Commit | Line | Data |
8c57b129 |
1 | =head1 NAME |
2 | |
3 | Catalyst::Upgrading - Instructions for upgrading to the latest Catalyst |
4 | |
eb7421fd |
5 | =head1 Upgrading to Catalyst 5.90121 |
6 | |
7 | A new C<log_stats> method has been added. This will only affect |
8 | subclasses that have a method with this name added. |
9 | |
5e7e6b27 |
10 | =head1 Upgrading to Catalyst 5.90100 |
11 | |
12 | We changed the way the middleware stash works so that it no longer localizes |
13 | the PSGI env hashref. This was done to fix bugs where people set PSGI ENV hash |
79fb8f95 |
14 | keys and found them to disappear in certain cases. It also means that now if |
5e7e6b27 |
15 | a sub applications sets stash variables, that stash will now bubble up to the |
16 | parent application. This may be a breaking change for you since previous |
17 | versions of this code did not allow that. A workaround is to explicitly delete |
18 | stash keys in your sub application before returning control to the parent |
19 | application. |
20 | |
6b9f9ef7 |
21 | =head1 Upgrading to Catalyst 5.90097 |
22 | |
23 | In older versions of Catalyst one could construct a L<URI> with a fragment (such as |
24 | https://localhost/foo/bar#fragment) by using a '#' in the path or final argument, for |
25 | example: |
26 | |
6b9f9ef7 |
27 | $c->uri_for($action, 'foo#fragment'); |
28 | |
29 | This behavior was never documented and would break if using the Unicode plugin, or when |
30 | adding a query to the arguments: |
31 | |
32 | $c->uri_for($action, 'foo#fragment', +{ a=>1, b=>2}); |
33 | |
34 | would define a fragment like "#fragment?a=1&b=2". |
35 | |
36 | When we introduced UTF-8 encoding by default in Catalyst 5.9008x this side effect behavior |
37 | was broken since we started encoding the '#' when it was part of the URI path. |
38 | |
39 | In version 5.90095 and 5.90096 we attempted to fix this, but all we managed to do was break |
40 | people with URIs that included '#' as part of the path data, when it was not expected to |
41 | be a fragment delimiter. |
42 | |
43 | In general L<Catalyst> prefers an explicit specification rather than relying on side effects |
44 | or domain specific mini languages. As a result we are now defining how to set a fragment |
45 | for a URI via ->uri_for: |
46 | |
47 | $c->uri_for($action_or_path, \@captures_or_args, @args, \$query, \$fragment); |
48 | |
49 | If you are relying on the previous side effect behavior your URLs will now encode the '#' |
50 | delimiter, which is going to be a breaking change for you. You need to alter your code |
51 | to match the new specification or modify uri_for for your local case. Patches to solve |
02336198 |
52 | this are very welcomed, as long as they don't break existing test cases. |
53 | |
a1dba4cf |
54 | B<NOTE> If you are using the string form of the first argument: |
02336198 |
55 | |
56 | $c->uri_for('/foo/bar#baz') |
57 | |
58 | construction, we do not attempt to encode this and it will make a URL with a |
59 | fragment of 'baz'. |
60 | |
6b9f9ef7 |
61 | |
62 | =head1 Upgrading to Catalyst 5.90095 |
e5ac67e5 |
63 | |
64 | The method C<last_error> in L</Catalyst> was actually returning the first error. This has |
65 | been fixed but there is a small chance it could be a breaking issue for you. If this gives |
66 | you trouble changing to C<shift_errors> is the easiest workaround (although that does |
67 | modify the error stack so if you are relying on that not being changed you should try something |
68 | like @{$c->errors}[-1] instead. Since this method is relatively new and the cases when the |
69 | error stack actually has more than one error in it, we feel the exposure is very low, but bug |
70 | reports are very welcomed. |
71 | |
ec4d7259 |
72 | =head1 Upgrading to Catalyst 5.90090 |
73 | |
74 | L<Catalyst::Utils> has a new method 'inject_component' which works the same as the method of |
75 | the same name in L<CatalystX::InjectComponent>. You should start converting any |
76 | use of the non core method in your code as future changes to Catalyst will be |
79fb8f95 |
77 | synchronized to the core method first. We reserve the right to cease support |
ec4d7259 |
78 | of the non core version should we reach a point in time where it cannot be |
79 | properly supported as an external module. Luckily this should be a trivial |
0fb9642d |
80 | search and replace. Change all occurrences of: |
ec4d7259 |
81 | |
82 | CatalystX::InjectComponent->inject(...) |
83 | |
84 | Into |
85 | |
86 | Catalyst::Utils::inject_component(...) |
87 | |
88 | and we expect everything to work the same (we'd consider it not working the same |
89 | to be a bug, and please report it.) |
90 | |
a791afa9 |
91 | We also cored features from L<CatalystX::RoleApplicator> to compose a role into the |
92 | request, response and stats classes. The main difference is that with L<CatalystX::RoleApplicator> |
93 | you did: |
94 | |
95 | package MyApp; |
96 | |
97 | use Catalyst; |
98 | use CatalystX::RoleApplicator; |
99 | |
100 | __PACKAGE__->apply_request_class_roles( |
101 | qw/My::Request::Role Other::Request::Role/); |
102 | |
103 | Whereas now we have three class attributes, 'request_class_traits', 'response_class_traits' |
104 | and 'stats_class_traits', so you use like this (note this value is an ArrayRef) |
105 | |
106 | |
107 | package MyApp; |
108 | |
109 | use Catalyst; |
110 | |
111 | __PACKAGE__->request_class_traits([qw/ |
112 | My::Request::Role |
113 | Other::Request::Role/]); |
114 | |
115 | (And the same for response_class_traits and stats_class_traits. We left off the |
116 | traits for Engine, since that class does a lot less nowadays, and dispatcher. If you |
117 | used those and can share a use case, we'd be likely to support them. |
118 | |
3e560748 |
119 | Lastly, we have some of the feature from L<CatalystX::ComponentsFromConfig> in |
120 | core. This should mostly work the same way in core, except for now the |
121 | core version does not create an automatic base wrapper class for your configured |
