1 package DBIx::Class::ResultSet;
5 use base qw/DBIx::Class/;
7 use DBIx::Class::ResultSetColumn;
8 use Scalar::Util qw/blessed weaken reftype/;
10 use Data::Compare (); # no imports!!! guard against insane architecture
12 # not importing first() as it will clash with our own method
16 # De-duplication in _merge_attr() is disabled, but left in for reference
17 # (the merger is used for other things that ought not to be de-duped)
18 *__HM_DEDUP = sub () { 0 };
28 # this is real - CDBICompat overrides it with insanity
29 # yes, prototype won't matter, but that's for now ;)
32 __PACKAGE__->mk_group_accessors('simple' => qw/_result_class result_source/);
36 DBIx::Class::ResultSet - Represents a query used for fetching a set of results.
40 my $users_rs = $schema->resultset('User');
41 while( $user = $users_rs->next) {
42 print $user->username;
45 my $registered_users_rs = $schema->resultset('User')->search({ registered => 1 });
46 my @cds_in_2005 = $schema->resultset('CD')->search({ year => 2005 })->all();
50 A ResultSet is an object which stores a set of conditions representing
51 a query. It is the backbone of DBIx::Class (i.e. the really
52 important/useful bit).
54 No SQL is executed on the database when a ResultSet is created, it
55 just stores all the conditions needed to create the query.
57 A basic ResultSet representing the data of an entire table is returned
58 by calling C<resultset> on a L<DBIx::Class::Schema> and passing in a
59 L<Source|DBIx::Class::Manual::Glossary/Source> name.
61 my $users_rs = $schema->resultset('User');
63 A new ResultSet is returned from calling L</search> on an existing
64 ResultSet. The new one will contain all the conditions of the
65 original, plus any new conditions added in the C<search> call.
67 A ResultSet also incorporates an implicit iterator. L</next> and L</reset>
68 can be used to walk through all the L<DBIx::Class::Row>s the ResultSet
71 The query that the ResultSet represents is B<only> executed against
72 the database when these methods are called:
73 L</find>, L</next>, L</all>, L</first>, L</single>, L</count>.
75 If a resultset is used in a numeric context it returns the L</count>.
76 However, if it is used in a boolean context it is B<always> true. So if
77 you want to check if a resultset has any results, you must use C<if $rs
80 =head1 CUSTOM ResultSet CLASSES THAT USE Moose
82 If you want to make your custom ResultSet classes with L<Moose>, use a template
85 package MyApp::Schema::ResultSet::User;
88 use namespace::autoclean;
90 extends 'DBIx::Class::ResultSet';
92 sub BUILDARGS { $_[2] }
96 __PACKAGE__->meta->make_immutable;
100 The L<MooseX::NonMoose> is necessary so that the L<Moose> constructor does not
101 clash with the regular ResultSet constructor. Alternatively, you can use:
103 __PACKAGE__->meta->make_immutable(inline_constructor => 0);
105 The L<BUILDARGS|Moose::Manual::Construction/BUILDARGS> is necessary because the
106 signature of the ResultSet C<new> is C<< ->new($source, \%args) >>.
110 =head2 Chaining resultsets
112 Let's say you've got a query that needs to be run to return some data
113 to the user. But, you have an authorization system in place that
114 prevents certain users from seeing certain information. So, you want
115 to construct the basic query in one method, but add constraints to it in
120 my $request = $self->get_request; # Get a request object somehow.
121 my $schema = $self->result_source->schema;
123 my $cd_rs = $schema->resultset('CD')->search({
124 title => $request->param('title'),
125 year => $request->param('year'),
128 $cd_rs = $self->apply_security_policy( $cd_rs );
130 return $cd_rs->all();
133 sub apply_security_policy {
142 =head3 Resolving conditions and attributes
144 When a resultset is chained from another resultset (ie:
145 C<my $new_rs = $old_rs->search(\%extra_cond, \%attrs)>), conditions
146 and attributes with the same keys need resolving.
148 If any of L</columns>, L</select>, L</as> are present, they reset the
149 original selection, and start the selection "clean".
151 The L</join>, L</prefetch>, L</+columns>, L</+select>, L</+as> attributes
152 are merged into the existing ones from the original resultset.
154 The L</where> and L</having> attributes, and any search conditions, are
155 merged with an SQL C<AND> to the existing condition from the original
158 All other attributes are overridden by any new ones supplied in the
161 =head2 Multiple queries
163 Since a resultset just defines a query, you can do all sorts of
164 things with it with the same object.
166 # Don't hit the DB yet.
167 my $cd_rs = $schema->resultset('CD')->search({
168 title => 'something',
172 # Each of these hits the DB individually.
173 my $count = $cd_rs->count;
174 my $most_recent = $cd_rs->get_column('date_released')->max();
175 my @records = $cd_rs->all;
177 And it's not just limited to SELECT statements.
183 $cd_rs->create({ artist => 'Fred' });
185 Which is the same as:
187 $schema->resultset('CD')->create({
188 title => 'something',
193 See: L</search>, L</count>, L</get_column>, L</all>, L</create>.
201 =item Arguments: L<$source|DBIx::Class::ResultSource>, L<\%attrs?|/ATTRIBUTES>
203 =item Return Value: L<$resultset|/search>
207 The resultset constructor. Takes a source object (usually a
208 L<DBIx::Class::ResultSourceProxy::Table>) and an attribute hash (see
209 L</ATTRIBUTES> below). Does not perform any queries -- these are
210 executed as needed by the other methods.
212 Generally you never construct a resultset manually. Instead you get one
214 C<< $schema->L<resultset|DBIx::Class::Schema/resultset>('$source_name') >>
215 or C<< $another_resultset->L<search|/search>(...) >> (the later called in
218 my $rs = $schema->resultset('CD')->search({ title => '100th Window' });
224 If called on an object, proxies to L</new_result> instead, so
226 my $cd = $schema->resultset('CD')->new({ title => 'Spoon' });
228 will return a CD object, not a ResultSet, and is equivalent to:
230 my $cd = $schema->resultset('CD')->new_result({ title => 'Spoon' });
232 Please also keep in mind that many internals call L</new_result> directly,
233 so overloading this method with the idea of intercepting new result object
234 creation B<will not work>. See also warning pertaining to L</create>.
242 return $class->new_result(@_) if ref $class;
244 my ($source, $attrs) = @_;
245 $source = $source->resolve
246 if $source->isa('DBIx::Class::ResultSourceHandle');
247 $attrs = { %{$attrs||{}} };
249 if ($attrs->{page}) {
250 $attrs->{rows} ||= 10;
253 $attrs->{alias} ||= 'me';
256 result_source => $source,
257 cond => $attrs->{where},
262 # if there is a dark selector, this means we are already in a
263 # chain and the cleanup/sanification was taken care of by
265 $self->_normalize_selection($attrs)
266 unless $attrs->{_dark_selector};
269 $attrs->{result_class} || $source->result_class
279 =item Arguments: L<$cond|DBIx::Class::SQLMaker> | undef, L<\%attrs?|/ATTRIBUTES>
281 =item Return Value: $resultset (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
285 my @cds = $cd_rs->search({ year => 2001 }); # "... WHERE year = 2001"
286 my $new_rs = $cd_rs->search({ year => 2005 });
288 my $new_rs = $cd_rs->search([ { year => 2005 }, { year => 2004 } ]);
289 # year = 2005 OR year = 2004
291 In list context, C<< ->all() >> is called implicitly on the resultset, thus
292 returning a list of L<result|DBIx::Class::Manual::ResultClass> objects instead.
293 To avoid that, use L</search_rs>.
295 If you need to pass in additional attributes but no additional condition,
296 call it as C<search(undef, \%attrs)>.
298 # "SELECT name, artistid FROM $artist_table"
299 my @all_artists = $schema->resultset('Artist')->search(undef, {
300 columns => [qw/name artistid/],
303 For a list of attributes that can be passed to C<search>, see
304 L</ATTRIBUTES>. For more examples of using this function, see
305 L<Searching|DBIx::Class::Manual::Cookbook/Searching>. For a complete
306 documentation for the first argument, see L<SQL::Abstract>
307 and its extension L<DBIx::Class::SQLMaker>.
309 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
313 Note that L</search> does not process/deflate any of the values passed in the
314 L<SQL::Abstract>-compatible search condition structure. This is unlike other
315 condition-bound methods L</new_result>, L</create> and L</find>. The user must ensure
316 manually that any value passed to this method will stringify to something the
317 RDBMS knows how to deal with. A notable example is the handling of L<DateTime>
318 objects, for more info see:
319 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
325 my $rs = $self->search_rs( @_ );
330 elsif (defined wantarray) {
334 # we can be called by a relationship helper, which in
335 # turn may be called in void context due to some braindead
336 # overload or whatever else the user decided to be clever
337 # at this particular day. Thus limit the exception to
338 # external code calls only
339 $self->throw_exception ('->search is *not* a mutator, calling it in void context makes no sense')
340 if (caller)[0] !~ /^\QDBIx::Class::/;
350 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
352 =item Return Value: L<$resultset|/search>
356 This method does the same exact thing as search() except it will
357 always return a resultset, even in list context.
364 my $rsrc = $self->result_source;
365 my ($call_cond, $call_attrs);
367 # Special-case handling for (undef, undef) or (undef)
368 # Note that (foo => undef) is valid deprecated syntax
369 @_ = () if not scalar grep { defined $_ } @_;
375 # fish out attrs in the ($condref, $attr) case
376 elsif (@_ == 2 and ( ! defined $_[0] or (ref $_[0]) ne '') ) {
377 ($call_cond, $call_attrs) = @_;
380 $self->throw_exception('Odd number of arguments to search')
384 carp_unique 'search( %condition ) is deprecated, use search( \%condition ) instead'
385 unless $rsrc->result_class->isa('DBIx::Class::CDBICompat');
387 for my $i (0 .. $#_) {
389 $self->throw_exception ('All keys in condition key/value pairs must be plain scalars')
390 if (! defined $_[$i] or ref $_[$i] ne '');
396 # see if we can keep the cache (no $rs changes)
398 my %safe = (alias => 1, cache => 1);
399 if ( ! List::Util::first { !$safe{$_} } keys %$call_attrs and (
402 ref $call_cond eq 'HASH' && ! keys %$call_cond
404 ref $call_cond eq 'ARRAY' && ! @$call_cond
406 $cache = $self->get_cache;
409 my $old_attrs = { %{$self->{attrs}} };
410 my $old_having = delete $old_attrs->{having};
411 my $old_where = delete $old_attrs->{where};
413 my $new_attrs = { %$old_attrs };
415 # take care of call attrs (only if anything is changing)
416 if ($call_attrs and keys %$call_attrs) {
418 # copy for _normalize_selection
419 $call_attrs = { %$call_attrs };
421 my @selector_attrs = qw/select as columns cols +select +as +columns include_columns/;
423 # reset the current selector list if new selectors are supplied
424 if (List::Util::first { exists $call_attrs->{$_} } qw/columns cols select as/) {
425 delete @{$old_attrs}{(@selector_attrs, '_dark_selector')};
428 # Normalize the new selector list (operates on the passed-in attr structure)
429 # Need to do it on every chain instead of only once on _resolved_attrs, in
430 # order to allow detection of empty vs partial 'as'
431 $call_attrs->{_dark_selector} = $old_attrs->{_dark_selector}
432 if $old_attrs->{_dark_selector};
433 $self->_normalize_selection ($call_attrs);
435 # start with blind overwriting merge, exclude selector attrs
436 $new_attrs = { %{$old_attrs}, %{$call_attrs} };
437 delete @{$new_attrs}{@selector_attrs};
439 for (@selector_attrs) {
440 $new_attrs->{$_} = $self->_merge_attr($old_attrs->{$_}, $call_attrs->{$_})
441 if ( exists $old_attrs->{$_} or exists $call_attrs->{$_} );
444 # older deprecated name, use only if {columns} is not there
445 if (my $c = delete $new_attrs->{cols}) {
446 if ($new_attrs->{columns}) {
447 carp "Resultset specifies both the 'columns' and the legacy 'cols' attributes - ignoring 'cols'";
450 $new_attrs->{columns} = $c;
455 # join/prefetch use their own crazy merging heuristics
456 foreach my $key (qw/join prefetch/) {
457 $new_attrs->{$key} = $self->_merge_joinpref_attr($old_attrs->{$key}, $call_attrs->{$key})
458 if exists $call_attrs->{$key};
461 # stack binds together
462 $new_attrs->{bind} = [ @{ $old_attrs->{bind} || [] }, @{ $call_attrs->{bind} || [] } ];
466 for ($old_where, $call_cond) {
468 $new_attrs->{where} = $self->_stack_cond (
469 $_, $new_attrs->{where}
474 if (defined $old_having) {
475 $new_attrs->{having} = $self->_stack_cond (
476 $old_having, $new_attrs->{having}
480 my $rs = (ref $self)->new($rsrc, $new_attrs);
482 $rs->set_cache($cache) if ($cache);
488 sub _normalize_selection {
489 my ($self, $attrs) = @_;
492 $attrs->{'+columns'} = $self->_merge_attr($attrs->{'+columns'}, delete $attrs->{include_columns})
493 if exists $attrs->{include_columns};
495 # columns are always placed first, however
497 # Keep the X vs +X separation until _resolved_attrs time - this allows to
498 # delay the decision on whether to use a default select list ($rsrc->columns)
499 # allowing stuff like the remove_columns helper to work
501 # select/as +select/+as pairs need special handling - the amount of select/as
502 # elements in each pair does *not* have to be equal (think multicolumn
503 # selectors like distinct(foo, bar) ). If the selector is bare (no 'as'
504 # supplied at all) - try to infer the alias, either from the -as parameter
505 # of the selector spec, or use the parameter whole if it looks like a column
506 # name (ugly legacy heuristic). If all fails - leave the selector bare (which
507 # is ok as well), but make sure no more additions to the 'as' chain take place
508 for my $pref ('', '+') {
510 my ($sel, $as) = map {
511 my $key = "${pref}${_}";
513 my $val = [ ref $attrs->{$key} eq 'ARRAY'
515 : $attrs->{$key} || ()
517 delete $attrs->{$key};
521 if (! @$as and ! @$sel ) {
524 elsif (@$as and ! @$sel) {
525 $self->throw_exception(
526 "Unable to handle ${pref}as specification (@$as) without a corresponding ${pref}select"
530 # no as part supplied at all - try to deduce (unless explicit end of named selection is declared)
531 # if any @$as has been supplied we assume the user knows what (s)he is doing
532 # and blindly keep stacking up pieces
533 unless ($attrs->{_dark_selector}) {
536 if ( ref $_ eq 'HASH' and exists $_->{-as} ) {
537 push @$as, $_->{-as};
539 # assume any plain no-space, no-parenthesis string to be a column spec
540 # FIXME - this is retarded but is necessary to support shit like 'count(foo)'
541 elsif ( ! ref $_ and $_ =~ /^ [^\s\(\)]+ $/x) {
544 # if all else fails - raise a flag that no more aliasing will be allowed
546 $attrs->{_dark_selector} = {
548 string => ($dark_sel_dumper ||= do {
549 require Data::Dumper::Concise;
550 Data::Dumper::Concise::DumperObject()->Indent(0);
551 })->Values([$_])->Dump
559 elsif (@$as < @$sel) {
560 $self->throw_exception(
561 "Unable to handle an ${pref}as specification (@$as) with less elements than the corresponding ${pref}select"
564 elsif ($pref and $attrs->{_dark_selector}) {
565 $self->throw_exception(
566 "Unable to process named '+select', resultset contains an unnamed selector $attrs->{_dark_selector}{string}"
572 $attrs->{"${pref}select"} = $self->_merge_attr($attrs->{"${pref}select"}, $sel);
573 $attrs->{"${pref}as"} = $self->_merge_attr($attrs->{"${pref}as"}, $as);
578 my ($self, $left, $right) = @_;
580 # collapse single element top-level conditions
581 # (single pass only, unlikely to need recursion)
582 for ($left, $right) {
583 if (ref $_ eq 'ARRAY') {
591 elsif (ref $_ eq 'HASH') {
592 my ($first, $more) = keys %$_;
595 if (! defined $first) {
599 elsif (! defined $more) {
600 if ($first eq '-and' and ref $_->{'-and'} eq 'HASH') {
603 elsif ($first eq '-or' and ref $_->{'-or'} eq 'ARRAY') {
610 # merge hashes with weeding out of duplicates (simple cases only)
611 if (ref $left eq 'HASH' and ref $right eq 'HASH') {
613 # shallow copy to destroy
614 $right = { %$right };
615 for (grep { exists $right->{$_} } keys %$left) {
616 # the use of eq_deeply here is justified - the rhs of an
617 # expression can contain a lot of twisted weird stuff
618 delete $right->{$_} if Data::Compare::Compare( $left->{$_}, $right->{$_} );
621 $right = undef unless keys %$right;
625 if (defined $left xor defined $right) {
626 return defined $left ? $left : $right;
628 elsif (! defined $left) {
632 return { -and => [ $left, $right ] };
636 =head2 search_literal
638 B<CAVEAT>: C<search_literal> is provided for Class::DBI compatibility and
639 should only be used in that context. C<search_literal> is a convenience
640 method. It is equivalent to calling C<< $schema->search(\[]) >>, but if you
641 want to ensure columns are bound correctly, use L</search>.
