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 (e.g.:
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');
248 $attrs = { %{$attrs||{}} };
249 delete @{$attrs}{qw(_sqlmaker_select_args _related_results_construction)};
251 if ($attrs->{page}) {
252 $attrs->{rows} ||= 10;
255 $attrs->{alias} ||= 'me';
258 result_source => $source,
259 cond => $attrs->{where},
264 # if there is a dark selector, this means we are already in a
265 # chain and the cleanup/sanification was taken care of by
267 $self->_normalize_selection($attrs)
268 unless $attrs->{_dark_selector};
271 $attrs->{result_class} || $source->result_class
281 =item Arguments: L<$cond|DBIx::Class::SQLMaker> | undef, L<\%attrs?|/ATTRIBUTES>
283 =item Return Value: $resultset (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
287 my @cds = $cd_rs->search({ year => 2001 }); # "... WHERE year = 2001"
288 my $new_rs = $cd_rs->search({ year => 2005 });
290 my $new_rs = $cd_rs->search([ { year => 2005 }, { year => 2004 } ]);
291 # year = 2005 OR year = 2004
293 In list context, C<< ->all() >> is called implicitly on the resultset, thus
294 returning a list of L<result|DBIx::Class::Manual::ResultClass> objects instead.
295 To avoid that, use L</search_rs>.
297 If you need to pass in additional attributes but no additional condition,
298 call it as C<search(undef, \%attrs)>.
300 # "SELECT name, artistid FROM $artist_table"
301 my @all_artists = $schema->resultset('Artist')->search(undef, {
302 columns => [qw/name artistid/],
305 For a list of attributes that can be passed to C<search>, see
306 L</ATTRIBUTES>. For more examples of using this function, see
307 L<Searching|DBIx::Class::Manual::Cookbook/Searching>. For a complete
308 documentation for the first argument, see L<SQL::Abstract/"WHERE CLAUSES">
309 and its extension L<DBIx::Class::SQLMaker>.
311 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
315 Note that L</search> does not process/deflate any of the values passed in the
316 L<SQL::Abstract>-compatible search condition structure. This is unlike other
317 condition-bound methods L</new_result>, L</create> and L</find>. The user must ensure
318 manually that any value passed to this method will stringify to something the
319 RDBMS knows how to deal with. A notable example is the handling of L<DateTime>
320 objects, for more info see:
321 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
327 my $rs = $self->search_rs( @_ );
332 elsif (defined wantarray) {
336 # we can be called by a relationship helper, which in
337 # turn may be called in void context due to some braindead
338 # overload or whatever else the user decided to be clever
339 # at this particular day. Thus limit the exception to
340 # external code calls only
341 $self->throw_exception ('->search is *not* a mutator, calling it in void context makes no sense')
342 if (caller)[0] !~ /^\QDBIx::Class::/;
352 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
354 =item Return Value: L<$resultset|/search>
358 This method does the same exact thing as search() except it will
359 always return a resultset, even in list context.
366 my $rsrc = $self->result_source;
367 my ($call_cond, $call_attrs);
369 # Special-case handling for (undef, undef) or (undef)
370 # Note that (foo => undef) is valid deprecated syntax
371 @_ = () if not scalar grep { defined $_ } @_;
377 # fish out attrs in the ($condref, $attr) case
378 elsif (@_ == 2 and ( ! defined $_[0] or (ref $_[0]) ne '') ) {
379 ($call_cond, $call_attrs) = @_;
382 $self->throw_exception('Odd number of arguments to search')
386 carp_unique 'search( %condition ) is deprecated, use search( \%condition ) instead'
387 unless $rsrc->result_class->isa('DBIx::Class::CDBICompat');
389 for my $i (0 .. $#_) {
391 $self->throw_exception ('All keys in condition key/value pairs must be plain scalars')
392 if (! defined $_[$i] or ref $_[$i] ne '');
398 # see if we can keep the cache (no $rs changes)
400 my %safe = (alias => 1, cache => 1);
401 if ( ! List::Util::first { !$safe{$_} } keys %$call_attrs and (
404 ref $call_cond eq 'HASH' && ! keys %$call_cond
406 ref $call_cond eq 'ARRAY' && ! @$call_cond
408 $cache = $self->get_cache;
411 my $old_attrs = { %{$self->{attrs}} };
412 my ($old_having, $old_where) = delete @{$old_attrs}{qw(having where)};
414 my $new_attrs = { %$old_attrs };
416 # take care of call attrs (only if anything is changing)
417 if ($call_attrs and keys %$call_attrs) {
419 # copy for _normalize_selection
420 $call_attrs = { %$call_attrs };
422 my @selector_attrs = qw/select as columns cols +select +as +columns include_columns/;
424 # reset the current selector list if new selectors are supplied
425 if (List::Util::first { exists $call_attrs->{$_} } qw/columns cols select as/) {
426 delete @{$old_attrs}{(@selector_attrs, '_dark_selector')};
429 # Normalize the new selector list (operates on the passed-in attr structure)
430 # Need to do it on every chain instead of only once on _resolved_attrs, in
431 # order to allow detection of empty vs partial 'as'
432 $call_attrs->{_dark_selector} = $old_attrs->{_dark_selector}
433 if $old_attrs->{_dark_selector};
434 $self->_normalize_selection ($call_attrs);
436 # start with blind overwriting merge, exclude selector attrs
437 $new_attrs = { %{$old_attrs}, %{$call_attrs} };
438 delete @{$new_attrs}{@selector_attrs};
440 for (@selector_attrs) {
441 $new_attrs->{$_} = $self->_merge_attr($old_attrs->{$_}, $call_attrs->{$_})
442 if ( exists $old_attrs->{$_} or exists $call_attrs->{$_} );
445 # older deprecated name, use only if {columns} is not there
446 if (my $c = delete $new_attrs->{cols}) {
447 carp_unique( "Resultset attribute 'cols' is deprecated, use 'columns' instead" );
448 if ($new_attrs->{columns}) {
449 carp "Resultset specifies both the 'columns' and the legacy 'cols' attributes - ignoring 'cols'";
452 $new_attrs->{columns} = $c;
457 # join/prefetch use their own crazy merging heuristics
458 foreach my $key (qw/join prefetch/) {
459 $new_attrs->{$key} = $self->_merge_joinpref_attr($old_attrs->{$key}, $call_attrs->{$key})
460 if exists $call_attrs->{$key};
463 # stack binds together
464 $new_attrs->{bind} = [ @{ $old_attrs->{bind} || [] }, @{ $call_attrs->{bind} || [] } ];
468 for ($old_where, $call_cond) {
470 $new_attrs->{where} = $self->_stack_cond (
471 $_, $new_attrs->{where}
476 if (defined $old_having) {
477 $new_attrs->{having} = $self->_stack_cond (
478 $old_having, $new_attrs->{having}
482 my $rs = (ref $self)->new($rsrc, $new_attrs);
484 $rs->set_cache($cache) if ($cache);
490 sub _normalize_selection {
491 my ($self, $attrs) = @_;
494 if ( exists $attrs->{include_columns} ) {
495 carp_unique( "Resultset attribute 'include_columns' is deprecated, use '+columns' instead" );
496 $attrs->{'+columns'} = $self->_merge_attr(
497 $attrs->{'+columns'}, delete $attrs->{include_columns}
501 # columns are always placed first, however
503 # Keep the X vs +X separation until _resolved_attrs time - this allows to
504 # delay the decision on whether to use a default select list ($rsrc->columns)
505 # allowing stuff like the remove_columns helper to work
507 # select/as +select/+as pairs need special handling - the amount of select/as
508 # elements in each pair does *not* have to be equal (think multicolumn
509 # selectors like distinct(foo, bar) ). If the selector is bare (no 'as'
510 # supplied at all) - try to infer the alias, either from the -as parameter
511 # of the selector spec, or use the parameter whole if it looks like a column
512 # name (ugly legacy heuristic). If all fails - leave the selector bare (which
513 # is ok as well), but make sure no more additions to the 'as' chain take place
514 for my $pref ('', '+') {
516 my ($sel, $as) = map {
517 my $key = "${pref}${_}";
519 my $val = [ ref $attrs->{$key} eq 'ARRAY'
521 : $attrs->{$key} || ()
523 delete $attrs->{$key};
527 if (! @$as and ! @$sel ) {
530 elsif (@$as and ! @$sel) {
531 $self->throw_exception(
532 "Unable to handle ${pref}as specification (@$as) without a corresponding ${pref}select"
536 # no as part supplied at all - try to deduce (unless explicit end of named selection is declared)
537 # if any @$as has been supplied we assume the user knows what (s)he is doing
538 # and blindly keep stacking up pieces
539 unless ($attrs->{_dark_selector}) {
542 if ( ref $_ eq 'HASH' and exists $_->{-as} ) {
543 push @$as, $_->{-as};
545 # assume any plain no-space, no-parenthesis string to be a column spec
546 # FIXME - this is retarded but is necessary to support shit like 'count(foo)'
547 elsif ( ! ref $_ and $_ =~ /^ [^\s\(\)]+ $/x) {
550 # if all else fails - raise a flag that no more aliasing will be allowed
552 $attrs->{_dark_selector} = {
554 string => ($dark_sel_dumper ||= do {
555 require Data::Dumper::Concise;
556 Data::Dumper::Concise::DumperObject()->Indent(0);
557 })->Values([$_])->Dump
565 elsif (@$as < @$sel) {
566 $self->throw_exception(
567 "Unable to handle an ${pref}as specification (@$as) with less elements than the corresponding ${pref}select"
570 elsif ($pref and $attrs->{_dark_selector}) {
571 $self->throw_exception(
572 "Unable to process named '+select', resultset contains an unnamed selector $attrs->{_dark_selector}{string}"
578 $attrs->{"${pref}select"} = $self->_merge_attr($attrs->{"${pref}select"}, $sel);
579 $attrs->{"${pref}as"} = $self->_merge_attr($attrs->{"${pref}as"}, $as);
584 my ($self, $left, $right) = @_;
586 # collapse single element top-level conditions
587 # (single pass only, unlikely to need recursion)
588 for ($left, $right) {
589 if (ref $_ eq 'ARRAY') {
597 elsif (ref $_ eq 'HASH') {
598 my ($first, $more) = keys %$_;
601 if (! defined $first) {
605 elsif (! defined $more) {
606 if ($first eq '-and' and ref $_->{'-and'} eq 'HASH') {
609 elsif ($first eq '-or' and ref $_->{'-or'} eq 'ARRAY') {
616 # merge hashes with weeding out of duplicates (simple cases only)
617 if (ref $left eq 'HASH' and ref $right eq 'HASH') {
619 # shallow copy to destroy
620 $right = { %$right };
621 for (grep { exists $right->{$_} } keys %$left) {
622 # the use of eq_deeply here is justified - the rhs of an
623 # expression can contain a lot of twisted weird stuff
624 delete $right->{$_} if Data::Compare::Compare( $left->{$_}, $right->{$_} );
627 $right = undef unless keys %$right;
631 if (defined $left xor defined $right) {
632 return defined $left ? $left : $right;
634 elsif (! defined $left) {
638 return { -and => [ $left, $right ] };
642 =head2 search_literal
644 B<CAVEAT>: C<search_literal> is provided for Class::DBI compatibility and
645 should only be used in that context. C<search_literal> is a convenience
646 method. It is equivalent to calling C<< $schema->search(\[]) >>, but if you
647 want to ensure columns are bound correctly, use L</search>.
649 See L<DBIx::Class::Manual::Cookbook/Searching> and
650 L<DBIx::Class::Manual::FAQ/Searching> for searching techniques that do not
651 require C<search_literal>.
655 =item Arguments: $sql_fragment, @standalone_bind_values
657 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
661 my @cds = $cd_rs->search_literal('year = ? AND title = ?', qw/2001 Reload/);
662 my $newrs = $artist_rs->search_literal('name = ?', 'Metallica');
664 Pass a literal chunk of SQL to be added to the conditional part of the
667 Example of how to use C<search> instead of C<search_literal>
669 my @cds = $cd_rs->search_literal('cdid = ? AND (artist = ? OR artist = ?)', (2, 1, 2));
670 my @cds = $cd_rs->search(\[ 'cdid = ? AND (artist = ? OR artist = ?)', [ 'cdid', 2 ], [ 'artist', 1 ], [ 'artist', 2 ] ]);
675 my ($self, $sql, @bind) = @_;
677 if ( @bind && ref($bind[-1]) eq 'HASH' ) {
680 return $self->search(\[ $sql, map [ {} => $_ ], @bind ], ($attr || () ));
687 =item Arguments: \%columns_values | @pk_values, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
689 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
693 Finds and returns a single row based on supplied criteria. Takes either a
694 hashref with the same format as L</create> (including inference of foreign
695 keys from related objects), or a list of primary key values in the same
696 order as the L<primary columns|DBIx::Class::ResultSource/primary_columns>
697 declaration on the L</result_source>.
699 In either case an attempt is made to combine conditions already existing on
700 the resultset with the condition passed to this method.