122 | components (it requires these to be catalyst components and injects them directly. |
123 | So if you make heavy use of custom base classes in L<CatalystX::ComponentsFromConfig> |
124 | you might need a bit of work to use the core version (although there is no reason |
125 | to stop using L<CatalystX::ComponentsFromConfig> since it should continue to work |
126 | fine and we'd consider issues with it to be bugs). Here's one way to map from |
127 | L<CatalystX::ComponentsFromConfig> to core: |
128 | |
129 | In L<CatalystX::ComponentsFromConfig>: |
130 | |
131 | MyApp->config( |
132 | 'Model::MyClass' => { |
044e7667 |
133 | class => 'MyClass', |
134 | args => { %args }, |
3e560748 |
135 | |
136 | }); |
137 | |
044e7667 |
138 | and now in core: |
139 | |
140 | MyApp->config( |
141 | inject_components => { |
142 | 'Model::MyClass' => { from_component => 'My::Class' }, |
143 | }, |
144 | 'Model::MyClass' => { |
145 | %args |
146 | }, |
147 | ); |
148 | |
79fb8f95 |
149 | Although the core behavior requires more code, it better separates concerns |
150 | as well as plays more into core Catalyst expectations of how configuration should |
044e7667 |
151 | look. |
3e560748 |
152 | |
153 | Also we added a new develop console mode only warning when you call a component |
154 | with arguments that don't expect or do anything meaningful with those args. Its |
88e5a8b0 |
155 | possible if you are logging debug mode in production (please don't...) this |
3e560748 |
156 | could add verbosity to those logs if you also happen to be calling for components |
157 | and passing pointless arguments. We added this warning to help people not make this |
158 | error and to better understand the component resolution flow. |
159 | |
7a504990 |
160 | =head1 Upgrading to Catalyst 5.90085 |
161 | |
162 | In this version of Catalyst we made a small change to Chained Dispatching so |
163 | that when two or more actions all have the same path specification AND they |
164 | all have Args(0), we break the tie by choosing the last action defined, and |
165 | not the first one defined. This was done to normalize Chaining to following |
166 | the 'longest Path wins, and when several actions match the same Path specification |
167 | we choose the last defined.' rule. Previously Args(0) was hard coded to be a special |
168 | case such that the first action defined would match (which is not the case when |
169 | Args is not zero.) |
170 | |
171 | Its possible that this could be a breaking change for you, if you had used |
172 | action roles (custom or otherwise) to add additional matching rules to differentiate |
173 | between several Args(0) actions that share the same root action chain. For |
174 | example if you have code now like this: |
175 | |
176 | sub check_default :Chained(/) CaptureArgs(0) { ... } |
177 | |
178 | sub default_get :Chained('check_default') PathPart('') Args(0) GET { |
179 | pop->res->body('get3'); |
180 | } |
181 | |
182 | sub default_post :Chained('check_default') PathPart('') Args(0) POST { |
183 | pop->res->body('post3'); |
184 | } |
185 | |
186 | sub chain_default :Chained('check_default') PathPart('') Args(0) { |
187 | pop->res->body('chain_default'); |
188 | } |
189 | |
190 | The way that chaining will work previous is that when two or more equal actions can |
191 | match, the 'top' one wins. So if the request is "GET .../check_default" BOTH |
192 | actions 'default_get' AND 'chain_default' would match. To break the tie in |
193 | the case when Args is 0, we'd previous take the 'top' (or first defined) action. |
194 | Unfortunately this treatment of Args(0) is special case. In all other cases |
195 | we choose the 'last defined' action to break a tie. So this version of |
196 | Catalyst changed the dispatcher to make Args(0) no longer a special case for |
197 | breaking ties. This means that the above code must now become: |
198 | |
199 | sub check_default :Chained(/) CaptureArgs(0) { ... } |
200 | |
201 | sub chain_default :Chained('check_default') PathPart('') Args(0) { |
202 | pop->res->body('chain_default'); |
203 | } |
204 | |
205 | sub default_get :Chained('check_default') PathPart('') Args(0) GET { |
206 | pop->res->body('get3'); |
207 | } |
208 | |
209 | sub default_post :Chained('check_default') PathPart('') Args(0) POST { |
210 | pop->res->body('post3'); |
211 | } |
212 | |
213 | If we want it to work as expected (for example we we GET to match 'default_get' and |
214 | POST to match 'default_post' and any other http Method to match 'chain_default'). |
215 | |
216 | In other words Arg(0) and chained actions must now follow the normal rule where |
217 | in a tie the last defined action wins and you should place all your less defined |
218 | or 'catch all' actions first. |
219 | |
220 | If this causes you trouble and you can't fix your code to conform, you may set the |
221 | application configuration setting "use_chained_args_0_special_case" to true and |
222 | that will revert you code to the previous behavior. |
223 | |
6cf77e11 |
224 | =head2 More backwards compatibility options with UTF-8 changes |
225 | |
79fb8f95 |
226 | In order to give better backwards compatibility with the 5.90080+ UTF-8 changes |
6cf77e11 |
227 | we've added several configuration options around control of how we try to decode |
228 | your URL keywords / query parameters. |
229 | |
230 | C<do_not_decode_query> |
231 | |
232 | If true, then do not try to character decode any wide characters in your |
79fb8f95 |
233 | request URL query or keywords. Most readings of the relevant specifications |
6cf77e11 |
234 | suggest these should be UTF-* encoded, which is the default that L<Catalyst> |
79fb8f95 |
235 | will use, however if you are creating a lot of URLs manually or have external |
6cf77e11 |
236 | evil clients, this might cause you trouble. If you find the changes introduced |
88e5a8b0 |
237 | in Catalyst version 5.90080+ break some of your query code, you may disable |
6cf77e11 |
