On Sun, 2008-07-13 at 06:35 +0000, John M. Dlugosz wrote: > Reading the docs, I'm a little confused. Is ModPerl::Registry > specifically for emulating the CGI environment, or should I do > something else if I plan to write to the API from scratch and don't > need the backward compatibility?
ModPerl::Registry is intended to provide some of the speed benefits of mod_perl to people with legacy CGI scripts, or for people who are more comfortable with the CGI idiom. As such, it does the job well, but there are also a number of gotchas that are introduced by suddenly making your scripts (a) persistent in memory and (b) potential closures - they don't always work as you previously expected. >From your posts on PerlMonks, I've seen that you are obviously not a beginner, and as you have no legacy system to support, I would recommend that you dive straight into mod_perl handlers. If you haven't already, start with: http://perl.apache.org/docs/2.0/user/intro/start_fast.html#Handler_Modules and read the rest of the docs thoroughly : http://perl.apache.org/docs/2.0/user/index.html Much of what is said in the mod_perl v1 books still stands. The things that have changed are detailed here: http://perl.apache.org/docs/2.0/user/porting/compat.html For the most part, you'll find that you set up a standard way of receiving the request from MP, and returning the result, and the rest will just be normal Perl. good luck :) clint
