3 Catalyst::Upgrading - Instructions for upgrading to the latest Catalyst
5 =head1 Upgrading to Catalyst 5.9
7 The major change is that L<Plack>, a toolkit for using the L<PSGI>
8 specification, now replaces most of the subclasses of L<Catalyst::Engine>. If
9 you are using one of the standard subclasses of L<Catalyst::Engine> this
10 should be a straightforward upgrade for you. It was a design goal for
11 this release to preserve as much backwards compatibility as possible.
12 However, since L<Plack> is different from L<Catalyst::Engine>, it is
13 possible that differences exist for edge cases. Therefore, we recommend
14 that care be taken with this upgrade and that testing should be greater
15 than would be the case with a minor point update. Please inform the
16 Catalyst developers of any problems so that we can fix them and
19 It is highly recommended that you become familiar with the L<Plack> ecosystem
20 and documentation. Being able to take advantage of L<Plack> development and
21 middleware is a major bonus to this upgrade. Documentation about how to
22 take advantage of L<Plack::Middleware> by writing your own C<< .psgi >> file
23 is contained in L<Catalyst::PSGI>.
25 If you have created a custom subclass of L<Catalyst:Engine>, you will
26 need to convert it to be a subclass of L<Plack::Handler>.
28 If you are using the L<Plack> engine, L<Catalyst::Engine::PSGI>, this new
29 release supersedes that code.
31 If you are using a subclass of L<Catalyst::Engine> that is aimed at
32 nonstandard or internal/testing uses, such as
33 L<Catalyst::Engine::Embeddable>, you should still be able to continue
36 Advice for specific subclasses of L<Catalyst::Engine> follows:
38 =head2 Upgrading the FastCGI Engine
40 No upgrade is needed if your myapp_fastcgi.pl script is already upgraded
41 to use L<Catalyst::Script::FastCGI>.
43 =head2 Upgrading the mod_perl / Apache Engines
45 The engines that are built upon the various iterations of mod_perl,
46 L<Catalyst::Engine::Apache::MP13> (for mod_perl 1, and Apache 1.x) and
47 L<Catalyst::Engine::Apache2::MP20> (for mod_perl 2, and Apache 2.x),
48 should be seamless upgrades and will work using using L<Plack::Handler::Apache1>
49 or L<Plack::Handler::Apache2> as required.
51 L<Catalyst::Engine::Apache2::MP19>, however, is no longer supported, as
52 Plack does not support mod_perl version 1.99. This is unlikely to be a
53 problem for anyone, as 1.99 was a brief beta-test release for mod_perl
54 2, and all users of mod_perl 1.99 are encouraged to upgrade to a
55 supported release of Apache 2 and mod_perl 2.
57 =head2 Upgrading the HTTP Engine
59 The default development server that comes with the L<Catalyst> distribution
60 should continue to work as expected with no changes as long as your C<myapp_server>
61 script is upgraded to use L<Catalyst::Script::HTTP>.
63 =head2 Upgrading the CGI Engine
65 If you were using L<Catalyst::Engine::CGI> there is no upgrade needed if your
66 myapp_cgi.pl script is already upgraded to use L<Catalyst::Script::CGI>.
68 =head2 Upgrading Catalyst::Engine::HTTP::Prefork
70 If you were using L<Catalyst::Engine::HTTP::Prefork> then L<Starman>
71 is automatically loaded. You should (at least) change your C<Makefile.PL>
74 You can regenerate your C<myapp_server.pl> script with C<catalyst.pl>
75 and implement a C<MyApp::Script::Server> class that looks like this:
77 package MyApp::Script::Server;
79 use namespace::autoclean;
81 extends 'CatalystX::Script::Server::Starman';
85 This takes advantage of the new script system, and will add a number of
86 options to the standard server script as extra options are added by
89 More information about these options can be seen at
90 L<CatalystX::Script::Server::Starman/SYNOPSIS>.
92 An alternate route to implement this functionality is to write a simple .psgi
93 file for your application, and then use the L<plackup> utility to start the
96 =head2 Upgrading the PSGI Engine
98 If you were using L<Catalyst::Engine::PSGI>, this new release supersedes
99 this engine in supporting L<Plack>. By default the Engine is now always
100 L<Plack>. As a result, you can remove the dependency on
101 L<Catalyst::Engine::PSGI> in your C<Makefile.PL>.
103 Applications that were using L<Catalyst::Engine::PSGI>
104 previously should entirely continue to work in this release with no changes.