643 See L<DBIx::Class::Manual::Cookbook/Searching> and
644 L<DBIx::Class::Manual::FAQ/Searching> for searching techniques that do not
645 require C<search_literal>.
649 =item Arguments: $sql_fragment, @standalone_bind_values
651 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
655 my @cds = $cd_rs->search_literal('year = ? AND title = ?', qw/2001 Reload/);
656 my $newrs = $artist_rs->search_literal('name = ?', 'Metallica');
658 Pass a literal chunk of SQL to be added to the conditional part of the
661 Example of how to use C<search> instead of C<search_literal>
663 my @cds = $cd_rs->search_literal('cdid = ? AND (artist = ? OR artist = ?)', (2, 1, 2));
664 my @cds = $cd_rs->search(\[ 'cdid = ? AND (artist = ? OR artist = ?)', [ 'cdid', 2 ], [ 'artist', 1 ], [ 'artist', 2 ] ]);
669 my ($self, $sql, @bind) = @_;
671 if ( @bind && ref($bind[-1]) eq 'HASH' ) {
674 return $self->search(\[ $sql, map [ {} => $_ ], @bind ], ($attr || () ));
681 =item Arguments: \%columns_values | @pk_values, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
683 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
687 Finds and returns a single row based on supplied criteria. Takes either a
688 hashref with the same format as L</create> (including inference of foreign
689 keys from related objects), or a list of primary key values in the same
690 order as the L<primary columns|DBIx::Class::ResultSource/primary_columns>
691 declaration on the L</result_source>.
693 In either case an attempt is made to combine conditions already existing on
694 the resultset with the condition passed to this method.
696 To aid with preparing the correct query for the storage you may supply the
697 C<key> attribute, which is the name of a
698 L<unique constraint|DBIx::Class::ResultSource/add_unique_constraint> (the
699 unique constraint corresponding to the
700 L<primary columns|DBIx::Class::ResultSource/primary_columns> is always named
701 C<primary>). If the C<key> attribute has been supplied, and DBIC is unable
702 to construct a query that satisfies the named unique constraint fully (
703 non-NULL values for each column member of the constraint) an exception is
706 If no C<key> is specified, the search is carried over all unique constraints
707 which are fully defined by the available condition.
709 If no such constraint is found, C<find> currently defaults to a simple
710 C<< search->(\%column_values) >> which may or may not do what you expect.
711 Note that this fallback behavior may be deprecated in further versions. If
712 you need to search with arbitrary conditions - use L</search>. If the query
713 resulting from this fallback produces more than one row, a warning to the
714 effect is issued, though only the first row is constructed and returned as
717 In addition to C<key>, L</find> recognizes and applies standard
718 L<resultset attributes|/ATTRIBUTES> in the same way as L</search> does.
720 Note that if you have extra concerns about the correctness of the resulting
721 query you need to specify the C<key> attribute and supply the entire condition
722 as an argument to find (since it is not always possible to perform the
723 combination of the resultset condition with the supplied one, especially if
724 the resultset condition contains literal sql).
726 For example, to find a row by its primary key:
728 my $cd = $schema->resultset('CD')->find(5);
730 You can also find a row by a specific unique constraint:
732 my $cd = $schema->resultset('CD')->find(
734 artist => 'Massive Attack',
735 title => 'Mezzanine',
737 { key => 'cd_artist_title' }
740 See also L</find_or_create> and L</update_or_create>.
746 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
748 my $rsrc = $self->result_source;
751 if (exists $attrs->{key}) {
752 $constraint_name = defined $attrs->{key}
754 : $self->throw_exception("An undefined 'key' resultset attribute makes no sense")
758 # Parse out the condition from input
761 if (ref $_[0] eq 'HASH') {
762 $call_cond = { %{$_[0]} };
765 # if only values are supplied we need to default to 'primary'
766 $constraint_name = 'primary' unless defined $constraint_name;
768 my @c_cols = $rsrc->unique_constraint_columns($constraint_name);
770 $self->throw_exception(
771 "No constraint columns, maybe a malformed '$constraint_name' constraint?"
774 $self->throw_exception (
775 'find() expects either a column/value hashref, or a list of values '
776 . "corresponding to the columns of the specified unique constraint '$constraint_name'"
777 ) unless @c_cols == @_;
780 @{$call_cond}{@c_cols} = @_;
784 for my $key (keys %$call_cond) {
786 my $keyref = ref($call_cond->{$key})
788 my $relinfo = $rsrc->relationship_info($key)
790 my $val = delete $call_cond->{$key};
792 next if $keyref eq 'ARRAY'; # has_many for multi_create
794 my $rel_q = $rsrc->_resolve_condition(
795 $relinfo->{cond}, $val, $key, $key
797 die "Can't handle complex relationship conditions in find" if ref($rel_q) ne 'HASH';
798 @related{keys %$rel_q} = values %$rel_q;
802 # relationship conditions take precedence (?)
803 @{$call_cond}{keys %related} = values %related;
805 my $alias = exists $attrs->{alias} ? $attrs->{alias} : $self->{attrs}{alias};
807 if (defined $constraint_name) {
808 $final_cond = $self->_qualify_cond_columns (
810 $self->_build_unique_cond (
818 elsif ($self->{attrs}{accessor} and $self->{attrs}{accessor} eq 'single') {
819 # This means that we got here after a merger of relationship conditions
820 # in ::Relationship::Base::search_related (the row method), and furthermore
821 # the relationship is of the 'single' type. This means that the condition
822 # provided by the relationship (already attached to $self) is sufficient,
823 # as there can be only one row in the database that would satisfy the
827 # no key was specified - fall down to heuristics mode:
828 # run through all unique queries registered on the resultset, and
829 # 'OR' all qualifying queries together
830 my (@unique_queries, %seen_column_combinations);
831 for my $c_name ($rsrc->unique_constraint_names) {
832 next if $seen_column_combinations{
833 join "\x00", sort $rsrc->unique_constraint_columns($c_name)
836 push @unique_queries, try {
837 $self->_build_unique_cond ($c_name, $call_cond, 'croak_on_nulls')
841 $final_cond = @unique_queries
842 ? [ map { $self->_qualify_cond_columns($_, $alias) } @unique_queries ]
843 : $self->_non_unique_find_fallback ($call_cond, $attrs)
847 # Run the query, passing the result_class since it should propagate for find
848 my $rs = $self->search ($final_cond, {result_class => $self->result_class, %$attrs});
849 if ($rs->_resolved_attrs->{collapse}) {
851 carp "Query returned more than one row" if $rs->next;
859 # This is a stop-gap method as agreed during the discussion on find() cleanup:
860 # http://lists.scsys.co.uk/pipermail/dbix-class/2010-October/009535.html
862 # It is invoked when find() is called in legacy-mode with insufficiently-unique
863 # condition. It is provided for overrides until a saner way forward is devised
865 # *NOTE* This is not a public method, and it's *GUARANTEED* to disappear down
866 # the road. Please adjust your tests accordingly to catch this situation early
867 # DBIx::Class::ResultSet->can('_non_unique_find_fallback') is reasonable
869 # The method will not be removed without an adequately complete replacement
870 # for strict-mode enforcement
871 sub _non_unique_find_fallback {
872 my ($self, $cond, $attrs) = @_;
874 return $self->_qualify_cond_columns(
876 exists $attrs->{alias}
878 : $self->{attrs}{alias}
883 sub _qualify_cond_columns {
884 my ($self, $cond, $alias) = @_;
886 my %aliased = %$cond;
887 for (keys %aliased) {
888 $aliased{"$alias.$_"} = delete $aliased{$_}
895 sub _build_unique_cond {
896 my ($self, $constraint_name, $extra_cond, $croak_on_null) = @_;
898 my @c_cols = $self->result_source->unique_constraint_columns($constraint_name);
900 # combination may fail if $self->{cond} is non-trivial
901 my ($final_cond) = try {
902 $self->_merge_with_rscond ($extra_cond)
907 # trim out everything not in $columns
908 $final_cond = { map {
909 exists $final_cond->{$_}
910 ? ( $_ => $final_cond->{$_} )
914 if (my @missing = grep
915 { ! ($croak_on_null ? defined $final_cond->{$_} : exists $final_cond->{$_}) }
918 $self->throw_exception( sprintf ( "Unable to satisfy requested constraint '%s', no values for column(s): %s",
920 join (', ', map { "'$_'" } @missing),
927 !$ENV{DBIC_NULLABLE_KEY_NOWARN}
929 my @undefs = sort grep { ! defined $final_cond->{$_} } (keys %$final_cond)
931 carp_unique ( sprintf (
932 "NULL/undef values supplied for requested unique constraint '%s' (NULL "
933 . 'values in column(s): %s). This is almost certainly not what you wanted, '
934 . 'though you can set DBIC_NULLABLE_KEY_NOWARN to disable this warning.',
936 join (', ', map { "'$_'" } @undefs),
943 =head2 search_related
947 =item Arguments: $rel_name, $cond?, L<\%attrs?|/ATTRIBUTES>
949 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
953 $new_rs = $cd_rs->search_related('artist', {
957 Searches the specified relationship, optionally specifying a condition and
958 attributes for matching records. See L</ATTRIBUTES> for more information.
960 In list context, C<< ->all() >> is called implicitly on the resultset, thus
961 returning a list of result objects instead. To avoid that, use L</search_related_rs>.
963 See also L</search_related_rs>.
968 return shift->related_resultset(shift)->search(@_);
971 =head2 search_related_rs
973 This method works exactly the same as search_related, except that
974 it guarantees a resultset, even in list context.
978 sub search_related_rs {
979 return shift->related_resultset(shift)->search_rs(@_);
986 =item Arguments: none
988 =item Return Value: L<$cursor|DBIx::Class::Cursor>
992 Returns a storage-driven cursor to the given resultset. See
993 L<DBIx::Class::Cursor> for more information.
1000 return $self->{cursor} ||= do {
1001 my $attrs = { %{$self->_resolved_attrs } };
1002 $self->result_source->storage->select(
1003 $attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs
1012 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1014 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1018 my $cd = $schema->resultset('CD')->single({ year => 2001 });
1020 Inflates the first result without creating a cursor if the resultset has
1021 any records in it; if not returns C<undef>. Used by L</find> as a lean version
1024 While this method can take an optional search condition (just like L</search>)
1025 being a fast-code-path it does not recognize search attributes. If you need to
1026 add extra joins or similar, call L</search> and then chain-call L</single> on the
1027 L<DBIx::Class::ResultSet> returned.
1033 As of 0.08100, this method enforces the assumption that the preceding
1034 query returns only one row. If more than one row is returned, you will receive
1037 Query returned more than one row
1039 In this case, you should be using L</next> or L</find> instead, or if you really
1040 know what you are doing, use the L</rows> attribute to explicitly limit the size
1043 This method will also throw an exception if it is called on a resultset prefetching
1044 has_many, as such a prefetch implies fetching multiple rows from the database in
1045 order to assemble the resulting object.
1052 my ($self, $where) = @_;
1054 $self->throw_exception('single() only takes search conditions, no attributes. You want ->search( $cond, $attrs )->single()');
1057 my $attrs = { %{$self->_resolved_attrs} };
1059 $self->throw_exception(
1060 'single() can not be used on resultsets prefetching has_many. Use find( \%cond ) or next() instead'
1061 ) if $attrs->{collapse};
1064 if (defined $attrs->{where}) {
1067 [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ }
1068 $where, delete $attrs->{where} ]
1071 $attrs->{where} = $where;
1075 my $data = [ $self->result_source->storage->select_single(
1076 $attrs->{from}, $attrs->{select},
1077 $attrs->{where}, $attrs
1079 return undef unless @$data;
1080 $self->{_stashed_rows} = [ $data ];
1081 $self->_construct_results->[0];
1087 # Recursively collapse the query, accumulating values for each column.
1089 sub _collapse_query {
1090 my ($self, $query, $collapsed) = @_;
1094 if (ref $query eq 'ARRAY') {
1095 foreach my $subquery (@$query) {
1096 next unless ref $subquery; # -or
1097 $collapsed = $self->_collapse_query($subquery, $collapsed);
1100 elsif (ref $query eq 'HASH') {
1101 if (keys %$query and (keys %$query)[0] eq '-and') {
1102 foreach my $subquery (@{$query->{-and}}) {
1103 $collapsed = $self->_collapse_query($subquery, $collapsed);
1107 foreach my $col (keys %$query) {
1108 my $value = $query->{$col};
1109 $collapsed->{$col}{$value}++;
1121 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1123 =item Return Value: L<$resultsetcolumn|DBIx::Class::ResultSetColumn>
1127 my $max_length = $rs->get_column('length')->max;
1129 Returns a L<DBIx::Class::ResultSetColumn> instance for a column of the ResultSet.
1134 my ($self, $column) = @_;
1135 my $new = DBIx::Class::ResultSetColumn->new($self, $column);
1143 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1145 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1149 # WHERE title LIKE '%blue%'
1150 $cd_rs = $rs->search_like({ title => '%blue%'});
1152 Performs a search, but uses C<LIKE> instead of C<=> as the condition. Note
1153 that this is simply a convenience method retained for ex Class::DBI users.
1154 You most likely want to use L</search> with specific operators.
1156 For more information, see L<DBIx::Class::Manual::Cookbook>.
1158 This method is deprecated and will be removed in 0.09. Use L</search()>
1159 instead. An example conversion is:
1161 ->search_like({ foo => 'bar' });
1165 ->search({ foo => { like => 'bar' } });
1172 'search_like() is deprecated and will be removed in DBIC version 0.09.'
