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 automatically
249 from a sequence or function (if supported by your Storage driver.) For a
250 sequence, if you do not use a trigger to get the nextval, you have to set the
251 L</sequence> value as well.
253 Also set this for MSSQL columns with the 'uniqueidentifier'
254 L<DBIx::Class::ResultSource/data_type> whose values you want to automatically
255 generate using C<NEWID()>, unless they are a primary key in which case this will
260 This is used by L<DBIx::Class::Schema/deploy> and L<SQL::Translator>
261 to add extra non-generic data to the column. For example: C<< extra
262 => { unsigned => 1} >> is used by the MySQL producer to set an integer
263 column to unsigned. For more details, see
264 L<SQL::Translator::Producer::MySQL>.
272 =item Arguments: $colname, \%columninfo?
274 =item Return value: 1/0 (true/false)
278 $source->add_column('col' => \%info);
280 Add a single column and optional column info. Uses the same column
281 info keys as L</add_columns>.
286 my ($self, @cols) = @_;
287 $self->_ordered_columns(\@cols) unless $self->_ordered_columns;
290 my $columns = $self->_columns;
291 while (my $col = shift @cols) {
292 # If next entry is { ... } use that for the column info, if not
293 # use an empty hashref
294 my $column_info = ref $cols[0] ? shift(@cols) : {};
295 push(@added, $col) unless exists $columns->{$col};
296 $columns->{$col} = $column_info;
298 push @{ $self->_ordered_columns }, @added;
302 sub add_column { shift->add_columns(@_); } # DO NOT CHANGE THIS TO GLOB
308 =item Arguments: $colname
310 =item Return value: 1/0 (true/false)
314 if ($source->has_column($colname)) { ... }
316 Returns true if the source has a column of this name, false otherwise.
321 my ($self, $column) = @_;
322 return exists $self->_columns->{$column};
329 =item Arguments: $colname
331 =item Return value: Hashref of info
335 my $info = $source->column_info($col);
337 Returns the column metadata hashref for a column, as originally passed
338 to L</add_columns>. See L</add_columns> above for information on the
339 contents of the hashref.
344 my ($self, $column) = @_;
345 $self->throw_exception("No such column $column")
346 unless exists $self->_columns->{$column};
347 #warn $self->{_columns_info_loaded}, "\n";
348 if ( ! $self->_columns->{$column}{data_type}
349 and $self->column_info_from_storage
350 and ! $self->{_columns_info_loaded}
351 and $self->schema and $self->storage )
353 $self->{_columns_info_loaded}++;
356 # eval for the case of storage without table
357 eval { $info = $self->storage->columns_info_for( $self->from ) };
359 for my $realcol ( keys %{$info} ) {
360 $lc_info->{lc $realcol} = $info->{$realcol};
362 foreach my $col ( keys %{$self->_columns} ) {
363 $self->_columns->{$col} = {
364 %{ $self->_columns->{$col} },
365 %{ $info->{$col} || $lc_info->{lc $col} || {} }
370 return $self->_columns->{$column};
377 =item Arguments: None
379 =item Return value: Ordered list of column names
383 my @column_names = $source->columns;
385 Returns all column names in the order they were declared to L</add_columns>.
391 $self->throw_exception(
392 "columns() is a read-only accessor, did you mean add_columns()?"
394 return @{$self->{_ordered_columns}||[]};
397 =head2 remove_columns
401 =item Arguments: @colnames
403 =item Return value: undefined
407 $source->remove_columns(qw/col1 col2 col3/);
409 Removes the given list of columns by name, from the result source.
411 B<Warning>: Removing a column that is also used in the sources primary
412 key, or in one of the sources unique constraints, B<will> result in a
413 broken result source.
419 =item Arguments: $colname
421 =item Return value: undefined
425 $source->remove_column('col');
427 Remove a single column by name from the result source, similar to
430 B<Warning>: Removing a column that is also used in the sources primary
431 key, or in one of the sources unique constraints, B<will> result in a
432 broken result source.
437 my ($self, @to_remove) = @_;
439 my $columns = $self->_columns
444 delete $columns->{$_};
448 $self->_ordered_columns([ grep { not $to_remove{$_} } @{$self->_ordered_columns} ]);
451 sub remove_column { shift->remove_columns(@_); } # DO NOT CHANGE THIS TO GLOB
453 =head2 set_primary_key
457 =item Arguments: @cols
459 =item Return value: undefined
463 Defines one or more columns as primary key for this source. Must be
464 called after L</add_columns>.
