1 package DBIx::Class::ResultSource;
6 use DBIx::Class::ResultSet;
7 use DBIx::Class::ResultSourceHandle;
8 use Carp::Clan qw/^DBIx::Class/;
11 use base qw/DBIx::Class/;
13 __PACKAGE__->mk_group_accessors('simple' => qw/_ordered_columns
14 _columns _primaries _unique_constraints name resultset_attributes
15 schema from _relationships column_info_from_storage source_info
16 source_name sqlt_deploy_callback/);
18 __PACKAGE__->mk_group_accessors('component_class' => qw/resultset_class
23 DBIx::Class::ResultSource - Result source object
27 # Create a table based result source, in a result class.
29 package MyDB::Schema::Result::Artist;
30 use base qw/DBIx::Class/;
32 __PACKAGE__->load_components(qw/Core/);
33 __PACKAGE__->table('artist');
34 __PACKAGE__->add_columns(qw/ artistid name /);
35 __PACKAGE__->set_primary_key('artistid');
36 __PACKAGE__->has_many(cds => 'MyDB::Schema::Result::CD');
40 # Create a query (view) based result source, in a result class
41 package MyDB::Schema::Result::Year2000CDs;
43 use DBIx::Class::ResultSource::View;
45 __PACKAGE__->load_components('Core');
46 __PACKAGE__->table_class('DBIx::Class::ResultSource::View');
48 __PACKAGE__->table('year2000cds');
49 __PACKAGE__->result_source_instance->is_virtual(1);
50 __PACKAGE__->result_source_instance->view_definition(
51 "SELECT cdid, artist, title FROM cd WHERE year ='2000'"
57 A ResultSource is an object that represents a source of data for querying.
59 This class is a base class for various specialised types of result
60 sources, for example L<DBIx::Class::ResultSource::Table>. Table is the
61 default result source type, so one is created for you when defining a
62 result class as described in the synopsis above.
64 More specifically, the L<DBIx::Class::Core> component pulls in the
65 L<DBIx::Class::ResultSourceProxy::Table> as a base class, which
66 defines the L<table|DBIx::Class::ResultSourceProxy::Table/table>
67 method. When called, C<table> creates and stores an instance of
68 L<DBIx::Class::ResultSoure::Table>. Luckily, to use tables as result
69 sources, you don't need to remember any of this.
71 Result sources representing select queries, or views, can also be
72 created, see L<DBIx::Class::ResultSource::View> for full details.
74 =head2 Finding result source objects
76 As mentioned above, a result source instance is created and stored for
77 you when you define a L<Result Class|DBIx::Class::Manual::Glossary/Result Class>.
79 You can retrieve the result source at runtime in the following ways:
83 =item From a Schema object:
85 $schema->source($source_name);
87 =item From a Row object:
91 =item From a ResultSet object:
104 my ($class, $attrs) = @_;
105 $class = ref $class if ref $class;
107 my $new = bless { %{$attrs || {}} }, $class;
108 $new->{resultset_class} ||= 'DBIx::Class::ResultSet';
109 $new->{resultset_attributes} = { %{$new->{resultset_attributes} || {}} };
110 $new->{_ordered_columns} = [ @{$new->{_ordered_columns}||[]}];
111 $new->{_columns} = { %{$new->{_columns}||{}} };
112 $new->{_relationships} = { %{$new->{_relationships}||{}} };
113 $new->{name} ||= "!!NAME NOT SET!!";
114 $new->{_columns_info_loaded} ||= 0;
115 $new->{sqlt_deploy_callback} ||= "default_sqlt_deploy_hook";
125 =item Arguments: @columns
127 =item Return value: The ResultSource object
131 $source->add_columns(qw/col1 col2 col3/);
133 $source->add_columns('col1' => \%col1_info, 'col2' => \%col2_info, ...);
135 Adds columns to the result source. If supplied colname => hashref
136 pairs, uses the hashref as the L</column_info> for that column. Repeated
137 calls of this method will add more columns, not replace them.
139 The column names given will be created as accessor methods on your
140 L<DBIx::Class::Row> objects. You can change the name of the accessor
141 by supplying an L</accessor> in the column_info hash.
143 The contents of the column_info are not set in stone. The following
144 keys are currently recognised/used by DBIx::Class:
150 { accessor => '_name' }
152 # example use, replace standard accessor with one of your own:
154 my ($self, $value) = @_;
156 die "Name cannot contain digits!" if($value =~ /\d/);
157 $self->_name($value);
159 return $self->_name();
162 Use this to set the name of the accessor method for this column. If unset,
163 the name of the column will be used.
167 { data_type => 'integer' }
169 This contains the column type. It is automatically filled if you use the
170 L<SQL::Translator::Producer::DBIx::Class::File> producer, or the
171 L<DBIx::Class::Schema::Loader> module.
173 Currently there is no standard set of values for the data_type. Use
174 whatever your database supports.
180 The length of your column, if it is a column type that can have a size
181 restriction. This is currently only used to create tables from your
182 schema, see L<DBIx::Class::Schema/deploy>.
188 Set this to a true value for a columns that is allowed to contain NULL
189 values, default is false. This is currently only used to create tables
190 from your schema, see L<DBIx::Class::Schema/deploy>.
