NAME
   Web::Simple - A quick and easy way to build simple web applications

SYNOPSIS
     #!/usr/bin/env perl

     package HelloWorld;
     use Web::Simple;

     sub dispatch_request {
       sub (GET) {
         [ 200, [ 'Content-type', 'text/plain' ], [ 'Hello world!' ] ]
       },
       sub () {
         [ 405, [ 'Content-type', 'text/plain' ], [ 'Method not allowed' ] ]
       }
     }

     HelloWorld->run_if_script;

   If you save this file into your cgi-bin as "hello-world.cgi" and then
   visit:

     http://my.server.name/cgi-bin/hello-world.cgi/

   you'll get the "Hello world!" string output to your browser. At the same
   time this file will also act as a class module, so you can save it as
   HelloWorld.pm and use it as-is in test scripts or other deployment
   mechanisms.

   Note that you should retain the ->run_if_script even if your app is a
   module, since this additionally makes it valid as a .psgi file, which
   can be extremely useful during development.

   For more complex examples and non-CGI deployment, see
   Web::Simple::Deployment. To get help with Web::Simple, please connect to
   the irc.perl.org IRC network and join #web-simple.

DESCRIPTION
   The philosophy of Web::Simple is to keep to an absolute bare minimum for
   everything. It is not designed to be used for large scale applications;
   the Catalyst web framework already works very nicely for that and is a
   far more mature, well supported piece of software.

   However, if you have an application that only does a couple of things,
   and want to not have to think about complexities of deployment, then
   Web::Simple might be just the thing for you.

   The only public interface the Web::Simple module itself provides is an
   "import" based one:

     use Web::Simple 'NameOfApplication';

   This sets up your package (in this case "NameOfApplication" is your
   package) so that it inherits from Web::Simple::Application and imports
   strictures, as well as installs a "PSGI_ENV" constant for convenience,
   as well as some other subroutines.

   Importing strictures will automatically make your code use the "strict"
   and "warnings" pragma, so you can skip the usual:

     use strict;
     use warnings FATAL => 'aa';

   provided you 'use Web::Simple' at the top of the file. Note that we turn
   on *fatal* warnings so if you have any warnings at any point from the
   file that you did 'use Web::Simple' in, then your application will die.
   This is, so far, considered a feature.

   When we inherit from Web::Simple::Application we also use Moo, which is
   the the equivalent of:

     {
       package NameOfApplication;
       use Moo;
       extends 'Web::Simple::Application';
     }

   So you can use Moo features in your application, such as creating
   attributes using the "has" subroutine, etc. Please see the documentation
   for Moo for more information.

   It also exports the following subroutines for use in dispatchers:

     response_filter { ... };

     redispatch_to '/somewhere';

   Finally, import sets

     $INC{"NameOfApplication.pm"} = 'Set by "use Web::Simple;" invocation';

   so that perl will not attempt to load the application again even if

     require NameOfApplication;

   is encountered in other code.

   One important thing to remember when using

     NameOfApplication->run_if_script;

   At the end of your app is that this call will create an instance of your
   app for you automatically, regardless of context. An easier way to think
   of this would be if the method were more verbosely named

    NameOfApplication->run_request_if_script_else_turn_coderef_for_psgi;

DISPATCH STRATEGY
   Web::Simple despite being straightforward to use, has a powerful system
   for matching all sorts of incoming URLs to one or more subroutines.
   These subroutines can be simple actions to take for a given URL, or
   something more complicated, including entire Plack applications,
   Plack::Middleware and nested subdispatchers.