1173 .' Instead use ->search({ x => { -like => "y%" } })'
1174 .' (note the outer pair of {}s - they are important!)'
1176 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
1177 my $query = ref $_[0] eq 'HASH' ? { %{shift()} }: {@_};
1178 $query->{$_} = { 'like' => $query->{$_} } for keys %$query;
1179 return $class->search($query, { %$attrs });
1186 =item Arguments: $first, $last
1188 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1192 Returns a resultset or object list representing a subset of elements from the
1193 resultset slice is called on. Indexes are from 0, i.e., to get the first
1194 three records, call:
1196 my ($one, $two, $three) = $rs->slice(0, 2);
1201 my ($self, $min, $max) = @_;
1202 my $attrs = {}; # = { %{ $self->{attrs} || {} } };
1203 $attrs->{offset} = $self->{attrs}{offset} || 0;
1204 $attrs->{offset} += $min;
1205 $attrs->{rows} = ($max ? ($max - $min + 1) : 1);
1206 return $self->search(undef, $attrs);
1207 #my $slice = (ref $self)->new($self->result_source, $attrs);
1208 #return (wantarray ? $slice->all : $slice);
1215 =item Arguments: none
1217 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1221 Returns the next element in the resultset (C<undef> is there is none).
1223 Can be used to efficiently iterate over records in the resultset:
1225 my $rs = $schema->resultset('CD')->search;
1226 while (my $cd = $rs->next) {
1230 Note that you need to store the resultset object, and call C<next> on it.
1231 Calling C<< resultset('Table')->next >> repeatedly will always return the
1232 first record from the resultset.
1239 if (my $cache = $self->get_cache) {
1240 $self->{all_cache_position} ||= 0;
1241 return $cache->[$self->{all_cache_position}++];
1244 if ($self->{attrs}{cache}) {
1245 delete $self->{pager};
1246 $self->{all_cache_position} = 1;
1247 return ($self->all)[0];
1250 return shift(@{$self->{_stashed_results}}) if @{ $self->{_stashed_results}||[] };
1252 $self->{_stashed_results} = $self->_construct_results
1255 return shift @{$self->{_stashed_results}};
1258 # Constructs as many results as it can in one pass while respecting
1259 # cursor laziness. Several modes of operation:
1261 # * Always builds everything present in @{$self->{_stashed_rows}}
1262 # * If called with $fetch_all true - pulls everything off the cursor and
1263 # builds all result structures (or objects) in one pass
1264 # * If $self->_resolved_attrs->{collapse} is true, checks the order_by
1265 # and if the resultset is ordered properly by the left side:
1266 # * Fetches stuff off the cursor until the "master object" changes,
1267 # and saves the last extra row (if any) in @{$self->{_stashed_rows}}
1269 # * Just fetches, and collapses/constructs everything as if $fetch_all
1270 # was requested (there is no other way to collapse except for an
1272 # * If no collapse is requested - just get the next row, construct and
1274 sub _construct_results {
1275 my ($self, $fetch_all) = @_;
1277 my $rsrc = $self->result_source;
1278 my $attrs = $self->_resolved_attrs;
1283 ! $attrs->{order_by}
1287 my @pcols = $rsrc->primary_columns
1289 # default order for collapsing unless the user asked for something
1290 $attrs->{order_by} = [ map { join '.', $attrs->{alias}, $_} @pcols ];
1291 $attrs->{_ordered_for_collapse} = 1;
1292 $attrs->{_order_is_artificial} = 1;
1295 my $cursor = $self->cursor;
1297 # this will be used as both initial raw-row collector AND as a RV of
1298 # _construct_results. Not regrowing the array twice matters a lot...
1299 # a surprising amount actually
1300 my $rows = delete $self->{_stashed_rows};
1302 my $did_fetch_all = $fetch_all;
1305 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1306 $rows = [ ($rows ? @$rows : ()), $cursor->all ];
1308 elsif( $attrs->{collapse} ) {
1310 $attrs->{_ordered_for_collapse} = (!$attrs->{order_by}) ? 0 : do {
1311 my $st = $rsrc->schema->storage;
1314 ( $st->_extract_order_criteria($attrs->{order_by}) )
1317 my $colinfos = $st->_resolve_column_info($attrs->{from}, \@ord_cols);
1319 for (0 .. $#ord_cols) {
1321 ! $colinfos->{$ord_cols[$_]}
1323 $colinfos->{$ord_cols[$_]}{-result_source} != $rsrc
1325 splice @ord_cols, $_;
1330 # since all we check here are the start of the order_by belonging to the
1331 # top level $rsrc, a present identifying set will mean that the resultset
1332 # is ordered by its leftmost table in a tsable manner
1333 (@ord_cols and $rsrc->_identifying_column_set({ map
1334 { $colinfos->{$_}{-colname} => $colinfos->{$_} }
1337 } unless defined $attrs->{_ordered_for_collapse};
1339 if (! $attrs->{_ordered_for_collapse}) {
1342 # instead of looping over ->next, use ->all in stealth mode
1343 # *without* calling a ->reset afterwards
1344 # FIXME ENCAPSULATION - encapsulation breach, cursor method additions pending
1345 if (! $cursor->{_done}) {
1346 $rows = [ ($rows ? @$rows : ()), $cursor->all ];
1347 $cursor->{_done} = 1;
1352 if (! $did_fetch_all and ! @{$rows||[]} ) {
1353 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1354 if (scalar (my @r = $cursor->next) ) {
1359 return undef unless @{$rows||[]};
1361 my @extra_collapser_args;
1362 if ($attrs->{collapse} and ! $did_fetch_all ) {
1364 @extra_collapser_args = (
1365 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1366 sub { my @r = $cursor->next or return; \@r }, # how the collapser gets more rows
1367 ($self->{_stashed_rows} = []), # where does it stuff excess
1371 # hotspot - skip the setter
1372 my $res_class = $self->_result_class;
1374 my $inflator_cref = $self->{_result_inflator}{cref} ||= do {
1375 $res_class->can ('inflate_result')
1376 or $self->throw_exception("Inflator $res_class does not provide an inflate_result() method");
1379 my $infmap = $attrs->{as};
1382 $self->{_result_inflator}{is_core_row} = ( (
1385 ( \&DBIx::Class::Row::inflate_result || die "No ::Row::inflate_result() - can't happen" )
1386 ) ? 1 : 0 ) unless defined $self->{_result_inflator}{is_core_row};
1388 $self->{_result_inflator}{is_hri} = ( (
1389 ! $self->{_result_inflator}{is_core_row}
1392 require DBIx::Class::ResultClass::HashRefInflator
1394 DBIx::Class::ResultClass::HashRefInflator->can('inflate_result')
1396 ) ? 1 : 0 ) unless defined $self->{_result_inflator}{is_hri};
1399 if (! $attrs->{_related_results_construction}) {
1400 # construct a much simpler array->hash folder for the one-table cases right here
1401 if ($self->{_result_inflator}{is_hri}) {
1402 for my $r (@$rows) {
1403 $r = { map { $infmap->[$_] => $r->[$_] } 0..$#$infmap };
1406 # FIXME SUBOPTIMAL this is a very very very hot spot
1407 # while rather optimal we can *still* do much better, by
1408 # building a smarter Row::inflate_result(), and
1409 # switch to feeding it data via a much leaner interface
1411 # crude unscientific benchmarking indicated the shortcut eval is not worth it for
1412 # this particular resultset size
1413 elsif (@$rows < 60) {
1414 for my $r (@$rows) {
1415 $r = $inflator_cref->($res_class, $rsrc, { map { $infmap->[$_] => $r->[$_] } (0..$#$infmap) } );
1420 '$_ = $inflator_cref->($res_class, $rsrc, { %s }) for @$rows',
1421 join (', ', map { "\$infmap->[$_] => \$_->[$_]" } 0..$#$infmap )
1425 # Special-case multi-object HRI (we always prune, and there is no $inflator_cref pass)
1426 elsif ($self->{_result_inflator}{is_hri}) {
1427 ( $self->{_row_parser}{hri} ||= $rsrc->_mk_row_parser({
1429 inflate_map => $infmap,
1430 selection => $attrs->{select},
1431 collapse => $attrs->{collapse},
1432 premultiplied => $attrs->{_main_source_premultiplied},
1434 prune_null_branches => 1,
1435 }) )->($rows, @extra_collapser_args);
1437 # Regular multi-object
1439 my $parser_type = $self->{_result_inflator}{is_core_row} ? 'classic_pruning' : 'classic_nonpruning';
1441 ( $self->{_row_parser}{$parser_type} ||= $rsrc->_mk_row_parser({
1443 inflate_map => $infmap,
1444 selection => $attrs->{select},
1445 collapse => $attrs->{collapse},
1446 premultiplied => $attrs->{_main_source_premultiplied},
1447 prune_null_branches => $self->{_result_inflator}{is_core_row},
1448 }) )->($rows, @extra_collapser_args);
1450 $_ = $inflator_cref->($res_class, $rsrc, @$_) for @$rows;
1453 # The @$rows check seems odd at first - why wouldn't we want to warn
1454 # regardless? The issue is things like find() etc, where the user
1455 # *knows* only one result will come back. In these cases the ->all
1456 # is not a pessimization, but rather something we actually want
1458 'Unable to properly collapse has_many results in iterator mode due '
1459 . 'to order criteria - performed an eager cursor slurp underneath. '
1460 . 'Consider using ->all() instead'
1461 ) if ( ! $fetch_all and @$rows > 1 );
1466 =head2 result_source
1470 =item Arguments: L<$result_source?|DBIx::Class::ResultSource>
1472 =item Return Value: L<$result_source|DBIx::Class::ResultSource>
1476 An accessor for the primary ResultSource object from which this ResultSet
1483 =item Arguments: $result_class?
1485 =item Return Value: $result_class
1489 An accessor for the class to use when creating result objects. Defaults to
1490 C<< result_source->result_class >> - which in most cases is the name of the
1491 L<"table"|DBIx::Class::Manual::Glossary/"ResultSource"> class.
1493 Note that changing the result_class will also remove any components
1494 that were originally loaded in the source class via
1495 L<DBIx::Class::ResultSource/load_components>. Any overloaded methods
1496 in the original source class will not run.
1501 my ($self, $result_class) = @_;
1502 if ($result_class) {
1504 # don't fire this for an object
1505 $self->ensure_class_loaded($result_class)
1506 unless ref($result_class);
1508 if ($self->get_cache) {
1509 carp_unique('Changing the result_class of a ResultSet instance with cached results is a noop - the cache contents will not be altered');
1511 # FIXME ENCAPSULATION - encapsulation breach, cursor method additions pending
1512 elsif ($self->{cursor} && $self->{cursor}{_pos}) {
1513 $self->throw_exception('Changing the result_class of a ResultSet instance with an active cursor is not supported');
1516 $self->_result_class($result_class);
1518 delete $self->{_result_inflator};
1520 $self->_result_class;
1527 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1529 =item Return Value: $count
1533 Performs an SQL C<COUNT> with the same query as the resultset was built
1534 with to find the number of elements. Passing arguments is equivalent to
1535 C<< $rs->search ($cond, \%attrs)->count >>
1541 return $self->search(@_)->count if @_ and defined $_[0];
1542 return scalar @{ $self->get_cache } if $self->get_cache;
1544 my $attrs = { %{ $self->_resolved_attrs } };
1546 # this is a little optimization - it is faster to do the limit
1547 # adjustments in software, instead of a subquery
1548 my ($rows, $offset) = delete @{$attrs}{qw/rows offset/};
1551 if ($self->_has_resolved_attr (qw/collapse group_by/)) {
1552 $crs = $self->_count_subq_rs ($attrs);
1555 $crs = $self->_count_rs ($attrs);
1557 my $count = $crs->next;
1559 $count -= $offset if $offset;
1560 $count = $rows if $rows and $rows < $count;
1561 $count = 0 if ($count < 0);
1570 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1572 =item Return Value: L<$count_rs|DBIx::Class::ResultSetColumn>
1576 Same as L</count> but returns a L<DBIx::Class::ResultSetColumn> object.
1577 This can be very handy for subqueries:
1579 ->search( { amount => $some_rs->count_rs->as_query } )
1581 As with regular resultsets the SQL query will be executed only after
1582 the resultset is accessed via L</next> or L</all>. That would return
1583 the same single value obtainable via L</count>.
1589 return $self->search(@_)->count_rs if @_;
1591 # this may look like a lack of abstraction (count() does about the same)
1592 # but in fact an _rs *must* use a subquery for the limits, as the
1593 # software based limiting can not be ported if this $rs is to be used
1594 # in a subquery itself (i.e. ->as_query)
1595 if ($self->_has_resolved_attr (qw/collapse group_by offset rows/)) {
1596 return $self->_count_subq_rs;
1599 return $self->_count_rs;
1604 # returns a ResultSetColumn object tied to the count query
1607 my ($self, $attrs) = @_;
1609 my $rsrc = $self->result_source;
1610 $attrs ||= $self->_resolved_attrs;
1612 my $tmp_attrs = { %$attrs };
1613 # take off any limits, record_filter is cdbi, and no point of ordering nor locking a count
1614 delete @{$tmp_attrs}{qw/rows offset order_by _related_results_construction record_filter for/};
1616 # overwrite the selector (supplied by the storage)
1617 $tmp_attrs->{select} = $rsrc->storage->_count_select ($rsrc, $attrs);
1618 $tmp_attrs->{as} = 'count';
1620 my $tmp_rs = $rsrc->resultset_class->new($rsrc, $tmp_attrs)->get_column ('count');
1626 # same as above but uses a subquery
1628 sub _count_subq_rs {
1629 my ($self, $attrs) = @_;
1631 my $rsrc = $self->result_source;
1632 $attrs ||= $self->_resolved_attrs;
1634 my $sub_attrs = { %$attrs };
1635 # extra selectors do not go in the subquery and there is no point of ordering it, nor locking it
1636 delete @{$sub_attrs}{qw/collapse columns as select _related_results_construction order_by for/};
1638 # if we multi-prefetch we group_by something unique, as this is what we would
1639 # get out of the rs via ->next/->all. We *DO WANT* to clobber old group_by regardless
1640 if ( $attrs->{collapse} ) {
1641 $sub_attrs->{group_by} = [ map { "$attrs->{alias}.$_" } @{
1642 $rsrc->_identifying_column_set || $self->throw_exception(
1643 'Unable to construct a unique group_by criteria properly collapsing the '
1644 . 'has_many prefetch before count()'
1649 # Calculate subquery selector
1650 if (my $g = $sub_attrs->{group_by}) {
1652 my $sql_maker = $rsrc->storage->sql_maker;
1654 # necessary as the group_by may refer to aliased functions
1656 for my $sel (@{$attrs->{select}}) {
1657 $sel_index->{$sel->{-as}} = $sel
1658 if (ref $sel eq 'HASH' and $sel->{-as});
1661 # anything from the original select mentioned on the group-by needs to make it to the inner selector
1662 # also look for named aggregates referred in the having clause
1663 # having often contains scalarrefs - thus parse it out entirely
1665 if ($attrs->{having}) {
1666 local $sql_maker->{having_bind};
1667 local $sql_maker->{quote_char} = $sql_maker->{quote_char};
1668 local $sql_maker->{name_sep} = $sql_maker->{name_sep};
1669 unless (defined $sql_maker->{quote_char} and length $sql_maker->{quote_char}) {
1670 $sql_maker->{quote_char} = [ "\x00", "\xFF" ];
1671 # if we don't unset it we screw up retarded but unfortunately working
1672 # 'MAX(foo.bar)' => { '>', 3 }
1673 $sql_maker->{name_sep} = '';
1676 my ($lquote, $rquote, $sep) = map { quotemeta $_ } ($sql_maker->_quote_chars, $sql_maker->name_sep);
1678 my $having_sql = $sql_maker->_parse_rs_attrs ({ having => $attrs->{having} });
1681 # search for both a proper quoted qualified string, for a naive unquoted scalarref
1682 # and if all fails for an utterly naive quoted scalar-with-function
1683 while ($having_sql =~ /
1684 $rquote $sep $lquote (.+?) $rquote
1686 [\s,] \w+ \. (\w+) [\s,]
1688 [\s,] $lquote (.+?) $rquote [\s,]
1690 my $part = $1 || $2 || $3; # one of them matched if we got here
1691 unless ($seen_having{$part}++) {
1698 my $colpiece = $sel_index->{$_} || $_;
1700 # unqualify join-based group_by's. Arcane but possible query
1701 # also horrible horrible hack to alias a column (not a func.)