238 | the UTF-8 decoding globally using this configuration. |
239 | |
240 | This setting takes precedence over C<default_query_encoding> and |
241 | C<decode_query_using_global_encoding> |
242 | |
243 | C<default_query_encoding> |
244 | |
245 | By default we decode query and keywords in your request URL using UTF-8, which |
79fb8f95 |
246 | is our reading of the relevant specifications. This setting allows one to |
6cf77e11 |
247 | specify a fixed value for how to decode your query. You might need this if |
248 | you are doing a lot of custom encoding of your URLs and not using UTF-8. |
249 | |
250 | This setting take precedence over C<decode_query_using_global_encoding>. |
251 | |
252 | C<decode_query_using_global_encoding> |
253 | |
254 | Setting this to true will default your query decoding to whatever your |
255 | general global encoding is (the default is UTF-8). |
256 | |
257 | |
b8b29bac |
258 | =head1 Upgrading to Catalyst 5.90080 |
259 | |
260 | UTF8 encoding is now default. For temporary backwards compatibility, if this |
261 | change is causing you trouble, you can disable it by setting the application |
262 | configuration option to undef: |
263 | |
264 | MyApp->config(encoding => undef); |
265 | |
266 | But please consider this a temporary measure since it is the intention that |
267 | UTF8 is enabled going forwards and the expectation is that other ecosystem |
268 | projects will assume this as well. At some point you application will not |
269 | correctly function without this setting. |
270 | |
0d94e986 |
271 | As of 5.90084 we've added two additional configuration flags for more selective |
272 | control over some encoding changes: 'skip_body_param_unicode_decoding' and |
273 | 'skip_complex_post_part_handling'. You may use these to more selectively |
274 | disable new features while you are seeking a long term fix. Please review |
275 | CONFIGURATION in L<Catalyst>. |
276 | |
d63cc9c8 |
277 | For further information, please see L<Catalyst::UTF8> |
278 | |
b8b29bac |
279 | A number of projects in the wider ecosystem required minor updates to be able |
280 | to work correctly. Here's the known list: |
281 | |
282 | L<Catalyst::View::TT>, L<Catalyst::View::Mason>, L<Catalyst::View::HTML::Mason>, |
283 | L<Catalyst::View::Xslate>, L<Test::WWW::Mechanize::Catalyst> |
284 | |
285 | You will need to update to modern versions in most cases, although quite a few |
286 | of these only needed minor test case and documentation changes so you will need |
287 | to review the changelog of each one that is relevant to you to determine your |
288 | true upgrade needs. |
289 | |
78acc1f7 |
290 | =head1 Upgrading to Catalyst 5.90060 |
291 | |
292 | Starting in the v5.90059_001 development release, the regexp dispatch type is |
293 | no longer automatically included as a dependency. If you are still using this |
294 | dispatch type, you need to add L<Catalyst::DispatchType::Regex> into your build |
295 | system. |
296 | |
297 | The standalone distribution of Regexp will be supported for the time being, but |
298 | should we find that supporting it prevents us from moving L<Catalyst> forward |
299 | in necessary ways, we reserve the right to drop that support. It is highly |
300 | recommended that you use this last stage of deprecation to change your code. |
301 | |
ba7766f8 |
302 | =head1 Upgrading to Catalyst 5.90040 |
717fc5c9 |
303 | |
8275d3b9 |
304 | =head2 Catalyst::Plugin::Unicode::Encoding is now core |
305 | |
306 | The previously stand alone Unicode support module L<Catalyst::Plugin::Unicode::Encoding> |
307 | has been brought into core as a default plugin. Going forward, all you need is |
308 | to add a configuration setting for the encoding type. For example: |
309 | |
310 | package Myapp::Web; |
311 | |
312 | use Catalyst; |
313 | |
314 | __PACKAGE__->config( encoding => 'UTF-8' ); |
315 | |
316 | Please note that this is different from the old stand alone plugin which applied |
317 | C<UTF-8> encoding by default (that is, if you did not set an explicit |
5fa5b709 |
318 | C<encoding> configuration value, it assumed you wanted UTF-8). In order to |
8275d3b9 |
319 | preserve backwards compatibility you will need to explicitly turn it on via the |
320 | configuration setting. THIS MIGHT CHANGE IN THE FUTURE, so please consider |
321 | starting to test your application with proper UTF-8 support and remove all those |
322 | crappy hacks you munged into the code because you didn't know the Plugin |
323 | existed :) |
324 | |
325 | For people that are using the Plugin, you will note a startup warning suggesting |
326 | that you can remove it from the plugin list. When you do so, please remember to |
327 | add the configuration setting, since you can no longer rely on the default being |
328 | UTF-8. We'll add it for you if you continue to use the stand alone plugin and |
329 | we detect this, but this backwards compatibility shim will likely be removed in |
330 | a few releases (trying to clean up the codebase after all). |
331 | |
332 | If you have trouble with any of this, please bring it to the attention of the |
333 | Catalyst maintainer group. |
334 | |
335 | =head2 basic async and event loop support |
336 | |
717fc5c9 |
337 | This version of L<Catalyst> offers some support for using L<AnyEvent> and |
e37f92f5 |
338 | L<IO::Async> event loops in your application. These changes should work |
339 | fine for most applications however if you are already trying to perform |
340 | some streaming, minor changes in this area of the code might affect your |
4e6e0ab2 |
341 | functionality. Please see L<Catalyst::Response\write_fh> for more and for a |
342 | basic example. |
8275d3b9 |
343 | |
344 | We consider this feature experimental. We will try not to break it, but we |
345 | reserve the right to make necessary changes to fix major issues that people |
346 | run into when the use this functionality in the wild. |
717fc5c9 |
347 | |
ba7766f8 |
348 | =head1 Upgrading to Catalyst 5.90030 |
349 | |
350 | =head2 Regex dispatch type is deprecated. |