702 To aid with preparing the correct query for the storage you may supply the
703 C<key> attribute, which is the name of a
704 L<unique constraint|DBIx::Class::ResultSource/add_unique_constraint> (the
705 unique constraint corresponding to the
706 L<primary columns|DBIx::Class::ResultSource/primary_columns> is always named
707 C<primary>). If the C<key> attribute has been supplied, and DBIC is unable
708 to construct a query that satisfies the named unique constraint fully (
709 non-NULL values for each column member of the constraint) an exception is
712 If no C<key> is specified, the search is carried over all unique constraints
713 which are fully defined by the available condition.
715 If no such constraint is found, C<find> currently defaults to a simple
716 C<< search->(\%column_values) >> which may or may not do what you expect.
717 Note that this fallback behavior may be deprecated in further versions. If
718 you need to search with arbitrary conditions - use L</search>. If the query
719 resulting from this fallback produces more than one row, a warning to the
720 effect is issued, though only the first row is constructed and returned as
723 In addition to C<key>, L</find> recognizes and applies standard
724 L<resultset attributes|/ATTRIBUTES> in the same way as L</search> does.
726 Note that if you have extra concerns about the correctness of the resulting
727 query you need to specify the C<key> attribute and supply the entire condition
728 as an argument to find (since it is not always possible to perform the
729 combination of the resultset condition with the supplied one, especially if
730 the resultset condition contains literal sql).
732 For example, to find a row by its primary key:
734 my $cd = $schema->resultset('CD')->find(5);
736 You can also find a row by a specific unique constraint:
738 my $cd = $schema->resultset('CD')->find(
740 artist => 'Massive Attack',
741 title => 'Mezzanine',
743 { key => 'cd_artist_title' }
746 See also L</find_or_create> and L</update_or_create>.
752 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
754 my $rsrc = $self->result_source;
757 if (exists $attrs->{key}) {
758 $constraint_name = defined $attrs->{key}
760 : $self->throw_exception("An undefined 'key' resultset attribute makes no sense")
764 # Parse out the condition from input
767 if (ref $_[0] eq 'HASH') {
768 $call_cond = { %{$_[0]} };
771 # if only values are supplied we need to default to 'primary'
772 $constraint_name = 'primary' unless defined $constraint_name;
774 my @c_cols = $rsrc->unique_constraint_columns($constraint_name);
776 $self->throw_exception(
777 "No constraint columns, maybe a malformed '$constraint_name' constraint?"
780 $self->throw_exception (
781 'find() expects either a column/value hashref, or a list of values '
782 . "corresponding to the columns of the specified unique constraint '$constraint_name'"
783 ) unless @c_cols == @_;
786 @{$call_cond}{@c_cols} = @_;
790 for my $key (keys %$call_cond) {
792 my $keyref = ref($call_cond->{$key})
794 my $relinfo = $rsrc->relationship_info($key)
796 my $val = delete $call_cond->{$key};
798 next if $keyref eq 'ARRAY'; # has_many for multi_create
800 my $rel_q = $rsrc->_resolve_condition(
801 $relinfo->{cond}, $val, $key, $key
803 die "Can't handle complex relationship conditions in find" if ref($rel_q) ne 'HASH';
804 @related{keys %$rel_q} = values %$rel_q;
808 # relationship conditions take precedence (?)
809 @{$call_cond}{keys %related} = values %related;
811 my $alias = exists $attrs->{alias} ? $attrs->{alias} : $self->{attrs}{alias};
813 if (defined $constraint_name) {
814 $final_cond = $self->_qualify_cond_columns (
816 $self->_build_unique_cond (
824 elsif ($self->{attrs}{accessor} and $self->{attrs}{accessor} eq 'single') {
825 # This means that we got here after a merger of relationship conditions
826 # in ::Relationship::Base::search_related (the row method), and furthermore
827 # the relationship is of the 'single' type. This means that the condition
828 # provided by the relationship (already attached to $self) is sufficient,
829 # as there can be only one row in the database that would satisfy the
833 # no key was specified - fall down to heuristics mode:
834 # run through all unique queries registered on the resultset, and
835 # 'OR' all qualifying queries together
836 my (@unique_queries, %seen_column_combinations);
837 for my $c_name ($rsrc->unique_constraint_names) {
838 next if $seen_column_combinations{
839 join "\x00", sort $rsrc->unique_constraint_columns($c_name)
842 push @unique_queries, try {
843 $self->_build_unique_cond ($c_name, $call_cond, 'croak_on_nulls')
847 $final_cond = @unique_queries
848 ? [ map { $self->_qualify_cond_columns($_, $alias) } @unique_queries ]
849 : $self->_non_unique_find_fallback ($call_cond, $attrs)
853 # Run the query, passing the result_class since it should propagate for find
854 my $rs = $self->search ($final_cond, {result_class => $self->result_class, %$attrs});
855 if ($rs->_resolved_attrs->{collapse}) {
857 carp "Query returned more than one row" if $rs->next;
865 # This is a stop-gap method as agreed during the discussion on find() cleanup:
866 # http://lists.scsys.co.uk/pipermail/dbix-class/2010-October/009535.html
868 # It is invoked when find() is called in legacy-mode with insufficiently-unique
869 # condition. It is provided for overrides until a saner way forward is devised
871 # *NOTE* This is not a public method, and it's *GUARANTEED* to disappear down
872 # the road. Please adjust your tests accordingly to catch this situation early
873 # DBIx::Class::ResultSet->can('_non_unique_find_fallback') is reasonable
875 # The method will not be removed without an adequately complete replacement
876 # for strict-mode enforcement
877 sub _non_unique_find_fallback {
878 my ($self, $cond, $attrs) = @_;
880 return $self->_qualify_cond_columns(
882 exists $attrs->{alias}
884 : $self->{attrs}{alias}
889 sub _qualify_cond_columns {
890 my ($self, $cond, $alias) = @_;
892 my %aliased = %$cond;
893 for (keys %aliased) {
894 $aliased{"$alias.$_"} = delete $aliased{$_}
901 sub _build_unique_cond {
902 my ($self, $constraint_name, $extra_cond, $croak_on_null) = @_;
904 my @c_cols = $self->result_source->unique_constraint_columns($constraint_name);
906 # combination may fail if $self->{cond} is non-trivial
907 my ($final_cond) = try {
908 $self->_merge_with_rscond ($extra_cond)
913 # trim out everything not in $columns
914 $final_cond = { map {
915 exists $final_cond->{$_}
916 ? ( $_ => $final_cond->{$_} )
920 if (my @missing = grep
921 { ! ($croak_on_null ? defined $final_cond->{$_} : exists $final_cond->{$_}) }
924 $self->throw_exception( sprintf ( "Unable to satisfy requested constraint '%s', no values for column(s): %s",
926 join (', ', map { "'$_'" } @missing),
933 !$ENV{DBIC_NULLABLE_KEY_NOWARN}
935 my @undefs = sort grep { ! defined $final_cond->{$_} } (keys %$final_cond)
937 carp_unique ( sprintf (
938 "NULL/undef values supplied for requested unique constraint '%s' (NULL "
939 . 'values in column(s): %s). This is almost certainly not what you wanted, '
940 . 'though you can set DBIC_NULLABLE_KEY_NOWARN to disable this warning.',
942 join (', ', map { "'$_'" } @undefs),
949 =head2 search_related
953 =item Arguments: $rel_name, $cond?, L<\%attrs?|/ATTRIBUTES>
955 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
959 $new_rs = $cd_rs->search_related('artist', {
963 Searches the specified relationship, optionally specifying a condition and
964 attributes for matching records. See L</ATTRIBUTES> for more information.
966 In list context, C<< ->all() >> is called implicitly on the resultset, thus
967 returning a list of result objects instead. To avoid that, use L</search_related_rs>.
969 See also L</search_related_rs>.
974 return shift->related_resultset(shift)->search(@_);
977 =head2 search_related_rs
979 This method works exactly the same as search_related, except that
980 it guarantees a resultset, even in list context.
984 sub search_related_rs {
985 return shift->related_resultset(shift)->search_rs(@_);
992 =item Arguments: none
994 =item Return Value: L<$cursor|DBIx::Class::Cursor>
998 Returns a storage-driven cursor to the given resultset. See
999 L<DBIx::Class::Cursor> for more information.
1006 return $self->{cursor} ||= do {
1007 my $attrs = $self->_resolved_attrs;
1008 $self->result_source->storage->select(
1009 $attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs
1018 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1020 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1024 my $cd = $schema->resultset('CD')->single({ year => 2001 });
1026 Inflates the first result without creating a cursor if the resultset has
1027 any records in it; if not returns C<undef>. Used by L</find> as a lean version
1030 While this method can take an optional search condition (just like L</search>)
1031 being a fast-code-path it does not recognize search attributes. If you need to
1032 add extra joins or similar, call L</search> and then chain-call L</single> on the
1033 L<DBIx::Class::ResultSet> returned.
1039 As of 0.08100, this method enforces the assumption that the preceding
1040 query returns only one row. If more than one row is returned, you will receive
1043 Query returned more than one row
1045 In this case, you should be using L</next> or L</find> instead, or if you really
1046 know what you are doing, use the L</rows> attribute to explicitly limit the size
1049 This method will also throw an exception if it is called on a resultset prefetching
1050 has_many, as such a prefetch implies fetching multiple rows from the database in
1051 order to assemble the resulting object.
1058 my ($self, $where) = @_;
1060 $self->throw_exception('single() only takes search conditions, no attributes. You want ->search( $cond, $attrs )->single()');
1063 my $attrs = { %{$self->_resolved_attrs} };
1065 $self->throw_exception(
1066 'single() can not be used on resultsets collapsing a has_many. Use find( \%cond ) or next() instead'
1067 ) if $attrs->{collapse};
1070 if (defined $attrs->{where}) {
1073 [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ }
1074 $where, delete $attrs->{where} ]
1077 $attrs->{where} = $where;
1081 my $data = [ $self->result_source->storage->select_single(
1082 $attrs->{from}, $attrs->{select},
1083 $attrs->{where}, $attrs
1086 return undef unless @$data;
1087 $self->{_stashed_rows} = [ $data ];
1088 $self->_construct_results->[0];
1094 # Recursively collapse the query, accumulating values for each column.
1096 sub _collapse_query {
1097 my ($self, $query, $collapsed) = @_;
1101 if (ref $query eq 'ARRAY') {
1102 foreach my $subquery (@$query) {
1103 next unless ref $subquery; # -or
1104 $collapsed = $self->_collapse_query($subquery, $collapsed);
1107 elsif (ref $query eq 'HASH') {
1108 if (keys %$query and (keys %$query)[0] eq '-and') {
1109 foreach my $subquery (@{$query->{-and}}) {
1110 $collapsed = $self->_collapse_query($subquery, $collapsed);
1114 foreach my $col (keys %$query) {
1115 my $value = $query->{$col};
1116 $collapsed->{$col}{$value}++;
1128 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1130 =item Return Value: L<$resultsetcolumn|DBIx::Class::ResultSetColumn>
1134 my $max_length = $rs->get_column('length')->max;
1136 Returns a L<DBIx::Class::ResultSetColumn> instance for a column of the ResultSet.
1141 my ($self, $column) = @_;
1142 my $new = DBIx::Class::ResultSetColumn->new($self, $column);
1150 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1152 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1156 # WHERE title LIKE '%blue%'
1157 $cd_rs = $rs->search_like({ title => '%blue%'});
1159 Performs a search, but uses C<LIKE> instead of C<=> as the condition. Note
1160 that this is simply a convenience method retained for ex Class::DBI users.
1161 You most likely want to use L</search> with specific operators.
1163 For more information, see L<DBIx::Class::Manual::Cookbook>.
1165 This method is deprecated and will be removed in 0.09. Use L</search()>
1166 instead. An example conversion is:
1168 ->search_like({ foo => 'bar' });
1172 ->search({ foo => { like => 'bar' } });
1179 'search_like() is deprecated and will be removed in DBIC version 0.09.'
1180 .' Instead use ->search({ x => { -like => "y%" } })'
1181 .' (note the outer pair of {}s - they are important!)'
1183 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
1184 my $query = ref $_[0] eq 'HASH' ? { %{shift()} }: {@_};
1185 $query->{$_} = { 'like' => $query->{$_} } for keys %$query;
1186 return $class->search($query, { %$attrs });
1193 =item Arguments: $first, $last
1195 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1199 Returns a resultset or object list representing a subset of elements from the
1200 resultset slice is called on. Indexes are from 0, i.e., to get the first
1201 three records, call:
1203 my ($one, $two, $three) = $rs->slice(0, 2);
1208 my ($self, $min, $max) = @_;
1209 my $attrs = {}; # = { %{ $self->{attrs} || {} } };
1210 $attrs->{offset} = $self->{attrs}{offset} || 0;
1211 $attrs->{offset} += $min;
1212 $attrs->{rows} = ($max ? ($max - $min + 1) : 1);
1213 return $self->search(undef, $attrs);
1214 #my $slice = (ref $self)->new($self->result_source, $attrs);
1215 #return (wantarray ? $slice->all : $slice);
1222 =item Arguments: none
1224 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1228 Returns the next element in the resultset (C<undef> is there is none).