106 However, if you have an C<app.psgi> script, then you no longer need to
107 specify the PSGI engine. Instead, the L<Catalyst> application class now
108 has a new method C<psgi_app> which returns a L<PSGI> compatible coderef
109 which you can wrap in the middleware of your choice.
111 Catalyst will use the .psgi for your application if it is located in the C<home>
112 directory of the application.
114 For example, if you were using L<Catalyst::Engine::PSGI> in the past, you will
115 have written (or generated) a C<script/myapp.psgi> file similar to this one:
120 MyCatalystApp->setup_engine('PSGI');
123 enable ... # enable your desired middleware
124 sub { MyCatalystApp->run(@_) };
127 Instead, you now say:
133 enable ... #enable your desired middleware
134 MyCatalystApp->psgi_app;
137 In the simplest case:
139 MyCatalystApp->setup_engine('PSGI');
140 my $app = sub { MyCatalystApp->run(@_) }
144 my $app = MyCatalystApp->psgi_app(@_);
148 my $app = sub { MyCatalystApp->psgi_app(@_) };
149 # If you make ^^ this mistake, your app won't work, and will confuse the hell out of you!
151 You can now move C<< script/myapp.psgi >> to C<< myapp.psgi >>, and the built-in
152 Catalyst scripts and your test suite will start using your .psgi file.
154 B<NOTE:> If you rename your .psgi file without these modifications, then
155 any tests run via L<Catalyst::Test> will not be compatible with the new
156 release, and will result in the development server starting, rather than
157 the expected test running.
159 B<NOTE:> If you are directly accessing C<< $c->req->env >> to get the PSGI
160 environment then this accessor is moved to C<< $c->engine->env >>,
161 you will need to update your code.
163 =head2 Engines which are known to be broken
165 The following engines B<DO NOT> work as of Catalyst version 5.9. The
166 core team will be happy to work with the developers and/or users of
167 these engines to help them port to the new Plack/Engine system, but for
168 now, applications which are currently using these engines B<WILL NOT>
169 run without modification to the engine code.
173 =item Catalyst::Engine::Wx
175 =item Catalyst::Engine::Zeus
177 =item Catalyst::Engine::JobQueue::POE
179 =item Catalyst::Engine::XMPP2
181 =item Catalyst::Engine::SCGI
185 =head2 Engines with unknown status
187 The following engines are untested or have unknown compatibility.
188 Reports are highly encouraged:
192 =item Catalyst::Engine::Mojo
194 =item Catalyst::Engine::Server (marked as Deprecated)
196 =item Catalyst::Engine::HTTP::POE (marked as Deprecated)
200 =head2 Plack functionality
202 See L<Catalyst::PSGI>.
206 Tests should generally work the same in Catalyst 5.9, but there are
209 Previously, if using L<Catalyst::Test> and doing local requests (against
210 a local server), if the application threw an exception then this
211 exception propagated into the test.
213 This behavior has been removed, and now a 500 response will be returned
214 to the test. This change standardizes behavior, so that local test
215 requests behave similarly to remote requests.
217 =head2 Regex dispatch type is deprecated.
219 The Regex dispatchtype (L<Catalyst::DispatchType::Regex>) has been deprecated.
221 You are encouraged to move your application to Chained dispatch (L<Catalyst::DispatchType::Chained>).
223 If you cannot do so, please add a dependency to Catalyst::DispatchType::Regex to your application's
226 =head1 Upgrading to Catalyst 5.80
228 Most applications and plugins should run unaltered on Catalyst 5.80.
230 However, a lot of refactoring work has taken place, and several changes have
231 been made which could cause incompatibilities. If your application or plugin
232 is using deprecated code, or relying on side effects, then you could have
233 issues upgrading to this release.
