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