Question

I’m using some ruby code wrapped in a begin – rescue block but somehow it manages to still crash.

the block of code looks like this:

# Retrieve messages from server
def get_messages
  @connection.select('INBOX')
  @connection.uid_search(['ALL']).each do |uid|
    msg = @connection.uid_fetch(uid,'RFC822').first.attr['RFC822']
    begin
      process_message(msg)
      add_to_processed_folder(uid) if @processed_folder
    rescue
       handle_bogus_message(msg)
    end
    # Mark message as deleted 
    @connection.uid_store(uid, "+FLAGS", [:Seen, :Deleted])
  end
end

Given this code i would assume that if process_message or add_to_processed_folder could not execute then rescue would kick in and call handle_bogus_message. That being said I’m running this code in a production environment and sometimes when i “get” an email message (this is run from a rake task) it dies with a SyntaxError.

For a look at the error message check out http://pastie.org/1028479 and not that process_message that it is referring to is the same process_message above. Is there any reason why begin – rescue won’t catch this exception?

Answer 1

rescue without a parameter just rescues exceptions that inherit from StandardError. To rescue a SyntaxError use rescue SyntaxError.

To rescue all exceptions you would use rescue Exception, but note that that’s a bad idea (which is why it’s not the default behavior of rescue) as explained here and here. Especially this part:

Rescuing Interrupt prevents the user from using CTRLC to exit the program.

Rescuing SignalException prevents the program from responding correctly to signals. It will be unkillable except by kill -9.