Commit | Line | Data |
77d892df |
1 | # Known Bugs: |
cdb34619 |
2 | |
3 | - Bug ->go or ->visit causes actions which have Args or CaptureArgs called |
4 | twice when called via ->go or ->visit. |
5 | |
6 | Test app: http://github.com/bobtfish/catalyst-app-bug-go_chain/tree/master |
7 | |
77d892df |
8 | # Compatibility warnings to add: |
9 | |
10 | - $self->config should warn as config should only ever be called as a |
5d94e8f9 |
11 | class method (TESTS). |
77d892df |
12 | |
13 | # Proposed functionality / feature additions: |
14 | |
15 | ## Log setup needs to be less lame |
16 | |
17 | So Catalyst::Plugin::Log::* can die |
18 | in a fire. Having $c->log_class would be a good start. kane volunteered |
19 | to do some of this. |
20 | |
21 | Simple example: Catalyst::Plugin::Log::Colorful should just be a |
22 | subclass of Catalyst::Log, no ::Plugin:: needed. |
23 | |
24 | See also: Catalyst::Plugin::Log::Dispatch and |
25 | http://github.com/willert/catalyst-plugin-log4perl-simple/tree |
26 | |
27 | # REFACTORING |
28 | |
29 | ## The horrible hack for plugin setup - replacing it: |
30 | |
31 | * Have a look at the Devel::REPL BEFORE_PLUGIN stuff |
32 | I wonder if what we need is that combined with plugins-as-roles |
33 | |
34 | ## App / ctx split: |
35 | |
36 | NOTE - these are notes that t0m thought up after doing back compat for |
5d94e8f9 |
37 | catalyst_component_class, may be inaccurate, wrong or missing things |
77d892df |
38 | bug mst (at least) to correct before trying more than the first 2 |
39 | steps. Please knock yourself out on the first two however :) |
40 | |
41 | - Eliminate actions in MyApp from the main test suite |
42 | - Uncomment warning in C::C::register_action_methods, add tests it works |
43 | by mocking out the logging.. |
44 | - Remove MyApp @ISA controller (ask metaclass if it has attributes, and if |
45 | so you need back compat :/) |
46 | - Make Catalyst::Context, move the per request stuff in there, handles from |
47 | main app class to delegate |
48 | - Make an instance of the app class which is a global variable |
49 | - Make new instance of the context class, not the app class per-request |
50 | - Remove the components as class data, move to instance data on the app |
51 | class (you probably have to do this for _all_ the class data, good luck!) |
52 | - Make it possible for users to spin up different instances of the app class |
53 | (with different config etc each) |
54 | - Profit! (Things like changing the complete app config per vhost, i.e. |
55 | writing a config loader / app class role which dispatches per vhost to |
56 | differently configured apps is piss easy) |
19c64905 |
57 | |
58 | ## GSOC |
59 | |
a43734f6 |
60 | ### Next large steps, planned: |
61 | |
62 | For all components that have been discovered, in whatever way, we create a service: |
63 | - that's a catalyst component service |
64 | - which is basically just a constructor injection |
65 | - except the constructor name is COMPONENT |
66 | - and we're "implicitly" passing along some constructor args |
67 | - Lifecycle => Singleton |
68 | |
f5dbaa05 |
69 | - Fix B::B so that Service::WithParametrs' parameters attribute has a builder |
70 | - Fix ConstructorInjection so that default parameters are supplied (for accept_context_args) |
71 | - Fix ConstructorInjection's 'suffix' - should be called 'config_key' or something, and |
72 | should be an attribute on the service (as it never changes), rather than a parameter |
73 | to the service |
74 | |
a43734f6 |
75 | - We make a 'components' sub container in the main container. |
76 | - This gets the ConstructorInjection COMPONENT services, as model_Foo. |
77 | - Lifecycle of these services is Singleton |
78 | - This simplifies the code for MyApp->components, as it need only list one sub-container |
79 | |
80 | - We create a second service (depending on the first one) for ACCEPT_CONTEXT |
81 | - This has a custom service which calls ACCEPT_CONTEXT when the instance is fetched |
82 | - Not Singleton lifecycle |
83 | |
84 | Note - ACCEPT_CONTEXT can be called on app values - if you have a Model::Foo, with an ACCEPT_CONTEXT |
85 | and you call MyApp->model('Foo') then ACCEPT_CONTEXT gets invoked with a $c of 'MyApp' (this is not\ |
