From: GOTOU Yuuzou <gotoyuzo@...>
Date: 2003-08-24T02:34:51+09:00
Subject: Re: Getting html page fields with webrick

Hi,

In message <oprucvpwswmwmsst@out.virgilio.it>,
 `JK <j.khaldi@virgilio.it>' wrote:
> Well, now instead of having the output of results (server side) to 
> "$stderr" I want to have it in a "log_file". How to initialize the BasicLog 
> class?

Here is a sample code:

  require 'webrick'

  logger    = WEBrick::Log::new("log_file", WEBrick::Log::DEBUG)
  accesslog = WEBrick::BasicLog::new("accesslog")
  referer   = WEBrick::BasicLog::new("referer")
  agent     = WEBrick::BasicLog::new("agent")
  customlog = WEBrick::BasicLog::new("customlog")

  s = WEBrick::HTTPServer.new(
    :BindAddress    => "0.0.0.0",
    :Port           => 2000,
    :DocumentRoot   => File::join(ENV["HOME"], "public_html"),
    :Logger         => logger,
    :AccessLog      => [
      [ accesslog, WEBrick::AccessLog::COMMON_LOG_FORMAT  ],
      [ referer,   WEBrick::AccessLog::REFERER_LOG_FORMAT ],
      [ agent,     WEBrick::AccessLog::AGENT_LOG_FORMAT   ],
      [ customlog, "%a %U %T" ]  # peeraddr, Request-URI, process time
    ]
  )
  trap(:INT){ s.shutdown }
  s.start

There are two logging options; :Logger is server's trace log
and :AccessLog is an array of access logs.
Each entry of :AccessLog includes BasicLog object and logging
format string. The specification of format string is a subset
of Apache's LogFormat.

-- 
gotoyuzo