tv-on-tpn/05-12-11-the-killing-of-bin-laden/"] | |
urls = ["http://owl.li/4TAMu", "http://nyti.ms/lXobRU", "http://youtu.be/ydbOwOpyF-o", "http://joint.im/beta/r/khPZl9sn4gT", "http://www.aim.com/av/", "http://nyti.ms/kMalpD", "http://nyti.ms/kushxM", "http://4sq.com/ls4SMM", "http://bzfd.it/lCFVTs", "http://bit.ly/juekXw"] | |
puts "Processing #{urls.length} URLs..." | |
puts "\nSyncronously..." | |
output0 = [] | |
time = Benchmark.realtime{ output0 = syncronous(urls).compact.sort } | |
pp output0.sort | |
puts "Took #{time}s (#{(time*1000.0/urls.length.to_f).to_i}ms/url)" | |
sleep 2 | |
puts "\nEM::Iterator(2)..." | |
output1 = [] | |
time = Benchmark.realtime{ output1 = iterator(urls, 2).compact.sort } | |
pp output1.sort | |
puts "Took #{time}s (#{(time*1000.0/urls.length.to_f).to_i}ms/url)" | |
sleep 2 | |
puts "\nEM::Iterator(10)..." | |
output2 = [] | |
time = Benchmark.realtime{ output2 = iterator(urls, 10).compact.sort } | |
pp output2.sort | |
puts "Took #{time}s (#{(time*1000.0/urls.length.to_f).to_i}ms/url)" | |
sleep 2 | |
puts "\nEm::MultiRequest..." | |
output3 = [] | |
time = Benchmark.realtime{ output3 = multi(urls).compact.sort } | |
pp output3.sort | |
puts "Took #{time}s (#{(time*1000.0/urls.length.to_f).to_i}ms/url)" | |
sleep 2 | |
puts "\nDone!" | |
puts "output0.length=#{output0.length} output1.length=#{output1.length} output2.length=#{output2.length} output3.length=#{output3.length}" | |
puts "0==1?#{output0 == output1} 1==2?#{output1 == output2} 2==3?#{output2 == output3}" |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment