3 Catalyst::Upgrading - Instructions for upgrading to the latest Catalyst
5 =head1 Upgrading to Catalyst TBA
7 =head2 Catalyst::Plugin::Unicode::Encoding is now core
9 The previously stand alone Unicode support module L<Catalyst::Plugin::Unicode::Encoding>
10 has been brought into core as a default plugin. Going forward, all you need is
11 to add a configuration setting for the encoding type. For example:
17 __PACKAGE__->config( encoding => 'UTF-8' );
19 Please note that this is different from the old stand alone plugin which applied
20 C<UTF-8> encoding by default (that is, if you did not set an explicit
21 C<encoding> configuration value, it assumed you wanted UTF-8). In order to
22 preserve backwards compatibility you will need to explicitly turn it on via the
23 configuration setting. THIS MIGHT CHANGE IN THE FUTURE, so please consider
24 starting to test your application with proper UTF-8 support and remove all those
25 crappy hacks you munged into the code because you didn't know the Plugin
28 For people that are using the Plugin, you will note a startup warning suggesting
29 that you can remove it from the plugin list. When you do so, please remember to
30 add the configuration setting, since you can no longer rely on the default being
31 UTF-8. We'll add it for you if you continue to use the stand alone plugin and
32 we detect this, but this backwards compatibility shim will likely be removed in
33 a few releases (trying to clean up the codebase after all).
35 B<NOTE>: One other difference between the cored plugin and the stand alone one
36 is that in core we no longer throw an exception when there's a decode failure
37 but instead log a warning. If you rely on exceptions for control flow, you
38 will need to override method C<handle_unicode_encoding_exception> to die instead
39 of warning. Please let the dev team know if this is a problem for you.
41 If you have trouble with any of this, please bring it to the attention of the
42 Catalyst maintainer group.
44 =head2 basic async and event loop support
46 This version of L<Catalyst> offers some support for using L<AnyEvent> and
47 L<IO::Async> event loops in your application. These changes should work
48 fine for most applications however if you are already trying to perform
49 some streaming, minor changes in this area of the code might affect your
50 functionality. Please see L<Catalyst::Response\write_fh> for more and for a
53 We consider this feature experimental. We will try not to break it, but we
54 reserve the right to make necessary changes to fix major issues that people
55 run into when the use this functionality in the wild.
57 =head1 Upgrading to Catalyst 5.9
59 The major change is that L<Plack>, a toolkit for using the L<PSGI>
60 specification, now replaces most of the subclasses of L<Catalyst::Engine>. If
61 you are using one of the standard subclasses of L<Catalyst::Engine> this
62 should be a straightforward upgrade for you. It was a design goal for
63 this release to preserve as much backwards compatibility as possible.
64 However, since L<Plack> is different from L<Catalyst::Engine>, it is
65 possible that differences exist for edge cases. Therefore, we recommend
66 that care be taken with this upgrade and that testing should be greater
67 than would be the case with a minor point update. Please inform the
68 Catalyst developers of any problems so that we can fix them and
71 It is highly recommended that you become familiar with the L<Plack> ecosystem
72 and documentation. Being able to take advantage of L<Plack> development and
73 middleware is a major bonus to this upgrade. Documentation about how to
74 take advantage of L<Plack::Middleware> by writing your own C<< .psgi >> file
75 is contained in L<Catalyst::PSGI>.
77 If you have created a custom subclass of L<Catalyst:Engine>, you will
78 need to convert it to be a subclass of L<Plack::Handler>.
80 If you are using the L<Plack> engine, L<Catalyst::Engine::PSGI>, this new
81 release supersedes that code.
83 If you are using a subclass of L<Catalyst::Engine> that is aimed at
84 nonstandard or internal/testing uses, such as
85 L<Catalyst::Engine::Embeddable>, you should still be able to continue
88 Advice for specific subclasses of L<Catalyst::Engine> follows:
90 =head2 Upgrading the FastCGI Engine
92 No upgrade is needed if your myapp_fastcgi.pl script is already upgraded
93 to use L<Catalyst::Script::FastCGI>.
95 =head2 Upgrading the mod_perl / Apache Engines
97 The engines that are built upon the various iterations of mod_perl,
98 L<Catalyst::Engine::Apache::MP13> (for mod_perl 1, and Apache 1.x) and
99 L<Catalyst::Engine::Apache2::MP20> (for mod_perl 2, and Apache 2.x),
100 should be seamless upgrades and will work using using L<Plack::Handler::Apache1>
101 or L<Plack::Handler::Apache2> as required.
103 L<Catalyst::Engine::Apache2::MP19>, however, is no longer supported, as
104 Plack does not support mod_perl version 1.99. This is unlikely to be a
105 problem for anyone, as 1.99 was a brief beta-test release for mod_perl
106 2, and all users of mod_perl 1.99 are encouraged to upgrade to a
107 supported release of Apache 2 and mod_perl 2.
109 =head2 Upgrading the HTTP Engine
111 The default development server that comes with the L<Catalyst> distribution
112 should continue to work as expected with no changes as long as your C<myapp_server>
113 script is upgraded to use L<Catalyst::Script::HTTP>.
115 =head2 Upgrading the CGI Engine
117 If you were using L<Catalyst::Engine::CGI> there is no upgrade needed if your
118 myapp_cgi.pl script is already upgraded to use L<Catalyst::Script::CGI>.
120 =head2 Upgrading Catalyst::Engine::HTTP::Prefork
122 If you were using L<Catalyst::Engine::HTTP::Prefork> then L<Starman>
123 is automatically loaded. You should (at least) change your C<Makefile.PL>
124 to depend on Starman.
126 You can regenerate your C<myapp_server.pl> script with C<catalyst.pl>
127 and implement a C<MyApp::Script::Server> class that looks like this:
129 package MyApp::Script::Server;
131 use namespace::autoclean;
133 extends 'CatalystX::Script::Server::Starman';
137 This takes advantage of the new script system, and will add a number of
138 options to the standard server script as extra options are added by
141 More information about these options can be seen at
142 L<CatalystX::Script::Server::Starman/SYNOPSIS>.
144 An alternate route to implement this functionality is to write a simple .psgi
145 file for your application, and then use the L<plackup> utility to start the
148 =head2 Upgrading the PSGI Engine
150 If you were using L<Catalyst::Engine::PSGI>, this new release supersedes
151 this engine in supporting L<Plack>. By default the Engine is now always
152 L<Plack>. As a result, you can remove the dependency on
153 L<Catalyst::Engine::PSGI> in your C<Makefile.PL>.
155 Applications that were using L<Catalyst::Engine::PSGI>
156 previously should entirely continue to work in this release with no changes.
158 However, if you have an C<app.psgi> script, then you no longer need to
159 specify the PSGI engine. Instead, the L<Catalyst> application class now
160 has a new method C<psgi_app> which returns a L<PSGI> compatible coderef
161 which you can wrap in the middleware of your choice.
163 Catalyst will use the .psgi for your application if it is located in the C<home>
164 directory of the application.
166 For example, if you were using L<Catalyst::Engine::PSGI> in the past, you will
167 have written (or generated) a C<script/myapp.psgi> file similar to this one:
172 MyCatalystApp->setup_engine('PSGI');
175 enable ... # enable your desired middleware
176 sub { MyCatalystApp->run(@_) };
179 Instead, you now say:
185 enable ... #enable your desired middleware
186 MyCatalystApp->psgi_app;
189 In the simplest case:
191 MyCatalystApp->setup_engine('PSGI');
192 my $app = sub { MyCatalystApp->run(@_) }
196 my $app = MyCatalystApp->psgi_app(@_);
200 my $app = sub { MyCatalystApp->psgi_app(@_) };
201 # If you make ^^ this mistake, your app won't work, and will confuse the hell out of you!
203 You can now move C<< script/myapp.psgi >> to C<< myapp.psgi >>, and the built-in
204 Catalyst scripts and your test suite will start using your .psgi file.
206 B<NOTE:> If you rename your .psgi file without these modifications, then
207 any tests run via L<Catalyst::Test> will not be compatible with the new
208 release, and will result in the development server starting, rather than
209 the expected test running.
211 B<NOTE:> If you are directly accessing C<< $c->req->env >> to get the PSGI
212 environment then this accessor is moved to C<< $c->engine->env >>,
213 you will need to update your code.
215 =head2 Engines which are known to be broken
217 The following engines B<DO NOT> work as of Catalyst version 5.9. The
218 core team will be happy to work with the developers and/or users of
219 these engines to help them port to the new Plack/Engine system, but for
220 now, applications which are currently using these engines B<WILL NOT>
221 run without modification to the engine code.
225 =item Catalyst::Engine::Wx
227 =item Catalyst::Engine::Zeus
229 =item Catalyst::Engine::JobQueue::POE
231 =item Catalyst::Engine::XMPP2
233 =item Catalyst::Engine::SCGI
237 =head2 Engines with unknown status
239 The following engines are untested or have unknown compatibility.
240 Reports are highly encouraged:
244 =item Catalyst::Engine::Mojo
246 =item Catalyst::Engine::Server (marked as Deprecated)
248 =item Catalyst::Engine::HTTP::POE (marked as Deprecated)
252 =head2 Plack functionality
254 See L<Catalyst::PSGI>.
258 Tests should generally work the same in Catalyst 5.9, but there are
261 Previously, if using L<Catalyst::Test> and doing local requests (against
262 a local server), if the application threw an exception then this
263 exception propagated into the test.
265 This behavior has been removed, and now a 500 response will be returned
266 to the test. This change standardizes behavior, so that local test
267 requests behave similarly to remote requests.
269 =head2 Regex dispatch type is deprecated.
271 The Regex dispatchtype (L<Catalyst::DispatchType::Regex>) has been deprecated.
273 You are encouraged to move your application to Chained dispatch (L<Catalyst::DispatchType::Chained>).
275 If you cannot do so, please add a dependency to Catalyst::DispatchType::Regex to your application's
278 =head1 Upgrading to Catalyst 5.80
280 Most applications and plugins should run unaltered on Catalyst 5.80.
282 However, a lot of refactoring work has taken place, and several changes have
283 been made which could cause incompatibilities. If your application or plugin
284 is using deprecated code, or relying on side effects, then you could have
285 issues upgrading to this release.
287 Most issues found with existing components have been easy to
288 solve. This document provides a complete description of behavior changes
289 which may cause compatibility issues, and of new Catalyst warnings which
292 If you think you have found an upgrade-related issue which is not covered in
293 this document, please email the Catalyst list to discuss the problem.
295 =head1 Moose features
297 =head2 Application class roles
299 You can only apply method modifiers after the application's C<< ->setup >>
300 method has been called. This means that modifiers will not work with methods
301 run during the call to C<< ->setup >>.
303 See L<Catalyst::Manual::ExtendingCatalyst> for more information about using
304 L<Moose> in your applications.
306 =head2 Controller actions in Moose roles
308 You can use L<MooseX::MethodAttributes::Role> if you want to declare actions
311 =head2 Using Moose in Components
313 The correct way to use Moose in a component in a both forward and backwards
316 package TestApp::Controller::Root;
318 BEGIN { extends 'Catalyst::Component' }; # Or ::Controller, or whatever
320 See L<Components which inherit from Moose::Object before Catalyst::Component>.
322 =head1 Known backwards compatibility breakages
324 =head2 Applications in a single file
326 Applications must be in their own file, and loaded at compile time. This
327 issue generally only affects the tests of CPAN distributions. Your
328 application will fail if you try to define an application inline in a
329 block, and use plugins which supply a C< new > method, then use that
330 application latter in tests within the same file.
332 This is due to the fact that Catalyst is inlining a new method on your
333 application class allowing it to be compatible with Moose. The method
334 used to do this changed in 5.80004 to avoid the possibility of reporting
335 an 'Unknown Error' if your application failed to compile.
337 =head2 Issues with Class::C3
339 Catalyst 5.80 uses the L<Algorithm::C3> method dispatch order. This is
340 built into Perl 5.10, and comes via L<Class::C3> for Perl 5.8. This
341 replaces L<NEXT> with L<Class::C3::Adopt::NEXT>, forcing all components
342 to resolve methods using C3, rather than the unpredictable dispatch
345 This issue manifests itself by your application failing to start due to an
346 error message about having a non-linear @ISA.
348 The Catalyst plugin most often causing this is
349 L<Catalyst::Plugin::Session::Store::FastMmap> - if you are using this
350 plugin and see issues, then please upgrade your plugins, as it has been
351 fixed. Note that Makefile.PL in the distribution will warn about known
352 incompatible components.
354 This issue can, however, be found in your own application - the only solution is
355 to go through each base class of the class the error was reported against, until
356 you identify the ones in conflict, and resolve them.
358 To be able to generate a linear @ISA, the list of superclasses for each
359 class must be resolvable using the C3 algorithm. Unfortunately, when
360 superclasses are being used as mixins (to add functionality used in your class),
361 and with multiple inheritance, it is easy to get this wrong.
363 Most common is the case of:
365 package Component1; # Note, this is the common case
366 use base qw/Class::Accessor::Fast Class::Data::Inheritable/;
368 package Component2; # Accidentally saying it this way causes a failure
369 use base qw/Class::Data::Inheritable Class::Accessor::Fast/;
372 use base qw/Component1 Component2/;
374 Any situation like this will cause your application to fail to start.
376 For additional documentation about this issue, and how to resolve it, see
377 L<Class::C3::Adopt::NEXT>.
379 =head2 Components which inherit from Moose::Object before Catalyst::Component
381 Moose components which say:
383 package TestApp::Controller::Example;
385 extends qw/Moose::Object Catalyst::Component/;
387 to use the constructor provided by Moose, while working (if you do some hacks
388 with the C< BUILDARGS > method), will not work with Catalyst 5.80 as
389 C<Catalyst::Component> inherits from C<Moose::Object>, and so C< @ISA > fails
392 The correct way to use Moose in a component in a both forward and backwards
395 package TestApp::Controller::Root;
397 BEGIN { extends 'Catalyst::Component' }; # Or ::Controller, or whatever
399 Note that the C< extends > declaration needs to occur in a begin block for
400 L<attributes> to operate correctly.
402 This way you do not inherit directly from C<Moose::Object>
403 yourself. Having components which do not inherit their constructor from
404 C<Catalyst::Component> is B<unsupported>, and has never been recommended,
405 therefore you're on your own if you're using this technique. You'll need
406 to detect the version of Catalyst your application is running, and deal
407 with it appropriately.
409 You also don't get the L<Moose::Object> constructor, and therefore attribute
410 initialization will not work as normally expected. If you want to use Moose
411 attributes, then they need to be made lazy to correctly initialize.
413 Note that this only applies if your component needs to maintain component
414 backwards compatibility for Catalyst versions before 5.71001 - in 5.71001
415 attributes work as expected, and the BUILD method is called normally
416 (although BUILDARGS is not).
418 If you depend on Catalyst 5.8, then B<all> Moose features work as expected.
420 You will also see this issue if you do the following:
422 package TestApp::Controller::Example;
424 use base 'Catalyst::Controller';
426 as C< use base > appends to @ISA.
428 =head3 use Moose in MyApp
430 Similar to the above, this will also fail:
439 If you need to use Moose in your application class (e.g. for method modifiers
440 etc.) then the correct technique is:
448 __PACKAGE__->config( name => 'MyApp' );
449 __PACKAGE__->setup(qw/
453 =head2 Anonymous closures installed directly into the symbol table
455 If you have any code which installs anonymous subroutine references directly
456 into the symbol table, you may encounter breakages. The simplest solution is
457 to use L<Sub::Name> to name the subroutine. Example:
459 # Original code, likely to break:
460 my $full_method_name = join('::', $package_name, $method_name);
461 *$full_method_name = sub { ... };
464 use Sub::Name 'subname';
465 my $full_method_name = join('::',$package_name, $method_name);
466 *$full_method_name = subname $full_method_name, sub { ... };
468 Additionally, you can take advantage of Catalyst's use of L<Class::MOP> and
469 install the closure using the appropriate metaclass. Example:
472 my $metaclass = Moose::Meta::Class->initialize($package_name);
473 $metaclass->add_method($method_name => sub { ... });
475 =head2 Hooking into application setup
477 To execute code during application start-up, the following snippet in MyApp.pm
481 my ($class, @args) = @_;
482 $class->NEXT::setup(@args);
483 ... # things to do after the actual setup
486 With Catalyst 5.80 this won't work anymore, because Catalyst no longer
487 uses NEXT.pm for method resolution. The functionality was only ever
488 originally operational as L<NEXT> remembers what methods have already
489 been called, and will not call them again.
491 Using this now causes infinite recursion between MyApp::setup and
492 Catalyst::setup, due to other backwards compatibility issues related to how
493 plugin setup works. Moose method modifiers like C<< before|after|around setup
494 => sub { ... }; >> also will not operate correctly on the setup method.
496 The right way to do it is this:
498 after setup_finalize => sub {
499 ... # things to do after the actual setup
502 The setup_finalize hook was introduced as a way to avoid this issue.
504 =head2 Components with a new method which returns false
506 Previously, if you had a component which inherited from Catalyst::COMPONENT,
507 but overrode the new method to return false, then your class's configuration
508 would be blessed into a hash on your behalf, and this would be returned from
509 the COMPONENT method.
511 This behavior makes no sense, and so has been removed. Implementing your own
512 C< new > method in components is B<highly> discouraged. Instead, you should
513 inherit the new method from Catalyst::Component, and use Moose's BUILD
514 functionality and/or Moose attributes to perform any construction work
515 necessary for your class.