466 Additionally, defines a L<unique constraint|add_unique_constraint>
469 The primary key columns are used by L<DBIx::Class::PK::Auto> to
470 retrieve automatically created values from the database. They are also
471 used as default joining columns when specifying relationships, see
472 L<DBIx::Class::Relationship>.
476 sub set_primary_key {
477 my ($self, @cols) = @_;
478 # check if primary key columns are valid columns
479 foreach my $col (@cols) {
480 $self->throw_exception("No such column $col on table " . $self->name)
481 unless $self->has_column($col);
483 $self->_primaries(\@cols);
485 $self->add_unique_constraint(primary => \@cols);
488 =head2 primary_columns
492 =item Arguments: None
494 =item Return value: Ordered list of primary column names
498 Read-only accessor which returns the list of primary keys, supplied by
503 sub primary_columns {
504 return @{shift->_primaries||[]};
507 =head2 add_unique_constraint
511 =item Arguments: $name?, \@colnames
513 =item Return value: undefined
517 Declare a unique constraint on this source. Call once for each unique
520 # For UNIQUE (column1, column2)
521 __PACKAGE__->add_unique_constraint(
522 constraint_name => [ qw/column1 column2/ ],
525 Alternatively, you can specify only the columns:
527 __PACKAGE__->add_unique_constraint([ qw/column1 column2/ ]);
529 This will result in a unique constraint named
530 C<table_column1_column2>, where C<table> is replaced with the table
533 Unique constraints are used, for example, when you pass the constraint
534 name as the C<key> attribute to L<DBIx::Class::ResultSet/find>. Then
535 only columns in the constraint are searched.
537 Throws an error if any of the given column names do not yet exist on
542 sub add_unique_constraint {
547 $name ||= $self->name_unique_constraint($cols);
549 foreach my $col (@$cols) {
550 $self->throw_exception("No such column $col on table " . $self->name)
551 unless $self->has_column($col);
554 my %unique_constraints = $self->unique_constraints;
555 $unique_constraints{$name} = $cols;
556 $self->_unique_constraints(\%unique_constraints);
559 =head2 name_unique_constraint
563 =item Arguments: @colnames
565 =item Return value: Constraint name
569 $source->table('mytable');
570 $source->name_unique_constraint('col1', 'col2');
574 Return a name for a unique constraint containing the specified
575 columns. The name is created by joining the table name and each column
576 name, using an underscore character.
578 For example, a constraint on a table named C<cd> containing the columns
579 C<artist> and C<title> would result in a constraint name of C<cd_artist_title>.
581 This is used by L</add_unique_constraint> if you do not specify the
582 optional constraint name.
586 sub name_unique_constraint {
587 my ($self, $cols) = @_;
589 return join '_', $self->name, @$cols;
592 =head2 unique_constraints
596 =item Arguments: None
598 =item Return value: Hash of unique constraint data
602 $source->unique_constraints();
604 Read-only accessor which returns a hash of unique constraints on this
607 The hash is keyed by constraint name, and contains an arrayref of
608 column names as values.
612 sub unique_constraints {
613 return %{shift->_unique_constraints||{}};
616 =head2 unique_constraint_names
620 =item Arguments: None
622 =item Return value: Unique constraint names
626 $source->unique_constraint_names();
628 Returns the list of unique constraint names defined on this source.
632 sub unique_constraint_names {
635 my %unique_constraints = $self->unique_constraints;
637 return keys %unique_constraints;
640 =head2 unique_constraint_columns
644 =item Arguments: $constraintname
646 =item Return value: List of constraint columns
650 $source->unique_constraint_columns('myconstraint');
652 Returns the list of columns that make up the specified unique constraint.
656 sub unique_constraint_columns {
657 my ($self, $constraint_name) = @_;
659 my %unique_constraints = $self->unique_constraints;
661 $self->throw_exception(
662 "Unknown unique constraint $constraint_name on '" . $self->name . "'"
663 ) unless exists $unique_constraints{$constraint_name};
665 return @{ $unique_constraints{$constraint_name} };
668 =head2 sqlt_deploy_callback
672 =item Arguments: $callback
676 __PACKAGE__->sqlt_deploy_callback('mycallbackmethod');
678 An accessor to set a callback to be called during deployment of
679 the schema via L<DBIx::Class::Schema/create_ddl_dir> or
680 L<DBIx::Class::Schema/deploy>.
682 The callback can be set as either a code reference or the name of a
683 method in the current result class.