1702 # (probably need to introduce SQLA syntax)
1703 if ($colpiece =~ /\./ && $colpiece !~ /^$attrs->{alias}\./) {
1706 $colpiece = \ sprintf ('%s AS %s', map { $sql_maker->_quote ($_) } ($colpiece, $as) );
1708 push @{$sub_attrs->{select}}, $colpiece;
1712 my @pcols = map { "$attrs->{alias}.$_" } ($rsrc->primary_columns);
1713 $sub_attrs->{select} = @pcols ? \@pcols : [ 1 ];
1716 return $rsrc->resultset_class
1717 ->new ($rsrc, $sub_attrs)
1719 ->search ({}, { columns => { count => $rsrc->storage->_count_select ($rsrc, $attrs) } })
1720 ->get_column ('count');
1724 =head2 count_literal
1726 B<CAVEAT>: C<count_literal> is provided for Class::DBI compatibility and
1727 should only be used in that context. See L</search_literal> for further info.
1731 =item Arguments: $sql_fragment, @standalone_bind_values
1733 =item Return Value: $count
1737 Counts the results in a literal query. Equivalent to calling L</search_literal>
1738 with the passed arguments, then L</count>.
1742 sub count_literal { shift->search_literal(@_)->count; }
1748 =item Arguments: none
1750 =item Return Value: L<@result_objs|DBIx::Class::Manual::ResultClass>
1754 Returns all elements in the resultset.
1761 $self->throw_exception("all() doesn't take any arguments, you probably wanted ->search(...)->all()");
1764 delete @{$self}{qw/_stashed_rows _stashed_results/};
1766 if (my $c = $self->get_cache) {
1770 $self->cursor->reset;
1772 my $objs = $self->_construct_results('fetch_all') || [];
1774 $self->set_cache($objs) if $self->{attrs}{cache};
1783 =item Arguments: none
1785 =item Return Value: $self
1789 Resets the resultset's cursor, so you can iterate through the elements again.
1790 Implicitly resets the storage cursor, so a subsequent L</next> will trigger
1798 delete @{$self}{qw/_stashed_rows _stashed_results/};
1799 $self->{all_cache_position} = 0;
1800 $self->cursor->reset;
1808 =item Arguments: none
1810 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1814 L<Resets|/reset> the resultset (causing a fresh query to storage) and returns
1815 an object for the first result (or C<undef> if the resultset is empty).
1820 return $_[0]->reset->next;
1826 # Determines whether and what type of subquery is required for the $rs operation.
1827 # If grouping is necessary either supplies its own, or verifies the current one
1828 # After all is done delegates to the proper storage method.
1830 sub _rs_update_delete {
1831 my ($self, $op, $values) = @_;
1833 my $rsrc = $self->result_source;
1834 my $storage = $rsrc->schema->storage;
1836 my $attrs = { %{$self->_resolved_attrs} };
1838 my $join_classifications;
1839 my ($existing_group_by) = delete @{$attrs}{qw(group_by _grouped_by_distinct)};
1841 # do we need a subquery for any reason?
1843 defined $existing_group_by
1845 # if {from} is unparseable wrap a subq
1846 ref($attrs->{from}) ne 'ARRAY'
1848 # limits call for a subq
1849 $self->_has_resolved_attr(qw/rows offset/)
1852 # simplify the joinmap, so we can further decide if a subq is necessary
1853 if (!$needs_subq and @{$attrs->{from}} > 1) {
1854 $attrs->{from} = $storage->_prune_unused_joins ($attrs->{from}, $attrs->{select}, $self->{cond}, $attrs);
1856 # check if there are any joins left after the prune
1857 if ( @{$attrs->{from}} > 1 ) {
1858 $join_classifications = $storage->_resolve_aliastypes_from_select_args (
1859 [ @{$attrs->{from}}[1 .. $#{$attrs->{from}}] ],
1865 # any non-pruneable joins imply subq
1866 $needs_subq = scalar keys %{ $join_classifications->{restricting} || {} };
1870 # check if the head is composite (by now all joins are thrown out unless $needs_subq)
1872 (ref $attrs->{from}[0]) ne 'HASH'
1874 ref $attrs->{from}[0]{ $attrs->{from}[0]{-alias} }
1878 # do we need anything like a subquery?
1879 if (! $needs_subq) {
1880 # Most databases do not allow aliasing of tables in UPDATE/DELETE. Thus
1881 # a condition containing 'me' or other table prefixes will not work
1882 # at all. Tell SQLMaker to dequalify idents via a gross hack.
1884 my $sqla = $rsrc->storage->sql_maker;
1885 local $sqla->{_dequalify_idents} = 1;
1886 \[ $sqla->_recurse_where($self->{cond}) ];
1890 # we got this far - means it is time to wrap a subquery
1891 my $idcols = $rsrc->_identifying_column_set || $self->throw_exception(
1893 "Unable to perform complex resultset %s() without an identifying set of columns on source '%s'",
1899 # make a new $rs selecting only the PKs (that's all we really need for the subq)
1900 delete $attrs->{$_} for qw/select as collapse _related_results_construction/;
1901 $attrs->{columns} = [ map { "$attrs->{alias}.$_" } @$idcols ];
1902 $attrs->{group_by} = \ ''; # FIXME - this is an evil hack, it causes the optimiser to kick in and throw away the LEFT joins
1903 my $subrs = (ref $self)->new($rsrc, $attrs);
1905 if (@$idcols == 1) {
1906 $cond = { $idcols->[0] => { -in => $subrs->as_query } };
1908 elsif ($storage->_use_multicolumn_in) {
1909 # no syntax for calling this properly yet
1910 # !!! EXPERIMENTAL API !!! WILL CHANGE !!!
1911 $cond = $storage->sql_maker->_where_op_multicolumn_in (
1912 $idcols, # how do I convey a list of idents...? can binds reside on lhs?
1917 # if all else fails - get all primary keys and operate over a ORed set
1918 # wrap in a transaction for consistency
1919 # this is where the group_by/multiplication starts to matter
1923 keys %{ $join_classifications->{multiplying} || {} }
1925 # make sure if there is a supplied group_by it matches the columns compiled above
1926 # perfectly. Anything else can not be sanely executed on most databases so croak
1927 # right then and there
1928 if ($existing_group_by) {
1929 my @current_group_by = map
1930 { $_ =~ /\./ ? $_ : "$attrs->{alias}.$_" }
1935 join ("\x00", sort @current_group_by)
1937 join ("\x00", sort @{$attrs->{columns}} )
1939 $self->throw_exception (
1940 "You have just attempted a $op operation on a resultset which does group_by"
1941 . ' on columns other than the primary keys, while DBIC internally needs to retrieve'
1942 . ' the primary keys in a subselect. All sane RDBMS engines do not support this'
1943 . ' kind of queries. Please retry the operation with a modified group_by or'
1944 . ' without using one at all.'
1949 $subrs = $subrs->search({}, { group_by => $attrs->{columns} });
1952 $guard = $storage->txn_scope_guard;
1955 for my $row ($subrs->cursor->all) {
1957 { $idcols->[$_] => $row->[$_] }
1964 my $res = $storage->$op (
1966 $op eq 'update' ? $values : (),
1970 $guard->commit if $guard;
1979 =item Arguments: \%values
1981 =item Return Value: $underlying_storage_rv
1985 Sets the specified columns in the resultset to the supplied values in a
1986 single query. Note that this will not run any accessor/set_column/update
1987 triggers, nor will it update any result object instances derived from this
1988 resultset (this includes the contents of the L<resultset cache|/set_cache>
1989 if any). See L</update_all> if you need to execute any on-update
1990 triggers or cascades defined either by you or a
1991 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
1993 The return value is a pass through of what the underlying
1994 storage backend returned, and may vary. See L<DBI/execute> for the most
1999 Note that L</update> does not process/deflate any of the values passed in.
2000 This is unlike the corresponding L<DBIx::Class::Row/update>. The user must
2001 ensure manually that any value passed to this method will stringify to
2002 something the RDBMS knows how to deal with. A notable example is the
2003 handling of L<DateTime> objects, for more info see:
2004 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
2009 my ($self, $values) = @_;
2010 $self->throw_exception('Values for update must be a hash')
2011 unless ref $values eq 'HASH';
2013 return $self->_rs_update_delete ('update', $values);
2020 =item Arguments: \%values
2022 =item Return Value: 1
2026 Fetches all objects and updates them one at a time via
2027 L<DBIx::Class::Row/update>. Note that C<update_all> will run DBIC defined
2028 triggers, while L</update> will not.
2033 my ($self, $values) = @_;
2034 $self->throw_exception('Values for update_all must be a hash')
2035 unless ref $values eq 'HASH';
2037 my $guard = $self->result_source->schema->txn_scope_guard;
2038 $_->update({%$values}) for $self->all; # shallow copy - update will mangle it
2047 =item Arguments: none
2049 =item Return Value: $underlying_storage_rv
2053 Deletes the rows matching this resultset in a single query. Note that this
2054 will not run any delete triggers, nor will it alter the
2055 L<in_storage|DBIx::Class::Row/in_storage> status of any result object instances
2056 derived from this resultset (this includes the contents of the
2057 L<resultset cache|/set_cache> if any). See L</delete_all> if you need to
2058 execute any on-delete triggers or cascades defined either by you or a
2059 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
2061 The return value is a pass through of what the underlying storage backend
2062 returned, and may vary. See L<DBI/execute> for the most common case.
2068 $self->throw_exception('delete does not accept any arguments')
2071 return $self->_rs_update_delete ('delete');
2078 =item Arguments: none
2080 =item Return Value: 1
2084 Fetches all objects and deletes them one at a time via
2085 L<DBIx::Class::Row/delete>. Note that C<delete_all> will run DBIC defined
2086 triggers, while L</delete> will not.
2092 $self->throw_exception('delete_all does not accept any arguments')
2095 my $guard = $self->result_source->schema->txn_scope_guard;
2096 $_->delete for $self->all;
2105 =item Arguments: [ \@column_list, \@row_values+ ] | [ \%col_data+ ]
2107 =item Return Value: L<\@result_objects|DBIx::Class::Manual::ResultClass> (scalar context) | L<@result_objects|DBIx::Class::Manual::ResultClass> (list context)
2111 Accepts either an arrayref of hashrefs or alternatively an arrayref of
2118 The context of this method call has an important effect on what is
2119 submitted to storage. In void context data is fed directly to fastpath
2120 insertion routines provided by the underlying storage (most often
2121 L<DBI/execute_for_fetch>), bypassing the L<new|DBIx::Class::Row/new> and
2122 L<insert|DBIx::Class::Row/insert> calls on the
2123 L<Result|DBIx::Class::Manual::ResultClass> class, including any
2124 augmentation of these methods provided by components. For example if you
2125 are using something like L<DBIx::Class::UUIDColumns> to create primary
2126 keys for you, you will find that your PKs are empty. In this case you
2127 will have to explicitly force scalar or list context in order to create
2132 In non-void (scalar or list) context, this method is simply a wrapper
2133 for L</create>. Depending on list or scalar context either a list of
2134 L<Result|DBIx::Class::Manual::ResultClass> objects or an arrayref
2135 containing these objects is returned.
2137 When supplying data in "arrayref of arrayrefs" invocation style, the
2138 first element should be a list of column names and each subsequent
2139 element should be a data value in the earlier specified column order.
2142 $Arstist_rs->populate([
2143 [ qw( artistid name ) ],
2144 [ 100, 'A Formally Unknown Singer' ],
2145 [ 101, 'A singer that jumped the shark two albums ago' ],
2146 [ 102, 'An actually cool singer' ],
2149 For the arrayref of hashrefs style each hashref should be a structure
2150 suitable for passing to L</create>. Multi-create is also permitted with
2153 $schema->resultset("Artist")->populate([
2154 { artistid => 4, name => 'Manufactured Crap', cds => [
2155 { title => 'My First CD', year => 2006 },
2156 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2159 { artistid => 5, name => 'Angsty-Whiny Girl', cds => [
2160 { title => 'My parents sold me to a record company', year => 2005 },
2161 { title => 'Why Am I So Ugly?', year => 2006 },
2162 { title => 'I Got Surgery and am now Popular', year => 2007 }
2167 If you attempt a void-context multi-create as in the example above (each
2168 Artist also has the related list of CDs), and B<do not> supply the
2169 necessary autoinc foreign key information, this method will proxy to the
2170 less efficient L</create>, and then throw the Result objects away. In this
2171 case there are obviously no benefits to using this method over L</create>.
2178 # cruft placed in standalone method
2179 my $data = $self->_normalize_populate_args(@_);
2181 return unless @$data;
2183 if(defined wantarray) {
2184 my @created = map { $self->create($_) } @$data;
2185 return wantarray ? @created : \@created;
2188 my $first = $data->[0];
2190 # if a column is a registered relationship, and is a non-blessed hash/array, consider
2191 # it relationship data
2192 my (@rels, @columns);
2193 my $rsrc = $self->result_source;
2194 my $rels = { map { $_ => $rsrc->relationship_info($_) } $rsrc->relationships };
2195 for (keys %$first) {
2196 my $ref = ref $first->{$_};
2197 $rels->{$_} && ($ref eq 'ARRAY' or $ref eq 'HASH')
2203 my @pks = $rsrc->primary_columns;
2205 ## do the belongs_to relationships
2206 foreach my $index (0..$#$data) {
2208 # delegate to create() for any dataset without primary keys with specified relationships
2209 if (grep { !defined $data->[$index]->{$_} } @pks ) {
2211 if (grep { ref $data->[$index]{$r} eq $_ } qw/HASH ARRAY/) { # a related set must be a HASH or AoH
2212 my @ret = $self->populate($data);
2218 foreach my $rel (@rels) {
2219 next unless ref $data->[$index]->{$rel} eq "HASH";
2220 my $result = $self->related_resultset($rel)->create($data->[$index]->{$rel});
2221 my ($reverse_relname, $reverse_relinfo) = %{$rsrc->reverse_relationship_info($rel)};
2222 my $related = $result->result_source->_resolve_condition(
2223 $reverse_relinfo->{cond},
2229 delete $data->[$index]->{$rel};
2230 $data->[$index] = {%{$data->[$index]}, %$related};
2232 push @columns, keys %$related if $index == 0;
2236 ## inherit the data locked in the conditions of the resultset
2237 my ($rs_data) = $self->_merge_with_rscond({});
2238 delete @{$rs_data}{@columns};
2240 ## do bulk insert on current row
2241 $rsrc->storage->insert_bulk(
2243 [@columns, keys %$rs_data],
2244 [ map { [ @$_{@columns}, values %$rs_data ] } @$data ],
2247 ## do the has_many relationships
2248 foreach my $item (@$data) {
2252 foreach my $rel (@rels) {
2253 next unless ref $item->{$rel} eq "ARRAY" && @{ $item->{$rel} };
2255 $main_row ||= $self->new_result({map { $_ => $item->{$_} } @pks});
2257 my $child = $main_row->$rel;
2259 my $related = $child->result_source->_resolve_condition(
2260 $rels->{$rel}{cond},
2266 my @rows_to_add = ref $item->{$rel} eq 'ARRAY' ? @{$item->{$rel}} : ($item->{$rel});
2267 my @populate = map { {%$_, %$related} } @rows_to_add;
2269 $child->populate( \@populate );
2276 # populate() argumnets went over several incarnations
2277 # What we ultimately support is AoH
2278 sub _normalize_populate_args {
2279 my ($self, $arg) = @_;
2281 if (ref $arg eq 'ARRAY') {
2285 elsif (ref $arg->[0] eq 'HASH') {
2288 elsif (ref $arg->[0] eq 'ARRAY') {
2290 my @colnames = @{$arg->[0]};
2291 foreach my $values (@{$arg}[1 .. $#$arg]) {
2292 push @ret, { map { $colnames[$_] => $values->[$_] } (0 .. $#colnames) };
2298 $self->throw_exception('Populate expects an arrayref of hashrefs or arrayref of arrayrefs');
2305 =item Arguments: none
2307 =item Return Value: L<$pager|Data::Page>
2311 Returns a L<Data::Page> object for the current resultset. Only makes
2312 sense for queries with a C<page> attribute.