235 Most issues found with existing components have been easy to
236 solve. This document provides a complete description of behavior changes
237 which may cause compatibility issues, and of new Catalyst warnings which
240 If you think you have found an upgrade-related issue which is not covered in
241 this document, please email the Catalyst list to discuss the problem.
243 =head1 Moose features
245 =head2 Application class roles
247 You can only apply method modifiers after the application's C<< ->setup >>
248 method has been called. This means that modifiers will not work with methods
249 run during the call to C<< ->setup >>.
251 See L<Catalyst::Manual::ExtendingCatalyst> for more information about using
252 L<Moose> in your applications.
254 =head2 Controller actions in Moose roles
256 You can use L<MooseX::MethodAttributes::Role> if you want to declare actions
259 =head2 Using Moose in Components
261 The correct way to use Moose in a component in a both forward and backwards
264 package TestApp::Controller::Root;
266 BEGIN { extends 'Catalyst::Component' }; # Or ::Controller, or whatever
268 See L<Components which inherit from Moose::Object before Catalyst::Component>.
270 =head1 Known backwards compatibility breakages
272 =head2 Applications in a single file
274 Applications must be in their own file, and loaded at compile time. This
275 issue generally only affects the tests of CPAN distributions. Your
276 application will fail if you try to define an application inline in a
277 block, and use plugins which supply a C< new > method, then use that
278 application latter in tests within the same file.
280 This is due to the fact that Catalyst is inlining a new method on your
281 application class allowing it to be compatible with Moose. The method
282 used to do this changed in 5.80004 to avoid the possibility of reporting
283 an 'Unknown Error' if your application failed to compile.
285 =head2 Issues with Class::C3
287 Catalyst 5.80 uses the L<Algorithm::C3> method dispatch order. This is
288 built into Perl 5.10, and comes via L<Class::C3> for Perl 5.8. This
289 replaces L<NEXT> with L<Class::C3::Adopt::NEXT>, forcing all components
290 to resolve methods using C3, rather than the unpredictable dispatch
293 This issue manifests itself by your application failing to start due to an
294 error message about having a non-linear @ISA.
296 The Catalyst plugin most often causing this is
297 L<Catalyst::Plugin::Session::Store::FastMmap> - if you are using this
298 plugin and see issues, then please upgrade your plugins, as it has been
299 fixed. Note that Makefile.PL in the distribution will warn about known
300 incompatible components.
302 This issue can, however, be found in your own application - the only solution is
303 to go through each base class of the class the error was reported against, until
304 you identify the ones in conflict, and resolve them.
306 To be able to generate a linear @ISA, the list of superclasses for each
307 class must be resolvable using the C3 algorithm. Unfortunately, when
308 superclasses are being used as mixins (to add functionality used in your class),
309 and with multiple inheritance, it is easy to get this wrong.
311 Most common is the case of:
313 package Component1; # Note, this is the common case
314 use base qw/Class::Accessor::Fast Class::Data::Inheritable/;
316 package Component2; # Accidentally saying it this way causes a failure
317 use base qw/Class::Data::Inheritable Class::Accessor::Fast/;
320 use base qw/Component1 Component2/;
322 Any situation like this will cause your application to fail to start.
324 For additional documentation about this issue, and how to resolve it, see
325 L<Class::C3::Adopt::NEXT>.
327 =head2 Components which inherit from Moose::Object before Catalyst::Component
329 Moose components which say:
331 package TestApp::Controller::Example;
333 extends qw/Moose::Object Catalyst::Component/;
335 to use the constructor provided by Moose, while working (if you do some hacks
336 with the C< BUILDARGS > method), will not work with Catalyst 5.80 as
337 C<Catalyst::Component> inherits from C<Moose::Object>, and so C< @ISA > fails
340 The correct way to use Moose in a component in a both forward and backwards
343 package TestApp::Controller::Root;
345 BEGIN { extends 'Catalyst::Component' }; # Or ::Controller, or whatever
347 Note that the C< extends > declaration needs to occur in a begin block for
348 L<attributes> to operate correctly.
350 This way you do not inherit directly from C<Moose::Object>
351 yourself. Having components which do not inherit their constructor from
352 C<Catalyst::Component> is B<unsupported>, and has never been recommended,
353 therefore you're on your own if you're using this technique. You'll need
354 to detect the version of Catalyst your application is running, and deal
355 with it appropriately.