1230 Can be used to efficiently iterate over records in the resultset:
1232 my $rs = $schema->resultset('CD')->search;
1233 while (my $cd = $rs->next) {
1237 Note that you need to store the resultset object, and call C<next> on it.
1238 Calling C<< resultset('Table')->next >> repeatedly will always return the
1239 first record from the resultset.
1246 if (my $cache = $self->get_cache) {
1247 $self->{all_cache_position} ||= 0;
1248 return $cache->[$self->{all_cache_position}++];
1251 if ($self->{attrs}{cache}) {
1252 delete $self->{pager};
1253 $self->{all_cache_position} = 1;
1254 return ($self->all)[0];
1257 return shift(@{$self->{_stashed_results}}) if @{ $self->{_stashed_results}||[] };
1259 $self->{_stashed_results} = $self->_construct_results
1262 return shift @{$self->{_stashed_results}};
1265 # Constructs as many results as it can in one pass while respecting
1266 # cursor laziness. Several modes of operation:
1268 # * Always builds everything present in @{$self->{_stashed_rows}}
1269 # * If called with $fetch_all true - pulls everything off the cursor and
1270 # builds all result structures (or objects) in one pass
1271 # * If $self->_resolved_attrs->{collapse} is true, checks the order_by
1272 # and if the resultset is ordered properly by the left side:
1273 # * Fetches stuff off the cursor until the "master object" changes,
1274 # and saves the last extra row (if any) in @{$self->{_stashed_rows}}
1276 # * Just fetches, and collapses/constructs everything as if $fetch_all
1277 # was requested (there is no other way to collapse except for an
1279 # * If no collapse is requested - just get the next row, construct and
1281 sub _construct_results {
1282 my ($self, $fetch_all) = @_;
1284 my $rsrc = $self->result_source;
1285 my $attrs = $self->_resolved_attrs;
1290 ! $attrs->{order_by}
1294 my @pcols = $rsrc->primary_columns
1296 # default order for collapsing unless the user asked for something
1297 $attrs->{order_by} = [ map { join '.', $attrs->{alias}, $_} @pcols ];
1298 $attrs->{_ordered_for_collapse} = 1;
1299 $attrs->{_order_is_artificial} = 1;
1302 # this will be used as both initial raw-row collector AND as a RV of
1303 # _construct_results. Not regrowing the array twice matters a lot...
1304 # a surprising amount actually
1305 my $rows = delete $self->{_stashed_rows};
1307 my $cursor; # we may not need one at all
1309 my $did_fetch_all = $fetch_all;
1312 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1313 $rows = [ ($rows ? @$rows : ()), $self->cursor->all ];
1315 elsif( $attrs->{collapse} ) {
1317 # a cursor will need to be closed over in case of collapse
1318 $cursor = $self->cursor;
1320 $attrs->{_ordered_for_collapse} = (
1326 ->_main_source_order_by_portion_is_stable($rsrc, $attrs->{order_by}, $attrs->{where})
1328 ) unless defined $attrs->{_ordered_for_collapse};
1330 if (! $attrs->{_ordered_for_collapse}) {
1333 # instead of looping over ->next, use ->all in stealth mode
1334 # *without* calling a ->reset afterwards
1335 # FIXME ENCAPSULATION - encapsulation breach, cursor method additions pending
1336 if (! $cursor->{_done}) {
1337 $rows = [ ($rows ? @$rows : ()), $cursor->all ];
1338 $cursor->{_done} = 1;
1343 if (! $did_fetch_all and ! @{$rows||[]} ) {
1344 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1345 $cursor ||= $self->cursor;
1346 if (scalar (my @r = $cursor->next) ) {
1351 return undef unless @{$rows||[]};
1353 # sanity check - people are too clever for their own good
1354 if ($attrs->{collapse} and my $aliastypes = $attrs->{_sqlmaker_select_args}[3]{_aliastypes} ) {
1356 my $multiplied_selectors;
1357 for my $sel_alias ( grep { $_ ne $attrs->{alias} } keys %{ $aliastypes->{selecting} } ) {
1359 $aliastypes->{multiplying}{$sel_alias}
1361 scalar grep { $aliastypes->{multiplying}{(values %$_)[0]} } @{ $aliastypes->{selecting}{$sel_alias}{-parents} }
1363 $multiplied_selectors->{$_} = 1 for values %{$aliastypes->{selecting}{$sel_alias}{-seen_columns}}
1367 for my $i (0 .. $#{$attrs->{as}} ) {
1368 my $sel = $attrs->{select}[$i];
1370 if (ref $sel eq 'SCALAR') {
1373 elsif( ref $sel eq 'REF' and ref $$sel eq 'ARRAY' ) {
1377 $self->throw_exception(
1378 'Result collapse not possible - selection from a has_many source redirected to the main object'
1379 ) if ($multiplied_selectors->{$sel} and $attrs->{as}[$i] !~ /\./);
1383 # hotspot - skip the setter
1384 my $res_class = $self->_result_class;
1386 my $inflator_cref = $self->{_result_inflator}{cref} ||= do {
1387 $res_class->can ('inflate_result')
1388 or $self->throw_exception("Inflator $res_class does not provide an inflate_result() method");
1391 my $infmap = $attrs->{as};
1393 $self->{_result_inflator}{is_core_row} = ( (
1396 ( \&DBIx::Class::Row::inflate_result || die "No ::Row::inflate_result() - can't happen" )
1397 ) ? 1 : 0 ) unless defined $self->{_result_inflator}{is_core_row};
1399 $self->{_result_inflator}{is_hri} = ( (
1400 ! $self->{_result_inflator}{is_core_row}
1403 require DBIx::Class::ResultClass::HashRefInflator
1405 DBIx::Class::ResultClass::HashRefInflator->can('inflate_result')
1407 ) ? 1 : 0 ) unless defined $self->{_result_inflator}{is_hri};
1410 if (! $attrs->{_related_results_construction}) {
1411 # construct a much simpler array->hash folder for the one-table cases right here
1412 if ($self->{_result_inflator}{is_hri}) {
1413 for my $r (@$rows) {
1414 $r = { map { $infmap->[$_] => $r->[$_] } 0..$#$infmap };
1417 # FIXME SUBOPTIMAL this is a very very very hot spot
1418 # while rather optimal we can *still* do much better, by
1419 # building a smarter Row::inflate_result(), and
1420 # switch to feeding it data via a much leaner interface
1422 # crude unscientific benchmarking indicated the shortcut eval is not worth it for
1423 # this particular resultset size
1424 elsif (@$rows < 60) {
1425 for my $r (@$rows) {
1426 $r = $inflator_cref->($res_class, $rsrc, { map { $infmap->[$_] => $r->[$_] } (0..$#$infmap) } );
1431 '$_ = $inflator_cref->($res_class, $rsrc, { %s }) for @$rows',
1432 join (', ', map { "\$infmap->[$_] => \$_->[$_]" } 0..$#$infmap )
1438 $self->{_result_inflator}{is_hri} ? 'hri'
1439 : $self->{_result_inflator}{is_core_row} ? 'classic_pruning'
1440 : 'classic_nonpruning'
1443 # $args and $attrs to _mk_row_parser are separated to delineate what is
1444 # core collapser stuff and what is dbic $rs specific
1445 @{$self->{_row_parser}{$parser_type}}{qw(cref nullcheck)} = $rsrc->_mk_row_parser({
1447 inflate_map => $infmap,
1448 collapse => $attrs->{collapse},
1449 premultiplied => $attrs->{_main_source_premultiplied},
1450 hri_style => $self->{_result_inflator}{is_hri},
1451 prune_null_branches => $self->{_result_inflator}{is_hri} || $self->{_result_inflator}{is_core_row},
1452 }, $attrs) unless $self->{_row_parser}{$parser_type}{cref};
1454 # column_info metadata historically hasn't been too reliable.
1455 # We need to start fixing this somehow (the collapse resolver
1456 # can't work without it). Add an explicit check for the *main*
1457 # result, hopefully this will gradually weed out such errors
1459 # FIXME - this is a temporary kludge that reduces performance
1460 # It is however necessary for the time being
1461 my ($unrolled_non_null_cols_to_check, $err);
1463 if (my $check_non_null_cols = $self->{_row_parser}{$parser_type}{nullcheck} ) {
1466 'Collapse aborted due to invalid ResultSource metadata - the following '
1467 . 'selections are declared non-nullable but NULLs were retrieved: '
1471 COL: for my $i (@$check_non_null_cols) {
1472 ! defined $_->[$i] and push @violating_idx, $i and next COL for @$rows;
1475 $self->throw_exception( $err . join (', ', map { "'$infmap->[$_]'" } @violating_idx ) )
1478 $unrolled_non_null_cols_to_check = join (',', @$check_non_null_cols);
1482 ($did_fetch_all or ! $attrs->{collapse}) ? undef
1483 : defined $unrolled_non_null_cols_to_check ? eval sprintf <<'EOS', $unrolled_non_null_cols_to_check
1485 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1486 my @r = $cursor->next or return;
1487 if (my @violating_idx = grep { ! defined $r[$_] } (%s) ) {
1488 $self->throw_exception( $err . join (', ', map { "'$infmap->[$_]'" } @violating_idx ) )
1494 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1495 my @r = $cursor->next or return;
1500 $self->{_row_parser}{$parser_type}{cref}->(
1502 $next_cref ? ( $next_cref, $self->{_stashed_rows} = [] ) : (),
1505 # Special-case multi-object HRI - there is no $inflator_cref pass
1506 unless ($self->{_result_inflator}{is_hri}) {
1507 $_ = $inflator_cref->($res_class, $rsrc, @$_) for @$rows
1511 # The @$rows check seems odd at first - why wouldn't we want to warn
1512 # regardless? The issue is things like find() etc, where the user
1513 # *knows* only one result will come back. In these cases the ->all
1514 # is not a pessimization, but rather something we actually want
1516 'Unable to properly collapse has_many results in iterator mode due '
1517 . 'to order criteria - performed an eager cursor slurp underneath. '
1518 . 'Consider using ->all() instead'
1519 ) if ( ! $fetch_all and @$rows > 1 );
1524 =head2 result_source
1528 =item Arguments: L<$result_source?|DBIx::Class::ResultSource>
1530 =item Return Value: L<$result_source|DBIx::Class::ResultSource>
1534 An accessor for the primary ResultSource object from which this ResultSet
1541 =item Arguments: $result_class?
1543 =item Return Value: $result_class
1547 An accessor for the class to use when creating result objects. Defaults to
1548 C<< result_source->result_class >> - which in most cases is the name of the
1549 L<"table"|DBIx::Class::Manual::Glossary/"ResultSource"> class.
1551 Note that changing the result_class will also remove any components
1552 that were originally loaded in the source class via
1553 L<DBIx::Class::ResultSource/load_components>. Any overloaded methods
1554 in the original source class will not run.
1559 my ($self, $result_class) = @_;
1560 if ($result_class) {
1562 # don't fire this for an object
1563 $self->ensure_class_loaded($result_class)
1564 unless ref($result_class);
1566 if ($self->get_cache) {
1567 carp_unique('Changing the result_class of a ResultSet instance with cached results is a noop - the cache contents will not be altered');
1569 # FIXME ENCAPSULATION - encapsulation breach, cursor method additions pending
1570 elsif ($self->{cursor} && $self->{cursor}{_pos}) {
1571 $self->throw_exception('Changing the result_class of a ResultSet instance with an active cursor is not supported');
1574 $self->_result_class($result_class);
1576 delete $self->{_result_inflator};
1578 $self->_result_class;
1585 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1587 =item Return Value: $count
1591 Performs an SQL C<COUNT> with the same query as the resultset was built
1592 with to find the number of elements. Passing arguments is equivalent to
1593 C<< $rs->search ($cond, \%attrs)->count >>
1599 return $self->search(@_)->count if @_ and defined $_[0];
1600 return scalar @{ $self->get_cache } if $self->get_cache;
1602 my $attrs = { %{ $self->_resolved_attrs } };
1604 # this is a little optimization - it is faster to do the limit
1605 # adjustments in software, instead of a subquery
1606 my ($rows, $offset) = delete @{$attrs}{qw/rows offset/};
1609 if ($self->_has_resolved_attr (qw/collapse group_by/)) {
1610 $crs = $self->_count_subq_rs ($attrs);
1613 $crs = $self->_count_rs ($attrs);
1615 my $count = $crs->next;
1617 $count -= $offset if $offset;
1618 $count = $rows if $rows and $rows < $count;
1619 $count = 0 if ($count < 0);
1628 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1630 =item Return Value: L<$count_rs|DBIx::Class::ResultSetColumn>
1634 Same as L</count> but returns a L<DBIx::Class::ResultSetColumn> object.
1635 This can be very handy for subqueries:
1637 ->search( { amount => $some_rs->count_rs->as_query } )
1639 As with regular resultsets the SQL query will be executed only after
1640 the resultset is accessed via L</next> or L</all>. That would return
1641 the same single value obtainable via L</count>.
1647 return $self->search(@_)->count_rs if @_;
1649 # this may look like a lack of abstraction (count() does about the same)
1650 # but in fact an _rs *must* use a subquery for the limits, as the
1651 # software based limiting can not be ported if this $rs is to be used
1652 # in a subquery itself (i.e. ->as_query)
1653 if ($self->_has_resolved_attr (qw/collapse group_by offset rows/)) {
1654 return $self->_count_subq_rs($self->{_attrs});
1657 return $self->_count_rs($self->{_attrs});
1662 # returns a ResultSetColumn object tied to the count query
1665 my ($self, $attrs) = @_;
1667 my $rsrc = $self->result_source;
1669 my $tmp_attrs = { %$attrs };
1670 # take off any limits, record_filter is cdbi, and no point of ordering nor locking a count
1671 delete @{$tmp_attrs}{qw/rows offset order_by record_filter for/};
1673 # overwrite the selector (supplied by the storage)
1674 $rsrc->resultset_class->new($rsrc, {
1676 select => $rsrc->storage->_count_select ($rsrc, $attrs),
1678 })->get_column ('count');
1682 # same as above but uses a subquery
1684 sub _count_subq_rs {
1685 my ($self, $attrs) = @_;
1687 my $rsrc = $self->result_source;
1689 my $sub_attrs = { %$attrs };
1690 # extra selectors do not go in the subquery and there is no point of ordering it, nor locking it
1691 delete @{$sub_attrs}{qw/collapse columns as select order_by for/};
1693 # if we multi-prefetch we group_by something unique, as this is what we would
1694 # get out of the rs via ->next/->all. We *DO WANT* to clobber old group_by regardless
1695 if ( $attrs->{collapse} ) {
1696 $sub_attrs->{group_by} = [ map { "$attrs->{alias}.$_" } @{
1697 $rsrc->_identifying_column_set || $self->throw_exception(
1698 'Unable to construct a unique group_by criteria properly collapsing the '
1699 . 'has_many prefetch before count()'
1704 # Calculate subquery selector
1705 if (my $g = $sub_attrs->{group_by}) {
1707 my $sql_maker = $rsrc->storage->sql_maker;
1709 # necessary as the group_by may refer to aliased functions
1711 for my $sel (@{$attrs->{select}}) {
1712 $sel_index->{$sel->{-as}} = $sel
1713 if (ref $sel eq 'HASH' and $sel->{-as});
1716 # anything from the original select mentioned on the group-by needs to make it to the inner selector
1717 # also look for named aggregates referred in the having clause
1718 # having often contains scalarrefs - thus parse it out entirely
1720 if ($attrs->{having}) {
1721 local $sql_maker->{having_bind};
1722 local $sql_maker->{quote_char} = $sql_maker->{quote_char};
1723 local $sql_maker->{name_sep} = $sql_maker->{name_sep};
1724 unless (defined $sql_maker->{quote_char} and length $sql_maker->{quote_char}) {
1725 $sql_maker->{quote_char} = [ "\x00", "\xFF" ];
1726 # if we don't unset it we screw up retarded but unfortunately working
1727 # 'MAX(foo.bar)' => { '>', 3 }
1728 $sql_maker->{name_sep} = '';
1731 my ($lquote, $rquote, $sep) = map { quotemeta $_ } ($sql_maker->_quote_chars, $sql_maker->name_sep);
1733 my $having_sql = $sql_maker->_parse_rs_attrs ({ having => $attrs->{having} });
1736 # search for both a proper quoted qualified string, for a naive unquoted scalarref
1737 # and if all fails for an utterly naive quoted scalar-with-function
1738 while ($having_sql =~ /
1739 $rquote $sep $lquote (.+?) $rquote
1741 [\s,] \w+ \. (\w+) [\s,]
1743 [\s,] $lquote (.+?) $rquote [\s,]
1745 my $part = $1 || $2 || $3; # one of them matched if we got here
1746 unless ($seen_having{$part}++) {
1753 my $colpiece = $sel_index->{$_} || $_;
1755 # unqualify join-based group_by's. Arcane but possible query
1756 # also horrible horrible hack to alias a column (not a func.)
1757 # (probably need to introduce SQLA syntax)
1758 if ($colpiece =~ /\./ && $colpiece !~ /^$attrs->{alias}\./) {
1761 $colpiece = \ sprintf ('%s AS %s', map { $sql_maker->_quote ($_) } ($colpiece, $as) );
1763 push @{$sub_attrs->{select}}, $colpiece;
1767 my @pcols = map { "$attrs->{alias}.$_" } ($rsrc->primary_columns);
1768 $sub_attrs->{select} = @pcols ? \@pcols : [ 1 ];
1771 return $rsrc->resultset_class
1772 ->new ($rsrc, $sub_attrs)
1774 ->search ({}, { columns => { count => $rsrc->storage->_count_select ($rsrc, $attrs) } })
1775 ->get_column ('count');
1779 =head2 count_literal
1781 B<CAVEAT>: C<count_literal> is provided for Class::DBI compatibility and
1782 should only be used in that context. See L</search_literal> for further info.
1786 =item Arguments: $sql_fragment, @standalone_bind_values
1788 =item Return Value: $count
1792 Counts the results in a literal query. Equivalent to calling L</search_literal>
1793 with the passed arguments, then L</count>.
1797 sub count_literal { shift->search_literal(@_)->count; }
1803 =item Arguments: none
1805 =item Return Value: L<@result_objs|DBIx::Class::Manual::ResultClass>
1809 Returns all elements in the resultset.
1816 $self->throw_exception("all() doesn't take any arguments, you probably wanted ->search(...)->all()");
1819 delete @{$self}{qw/_stashed_rows _stashed_results/};
1821 if (my $c = $self->get_cache) {
1825 $self->cursor->reset;
1827 my $objs = $self->_construct_results('fetch_all') || [];
1829 $self->set_cache($objs) if $self->{attrs}{cache};
1838 =item Arguments: none
1840 =item Return Value: $self
1844 Resets the resultset's cursor, so you can iterate through the elements again.
1845 Implicitly resets the storage cursor, so a subsequent L</next> will trigger
1853 delete @{$self}{qw/_stashed_rows _stashed_results/};
1854 $self->{all_cache_position} = 0;
1855 $self->cursor->reset;
1863 =item Arguments: none
1865 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1869 L<Resets|/reset> the resultset (causing a fresh query to storage) and returns
1870 an object for the first result (or C<undef> if the resultset is empty).
1875 return $_[0]->reset->next;
1881 # Determines whether and what type of subquery is required for the $rs operation.
1882 # If grouping is necessary either supplies its own, or verifies the current one
1883 # After all is done delegates to the proper storage method.
1885 sub _rs_update_delete {
1886 my ($self, $op, $values) = @_;
1888 my $rsrc = $self->result_source;
1889 my $storage = $rsrc->schema->storage;
1891 my $attrs = { %{$self->_resolved_attrs} };
1893 my $join_classifications;
1894 my ($existing_group_by) = delete @{$attrs}{qw(group_by _grouped_by_distinct)};
1896 # do we need a subquery for any reason?
1898 defined $existing_group_by
1900 # if {from} is unparseable wrap a subq
1901 ref($attrs->{from}) ne 'ARRAY'
1903 # limits call for a subq
1904 $self->_has_resolved_attr(qw/rows offset/)
1907 # simplify the joinmap, so we can further decide if a subq is necessary
1908 if (!$needs_subq and @{$attrs->{from}} > 1) {
1910 ($attrs->{from}, $join_classifications) =
1911 $storage->_prune_unused_joins ($attrs->{from}, $attrs->{select}, $self->{cond}, $attrs);
1913 # any non-pruneable non-local restricting joins imply subq
1914 $needs_subq = defined List::Util::first { $_ ne $attrs->{alias} } keys %{ $join_classifications->{restricting} || {} };
1917 # check if the head is composite (by now all joins are thrown out unless $needs_subq)
1919 (ref $attrs->{from}[0]) ne 'HASH'
1921 ref $attrs->{from}[0]{ $attrs->{from}[0]{-alias} }
1925 # do we need anything like a subquery?
1926 if (! $needs_subq) {
1927 # Most databases do not allow aliasing of tables in UPDATE/DELETE. Thus
1928 # a condition containing 'me' or other table prefixes will not work
1929 # at all. Tell SQLMaker to dequalify idents via a gross hack.
1930 if ($self->{cond}) {
1932 my $sqla = $rsrc->storage->sql_maker;
1933 local $sqla->{_dequalify_idents} = 1;
1934 \[ $sqla->_recurse_where($self->{cond}) ];
1939 # we got this far - means it is time to wrap a subquery
1940 my $idcols = $rsrc->_identifying_column_set || $self->throw_exception(
1942 "Unable to perform complex resultset %s() without an identifying set of columns on source '%s'",
1948 # make a new $rs selecting only the PKs (that's all we really need for the subq)
1949 delete $attrs->{$_} for qw/select as collapse/;
1950 $attrs->{columns} = [ map { "$attrs->{alias}.$_" } @$idcols ];
1952 # this will be consumed by the pruner waaaaay down the stack
1953 $attrs->{_force_prune_multiplying_joins} = 1;
1955 my $subrs = (ref $self)->new($rsrc, $attrs);
1957 if (@$idcols == 1) {
1958 $cond = { $idcols->[0] => { -in => $subrs->as_query } };
1960 elsif ($storage->_use_multicolumn_in) {
1961 # no syntax for calling this properly yet
1962 # !!! EXPERIMENTAL API !!! WILL CHANGE !!!
1963 $cond = $storage->sql_maker->_where_op_multicolumn_in (
1964 $idcols, # how do I convey a list of idents...? can binds reside on lhs?
1969 # if all else fails - get all primary keys and operate over a ORed set
1970 # wrap in a transaction for consistency
1971 # this is where the group_by/multiplication starts to matter
1975 keys %{ $join_classifications->{multiplying} || {} }
1977 # make sure if there is a supplied group_by it matches the columns compiled above
1978 # perfectly. Anything else can not be sanely executed on most databases so croak
1979 # right then and there
1980 if ($existing_group_by) {
1981 my @current_group_by = map
1982 { $_ =~ /\./ ? $_ : "$attrs->{alias}.$_" }
1987 join ("\x00", sort @current_group_by)
1989 join ("\x00", sort @{$attrs->{columns}} )
1991 $self->throw_exception (
1992 "You have just attempted a $op operation on a resultset which does group_by"
1993 . ' on columns other than the primary keys, while DBIC internally needs to retrieve'
1994 . ' the primary keys in a subselect. All sane RDBMS engines do not support this'
1995 . ' kind of queries. Please retry the operation with a modified group_by or'
1996 . ' without using one at all.'
2001 $subrs = $subrs->search({}, { group_by => $attrs->{columns} });
2004 $guard = $storage->txn_scope_guard;
2007 for my $row ($subrs->cursor->all) {
2009 { $idcols->[$_] => $row->[$_] }
2016 my $res = $storage->$op (
2018 $op eq 'update' ? $values : (),
2022 $guard->commit if $guard;
2031 =item Arguments: \%values
2033 =item Return Value: $underlying_storage_rv
2037 Sets the specified columns in the resultset to the supplied values in a
2038 single query. Note that this will not run any accessor/set_column/update
2039 triggers, nor will it update any result object instances derived from this
2040 resultset (this includes the contents of the L<resultset cache|/set_cache>
2041 if any). See L</update_all> if you need to execute any on-update
2042 triggers or cascades defined either by you or a
2043 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
2045 The return value is a pass through of what the underlying
2046 storage backend returned, and may vary. See L<DBI/execute> for the most
2051 Note that L</update> does not process/deflate any of the values passed in.
2052 This is unlike the corresponding L<DBIx::Class::Row/update>. The user must
2053 ensure manually that any value passed to this method will stringify to
2054 something the RDBMS knows how to deal with. A notable example is the
2055 handling of L<DateTime> objects, for more info see:
2056 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
2061 my ($self, $values) = @_;
2062 $self->throw_exception('Values for update must be a hash')
2063 unless ref $values eq 'HASH';
2065 return $self->_rs_update_delete ('update', $values);
2072 =item Arguments: \%values
2074 =item Return Value: 1
2078 Fetches all objects and updates them one at a time via
2079 L<DBIx::Class::Row/update>. Note that C<update_all> will run DBIC defined
2080 triggers, while L</update> will not.
2085 my ($self, $values) = @_;
2086 $self->throw_exception('Values for update_all must be a hash')
2087 unless ref $values eq 'HASH';
2089 my $guard = $self->result_source->schema->txn_scope_guard;
2090 $_->update({%$values}) for $self->all; # shallow copy - update will mangle it
2099 =item Arguments: none
2101 =item Return Value: $underlying_storage_rv
2105 Deletes the rows matching this resultset in a single query. Note that this
2106 will not run any delete triggers, nor will it alter the
2107 L<in_storage|DBIx::Class::Row/in_storage> status of any result object instances
2108 derived from this resultset (this includes the contents of the
2109 L<resultset cache|/set_cache> if any). See L</delete_all> if you need to
2110 execute any on-delete triggers or cascades defined either by you or a
2111 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
2113 The return value is a pass through of what the underlying storage backend
2114 returned, and may vary. See L<DBI/execute> for the most common case.
2120 $self->throw_exception('delete does not accept any arguments')
2123 return $self->_rs_update_delete ('delete');
2130 =item Arguments: none
2132 =item Return Value: 1
2136 Fetches all objects and deletes them one at a time via
2137 L<DBIx::Class::Row/delete>. Note that C<delete_all> will run DBIC defined
2138 triggers, while L</delete> will not.
2144 $self->throw_exception('delete_all does not accept any arguments')
2147 my $guard = $self->result_source->schema->txn_scope_guard;
2148 $_->delete for $self->all;
2157 =item Arguments: [ \@column_list, \@row_values+ ] | [ \%col_data+ ]
2159 =item Return Value: L<\@result_objects|DBIx::Class::Manual::ResultClass> (scalar context) | L<@result_objects|DBIx::Class::Manual::ResultClass> (list context)
2163 Accepts either an arrayref of hashrefs or alternatively an arrayref of
2170 The context of this method call has an important effect on what is
2171 submitted to storage. In void context data is fed directly to fastpath
2172 insertion routines provided by the underlying storage (most often
2173 L<DBI/execute_for_fetch>), bypassing the L<new|DBIx::Class::Row/new> and
2174 L<insert|DBIx::Class::Row/insert> calls on the
2175 L<Result|DBIx::Class::Manual::ResultClass> class, including any
2176 augmentation of these methods provided by components. For example if you
2177 are using something like L<DBIx::Class::UUIDColumns> to create primary
2178 keys for you, you will find that your PKs are empty. In this case you
2179 will have to explicitly force scalar or list context in order to create
2184 In non-void (scalar or list) context, this method is simply a wrapper
2185 for L</create>. Depending on list or scalar context either a list of
2186 L<Result|DBIx::Class::Manual::ResultClass> objects or an arrayref
2187 containing these objects is returned.
2189 When supplying data in "arrayref of arrayrefs" invocation style, the
2190 first element should be a list of column names and each subsequent
2191 element should be a data value in the earlier specified column order.
2194 $schema->resultset("Artist")->populate([
2195 [ qw( artistid name ) ],
2196 [ 100, 'A Formally Unknown Singer' ],
2197 [ 101, 'A singer that jumped the shark two albums ago' ],
2198 [ 102, 'An actually cool singer' ],
2201 For the arrayref of hashrefs style each hashref should be a structure
2202 suitable for passing to L</create>. Multi-create is also permitted with
2205 $schema->resultset("Artist")->populate([
2206 { artistid => 4, name => 'Manufactured Crap', cds => [
2207 { title => 'My First CD', year => 2006 },
2208 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2211 { artistid => 5, name => 'Angsty-Whiny Girl', cds => [
2212 { title => 'My parents sold me to a record company', year => 2005 },
2213 { title => 'Why Am I So Ugly?', year => 2006 },
2214 { title => 'I Got Surgery and am now Popular', year => 2007 }
2219 If you attempt a void-context multi-create as in the example above (each
2220 Artist also has the related list of CDs), and B<do not> supply the
2221 necessary autoinc foreign key information, this method will proxy to the
2222 less efficient L</create>, and then throw the Result objects away. In this
2223 case there are obviously no benefits to using this method over L</create>.
2230 # cruft placed in standalone method
2231 my $data = $self->_normalize_populate_args(@_);
2233 return unless @$data;
2235 if(defined wantarray) {
2236 my @created = map { $self->create($_) } @$data;
2237 return wantarray ? @created : \@created;
2240 my $first = $data->[0];
2242 # if a column is a registered relationship, and is a non-blessed hash/array, consider
2243 # it relationship data
2244 my (@rels, @columns);
2245 my $rsrc = $self->result_source;
2246 my $rels = { map { $_ => $rsrc->relationship_info($_) } $rsrc->relationships };
2247 for (keys %$first) {
2248 my $ref = ref $first->{$_};
2249 $rels->{$_} && ($ref eq 'ARRAY' or $ref eq 'HASH')
2255 my @pks = $rsrc->primary_columns;
2257 ## do the belongs_to relationships
2258 foreach my $index (0..$#$data) {
2260 # delegate to create() for any dataset without primary keys with specified relationships
2261 if (grep { !defined $data->[$index]->{$_} } @pks ) {
2263 if (grep { ref $data->[$index]{$r} eq $_ } qw/HASH ARRAY/) { # a related set must be a HASH or AoH
2264 my @ret = $self->populate($data);
2270 foreach my $rel (@rels) {
2271 next unless ref $data->[$index]->{$rel} eq "HASH";
2272 my $result = $self->related_resultset($rel)->create($data->[$index]->{$rel});
2273 my ($reverse_relname, $reverse_relinfo) = %{$rsrc->reverse_relationship_info($rel)};
2274 my $related = $result->result_source->_resolve_condition(
2275 $reverse_relinfo->{cond},
2281 delete $data->[$index]->{$rel};
2282 $data->[$index] = {%{$data->[$index]}, %$related};
2284 push @columns, keys %$related if $index == 0;
2288 ## inherit the data locked in the conditions of the resultset
2289 my ($rs_data) = $self->_merge_with_rscond({});
2290 delete @{$rs_data}{@columns};
2292 ## do bulk insert on current row
2293 $rsrc->storage->insert_bulk(
2295 [@columns, keys %$rs_data],
2296 [ map { [ @$_{@columns}, values %$rs_data ] } @$data ],
2299 ## do the has_many relationships
2300 foreach my $item (@$data) {
2304 foreach my $rel (@rels) {
2305 next unless ref $item->{$rel} eq "ARRAY" && @{ $item->{$rel} };
2307 $main_row ||= $self->new_result({map { $_ => $item->{$_} } @pks});
2309 my $child = $main_row->$rel;
2311 my $related = $child->result_source->_resolve_condition(
2312 $rels->{$rel}{cond},
2318 my @rows_to_add = ref $item->{$rel} eq 'ARRAY' ? @{$item->{$rel}} : ($item->{$rel});
2319 my @populate = map { {%$_, %$related} } @rows_to_add;
2321 $child->populate( \@populate );
2328 # populate() arguments went over several incarnations
2329 # What we ultimately support is AoH
2330 sub _normalize_populate_args {
2331 my ($self, $arg) = @_;
2333 if (ref $arg eq 'ARRAY') {
2337 elsif (ref $arg->[0] eq 'HASH') {
2340 elsif (ref $arg->[0] eq 'ARRAY') {
2342 my @colnames = @{$arg->[0]};
2343 foreach my $values (@{$arg}[1 .. $#$arg]) {
2344 push @ret, { map { $colnames[$_] => $values->[$_] } (0 .. $#colnames) };
2350 $self->throw_exception('Populate expects an arrayref of hashrefs or arrayref of arrayrefs');
2357 =item Arguments: none
2359 =item Return Value: L<$pager|Data::Page>
2363 Returns a L<Data::Page> object for the current resultset. Only makes
2364 sense for queries with a C<page> attribute.
2366 To get the full count of entries for a paged resultset, call
2367 C<total_entries> on the L<Data::Page> object.
2374 return $self->{pager} if $self->{pager};
2376 my $attrs = $self->{attrs};
2377 if (!defined $attrs->{page}) {
2378 $self->throw_exception("Can't create pager for non-paged rs");
2380 elsif ($attrs->{page} <= 0) {
2381 $self->throw_exception('Invalid page number (page-numbers are 1-based)');
2383 $attrs->{rows} ||= 10;
2385 # throw away the paging flags and re-run the count (possibly
2386 # with a subselect) to get the real total count
2387 my $count_attrs = { %$attrs };
2388 delete @{$count_attrs}{qw/rows offset page pager/};
2390 my $total_rs = (ref $self)->new($self->result_source, $count_attrs);
2392 require DBIx::Class::ResultSet::Pager;
2393 return $self->{pager} = DBIx::Class::ResultSet::Pager->new(
2394 sub { $total_rs->count }, #lazy-get the total
2396 $self->{attrs}{page},
2404 =item Arguments: $page_number
2406 =item Return Value: L<$resultset|/search>
2410 Returns a resultset for the $page_number page of the resultset on which page
2411 is called, where each page contains a number of rows equal to the 'rows'
2412 attribute set on the resultset (10 by default).
2417 my ($self, $page) = @_;
2418 return (ref $self)->new($self->result_source, { %{$self->{attrs}}, page => $page });
2425 =item Arguments: \%col_data
2427 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2431 Creates a new result object in the resultset's result class and returns
2432 it. The row is not inserted into the database at this point, call
2433 L<DBIx::Class::Row/insert> to do that. Calling L<DBIx::Class::Row/in_storage>
2434 will tell you whether the result object has been inserted or not.
2436 Passes the hashref of input on to L<DBIx::Class::Row/new>.
2441 my ($self, $values) = @_;
2443 $self->throw_exception( "new_result takes only one argument - a hashref of values" )
2446 $self->throw_exception( "new_result expects a hashref" )
2447 unless (ref $values eq 'HASH');
2449 my ($merged_cond, $cols_from_relations) = $self->_merge_with_rscond($values);
2451 my $new = $self->result_class->new({
2453 ( @$cols_from_relations
2454 ? (-cols_from_relations => $cols_from_relations)
2457 -result_source => $self->result_source, # DO NOT REMOVE THIS, REQUIRED
2461 reftype($new) eq 'HASH'
2467 carp_unique (sprintf (
2468 "%s->new returned a blessed empty hashref - a strong indicator something is wrong with its inheritance chain",
2469 $self->result_class,
2476 # _merge_with_rscond
2478 # Takes a simple hash of K/V data and returns its copy merged with the
2479 # condition already present on the resultset. Additionally returns an
2480 # arrayref of value/condition names, which were inferred from related
2481 # objects (this is needed for in-memory related objects)
2482 sub _merge_with_rscond {
2483 my ($self, $data) = @_;
2485 my (%new_data, @cols_from_relations);
2487 my $alias = $self->{attrs}{alias};
2489 if (! defined $self->{cond}) {
2490 # just massage $data below
2492 elsif ($self->{cond} eq $DBIx::Class::ResultSource::UNRESOLVABLE_CONDITION) {
2493 %new_data = %{ $self->{attrs}{related_objects} || {} }; # nothing might have been inserted yet
2494 @cols_from_relations = keys %new_data;
2496 elsif (ref $self->{cond} ne 'HASH') {
2497 $self->throw_exception(
2498 "Can't abstract implicit construct, resultset condition not a hash"
2502 # precedence must be given to passed values over values inherited from
2503 # the cond, so the order here is important.
2504 my $collapsed_cond = $self->_collapse_cond($self->{cond});
2505 my %implied = %{$self->_remove_alias($collapsed_cond, $alias)};
2507 while ( my($col, $value) = each %implied ) {
2508 my $vref = ref $value;
2514 (keys %$value)[0] eq '='
2516 $new_data{$col} = $value->{'='};
2518 elsif( !$vref or $vref eq 'SCALAR' or blessed($value) ) {
2519 $new_data{$col} = $value;
2526 %{ $self->_remove_alias($data, $alias) },
2529 return (\%new_data, \@cols_from_relations);
2532 # _has_resolved_attr
2534 # determines if the resultset defines at least one
2535 # of the attributes supplied
2537 # used to determine if a subquery is necessary
2539 # supports some virtual attributes:
2541 # This will scan for any joins being present on the resultset.
2542 # It is not a mere key-search but a deep inspection of {from}
2545 sub _has_resolved_attr {
2546 my ($self, @attr_names) = @_;
2548 my $attrs = $self->_resolved_attrs;
2552 for my $n (@attr_names) {
2553 if (grep { $n eq $_ } (qw/-join/) ) {
2554 $extra_checks{$n}++;
2558 my $attr = $attrs->{$n};
2560 next if not defined $attr;
2562 if (ref $attr eq 'HASH') {
2563 return 1 if keys %$attr;
2565 elsif (ref $attr eq 'ARRAY') {
2573 # a resolved join is expressed as a multi-level from
2575 $extra_checks{-join}
2577 ref $attrs->{from} eq 'ARRAY'
2579 @{$attrs->{from}} > 1
2587 # Recursively collapse the condition.
2589 sub _collapse_cond {
2590 my ($self, $cond, $collapsed) = @_;
2594 if (ref $cond eq 'ARRAY') {
2595 foreach my $subcond (@$cond) {
2596 next unless ref $subcond; # -or
2597 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2600 elsif (ref $cond eq 'HASH') {
2601 if (keys %$cond and (keys %$cond)[0] eq '-and') {
2602 foreach my $subcond (@{$cond->{-and}}) {
2603 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2607 foreach my $col (keys %$cond) {
2608 my $value = $cond->{$col};
2609 $collapsed->{$col} = $value;
2619 # Remove the specified alias from the specified query hash. A copy is made so
2620 # the original query is not modified.