685 If not set, the L</default_sqlt_deploy_hook> is called.
687 Your callback will be passed the $source object representing the
688 ResultSource instance being deployed, and the
689 L<SQL::Translator::Schema::Table> object being created from it. The
690 callback can be used to manipulate the table object or add your own
691 customised indexes. If you need to manipulate a non-table object, use
692 the L<DBIx::Class::Schema/sqlt_deploy_hook>.
694 See L<DBIx::Class::Manual::Cookbook/Adding Indexes And Functions To
695 Your SQL> for examples.
697 This sqlt deployment callback can only be used to manipulate
698 SQL::Translator objects as they get turned into SQL. To execute
699 post-deploy statements which SQL::Translator does not currently
700 handle, override L<DBIx::Class::Schema/deploy> in your Schema class
701 and call L<dbh_do|DBIx::Class::Storage::DBI/dbh_do>.
703 =head2 default_sqlt_deploy_hook
707 =item Arguments: $source, $sqlt_table
709 =item Return value: undefined
713 This is the sensible default for L</sqlt_deploy_callback>.
715 If a method named C<sqlt_deploy_hook> exists in your Result class, it
716 will be called and passed the current C<$source> and the
717 C<$sqlt_table> being deployed.
721 sub default_sqlt_deploy_hook {
724 my $class = $self->result_class;
726 if ($class and $class->can('sqlt_deploy_hook')) {
727 $class->sqlt_deploy_hook(@_);
731 sub _invoke_sqlt_deploy_hook {
733 if ( my $hook = $self->sqlt_deploy_callback) {
742 =item Arguments: None
744 =item Return value: $resultset
748 Returns a resultset for the given source. This will initially be created
751 $self->resultset_class->new($self, $self->resultset_attributes)
753 but is cached from then on unless resultset_class changes.
755 =head2 resultset_class
759 =item Arguments: $classname
761 =item Return value: $classname
765 package My::Schema::ResultSet::Artist;
766 use base 'DBIx::Class::ResultSet';
769 # In the result class
770 __PACKAGE__->resultset_class('My::Schema::ResultSet::Artist');
773 $source->resultset_class('My::Schema::ResultSet::Artist');
775 Set the class of the resultset. This is useful if you want to create your
776 own resultset methods. Create your own class derived from
777 L<DBIx::Class::ResultSet>, and set it here. If called with no arguments,
778 this method returns the name of the existing resultset class, if one
781 =head2 resultset_attributes
785 =item Arguments: \%attrs
787 =item Return value: \%attrs
791 # In the result class
792 __PACKAGE__->resultset_attributes({ order_by => [ 'id' ] });
795 $source->resultset_attributes({ order_by => [ 'id' ] });
797 Store a collection of resultset attributes, that will be set on every
798 L<DBIx::Class::ResultSet> produced from this result source. For a full
799 list see L<DBIx::Class::ResultSet/ATTRIBUTES>.
805 $self->throw_exception(
806 'resultset does not take any arguments. If you want another resultset, '.
807 'call it on the schema instead.'
810 return $self->resultset_class->new(
813 %{$self->{resultset_attributes}},
814 %{$self->schema->default_resultset_attributes}
823 =item Arguments: $source_name
825 =item Result value: $source_name
829 Set an alternate name for the result source when it is loaded into a schema.
830 This is useful if you want to refer to a result source by a name other than
833 package ArchivedBooks;
834 use base qw/DBIx::Class/;
835 __PACKAGE__->table('books_archive');
836 __PACKAGE__->source_name('Books');
838 # from your schema...
839 $schema->resultset('Books')->find(1);
845 =item Arguments: None
847 =item Return value: FROM clause
851 my $from_clause = $source->from();
853 Returns an expression of the source to be supplied to storage to specify
854 retrieval from this source. In the case of a database, the required FROM
861 =item Arguments: None
863 =item Return value: A schema object
867 my $schema = $source->schema();
869 Returns the L<DBIx::Class::Schema> object that this result source
876 =item Arguments: None
878 =item Return value: A Storage object
882 $source->storage->debug(1);
884 Returns the storage handle for the current schema.
886 See also: L<DBIx::Class::Storage>
890 sub storage { shift->schema->storage; }
892 =head2 add_relationship
896 =item Arguments: $relname, $related_source_name, \%cond, [ \%attrs ]
898 =item Return value: 1/true if it succeeded
902 $source->add_relationship('relname', 'related_source', $cond, $attrs);
904 L<DBIx::Class::Relationship> describes a series of methods which
905 create pre-defined useful types of relationships. Look there first
906 before using this method directly.