351 | |
352 | The Regex dispatchtype (L<Catalyst::DispatchType::Regex>) has been deprecated. |
353 | |
354 | You are encouraged to move your application to Chained dispatch (L<Catalyst::DispatchType::Chained>). |
355 | |
356 | If you cannot do so, please add a dependency to Catalyst::DispatchType::Regex to your application's |
357 | Makefile.PL |
358 | |
dacd8b0e |
359 | =head1 Upgrading to Catalyst 5.9 |
5d5f4a73 |
360 | |
e6006848 |
361 | The major change is that L<Plack>, a toolkit for using the L<PSGI> |
862a7989 |
362 | specification, now replaces most of the subclasses of L<Catalyst::Engine>. If |
e6006848 |
363 | you are using one of the standard subclasses of L<Catalyst::Engine> this |
364 | should be a straightforward upgrade for you. It was a design goal for |
365 | this release to preserve as much backwards compatibility as possible. |
366 | However, since L<Plack> is different from L<Catalyst::Engine>, it is |
367 | possible that differences exist for edge cases. Therefore, we recommend |
368 | that care be taken with this upgrade and that testing should be greater |
369 | than would be the case with a minor point update. Please inform the |
370 | Catalyst developers of any problems so that we can fix them and |
371 | incorporate tests. |
5d5f4a73 |
372 | |
773b3b08 |
373 | It is highly recommended that you become familiar with the L<Plack> ecosystem |
ae908e7e |
374 | and documentation. Being able to take advantage of L<Plack> development and |
375 | middleware is a major bonus to this upgrade. Documentation about how to |
376 | take advantage of L<Plack::Middleware> by writing your own C<< .psgi >> file |
377 | is contained in L<Catalyst::PSGI>. |
5d5f4a73 |
378 | |
e6006848 |
379 | If you have created a custom subclass of L<Catalyst:Engine>, you will |
380 | need to convert it to be a subclass of L<Plack::Handler>. |
5d5f4a73 |
381 | |
382 | If you are using the L<Plack> engine, L<Catalyst::Engine::PSGI>, this new |
773b3b08 |
383 | release supersedes that code. |
5d5f4a73 |
384 | |
e6006848 |
385 | If you are using a subclass of L<Catalyst::Engine> that is aimed at |
386 | nonstandard or internal/testing uses, such as |
387 | L<Catalyst::Engine::Embeddable>, you should still be able to continue |
388 | using that engine. |
5d5f4a73 |
389 | |
390 | Advice for specific subclasses of L<Catalyst::Engine> follows: |
391 | |
93d60cae |
392 | =head2 Upgrading the FastCGI Engine |
5d5f4a73 |
393 | |
e6006848 |
394 | No upgrade is needed if your myapp_fastcgi.pl script is already upgraded |
395 | to use L<Catalyst::Script::FastCGI>. |
5d5f4a73 |
396 | |
93d60cae |
397 | =head2 Upgrading the mod_perl / Apache Engines |
5d5f4a73 |
398 | |
e6006848 |
399 | The engines that are built upon the various iterations of mod_perl, |
14148e06 |
400 | L<Catalyst::Engine::Apache::MP13> (for mod_perl 1, and Apache 1.x) and |
862a7989 |
401 | L<Catalyst::Engine::Apache2::MP20> (for mod_perl 2, and Apache 2.x), |
bd85860b |
402 | should be seamless upgrades and will work using L<Plack::Handler::Apache1> |
14148e06 |
403 | or L<Plack::Handler::Apache2> as required. |
5d5f4a73 |
404 | |
e6006848 |
405 | L<Catalyst::Engine::Apache2::MP19>, however, is no longer supported, as |
862a7989 |
406 | Plack does not support mod_perl version 1.99. This is unlikely to be a |
407 | problem for anyone, as 1.99 was a brief beta-test release for mod_perl |
408 | 2, and all users of mod_perl 1.99 are encouraged to upgrade to a |
409 | supported release of Apache 2 and mod_perl 2. |
5d5f4a73 |
410 | |
93d60cae |
411 | =head2 Upgrading the HTTP Engine |
5d5f4a73 |
412 | |
040835f0 |
413 | The default development server that comes with the L<Catalyst> distribution |
414 | should continue to work as expected with no changes as long as your C<myapp_server> |
415 | script is upgraded to use L<Catalyst::Script::HTTP>. |
5d5f4a73 |
416 | |
93d60cae |
417 | =head2 Upgrading the CGI Engine |
5d5f4a73 |
418 | |
697a3e9e |
419 | If you were using L<Catalyst::Engine::CGI> there is no upgrade needed if your |
e6006848 |
420 | myapp_cgi.pl script is already upgraded to use L<Catalyst::Script::CGI>. |
5d5f4a73 |
421 | |
cf8eab35 |
422 | =head2 Upgrading Catalyst::Engine::HTTP::Prefork |
5d5f4a73 |
423 | |
040835f0 |
424 | If you were using L<Catalyst::Engine::HTTP::Prefork> then L<Starman> |
da9eab5a |
425 | is automatically loaded. You should (at least) change your C<Makefile.PL> |
426 | to depend on Starman. |
0ea8962d |
427 | |
da9eab5a |
428 | You can regenerate your C<myapp_server.pl> script with C<catalyst.pl> |
429 | and implement a C<MyApp::Script::Server> class that looks like this: |
430 | |
431 | package MyApp::Script::Server; |
432 | use Moose; |
433 | use namespace::autoclean; |
434 | |
435 | extends 'CatalystX::Script::Server::Starman'; |
436 | |
437 | 1; |
438 | |
e6006848 |
439 | This takes advantage of the new script system, and will add a number of |
440 | options to the standard server script as extra options are added by |
441 | Starman. |
da9eab5a |
442 | |
443 | More information about these options can be seen at |
444 | L<CatalystX::Script::Server::Starman/SYNOPSIS>. |
445 | |
446 | An alternate route to implement this functionality is to write a simple .psgi |
e6006848 |
447 | file for your application, and then use the L<plackup> utility to start the |
da9eab5a |
448 | server. |
5d5f4a73 |
449 | |
93d60cae |
450 | =head2 Upgrading the PSGI Engine |
5d5f4a73 |
451 | |
e6006848 |
452 | If you were using L<Catalyst::Engine::PSGI>, this new release supersedes |
453 | this engine in supporting L<Plack>. By default the Engine is now always |
454 | L<Plack>. As a result, you can remove the dependency on |
455 | L<Catalyst::Engine::PSGI> in your C<Makefile.PL>. |
8f912f0b |
456 | |
457 | Applications that were using L<Catalyst::Engine::PSGI> |
458 | previously should entirely continue to work in this release with no changes. |
459 | |
e6006848 |
460 | However, if you have an C<app.psgi> script, then you no longer need to |
