1 package Catalyst::Model::DBIC::Schema;
5 extends 'Catalyst::Model';
6 with 'CatalystX::Component::Traits';
9 $VERSION = eval $VERSION;
11 use namespace::autoclean;
12 use Carp::Clan '^Catalyst::Model::DBIC::Schema';
16 use Catalyst::Model::DBIC::Schema::Types
17 qw/ConnectInfo LoadedClass SchemaClass/;
19 use MooseX::Types::Moose qw/ArrayRef Str ClassName Undef/;
23 Catalyst::Model::DBIC::Schema - DBIx::Class::Schema Model Class
31 First, prepare your database schema using C<DBIx::Class>.
33 (If you're not sure how to do this, then read L<DBIx::Class::Manual::Intro> first!)
35 This example assumes that you already have a schema called
36 C<MyApp::Schema::FilmDB>,
37 which defines some tables in C<MyApp::Schema::FilmDB::Actor> and
38 C<MyApp::Schema::FilmDB::Film>.
42 Now, to expose it to Catalyst as a model, you should create a DBIC Model in
43 MyApp/Model/FilmDB.pm:
49 =item * With the helper script
51 script/create.pl model FilmDB DBIC::Schema MyApp::Schema::FilmDB ...options...
53 See L<Catalyst::Helper::Model::DBIC::Schema> for details.
57 package MyApp::Model::FilmDB;
58 use base qw/Catalyst::Model::DBIC::Schema/;
61 schema_class => 'MyApp::Schema::FilmDB',
65 password => "password",
69 See below for a full list of the possible config parameters.
75 Now you have a working Model which accesses your separate DBIC Schema. This can
76 be used/accessed in the normal Catalyst manner, via C<< $c->model() >>:
78 my $db_model = $c->model('FilmDB'); # a Catalyst::Model
79 my $dbic = $c->model('FilmDB')->schema; # the actual DBIC object
81 The Model proxies the DBIC instance so you can do
83 my $rs = $db_model->resultset('Actor'); # ... or ...
84 my $rs = $dbic ->resultset('Actor'); # same!
86 There is also a shortcut, which returns a L<DBIx::Class::ResultSet> directly,
87 instead of a L<Catalyst::Model>:
89 my $rs = $c->model('FilmDB::Actor');
91 To find out more about which methods can be called on a ResultSet, or how to
92 add your own methods to it, please see the ResultSet documentation in the
93 L<DBIx::Class> distribution.
97 # to access schema methods directly:
98 $c->model('FilmDB')->schema->source(...);
100 # to access the source object, resultset, and class:
101 $c->model('FilmDB')->source(...);
102 $c->model('FilmDB')->resultset(...);
103 $c->model('FilmDB')->class(...);
105 # For resultsets, there's an even quicker shortcut:
106 $c->model('FilmDB::Actor')
107 # is the same as $c->model('FilmDB')->resultset('Actor')
109 # To get the composed schema for making new connections:
110 my $newconn = $c->model('FilmDB')->composed_schema->connect(...);
112 # Or the same thing via a convenience shortcut:
113 my $newconn = $c->model('FilmDB')->connect(...);
115 # or, if your schema works on different storage drivers:
116 my $newconn = $c->model('FilmDB')->composed_schema->clone();
117 $newconn->storage_type('::LDAP');
118 $newconn->connection(...);
120 # and again, a convenience shortcut
121 my $newconn = $c->model('FilmDB')->clone();
122 $newconn->storage_type('::LDAP');
123 $newconn->connection(...);
125 To set up authentication, see L</"Setting up DBIC authentication"> below.
129 This is a Catalyst Model for L<DBIx::Class::Schema>-based Models. See
130 the documentation for L<Catalyst::Helper::Model::DBIC::Schema> for
131 information on generating these Models via Helper scripts.
133 When your Catalyst app starts up, a thin Model layer is created as an
134 interface to your DBIC Schema. It should be clearly noted that the model
135 object returned by C<< $c->model('FilmDB') >> is NOT itself a DBIC schema or
136 resultset object, but merely a wrapper proving L<methods|/METHODS> to access
137 the underlying schema.