 Examples
    sub dispatch_request {
      # matches: GET /user/1.htm?show_details=1
      #          GET /user/1.htm
      sub (GET + /user/* + ?show_details~ + .htm|.html|.xhtml) {
        my ($self, $user_id, $show_details) = @_;
        ...
      },
      # matches: POST /user?username=frew
      #          POST /user?username=mst&first_name=matt&last_name=trout
      sub (POST + /user + ?username=&*) {
         my ($self, $username, $misc_params) = @_;
        ...
      },
      # matches: DELETE /user/1/friend/2
      sub (DELETE + /user/*/friend/*) {
        my ($self, $user_id, $friend_id) = @_;
        ...
      },
      # matches: PUT /user/1?first_name=Matt&last_name=Trout
      sub (PUT + /user/* + ?first_name~&last_name~) {
        my ($self, $user_id, $first_name, $last_name) = @_;
        ...
      },
      sub (/user/*/...) {
        my $user_id = $_[1];
        # matches: PUT /user/1/role/1
        sub (PUT + /role/*) {
          my $role_id = $_[1];
          ...
        },
        # matches: DELETE /user/1/role/1
        sub (DELETE + /role/*) {
          my $role_id = $_[1];
          ...
        },
      },
    }

 The dispatch cycle
   At the beginning of a request, your app's dispatch_request method is
   called with the PSGI $env as an argument. You can handle the request
   entirely in here and return a PSGI response arrayref if you want:

     sub dispatch_request {
       my ($self, $env) = @_;
       [ 404, [ 'Content-type' => 'text/plain' ], [ 'Amnesia == fail' ] ]
     }

   However, generally, instead of that, you return a set of dispatch subs:

     sub dispatch_request {
       my $self = shift;
       sub (/) { redispatch_to '/index.html' },
       sub (/user/*) { $self->show_user($_[1]) },
       ...
     }

   Well, a sub is a valid PSGI response too (for ultimate streaming and
   async cleverness). If you want to return a PSGI sub you have to wrap it
   into an array ref.

     sub dispatch_request {
       [ sub {
           my $respond = shift;
           # This is pure PSGI here, so read perldoc PSGI
       } ]
     }

   If you return a subroutine with a prototype, the prototype is treated as
   a match specification - and if the test is passed, the body of the sub
   is called as a method and passed any matched arguments (see below for
   more details).

   You can also return a plain subroutine which will be called with just
   $env - remember that in this case if you need $self you must close over
   it.

   If you return a normal object, Web::Simple will simply return it upwards
   on the assumption that a response_filter (or some arbitrary
   Plack::Middleware) somewhere will convert it to something useful. This
   allows:

     sub dispatch_request {
       my $self = shift;
       sub (.html) { response_filter { $self->render_zoom($_[0]) } },
       sub (/user/*) { $self->users->get($_[1]) },
     }

   An alternative to using prototypes to declare a match specification for
   a given route is to provide a Dancer like key-value list:

     sub dispatch_request {
       my $self = shift;
       (
         '.html' => sub { response_filter { $self->render_zoom($_[0]) } },
         '/user/*' => sub { $self->users->get($_[1]) },
         'POST + %*' => 'handle_post',
       )
     }

   This can be useful in situations where you are generating a dispatch
   table programmatically, where setting a subroutines protoype is
   difficult. Note that in the example above, "handle_post" is a method
   that would be called.

   to render a user object to HTML, if there is an incoming URL such as:

     http://myweb.org/user/111.html

   This works because as we descend down the dispachers, we first match
   "sub (.html)", which adds a "response_filter" (basically a specialized
   routine that follows the Plack::Middleware specification), and then
   later we also match "sub (/user/*)" which gets a user and returns that
   as the response. This user object 'bubbles up' through all the wrapping
   middleware until it hits the "response_filter" we defined, after which
   the return is converted to a true html response.

   However, two types of objects are treated specially - a
   "Plack::Component" object will have its "to_app" method called and be
   used as a dispatcher:

     sub dispatch_request {
       my $self = shift;
       sub (/static/...) { Plack::App::File->new(...) },
       ...
     }

   A Plack::Middleware object will be used as a filter for the rest of the
   dispatch being returned into:

     ## responds to /admin/track_usage AND /admin/delete_accounts

     sub dispatch_request {
       my $self = shift;
       sub (/admin/**) {
         Plack::Middleware::Session->new(%opts);
       },
       sub (/admin/track_usage) {
         ## something that needs a session
       },
       sub (/admin/delete_accounts) {
         ## something else that needs a session
       },
     }