461 | specify the PSGI engine. Instead, the L<Catalyst> application class now |
462 | has a new method C<psgi_app> which returns a L<PSGI> compatible coderef |
463 | which you can wrap in the middleware of your choice. |
8f912f0b |
464 | |
465 | Catalyst will use the .psgi for your application if it is located in the C<home> |
e6006848 |
466 | directory of the application. |
697a3e9e |
467 | |
93a57b4b |
468 | For example, if you were using L<Catalyst::Engine::PSGI> in the past, you will |
8f912f0b |
469 | have written (or generated) a C<script/myapp.psgi> file similar to this one: |
697a3e9e |
470 | |
471 | use Plack::Builder; |
472 | use MyCatalytApp; |
473 | |
474 | MyCatalystApp->setup_engine('PSGI'); |
475 | |
476 | builder { |
477 | enable ... # enable your desired middleware |
478 | sub { MyCatalystApp->run(@_) }; |
479 | }; |
480 | |
8f912f0b |
481 | Instead, you now say: |
697a3e9e |
482 | |
483 | use Plack::Builder; |
484 | use MyCatalystApp; |
485 | |
486 | builder { |
487 | enable ... #enable your desired middleware |
75d68821 |
488 | MyCatalystApp->psgi_app; |
697a3e9e |
489 | }; |
5d5f4a73 |
490 | |
34effbc7 |
491 | In the simplest case: |
8f912f0b |
492 | |
34effbc7 |
493 | MyCatalystApp->setup_engine('PSGI'); |
494 | my $app = sub { MyCatalystApp->run(@_) } |
495 | |
496 | becomes |
497 | |
34effbc7 |
498 | my $app = MyCatalystApp->psgi_app(@_); |
499 | |
500 | B<NOT>: |
501 | |
502 | my $app = sub { MyCatalystApp->psgi_app(@_) }; |
503 | # If you make ^^ this mistake, your app won't work, and will confuse the hell out of you! |
504 | |
e6006848 |
505 | You can now move C<< script/myapp.psgi >> to C<< myapp.psgi >>, and the built-in |
773b3b08 |
506 | Catalyst scripts and your test suite will start using your .psgi file. |
ad15c817 |
507 | |
e6006848 |
508 | B<NOTE:> If you rename your .psgi file without these modifications, then |
509 | any tests run via L<Catalyst::Test> will not be compatible with the new |
510 | release, and will result in the development server starting, rather than |
511 | the expected test running. |
93a57b4b |
512 | |
c47cd2ce |
513 | B<NOTE:> If you are directly accessing C<< $c->req->env >> to get the PSGI |
514 | environment then this accessor is moved to C<< $c->engine->env >>, |
515 | you will need to update your code. |
516 | |
e6006848 |
517 | =head2 Engines which are known to be broken |
93a57b4b |
518 | |
e6006848 |
519 | The following engines B<DO NOT> work as of Catalyst version 5.9. The |
520 | core team will be happy to work with the developers and/or users of |
521 | these engines to help them port to the new Plack/Engine system, but for |
522 | now, applications which are currently using these engines B<WILL NOT> |
523 | run without modification to the engine code. |
93a57b4b |
524 | |
525 | =over |
526 | |
527 | =item Catalyst::Engine::Wx |
528 | |
ad15c817 |
529 | =item Catalyst::Engine::Zeus |
530 | |
531 | =item Catalyst::Engine::JobQueue::POE |
532 | |
533 | =item Catalyst::Engine::XMPP2 |
534 | |
535 | =item Catalyst::Engine::SCGI |
536 | |
93a57b4b |
537 | =back |
538 | |
5d5f4a73 |
539 | =head2 Engines with unknown status |
540 | |
e6006848 |
541 | The following engines are untested or have unknown compatibility. |
542 | Reports are highly encouraged: |
5d5f4a73 |
543 | |
ad15c817 |
544 | =over |
545 | |
546 | =item Catalyst::Engine::Mojo |
547 | |
e6006848 |
548 | =item Catalyst::Engine::Server (marked as Deprecated) |
ad15c817 |
549 | |
e6006848 |
550 | =item Catalyst::Engine::HTTP::POE (marked as Deprecated) |
ad15c817 |
551 | |
552 | =back |
5d5f4a73 |
553 | |
3f22de0b |
554 | =head2 Plack functionality |
040835f0 |
555 | |
3f22de0b |
556 | See L<Catalyst::PSGI>. |
0aafa77a |
557 | |
dacd8b0e |
558 | =head2 Tests in 5.9 |
4db14a9a |
559 | |
e6006848 |
560 | Tests should generally work the same in Catalyst 5.9, but there are |
561 | some differences. |
4db14a9a |
562 | |
e6006848 |
563 | Previously, if using L<Catalyst::Test> and doing local requests (against |
564 | a local server), if the application threw an exception then this |
565 | exception propagated into the test. |
4db14a9a |
566 | |
e6006848 |
567 | This behavior has been removed, and now a 500 response will be returned |
568 | to the test. This change standardizes behavior, so that local test |
569 | requests behave similarly to remote requests. |
4db14a9a |
570 | |
7e2ec16e |
571 | =head1 Upgrading to Catalyst 5.80 |
572 | |
5687c7f9 |
573 | Most applications and plugins should run unaltered on Catalyst 5.80. |
7e2ec16e |
574 | |
8f61d649 |
575 | However, a lot of refactoring work has taken place, and several changes have |
1a98f036 |
576 | been made which could cause incompatibilities. If your application or plugin |
8f61d649 |
577 | is using deprecated code, or relying on side effects, then you could have |
ba03ccca |
578 | issues upgrading to this release. |
5687c7f9 |
579 | |
cf8eab35 |
580 | Most issues found with existing components have been easy to |
8f61d649 |
581 | solve. This document provides a complete description of behavior changes |
582 | which may cause compatibility issues, and of new Catalyst warnings which |
773b3b08 |
583 | might be unclear. |
7e2ec16e |
584 | |
8f61d649 |
585 | If you think you have found an upgrade-related issue which is not covered in |
586 | this document, please email the Catalyst list to discuss the problem. |
7e2ec16e |
587 | |
85f0a66f |
588 | =head1 Moose features |
589 | |
8f61d649 |
590 | =head2 Application class roles |
85f0a66f |
591 | |
8f61d649 |
592 | You can only apply method modifiers after the application's C<< ->setup >> |
85f0a66f |
593 | method has been called. This means that modifiers will not work with methods |
773b3b08 |
594 | run during the call to C<< ->setup >>. |
85f0a66f |
595 | |
a6eb852a |
596 | See L<Catalyst::Manual::ExtendingCatalyst> for more information about using |
597 | L<Moose> in your applications. |
598 | |
85f0a66f |
599 | =head2 Controller actions in Moose roles |
600 | |