192 =item is_auto_increment
194 { is_auto_increment => 1 }
196 Set this to a true value for a column whose value is somehow
197 automatically set, defaults to false. This is used to determine which
198 columns to empty when cloning objects using
199 L<DBIx::Class::Row/copy>. It is also used by
200 L<DBIx::Class::Schema/deploy>.
206 Set this to a true or false value (not C<undef>) to explicitly specify
207 if this column contains numeric data. This controls how set_column
208 decides whether to consider a column dirty after an update: if
209 C<is_numeric> is true a numeric comparison C<< != >> will take place
210 instead of the usual C<eq>
212 If not specified the storage class will attempt to figure this out on
213 first access to the column, based on the column C<data_type>. The
214 result will be cached in this attribute.
218 { is_foreign_key => 1 }
220 Set this to a true value for a column that contains a key from a
221 foreign table, defaults to false. This is currently only used to
222 create tables from your schema, see L<DBIx::Class::Schema/deploy>.
226 { default_value => \'now()' }
228 Set this to the default value which will be inserted into a column by
229 the database. Can contain either a value or a function (use a
230 reference to a scalar e.g. C<\'now()'> if you want a function). This
231 is currently only used to create tables from your schema, see
232 L<DBIx::Class::Schema/deploy>.
234 See the note on L<DBIx::Class::Row/new> for more information about possible
235 issues related to db-side default values.
239 { sequence => 'my_table_seq' }
241 Set this on a primary key column to the name of the sequence used to
242 generate a new key value. If not specified, L<DBIx::Class::PK::Auto>
243 will attempt to retrieve the name of the sequence from the database
248 Set this to a true value for a column whose value is retrieved
249 automatically from an oracle sequence. If you do not use an Oracle
250 trigger to get the nextval, you have to set sequence as well.
254 This is used by L<DBIx::Class::Schema/deploy> and L<SQL::Translator>
255 to add extra non-generic data to the column. For example: C<< extra
256 => { unsigned => 1} >> is used by the MySQL producer to set an integer
257 column to unsigned. For more details, see
258 L<SQL::Translator::Producer::MySQL>.
266 =item Arguments: $colname, \%columninfo?
268 =item Return value: 1/0 (true/false)
272 $source->add_column('col' => \%info);
274 Add a single column and optional column info. Uses the same column
275 info keys as L</add_columns>.
280 my ($self, @cols) = @_;
281 $self->_ordered_columns(\@cols) unless $self->_ordered_columns;
284 my $columns = $self->_columns;
285 while (my $col = shift @cols) {
286 # If next entry is { ... } use that for the column info, if not
287 # use an empty hashref
288 my $column_info = ref $cols[0] ? shift(@cols) : {};
289 push(@added, $col) unless exists $columns->{$col};
290 $columns->{$col} = $column_info;
292 push @{ $self->_ordered_columns }, @added;
296 sub add_column { shift->add_columns(@_); } # DO NOT CHANGE THIS TO GLOB
302 =item Arguments: $colname
304 =item Return value: 1/0 (true/false)
308 if ($source->has_column($colname)) { ... }
310 Returns true if the source has a column of this name, false otherwise.
315 my ($self, $column) = @_;
316 return exists $self->_columns->{$column};
323 =item Arguments: $colname
325 =item Return value: Hashref of info
329 my $info = $source->column_info($col);
331 Returns the column metadata hashref for a column, as originally passed
332 to L</add_columns>. See L</add_columns> above for information on the
333 contents of the hashref.
338 my ($self, $column) = @_;
339 $self->throw_exception("No such column $column")
340 unless exists $self->_columns->{$column};
341 #warn $self->{_columns_info_loaded}, "\n";
342 if ( ! $self->_columns->{$column}{data_type}
343 and $self->column_info_from_storage
344 and ! $self->{_columns_info_loaded}
345 and $self->schema and $self->storage )
347 $self->{_columns_info_loaded}++;
350 # eval for the case of storage without table
351 eval { $info = $self->storage->columns_info_for( $self->from ) };
353 for my $realcol ( keys %{$info} ) {
354 $lc_info->{lc $realcol} = $info->{$realcol};
356 foreach my $col ( keys %{$self->_columns} ) {
357 $self->_columns->{$col} = {
358 %{ $self->_columns->{$col} },
359 %{ $info->{$col} || $lc_info->{lc $col} || {} }
364 return $self->_columns->{$column};
371 =item Arguments: None
373 =item Return value: Ordered list of column names
377 my @column_names = $source->columns;
379 Returns all column names in the order they were declared to L</add_columns>.
385 $self->throw_exception(
386 "columns() is a read-only accessor, did you mean add_columns()?"
388 return @{$self->{_ordered_columns}||[]};
391 =head2 remove_columns
395 =item Arguments: @colnames
397 =item Return value: undefined
401 $source->remove_columns(qw/col1 col2 col3/);
403 Removes the given list of columns by name, from the result source.
405 B<Warning>: Removing a column that is also used in the sources primary
406 key, or in one of the sources unique constraints, B<will> result in a
407 broken result source.
413 =item Arguments: $colname
415 =item Return value: undefined
419 $source->remove_column('col');
421 Remove a single column by name from the result source, similar to
424 B<Warning>: Removing a column that is also used in the sources primary
425 key, or in one of the sources unique constraints, B<will> result in a
426 broken result source.