2314 To get the full count of entries for a paged resultset, call
2315 C<total_entries> on the L<Data::Page> object.
2322 return $self->{pager} if $self->{pager};
2324 my $attrs = $self->{attrs};
2325 if (!defined $attrs->{page}) {
2326 $self->throw_exception("Can't create pager for non-paged rs");
2328 elsif ($attrs->{page} <= 0) {
2329 $self->throw_exception('Invalid page number (page-numbers are 1-based)');
2331 $attrs->{rows} ||= 10;
2333 # throw away the paging flags and re-run the count (possibly
2334 # with a subselect) to get the real total count
2335 my $count_attrs = { %$attrs };
2336 delete @{$count_attrs}{qw/rows offset page pager/};
2338 my $total_rs = (ref $self)->new($self->result_source, $count_attrs);
2340 require DBIx::Class::ResultSet::Pager;
2341 return $self->{pager} = DBIx::Class::ResultSet::Pager->new(
2342 sub { $total_rs->count }, #lazy-get the total
2344 $self->{attrs}{page},
2352 =item Arguments: $page_number
2354 =item Return Value: L<$resultset|/search>
2358 Returns a resultset for the $page_number page of the resultset on which page
2359 is called, where each page contains a number of rows equal to the 'rows'
2360 attribute set on the resultset (10 by default).
2365 my ($self, $page) = @_;
2366 return (ref $self)->new($self->result_source, { %{$self->{attrs}}, page => $page });
2373 =item Arguments: \%col_data
2375 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2379 Creates a new result object in the resultset's result class and returns
2380 it. The row is not inserted into the database at this point, call
2381 L<DBIx::Class::Row/insert> to do that. Calling L<DBIx::Class::Row/in_storage>
2382 will tell you whether the result object has been inserted or not.
2384 Passes the hashref of input on to L<DBIx::Class::Row/new>.
2389 my ($self, $values) = @_;
2391 $self->throw_exception( "new_result takes only one argument - a hashref of values" )
2394 $self->throw_exception( "new_result expects a hashref" )
2395 unless (ref $values eq 'HASH');
2397 my ($merged_cond, $cols_from_relations) = $self->_merge_with_rscond($values);
2399 my $new = $self->result_class->new({
2401 ( @$cols_from_relations
2402 ? (-cols_from_relations => $cols_from_relations)
2405 -result_source => $self->result_source, # DO NOT REMOVE THIS, REQUIRED
2409 reftype($new) eq 'HASH'
2415 carp_unique (sprintf (
2416 "%s->new returned a blessed empty hashref - a strong indicator something is wrong with its inheritance chain",
2417 $self->result_class,
2424 # _merge_with_rscond
2426 # Takes a simple hash of K/V data and returns its copy merged with the
2427 # condition already present on the resultset. Additionally returns an
2428 # arrayref of value/condition names, which were inferred from related
2429 # objects (this is needed for in-memory related objects)
2430 sub _merge_with_rscond {
2431 my ($self, $data) = @_;
2433 my (%new_data, @cols_from_relations);
2435 my $alias = $self->{attrs}{alias};
2437 if (! defined $self->{cond}) {
2438 # just massage $data below
2440 elsif ($self->{cond} eq $DBIx::Class::ResultSource::UNRESOLVABLE_CONDITION) {
2441 %new_data = %{ $self->{attrs}{related_objects} || {} }; # nothing might have been inserted yet
2442 @cols_from_relations = keys %new_data;
2444 elsif (ref $self->{cond} ne 'HASH') {
2445 $self->throw_exception(
2446 "Can't abstract implicit construct, resultset condition not a hash"
2450 # precendence must be given to passed values over values inherited from
2451 # the cond, so the order here is important.
2452 my $collapsed_cond = $self->_collapse_cond($self->{cond});
2453 my %implied = %{$self->_remove_alias($collapsed_cond, $alias)};
2455 while ( my($col, $value) = each %implied ) {
2456 my $vref = ref $value;
2462 (keys %$value)[0] eq '='
2464 $new_data{$col} = $value->{'='};
2466 elsif( !$vref or $vref eq 'SCALAR' or blessed($value) ) {
2467 $new_data{$col} = $value;
2474 %{ $self->_remove_alias($data, $alias) },
2477 return (\%new_data, \@cols_from_relations);
2480 # _has_resolved_attr
2482 # determines if the resultset defines at least one
2483 # of the attributes supplied
2485 # used to determine if a subquery is neccessary
2487 # supports some virtual attributes:
2489 # This will scan for any joins being present on the resultset.
2490 # It is not a mere key-search but a deep inspection of {from}
2493 sub _has_resolved_attr {
2494 my ($self, @attr_names) = @_;
2496 my $attrs = $self->_resolved_attrs;
2500 for my $n (@attr_names) {
2501 if (grep { $n eq $_ } (qw/-join/) ) {
2502 $extra_checks{$n}++;
2506 my $attr = $attrs->{$n};
2508 next if not defined $attr;
2510 if (ref $attr eq 'HASH') {
2511 return 1 if keys %$attr;
2513 elsif (ref $attr eq 'ARRAY') {
2521 # a resolved join is expressed as a multi-level from
2523 $extra_checks{-join}
2525 ref $attrs->{from} eq 'ARRAY'
2527 @{$attrs->{from}} > 1
2535 # Recursively collapse the condition.
2537 sub _collapse_cond {
2538 my ($self, $cond, $collapsed) = @_;
2542 if (ref $cond eq 'ARRAY') {
2543 foreach my $subcond (@$cond) {
2544 next unless ref $subcond; # -or
2545 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2548 elsif (ref $cond eq 'HASH') {
2549 if (keys %$cond and (keys %$cond)[0] eq '-and') {
2550 foreach my $subcond (@{$cond->{-and}}) {
2551 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2555 foreach my $col (keys %$cond) {
2556 my $value = $cond->{$col};
2557 $collapsed->{$col} = $value;
2567 # Remove the specified alias from the specified query hash. A copy is made so
2568 # the original query is not modified.
2571 my ($self, $query, $alias) = @_;
2573 my %orig = %{ $query || {} };
2576 foreach my $key (keys %orig) {
2578 $unaliased{$key} = $orig{$key};
2581 $unaliased{$1} = $orig{$key}
2582 if $key =~ m/^(?:\Q$alias\E\.)?([^.]+)$/;
2592 =item Arguments: none
2594 =item Return Value: \[ $sql, L<@bind_values|/DBIC BIND VALUES> ]
2598 Returns the SQL query and bind vars associated with the invocant.
2600 This is generally used as the RHS for a subquery.
2607 my $attrs = { %{ $self->_resolved_attrs } };
2612 # my ($sql, \@bind, \%dbi_bind_attrs) = _select_args_to_query (...)
2613 # $sql also has no wrapping parenthesis in list ctx
2615 my $sqlbind = $self->result_source->storage
2616 ->_select_args_to_query ($attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs);
2625 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2627 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2631 my $artist = $schema->resultset('Artist')->find_or_new(
2632 { artist => 'fred' }, { key => 'artists' });
2634 $cd->cd_to_producer->find_or_new({ producer => $producer },
2635 { key => 'primary });
2637 Find an existing record from this resultset using L</find>. if none exists,
2638 instantiate a new result object and return it. The object will not be saved
2639 into your storage until you call L<DBIx::Class::Row/insert> on it.
2641 You most likely want this method when looking for existing rows using a unique
2642 constraint that is not the primary key, or looking for related rows.
2644 If you want objects to be saved immediately, use L</find_or_create> instead.
2646 B<Note>: Make sure to read the documentation of L</find> and understand the
2647 significance of the C<key> attribute, as its lack may skew your search, and
2648 subsequently result in spurious new objects.
2650 B<Note>: Take care when using C<find_or_new> with a table having
2651 columns with default values that you intend to be automatically
2652 supplied by the database (e.g. an auto_increment primary key column).
2653 In normal usage, the value of such columns should NOT be included at
2654 all in the call to C<find_or_new>, even when set to C<undef>.
2660 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2661 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2662 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2665 return $self->new_result($hash);
2672 =item Arguments: \%col_data
2674 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2678 Attempt to create a single new row or a row with multiple related rows
2679 in the table represented by the resultset (and related tables). This
2680 will not check for duplicate rows before inserting, use
2681 L</find_or_create> to do that.
2683 To create one row for this resultset, pass a hashref of key/value
2684 pairs representing the columns of the table and the values you wish to
2685 store. If the appropriate relationships are set up, foreign key fields
2686 can also be passed an object representing the foreign row, and the
2687 value will be set to its primary key.
2689 To create related objects, pass a hashref of related-object column values
2690 B<keyed on the relationship name>. If the relationship is of type C<multi>
2691 (L<DBIx::Class::Relationship/has_many>) - pass an arrayref of hashrefs.
2692 The process will correctly identify columns holding foreign keys, and will
2693 transparently populate them from the keys of the corresponding relation.
2694 This can be applied recursively, and will work correctly for a structure
2695 with an arbitrary depth and width, as long as the relationships actually
2696 exists and the correct column data has been supplied.
2698 Instead of hashrefs of plain related data (key/value pairs), you may
2699 also pass new or inserted objects. New objects (not inserted yet, see
2700 L</new_result>), will be inserted into their appropriate tables.
2702 Effectively a shortcut for C<< ->new_result(\%col_data)->insert >>.
2704 Example of creating a new row.
2706 $person_rs->create({
2707 name=>"Some Person",
2708 email=>"somebody@someplace.com"
2711 Example of creating a new row and also creating rows in a related C<has_many>
2712 or C<has_one> resultset. Note Arrayref.
2715 { artistid => 4, name => 'Manufactured Crap', cds => [
2716 { title => 'My First CD', year => 2006 },
2717 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2722 Example of creating a new row and also creating a row in a related
2723 C<belongs_to> resultset. Note Hashref.
2726 title=>"Music for Silly Walks",
2729 name=>"Silly Musician",
2737 When subclassing ResultSet never attempt to override this method. Since
2738 it is a simple shortcut for C<< $self->new_result($attrs)->insert >>, a
2739 lot of the internals simply never call it, so your override will be
2740 bypassed more often than not. Override either L<DBIx::Class::Row/new>
2741 or L<DBIx::Class::Row/insert> depending on how early in the
2742 L</create> process you need to intervene. See also warning pertaining to
2750 my ($self, $col_data) = @_;
2751 $self->throw_exception( "create needs a hashref" )
2752 unless ref $col_data eq 'HASH';
2753 return $self->new_result($col_data)->insert;
2756 =head2 find_or_create
2760 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2762 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2766 $cd->cd_to_producer->find_or_create({ producer => $producer },
2767 { key => 'primary' });
2769 Tries to find a record based on its primary key or unique constraints; if none
2770 is found, creates one and returns that instead.
2772 my $cd = $schema->resultset('CD')->find_or_create({
2774 artist => 'Massive Attack',
2775 title => 'Mezzanine',
2779 Also takes an optional C<key> attribute, to search by a specific key or unique
2780 constraint. For example:
2782 my $cd = $schema->resultset('CD')->find_or_create(
2784 artist => 'Massive Attack',
2785 title => 'Mezzanine',
2787 { key => 'cd_artist_title' }
2790 B<Note>: Make sure to read the documentation of L</find> and understand the
2791 significance of the C<key> attribute, as its lack may skew your search, and
2792 subsequently result in spurious row creation.
2794 B<Note>: Because find_or_create() reads from the database and then
2795 possibly inserts based on the result, this method is subject to a race
2796 condition. Another process could create a record in the table after
2797 the find has completed and before the create has started. To avoid
2798 this problem, use find_or_create() inside a transaction.
2800 B<Note>: Take care when using C<find_or_create> with a table having
2801 columns with default values that you intend to be automatically
2802 supplied by the database (e.g. an auto_increment primary key column).
2803 In normal usage, the value of such columns should NOT be included at
2804 all in the call to C<find_or_create>, even when set to C<undef>.
2806 See also L</find> and L</update_or_create>. For information on how to declare
2807 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2809 If you need to know if an existing row was found or a new one created use
2810 L</find_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
2811 to call L<DBIx::Class::Row/insert> to save the newly created row to the
2814 my $cd = $schema->resultset('CD')->find_or_new({
2816 artist => 'Massive Attack',
2817 title => 'Mezzanine',
2821 if( !$cd->in_storage ) {
2828 sub find_or_create {
2830 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2831 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2832 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2835 return $self->create($hash);
2838 =head2 update_or_create
2842 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2844 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2848 $resultset->update_or_create({ col => $val, ... });
2850 Like L</find_or_create>, but if a row is found it is immediately updated via
2851 C<< $found_row->update (\%col_data) >>.
2854 Takes an optional C<key> attribute to search on a specific unique constraint.
2857 # In your application
2858 my $cd = $schema->resultset('CD')->update_or_create(
2860 artist => 'Massive Attack',
2861 title => 'Mezzanine',
2864 { key => 'cd_artist_title' }
2867 $cd->cd_to_producer->update_or_create({
2868 producer => $producer,
2874 B<Note>: Make sure to read the documentation of L</find> and understand the
2875 significance of the C<key> attribute, as its lack may skew your search, and
2876 subsequently result in spurious row creation.
2878 B<Note>: Take care when using C<update_or_create> with a table having
2879 columns with default values that you intend to be automatically
2880 supplied by the database (e.g. an auto_increment primary key column).
2881 In normal usage, the value of such columns should NOT be included at
2882 all in the call to C<update_or_create>, even when set to C<undef>.