d76c88f3 |
601 | You can use L<MooseX::MethodAttributes::Role> if you want to declare actions |
602 | inside Moose roles. |
85f0a66f |
603 | |
d935773d |
604 | =head2 Using Moose in Components |
605 | |
606 | The correct way to use Moose in a component in a both forward and backwards |
607 | compatible way is: |
608 | |
609 | package TestApp::Controller::Root; |
610 | use Moose; |
611 | BEGIN { extends 'Catalyst::Component' }; # Or ::Controller, or whatever |
612 | |
613 | See L<Components which inherit from Moose::Object before Catalyst::Component>. |
614 | |
8f61d649 |
615 | =head1 Known backwards compatibility breakages |
7e2ec16e |
616 | |
8f61d649 |
617 | =head2 Applications in a single file |
85f0a66f |
618 | |
619 | Applications must be in their own file, and loaded at compile time. This |
8f61d649 |
620 | issue generally only affects the tests of CPAN distributions. Your |
621 | application will fail if you try to define an application inline in a |
622 | block, and use plugins which supply a C< new > method, then use that |
623 | application latter in tests within the same file. |
85f0a66f |
624 | |
625 | This is due to the fact that Catalyst is inlining a new method on your |
8f61d649 |
626 | application class allowing it to be compatible with Moose. The method |
627 | used to do this changed in 5.80004 to avoid the possibility of reporting |
628 | an 'Unknown Error' if your application failed to compile. |
85f0a66f |
629 | |
38f90e49 |
630 | =head2 Issues with Class::C3 |
631 | |
8f61d649 |
632 | Catalyst 5.80 uses the L<Algorithm::C3> method dispatch order. This is |
633 | built into Perl 5.10, and comes via L<Class::C3> for Perl 5.8. This |
634 | replaces L<NEXT> with L<Class::C3::Adopt::NEXT>, forcing all components |
635 | to resolve methods using C3, rather than the unpredictable dispatch |
636 | order of L<NEXT>. |
38f90e49 |
637 | |
cf8eab35 |
638 | This issue manifests itself by your application failing to start due to an |
5d06547d |
639 | error message about having a non-linear @ISA. |
640 | |
8f61d649 |
641 | The Catalyst plugin most often causing this is |
642 | L<Catalyst::Plugin::Session::Store::FastMmap> - if you are using this |
643 | plugin and see issues, then please upgrade your plugins, as it has been |
644 | fixed. Note that Makefile.PL in the distribution will warn about known |
645 | incompatible components. |
5d06547d |
646 | |
647 | This issue can, however, be found in your own application - the only solution is |
648 | to go through each base class of the class the error was reported against, until |
649 | you identify the ones in conflict, and resolve them. |
650 | |
651 | To be able to generate a linear @ISA, the list of superclasses for each |
652 | class must be resolvable using the C3 algorithm. Unfortunately, when |
653 | superclasses are being used as mixins (to add functionality used in your class), |
ae7da8f5 |
654 | and with multiple inheritance, it is easy to get this wrong. |
38f90e49 |
655 | |
656 | Most common is the case of: |
657 | |
658 | package Component1; # Note, this is the common case |
659 | use base qw/Class::Accessor::Fast Class::Data::Inheritable/; |
660 | |
8f61d649 |
661 | package Component2; # Accidentally saying it this way causes a failure |
38f90e49 |
662 | use base qw/Class::Data::Inheritable Class::Accessor::Fast/; |
663 | |
664 | package GoesBang; |
665 | use base qw/Component1 Component2/; |
666 | |
5d06547d |
667 | Any situation like this will cause your application to fail to start. |
38f90e49 |
668 | |
8f61d649 |
669 | For additional documentation about this issue, and how to resolve it, see |
5d06547d |
670 | L<Class::C3::Adopt::NEXT>. |
38f90e49 |
671 | |
6f04e56a |
672 | =head2 Components which inherit from Moose::Object before Catalyst::Component |
7e2ec16e |
673 | |
6f04e56a |
674 | Moose components which say: |
7e2ec16e |
675 | |
6f04e56a |
676 | package TestApp::Controller::Example; |
677 | use Moose; |
845bfcd2 |
678 | extends qw/Moose::Object Catalyst::Component/; |
7e2ec16e |
679 | |
8f61d649 |
680 | to use the constructor provided by Moose, while working (if you do some hacks |
1a98f036 |
681 | with the C< BUILDARGS > method), will not work with Catalyst 5.80 as |
6f04e56a |
682 | C<Catalyst::Component> inherits from C<Moose::Object>, and so C< @ISA > fails |
25f61108 |
683 | to linearize. |
6f04e56a |
684 | |
6f04e56a |
685 | The correct way to use Moose in a component in a both forward and backwards |
686 | compatible way is: |
687 | |
688 | package TestApp::Controller::Root; |
689 | use Moose; |
690 | BEGIN { extends 'Catalyst::Component' }; # Or ::Controller, or whatever |
691 | |
ba03ccca |
692 | Note that the C< extends > declaration needs to occur in a begin block for |
3df46b1b |
693 | L<attributes> to operate correctly. |
694 | |
d935773d |
695 | This way you do not inherit directly from C<Moose::Object> |
696 | yourself. Having components which do not inherit their constructor from |
697 | C<Catalyst::Component> is B<unsupported>, and has never been recommended, |
698 | therefore you're on your own if you're using this technique. You'll need |
699 | to detect the version of Catalyst your application is running, and deal |
700 | with it appropriately. |
701 | |
eaae9a92 |
702 | You also don't get the L<Moose::Object> constructor, and therefore attribute |
703 | initialization will not work as normally expected. If you want to use Moose |
3df46b1b |
704 | attributes, then they need to be made lazy to correctly initialize. |
705 | |
706 | Note that this only applies if your component needs to maintain component |
707 | backwards compatibility for Catalyst versions before 5.71001 - in 5.71001 |
708 | attributes work as expected, and the BUILD method is called normally |
eaae9a92 |
709 | (although BUILDARGS is not). |
3df46b1b |
710 | |
711 | If you depend on Catalyst 5.8, then B<all> Moose features work as expected. |
8566c0de |
712 | |
d935773d |
713 | You will also see this issue if you do the following: |
714 | |