908 The relationship name can be arbitrary, but must be unique for each
909 relationship attached to this result source. 'related_source' should
910 be the name with which the related result source was registered with
911 the current schema. For example:
913 $schema->source('Book')->add_relationship('reviews', 'Review', {
914 'foreign.book_id' => 'self.id',
917 The condition C<$cond> needs to be an L<SQL::Abstract>-style
918 representation of the join between the tables. For example, if you're
919 creating a relation from Author to Book,
921 { 'foreign.author_id' => 'self.id' }
923 will result in the JOIN clause
925 author me JOIN book foreign ON foreign.author_id = me.id
927 You can specify as many foreign => self mappings as necessary.
929 Valid attributes are as follows:
935 Explicitly specifies the type of join to use in the relationship. Any
936 SQL join type is valid, e.g. C<LEFT> or C<RIGHT>. It will be placed in
937 the SQL command immediately before C<JOIN>.
941 An arrayref containing a list of accessors in the foreign class to proxy in
942 the main class. If, for example, you do the following:
944 CD->might_have(liner_notes => 'LinerNotes', undef, {
945 proxy => [ qw/notes/ ],
948 Then, assuming LinerNotes has an accessor named notes, you can do:
950 my $cd = CD->find(1);
951 # set notes -- LinerNotes object is created if it doesn't exist
952 $cd->notes('Notes go here');
956 Specifies the type of accessor that should be created for the
957 relationship. Valid values are C<single> (for when there is only a single
958 related object), C<multi> (when there can be many), and C<filter> (for
959 when there is a single related object, but you also want the relationship
960 accessor to double as a column accessor). For C<multi> accessors, an
961 add_to_* method is also created, which calls C<create_related> for the
966 Throws an exception if the condition is improperly supplied, or cannot
971 sub add_relationship {
972 my ($self, $rel, $f_source_name, $cond, $attrs) = @_;
973 $self->throw_exception("Can't create relationship without join condition")
977 # Check foreign and self are right in cond
978 if ( (ref $cond ||'') eq 'HASH') {
980 $self->throw_exception("Keys of condition should be of form 'foreign.col', not '$_'")
981 if /\./ && !/^foreign\./;
985 my %rels = %{ $self->_relationships };
986 $rels{$rel} = { class => $f_source_name,
987 source => $f_source_name,
990 $self->_relationships(\%rels);
994 # XXX disabled. doesn't work properly currently. skip in tests.
996 my $f_source = $self->schema->source($f_source_name);
998 $self->ensure_class_loaded($f_source_name);
999 $f_source = $f_source_name->result_source;
1000 #my $s_class = ref($self->schema);
1001 #$f_source_name =~ m/^${s_class}::(.*)$/;
1002 #$self->schema->register_class(($1 || $f_source_name), $f_source_name);
1003 #$f_source = $self->schema->source($f_source_name);
1005 return unless $f_source; # Can't test rel without f_source
1007 eval { $self->_resolve_join($rel, 'me', {}, []) };
1009 if ($@) { # If the resolve failed, back out and re-throw the error
1010 delete $rels{$rel}; #
1011 $self->_relationships(\%rels);
1012 $self->throw_exception("Error creating relationship $rel: $@");
1017 =head2 relationships
1021 =item Arguments: None
1023 =item Return value: List of relationship names
1027 my @relnames = $source->relationships();
1029 Returns all relationship names for this source.
1034 return keys %{shift->_relationships};
1037 =head2 relationship_info
1041 =item Arguments: $relname
1043 =item Return value: Hashref of relation data,
1047 Returns a hash of relationship information for the specified relationship
1048 name. The keys/values are as specified for L</add_relationship>.
1052 sub relationship_info {
1053 my ($self, $rel) = @_;
1054 return $self->_relationships->{$rel};
1057 =head2 has_relationship
1061 =item Arguments: $rel
1063 =item Return value: 1/0 (true/false)
1067 Returns true if the source has a relationship of this name, false otherwise.
1071 sub has_relationship {
1072 my ($self, $rel) = @_;
1073 return exists $self->_relationships->{$rel};
1076 =head2 reverse_relationship_info
1080 =item Arguments: $relname
1082 =item Return value: Hashref of relationship data
1086 Looks through all the relationships on the source this relationship
1087 points to, looking for one whose condition is the reverse of the
1088 condition on this relationship.