431 my ($self, @to_remove) = @_;
433 my $columns = $self->_columns
438 delete $columns->{$_};
442 $self->_ordered_columns([ grep { not $to_remove{$_} } @{$self->_ordered_columns} ]);
445 sub remove_column { shift->remove_columns(@_); } # DO NOT CHANGE THIS TO GLOB
447 =head2 set_primary_key
451 =item Arguments: @cols
453 =item Return value: undefined
457 Defines one or more columns as primary key for this source. Must be
458 called after L</add_columns>.
460 Additionally, defines a L<unique constraint|add_unique_constraint>
463 The primary key columns are used by L<DBIx::Class::PK::Auto> to
464 retrieve automatically created values from the database. They are also
465 used as default joining columns when specifying relationships, see
466 L<DBIx::Class::Relationship>.
470 sub set_primary_key {
471 my ($self, @cols) = @_;
472 # check if primary key columns are valid columns
473 foreach my $col (@cols) {
474 $self->throw_exception("No such column $col on table " . $self->name)
475 unless $self->has_column($col);
477 $self->_primaries(\@cols);
479 $self->add_unique_constraint(primary => \@cols);
482 =head2 primary_columns
486 =item Arguments: None
488 =item Return value: Ordered list of primary column names
492 Read-only accessor which returns the list of primary keys, supplied by
497 sub primary_columns {
498 return @{shift->_primaries||[]};
501 =head2 add_unique_constraint
505 =item Arguments: $name?, \@colnames
507 =item Return value: undefined
511 Declare a unique constraint on this source. Call once for each unique
514 # For UNIQUE (column1, column2)
515 __PACKAGE__->add_unique_constraint(
516 constraint_name => [ qw/column1 column2/ ],
519 Alternatively, you can specify only the columns:
521 __PACKAGE__->add_unique_constraint([ qw/column1 column2/ ]);
523 This will result in a unique constraint named
524 C<table_column1_column2>, where C<table> is replaced with the table
527 Unique constraints are used, for example, when you pass the constraint
528 name as the C<key> attribute to L<DBIx::Class::ResultSet/find>. Then
529 only columns in the constraint are searched.
531 Throws an error if any of the given column names do not yet exist on
536 sub add_unique_constraint {
541 $name ||= $self->name_unique_constraint($cols);
543 foreach my $col (@$cols) {
544 $self->throw_exception("No such column $col on table " . $self->name)
545 unless $self->has_column($col);
548 my %unique_constraints = $self->unique_constraints;
549 $unique_constraints{$name} = $cols;
550 $self->_unique_constraints(\%unique_constraints);
553 =head2 name_unique_constraint
557 =item Arguments: @colnames
559 =item Return value: Constraint name
563 $source->table('mytable');
564 $source->name_unique_constraint('col1', 'col2');
568 Return a name for a unique constraint containing the specified
569 columns. The name is created by joining the table name and each column
570 name, using an underscore character.
572 For example, a constraint on a table named C<cd> containing the columns
573 C<artist> and C<title> would result in a constraint name of C<cd_artist_title>.
575 This is used by L</add_unique_constraint> if you do not specify the
576 optional constraint name.
580 sub name_unique_constraint {
581 my ($self, $cols) = @_;
583 return join '_', $self->name, @$cols;
586 =head2 unique_constraints
590 =item Arguments: None
592 =item Return value: Hash of unique constraint data
596 $source->unique_constraints();
598 Read-only accessor which returns a hash of unique constraints on this
601 The hash is keyed by constraint name, and contains an arrayref of
602 column names as values.
606 sub unique_constraints {
607 return %{shift->_unique_constraints||{}};
610 =head2 unique_constraint_names
614 =item Arguments: None
616 =item Return value: Unique constraint names
620 $source->unique_constraint_names();
622 Returns the list of unique constraint names defined on this source.
626 sub unique_constraint_names {
629 my %unique_constraints = $self->unique_constraints;
631 return keys %unique_constraints;
634 =head2 unique_constraint_columns
638 =item Arguments: $constraintname
640 =item Return value: List of constraint columns
644 $source->unique_constraint_columns('myconstraint');
646 Returns the list of columns that make up the specified unique constraint.
650 sub unique_constraint_columns {
651 my ($self, $constraint_name) = @_;
653 my %unique_constraints = $self->unique_constraints;
655 $self->throw_exception(
656 "Unknown unique constraint $constraint_name on '" . $self->name . "'"
657 ) unless exists $unique_constraints{$constraint_name};
659 return @{ $unique_constraints{$constraint_name} };
662 =head2 sqlt_deploy_callback
666 =item Arguments: $callback
670 __PACKAGE__->sqlt_deploy_callback('mycallbackmethod');
672 An accessor to set a callback to be called during deployment of
673 the schema via L<DBIx::Class::Schema/create_ddl_dir> or
674 L<DBIx::Class::Schema/deploy>.
676 The callback can be set as either a code reference or the name of a
677 method in the current result class.
679 If not set, the L</default_sqlt_deploy_hook> is called.
681 Your callback will be passed the $source object representing the
682 ResultSource instance being deployed, and the
683 L<SQL::Translator::Schema::Table> object being created from it. The
684 callback can be used to manipulate the table object or add your own
685 customised indexes. If you need to manipulate a non-table object, use
686 the L<DBIx::Class::Schema/sqlt_deploy_hook>.