715 | package TestApp::Controller::Example; |
716 | use Moose; |
717 | use base 'Catalyst::Controller'; |
718 | |
719 | as C< use base > appends to @ISA. |
720 | |
e11cac87 |
721 | =head3 use Moose in MyApp |
722 | |
723 | Similar to the above, this will also fail: |
724 | |
725 | package MyApp; |
726 | use Moose; |
727 | use Catalyst qw/ |
728 | ConfigLoader |
729 | /; |
730 | __PACKAGE__->setup; |
731 | |
732 | If you need to use Moose in your application class (e.g. for method modifiers |
8f61d649 |
733 | etc.) then the correct technique is: |
e11cac87 |
734 | |
735 | package MyApp; |
736 | use Moose; |
5b6f82d2 |
737 | use Catalyst; |
738 | |
e11cac87 |
739 | extends 'Catalyst'; |
5b6f82d2 |
740 | |
741 | __PACKAGE__->config( name => 'MyApp' ); |
e11cac87 |
742 | __PACKAGE__->setup(qw/ |
743 | ConfigLoader |
744 | /); |
745 | |
04a48104 |
746 | =head2 Anonymous closures installed directly into the symbol table |
747 | |
748 | If you have any code which installs anonymous subroutine references directly |
749 | into the symbol table, you may encounter breakages. The simplest solution is |
750 | to use L<Sub::Name> to name the subroutine. Example: |
751 | |
e11cac87 |
752 | # Original code, likely to break: |
1a98f036 |
753 | my $full_method_name = join('::', $package_name, $method_name); |
04a48104 |
754 | *$full_method_name = sub { ... }; |
755 | |
e11cac87 |
756 | # Fixed Code |
04a48104 |
757 | use Sub::Name 'subname'; |
758 | my $full_method_name = join('::',$package_name, $method_name); |
759 | *$full_method_name = subname $full_method_name, sub { ... }; |
760 | |
8f61d649 |
761 | Additionally, you can take advantage of Catalyst's use of L<Class::MOP> and |
762 | install the closure using the appropriate metaclass. Example: |
04a48104 |
763 | |
764 | use Class::MOP; |
765 | my $metaclass = Moose::Meta::Class->initialize($package_name); |
766 | $metaclass->add_method($method_name => sub { ... }); |
767 | |
780654ad |
768 | =head2 Hooking into application setup |
769 | |
8f61d649 |
770 | To execute code during application start-up, the following snippet in MyApp.pm |
780654ad |
771 | used to work: |
772 | |
773 | sub setup { |
774 | my ($class, @args) = @_; |
775 | $class->NEXT::setup(@args); |
776 | ... # things to do after the actual setup |
777 | } |
778 | |
8f61d649 |
779 | With Catalyst 5.80 this won't work anymore, because Catalyst no longer |
780 | uses NEXT.pm for method resolution. The functionality was only ever |
781 | originally operational as L<NEXT> remembers what methods have already |
782 | been called, and will not call them again. |
780654ad |
783 | |
1a98f036 |
784 | Using this now causes infinite recursion between MyApp::setup and |
785 | Catalyst::setup, due to other backwards compatibility issues related to how |
e6c5b548 |
786 | plugin setup works. Moose method modifiers like C<< before|after|around setup |
1a98f036 |
787 | => sub { ... }; >> also will not operate correctly on the setup method. |
780654ad |
788 | |
789 | The right way to do it is this: |
790 | |
791 | after setup_finalize => sub { |
792 | ... # things to do after the actual setup |
793 | }; |
794 | |
ade00972 |
795 | The setup_finalize hook was introduced as a way to avoid this issue. |
1a98f036 |
796 | |
e11cac87 |
797 | =head2 Components with a new method which returns false |
7e2ec16e |
798 | |
8dd2f514 |
799 | Previously, if you had a component which inherited from Catalyst::COMPONENT, |
8f61d649 |
800 | but overrode the new method to return false, then your class's configuration |
8dd2f514 |
801 | would be blessed into a hash on your behalf, and this would be returned from |
a87f5aa5 |
802 | the COMPONENT method. |
7e2ec16e |
803 | |
8f61d649 |
804 | This behavior makes no sense, and so has been removed. Implementing your own |
805 | C< new > method in components is B<highly> discouraged. Instead, you should |
806 | inherit the new method from Catalyst::Component, and use Moose's BUILD |
1a98f036 |
807 | functionality and/or Moose attributes to perform any construction work |
808 | necessary for your class. |
7e2ec16e |
809 | |
810 | =head2 __PACKAGE__->mk_accessor('meta'); |
811 | |
e11cac87 |
812 | Won't work due to a limitation of L<Moose>. This is currently being fixed |
813 | inside Moose. |
7e2ec16e |
814 | |
815 | =head2 Class::Data::Inheritable side effects |
816 | |
8dd2f514 |
817 | Previously, writing to a class data accessor would copy the accessor method |
818 | down into your package. |
819 | |
8f61d649 |
820 | This behavior has been removed. While the class data is still stored |
8dd2f514 |
821 | per-class, it is stored on the metaclass of the class defining the accessor. |
7e2ec16e |
822 | |
8f61d649 |
823 | Therefore anything relying on the side effect of the accessor being copied down |
8dd2f514 |
824 | will be broken. |
7e2ec16e |
825 | |
1a98f036 |
826 | The following test demonstrates the problem: |
8dd2f514 |
827 | |
828 | { |
829 | package BaseClass; |
830 | use base qw/Class::Data::Inheritable/; |
831 | __PACKAGE__->mk_classdata('foo'); |
832 | } |
833 | |
834 | { |
835 | package Child; |
836 | use base qw/BaseClass/; |
837 | } |
838 | |
839 | BaseClass->foo('base class'); |
840 | Child->foo('sub class'); |
eaae9a92 |
841 | |
e11cac87 |
842 | use Test::More; |
8dd2f514 |
843 | isnt(BaseClass->can('foo'), Child->can('foo')); |
7e2ec16e |
844 | |
f4dda4a8 |
845 | =head2 Extending Catalyst::Request or other classes in an ad hoc manner using mk_accessors |
7e2ec16e |
846 | |
8dd2f514 |
847 | Previously, it was possible to add additional accessors to Catalyst::Request |
848 | (or other classes) by calling the mk_accessors class method. |
7e2ec16e |
849 | |
8f61d649 |
850 | This is no longer supported - users should make a subclass of the class whose |
851 | behavior they would like to change, rather than globally polluting the |
e11cac87 |
852 | Catalyst objects. |
8be895a7 |
853 | |
10011c19 |
854 | =head2 Confused multiple inheritance with Catalyst::Component::COMPONENT |
8be895a7 |
855 | |