2884 See also L</find> and L</find_or_create>. For information on how to declare
2885 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2887 If you need to know if an existing row was updated or a new one created use
2888 L</update_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
2889 to call L<DBIx::Class::Row/insert> to save the newly created row to the
2894 sub update_or_create {
2896 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2897 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2899 my $row = $self->find($cond, $attrs);
2901 $row->update($cond);
2905 return $self->create($cond);
2908 =head2 update_or_new
2912 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2914 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2918 $resultset->update_or_new({ col => $val, ... });
2920 Like L</find_or_new> but if a row is found it is immediately updated via
2921 C<< $found_row->update (\%col_data) >>.
2925 # In your application
2926 my $cd = $schema->resultset('CD')->update_or_new(
2928 artist => 'Massive Attack',
2929 title => 'Mezzanine',
2932 { key => 'cd_artist_title' }
2935 if ($cd->in_storage) {
2936 # the cd was updated
2939 # the cd is not yet in the database, let's insert it
2943 B<Note>: Make sure to read the documentation of L</find> and understand the
2944 significance of the C<key> attribute, as its lack may skew your search, and
2945 subsequently result in spurious new objects.
2947 B<Note>: Take care when using C<update_or_new> with a table having
2948 columns with default values that you intend to be automatically
2949 supplied by the database (e.g. an auto_increment primary key column).
2950 In normal usage, the value of such columns should NOT be included at
2951 all in the call to C<update_or_new>, even when set to C<undef>.
2953 See also L</find>, L</find_or_create> and L</find_or_new>.
2959 my $attrs = ( @_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {} );
2960 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2962 my $row = $self->find( $cond, $attrs );
2963 if ( defined $row ) {
2964 $row->update($cond);
2968 return $self->new_result($cond);
2975 =item Arguments: none
2977 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass> | undef
2981 Gets the contents of the cache for the resultset, if the cache is set.
2983 The cache is populated either by using the L</prefetch> attribute to
2984 L</search> or by calling L</set_cache>.
2996 =item Arguments: L<\@result_objs|DBIx::Class::Manual::ResultClass>
2998 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass>
3002 Sets the contents of the cache for the resultset. Expects an arrayref
3003 of objects of the same class as those produced by the resultset. Note that
3004 if the cache is set, the resultset will return the cached objects rather
3005 than re-querying the database even if the cache attr is not set.
3007 The contents of the cache can also be populated by using the
3008 L</prefetch> attribute to L</search>.
3013 my ( $self, $data ) = @_;
3014 $self->throw_exception("set_cache requires an arrayref")
3015 if defined($data) && (ref $data ne 'ARRAY');
3016 $self->{all_cache} = $data;
3023 =item Arguments: none
3025 =item Return Value: undef
3029 Clears the cache for the resultset.
3034 shift->set_cache(undef);
3041 =item Arguments: none
3043 =item Return Value: true, if the resultset has been paginated
3051 return !!$self->{attrs}{page};
3058 =item Arguments: none
3060 =item Return Value: true, if the resultset has been ordered with C<order_by>.
3068 return scalar $self->result_source->storage->_extract_order_criteria($self->{attrs}{order_by});
3071 =head2 related_resultset
3075 =item Arguments: $rel_name
3077 =item Return Value: L<$resultset|/search>
3081 Returns a related resultset for the supplied relationship name.
3083 $artist_rs = $schema->resultset('CD')->related_resultset('Artist');
3087 sub related_resultset {
3088 my ($self, $rel) = @_;
3090 return $self->{related_resultsets}{$rel} ||= do {
3091 my $rsrc = $self->result_source;
3092 my $rel_info = $rsrc->relationship_info($rel);
3094 $self->throw_exception(
3095 "search_related: result source '" . $rsrc->source_name .
3096 "' has no such relationship $rel")
3099 my $attrs = $self->_chain_relationship($rel);
3101 my $join_count = $attrs->{seen_join}{$rel};
3103 my $alias = $self->result_source->storage
3104 ->relname_to_table_alias($rel, $join_count);
3106 # since this is search_related, and we already slid the select window inwards
3107 # (the select/as attrs were deleted in the beginning), we need to flip all
3108 # left joins to inner, so we get the expected results
3109 # read the comment on top of the actual function to see what this does
3110 $attrs->{from} = $rsrc->schema->storage->_inner_join_to_node ($attrs->{from}, $alias);
3113 #XXX - temp fix for result_class bug. There likely is a more elegant fix -groditi
3114 delete @{$attrs}{qw(result_class alias)};
3118 if (my $cache = $self->get_cache) {
3119 $related_cache = [ map
3120 { @{$_->related_resultset($rel)->get_cache||[]} }
3125 my $rel_source = $rsrc->related_source($rel);
3129 # The reason we do this now instead of passing the alias to the
3130 # search_rs below is that if you wrap/overload resultset on the
3131 # source you need to know what alias it's -going- to have for things
3132 # to work sanely (e.g. RestrictWithObject wants to be able to add
3133 # extra query restrictions, and these may need to be $alias.)
3135 my $rel_attrs = $rel_source->resultset_attributes;
3136 local $rel_attrs->{alias} = $alias;
3138 $rel_source->resultset
3142 where => $attrs->{where},
3145 $new->set_cache($related_cache) if $related_cache;
3150 =head2 current_source_alias
3154 =item Arguments: none
3156 =item Return Value: $source_alias
3160 Returns the current table alias for the result source this resultset is built
3161 on, that will be used in the SQL query. Usually it is C<me>.
3163 Currently the source alias that refers to the result set returned by a
3164 L</search>/L</find> family method depends on how you got to the resultset: it's
3165 C<me> by default, but eg. L</search_related> aliases it to the related result
3166 source name (and keeps C<me> referring to the original result set). The long
3167 term goal is to make L<DBIx::Class> always alias the current resultset as C<me>
3168 (and make this method unnecessary).
3170 Thus it's currently necessary to use this method in predefined queries (see
3171 L<DBIx::Class::Manual::Cookbook/Predefined searches>) when referring to the
3172 source alias of the current result set:
3174 # in a result set class
3176 my ($self, $user) = @_;
3178 my $me = $self->current_source_alias;
3180 return $self->search({
3181 "$me.modified" => $user->id,
3187 sub current_source_alias {
3188 return (shift->{attrs} || {})->{alias} || 'me';
3191 =head2 as_subselect_rs
3195 =item Arguments: none
3197 =item Return Value: L<$resultset|/search>
3201 Act as a barrier to SQL symbols. The resultset provided will be made into a
3202 "virtual view" by including it as a subquery within the from clause. From this
3203 point on, any joined tables are inaccessible to ->search on the resultset (as if
3204 it were simply where-filtered without joins). For example:
3206 my $rs = $schema->resultset('Bar')->search({'x.name' => 'abc'},{ join => 'x' });
3208 # 'x' now pollutes the query namespace
3210 # So the following works as expected
3211 my $ok_rs = $rs->search({'x.other' => 1});
3213 # But this doesn't: instead of finding a 'Bar' related to two x rows (abc and
3214 # def) we look for one row with contradictory terms and join in another table
3215 # (aliased 'x_2') which we never use
3216 my $broken_rs = $rs->search({'x.name' => 'def'});
3218 my $rs2 = $rs->as_subselect_rs;
3220 # doesn't work - 'x' is no longer accessible in $rs2, having been sealed away
3221 my $not_joined_rs = $rs2->search({'x.other' => 1});
3223 # works as expected: finds a 'table' row related to two x rows (abc and def)
3224 my $correctly_joined_rs = $rs2->search({'x.name' => 'def'});
3226 Another example of when one might use this would be to select a subset of
3227 columns in a group by clause:
3229 my $rs = $schema->resultset('Bar')->search(undef, {
3230 group_by => [qw{ id foo_id baz_id }],
3231 })->as_subselect_rs->search(undef, {
3232 columns => [qw{ id foo_id }]
3235 In the above example normally columns would have to be equal to the group by,
3236 but because we isolated the group by into a subselect the above works.
3240 sub as_subselect_rs {
3243 my $attrs = $self->_resolved_attrs;
3245 my $fresh_rs = (ref $self)->new (
3246 $self->result_source
3249 # these pieces will be locked in the subquery
3250 delete $fresh_rs->{cond};
3251 delete @{$fresh_rs->{attrs}}{qw/where bind/};
3253 return $fresh_rs->search( {}, {
3255 $attrs->{alias} => $self->as_query,
3256 -alias => $attrs->{alias},
3257 -rsrc => $self->result_source,
3259 alias => $attrs->{alias},
3263 # This code is called by search_related, and makes sure there
3264 # is clear separation between the joins before, during, and
3265 # after the relationship. This information is needed later
3266 # in order to properly resolve prefetch aliases (any alias
3267 # with a relation_chain_depth less than the depth of the
3268 # current prefetch is not considered)
3270 # The increments happen twice per join. An even number means a
3271 # relationship specified via a search_related, whereas an odd
3272 # number indicates a join/prefetch added via attributes
3274 # Also this code will wrap the current resultset (the one we
3275 # chain to) in a subselect IFF it contains limiting attributes
3276 sub _chain_relationship {
3277 my ($self, $rel) = @_;
3278 my $source = $self->result_source;
3279 my $attrs = { %{$self->{attrs}||{}} };
3281 # we need to take the prefetch the attrs into account before we
3282 # ->_resolve_join as otherwise they get lost - captainL
3283 my $join = $self->_merge_joinpref_attr( $attrs->{join}, $attrs->{prefetch} );
3285 delete @{$attrs}{qw/join prefetch collapse group_by distinct _grouped_by_distinct select as columns +select +as +columns/};
3287 my $seen = { %{ (delete $attrs->{seen_join}) || {} } };
3290 my @force_subq_attrs = qw/offset rows group_by having/;
3293 ($attrs->{from} && ref $attrs->{from} ne 'ARRAY')
3295 $self->_has_resolved_attr (@force_subq_attrs)
3297 # Nuke the prefetch (if any) before the new $rs attrs
3298 # are resolved (prefetch is useless - we are wrapping
3299 # a subquery anyway).
3300 my $rs_copy = $self->search;
3301 $rs_copy->{attrs}{join} = $self->_merge_joinpref_attr (
3302 $rs_copy->{attrs}{join},
3303 delete $rs_copy->{attrs}{prefetch},
3308 -alias => $attrs->{alias},
3309 $attrs->{alias} => $rs_copy->as_query,
3311 delete @{$attrs}{@force_subq_attrs, qw/where bind/};
3312 $seen->{-relation_chain_depth} = 0;
3314 elsif ($attrs->{from}) { #shallow copy suffices
3315 $from = [ @{$attrs->{from}} ];
3320 -alias => $attrs->{alias},
3321 $attrs->{alias} => $source->from,
3325 my $jpath = ($seen->{-relation_chain_depth})
3326 ? $from->[-1][0]{-join_path}
3329 my @requested_joins = $source->_resolve_join(
3336 push @$from, @requested_joins;
3338 $seen->{-relation_chain_depth}++;
3340 # if $self already had a join/prefetch specified on it, the requested
3341 # $rel might very well be already included. What we do in this case
3342 # is effectively a no-op (except that we bump up the chain_depth on
3343 # the join in question so we could tell it *is* the search_related)
3346 # we consider the last one thus reverse
3347 for my $j (reverse @requested_joins) {
3348 my ($last_j) = keys %{$j->[0]{-join_path}[-1]};
3349 if ($rel eq $last_j) {
3350 $j->[0]{-relation_chain_depth}++;
3356 unless ($already_joined) {
3357 push @$from, $source->_resolve_join(
3365 $seen->{-relation_chain_depth}++;
3367 return {%$attrs, from => $from, seen_join => $seen};
3370 sub _resolved_attrs {
3372 return $self->{_attrs} if $self->{_attrs};
3374 my $attrs = { %{ $self->{attrs} || {} } };
3375 my $source = $self->result_source;
3376 my $alias = $attrs->{alias};
3378 # default selection list
3379 $attrs->{columns} = [ $source->columns ]
3380 unless List::Util::first { exists $attrs->{$_} } qw/columns cols select as/;
3382 # merge selectors together
3383 for (qw/columns select as/) {
3384 $attrs->{$_} = $self->_merge_attr($attrs->{$_}, delete $attrs->{"+$_"})
3385 if $attrs->{$_} or $attrs->{"+$_"};
3388 # disassemble columns
3390 if (my $cols = delete $attrs->{columns}) {
3391 for my $c (ref $cols eq 'ARRAY' ? @$cols : $cols) {
3392 if (ref $c eq 'HASH') {
3393 for my $as (sort keys %$c) {
3394 push @sel, $c->{$as};
3405 # when trying to weed off duplicates later do not go past this point -
3406 # everything added from here on is unbalanced "anyone's guess" stuff
3407 my $dedup_stop_idx = $#as;
3409 push @as, @{ ref $attrs->{as} eq 'ARRAY' ? $attrs->{as} : [ $attrs->{as} ] }
3411 push @sel, @{ ref $attrs->{select} eq 'ARRAY' ? $attrs->{select} : [ $attrs->{select} ] }
3412 if $attrs->{select};
3414 # assume all unqualified selectors to apply to the current alias (legacy stuff)
3415 $_ = (ref $_ or $_ =~ /\./) ? $_ : "$alias.$_" for @sel;
3417 # disqualify all $alias.col as-bits (inflate-map mandated)
3418 $_ = ($_ =~ /^\Q$alias.\E(.+)$/) ? $1 : $_ for @as;
3420 # de-duplicate the result (remove *identical* select/as pairs)
3421 # and also die on duplicate {as} pointing to different {select}s
3422 # not using a c-style for as the condition is prone to shrinkage
3425 while ($i <= $dedup_stop_idx) {
3426 if ($seen->{"$sel[$i] \x00\x00 $as[$i]"}++) {
3431 elsif ($seen->{$as[$i]}++) {
3432 $self->throw_exception(
3433 "inflate_result() alias '$as[$i]' specified twice with different SQL-side {select}-ors"
3441 $attrs->{select} = \@sel;
3442 $attrs->{as} = \@as;
3444 $attrs->{from} ||= [{
3446 -alias => $self->{attrs}{alias},
3447 $self->{attrs}{alias} => $source->from,
3450 if ( $attrs->{join} || $attrs->{prefetch} ) {
3452 $self->throw_exception ('join/prefetch can not be used with a custom {from}')
3453 if ref $attrs->{from} ne 'ARRAY';
3455 my $join = (delete $attrs->{join}) || {};
3457 if ( defined $attrs->{prefetch} ) {
3458 $join = $self->_merge_joinpref_attr( $join, $attrs->{prefetch} );
3461 $attrs->{from} = # have to copy here to avoid corrupting the original
3463 @{ $attrs->{from} },
3464 $source->_resolve_join(
3467 { %{ $attrs->{seen_join} || {} } },
3468 ( $attrs->{seen_join} && keys %{$attrs->{seen_join}})
3469 ? $attrs->{from}[-1][0]{-join_path}
3476 if ( defined $attrs->{order_by} ) {
3477 $attrs->{order_by} = (
3478 ref( $attrs->{order_by} ) eq 'ARRAY'
3479 ? [ @{ $attrs->{order_by} } ]
3480 : [ $attrs->{order_by} || () ]
3484 if ($attrs->{group_by} and ref $attrs->{group_by} ne 'ARRAY') {
3485 $attrs->{group_by} = [ $attrs->{group_by} ];
3488 # generate the distinct induced group_by early, as prefetch will be carried via a
3489 # subquery (since a group_by is present)
3490 if (delete $attrs->{distinct}) {
3491 if ($attrs->{group_by}) {
3492 carp_unique ("Useless use of distinct on a grouped resultset ('distinct' is ignored when a 'group_by' is present)");
3495 $attrs->{_grouped_by_distinct} = 1;
