and WEBrick::CGI#config. these are necessary to use an instance of
WEBrick::CGI as the first argument of HTTPServlet#get_instance.
(suggested by Tatsuki Sugiura)
* lib/webrick/cgi.rb
(WEBrick::CGI#initalize): set a dummy to @config[:ServerSoftware]
if SERVER_SOFTWARE environment variable is not given.
(WEBrick::CGI#start): req.path_info must be a String.
(WEBrick::CGI::Socket#request_line): treat REQUEST_METHOD, PATH_INFO
and SCRIPT_NAME to run in console.
* lib/webrick/httputils.rb (WEBrick::HTTPUtils.escape_path): should
not use String#split("/"). it removes trailing empty path component.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@8393 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
should delete trailing LF from the result of pack("m*").
* lib/webrick/httpproxy.rb (WEBrick::HTTPProxyServer#proxy_connect):
- should delete trailing LF from the result of pack("m*").
- clear Request-Line not to send the sesponse by HTTPServer#run.
* lib/webrick/httputils (WEBrick::HTTPUtils.parse_qvalues):
refine regexp (and change the name of a local variable).
* lib/webrick/server.rb (WEBrick::Daemon.start): prepared stdio
don't allow changing its mode.
* test/webrick/*, sample/webrick/httpproxy.rb: add new files.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@7743 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