139 In addition to this model class, a shortcut class is generated for each
140 source in the schema, allowing easy and direct access to a resultset of the
141 corresponding type. These generated classes are even thinner than the model
142 class, providing no public methods but simply hooking into Catalyst's
143 model() accessor via the
144 L<ACCEPT_CONTEXT|Catalyst::Component/ACCEPT_CONTEXT> mechanism. The complete
145 contents of each generated class is roughly equivalent to the following:
147 package MyApp::Model::FilmDB::Actor
150 $c->model('FilmDB')->resultset('Actor');
153 In short, there are three techniques available for obtaining a DBIC
157 my $rs = $c->model('FilmDB')->schema->resultset('Actor');
159 # using the shortcut method on the model object
160 my $rs = $c->model('FilmDB')->resultset('Actor');
162 # using the generated class directly
163 my $rs = $c->model('FilmDB::Actor');
165 In order to add methods to a DBIC resultset, you cannot simply add them to
166 the source (row, table) definition class; you must define a separate custom
167 resultset class. See L<DBIx::Class::Manual::Cookbook/"Predefined searches">
170 =head1 CONFIG PARAMETERS
172 Any options in your config not listed here are passed to your schema.
176 This is the classname of your L<DBIx::Class::Schema> Schema. It needs
177 to be findable in C<@INC>, but it does not need to be inside the
178 C<Catalyst::Model::> namespace. This parameter is required.
182 This is an arrayref of connection parameters, which are specific to your
183 C<storage_type> (see your storage type documentation for more details).
184 If you only need one parameter (e.g. the DSN), you can just pass a string
185 instead of an arrayref.
187 This is not required if C<schema_class> already has connection information
188 defined inside itself (which isn't highly recommended, but can be done)
190 For L<DBIx::Class::Storage::DBI>, which is the only supported
191 C<storage_type> in L<DBIx::Class> at the time of this writing, the
192 parameters are your dsn, username, password, and connect options hashref.
194 See L<DBIx::Class::Storage::DBI/connect_info> for a detailed explanation
195 of the arguments supported.
200 dsn => 'dbi:Pg:dbname=mypgdb',
206 dsn => 'dbi:SQLite:dbname=foo.db',
208 'PRAGMA synchronous = OFF',
213 dsn => 'dbi:Pg:dbname=mypgdb',
218 'some SQL statement',
219 'another SQL statement',
223 Or using L<Config::General>:
226 schema_class MyApp::Schema::FilmDB
229 dsn dbi:Pg:dbname=mypgdb
234 on_connect_do some SQL statement
235 on_connect_do another SQL statement
237 user_defined_schema_accessor foo
243 schema_class MyApp::Schema::FilmDB
244 connect_info dbi:SQLite:dbname=foo.db
258 on_connect_call: 'datetime_setup'
261 The old arrayref style with hashrefs for L<DBI> then L<DBIx::Class> options is also
265 'dbi:Pg:dbname=mypgdb',
274 'some SQL statement',
275 'another SQL statement',
282 Array of Traits to apply to the instance. Traits are L<Moose::Role>s.
284 They are relative to the C<< MyApp::TraitFor::Model::DBIC::Schema:: >>, then the C<<
285 Catalyst::TraitFor::Model::DBIC::Schema:: >> namespaces, unless prefixed with C<+>
286 in which case they are taken to be a fully qualified name. E.g.:
289 traits +MyApp::TraitFor::Model::Foo
291 A new instance is created at application time, so any consumed required
292 attributes, coercions and modifiers will work.
294 Traits are applied at L<Catalyst::Component/COMPONENT> time using
295 L<CatalystX::Component::Traits>.
297 C<ref $self> will be an anon class if any traits are applied, C<<
298 $self->_original_class_name >> will be the original class.
300 When writing a Trait, interesting points to modify are C<BUILD>, L</setup> and
303 Traits that come with the distribution:
307 =item L<Catalyst::TraitFor::Model::DBIC::Schema::Caching>
309 =item L<Catalyst::TraitFor::Model::DBIC::Schema::Replicated>
315 Allows the use of a different C<storage_type> than what is set in your
316 C<schema_class> (which in turn defaults to C<::DBI> if not set in current
317 L<DBIx::Class>). Completely optional, and probably unnecessary for most
318 people until other storage backends become available for L<DBIx::Class>.