8f61d649 |
856 | Previously, Catalyst's COMPONENT method would delegate to the method on |
857 | the right hand side, which could then delegate back again with |
858 | NEXT. This is poor practice, and in addition, makes no sense with C3 |
859 | method dispatch order, and is therefore no longer supported. |
bcc773b9 |
860 | |
ba03ccca |
861 | If a COMPONENT method is detected in the inheritance hierarchy to the right |
bcc773b9 |
862 | hand side of Catalyst::Component::COMPONENT, then the following warning |
863 | message will be emitted: |
7e2ec16e |
864 | |
8dd2f514 |
865 | There is a COMPONENT method resolving after Catalyst::Component |
5687c7f9 |
866 | in ${next_package}. |
8dd2f514 |
867 | |
8f61d649 |
868 | The correct fix is to re-arrange your class's inheritance hierarchy so that the |
bcc773b9 |
869 | COMPONENT method you would like to inherit is the first (left-hand most) |
870 | COMPONENT method in your @ISA. |
7e2ec16e |
871 | |
7e9340de |
872 | =head2 Development server relying on environment variables |
873 | |
874 | Previously, the development server would allow propagation of system |
875 | environment variables into the request environment, this has changed with the |
876 | adoption of Plack. You can use L<Plack::Middleware::ForceEnv> to achieve the |
877 | same effect. |
878 | |
c571d2c8 |
879 | =head1 WARNINGS |
880 | |
63b546b1 |
881 | =head2 Actions in your application class |
882 | |
883 | Having actions in your application class will now emit a warning at application |
e256d0e1 |
884 | startup as this is deprecated. It is highly recommended that these actions are moved |
63b546b1 |
885 | into a MyApp::Controller::Root (as demonstrated by the scaffold application |
5fa5b709 |
886 | generated by catalyst.pl). |
da73c6af |
887 | |
e256d0e1 |
888 | This warning, also affects tests. You should move actions in your test, |
889 | creating a myTest::Controller::Root, like the following example: |
da73c6af |
890 | |
891 | package MyTest::Controller::Root; |
95a52a01 |
892 | |
da73c6af |
893 | use strict; |
894 | use warnings; |
95a52a01 |
895 | |
da73c6af |
896 | use parent 'Catalyst::Controller'; |
95a52a01 |
897 | |
da73c6af |
898 | __PACKAGE__->config(namespace => ''); |
95a52a01 |
899 | |
da73c6af |
900 | sub action : Local { |
901 | my ( $self, $c ) = @_; |
5fa5b709 |
902 | $c->do_something; |
da73c6af |
903 | } |
95a52a01 |
904 | |
da73c6af |
905 | 1; |
63b546b1 |
906 | |
ac9279b0 |
907 | =head2 ::[MVC]:: naming scheme |
908 | |
909 | Having packages called MyApp::[MVC]::XX is deprecated and can no longer be generated |
910 | by catalyst.pl |
911 | |
912 | This is still supported, but it is recommended that you rename your application |
913 | components to Model/View/Controller. |
914 | |
915 | A warning will be issued at application startup if the ::[MVC]:: naming scheme is |
916 | in use. |
917 | |
ade00972 |
918 | =head2 Catalyst::Base |
919 | |
8f61d649 |
920 | Any code using L<Catalyst::Base> will now emit a warning; this |
921 | module will be removed in a future release. |
ade00972 |
922 | |
c571d2c8 |
923 | =head2 Methods in Catalyst::Dispatcher |
924 | |
8f61d649 |
925 | The following methods in Catalyst::Dispatcher are implementation |
926 | details, which may change in the 5.8X release series, and therefore their use |
bcc773b9 |
927 | is highly deprecated. |
c571d2c8 |
928 | |
929 | =over |
930 | |
8dd2f514 |
931 | =item tree |
c571d2c8 |
932 | |
8dd2f514 |
933 | =item dispatch_types |
c571d2c8 |
934 | |
8dd2f514 |
935 | =item registered_dispatch_types |
c571d2c8 |
936 | |
8dd2f514 |
937 | =item method_action_class |
c571d2c8 |
938 | |
8dd2f514 |
939 | =item action_hash |
c571d2c8 |
940 | |
941 | =item container_hash |
942 | |
943 | =back |
944 | |
945 | The first time one of these methods is called, a warning will be emitted: |
7e2ec16e |
946 | |
bcc773b9 |
947 | Class $class is calling the deprecated method Catalyst::Dispatcher::$public_method_name, |
dacd8b0e |
948 | this will be removed in Catalyst 5.9 |
7e2ec16e |
949 | |
c571d2c8 |
950 | You should B<NEVER> be calling any of these methods from application code. |
951 | |
8f61d649 |
952 | Plugin authors and maintainers whose plugins currently call these methods |
8f5a2bd9 |
953 | should change to using the public API, or, if you do not feel the public API |
8f61d649 |
954 | adequately supports your use case, please email the development list to |
8f5a2bd9 |
955 | discuss what API features you need so that you can be appropriately supported. |
7e2ec16e |
956 | |
95b20422 |
957 | =head2 Class files with names that don't correspond to the packages they define |
7e2ec16e |
958 | |
e11cac87 |
959 | In this version of Catalyst, if a component is loaded from disk, but no |
ba03ccca |
960 | symbols are defined in that component's name space after it is loaded, this |
bcc773b9 |
961 | warning will be issued: |
7e2ec16e |
962 | |
bcc773b9 |
963 | require $class was successful but the package is not defined. |
7e2ec16e |
964 | |
8f61d649 |
965 | This is to protect against confusing bugs caused by mistyping package names, |
bcc773b9 |
966 | and will become a fatal error in a future version. |
967 | |
968 | Please note that 'inner packages' (via L<Devel::InnerPackage>) are still fully |
8f61d649 |
969 | supported; this warning is only issued when component file naming does not map |
bcc773b9 |
970 | to B<any> of the packages defined within that component. |
7e2ec16e |
971 | |
5687c7f9 |
972 | =head2 $c->plugin method |
973 | |
25f61108 |
974 | Calling the plugin method is deprecated, and calling it at run time is B<highly |
8dd2f514 |
975 | deprecated>. |
7e2ec16e |
976 | |
95a52a01 |
977 | Instead you are recommended to use L<Catalyst::Model::Adaptor> or similar to |
ba03ccca |
978 | compose the functionality you need outside of the main application name space. |
7e2ec16e |
979 | |
4e68badc |
980 | Calling the plugin method will not be supported past Catalyst 5.81. |
bcc773b9 |
981 | |
7e2ec16e |
982 | =cut |
4e68badc |
983 | |