2623 my ($self, $query, $alias) = @_;
2625 my %orig = %{ $query || {} };
2628 foreach my $key (keys %orig) {
2630 $unaliased{$key} = $orig{$key};
2633 $unaliased{$1} = $orig{$key}
2634 if $key =~ m/^(?:\Q$alias\E\.)?([^.]+)$/;
2644 =item Arguments: none
2646 =item Return Value: \[ $sql, L<@bind_values|/DBIC BIND VALUES> ]
2650 Returns the SQL query and bind vars associated with the invocant.
2652 This is generally used as the RHS for a subquery.
2659 my $attrs = { %{ $self->_resolved_attrs } };
2661 my $aq = $self->result_source->storage->_select_args_to_query (
2662 $attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs
2665 $self->{_attrs}{_sqlmaker_select_args} = $attrs->{_sqlmaker_select_args};
2674 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2676 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2680 my $artist = $schema->resultset('Artist')->find_or_new(
2681 { artist => 'fred' }, { key => 'artists' });
2683 $cd->cd_to_producer->find_or_new({ producer => $producer },
2684 { key => 'primary' });
2686 Find an existing record from this resultset using L</find>. if none exists,
2687 instantiate a new result object and return it. The object will not be saved
2688 into your storage until you call L<DBIx::Class::Row/insert> on it.
2690 You most likely want this method when looking for existing rows using a unique
2691 constraint that is not the primary key, or looking for related rows.
2693 If you want objects to be saved immediately, use L</find_or_create> instead.
2695 B<Note>: Make sure to read the documentation of L</find> and understand the
2696 significance of the C<key> attribute, as its lack may skew your search, and
2697 subsequently result in spurious new objects.
2699 B<Note>: Take care when using C<find_or_new> with a table having
2700 columns with default values that you intend to be automatically
2701 supplied by the database (e.g. an auto_increment primary key column).
2702 In normal usage, the value of such columns should NOT be included at
2703 all in the call to C<find_or_new>, even when set to C<undef>.
2709 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2710 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2711 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2714 return $self->new_result($hash);
2721 =item Arguments: \%col_data
2723 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2727 Attempt to create a single new row or a row with multiple related rows
2728 in the table represented by the resultset (and related tables). This
2729 will not check for duplicate rows before inserting, use
2730 L</find_or_create> to do that.
2732 To create one row for this resultset, pass a hashref of key/value
2733 pairs representing the columns of the table and the values you wish to
2734 store. If the appropriate relationships are set up, foreign key fields
2735 can also be passed an object representing the foreign row, and the
2736 value will be set to its primary key.
2738 To create related objects, pass a hashref of related-object column values
2739 B<keyed on the relationship name>. If the relationship is of type C<multi>
2740 (L<DBIx::Class::Relationship/has_many>) - pass an arrayref of hashrefs.
2741 The process will correctly identify columns holding foreign keys, and will
2742 transparently populate them from the keys of the corresponding relation.
2743 This can be applied recursively, and will work correctly for a structure
2744 with an arbitrary depth and width, as long as the relationships actually
2745 exists and the correct column data has been supplied.
2747 Instead of hashrefs of plain related data (key/value pairs), you may
2748 also pass new or inserted objects. New objects (not inserted yet, see
2749 L</new_result>), will be inserted into their appropriate tables.
2751 Effectively a shortcut for C<< ->new_result(\%col_data)->insert >>.
2753 Example of creating a new row.
2755 $person_rs->create({
2756 name=>"Some Person",
2757 email=>"somebody@someplace.com"
2760 Example of creating a new row and also creating rows in a related C<has_many>
2761 or C<has_one> resultset. Note Arrayref.
2764 { artistid => 4, name => 'Manufactured Crap', cds => [
2765 { title => 'My First CD', year => 2006 },
2766 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2771 Example of creating a new row and also creating a row in a related
2772 C<belongs_to> resultset. Note Hashref.
2775 title=>"Music for Silly Walks",
2778 name=>"Silly Musician",
2786 When subclassing ResultSet never attempt to override this method. Since
2787 it is a simple shortcut for C<< $self->new_result($attrs)->insert >>, a
2788 lot of the internals simply never call it, so your override will be
2789 bypassed more often than not. Override either L<DBIx::Class::Row/new>
2790 or L<DBIx::Class::Row/insert> depending on how early in the
2791 L</create> process you need to intervene. See also warning pertaining to
2799 my ($self, $col_data) = @_;
2800 $self->throw_exception( "create needs a hashref" )
2801 unless ref $col_data eq 'HASH';
2802 return $self->new_result($col_data)->insert;
2805 =head2 find_or_create
2809 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2811 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2815 $cd->cd_to_producer->find_or_create({ producer => $producer },
2816 { key => 'primary' });
2818 Tries to find a record based on its primary key or unique constraints; if none
2819 is found, creates one and returns that instead.
2821 my $cd = $schema->resultset('CD')->find_or_create({
2823 artist => 'Massive Attack',
2824 title => 'Mezzanine',
2828 Also takes an optional C<key> attribute, to search by a specific key or unique
2829 constraint. For example:
2831 my $cd = $schema->resultset('CD')->find_or_create(
2833 artist => 'Massive Attack',
2834 title => 'Mezzanine',
2836 { key => 'cd_artist_title' }
2839 B<Note>: Make sure to read the documentation of L</find> and understand the
2840 significance of the C<key> attribute, as its lack may skew your search, and
2841 subsequently result in spurious row creation.
2843 B<Note>: Because find_or_create() reads from the database and then
2844 possibly inserts based on the result, this method is subject to a race
2845 condition. Another process could create a record in the table after
2846 the find has completed and before the create has started. To avoid
2847 this problem, use find_or_create() inside a transaction.
2849 B<Note>: Take care when using C<find_or_create> with a table having
2850 columns with default values that you intend to be automatically
2851 supplied by the database (e.g. an auto_increment primary key column).
2852 In normal usage, the value of such columns should NOT be included at
2853 all in the call to C<find_or_create>, even when set to C<undef>.
2855 See also L</find> and L</update_or_create>. For information on how to declare
2856 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2858 If you need to know if an existing row was found or a new one created use
2859 L</find_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
2860 to call L<DBIx::Class::Row/insert> to save the newly created row to the
2863 my $cd = $schema->resultset('CD')->find_or_new({
2865 artist => 'Massive Attack',
2866 title => 'Mezzanine',
2870 if( !$cd->in_storage ) {
2877 sub find_or_create {
2879 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2880 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2881 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2884 return $self->create($hash);
2887 =head2 update_or_create
2891 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2893 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2897 $resultset->update_or_create({ col => $val, ... });
2899 Like L</find_or_create>, but if a row is found it is immediately updated via
2900 C<< $found_row->update (\%col_data) >>.
2903 Takes an optional C<key> attribute to search on a specific unique constraint.
2906 # In your application
2907 my $cd = $schema->resultset('CD')->update_or_create(
2909 artist => 'Massive Attack',
2910 title => 'Mezzanine',
2913 { key => 'cd_artist_title' }
2916 $cd->cd_to_producer->update_or_create({
2917 producer => $producer,
2923 B<Note>: Make sure to read the documentation of L</find> and understand the
2924 significance of the C<key> attribute, as its lack may skew your search, and
2925 subsequently result in spurious row creation.
2927 B<Note>: Take care when using C<update_or_create> with a table having
2928 columns with default values that you intend to be automatically
2929 supplied by the database (e.g. an auto_increment primary key column).
2930 In normal usage, the value of such columns should NOT be included at
2931 all in the call to C<update_or_create>, even when set to C<undef>.
2933 See also L</find> and L</find_or_create>. For information on how to declare
2934 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2936 If you need to know if an existing row was updated or a new one created use
2937 L</update_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
2938 to call L<DBIx::Class::Row/insert> to save the newly created row to the
2943 sub update_or_create {
2945 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2946 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2948 my $row = $self->find($cond, $attrs);
2950 $row->update($cond);
2954 return $self->create($cond);
2957 =head2 update_or_new
2961 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2963 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2967 $resultset->update_or_new({ col => $val, ... });
2969 Like L</find_or_new> but if a row is found it is immediately updated via
2970 C<< $found_row->update (\%col_data) >>.
2974 # In your application
2975 my $cd = $schema->resultset('CD')->update_or_new(
2977 artist => 'Massive Attack',
2978 title => 'Mezzanine',
2981 { key => 'cd_artist_title' }
2984 if ($cd->in_storage) {
2985 # the cd was updated
2988 # the cd is not yet in the database, let's insert it
2992 B<Note>: Make sure to read the documentation of L</find> and understand the
2993 significance of the C<key> attribute, as its lack may skew your search, and
2994 subsequently result in spurious new objects.
2996 B<Note>: Take care when using C<update_or_new> with a table having
2997 columns with default values that you intend to be automatically
2998 supplied by the database (e.g. an auto_increment primary key column).
2999 In normal usage, the value of such columns should NOT be included at
3000 all in the call to C<update_or_new>, even when set to C<undef>.
3002 See also L</find>, L</find_or_create> and L</find_or_new>.
3008 my $attrs = ( @_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {} );
3009 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
3011 my $row = $self->find( $cond, $attrs );
3012 if ( defined $row ) {
3013 $row->update($cond);
3017 return $self->new_result($cond);
3024 =item Arguments: none
3026 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass> | undef
3030 Gets the contents of the cache for the resultset, if the cache is set.
3032 The cache is populated either by using the L</prefetch> attribute to
3033 L</search> or by calling L</set_cache>.
3045 =item Arguments: L<\@result_objs|DBIx::Class::Manual::ResultClass>
3047 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass>
3051 Sets the contents of the cache for the resultset. Expects an arrayref
3052 of objects of the same class as those produced by the resultset. Note that
3053 if the cache is set, the resultset will return the cached objects rather
3054 than re-querying the database even if the cache attr is not set.
3056 The contents of the cache can also be populated by using the
3057 L</prefetch> attribute to L</search>.
3062 my ( $self, $data ) = @_;
3063 $self->throw_exception("set_cache requires an arrayref")
3064 if defined($data) && (ref $data ne 'ARRAY');
3065 $self->{all_cache} = $data;
3072 =item Arguments: none
3074 =item Return Value: undef
3078 Clears the cache for the resultset.
3083 shift->set_cache(undef);
3090 =item Arguments: none
3092 =item Return Value: true, if the resultset has been paginated
3100 return !!$self->{attrs}{page};
3107 =item Arguments: none
3109 =item Return Value: true, if the resultset has been ordered with C<order_by>.
3117 return scalar $self->result_source->storage->_extract_order_criteria($self->{attrs}{order_by});
3120 =head2 related_resultset
3124 =item Arguments: $rel_name
3126 =item Return Value: L<$resultset|/search>
3130 Returns a related resultset for the supplied relationship name.
3132 $artist_rs = $schema->resultset('CD')->related_resultset('Artist');
3136 sub related_resultset {
3137 my ($self, $rel) = @_;
3139 return $self->{related_resultsets}{$rel}
3140 if defined $self->{related_resultsets}{$rel};
3142 return $self->{related_resultsets}{$rel} = do {
3143 my $rsrc = $self->result_source;
3144 my $rel_info = $rsrc->relationship_info($rel);
3146 $self->throw_exception(
3147 "search_related: result source '" . $rsrc->source_name .
3148 "' has no such relationship $rel")
3151 my $attrs = $self->_chain_relationship($rel);
3153 my $join_count = $attrs->{seen_join}{$rel};
3155 my $alias = $self->result_source->storage
3156 ->relname_to_table_alias($rel, $join_count);
3158 # since this is search_related, and we already slid the select window inwards
3159 # (the select/as attrs were deleted in the beginning), we need to flip all
3160 # left joins to inner, so we get the expected results
3161 # read the comment on top of the actual function to see what this does
3162 $attrs->{from} = $rsrc->schema->storage->_inner_join_to_node ($attrs->{from}, $alias);
3165 #XXX - temp fix for result_class bug. There likely is a more elegant fix -groditi
3166 delete @{$attrs}{qw(result_class alias)};
3168 my $rel_source = $rsrc->related_source($rel);
3172 # The reason we do this now instead of passing the alias to the
3173 # search_rs below is that if you wrap/overload resultset on the
3174 # source you need to know what alias it's -going- to have for things
3175 # to work sanely (e.g. RestrictWithObject wants to be able to add
3176 # extra query restrictions, and these may need to be $alias.)
