Linux and UNIX Man Pages

Linux & Unix Commands - Search Man Pages

plack::app::cgibin(3pm) [debian man page]

Plack::App::CGIBin(3pm) 				User Contributed Perl Documentation				   Plack::App::CGIBin(3pm)

NAME
Plack::App::CGIBin - cgi-bin replacement for Plack servers SYNOPSIS
use Plack::App::CGIBin; use Plack::Builder; my $app = Plack::App::CGIBin->new(root => "/path/to/cgi-bin")->to_app; builder { mount "/cgi-bin" => $app; }; # Or from the command line plackup -MPlack::App::CGIBin -e 'Plack::App::CGIBin->new(root => "/path/to/cgi-bin")->to_app' DESCRIPTION
Plack::App::CGIBin allows you to load CGI scripts from a directory and convert them into a PSGI application. This would give you the extreme easiness when you have bunch of old CGI scripts that is loaded using cgi-bin of Apache web server. HOW IT WORKS
This application checks if a given file path is a perl script and if so, uses CGI::Compile to compile a CGI script into a sub (like ModPerl::Registry) and then run it as a persistent application using CGI::Emulate::PSGI. If the given file is not a perl script, it executes the script just like a normal CGI script with fork & exec. This is like a normal web server mode and no performance benefit is achieved. The default mechanism to determine if a given file is a Perl script is as follows: o Check if the filename ends with ".pl". If yes, it is a Perl script. o Open the file and see if the shebang (first line of the file) contains the word "perl" (like "#!/usr/bin/perl"). If yes, it is a Perl script. You can customize this behavior by passing "exec_cb" callback, which takes a file path to its first argument. For example, if your perl-based CGI script uses lots of global variables and such and are not ready to run on a persistent environment, you can do: my $app = Plack::App::CGIBin->new( root => "/path/to/cgi-bin", exec_cb => sub { 1 }, )->to_app; to always force the execute option for any files. AUTHOR
Tatsuhiko Miyagawa SEE ALSO
Plack::App::File CGI::Emulate::PSGI CGI::Compile Plack::App::WrapCGI See also Plack::App::WrapCGI if you compile one CGI script into a PSGI application without serving CGI scripts from a directory, to remove overhead of filesystem lookups, etc. perl v5.14.2 2011-11-02 Plack::App::CGIBin(3pm)

Check Out this Related Man Page

Plack::Builder(3pm)					User Contributed Perl Documentation				       Plack::Builder(3pm)

NAME
Plack::Builder - OO and DSL to enable Plack Middlewares SYNOPSIS
# in .psgi use Plack::Builder; my $app = sub { ... }; builder { enable "Deflater"; enable "Session", store => "File"; enable "Debug", panels => [ qw(DBITrace Memory Timer) ]; enable "+My::Plack::Middleware"; $app; }; # use URLMap builder { mount "/foo" => builder { enable "Foo"; $app; }; mount "/bar" => $app2; mount "http://example.com/" => builder { $app3 }; }; # using OO interface my $builder = Plack::Builder->new(); $builder->add_middleware('Foo', opt => 1); $app = $builder->mount('/app' => $app); $app = $builder->to_app($app); DESCRIPTION
Plack::Builder gives you a quick domain specific language (DSL) to wrap your application with Plack::Middleware subclasses. The middleware you're trying to use should use Plack::Middleware as a base class to use this DSL, inspired by Rack::Builder. Whenever you call "enable" on any middleware, the middleware app is pushed to the stack inside the builder, and then reversed when it actually creates a wrapped application handler. "Plack::Middleware::" is added as a prefix by default. So: builder { enable "Foo"; enable "Bar", opt => "val"; $app; }; is syntactically equal to: $app = Plack::Middleware::Bar->wrap($app, opt => "val"); $app = Plack::Middleware::Foo->wrap($app); In other words, you're supposed to "enable" middleware from outer to inner. INLINE MIDDLEWARE
Plack::Builder allows you to code middleware inline using a nested code reference. If the first argument to "enable" is a code reference, it will be passed an $app and is supposed to return another code reference which is PSGI application that consumes $env in runtime. So: builder { enable sub { my $app = shift; sub { my $env = shift; # do preprocessing my $res = $app->($env); # do postprocessing return $res; }; }; $app; }; is equal to: my $mw = sub { my $app = shift; sub { my $env = shift; $app->($env) }; }; $app = $mw->($app); URLMap support Plack::Builder has a native support for Plack::App::URLMap with "mount" method. use Plack::Builder; my $app = builder { mount "/foo" => $app1; mount "/bar" => builder { enable "Foo"; $app2; }; }; See Plack::App::URLMap's "map" method to see what they mean. With builder you can't use "map" as a DSL, for the obvious reason :) NOTE: Once you use "mount" in your builder code, you have to use "mount" for all the paths, including the root path ("/"). You can't have the default app in the last line of "builder" like: my $app = sub { my $env = shift; ... }; builder { mount "/foo" => sub { ... }; $app; # THIS DOESN'T WORK }; You'll get warnings saying that your mount configuration will be ignored. Instead you should use "mount "/" => ..." in the last line to set the default fallback app. builder { mount "/foo" => sub { ... }; mount "/" => $app; } Note that the "builder" DSL returns a whole new PSGI application, which means o "builder { ... }" should normally the last statement of a ".psgi" file, because the return value of "builder" is the application that actually is executed. o You can nest your "builder" block, mixed with "mount" (see URLMap support above): builder { mount "/foo" => builder { mount "/bar" => $app; } } will locate the $app under "/foo/bar" since the inner "builder" block puts it under "/bar" and it results a new PSGI application which is located under "/foo" because of the outer "builder" block. CONDITIONAL MIDDLEWARE SUPPORT
You can use "enable_if" to conditionally enable middleware based on the runtime environment. See Plack::Middleware::Conditional for details. SEE ALSO
Plack::Middleware Plack::App::URLMap Plack::Middleware::Conditional perl v5.14.2 2012-05-17 Plack::Builder(3pm)
Man Page