From: Luka Stolyarov <lukich@...>
Date: 2010-05-31T07:27:48+09:00
Subject: Trouble with timeout

Hi.  I have a small script that I use to download archives from the web.
The code looks like this:

require 'rubygems'
require 'mechanize'

def downloader()
  agent = WWW::Mechanize.new

  # create agent object
  page = agent.get('http://www.myurl.com')

  #authenticate
  form = page.forms.first
  form.username = '####'
  form.password = '####'

  #submit form
  page = agent.submit form

  #grab all links that have zip in them
  mylinks = page.links_with(:href => /zip\//)

  #for each link download and rename
  mylinks.each do |archive|

    name = archive.href.split('zip/')[1].sub('/','').strip

    puts "Archive #{name} is saving..."

    myfile = agent.click(archive)

    output = File.open("/Users/luka/Desktop/#{name}.zip", 'w') {|file|
file << myfile.body }

    puts "Done!"

  end
end

begin
  downloader
rescue Timeout::Error
  puts 'Timeout was detected.  Trying again...'
  downloader
end


However, I've been having problems with timeouts from time to time, so I
ended up making this code into a function and upon timeout invoke this
function once again, as you can see at the bottom of the code. However,
I still get an error, which looks like this on timeout:


Archive 914188 is saving...
Timeout was detected.  Trying again...
/System/Library/Frameworks/Ruby.framework/Versions/1.8/usr/lib/ruby/1.8/timeout.rb:60:in
`rbuf_fill': execution expired (Timeout::Error)

So it tries to rescue the timeout but obviously calling the same
function again is a bad idea.  I was wondering why is it, and what would
be a proper solution.
Thanks!

Luka
-- 
Posted via http://www.ruby-forum.com/.