   Note that this is for the dispatch being returned to, so if you want to
   provide it inline you need to do:

     ## ALSO responds to /admin/track_usage AND /admin/delete_accounts

     sub dispatch_request {
       my $self = shift;
       sub (/admin/...) {
         sub {
           Plack::Middleware::Session->new(%opts);
         },
         sub (/track_usage) {
           ## something that needs a session
         },
         sub (/delete_accounts) {
           ## something else that needs a session
         },
       }
     }

   And that's it - but remember that all this happens recursively - it's
   dispatchers all the way down. A URL incoming pattern will run all
   matching dispatchers and then hit all added filters or
   Plack::Middleware.

 Web::Simple match specifications
  Method matches
     sub (GET) {

   A match specification beginning with a capital letter matches HTTP
   requests with that request method.

  Path matches
     sub (/login) {

   A match specification beginning with a / is a path match. In the
   simplest case it matches a specific path. To match a path with a
   wildcard part, you can do:

     sub (/user/*) {
       $self->handle_user($_[1])

   This will match /user/<anything> where <anything> does not include a
   literal / character. The matched part becomes part of the match
   arguments. You can also match more than one part:

     sub (/user/*/*) {
       my ($self, $user_1, $user_2) = @_;

     sub (/domain/*/user/*) {
       my ($self, $domain, $user) = @_;

   and so on. To match an arbitrary number of parts, use "**":

     sub (/page/**) {
       my ($self, $match) = @_;

   This will result in a single element for the entire match. Note that you
   can do

     sub (/page/**/edit) {

   to match an arbitrary number of parts up to but not including some final
   part.

   Note: Since Web::Simple handles a concept of file extensions, "*" and
   "**" matchers will not by default match things after a final dot, and
   this can be modified by using "*.*" and "**.*" in the final position,
   e.g.:

     /one/*       matches /one/two.three    and captures "two"
     /one/*.*     matches /one/two.three    and captures "two.three"
     /**          matches /one/two.three    and captures "one/two"
     /**.*        matches /one/two.three    and captures "one/two.three"

   Finally,

     sub (/foo/...) {

   Will match "/foo/" on the beginning of the path and strip it. This is
   designed to be used to construct nested dispatch structures, but can
   also prove useful for having e.g. an optional language specification at
   the start of a path.

   Note that the '...' is a "maybe something here, maybe not" so the above
   specification will match like this:

     /foo         # no match
     /foo/        # match and strip path to '/'
     /foo/bar/baz # match and strip path to '/bar/baz'

   Almost the same,

     sub (/foo...) {

   Will match on "/foo/bar/baz", but also include "/foo". Otherwise it
   operates the same way as "/foo/...".

     /foo         # match and strip path to ''
     /foo/        # match and strip path to '/'
     /foo/bar/baz # match and strip path to '/bar/baz'

   Please note the difference between "sub(/foo/...)" and "sub(/foo...)".
   In the first case, this is expecting to find something after "/foo" (and
   fails to match if nothing is found), while in the second case we can
   match both "/foo" and "/foo/more/to/come". The following are roughly the
   same:

     sub (/foo)   { 'I match /foo' },
     sub (/foo/...) {
       sub (/bar) { 'I match /foo/bar' },
       sub (/*)   { 'I match /foo/{id}' },
     }

   Versus

     sub (/foo...) {
       sub (~)    { 'I match /foo' },
       sub (/bar) { 'I match /foo/bar' },
       sub (/*)   { 'I match /foo/{id}' },
     }

   You may prefer the latter example should you wish to take advantage of
   subdispatchers to scope common activities. For example:

     sub (/user...) {
       my $user_rs = $schema->resultset('User');
       sub (~) { $user_rs },
       sub (/*) { $user_rs->find($_[1]) },
     }

   You should note the special case path match "sub (~)" which is only
   meaningful when it is contained in this type of path match. It matches
   to an empty path.