86 | the normal case, but we need to preserve for compat) |
87 | |
88 | ### Next steps - less planned: |
731a4757 |
89 | |
334eb9fb |
90 | - Creating service()-like sugar for component |
91 | |
731a4757 |
92 | - Test cases for extending the container in an application. |
334eb9fb |
93 | - Using the sugar added in the previous item |
94 | - Test when Model::Foo depends_on Model::Bar |
731a4757 |
95 | |
4634d3ad |
96 | |
f5dbaa05 |
97 | #### Extending my app, notes |
98 | |
99 | Basically try to implement something like this (starting out without the sugar!), and see how it breaks |
100 | and what needs to be done to fix it! |
101 | |
102 | ##### Eventual syntax |
103 | |
104 | package MyApp::Container; |
105 | use Catalyst::IOC; |
106 | |
107 | container $self, as { |
108 | container model => as { |
109 | component Foo => (); # As per default! |
110 | component Bar => (dependencies => ['/model/Foo']); # Magic! |
111 | component Baz => ( lifecycle => 'InstancePerContext ); |
112 | component Quux => ( lifecycle => 'Singleton' ); # ACCEPT_CONTEXT not called |
113 | }; |
114 | # Note - implementation of BB may need to be changed to support making sure existing |
115 | # services actually get overridden. not sure how the default container behaves when doing that |
116 | # above code would build the constructor injection as it currently does, |
117 | # defaulting to the class name in the right namespace as declared by the surrounding container |
118 | # as well as adding using the catalyst-specific service class |
119 | }; |
120 | |
121 | 1; |
122 | |
123 | ##### To start with |
124 | |
125 | package MyApp::Container; |
126 | use Moose; |
127 | |
128 | extends 'Catalyst::Container; |
129 | |
130 | after BUILD => sub { |
131 | my $self = shift; |
132 | my $model_container = $self->get_sub_container('model'); |
133 | my $service = Catalyst::IOC::ConstructorInjection->new( |
134 | name => 'Baz', |
135 | class => 'MyApp::Model::Baz', |
136 | dependencies => [ |
137 | depends_on( '/application_name' ), |
138 | depends_on( '/config' ), |
139 | depends_on( '/model/Foo' ), |
140 | ], |
141 | lifecycle => 'InstancePerContext', |
142 | ); |
143 | $model_container->add_service( 'Foo', $service ); |
144 | }; |
19c64905 |
145 | |
a43734f6 |
146 | ### To polish off / t0m review |
19c64905 |
147 | |
19c64905 |
148 | - + $class->container->get_sub_container('model')->make_single_default; |
a43734f6 |
149 | + $class->container->get_sub_container('view')->make_single_default; |
150 | |
5bc06bcb |
151 | locate_components service vs setup_components method |
152 | - can we be more lazy? |
153 | - should setup_components be a service that things like the ->component lookup |
154 | can depend on? |
155 | |
a43734f6 |
156 | get_components_names_types |
157 | |
a43734f6 |
158 | +# FIXME - t0m, how do you feel about this name? |
159 | +# also, do you think I should draw it here, or just return the data structure? |
160 | +sub get_components_names_types { |
161 | |
c3125988 |
162 | +Same as L<build_model_subcontainer>, but for controllers. The difference is |
163 | +that there is no ACCEPT_CONTEXT for controllers. |
164 | ^^ This is wrong!! |
165 | |
166 | - my $accept_context_args = $self->param('accept_context_args'); |
167 | + my $accept_context_args = $params{accept_context_args}; |
d9661bfa |
168 | ^^ This (may be) wrong! I am thinking the service should be allowed to mangle the |
169 | accept_context args, no? |
c3125988 |
170 | Without this change, the user could make a custom service which mangled the param, and use |
171 | Catalyst/IOC/Service/WithAcceptContext.pm, with this change, that module will always see the |
d9661bfa |
172 | un-mangled version?? However, without this change, shit doesn't work... |
c3125988 |
173 | |
a43734f6 |
174 | accept_context_args - where does this come from? |
c3125988 |
175 | |
176 | + $purity = $purity // 1; |
177 | ^^ perl 5.10 only |
4634d3ad |
178 | |
179 | ### Known issues |
180 | |
181 | - Broken $instance->expand_modules() in setup_components and figure |
182 | out later how to bring it back |
183 | |
184 | - expand_component_module |
185 | |
5bc06bcb |
186 | - People wrapping locate_components in Catalyst.pm |