322 The keys you pass in the model configuration are available as attributes.
324 Other attributes available:
328 Your connect_info args normalized to hashref form (with dsn/user/password.) See
329 L<DBIx::Class::Storage::DBI/connect_info> for more info on the hashref form of
334 The model name L<Catalyst> uses to resolve this model, the part after
335 C<::Model::> or C<::M::> in your class name. E.g. if your class name is
336 C<MyApp::Model::DB> the L</model_name> will be C<DB>.
338 =head2 _default_cursor_class
340 What to reset your L<DBIx::Class::Storage::DBI/cursor_class> to if a custom one
341 doesn't work out. Defaults to L<DBIx::Class::Storage::DBI::Cursor>.
343 =head1 ATTRIBUTES FROM L<MooseX::Traits::Pluggable>
345 =head2 _original_class_name
347 The class name of your model before any L</traits> are applied. E.g.
352 Unresolved arrayref of traits passed in the config.
354 =head2 _resolved_traits
356 Traits you used resolved to full class names.
360 Methods not listed here are delegated to the connected schema used by the model
361 instance, so the following are equivalent:
363 $c->model('DB')->schema->my_accessor('foo');
365 $c->model('DB')->my_accessor('foo');
367 Methods on the model take precedence over schema methods.
371 Instantiates the Model based on the above-documented ->config parameters.
372 The only required parameter is C<schema_class>. C<connect_info> is
373 required in the case that C<schema_class> does not already have connection
374 information defined for it.
378 Accessor which returns the connected schema being used by the this model.
379 There are direct shortcuts on the model class itself for
380 schema->resultset, schema->source, and schema->class.
382 =head2 composed_schema
384 Accessor which returns the composed schema, which has no connection info,
385 which was used in constructing the C<schema> above. Useful for creating
386 new connections based on the same schema/model. There are direct shortcuts
387 from the model object for composed_schema->clone and composed_schema->connect
391 Shortcut for ->composed_schema->clone
395 Shortcut for ->composed_schema->connect
399 Shortcut for ->schema->source
403 Shortcut for ->schema->class
407 Shortcut for ->schema->resultset
411 Provides an accessor for the connected schema's storage object.
412 Used often for debugging and controlling transactions.
416 has schema_class => (
423 has storage_type => (is => 'rw', isa => Str);
425 has connect_info => (is => 'rw', isa => ConnectInfo, coerce => 1);
434 has _default_cursor_class => (
437 default => 'DBIx::Class::Storage::DBI::Cursor',
442 my ($self, $args) = @_;
443 my $class = $self->_original_class_name;
444 my $schema_class = $self->schema_class;
446 if( !$self->connect_info ) {
447 if($schema_class->storage && $schema_class->storage->connect_info) {
448 $self->connect_info($schema_class->storage->connect_info);
451 die "Either ->config->{connect_info} must be defined for $class"
452 . " or $schema_class must have connect info defined on it."
453 . " Here's what we got:\n"
458 if (exists $self->connect_info->{cursor_class}) {
459 eval { Class::MOP::load_class($self->connect_info->{cursor_class}) }
460 or croak "invalid connect_info: Cannot load your cursor_class"
461 . " ".$self->connect_info->{cursor_class}.": $@";
466 $self->composed_schema($schema_class->compose_namespace($class));
468 $self->meta->make_mutable;
469 $self->meta->add_attribute('schema',
471 isa => 'DBIx::Class::Schema',
472 handles => $self->_delegates
474 $self->meta->make_immutable;
476 $self->schema($self->composed_schema->clone);
478 $self->_pass_options_to_schema($args);
480 $self->schema->storage_type($self->storage_type)
481 if $self->storage_type;
483 $self->schema->connection($self->connect_info);
485 $self->_install_rs_models;
488 sub clone { shift->composed_schema->clone(@_); }
490 sub connect { shift->composed_schema->connect(@_); }
494 Called at C<BUILD> time before configuration, but after L</connect_info> is
495 set. To do something after configuuration use C<< after BUILD => >>.