688 See L<DBIx::Class::Manual::Cookbook/Adding Indexes And Functions To
689 Your SQL> for examples.
691 This sqlt deployment callback can only be used to manipulate
692 SQL::Translator objects as they get turned into SQL. To execute
693 post-deploy statements which SQL::Translator does not currently
694 handle, override L<DBIx::Class::Schema/deploy> in your Schema class
695 and call L<dbh_do|DBIx::Class::Storage::DBI/dbh_do>.
697 =head2 default_sqlt_deploy_hook
701 =item Arguments: $source, $sqlt_table
703 =item Return value: undefined
707 This is the sensible default for L</sqlt_deploy_callback>.
709 If a method named C<sqlt_deploy_hook> exists in your Result class, it
710 will be called and passed the current C<$source> and the
711 C<$sqlt_table> being deployed.
715 sub default_sqlt_deploy_hook {
718 my $class = $self->result_class;
720 if ($class and $class->can('sqlt_deploy_hook')) {
721 $class->sqlt_deploy_hook(@_);
725 sub _invoke_sqlt_deploy_hook {
727 if ( my $hook = $self->sqlt_deploy_callback) {
736 =item Arguments: None
738 =item Return value: $resultset
742 Returns a resultset for the given source. This will initially be created
745 $self->resultset_class->new($self, $self->resultset_attributes)
747 but is cached from then on unless resultset_class changes.
749 =head2 resultset_class
753 =item Arguments: $classname
755 =item Return value: $classname
759 package My::Schema::ResultSet::Artist;
760 use base 'DBIx::Class::ResultSet';
763 # In the result class
764 __PACKAGE__->resultset_class('My::Schema::ResultSet::Artist');
767 $source->resultset_class('My::Schema::ResultSet::Artist');
769 Set the class of the resultset. This is useful if you want to create your
770 own resultset methods. Create your own class derived from
771 L<DBIx::Class::ResultSet>, and set it here. If called with no arguments,
772 this method returns the name of the existing resultset class, if one
775 =head2 resultset_attributes
779 =item Arguments: \%attrs
781 =item Return value: \%attrs
785 # In the result class
786 __PACKAGE__->resultset_attributes({ order_by => [ 'id' ] });
789 $source->resultset_attributes({ order_by => [ 'id' ] });
791 Store a collection of resultset attributes, that will be set on every
792 L<DBIx::Class::ResultSet> produced from this result source. For a full
793 list see L<DBIx::Class::ResultSet/ATTRIBUTES>.
799 $self->throw_exception(
800 'resultset does not take any arguments. If you want another resultset, '.
801 'call it on the schema instead.'
804 return $self->resultset_class->new(
807 %{$self->{resultset_attributes}},
808 %{$self->schema->default_resultset_attributes}
817 =item Arguments: $source_name
819 =item Result value: $source_name
823 Set an alternate name for the result source when it is loaded into a schema.
824 This is useful if you want to refer to a result source by a name other than
827 package ArchivedBooks;
828 use base qw/DBIx::Class/;
829 __PACKAGE__->table('books_archive');
830 __PACKAGE__->source_name('Books');
832 # from your schema...
833 $schema->resultset('Books')->find(1);
839 =item Arguments: None
841 =item Return value: FROM clause
845 my $from_clause = $source->from();
847 Returns an expression of the source to be supplied to storage to specify
848 retrieval from this source. In the case of a database, the required FROM
855 =item Arguments: None
857 =item Return value: A schema object
861 my $schema = $source->schema();
863 Returns the L<DBIx::Class::Schema> object that this result source
870 =item Arguments: None
872 =item Return value: A Storage object
876 $source->storage->debug(1);
878 Returns the storage handle for the current schema.
880 See also: L<DBIx::Class::Storage>
884 sub storage { shift->schema->storage; }
886 =head2 add_relationship
890 =item Arguments: $relname, $related_source_name, \%cond, [ \%attrs ]
892 =item Return value: 1/true if it succeeded
896 $source->add_relationship('relname', 'related_source', $cond, $attrs);
898 L<DBIx::Class::Relationship> describes a series of methods which
899 create pre-defined useful types of relationships. Look there first
900 before using this method directly.
902 The relationship name can be arbitrary, but must be unique for each
903 relationship attached to this result source. 'related_source' should
904 be the name with which the related result source was registered with
905 the current schema. For example:
907 $schema->source('Book')->add_relationship('reviews', 'Review', {
908 'foreign.book_id' => 'self.id',
911 The condition C<$cond> needs to be an L<SQL::Abstract>-style
912 representation of the join between the tables. For example, if you're
913 creating a relation from Author to Book,
915 { 'foreign.author_id' => 'self.id' }
917 will result in the JOIN clause
919 author me JOIN book foreign ON foreign.author_id = me.id
921 You can specify as many foreign => self mappings as necessary.
923 Valid attributes are as follows:
929 Explicitly specifies the type of join to use in the relationship. Any
930 SQL join type is valid, e.g. C<LEFT> or C<RIGHT>. It will be placed in
931 the SQL command immediately before C<JOIN>.