357 You also don't get the L<Moose::Object> constructor, and therefore attribute
358 initialization will not work as normally expected. If you want to use Moose
359 attributes, then they need to be made lazy to correctly initialize.
361 Note that this only applies if your component needs to maintain component
362 backwards compatibility for Catalyst versions before 5.71001 - in 5.71001
363 attributes work as expected, and the BUILD method is called normally
364 (although BUILDARGS is not).
366 If you depend on Catalyst 5.8, then B<all> Moose features work as expected.
368 You will also see this issue if you do the following:
370 package TestApp::Controller::Example;
372 use base 'Catalyst::Controller';
374 as C< use base > appends to @ISA.
376 =head3 use Moose in MyApp
378 Similar to the above, this will also fail:
387 If you need to use Moose in your application class (e.g. for method modifiers
388 etc.) then the correct technique is:
396 __PACKAGE__->config( name => 'MyApp' );
397 __PACKAGE__->setup(qw/
401 =head2 Anonymous closures installed directly into the symbol table
403 If you have any code which installs anonymous subroutine references directly
404 into the symbol table, you may encounter breakages. The simplest solution is
405 to use L<Sub::Name> to name the subroutine. Example:
407 # Original code, likely to break:
408 my $full_method_name = join('::', $package_name, $method_name);
409 *$full_method_name = sub { ... };
412 use Sub::Name 'subname';
413 my $full_method_name = join('::',$package_name, $method_name);
414 *$full_method_name = subname $full_method_name, sub { ... };
416 Additionally, you can take advantage of Catalyst's use of L<Class::MOP> and
417 install the closure using the appropriate metaclass. Example:
420 my $metaclass = Moose::Meta::Class->initialize($package_name);
421 $metaclass->add_method($method_name => sub { ... });
423 =head2 Hooking into application setup
425 To execute code during application start-up, the following snippet in MyApp.pm
429 my ($class, @args) = @_;
430 $class->NEXT::setup(@args);
431 ... # things to do after the actual setup
434 With Catalyst 5.80 this won't work anymore, because Catalyst no longer
435 uses NEXT.pm for method resolution. The functionality was only ever
436 originally operational as L<NEXT> remembers what methods have already
437 been called, and will not call them again.
439 Using this now causes infinite recursion between MyApp::setup and
440 Catalyst::setup, due to other backwards compatibility issues related to how
441 plugin setup works. Moose method modifiers like C<< before|after|around setup
442 => sub { ... }; >> also will not operate correctly on the setup method.
444 The right way to do it is this:
446 after setup_finalize => sub {
447 ... # things to do after the actual setup
450 The setup_finalize hook was introduced as a way to avoid this issue.
452 =head2 Components with a new method which returns false
454 Previously, if you had a component which inherited from Catalyst::COMPONENT,
455 but overrode the new method to return false, then your class's configuration
456 would be blessed into a hash on your behalf, and this would be returned from
457 the COMPONENT method.
459 This behavior makes no sense, and so has been removed. Implementing your own
460 C< new > method in components is B<highly> discouraged. Instead, you should
461 inherit the new method from Catalyst::Component, and use Moose's BUILD
462 functionality and/or Moose attributes to perform any construction work
463 necessary for your class.
465 =head2 __PACKAGE__->mk_accessor('meta');
467 Won't work due to a limitation of L<Moose>. This is currently being fixed
470 =head2 Class::Data::Inheritable side effects
472 Previously, writing to a class data accessor would copy the accessor method
473 down into your package.
475 This behavior has been removed. While the class data is still stored
476 per-class, it is stored on the metaclass of the class defining the accessor.