1090 A common use of this is to find the name of the C<belongs_to> relation
1091 opposing a C<has_many> relation. For definition of these look in
1092 L<DBIx::Class::Relationship>.
1094 The returned hashref is keyed by the name of the opposing
1095 relationship, and contains its data in the same manner as
1096 L</relationship_info>.
1100 sub reverse_relationship_info {
1101 my ($self, $rel) = @_;
1102 my $rel_info = $self->relationship_info($rel);
1105 return $ret unless ((ref $rel_info->{cond}) eq 'HASH');
1107 my @cond = keys(%{$rel_info->{cond}});
1108 my @refkeys = map {/^\w+\.(\w+)$/} @cond;
1109 my @keys = map {$rel_info->{cond}->{$_} =~ /^\w+\.(\w+)$/} @cond;
1111 # Get the related result source for this relationship
1112 my $othertable = $self->related_source($rel);
1114 # Get all the relationships for that source that related to this source
1115 # whose foreign column set are our self columns on $rel and whose self
1116 # columns are our foreign columns on $rel.
1117 my @otherrels = $othertable->relationships();
1118 my $otherrelationship;
1119 foreach my $otherrel (@otherrels) {
1120 my $otherrel_info = $othertable->relationship_info($otherrel);
1122 my $back = $othertable->related_source($otherrel);
1123 next unless $back->source_name eq $self->source_name;
1127 if (ref $otherrel_info->{cond} eq 'HASH') {
1128 @othertestconds = ($otherrel_info->{cond});
1130 elsif (ref $otherrel_info->{cond} eq 'ARRAY') {
1131 @othertestconds = @{$otherrel_info->{cond}};
1137 foreach my $othercond (@othertestconds) {
1138 my @other_cond = keys(%$othercond);
1139 my @other_refkeys = map {/^\w+\.(\w+)$/} @other_cond;
1140 my @other_keys = map {$othercond->{$_} =~ /^\w+\.(\w+)$/} @other_cond;
1141 next if (!$self->_compare_relationship_keys(\@refkeys, \@other_keys) ||
1142 !$self->_compare_relationship_keys(\@other_refkeys, \@keys));
1143 $ret->{$otherrel} = $otherrel_info;
1149 sub compare_relationship_keys {
1150 carp 'compare_relationship_keys is a private method, stop calling it';
1152 $self->_compare_relationship_keys (@_);
1155 # Returns true if both sets of keynames are the same, false otherwise.
1156 sub _compare_relationship_keys {
1157 my ($self, $keys1, $keys2) = @_;
1159 # Make sure every keys1 is in keys2
1161 foreach my $key (@$keys1) {
1163 foreach my $prim (@$keys2) {
1164 if ($prim eq $key) {
1172 # Make sure every key2 is in key1
1174 foreach my $prim (@$keys2) {
1176 foreach my $key (@$keys1) {
1177 if ($prim eq $key) {
1190 carp 'resolve_join is a private method, stop calling it';
1192 $self->_resolve_join (@_);
1195 # Returns the {from} structure used to express JOIN conditions
1197 my ($self, $join, $alias, $seen, $jpath, $force_left) = @_;
1199 # we need a supplied one, because we do in-place modifications, no returns
1200 $self->throw_exception ('You must supply a seen hashref as the 3rd argument to _resolve_join')
1201 unless ref $seen eq 'HASH';
1203 $self->throw_exception ('You must supply a joinpath arrayref as the 4th argument to _resolve_join')
1204 unless ref $jpath eq 'ARRAY';
1208 if (ref $join eq 'ARRAY') {
1211 $self->_resolve_join($_, $alias, $seen, $jpath, $force_left);
1213 } elsif (ref $join eq 'HASH') {
1216 my $as = ($seen->{$_} ? join ('_', $_, $seen->{$_} + 1) : $_); # the actual seen value will be incremented below
1217 local $force_left->{force} = $force_left->{force};
1219 $self->_resolve_join($_, $alias, $seen, [@$jpath], $force_left),
1220 $self->related_source($_)->_resolve_join(
1221 $join->{$_}, $as, $seen, [@$jpath, $_], $force_left
1225 } elsif (ref $join) {
1226 $self->throw_exception("No idea how to resolve join reftype ".ref $join);
1229 return() unless defined $join;
1231 my $count = ++$seen->{$join};
1232 my $as = ($count > 1 ? "${join}_${count}" : $join);
1234 my $rel_info = $self->relationship_info($join);
1235 $self->throw_exception("No such relationship ${join}") unless $rel_info;
1240 $type = $rel_info->{attrs}{join_type} || '';
1241 $force_left = 1 if lc($type) eq 'left';
1244 my $rel_src = $self->related_source($join);
1245 return [ { $as => $rel_src->from,
1246 -source_handle => $rel_src->handle,
1247 -join_type => $type,
1248 -join_path => [@$jpath, $join],
1250 -relation_chain_depth => $seen->{-relation_chain_depth} || 0,
1252 $self->_resolve_condition($rel_info->{cond}, $as, $alias) ];
1257 carp 'pk_depends_on is a private method, stop calling it';
1259 $self->_pk_depends_on (@_);
1262 # Determines whether a relation is dependent on an object from this source
1263 # having already been inserted. Takes the name of the relationship and a
1264 # hashref of columns of the related object.