3178 my $rel_attrs = $rel_source->resultset_attributes;
3179 local $rel_attrs->{alias} = $alias;
3181 $rel_source->resultset
3185 where => $attrs->{where},
3189 if (my $cache = $self->get_cache) {
3190 my @related_cache = map
3191 { @{$_->related_resultset($rel)->get_cache||[]} }
3195 $new->set_cache(\@related_cache) if @related_cache;
3202 =head2 current_source_alias
3206 =item Arguments: none
3208 =item Return Value: $source_alias
3212 Returns the current table alias for the result source this resultset is built
3213 on, that will be used in the SQL query. Usually it is C<me>.
3215 Currently the source alias that refers to the result set returned by a
3216 L</search>/L</find> family method depends on how you got to the resultset: it's
3217 C<me> by default, but eg. L</search_related> aliases it to the related result
3218 source name (and keeps C<me> referring to the original result set). The long
3219 term goal is to make L<DBIx::Class> always alias the current resultset as C<me>
3220 (and make this method unnecessary).
3222 Thus it's currently necessary to use this method in predefined queries (see
3223 L<DBIx::Class::Manual::Cookbook/Predefined searches>) when referring to the
3224 source alias of the current result set:
3226 # in a result set class
3228 my ($self, $user) = @_;
3230 my $me = $self->current_source_alias;
3232 return $self->search({
3233 "$me.modified" => $user->id,
3239 sub current_source_alias {
3240 return (shift->{attrs} || {})->{alias} || 'me';
3243 =head2 as_subselect_rs
3247 =item Arguments: none
3249 =item Return Value: L<$resultset|/search>
3253 Act as a barrier to SQL symbols. The resultset provided will be made into a
3254 "virtual view" by including it as a subquery within the from clause. From this
3255 point on, any joined tables are inaccessible to ->search on the resultset (as if
3256 it were simply where-filtered without joins). For example:
3258 my $rs = $schema->resultset('Bar')->search({'x.name' => 'abc'},{ join => 'x' });
3260 # 'x' now pollutes the query namespace
3262 # So the following works as expected
3263 my $ok_rs = $rs->search({'x.other' => 1});
3265 # But this doesn't: instead of finding a 'Bar' related to two x rows (abc and
3266 # def) we look for one row with contradictory terms and join in another table
3267 # (aliased 'x_2') which we never use
3268 my $broken_rs = $rs->search({'x.name' => 'def'});
3270 my $rs2 = $rs->as_subselect_rs;
3272 # doesn't work - 'x' is no longer accessible in $rs2, having been sealed away
3273 my $not_joined_rs = $rs2->search({'x.other' => 1});
3275 # works as expected: finds a 'table' row related to two x rows (abc and def)
3276 my $correctly_joined_rs = $rs2->search({'x.name' => 'def'});
3278 Another example of when one might use this would be to select a subset of
3279 columns in a group by clause:
3281 my $rs = $schema->resultset('Bar')->search(undef, {
3282 group_by => [qw{ id foo_id baz_id }],
3283 })->as_subselect_rs->search(undef, {
3284 columns => [qw{ id foo_id }]
3287 In the above example normally columns would have to be equal to the group by,
3288 but because we isolated the group by into a subselect the above works.
3292 sub as_subselect_rs {
3295 my $attrs = $self->_resolved_attrs;
3297 my $fresh_rs = (ref $self)->new (
3298 $self->result_source
3301 # these pieces will be locked in the subquery
3302 delete $fresh_rs->{cond};
3303 delete @{$fresh_rs->{attrs}}{qw/where bind/};
3305 return $fresh_rs->search( {}, {
3307 $attrs->{alias} => $self->as_query,
3308 -alias => $attrs->{alias},
3309 -rsrc => $self->result_source,
3311 alias => $attrs->{alias},
3315 # This code is called by search_related, and makes sure there
3316 # is clear separation between the joins before, during, and
3317 # after the relationship. This information is needed later
3318 # in order to properly resolve prefetch aliases (any alias
3319 # with a relation_chain_depth less than the depth of the
3320 # current prefetch is not considered)
3322 # The increments happen twice per join. An even number means a
3323 # relationship specified via a search_related, whereas an odd
3324 # number indicates a join/prefetch added via attributes
3326 # Also this code will wrap the current resultset (the one we
3327 # chain to) in a subselect IFF it contains limiting attributes
3328 sub _chain_relationship {
3329 my ($self, $rel) = @_;
3330 my $source = $self->result_source;
3331 my $attrs = { %{$self->{attrs}||{}} };
3333 # we need to take the prefetch the attrs into account before we
3334 # ->_resolve_join as otherwise they get lost - captainL
3335 my $join = $self->_merge_joinpref_attr( $attrs->{join}, $attrs->{prefetch} );
3337 delete @{$attrs}{qw/join prefetch collapse group_by distinct _grouped_by_distinct select as columns +select +as +columns/};
3339 my $seen = { %{ (delete $attrs->{seen_join}) || {} } };
3342 my @force_subq_attrs = qw/offset rows group_by having/;
3345 ($attrs->{from} && ref $attrs->{from} ne 'ARRAY')
3347 $self->_has_resolved_attr (@force_subq_attrs)
3349 # Nuke the prefetch (if any) before the new $rs attrs
3350 # are resolved (prefetch is useless - we are wrapping
3351 # a subquery anyway).
3352 my $rs_copy = $self->search;
3353 $rs_copy->{attrs}{join} = $self->_merge_joinpref_attr (
3354 $rs_copy->{attrs}{join},
3355 delete $rs_copy->{attrs}{prefetch},
3360 -alias => $attrs->{alias},
3361 $attrs->{alias} => $rs_copy->as_query,
3363 delete @{$attrs}{@force_subq_attrs, qw/where bind/};
3364 $seen->{-relation_chain_depth} = 0;
3366 elsif ($attrs->{from}) { #shallow copy suffices
3367 $from = [ @{$attrs->{from}} ];
3372 -alias => $attrs->{alias},
3373 $attrs->{alias} => $source->from,
3377 my $jpath = ($seen->{-relation_chain_depth})
3378 ? $from->[-1][0]{-join_path}
3381 my @requested_joins = $source->_resolve_join(
3388 push @$from, @requested_joins;
3390 $seen->{-relation_chain_depth}++;
3392 # if $self already had a join/prefetch specified on it, the requested
3393 # $rel might very well be already included. What we do in this case
3394 # is effectively a no-op (except that we bump up the chain_depth on
3395 # the join in question so we could tell it *is* the search_related)
3398 # we consider the last one thus reverse
3399 for my $j (reverse @requested_joins) {
3400 my ($last_j) = keys %{$j->[0]{-join_path}[-1]};
3401 if ($rel eq $last_j) {
3402 $j->[0]{-relation_chain_depth}++;
3408 unless ($already_joined) {
3409 push @$from, $source->_resolve_join(
3417 $seen->{-relation_chain_depth}++;
3419 return {%$attrs, from => $from, seen_join => $seen};
3422 sub _resolved_attrs {
3424 return $self->{_attrs} if $self->{_attrs};
3426 my $attrs = { %{ $self->{attrs} || {} } };
3427 my $source = $self->result_source;
3428 my $alias = $attrs->{alias};
3430 # default selection list
3431 $attrs->{columns} = [ $source->columns ]
3432 unless List::Util::first { exists $attrs->{$_} } qw/columns cols select as/;
3434 # merge selectors together
3435 for (qw/columns select as/) {
3436 $attrs->{$_} = $self->_merge_attr($attrs->{$_}, delete $attrs->{"+$_"})
3437 if $attrs->{$_} or $attrs->{"+$_"};
3440 # disassemble columns
3442 if (my $cols = delete $attrs->{columns}) {
3443 for my $c (ref $cols eq 'ARRAY' ? @$cols : $cols) {
3444 if (ref $c eq 'HASH') {
3445 for my $as (sort keys %$c) {
3446 push @sel, $c->{$as};
3457 # when trying to weed off duplicates later do not go past this point -
3458 # everything added from here on is unbalanced "anyone's guess" stuff
3459 my $dedup_stop_idx = $#as;
3461 push @as, @{ ref $attrs->{as} eq 'ARRAY' ? $attrs->{as} : [ $attrs->{as} ] }
3463 push @sel, @{ ref $attrs->{select} eq 'ARRAY' ? $attrs->{select} : [ $attrs->{select} ] }
3464 if $attrs->{select};
3466 # assume all unqualified selectors to apply to the current alias (legacy stuff)
3467 $_ = (ref $_ or $_ =~ /\./) ? $_ : "$alias.$_" for @sel;
3469 # disqualify all $alias.col as-bits (inflate-map mandated)
3470 $_ = ($_ =~ /^\Q$alias.\E(.+)$/) ? $1 : $_ for @as;
3472 # de-duplicate the result (remove *identical* select/as pairs)
3473 # and also die on duplicate {as} pointing to different {select}s
3474 # not using a c-style for as the condition is prone to shrinkage
3477 while ($i <= $dedup_stop_idx) {
3478 if ($seen->{"$sel[$i] \x00\x00 $as[$i]"}++) {
3483 elsif ($seen->{$as[$i]}++) {
3484 $self->throw_exception(
3485 "inflate_result() alias '$as[$i]' specified twice with different SQL-side {select}-ors"
3493 $attrs->{select} = \@sel;
3494 $attrs->{as} = \@as;
3496 $attrs->{from} ||= [{
3498 -alias => $self->{attrs}{alias},
3499 $self->{attrs}{alias} => $source->from,
3502 if ( $attrs->{join} || $attrs->{prefetch} ) {
3504 $self->throw_exception ('join/prefetch can not be used with a custom {from}')
3505 if ref $attrs->{from} ne 'ARRAY';
3507 my $join = (delete $attrs->{join}) || {};
3509 if ( defined $attrs->{prefetch} ) {
3510 $join = $self->_merge_joinpref_attr( $join, $attrs->{prefetch} );
3513 $attrs->{from} = # have to copy here to avoid corrupting the original
3515 @{ $attrs->{from} },
3516 $source->_resolve_join(
3519 { %{ $attrs->{seen_join} || {} } },
3520 ( $attrs->{seen_join} && keys %{$attrs->{seen_join}})
3521 ? $attrs->{from}[-1][0]{-join_path}
3528 if ( defined $attrs->{order_by} ) {
3529 $attrs->{order_by} = (
3530 ref( $attrs->{order_by} ) eq 'ARRAY'
3531 ? [ @{ $attrs->{order_by} } ]
3532 : [ $attrs->{order_by} || () ]
3536 if ($attrs->{group_by} and ref $attrs->{group_by} ne 'ARRAY') {
3537 $attrs->{group_by} = [ $attrs->{group_by} ];
3540 # generate the distinct induced group_by early, as prefetch will be carried via a
3541 # subquery (since a group_by is present)
3542 if (delete $attrs->{distinct}) {
3543 if ($attrs->{group_by}) {
3544 carp_unique ("Useless use of distinct on a grouped resultset ('distinct' is ignored when a 'group_by' is present)");
3547 $attrs->{_grouped_by_distinct} = 1;
3548 # distinct affects only the main selection part, not what prefetch may
3550 $attrs->{group_by} = $source->storage->_group_over_selection($attrs);
3554 # generate selections based on the prefetch helper
3556 $prefetch = $self->_merge_joinpref_attr( {}, delete $attrs->{prefetch} )
3557 if defined $attrs->{prefetch};
3561 $self->throw_exception("Unable to prefetch, resultset contains an unnamed selector $attrs->{_dark_selector}{string}")
3562 if $attrs->{_dark_selector};
3564 $attrs->{collapse} = 1;
3566 # this is a separate structure (we don't look in {from} directly)
3567 # as the resolver needs to shift things off the lists to work
3568 # properly (identical-prefetches on different branches)
3570 if (ref $attrs->{from} eq 'ARRAY') {
3572 my $start_depth = $attrs->{seen_join}{-relation_chain_depth} || 0;
3574 for my $j ( @{$attrs->{from}}[1 .. $#{$attrs->{from}} ] ) {
3575 next unless $j->[0]{-alias};
3576 next unless $j->[0]{-join_path};
3577 next if ($j->[0]{-relation_chain_depth} || 0) < $start_depth;
3579 my @jpath = map { keys %$_ } @{$j->[0]{-join_path}};
3582 $p = $p->{$_} ||= {} for @jpath[ ($start_depth/2) .. $#jpath]; #only even depths are actual jpath boundaries
3583 push @{$p->{-join_aliases} }, $j->[0]{-alias};
3587 my @prefetch = $source->_resolve_prefetch( $prefetch, $alias, $join_map );
3589 push @{ $attrs->{select} }, (map { $_->[0] } @prefetch);
3590 push @{ $attrs->{as} }, (map { $_->[1] } @prefetch);
3593 if ( List::Util::first { $_ =~ /\./ } @{$attrs->{as}} ) {
3594 $attrs->{_related_results_construction} = 1;
3597 # run through the resulting joinstructure (starting from our current slot)
3598 # and unset collapse if proven unnecessary
3600 # also while we are at it find out if the current root source has
3601 # been premultiplied by previous related_source chaining
3603 # this allows to predict whether a root object with all other relation
3604 # data set to NULL is in fact unique
3605 if ($attrs->{collapse}) {