  Naming your patch matches
   Any "*", "**", "*.*", or "**.*" match can be followed with ":name" to
   make it into a named match, so:

     sub (/*:one/*:two/*:three/*:four) {
       "I match /1/2/3/4 capturing { one => 1, two =>  2, three => 3, four => 4 }"
     }

     sub (/**.*:allofit) {
       "I match anything capturing { allofit => \$whole_path }"
     }

   In the specific case of a simple single-* match, the * may be omitted,
   to allow you to write:

     sub (/:one/:two/:three/:four) {
       "I match /1/2/3/4 capturing { one => 1, two =>  2, three => 3, four => 4 }"
     }

  "/foo" and "/foo/" are different specs
   As you may have noticed with the difference between "sub(/foo/...)" and
   "sub(/foo...)", trailing slashes in path specs are significant. This is
   intentional and necessary to retain the ability to use relative links on
   websites. Let's demonstrate on this link:

     <a href="bar">bar</a>

   If the user loads the url "/foo/" and clicks on this link, they will be
   sent to "/foo/bar". However when they are on the url "/foo" and click
   this link, then they will be sent to "/bar".

   This makes it necessary to be explicit about the trailing slash.

  Extension matches
     sub (.html) {

   will match .html from the path (assuming the subroutine itself returns
   something, of course). This is normally used for rendering - e.g.:

     sub (.html) {
       response_filter { $self->render_html($_[1]) }
     }

   Additionally,

     sub (.*) {

   will match any extension and supplies the extension as a match argument.

  Query and body parameter matches
   Query and body parameters can be match via

     sub (?<param spec>) { # match URI query
     sub (%<param spec>) { # match body params

   The body spec will match if the request content is either
   application/x-www-form-urlencoded or multipart/form-data - the latter of
   which is required for uploads - see below.

   The param spec is elements of one of the following forms:

     param~        # optional parameter
     param=        # required parameter
     @param~       # optional multiple parameter
     @param=       # required multiple parameter
     :param~       # optional parameter in hashref
     :param=       # required parameter in hashref
     :@param~      # optional multiple in hashref
     :@param=      # required multiple in hashref
     *             # include all other parameters in hashref
     @*            # include all other parameters as multiple in hashref

   separated by the "&" character. The arguments added to the request are
   one per non-":"/"*" parameter (scalar for normal, arrayref for
   multiple), plus if any ":"/"*" specs exist a hashref containing those
   values.

   Please note that if you specify a multiple type parameter match, you are
   ensured of getting an arrayref for the value, EVEN if the current
   incoming request has only one value. However if a parameter is specified
   as single and multiple values are found, the last one will be used.

   For example to match a "page" parameter with an optional "order_by"
   parameter one would write:

     sub (?page=&order_by~) {
       my ($self, $page, $order_by) = @_;
       return unless $page =~ /^\d+$/;
       $order_by ||= 'id';
       response_filter {
         $_[1]->search_rs({}, { page => $page, order_by => $order_by });
       }
     }

   to implement paging and ordering against a DBIx::Class::ResultSet
   object.

   Another Example: To get all parameters as a hashref of arrayrefs, write:

     sub(?@*) {
       my ($self, $params) = @_;
       ...

   To get two parameters as a hashref, write:

     sub(?:user~&:domain~) {
       my ($self, $params) = @_; # params contains only 'user' and 'domain' keys

   You can also mix these, so:

     sub (?foo=&@bar~&:coffee=&@*) {
        my ($self, $foo, $bar, $params);

   where $bar is an arrayref (possibly an empty one), and $params contains
   arrayref values for all parameters not mentioned and a scalar value for
   the 'coffee' parameter.

   Note, in the case where you combine arrayref, single parameter and named
   hashref style, the arrayref and single parameters will appear in @_ in
   the order you defined them in the protoype, but all hashrefs will merge
   into a single $params, as in the example above.

  Upload matches
     sub (*foo=) { # param specifier can be anything valid for query or body

   The upload match system functions exactly like a query/body match,
   except that the values returned (if any) are "Web::Dispatch::Upload"
   objects.