935 An arrayref containing a list of accessors in the foreign class to proxy in
936 the main class. If, for example, you do the following:
938 CD->might_have(liner_notes => 'LinerNotes', undef, {
939 proxy => [ qw/notes/ ],
942 Then, assuming LinerNotes has an accessor named notes, you can do:
944 my $cd = CD->find(1);
945 # set notes -- LinerNotes object is created if it doesn't exist
946 $cd->notes('Notes go here');
950 Specifies the type of accessor that should be created for the
951 relationship. Valid values are C<single> (for when there is only a single
952 related object), C<multi> (when there can be many), and C<filter> (for
953 when there is a single related object, but you also want the relationship
954 accessor to double as a column accessor). For C<multi> accessors, an
955 add_to_* method is also created, which calls C<create_related> for the
960 Throws an exception if the condition is improperly supplied, or cannot
965 sub add_relationship {
966 my ($self, $rel, $f_source_name, $cond, $attrs) = @_;
967 $self->throw_exception("Can't create relationship without join condition")
971 # Check foreign and self are right in cond
972 if ( (ref $cond ||'') eq 'HASH') {
974 $self->throw_exception("Keys of condition should be of form 'foreign.col', not '$_'")
975 if /\./ && !/^foreign\./;
979 my %rels = %{ $self->_relationships };
980 $rels{$rel} = { class => $f_source_name,
981 source => $f_source_name,
984 $self->_relationships(\%rels);
988 # XXX disabled. doesn't work properly currently. skip in tests.
990 my $f_source = $self->schema->source($f_source_name);
992 $self->ensure_class_loaded($f_source_name);
993 $f_source = $f_source_name->result_source;
994 #my $s_class = ref($self->schema);
995 #$f_source_name =~ m/^${s_class}::(.*)$/;
996 #$self->schema->register_class(($1 || $f_source_name), $f_source_name);
997 #$f_source = $self->schema->source($f_source_name);
999 return unless $f_source; # Can't test rel without f_source
1001 eval { $self->_resolve_join($rel, 'me', {}, []) };
1003 if ($@) { # If the resolve failed, back out and re-throw the error
1004 delete $rels{$rel}; #
1005 $self->_relationships(\%rels);
1006 $self->throw_exception("Error creating relationship $rel: $@");
1011 =head2 relationships
1015 =item Arguments: None
1017 =item Return value: List of relationship names
1021 my @relnames = $source->relationships();
1023 Returns all relationship names for this source.
1028 return keys %{shift->_relationships};
1031 =head2 relationship_info
1035 =item Arguments: $relname
1037 =item Return value: Hashref of relation data,
1041 Returns a hash of relationship information for the specified relationship
1042 name. The keys/values are as specified for L</add_relationship>.
1046 sub relationship_info {
1047 my ($self, $rel) = @_;
1048 return $self->_relationships->{$rel};
1051 =head2 has_relationship
1055 =item Arguments: $rel
1057 =item Return value: 1/0 (true/false)
1061 Returns true if the source has a relationship of this name, false otherwise.
1065 sub has_relationship {
1066 my ($self, $rel) = @_;
1067 return exists $self->_relationships->{$rel};
1070 =head2 reverse_relationship_info
1074 =item Arguments: $relname
1076 =item Return value: Hashref of relationship data
1080 Looks through all the relationships on the source this relationship
1081 points to, looking for one whose condition is the reverse of the
1082 condition on this relationship.
1084 A common use of this is to find the name of the C<belongs_to> relation
1085 opposing a C<has_many> relation. For definition of these look in
1086 L<DBIx::Class::Relationship>.
1088 The returned hashref is keyed by the name of the opposing
1089 relationship, and contains its data in the same manner as
1090 L</relationship_info>.
1094 sub reverse_relationship_info {
1095 my ($self, $rel) = @_;
1096 my $rel_info = $self->relationship_info($rel);
1099 return $ret unless ((ref $rel_info->{cond}) eq 'HASH');
1101 my @cond = keys(%{$rel_info->{cond}});
1102 my @refkeys = map {/^\w+\.(\w+)$/} @cond;
1103 my @keys = map {$rel_info->{cond}->{$_} =~ /^\w+\.(\w+)$/} @cond;
1105 # Get the related result source for this relationship
1106 my $othertable = $self->related_source($rel);
1108 # Get all the relationships for that source that related to this source
1109 # whose foreign column set are our self columns on $rel and whose self
1110 # columns are our foreign columns on $rel.
1111 my @otherrels = $othertable->relationships();
1112 my $otherrelationship;
1113 foreach my $otherrel (@otherrels) {
1114 my $otherrel_info = $othertable->relationship_info($otherrel);
1116 my $back = $othertable->related_source($otherrel);
1117 next unless $back->source_name eq $self->source_name;
1121 if (ref $otherrel_info->{cond} eq 'HASH') {
1122 @othertestconds = ($otherrel_info->{cond});
1124 elsif (ref $otherrel_info->{cond} eq 'ARRAY') {
1125 @othertestconds = @{$otherrel_info->{cond}};
1131 foreach my $othercond (@othertestconds) {
1132 my @other_cond = keys(%$othercond);
1133 my @other_refkeys = map {/^\w+\.(\w+)$/} @other_cond;
1134 my @other_keys = map {$othercond->{$_} =~ /^\w+\.(\w+)$/} @other_cond;
1135 next if (!$self->_compare_relationship_keys(\@refkeys, \@other_keys) ||
1136 !$self->_compare_relationship_keys(\@other_refkeys, \@keys));
1137 $ret->{$otherrel} = $otherrel_info;
1143 sub compare_relationship_keys {
1144 carp 'compare_relationship_keys is a private method, stop calling it';
1146 $self->_compare_relationship_keys (@_);
1149 # Returns true if both sets of keynames are the same, false otherwise.