1265 sub _pk_depends_on {
1266 my ($self, $relname, $rel_data) = @_;
1267 my $cond = $self->relationship_info($relname)->{cond};
1269 return 0 unless ref($cond) eq 'HASH';
1271 # map { foreign.foo => 'self.bar' } to { bar => 'foo' }
1273 my $keyhash = { map { my $x = $_; $x =~ s/.*\.//; $x; } reverse %$cond };
1275 # assume anything that references our PK probably is dependent on us
1276 # rather than vice versa, unless the far side is (a) defined or (b)
1279 my $rel_source = $self->related_source($relname);
1281 foreach my $p ($self->primary_columns) {
1282 if (exists $keyhash->{$p}) {
1283 unless (defined($rel_data->{$keyhash->{$p}})
1284 || $rel_source->column_info($keyhash->{$p})
1285 ->{is_auto_increment}) {
1294 sub resolve_condition {
1295 carp 'resolve_condition is a private method, stop calling it';
1297 $self->_resolve_condition (@_);
1300 # Resolves the passed condition to a concrete query fragment. If given an alias,
1301 # returns a join condition; if given an object, inverts that object to produce
1302 # a related conditional from that object.
1303 our $UNRESOLVABLE_CONDITION = \'1 = 0';
1305 sub _resolve_condition {
1306 my ($self, $cond, $as, $for) = @_;
1307 if (ref $cond eq 'HASH') {
1309 foreach my $k (keys %{$cond}) {
1310 my $v = $cond->{$k};
1311 # XXX should probably check these are valid columns
1312 $k =~ s/^foreign\.// ||
1313 $self->throw_exception("Invalid rel cond key ${k}");
1314 $v =~ s/^self\.// ||
1315 $self->throw_exception("Invalid rel cond val ${v}");
1316 if (ref $for) { # Object
1317 #warn "$self $k $for $v";
1318 unless ($for->has_column_loaded($v)) {
1319 if ($for->in_storage) {
1320 $self->throw_exception(
1321 "Column ${v} not loaded or not passed to new() prior to insert()"
1322 ." on ${for} trying to resolve relationship (maybe you forgot "
1323 ."to call ->discard_changes to get defaults from the db)"
1326 return $UNRESOLVABLE_CONDITION;
1328 $ret{$k} = $for->get_column($v);
1329 #$ret{$k} = $for->get_column($v) if $for->has_column_loaded($v);
1331 } elsif (!defined $for) { # undef, i.e. "no object"
1333 } elsif (ref $as eq 'HASH') { # reverse hashref
1334 $ret{$v} = $as->{$k};
1335 } elsif (ref $as) { # reverse object
1336 $ret{$v} = $as->get_column($k);
1337 } elsif (!defined $as) { # undef, i.e. "no reverse object"
1340 $ret{"${as}.${k}"} = "${for}.${v}";
1344 } elsif (ref $cond eq 'ARRAY') {
1345 return [ map { $self->_resolve_condition($_, $as, $for) } @$cond ];
1347 die("Can't handle condition $cond yet :(");
1351 # Legacy code, needs to go entirely away (fully replaced by _resolve_prefetch)
1352 sub resolve_prefetch {
1353 carp 'resolve_prefetch is a private method, stop calling it';
1355 my ($self, $pre, $alias, $seen, $order, $collapse) = @_;
1357 if( ref $pre eq 'ARRAY' ) {
1359 map { $self->resolve_prefetch( $_, $alias, $seen, $order, $collapse ) }
1362 elsif( ref $pre eq 'HASH' ) {
1365 $self->resolve_prefetch($_, $alias, $seen, $order, $collapse),
1366 $self->related_source($_)->resolve_prefetch(
1367 $pre->{$_}, "${alias}.$_", $seen, $order, $collapse)
1372 $self->throw_exception(
1373 "don't know how to resolve prefetch reftype ".ref($pre));
1376 my $count = ++$seen->{$pre};
1377 my $as = ($count > 1 ? "${pre}_${count}" : $pre);
1378 my $rel_info = $self->relationship_info( $pre );
1379 $self->throw_exception( $self->name . " has no such relationship '$pre'" )
1381 my $as_prefix = ($alias =~ /^.*?\.(.+)$/ ? $1.'.' : '');
1382 my $rel_source = $self->related_source($pre);
1384 if (exists $rel_info->{attrs}{accessor}