501 =head2 ACCEPT_CONTEXT
503 Point of extension for doing things at C<< $c->model >> time with context,
504 returns the model instance, see L<Catalyst::Manual::Intro/ACCEPT_CONTEXT> for
509 sub ACCEPT_CONTEXT { shift }
511 sub _install_rs_models {
513 my $class = $self->_original_class_name;
517 my @sources = $self->schema->sources;
520 warn <<'EOF' unless $ENV{CMDS_NO_SOURCES};
521 ******************************* WARNING ***************************************
522 * No sources found (did you forget to define your tables?) *
524 * To turn off this warning, set the CMDS_NO_SOURCES environment variable. *
525 *******************************************************************************
529 foreach my $moniker (@sources) {
530 my $classname = "${class}::$moniker";
531 *{"${classname}::ACCEPT_CONTEXT"} = sub {
533 shift->model($self->model_name)->resultset($moniker);
538 sub _reset_cursor_class {
541 if ($self->storage->can('cursor_class')) {
542 $self->storage->cursor_class($self->_default_cursor_class)
543 if $self->storage->cursor_class ne $self->_default_cursor_class;
550 sub composed_schema {
552 my $class = $self->_original_class_name;
553 my $store = \$COMPOSED_CACHE{$class}{$self->schema_class};
555 $$store = shift if @_;
561 sub _build_model_name {
563 my $class = $self->_original_class_name;
564 (my $model_name = $class) =~ s/^[\w:]+::(?:Model|M):://;
572 my $schema_meta = Class::MOP::Class->initialize($self->schema_class);
573 my @schema_methods = $schema_meta->get_all_method_names;
575 # combine with any already added by other schemas
577 @{ $self->meta->find_attribute_by_name('schema')->handles }
580 # now kill the attribute, otherwise add_attribute in BUILD will not do the right
581 # thing (it clears the handles for some reason.) May be a Moose bug.
582 eval { $self->meta->remove_attribute('schema') };
585 @schema_methods{ @schema_methods, @handles } = ();
586 @schema_methods = keys %schema_methods;
588 my @my_methods = $self->meta->get_all_method_names;
590 @my_methods{@my_methods} = ();
593 for my $method (@schema_methods) {
594 push @delegates, $method unless exists $my_methods{$method};
600 sub _pass_options_to_schema {
601 my ($self, $args) = @_;
603 my @attributes = map {
605 } $self->meta->get_all_attributes;
608 @attributes{@attributes} = ();
610 for my $opt (keys %$args) {
611 if (not exists $attributes{$opt}) {
612 next unless $self->schema->can($opt);
613 $self->schema->$opt($self->{$opt});
618 __PACKAGE__->meta->make_immutable;
624 =item CMDS_NO_SOURCES
626 Set this variable if you will be using schemas with no sources (tables) to
627 disable the warning. The warning is there because this is usually a mistake.
631 =head1 Setting up DBIC authentication
633 You can set this up with
634 L<Catalyst::Authentication::Store::DBIx::Class> in MyApp.pm:
638 use Catalyst qw/... Authentication .../;
642 __PACKAGE__->config->{authentication} =
644 default_realm => 'members',
649 password_field => 'password',
650 password_type => 'hashed'
651 password_hash_type => 'SHA-256'
654 class => 'DBIx::Class',
655 user_model => 'DB::User',
656 role_relation => 'roles',
657 role_field => 'rolename',
665 General Catalyst Stuff:
667 L<Catalyst::Manual>, L<Catalyst::Test>, L<Catalyst::Request>,
668 L<Catalyst::Response>, L<Catalyst::Helper>, L<Catalyst>,
670 Stuff related to DBIC and this Model style:
672 L<DBIx::Class>, L<DBIx::Class::Schema>,
673 L<DBIx::Class::Schema::Loader>, L<Catalyst::Helper::Model::DBIC::Schema>,
674 L<CatalystX::Component::Traits>, L<MooseX::Traits::Pluggable>
678 L<Catalyst::TraitFor::Model::DBIC::Schema::Caching>,
679 L<Catalyst::TraitFor::Model::DBIC::Schema::Replicated>
683 Brandon L Black C<blblack at gmail.com>
687 caelum: Rafael Kitover C<rkitover at cpan.org>
689 Dan Dascalescu C<dandv at cpan.org>
691 Aran Deltac C<bluefeet@cpan.org>
695 This program is free software. You can redistribute it and/or modify it
696 under the same terms as Perl itself.