1150 sub _compare_relationship_keys {
1151 my ($self, $keys1, $keys2) = @_;
1153 # Make sure every keys1 is in keys2
1155 foreach my $key (@$keys1) {
1157 foreach my $prim (@$keys2) {
1158 if ($prim eq $key) {
1166 # Make sure every key2 is in key1
1168 foreach my $prim (@$keys2) {
1170 foreach my $key (@$keys1) {
1171 if ($prim eq $key) {
1184 carp 'resolve_join is a private method, stop calling it';
1186 $self->_resolve_join (@_);
1189 # Returns the {from} structure used to express JOIN conditions
1191 my ($self, $join, $alias, $seen, $jpath, $force_left) = @_;
1193 # we need a supplied one, because we do in-place modifications, no returns
1194 $self->throw_exception ('You must supply a seen hashref as the 3rd argument to _resolve_join')
1195 unless ref $seen eq 'HASH';
1197 $self->throw_exception ('You must supply a joinpath arrayref as the 4th argument to _resolve_join')
1198 unless ref $jpath eq 'ARRAY';
1202 if (ref $join eq 'ARRAY') {
1205 $self->_resolve_join($_, $alias, $seen, $jpath, $force_left);
1207 } elsif (ref $join eq 'HASH') {
1210 my $as = ($seen->{$_} ? join ('_', $_, $seen->{$_} + 1) : $_); # the actual seen value will be incremented below
1211 local $force_left->{force} = $force_left->{force};
1213 $self->_resolve_join($_, $alias, $seen, [@$jpath], $force_left),
1214 $self->related_source($_)->_resolve_join(
1215 $join->{$_}, $as, $seen, [@$jpath, $_], $force_left
1219 } elsif (ref $join) {
1220 $self->throw_exception("No idea how to resolve join reftype ".ref $join);
1223 return() unless defined $join;
1225 my $count = ++$seen->{$join};
1226 my $as = ($count > 1 ? "${join}_${count}" : $join);
1228 my $rel_info = $self->relationship_info($join);
1229 $self->throw_exception("No such relationship ${join}") unless $rel_info;
1234 $type = $rel_info->{attrs}{join_type} || '';
1235 $force_left = 1 if lc($type) eq 'left';
1238 my $rel_src = $self->related_source($join);
1239 return [ { $as => $rel_src->from,
1240 -source_handle => $rel_src->handle,
1241 -join_type => $type,
1242 -join_path => [@$jpath, $join],
1244 -relation_chain_depth => $seen->{-relation_chain_depth} || 0,
1246 $self->_resolve_condition($rel_info->{cond}, $as, $alias) ];
1251 carp 'pk_depends_on is a private method, stop calling it';
1253 $self->_pk_depends_on (@_);
1256 # Determines whether a relation is dependent on an object from this source
1257 # having already been inserted. Takes the name of the relationship and a
1258 # hashref of columns of the related object.
1259 sub _pk_depends_on {
1260 my ($self, $relname, $rel_data) = @_;
1262 my $relinfo = $self->relationship_info($relname);
1264 # don't assume things if the relationship direction is specified
1265 return $relinfo->{attrs}{is_foreign_key_constraint}
1266 if exists ($relinfo->{attrs}{is_foreign_key_constraint});
1268 my $cond = $relinfo->{cond};
1269 return 0 unless ref($cond) eq 'HASH';
1271 # map { foreign.foo => 'self.bar' } to { bar => 'foo' }
1272 my $keyhash = { map { my $x = $_; $x =~ s/.*\.//; $x; } reverse %$cond };
1274 # assume anything that references our PK probably is dependent on us
1275 # rather than vice versa, unless the far side is (a) defined or (b)
1277 my $rel_source = $self->related_source($relname);
1279 foreach my $p ($self->primary_columns) {
1280 if (exists $keyhash->{$p}) {
1281 unless (defined($rel_data->{$keyhash->{$p}})
1282 || $rel_source->column_info($keyhash->{$p})
1283 ->{is_auto_increment}) {
1292 sub resolve_condition {
1293 carp 'resolve_condition is a private method, stop calling it';
1295 $self->_resolve_condition (@_);
1298 # Resolves the passed condition to a concrete query fragment. If given an alias,
1299 # returns a join condition; if given an object, inverts that object to produce
1300 # a related conditional from that object.