   Note that this match type will succeed in two circumstances where you
   might not expect it to - first, when the field exists but is not an
   upload field and second, when the field exists but the form is not an
   upload form (i.e. content type "application/x-www-form-urlencoded"
   rather than "multipart/form-data"). In either of these cases, what
   you'll get back is a "Web::Dispatch::NotAnUpload" object, which will
   "die" with an error pointing out the problem if you try and use it. To
   be sure you have a real upload object, call

     $upload->is_upload # returns 1 on a valid upload, 0 on a non-upload field

   and to get the reason why such an object is not an upload, call

     $upload->reason # returns a reason or '' on a valid upload.

   Other than these two methods, the upload object provides the same
   interface as Plack::Request::Upload with the addition of a stringify to
   the temporary filename to make copying it somewhere else easier to
   handle.

  Combining matches
   Matches may be combined with the + character - e.g.

     sub (GET + /user/*) {

   to create an AND match. They may also be combined withe the | character
   - e.g.

     sub (GET|POST) {

   to create an OR match. Matches can be nested with () - e.g.

     sub ((GET|POST) + /user/*) {

   and negated with ! - e.g.

     sub (!/user/foo + /user/*) {

   ! binds to the immediate rightmost match specification, so if you want
   to negate a combination you will need to use

     sub ( !(POST|PUT|DELETE) ) {

   and | binds tighter than +, so

     sub ((GET|POST) + /user/*) {

   and

     sub (GET|POST + /user/*) {

   are equivalent, but

     sub ((GET + /admin/...) | (POST + /admin/...)) {

   and

     sub (GET + /admin/... | POST + /admin/...) {

   are not - the latter is equivalent to

     sub (GET + (/admin/...|POST) + /admin/...) {

   which will never match!

  Whitespace
   Note that for legibility you are permitted to use whitespace:

     sub (GET + /user/*) {

   but it will be ignored. This is because the perl parser strips
   whitespace from subroutine prototypes, so this is equivalent to

     sub (GET+/user/*) {

  Accessing parameters via %_
   If your dispatch specification causes your dispatch subroutine to
   receive a hash reference as its first argument, the contained named
   parameters will be accessible via %_.

   This can be used to access your path matches, if they are named:

     sub (GET + /foo/:path_part) {
       [ 200,
         ['Content-type' => 'text/plain'],
         ["We are in $_{path_part}"],
       ];
     }

   Or, if your first argument would be a hash reference containing named
   query parameters:

     sub (GET + /foo + ?:some_param=) {
       [ 200,
         ['Content-type' => 'text/plain'],
         ["We received $_{some_param} as parameter"],
       ];
     }

   Of course this also works when all you are doing is slurping the whole
   set of parameters by their name:

     sub (GET + /foo + ?*) {
       [ 200,
         ['Content-type' => 'text/plain'],
         [exists($_{foo}) ? "Received a foo: $_{foo}" : "No foo!"],
       ],
     }

   Note that only the first hash reference will be available via %_. If you
   receive additional hash references, you will need to access them as
   usual.

  Accessing the PSGI env hash
   In some cases you may wish to get the raw PSGI env hash - to do this,
   you can either use a plain sub:

     sub {
       my ($env) = @_;
       ...
     }

   or use the "PSGI_ENV" constant exported to retrieve it from @_:

     sub (GET + /foo + ?some_param=) {
       my $param = $_[1];
       my $env = $_[PSGI_ENV];
     }

   but note that if you're trying to add a middleware, you should simply
   use Web::Simple's direct support for doing so.

EXPORTED SUBROUTINES
 response_filter
     response_filter {
       # Hide errors from the user because we hates them, preciousss
       if (ref($_[0]) eq 'ARRAY' && $_[0]->[0] == 500) {
         $_[0] = [ 200, @{$_[0]}[1..$#{$_[0]}] ];
       }
       return $_[0];
     };

   The response_filter subroutine is designed for use inside dispatch
   subroutines.

   It creates and returns a special dispatcher that always matches, and
   calls the block passed to it as a filter on the result of running the
   rest of the current dispatch chain.