517 =head2 __PACKAGE__->mk_accessor('meta');
519 Won't work due to a limitation of L<Moose>. This is currently being fixed
522 =head2 Class::Data::Inheritable side effects
524 Previously, writing to a class data accessor would copy the accessor method
525 down into your package.
527 This behavior has been removed. While the class data is still stored
528 per-class, it is stored on the metaclass of the class defining the accessor.
530 Therefore anything relying on the side effect of the accessor being copied down
533 The following test demonstrates the problem:
537 use base qw/Class::Data::Inheritable/;
538 __PACKAGE__->mk_classdata('foo');
543 use base qw/BaseClass/;
546 BaseClass->foo('base class');
547 Child->foo('sub class');
550 isnt(BaseClass->can('foo'), Child->can('foo'));
552 =head2 Extending Catalyst::Request or other classes in an ad hoc manner using mk_accessors
554 Previously, it was possible to add additional accessors to Catalyst::Request
555 (or other classes) by calling the mk_accessors class method.
557 This is no longer supported - users should make a subclass of the class whose
558 behavior they would like to change, rather than globally polluting the
561 =head2 Confused multiple inheritance with Catalyst::Component::COMPONENT
563 Previously, Catalyst's COMPONENT method would delegate to the method on
564 the right hand side, which could then delegate back again with
565 NEXT. This is poor practice, and in addition, makes no sense with C3
566 method dispatch order, and is therefore no longer supported.
568 If a COMPONENT method is detected in the inheritance hierarchy to the right
569 hand side of Catalyst::Component::COMPONENT, then the following warning
570 message will be emitted:
572 There is a COMPONENT method resolving after Catalyst::Component
575 The correct fix is to re-arrange your class's inheritance hierarchy so that the
576 COMPONENT method you would like to inherit is the first (left-hand most)
577 COMPONENT method in your @ISA.
579 =head2 Development server relying on environment variables
581 Previously, the development server would allow propagation of system
582 environment variables into the request environment, this has changed with the
583 adoption of Plack. You can use L<Plack::Middleware::ForceEnv> to achieve the
588 =head2 Actions in your application class
590 Having actions in your application class will now emit a warning at application
591 startup as this is deprecated. It is highly recommended that these actions are moved
592 into a MyApp::Controller::Root (as demonstrated by the scaffold application
593 generated by catalyst.pl).
595 This warning, also affects tests. You should move actions in your test,
596 creating a myTest::Controller::Root, like the following example:
598 package MyTest::Controller::Root;
603 use parent 'Catalyst::Controller';
605 __PACKAGE__->config(namespace => '');
608 my ( $self, $c ) = @_;
614 =head2 ::[MVC]:: naming scheme
616 Having packages called MyApp::[MVC]::XX is deprecated and can no longer be generated
619 This is still supported, but it is recommended that you rename your application
620 components to Model/View/Controller.
622 A warning will be issued at application startup if the ::[MVC]:: naming scheme is
625 =head2 Catalyst::Base
627 Any code using L<Catalyst::Base> will now emit a warning; this
628 module will be removed in a future release.
630 =head2 Methods in Catalyst::Dispatcher
632 The following methods in Catalyst::Dispatcher are implementation
633 details, which may change in the 5.8X release series, and therefore their use
634 is highly deprecated.
642 =item registered_dispatch_types
644 =item method_action_class
652 The first time one of these methods is called, a warning will be emitted:
654 Class $class is calling the deprecated method Catalyst::Dispatcher::$public_method_name,
655 this will be removed in Catalyst 5.9
657 You should B<NEVER> be calling any of these methods from application code.
659 Plugin authors and maintainers whose plugins currently call these methods
660 should change to using the public API, or, if you do not feel the public API
661 adequately supports your use case, please email the development list to
662 discuss what API features you need so that you can be appropriately supported.
664 =head2 Class files with names that don't correspond to the packages they define
666 In this version of Catalyst, if a component is loaded from disk, but no
667 symbols are defined in that component's name space after it is loaded, this
668 warning will be issued:
670 require $class was successful but the package is not defined.
672 This is to protect against confusing bugs caused by mistyping package names,
673 and will become a fatal error in a future version.
675 Please note that 'inner packages' (via L<Devel::InnerPackage>) are still fully
676 supported; this warning is only issued when component file naming does not map
677 to B<any> of the packages defined within that component.
679 =head2 $c->plugin method
681 Calling the plugin method is deprecated, and calling it at run time is B<highly
684 Instead you are recommended to use L<Catalyst::Model::Adaptor> or similar to
685 compose the functionality you need outside of the main application name space.
687 Calling the plugin method will not be supported past Catalyst 5.81.