1301 our $UNRESOLVABLE_CONDITION = \'1 = 0';
1303 sub _resolve_condition {
1304 my ($self, $cond, $as, $for) = @_;
1305 if (ref $cond eq 'HASH') {
1307 foreach my $k (keys %{$cond}) {
1308 my $v = $cond->{$k};
1309 # XXX should probably check these are valid columns
1310 $k =~ s/^foreign\.// ||
1311 $self->throw_exception("Invalid rel cond key ${k}");
1312 $v =~ s/^self\.// ||
1313 $self->throw_exception("Invalid rel cond val ${v}");
1314 if (ref $for) { # Object
1315 #warn "$self $k $for $v";
1316 unless ($for->has_column_loaded($v)) {
1317 if ($for->in_storage) {
1318 $self->throw_exception(
1319 "Column ${v} not loaded or not passed to new() prior to insert()"
1320 ." on ${for} trying to resolve relationship (maybe you forgot "
1321 ."to call ->discard_changes to get defaults from the db)"
1324 return $UNRESOLVABLE_CONDITION;
1326 $ret{$k} = $for->get_column($v);
1327 #$ret{$k} = $for->get_column($v) if $for->has_column_loaded($v);
1329 } elsif (!defined $for) { # undef, i.e. "no object"
1331 } elsif (ref $as eq 'HASH') { # reverse hashref
1332 $ret{$v} = $as->{$k};
1333 } elsif (ref $as) { # reverse object
1334 $ret{$v} = $as->get_column($k);
1335 } elsif (!defined $as) { # undef, i.e. "no reverse object"
1338 $ret{"${as}.${k}"} = "${for}.${v}";
1342 } elsif (ref $cond eq 'ARRAY') {
1343 return [ map { $self->_resolve_condition($_, $as, $for) } @$cond ];
1345 die("Can't handle condition $cond yet :(");
1349 # Legacy code, needs to go entirely away (fully replaced by _resolve_prefetch)
1350 sub resolve_prefetch {
1351 carp 'resolve_prefetch is a private method, stop calling it';
1353 my ($self, $pre, $alias, $seen, $order, $collapse) = @_;
1355 if( ref $pre eq 'ARRAY' ) {
1357 map { $self->resolve_prefetch( $_, $alias, $seen, $order, $collapse ) }
1360 elsif( ref $pre eq 'HASH' ) {
1363 $self->resolve_prefetch($_, $alias, $seen, $order, $collapse),
1364 $self->related_source($_)->resolve_prefetch(
1365 $pre->{$_}, "${alias}.$_", $seen, $order, $collapse)
1370 $self->throw_exception(
1371 "don't know how to resolve prefetch reftype ".ref($pre));
1374 my $count = ++$seen->{$pre};
1375 my $as = ($count > 1 ? "${pre}_${count}" : $pre);
1376 my $rel_info = $self->relationship_info( $pre );
1377 $self->throw_exception( $self->name . " has no such relationship '$pre'" )
1379 my $as_prefix = ($alias =~ /^.*?\.(.+)$/ ? $1.'.' : '');
1380 my $rel_source = $self->related_source($pre);
1382 if (exists $rel_info->{attrs}{accessor}
1383 && $rel_info->{attrs}{accessor} eq 'multi') {
1384 $self->throw_exception(
1385 "Can't prefetch has_many ${pre} (join cond too complex)")
1386 unless ref($rel_info->{cond}) eq 'HASH';
1387 my $dots = @{[$as_prefix =~ m/\./g]} + 1; # +1 to match the ".${as_prefix}"
1388 if (my ($fail) = grep { @{[$_ =~ m/\./g]} == $dots }
1389 keys %{$collapse}) {
1390 my ($last) = ($fail =~ /([^\.]+)$/);
1392 "Prefetching multiple has_many rels ${last} and ${pre} "
1393 .(length($as_prefix)
1394 ? "at the same level (${as_prefix}) "
1397 . 'will explode the number of row objects retrievable via ->next or ->all. '
1398 . 'Use at your own risk.'
1401 #my @col = map { (/^self\.(.+)$/ ? ("${as_prefix}.$1") : ()); }
1402 # values %{$rel_info->{cond}};
1403 $collapse->{".${as_prefix}${pre}"} = [ $rel_source->primary_columns ];
1404 # action at a distance. prepending the '.' allows simpler code
1405 # in ResultSet->_collapse_result
1406 my @key = map { (/^foreign\.(.+)$/ ? ($1) : ()); }
1407 keys %{$rel_info->{cond}};
1408 my @ord = (ref($rel_info->{attrs}{order_by}) eq 'ARRAY'
1409 ? @{$rel_info->{attrs}{order_by}}
1410 : (defined $rel_info->{attrs}{order_by}
1411 ? ($rel_info->{attrs}{order_by})
1413 push(@$order, map { "${as}.$_" } (@key, @ord));
1416 return map { [ "${as}.$_", "${as_prefix}${pre}.$_", ] }
1417 $rel_source->columns;
1421 # Accepts one or more relationships for the current source and returns an
1422 # array of column names for each of those relationships. Column names are
1423 # prefixed relative to the current source, in accordance with where they appear
1424 # in the supplied relationships. Needs an alias_map generated by
1425 # $rs->_joinpath_aliases
1427 sub _resolve_prefetch {
1428 my ($self, $pre, $alias, $alias_map, $order, $collapse, $pref_path) = @_;
1431 if( ref $pre eq 'ARRAY' ) {
1433 map { $self->_resolve_prefetch( $_, $alias, $alias_map, $order, $collapse, [ @$pref_path ] ) }