   Thus the filter above runs further dispatch as normal, but if the result
   of dispatch is a 500 (Internal Server Error) response, changes this to a
   200 (OK) response without altering the headers or body.

 redispatch_to
     redispatch_to '/other/url';

   The redispatch_to subroutine is designed for use inside dispatch
   subroutines.

   It creates and returns a special dispatcher that always matches, and
   instead of continuing dispatch re-delegates it to the start of the
   dispatch process, but with the path of the request altered to the
   supplied URL.

   Thus if you receive a POST to "/some/url" and return a redispatch to
   "/other/url", the dispatch behaviour will be exactly as if the same POST
   request had been made to "/other/url" instead.

   Note, this is not the same as returning an HTTP 3xx redirect as a
   response; rather it is a much more efficient internal process.

CHANGES BETWEEN RELEASES
 Changes between 0.004 and 0.005
   *   dispatch {} replaced by declaring a dispatch_request method

       dispatch {} has gone away - instead, you write:

         sub dispatch_request {
           my $self = shift;
           sub (GET /foo/) { ... },
           ...
         }

       Note that this method is still returning the dispatch code - just
       like "dispatch" did.

       Also note that you need the "my $self = shift" since the magic $self
       variable went away.

   *   the magic $self variable went away.

       Just add "my $self = shift;" while writing your "sub
       dispatch_request {" like a normal perl method.

   *   subdispatch deleted - all dispatchers can now subdispatch

       In earlier releases you needed to write:

         subdispatch sub (/foo/...) {
           ...
           [
             sub (GET /bar/) { ... },
             ...
           ]
         }

       As of 0.005, you can instead write simply:

         sub (/foo/...) {
           ...
           (
             sub (GET /bar/) { ... },
             ...
           )
         }

 Changes since Antiquated Perl
   *   filter_response renamed to response_filter

       This is a pure rename; a global search and replace should fix it.

   *   dispatch [] changed to dispatch {}

       Simply changing

         dispatch [ sub(...) { ... }, ... ];

       to

         dispatch { sub(...) { ... }, ... };

       should work fine.

DEVELOPMENT HISTORY
   Web::Simple was originally written to form part of my Antiquated Perl
   talk for Italian Perl Workshop 2009, but in writing the bloggery example
   I realised that having a bare minimum system for writing web
   applications that doesn't drive me insane was rather nice and decided to
   spend my attempt at nanowrimo for 2009 improving and documenting it to
   the point where others could use it.

   The Antiquated Perl talk can be found at
   <http://www.shadowcat.co.uk/archive/conference-video/> and the slides
   are reproduced in this distribution under Web::Simple::AntiquatedPerl.

COMMUNITY AND SUPPORT
 IRC channel
   irc.perl.org #web-simple

 No mailing list yet
   Because mst's non-work email is a bombsite so he'd never read it anyway.

 Git repository
   Gitweb is on http://git.shadowcat.co.uk/ and the clone URL is:

     git clone git://git.shadowcat.co.uk/catagits/Web-Simple.git

AUTHOR
   Matt S. Trout (mst) <[email protected]>

CONTRIBUTORS
   Devin Austin (dhoss) <[email protected]>

   Arthur Axel 'fREW' Schmidt <[email protected]>

   gregor herrmann (gregoa) <[email protected]>

   John Napiorkowski (jnap) <[email protected]>

   Josh McMichael <[email protected]>

   Justin Hunter (arcanez) <[email protected]>

   Kjetil Kjernsmo <[email protected]>

   markie <[email protected]>

   Christian Walde (Mithaldu) <[email protected]>

   nperez <[email protected]>

   Robin Edwards <[email protected]>

   Andrew Rodland (hobbs) <[email protected]>

   Robert Sedlacek (phaylon) <[email protected]>

   Hakim Cassimally (osfameron) <[email protected]>

   Karen Etheridge (ether) <[email protected]>

COPYRIGHT
   Copyright (c) 2011 the Web::Simple "AUTHOR" and "CONTRIBUTORS" as listed
   above.

LICENSE
   This library is free software and may be distributed under the same
   terms as perl itself.

You are viewing proxied material from ftp.icm.edu.pl. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.