1385 && $rel_info->{attrs}{accessor} eq 'multi') {
1386 $self->throw_exception(
1387 "Can't prefetch has_many ${pre} (join cond too complex)")
1388 unless ref($rel_info->{cond}) eq 'HASH';
1389 my $dots = @{[$as_prefix =~ m/\./g]} + 1; # +1 to match the ".${as_prefix}"
1390 if (my ($fail) = grep { @{[$_ =~ m/\./g]} == $dots }
1391 keys %{$collapse}) {
1392 my ($last) = ($fail =~ /([^\.]+)$/);
1394 "Prefetching multiple has_many rels ${last} and ${pre} "
1395 .(length($as_prefix)
1396 ? "at the same level (${as_prefix}) "
1399 . 'will explode the number of row objects retrievable via ->next or ->all. '
1400 . 'Use at your own risk.'
1403 #my @col = map { (/^self\.(.+)$/ ? ("${as_prefix}.$1") : ()); }
1404 # values %{$rel_info->{cond}};
1405 $collapse->{".${as_prefix}${pre}"} = [ $rel_source->primary_columns ];
1406 # action at a distance. prepending the '.' allows simpler code
1407 # in ResultSet->_collapse_result
1408 my @key = map { (/^foreign\.(.+)$/ ? ($1) : ()); }
1409 keys %{$rel_info->{cond}};
1410 my @ord = (ref($rel_info->{attrs}{order_by}) eq 'ARRAY'
1411 ? @{$rel_info->{attrs}{order_by}}
1412 : (defined $rel_info->{attrs}{order_by}
1413 ? ($rel_info->{attrs}{order_by})
1415 push(@$order, map { "${as}.$_" } (@key, @ord));
1418 return map { [ "${as}.$_", "${as_prefix}${pre}.$_", ] }
1419 $rel_source->columns;
1423 # Accepts one or more relationships for the current source and returns an
1424 # array of column names for each of those relationships. Column names are
1425 # prefixed relative to the current source, in accordance with where they appear
1426 # in the supplied relationships. Needs an alias_map generated by
1427 # $rs->_joinpath_aliases
1429 sub _resolve_prefetch {
1430 my ($self, $pre, $alias, $alias_map, $order, $collapse, $pref_path) = @_;
1433 if( ref $pre eq 'ARRAY' ) {
1435 map { $self->_resolve_prefetch( $_, $alias, $alias_map, $order, $collapse, [ @$pref_path ] ) }
1438 elsif( ref $pre eq 'HASH' ) {
1441 $self->_resolve_prefetch($_, $alias, $alias_map, $order, $collapse, [ @$pref_path ] ),
1442 $self->related_source($_)->_resolve_prefetch(
1443 $pre->{$_}, "${alias}.$_", $alias_map, $order, $collapse, [ @$pref_path, $_] )
1448 $self->throw_exception(
1449 "don't know how to resolve prefetch reftype ".ref($pre));
1453 $p = $p->{$_} for (@$pref_path, $pre);
1455 $self->throw_exception (
1456 "Unable to resolve prefetch $pre - join alias map does not contain an entry for path: "
1457 . join (' -> ', @$pref_path, $pre)
1458 ) if (ref $p->{-join_aliases} ne 'ARRAY' or not @{$p->{-join_aliases}} );
1460 my $as = shift @{$p->{-join_aliases}};
1462 my $rel_info = $self->relationship_info( $pre );
1463 $self->throw_exception( $self->name . " has no such relationship '$pre'" )
1465 my $as_prefix = ($alias =~ /^.*?\.(.+)$/ ? $1.'.' : '');
1466 my $rel_source = $self->related_source($pre);
1468 if (exists $rel_info->{attrs}{accessor}
1469 && $rel_info->{attrs}{accessor} eq 'multi') {
1470 $self->throw_exception(
1471 "Can't prefetch has_many ${pre} (join cond too complex)")
1472 unless ref($rel_info->{cond}) eq 'HASH';
1473 my $dots = @{[$as_prefix =~ m/\./g]} + 1; # +1 to match the ".${as_prefix}"
1474 if (my ($fail) = grep { @{[$_ =~ m/\./g]} == $dots }
1475 keys %{$collapse}) {
1476 my ($last) = ($fail =~ /([^\.]+)$/);
1478 "Prefetching multiple has_many rels ${last} and ${pre} "
1479 .(length($as_prefix)
1480 ? "at the same level (${as_prefix}) "
1483 . 'will explode the number of row objects retrievable via ->next or ->all. '
1484 . 'Use at your own risk.'