1436 elsif( ref $pre eq 'HASH' ) {
1439 $self->_resolve_prefetch($_, $alias, $alias_map, $order, $collapse, [ @$pref_path ] ),
1440 $self->related_source($_)->_resolve_prefetch(
1441 $pre->{$_}, "${alias}.$_", $alias_map, $order, $collapse, [ @$pref_path, $_] )
1446 $self->throw_exception(
1447 "don't know how to resolve prefetch reftype ".ref($pre));
1451 $p = $p->{$_} for (@$pref_path, $pre);
1453 $self->throw_exception (
1454 "Unable to resolve prefetch $pre - join alias map does not contain an entry for path: "
1455 . join (' -> ', @$pref_path, $pre)
1456 ) if (ref $p->{-join_aliases} ne 'ARRAY' or not @{$p->{-join_aliases}} );
1458 my $as = shift @{$p->{-join_aliases}};
1460 my $rel_info = $self->relationship_info( $pre );
1461 $self->throw_exception( $self->name . " has no such relationship '$pre'" )
1463 my $as_prefix = ($alias =~ /^.*?\.(.+)$/ ? $1.'.' : '');
1464 my $rel_source = $self->related_source($pre);
1466 if (exists $rel_info->{attrs}{accessor}
1467 && $rel_info->{attrs}{accessor} eq 'multi') {
1468 $self->throw_exception(
1469 "Can't prefetch has_many ${pre} (join cond too complex)")
1470 unless ref($rel_info->{cond}) eq 'HASH';
1471 my $dots = @{[$as_prefix =~ m/\./g]} + 1; # +1 to match the ".${as_prefix}"
1472 if (my ($fail) = grep { @{[$_ =~ m/\./g]} == $dots }
1473 keys %{$collapse}) {
1474 my ($last) = ($fail =~ /([^\.]+)$/);
1476 "Prefetching multiple has_many rels ${last} and ${pre} "
1477 .(length($as_prefix)
1478 ? "at the same level (${as_prefix}) "
1481 . 'will explode the number of row objects retrievable via ->next or ->all. '
1482 . 'Use at your own risk.'
1485 #my @col = map { (/^self\.(.+)$/ ? ("${as_prefix}.$1") : ()); }
1486 # values %{$rel_info->{cond}};
1487 $collapse->{".${as_prefix}${pre}"} = [ $rel_source->primary_columns ];
1488 # action at a distance. prepending the '.' allows simpler code
1489 # in ResultSet->_collapse_result
1490 my @key = map { (/^foreign\.(.+)$/ ? ($1) : ()); }
1491 keys %{$rel_info->{cond}};
1492 my @ord = (ref($rel_info->{attrs}{order_by}) eq 'ARRAY'
1493 ? @{$rel_info->{attrs}{order_by}}
1494 : (defined $rel_info->{attrs}{order_by}
1495 ? ($rel_info->{attrs}{order_by})
1497 push(@$order, map { "${as}.$_" } (@key, @ord));
1500 return map { [ "${as}.$_", "${as_prefix}${pre}.$_", ] }
1501 $rel_source->columns;
1505 =head2 related_source
1509 =item Arguments: $relname
1511 =item Return value: $source
1515 Returns the result source object for the given relationship.
1519 sub related_source {
1520 my ($self, $rel) = @_;
1521 if( !$self->has_relationship( $rel ) ) {
1522 $self->throw_exception("No such relationship '$rel'");
1524 return $self->schema->source($self->relationship_info($rel)->{source});
1527 =head2 related_class
1531 =item Arguments: $relname
1533 =item Return value: $classname
1537 Returns the class name for objects in the given relationship.
1542 my ($self, $rel) = @_;
1543 if( !$self->has_relationship( $rel ) ) {
1544 $self->throw_exception("No such relationship '$rel'");
1546 return $self->schema->class($self->relationship_info($rel)->{source});
1551 Obtain a new handle to this source. Returns an instance of a
1552 L<DBIx::Class::ResultSourceHandle>.
1557 return new DBIx::Class::ResultSourceHandle({
1558 schema => $_[0]->schema,
1559 source_moniker => $_[0]->source_name
1563 =head2 throw_exception
1565 See L<DBIx::Class::Schema/"throw_exception">.
1569 sub throw_exception {
1571 if (defined $self->schema) {
1572 $self->schema->throw_exception(@_);
1580 Stores a hashref of per-source metadata. No specific key names
1581 have yet been standardized, the examples below are purely hypothetical
1582 and don't actually accomplish anything on their own:
1584 __PACKAGE__->source_info({
1585 "_tablespace" => 'fast_disk_array_3',
1586 "_engine" => 'InnoDB',
1593 $class->new({attribute_name => value});
1595 Creates a new ResultSource object. Not normally called directly by end users.
1597 =head2 column_info_from_storage
1601 =item Arguments: 1/0 (default: 0)
1603 =item Return value: 1/0
1607 __PACKAGE__->column_info_from_storage(1);
1609 Enables the on-demand automatic loading of the above column
1610 metadata from storage as neccesary. This is *deprecated*, and
1611 should not be used. It will be removed before 1.0.
1616 Matt S. Trout <mst@shadowcatsystems.co.uk>
1620 You may distribute this code under the same terms as Perl itself.