3496 # distinct affects only the main selection part, not what prefetch may
3498 $attrs->{group_by} = $source->storage->_group_over_selection (
3506 # generate selections based on the prefetch helper
3508 $prefetch = $self->_merge_joinpref_attr( {}, delete $attrs->{prefetch} )
3509 if defined $attrs->{prefetch};
3513 $self->throw_exception("Unable to prefetch, resultset contains an unnamed selector $attrs->{_dark_selector}{string}")
3514 if $attrs->{_dark_selector};
3516 $attrs->{collapse} = 1;
3518 # this is a separate structure (we don't look in {from} directly)
3519 # as the resolver needs to shift things off the lists to work
3520 # properly (identical-prefetches on different branches)
3522 if (ref $attrs->{from} eq 'ARRAY') {
3524 my $start_depth = $attrs->{seen_join}{-relation_chain_depth} || 0;
3526 for my $j ( @{$attrs->{from}}[1 .. $#{$attrs->{from}} ] ) {
3527 next unless $j->[0]{-alias};
3528 next unless $j->[0]{-join_path};
3529 next if ($j->[0]{-relation_chain_depth} || 0) < $start_depth;
3531 my @jpath = map { keys %$_ } @{$j->[0]{-join_path}};
3534 $p = $p->{$_} ||= {} for @jpath[ ($start_depth/2) .. $#jpath]; #only even depths are actual jpath boundaries
3535 push @{$p->{-join_aliases} }, $j->[0]{-alias};
3539 my @prefetch = $source->_resolve_prefetch( $prefetch, $alias, $join_map );
3541 push @{ $attrs->{select} }, (map { $_->[0] } @prefetch);
3542 push @{ $attrs->{as} }, (map { $_->[1] } @prefetch);
3545 if ( List::Util::first { $_ =~ /\./ } @{$attrs->{as}} ) {
3546 $attrs->{_related_results_construction} = 1;
3549 $attrs->{collapse} = 0;
3552 # run through the resulting joinstructure (starting from our current slot)
3553 # and unset collapse if proven unnesessary
3555 # also while we are at it find out if the current root source has
3556 # been premultiplied by previous related_source chaining
3558 # this allows to predict whether a root object with all other relation
3559 # data set to NULL is in fact unique
3560 if ($attrs->{collapse}) {
3562 if (ref $attrs->{from} eq 'ARRAY') {
3564 if (@{$attrs->{from}} <= 1) {
3565 # no joins - no collapse
3566 $attrs->{collapse} = 0;
3569 # find where our table-spec starts
3570 my @fromlist = @{$attrs->{from}};
3572 my $t = shift @fromlist;
3575 # me vs join from-spec distinction - a ref means non-root
3576 if (ref $t eq 'ARRAY') {
3578 $is_multi ||= ! $t->{-is_single};
3580 last if ($t->{-alias} && $t->{-alias} eq $alias);
3581 $attrs->{_main_source_premultiplied} ||= $is_multi;
3584 # no non-singles remaining, nor any premultiplication - nothing to collapse
3586 ! $attrs->{_main_source_premultiplied}
3588 ! List::Util::first { ! $_->[0]{-is_single} } @fromlist
3590 $attrs->{collapse} = 0;
3596 # if we can not analyze the from - err on the side of safety
3597 $attrs->{_main_source_premultiplied} = 1;
3601 # if both page and offset are specified, produce a combined offset
3602 # even though it doesn't make much sense, this is what pre 081xx has
3604 if (my $page = delete $attrs->{page}) {
3606 ($attrs->{rows} * ($page - 1))
3608 ($attrs->{offset} || 0)
3612 return $self->{_attrs} = $attrs;
3616 my ($self, $attr) = @_;
3618 if (ref $attr eq 'HASH') {
3619 return $self->_rollout_hash($attr);
3620 } elsif (ref $attr eq 'ARRAY') {
3621 return $self->_rollout_array($attr);
3627 sub _rollout_array {
3628 my ($self, $attr) = @_;
3631 foreach my $element (@{$attr}) {
3632 if (ref $element eq 'HASH') {
3633 push( @rolled_array, @{ $self->_rollout_hash( $element ) } );
3634 } elsif (ref $element eq 'ARRAY') {
3635 # XXX - should probably recurse here
3636 push( @rolled_array, @{$self->_rollout_array($element)} );
3638 push( @rolled_array, $element );
3641 return \@rolled_array;
3645 my ($self, $attr) = @_;
3648 foreach my $key (keys %{$attr}) {
3649 push( @rolled_array, { $key => $attr->{$key} } );
3651 return \@rolled_array;
3654 sub _calculate_score {
3655 my ($self, $a, $b) = @_;
3657 if (defined $a xor defined $b) {
3660 elsif (not defined $a) {
3664 if (ref $b eq 'HASH') {
3665 my ($b_key) = keys %{$b};
3666 if (ref $a eq 'HASH') {
3667 my ($a_key) = keys %{$a};
3668 if ($a_key eq $b_key) {
3669 return (1 + $self->_calculate_score( $a->{$a_key}, $b->{$b_key} ));
3674 return ($a eq $b_key) ? 1 : 0;
3677 if (ref $a eq 'HASH') {
3678 my ($a_key) = keys %{$a};
3679 return ($b eq $a_key) ? 1 : 0;
3681 return ($b eq $a) ? 1 : 0;
3686 sub _merge_joinpref_attr {
3687 my ($self, $orig, $import) = @_;
3689 return $import unless defined($orig);
3690 return $orig unless defined($import);
3692 $orig = $self->_rollout_attr($orig);
3693 $import = $self->_rollout_attr($import);
3696 foreach my $import_element ( @{$import} ) {
3697 # find best candidate from $orig to merge $b_element into
3698 my $best_candidate = { position => undef, score => 0 }; my $position = 0;
3699 foreach my $orig_element ( @{$orig} ) {
3700 my $score = $self->_calculate_score( $orig_element, $import_element );
3701 if ($score > $best_candidate->{score}) {
3702 $best_candidate->{position} = $position;
3703 $best_candidate->{score} = $score;
3707 my ($import_key) = ( ref $import_element eq 'HASH' ) ? keys %{$import_element} : ($import_element);
3708 $import_key = '' if not defined $import_key;
3710 if ($best_candidate->{score} == 0 || exists $seen_keys->{$import_key}) {
3711 push( @{$orig}, $import_element );
3713 my $orig_best = $orig->[$best_candidate->{position}];
3714 # merge orig_best and b_element together and replace original with merged
3715 if (ref $orig_best ne 'HASH') {
3716 $orig->[$best_candidate->{position}] = $import_element;
3717 } elsif (ref $import_element eq 'HASH') {
3718 my ($key) = keys %{$orig_best};
3719 $orig->[$best_candidate->{position}] = { $key => $self->_merge_joinpref_attr($orig_best->{$key}, $import_element->{$key}) };
3722 $seen_keys->{$import_key} = 1; # don't merge the same key twice
3725 return @$orig ? $orig : ();
3733 require Hash::Merge;
3734 my $hm = Hash::Merge->new;
3736 $hm->specify_behavior({
3739 my ($defl, $defr) = map { defined $_ } (@_[0,1]);
3741 if ($defl xor $defr) {
3742 return [ $defl ? $_[0] : $_[1] ];
3747 elsif (__HM_DEDUP and $_[0] eq $_[1]) {
3751 return [$_[0], $_[1]];
3755 return $_[1] if !defined $_[0];
3756 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3757 return [$_[0], @{$_[1]}]
3760 return [] if !defined $_[0] and !keys %{$_[1]};
3761 return [ $_[1] ] if !defined $_[0];
3762 return [ $_[0] ] if !keys %{$_[1]};
3763 return [$_[0], $_[1]]
3768 return $_[0] if !defined $_[1];
3769 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3770 return [@{$_[0]}, $_[1]]
3773 my @ret = @{$_[0]} or return $_[1];
3774 return [ @ret, @{$_[1]} ] unless __HM_DEDUP;
3775 my %idx = map { $_ => 1 } @ret;
3776 push @ret, grep { ! defined $idx{$_} } (@{$_[1]});
3780 return [ $_[1] ] if ! @{$_[0]};
3781 return $_[0] if !keys %{$_[1]};
3782 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3783 return [ @{$_[0]}, $_[1] ];
3788 return [] if !keys %{$_[0]} and !defined $_[1];
3789 return [ $_[0] ] if !defined $_[1];
3790 return [ $_[1] ] if !keys %{$_[0]};
3791 return [$_[0], $_[1]]
3794 return [] if !keys %{$_[0]} and !@{$_[1]};
3795 return [ $_[0] ] if !@{$_[1]};
3796 return $_[1] if !keys %{$_[0]};
3797 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3798 return [ $_[0], @{$_[1]} ];
3801 return [] if !keys %{$_[0]} and !keys %{$_[1]};
3802 return [ $_[0] ] if !keys %{$_[1]};
3803 return [ $_[1] ] if !keys %{$_[0]};
3804 return [ $_[0] ] if $_[0] eq $_[1];
3805 return [ $_[0], $_[1] ];
3808 } => 'DBIC_RS_ATTR_MERGER');
3812 return $hm->merge ($_[1], $_[2]);
3816 sub STORABLE_freeze {
3817 my ($self, $cloning) = @_;
3818 my $to_serialize = { %$self };
3820 # A cursor in progress can't be serialized (and would make little sense anyway)
3821 # the parser can be regenerated (and can't be serialized)
3822 delete @{$to_serialize}{qw/cursor _row_parser _result_inflator/};
3824 # nor is it sensical to store a not-yet-fired-count pager
3825 if ($to_serialize->{pager} and ref $to_serialize->{pager}{total_entries} eq 'CODE') {
3826 delete $to_serialize->{pager};
3829 Storable::nfreeze($to_serialize);
3832 # need this hook for symmetry
3834 my ($self, $cloning, $serialized) = @_;
3836 %$self = %{ Storable::thaw($serialized) };
3842 =head2 throw_exception
3844 See L<DBIx::Class::Schema/throw_exception> for details.
3848 sub throw_exception {
3851 if (ref $self and my $rsrc = $self->result_source) {
3852 $rsrc->throw_exception(@_)
3855 DBIx::Class::Exception->throw(@_);
3863 # XXX: FIXME: Attributes docs need clearing up
3867 Attributes are used to refine a ResultSet in various ways when
3868 searching for data. They can be passed to any method which takes an
3869 C<\%attrs> argument. See L</search>, L</search_rs>, L</find>,
3872 Default attributes can be set on the result class using
3873 L<DBIx::Class::ResultSource/resultset_attributes>. (Please read
3874 the CAVEATS on that feature before using it!)
3876 These are in no particular order:
3882 =item Value: ( $order_by | \@order_by | \%order_by )
3886 Which column(s) to order the results by.
3888 [The full list of suitable values is documented in
3889 L<SQL::Abstract/"ORDER BY CLAUSES">; the following is a summary of
3892 If a single column name, or an arrayref of names is supplied, the
3893 argument is passed through directly to SQL. The hashref syntax allows
3894 for connection-agnostic specification of ordering direction:
3896 For descending order:
3898 order_by => { -desc => [qw/col1 col2 col3/] }
3900 For explicit ascending order:
3902 order_by => { -asc => 'col' }
3904 The old scalarref syntax (i.e. order_by => \'year DESC') is still
3905 supported, although you are strongly encouraged to use the hashref
3906 syntax as outlined above.
3912 =item Value: \@columns | \%columns | $column
3916 Shortcut to request a particular set of columns to be retrieved. Each
3917 column spec may be a string (a table column name), or a hash (in which
3918 case the key is the C<as> value, and the value is used as the C<select>
3919 expression). Adds C<me.> onto the start of any column without a C<.> in
3920 it and sets C<select> from that, then auto-populates C<as> from
3921 C<select> as normal. (You may also use the C<cols> attribute, as in
3922 earlier versions of DBIC.)
3924 Essentially C<columns> does the same as L</select> and L</as>.
3926 columns => [ 'foo', { bar => 'baz' } ]
3930 select => [qw/foo baz/],
3937 =item Value: \@columns
3941 Indicates additional columns to be selected from storage. Works the same
3942 as L</columns> but adds columns to the selection. (You may also use the
3943 C<include_columns> attribute, as in earlier versions of DBIC). For
3946 $schema->resultset('CD')->search(undef, {
3947 '+columns' => ['artist.name'],
3951 would return all CDs and include a 'name' column to the information
3952 passed to object inflation. Note that the 'artist' is the name of the
3953 column (or relationship) accessor, and 'name' is the name of the column
3954 accessor in the related table.
3956 B<NOTE:> You need to explicitly quote '+columns' when defining the attribute.
3957 Not doing so causes Perl to incorrectly interpret +columns as a bareword with a
3958 unary plus operator before it.
3960 =head2 include_columns
3964 =item Value: \@columns
3968 Deprecated. Acts as a synonym for L</+columns> for backward compatibility.
3974 =item Value: \@select_columns
3978 Indicates which columns should be selected from the storage. You can use
3979 column names, or in the case of RDBMS back ends, function or stored procedure
3982 $rs = $schema->resultset('Employee')->search(undef, {
3985 { count => 'employeeid' },
3986 { max => { length => 'name' }, -as => 'longest_name' }
3991 SELECT name, COUNT( employeeid ), MAX( LENGTH( name ) ) AS longest_name FROM employee
3993 B<NOTE:> You will almost always need a corresponding L</as> attribute when you
3994 use L</select>, to instruct DBIx::Class how to store the result of the column.
3995 Also note that the L</as> attribute has nothing to do with the SQL-side 'AS'
3996 identifier aliasing. You can however alias a function, so you can use it in
3997 e.g. an C<ORDER BY> clause. This is done via the C<-as> B<select function
3998 attribute> supplied as shown in the example above.
4000 B<NOTE:> You need to explicitly quote '+select'/'+as' when defining the attributes.
4001 Not doing so causes Perl to incorrectly interpret them as a bareword with a
4002 unary plus operator before it.
4008 Indicates additional columns to be selected from storage. Works the same as
4009 L</select> but adds columns to the default selection, instead of specifying
4018 =item Value: \@inflation_names
4022 Indicates column names for object inflation. That is L</as> indicates the
4023 slot name in which the column value will be stored within the
4024 L<Row|DBIx::Class::Row> object. The value will then be accessible via this
4025 identifier by the C<get_column> method (or via the object accessor B<if one
4026 with the same name already exists>) as shown below. The L</as> attribute has
4027 B<nothing to do> with the SQL-side C<AS>. See L</select> for details.
4029 $rs = $schema->resultset('Employee')->search(undef, {
4032 { count => 'employeeid' },
4033 { max => { length => 'name' }, -as => 'longest_name' }
4042 If the object against which the search is performed already has an accessor
4043 matching a column name specified in C<as>, the value can be retrieved using
4044 the accessor as normal:
4046 my $name = $employee->name();
4048 If on the other hand an accessor does not exist in the object, you need to
4049 use C<get_column> instead:
4051 my $employee_count = $employee->get_column('employee_count');
4053 You can create your own accessors if required - see
4054 L<DBIx::Class::Manual::Cookbook> for details.
4060 Indicates additional column names for those added via L</+select>. See L</as>.
4068 =item Value: ($rel_name | \@rel_names | \%rel_names)
4072 Contains a list of relationships that should be joined for this query. For
4075 # Get CDs by Nine Inch Nails
4076 my $rs = $schema->resultset('CD')->search(
4077 { 'artist.name' => 'Nine Inch Nails' },
4078 { join => 'artist' }
4081 Can also contain a hash reference to refer to the other relation's relations.
4084 package MyApp::Schema::Track;
4085 use base qw/DBIx::Class/;
4086 __PACKAGE__->table('track');
4087 __PACKAGE__->add_columns(qw/trackid cd position title/);
4088 __PACKAGE__->set_primary_key('trackid');
4089 __PACKAGE__->belongs_to(cd => 'MyApp::Schema::CD');
4092 # In your application
4093 my $rs = $schema->resultset('Artist')->search(
4094 { 'track.title' => 'Teardrop' },
4096 join => { cd => 'track' },
4097 order_by => 'artist.name',
4101 You need to use the relationship (not the table) name in conditions,
4102 because they are aliased as such. The current table is aliased as "me", so
4103 you need to use me.column_name in order to avoid ambiguity. For example:
4105 # Get CDs from 1984 with a 'Foo' track
4106 my $rs = $schema->resultset('CD')->search(
4109 'tracks.name' => 'Foo'
4111 { join => 'tracks' }
4114 If the same join is supplied twice, it will be aliased to <rel>_2 (and
4115 similarly for a third time). For e.g.