1487 #my @col = map { (/^self\.(.+)$/ ? ("${as_prefix}.$1") : ()); }
1488 # values %{$rel_info->{cond}};
1489 $collapse->{".${as_prefix}${pre}"} = [ $rel_source->primary_columns ];
1490 # action at a distance. prepending the '.' allows simpler code
1491 # in ResultSet->_collapse_result
1492 my @key = map { (/^foreign\.(.+)$/ ? ($1) : ()); }
1493 keys %{$rel_info->{cond}};
1494 my @ord = (ref($rel_info->{attrs}{order_by}) eq 'ARRAY'
1495 ? @{$rel_info->{attrs}{order_by}}
1496 : (defined $rel_info->{attrs}{order_by}
1497 ? ($rel_info->{attrs}{order_by})
1499 push(@$order, map { "${as}.$_" } (@key, @ord));
1502 return map { [ "${as}.$_", "${as_prefix}${pre}.$_", ] }
1503 $rel_source->columns;
1507 =head2 related_source
1511 =item Arguments: $relname
1513 =item Return value: $source
1517 Returns the result source object for the given relationship.
1521 sub related_source {
1522 my ($self, $rel) = @_;
1523 if( !$self->has_relationship( $rel ) ) {
1524 $self->throw_exception("No such relationship '$rel'");
1526 return $self->schema->source($self->relationship_info($rel)->{source});
1529 =head2 related_class
1533 =item Arguments: $relname
1535 =item Return value: $classname
1539 Returns the class name for objects in the given relationship.
1544 my ($self, $rel) = @_;
1545 if( !$self->has_relationship( $rel ) ) {
1546 $self->throw_exception("No such relationship '$rel'");
1548 return $self->schema->class($self->relationship_info($rel)->{source});
1553 Obtain a new handle to this source. Returns an instance of a
1554 L<DBIx::Class::ResultSourceHandle>.
1559 return new DBIx::Class::ResultSourceHandle({
1560 schema => $_[0]->schema,
1561 source_moniker => $_[0]->source_name
1565 =head2 throw_exception
1567 See L<DBIx::Class::Schema/"throw_exception">.
1571 sub throw_exception {
1573 if (defined $self->schema) {
1574 $self->schema->throw_exception(@_);
1582 Stores a hashref of per-source metadata. No specific key names
1583 have yet been standardized, the examples below are purely hypothetical
1584 and don't actually accomplish anything on their own:
1586 __PACKAGE__->source_info({
1587 "_tablespace" => 'fast_disk_array_3',
1588 "_engine" => 'InnoDB',
1595 $class->new({attribute_name => value});
1597 Creates a new ResultSource object. Not normally called directly by end users.
1599 =head2 column_info_from_storage
1603 =item Arguments: 1/0 (default: 0)
1605 =item Return value: 1/0
1609 __PACKAGE__->column_info_from_storage(1);
1611 Enables the on-demand automatic loading of the above column
1612 metadata from storage as neccesary. This is *deprecated*, and
1613 should not be used. It will be removed before 1.0.
1618 Matt S. Trout <mst@shadowcatsystems.co.uk>
1622 You may distribute this code under the same terms as Perl itself.