478 Therefore anything relying on the side effect of the accessor being copied down
481 The following test demonstrates the problem:
485 use base qw/Class::Data::Inheritable/;
486 __PACKAGE__->mk_classdata('foo');
491 use base qw/BaseClass/;
494 BaseClass->foo('base class');
495 Child->foo('sub class');
498 isnt(BaseClass->can('foo'), Child->can('foo'));
500 =head2 Extending Catalyst::Request or other classes in an ad hoc manner using mk_accessors
502 Previously, it was possible to add additional accessors to Catalyst::Request
503 (or other classes) by calling the mk_accessors class method.
505 This is no longer supported - users should make a subclass of the class whose
506 behavior they would like to change, rather than globally polluting the
509 =head2 Confused multiple inheritance with Catalyst::Component::COMPONENT
511 Previously, Catalyst's COMPONENT method would delegate to the method on
512 the right hand side, which could then delegate back again with
513 NEXT. This is poor practice, and in addition, makes no sense with C3
514 method dispatch order, and is therefore no longer supported.
516 If a COMPONENT method is detected in the inheritance hierarchy to the right
517 hand side of Catalyst::Component::COMPONENT, then the following warning
518 message will be emitted:
520 There is a COMPONENT method resolving after Catalyst::Component
523 The correct fix is to re-arrange your class's inheritance hierarchy so that the
524 COMPONENT method you would like to inherit is the first (left-hand most)
525 COMPONENT method in your @ISA.
527 =head2 Development server relying on environment variables
529 Previously, the development server would allow propagation of system
530 environment variables into the request environment, this has changed with the
531 adoption of Plack. You can use L<Plack::Middleware::ForceEnv> to achieve the
536 =head2 Actions in your application class
538 Having actions in your application class will now emit a warning at application
539 startup as this is deprecated. It is highly recommended that these actions are moved
540 into a MyApp::Controller::Root (as demonstrated by the scaffold application
541 generated by catalyst.pl).
543 This warning, also affects tests. You should move actions in your test,
544 creating a myTest::Controller::Root, like the following example:
546 package MyTest::Controller::Root;
551 use parent 'Catalyst::Controller';
553 __PACKAGE__->config(namespace => '');
556 my ( $self, $c ) = @_;
562 =head2 ::[MVC]:: naming scheme
564 Having packages called MyApp::[MVC]::XX is deprecated and can no longer be generated
567 This is still supported, but it is recommended that you rename your application
568 components to Model/View/Controller.
570 A warning will be issued at application startup if the ::[MVC]:: naming scheme is
573 =head2 Catalyst::Base
575 Any code using L<Catalyst::Base> will now emit a warning; this
576 module will be removed in a future release.
578 =head2 Methods in Catalyst::Dispatcher
580 The following methods in Catalyst::Dispatcher are implementation
581 details, which may change in the 5.8X release series, and therefore their use
582 is highly deprecated.
590 =item registered_dispatch_types
592 =item method_action_class
600 The first time one of these methods is called, a warning will be emitted:
602 Class $class is calling the deprecated method Catalyst::Dispatcher::$public_method_name,
603 this will be removed in Catalyst 5.9
605 You should B<NEVER> be calling any of these methods from application code.
607 Plugin authors and maintainers whose plugins currently call these methods
608 should change to using the public API, or, if you do not feel the public API
609 adequately supports your use case, please email the development list to
610 discuss what API features you need so that you can be appropriately supported.
612 =head2 Class files with names that don't correspond to the packages they define
614 In this version of Catalyst, if a component is loaded from disk, but no
615 symbols are defined in that component's name space after it is loaded, this
616 warning will be issued:
618 require $class was successful but the package is not defined.
620 This is to protect against confusing bugs caused by mistyping package names,
621 and will become a fatal error in a future version.
623 Please note that 'inner packages' (via L<Devel::InnerPackage>) are still fully
624 supported; this warning is only issued when component file naming does not map
625 to B<any> of the packages defined within that component.
627 =head2 $c->plugin method
629 Calling the plugin method is deprecated, and calling it at run time is B<highly
632 Instead you are recommended to use L<Catalyst::Model::Adaptor> or similar to
633 compose the functionality you need outside of the main application name space.
635 Calling the plugin method will not be supported past Catalyst 5.81.