4117 my $rs = $schema->resultset('Artist')->search({
4118 'cds.title' => 'Down to Earth',
4119 'cds_2.title' => 'Popular',
4121 join => [ qw/cds cds/ ],
4124 will return a set of all artists that have both a cd with title 'Down
4125 to Earth' and a cd with title 'Popular'.
4127 If you want to fetch related objects from other tables as well, see L</prefetch>
4130 NOTE: An internal join-chain pruner will discard certain joins while
4131 constructing the actual SQL query, as long as the joins in question do not
4132 affect the retrieved result. This for example includes 1:1 left joins
4133 that are not part of the restriction specification (WHERE/HAVING) nor are
4134 a part of the query selection.
4136 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
4142 =item Value: (0 | 1)
4146 When set to a true value, indicates that any rows fetched from joined has_many
4147 relationships are to be aggregated into the corresponding "parent" object. For
4148 example, the resultset:
4150 my $rs = $schema->resultset('CD')->search({}, {
4151 '+columns' => [ qw/ tracks.title tracks.position / ],
4156 While executing the following query:
4158 SELECT me.*, tracks.title, tracks.position
4160 LEFT JOIN track tracks
4161 ON tracks.cdid = me.cdid
4163 Will return only as many objects as there are rows in the CD source, even
4164 though the result of the query may span many rows. Each of these CD objects
4165 will in turn have multiple "Track" objects hidden behind the has_many
4166 generated accessor C<tracks>. Without C<< collapse => 1 >>, the return values
4167 of this resultset would be as many CD objects as there are tracks (a "Cartesian
4168 product"), with each CD object containing exactly one of all fetched Track data.
4170 When a collapse is requested on a non-ordered resultset, an order by some
4171 unique part of the main source (the left-most table) is inserted automatically.
4172 This is done so that the resultset is allowed to be "lazy" - calling
4173 L<< $rs->next|/next >> will fetch only as many rows as it needs to build the next
4174 object with all of its related data.
4176 If an L</order_by> is already declared, and orders the resultset in a way that
4177 makes collapsing as described above impossible (e.g. C<< ORDER BY
4178 has_many_rel.column >> or C<ORDER BY RANDOM()>), DBIC will automatically
4179 switch to "eager" mode and slurp the entire resultset before consturcting the
4180 first object returned by L</next>.
4182 Setting this attribute on a resultset that does not join any has_many
4183 relations is a no-op.
4185 For a more in-depth discussion, see L</PREFETCHING>.
4191 =item Value: ($rel_name | \@rel_names | \%rel_names)
4195 This attribute is a shorthand for specifying a L</join> spec, adding all
4196 columns from the joined related sources as L</+columns> and setting
4197 L</collapse> to a true value. For example, the following two queries are
4200 my $rs = $schema->resultset('Artist')->search({}, {
4201 prefetch => { cds => ['genre', 'tracks' ] },
4206 my $rs = $schema->resultset('Artist')->search({}, {
4207 join => { cds => ['genre', 'tracks' ] },
4211 { +{ "cds.$_" => "cds.$_" } }
4212 $schema->source('Artist')->related_source('cds')->columns
4215 { +{ "cds.genre.$_" => "genre.$_" } }
4216 $schema->source('Artist')->related_source('cds')->related_source('genre')->columns
4219 { +{ "cds.tracks.$_" => "tracks.$_" } }
4220 $schema->source('Artist')->related_source('cds')->related_source('tracks')->columns
4225 Both producing the following SQL:
4227 SELECT me.artistid, me.name, me.rank, me.charfield,
4228 cds.cdid, cds.artist, cds.title, cds.year, cds.genreid, cds.single_track,
4229 genre.genreid, genre.name,
4230 tracks.trackid, tracks.cd, tracks.position, tracks.title, tracks.last_updated_on, tracks.last_updated_at
4233 ON cds.artist = me.artistid
4234 LEFT JOIN genre genre
4235 ON genre.genreid = cds.genreid
4236 LEFT JOIN track tracks
4237 ON tracks.cd = cds.cdid
4238 ORDER BY me.artistid
4240 While L</prefetch> implies a L</join>, it is ok to mix the two together, as
4241 the arguments are properly merged and generally do the right thing. For
4242 example, you may want to do the following:
4244 my $artists_and_cds_without_genre = $schema->resultset('Artist')->search(
4245 { 'genre.genreid' => undef },
4247 join => { cds => 'genre' },
4252 Which generates the following SQL:
4254 SELECT me.artistid, me.name, me.rank, me.charfield,
4255 cds.cdid, cds.artist, cds.title, cds.year, cds.genreid, cds.single_track
4258 ON cds.artist = me.artistid
4259 LEFT JOIN genre genre
4260 ON genre.genreid = cds.genreid
4261 WHERE genre.genreid IS NULL
4262 ORDER BY me.artistid
4264 For a more in-depth discussion, see L</PREFETCHING>.
4270 =item Value: $source_alias
4274 Sets the source alias for the query. Normally, this defaults to C<me>, but
4275 nested search queries (sub-SELECTs) might need specific aliases set to
4276 reference inner queries. For example:
4279 ->related_resultset('CDs')
4280 ->related_resultset('Tracks')
4282 'track.id' => { -ident => 'none_search.id' },
4286 my $ids = $self->search({
4289 alias => 'none_search',
4290 group_by => 'none_search.id',
4291 })->get_column('id')->as_query;
4293 $self->search({ id => { -in => $ids } })
4295 This attribute is directly tied to L</current_source_alias>.
4305 Makes the resultset paged and specifies the page to retrieve. Effectively
4306 identical to creating a non-pages resultset and then calling ->page($page)
4309 If L</rows> attribute is not specified it defaults to 10 rows per page.
4311 When you have a paged resultset, L</count> will only return the number
4312 of rows in the page. To get the total, use the L</pager> and call
4313 C<total_entries> on it.
4323 Specifies the maximum number of rows for direct retrieval or the number of
4324 rows per page if the page attribute or method is used.
4330 =item Value: $offset
4334 Specifies the (zero-based) row number for the first row to be returned, or the
4335 of the first row of the first page if paging is used.
4337 =head2 software_limit
4341 =item Value: (0 | 1)
4345 When combined with L</rows> and/or L</offset> the generated SQL will not
4346 include any limit dialect stanzas. Instead the entire result will be selected
4347 as if no limits were specified, and DBIC will perform the limit locally, by
4348 artificially advancing and finishing the resulting L</cursor>.
4350 This is the recommended way of performing resultset limiting when no sane RDBMS
4351 implementation is available (e.g.
4352 L<Sybase ASE|DBIx::Class::Storage::DBI::Sybase::ASE> using the
4353 L<Generic Sub Query|DBIx::Class::SQLMaker::LimitDialects/GenericSubQ> hack)
4359 =item Value: \@columns
4363 A arrayref of columns to group by. Can include columns of joined tables.
4365 group_by => [qw/ column1 column2 ... /]
4371 =item Value: $condition
4375 HAVING is a select statement attribute that is applied between GROUP BY and
4376 ORDER BY. It is applied to the after the grouping calculations have been
4379 having => { 'count_employee' => { '>=', 100 } }
4381 or with an in-place function in which case literal SQL is required:
4383 having => \[ 'count(employee) >= ?', [ count => 100 ] ]
4389 =item Value: (0 | 1)
4393 Set to 1 to group by all columns. If the resultset already has a group_by
4394 attribute, this setting is ignored and an appropriate warning is issued.
4400 Adds to the WHERE clause.
4402 # only return rows WHERE deleted IS NULL for all searches
4403 __PACKAGE__->resultset_attributes({ where => { deleted => undef } });
4405 Can be overridden by passing C<< { where => undef } >> as an attribute
4408 For more complicated where clauses see L<SQL::Abstract/WHERE CLAUSES>.
4414 Set to 1 to cache search results. This prevents extra SQL queries if you
4415 revisit rows in your ResultSet:
4417 my $resultset = $schema->resultset('Artist')->search( undef, { cache => 1 } );
4419 while( my $artist = $resultset->next ) {
4423 $rs->first; # without cache, this would issue a query
4425 By default, searches are not cached.
4427 For more examples of using these attributes, see
4428 L<DBIx::Class::Manual::Cookbook>.
4434 =item Value: ( 'update' | 'shared' | \$scalar )
4438 Set to 'update' for a SELECT ... FOR UPDATE or 'shared' for a SELECT
4439 ... FOR SHARED. If \$scalar is passed, this is taken directly and embedded in the
4444 DBIx::Class supports arbitrary related data prefetching from multiple related
4445 sources. Any combination of relationship types and column sets are supported.
4446 If L<collapsing|/collapse> is requested, there is an additional requirement of
4447 selecting enough data to make every individual object uniquely identifiable.
4449 Here are some more involved examples, based on the following relationship map:
4452 My::Schema::CD->belongs_to( artist => 'My::Schema::Artist' );
4453 My::Schema::CD->might_have( liner_note => 'My::Schema::LinerNotes' );
4454 My::Schema::CD->has_many( tracks => 'My::Schema::Track' );
4456 My::Schema::Artist->belongs_to( record_label => 'My::Schema::RecordLabel' );
4458 My::Schema::Track->has_many( guests => 'My::Schema::Guest' );
4462 my $rs = $schema->resultset('Tag')->search(
4471 The initial search results in SQL like the following:
4473 SELECT tag.*, cd.*, artist.* FROM tag
4474 JOIN cd ON tag.cd = cd.cdid
4475 JOIN artist ON cd.artist = artist.artistid
4477 L<DBIx::Class> has no need to go back to the database when we access the
4478 C<cd> or C<artist> relationships, which saves us two SQL statements in this
4481 Simple prefetches will be joined automatically, so there is no need
4482 for a C<join> attribute in the above search.
4484 The L</prefetch> attribute can be used with any of the relationship types
4485 and multiple prefetches can be specified together. Below is a more complex
4486 example that prefetches a CD's artist, its liner notes (if present),
4487 the cover image, the tracks on that CD, and the guests on those
4490 my $rs = $schema->resultset('CD')->search(
4494 { artist => 'record_label'}, # belongs_to => belongs_to
4495 'liner_note', # might_have
4496 'cover_image', # has_one
4497 { tracks => 'guests' }, # has_many => has_many
4502 This will produce SQL like the following:
4504 SELECT cd.*, artist.*, record_label.*, liner_note.*, cover_image.*,
4508 ON artist.artistid = me.artistid
4509 JOIN record_label record_label
4510 ON record_label.labelid = artist.labelid
4511 LEFT JOIN track tracks
4512 ON tracks.cdid = me.cdid
4513 LEFT JOIN guest guests
4514 ON guests.trackid = track.trackid
4515 LEFT JOIN liner_notes liner_note
4516 ON liner_note.cdid = me.cdid
4517 JOIN cd_artwork cover_image
4518 ON cover_image.cdid = me.cdid
4521 Now the C<artist>, C<record_label>, C<liner_note>, C<cover_image>,
4522 C<tracks>, and C<guests> of the CD will all be available through the
4523 relationship accessors without the need for additional queries to the
4528 Prefetch does a lot of deep magic. As such, it may not behave exactly
4529 as you might expect.
4535 Prefetch uses the L</cache> to populate the prefetched relationships. This
4536 may or may not be what you want.
4540 If you specify a condition on a prefetched relationship, ONLY those
4541 rows that match the prefetched condition will be fetched into that relationship.
4542 This means that adding prefetch to a search() B<may alter> what is returned by
4543 traversing a relationship. So, if you have C<< Artist->has_many(CDs) >> and you do
4545 my $artist_rs = $schema->resultset('Artist')->search({
4551 my $count = $artist_rs->first->cds->count;
4553 my $artist_rs_prefetch = $artist_rs->search( {}, { prefetch => 'cds' } );
4555 my $prefetch_count = $artist_rs_prefetch->first->cds->count;
4557 cmp_ok( $count, '==', $prefetch_count, "Counts should be the same" );
4559 That cmp_ok() may or may not pass depending on the datasets involved. In other
4560 words the C<WHERE> condition would apply to the entire dataset, just like
4561 it would in regular SQL. If you want to add a condition only to the "right side"
4562 of a C<LEFT JOIN> - consider declaring and using a L<relationship with a custom
4563 condition|DBIx::Class::Relationship::Base/condition>
4567 =head1 DBIC BIND VALUES
4569 Because DBIC may need more information to bind values than just the column name
4570 and value itself, it uses a special format for both passing and receiving bind
4571 values. Each bind value should be composed of an arrayref of
4572 C<< [ \%args => $val ] >>. The format of C<< \%args >> is currently:
4578 If present (in any form), this is what is being passed directly to bind_param.
4579 Note that different DBD's expect different bind args. (e.g. DBD::SQLite takes
4580 a single numerical type, while DBD::Pg takes a hashref if bind options.)
4582 If this is specified, all other bind options described below are ignored.
4586 If present, this is used to infer the actual bind attribute by passing to
4587 C<< $resolved_storage->bind_attribute_by_data_type() >>. Defaults to the
4588 "data_type" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>.
4590 Note that the data type is somewhat freeform (hence the sqlt_ prefix);
4591 currently drivers are expected to "Do the Right Thing" when given a common
4592 datatype name. (Not ideal, but that's what we got at this point.)
4596 Currently used to correctly allocate buffers for bind_param_inout().
4597 Defaults to "size" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>,
4598 or to a sensible value based on the "data_type".
4602 Used to fill in missing sqlt_datatype and sqlt_size attributes (if they are
4603 explicitly specified they are never overriden). Also used by some weird DBDs,
4604 where the column name should be available at bind_param time (e.g. Oracle).
4608 For backwards compatibility and convenience, the following shortcuts are
4611 [ $name => $val ] === [ { dbic_colname => $name }, $val ]
4612 [ \$dt => $val ] === [ { sqlt_datatype => $dt }, $val ]
4613 [ undef, $val ] === [ {}, $val ]
4615 =head1 AUTHOR AND CONTRIBUTORS
4617 See L<AUTHOR|DBIx::Class/AUTHOR> and L<CONTRIBUTORS|DBIx::Class/CONTRIBUTORS> in DBIx::Class
4621 You may distribute this code under the same terms as Perl itself.