3607 if (ref $attrs->{from} eq 'ARRAY') {
3609 if (@{$attrs->{from}} == 1) {
3610 # no joins - no collapse
3611 $attrs->{collapse} = 0;
3614 # find where our table-spec starts
3615 my @fromlist = @{$attrs->{from}};
3617 my $t = shift @fromlist;
3620 # me vs join from-spec distinction - a ref means non-root
3621 if (ref $t eq 'ARRAY') {
3623 $is_multi ||= ! $t->{-is_single};
3625 last if ($t->{-alias} && $t->{-alias} eq $alias);
3626 $attrs->{_main_source_premultiplied} ||= $is_multi;
3629 # no non-singles remaining, nor any premultiplication - nothing to collapse
3631 ! $attrs->{_main_source_premultiplied}
3633 ! List::Util::first { ! $_->[0]{-is_single} } @fromlist
3635 $attrs->{collapse} = 0;
3641 # if we can not analyze the from - err on the side of safety
3642 $attrs->{_main_source_premultiplied} = 1;
3646 # if both page and offset are specified, produce a combined offset
3647 # even though it doesn't make much sense, this is what pre 081xx has
3649 if (my $page = delete $attrs->{page}) {
3651 ($attrs->{rows} * ($page - 1))
3653 ($attrs->{offset} || 0)
3657 return $self->{_attrs} = $attrs;
3661 my ($self, $attr) = @_;
3663 if (ref $attr eq 'HASH') {
3664 return $self->_rollout_hash($attr);
3665 } elsif (ref $attr eq 'ARRAY') {
3666 return $self->_rollout_array($attr);
3672 sub _rollout_array {
3673 my ($self, $attr) = @_;
3676 foreach my $element (@{$attr}) {
3677 if (ref $element eq 'HASH') {
3678 push( @rolled_array, @{ $self->_rollout_hash( $element ) } );
3679 } elsif (ref $element eq 'ARRAY') {
3680 # XXX - should probably recurse here
3681 push( @rolled_array, @{$self->_rollout_array($element)} );
3683 push( @rolled_array, $element );
3686 return \@rolled_array;
3690 my ($self, $attr) = @_;
3693 foreach my $key (keys %{$attr}) {
3694 push( @rolled_array, { $key => $attr->{$key} } );
3696 return \@rolled_array;
3699 sub _calculate_score {
3700 my ($self, $a, $b) = @_;
3702 if (defined $a xor defined $b) {
3705 elsif (not defined $a) {
3709 if (ref $b eq 'HASH') {
3710 my ($b_key) = keys %{$b};
3711 if (ref $a eq 'HASH') {
3712 my ($a_key) = keys %{$a};
3713 if ($a_key eq $b_key) {
3714 return (1 + $self->_calculate_score( $a->{$a_key}, $b->{$b_key} ));
3719 return ($a eq $b_key) ? 1 : 0;
3722 if (ref $a eq 'HASH') {
3723 my ($a_key) = keys %{$a};
3724 return ($b eq $a_key) ? 1 : 0;
3726 return ($b eq $a) ? 1 : 0;
3731 sub _merge_joinpref_attr {
3732 my ($self, $orig, $import) = @_;
3734 return $import unless defined($orig);
3735 return $orig unless defined($import);
3737 $orig = $self->_rollout_attr($orig);
3738 $import = $self->_rollout_attr($import);
3741 foreach my $import_element ( @{$import} ) {
3742 # find best candidate from $orig to merge $b_element into
3743 my $best_candidate = { position => undef, score => 0 }; my $position = 0;
3744 foreach my $orig_element ( @{$orig} ) {
3745 my $score = $self->_calculate_score( $orig_element, $import_element );
3746 if ($score > $best_candidate->{score}) {
3747 $best_candidate->{position} = $position;
3748 $best_candidate->{score} = $score;
3752 my ($import_key) = ( ref $import_element eq 'HASH' ) ? keys %{$import_element} : ($import_element);
3753 $import_key = '' if not defined $import_key;
3755 if ($best_candidate->{score} == 0 || exists $seen_keys->{$import_key}) {
3756 push( @{$orig}, $import_element );
3758 my $orig_best = $orig->[$best_candidate->{position}];
3759 # merge orig_best and b_element together and replace original with merged
3760 if (ref $orig_best ne 'HASH') {
3761 $orig->[$best_candidate->{position}] = $import_element;
3762 } elsif (ref $import_element eq 'HASH') {
3763 my ($key) = keys %{$orig_best};
3764 $orig->[$best_candidate->{position}] = { $key => $self->_merge_joinpref_attr($orig_best->{$key}, $import_element->{$key}) };
3767 $seen_keys->{$import_key} = 1; # don't merge the same key twice
3770 return @$orig ? $orig : ();
3778 require Hash::Merge;
3779 my $hm = Hash::Merge->new;
3781 $hm->specify_behavior({
3784 my ($defl, $defr) = map { defined $_ } (@_[0,1]);
3786 if ($defl xor $defr) {
3787 return [ $defl ? $_[0] : $_[1] ];
3792 elsif (__HM_DEDUP and $_[0] eq $_[1]) {
3796 return [$_[0], $_[1]];
3800 return $_[1] if !defined $_[0];
3801 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3802 return [$_[0], @{$_[1]}]
3805 return [] if !defined $_[0] and !keys %{$_[1]};
3806 return [ $_[1] ] if !defined $_[0];
3807 return [ $_[0] ] if !keys %{$_[1]};
3808 return [$_[0], $_[1]]
3813 return $_[0] if !defined $_[1];
3814 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3815 return [@{$_[0]}, $_[1]]
3818 my @ret = @{$_[0]} or return $_[1];
3819 return [ @ret, @{$_[1]} ] unless __HM_DEDUP;
3820 my %idx = map { $_ => 1 } @ret;
3821 push @ret, grep { ! defined $idx{$_} } (@{$_[1]});
3825 return [ $_[1] ] if ! @{$_[0]};
3826 return $_[0] if !keys %{$_[1]};
3827 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3828 return [ @{$_[0]}, $_[1] ];
3833 return [] if !keys %{$_[0]} and !defined $_[1];
3834 return [ $_[0] ] if !defined $_[1];
3835 return [ $_[1] ] if !keys %{$_[0]};
3836 return [$_[0], $_[1]]
3839 return [] if !keys %{$_[0]} and !@{$_[1]};
3840 return [ $_[0] ] if !@{$_[1]};
3841 return $_[1] if !keys %{$_[0]};
3842 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3843 return [ $_[0], @{$_[1]} ];
3846 return [] if !keys %{$_[0]} and !keys %{$_[1]};
3847 return [ $_[0] ] if !keys %{$_[1]};
3848 return [ $_[1] ] if !keys %{$_[0]};
3849 return [ $_[0] ] if $_[0] eq $_[1];
3850 return [ $_[0], $_[1] ];
3853 } => 'DBIC_RS_ATTR_MERGER');
3857 return $hm->merge ($_[1], $_[2]);
3861 sub STORABLE_freeze {
3862 my ($self, $cloning) = @_;
3863 my $to_serialize = { %$self };
3865 # A cursor in progress can't be serialized (and would make little sense anyway)
3866 # the parser can be regenerated (and can't be serialized)
3867 delete @{$to_serialize}{qw/cursor _row_parser _result_inflator/};
3869 # nor is it sensical to store a not-yet-fired-count pager
3870 if ($to_serialize->{pager} and ref $to_serialize->{pager}{total_entries} eq 'CODE') {
3871 delete $to_serialize->{pager};
3874 Storable::nfreeze($to_serialize);
3877 # need this hook for symmetry
3879 my ($self, $cloning, $serialized) = @_;
3881 %$self = %{ Storable::thaw($serialized) };
3887 =head2 throw_exception
3889 See L<DBIx::Class::Schema/throw_exception> for details.
3893 sub throw_exception {
3896 if (ref $self and my $rsrc = $self->result_source) {
3897 $rsrc->throw_exception(@_)
3900 DBIx::Class::Exception->throw(@_);
3908 # XXX: FIXME: Attributes docs need clearing up
3912 Attributes are used to refine a ResultSet in various ways when
3913 searching for data. They can be passed to any method which takes an
3914 C<\%attrs> argument. See L</search>, L</search_rs>, L</find>,
3917 Default attributes can be set on the result class using
3918 L<DBIx::Class::ResultSource/resultset_attributes>. (Please read
3919 the CAVEATS on that feature before using it!)
3921 These are in no particular order:
3927 =item Value: ( $order_by | \@order_by | \%order_by )
3931 Which column(s) to order the results by.
3933 [The full list of suitable values is documented in
3934 L<SQL::Abstract/"ORDER BY CLAUSES">; the following is a summary of
3937 If a single column name, or an arrayref of names is supplied, the
3938 argument is passed through directly to SQL. The hashref syntax allows
3939 for connection-agnostic specification of ordering direction:
3941 For descending order:
3943 order_by => { -desc => [qw/col1 col2 col3/] }
3945 For explicit ascending order:
3947 order_by => { -asc => 'col' }
3949 The old scalarref syntax (i.e. order_by => \'year DESC') is still
3950 supported, although you are strongly encouraged to use the hashref
3951 syntax as outlined above.
3957 =item Value: \@columns | \%columns | $column
3961 Shortcut to request a particular set of columns to be retrieved. Each
3962 column spec may be a string (a table column name), or a hash (in which
3963 case the key is the C<as> value, and the value is used as the C<select>
3964 expression). Adds C<me.> onto the start of any column without a C<.> in
3965 it and sets C<select> from that, then auto-populates C<as> from
3966 C<select> as normal. (You may also use the C<cols> attribute, as in
3967 earlier versions of DBIC, but this is deprecated.)
3969 Essentially C<columns> does the same as L</select> and L</as>.
3971 columns => [ 'foo', { bar => 'baz' } ]
3975 select => [qw/foo baz/],
3982 =item Value: \@columns
3986 Indicates additional columns to be selected from storage. Works the same as
3987 L</columns> but adds columns to the selection. (You may also use the
3988 C<include_columns> attribute, as in earlier versions of DBIC, but this is
3989 deprecated). For example:-
3991 $schema->resultset('CD')->search(undef, {
3992 '+columns' => ['artist.name'],
3996 would return all CDs and include a 'name' column to the information
3997 passed to object inflation. Note that the 'artist' is the name of the
3998 column (or relationship) accessor, and 'name' is the name of the column
3999 accessor in the related table.
4001 B<NOTE:> You need to explicitly quote '+columns' when defining the attribute.
4002 Not doing so causes Perl to incorrectly interpret +columns as a bareword with a
4003 unary plus operator before it.
4005 =head2 include_columns
4009 =item Value: \@columns
4013 Deprecated. Acts as a synonym for L</+columns> for backward compatibility.
4019 =item Value: \@select_columns
4023 Indicates which columns should be selected from the storage. You can use
4024 column names, or in the case of RDBMS back ends, function or stored procedure
4027 $rs = $schema->resultset('Employee')->search(undef, {
4030 { count => 'employeeid' },
4031 { max => { length => 'name' }, -as => 'longest_name' }
4036 SELECT name, COUNT( employeeid ), MAX( LENGTH( name ) ) AS longest_name FROM employee
4038 B<NOTE:> You will almost always need a corresponding L</as> attribute when you
4039 use L</select>, to instruct DBIx::Class how to store the result of the column.
4040 Also note that the L</as> attribute has nothing to do with the SQL-side 'AS'
4041 identifier aliasing. You can however alias a function, so you can use it in
4042 e.g. an C<ORDER BY> clause. This is done via the C<-as> B<select function
4043 attribute> supplied as shown in the example above.
4045 B<NOTE:> You need to explicitly quote '+select'/'+as' when defining the attributes.
4046 Not doing so causes Perl to incorrectly interpret them as a bareword with a
4047 unary plus operator before it.
4053 Indicates additional columns to be selected from storage. Works the same as
4054 L</select> but adds columns to the default selection, instead of specifying
4063 =item Value: \@inflation_names
4067 Indicates column names for object inflation. That is L</as> indicates the
4068 slot name in which the column value will be stored within the
4069 L<Row|DBIx::Class::Row> object. The value will then be accessible via this
4070 identifier by the C<get_column> method (or via the object accessor B<if one
4071 with the same name already exists>) as shown below. The L</as> attribute has
4072 B<nothing to do> with the SQL-side C<AS>. See L</select> for details.
4074 $rs = $schema->resultset('Employee')->search(undef, {
4077 { count => 'employeeid' },
4078 { max => { length => 'name' }, -as => 'longest_name' }
4087 If the object against which the search is performed already has an accessor
4088 matching a column name specified in C<as>, the value can be retrieved using
4089 the accessor as normal:
4091 my $name = $employee->name();
4093 If on the other hand an accessor does not exist in the object, you need to
4094 use C<get_column> instead:
4096 my $employee_count = $employee->get_column('employee_count');
4098 You can create your own accessors if required - see
4099 L<DBIx::Class::Manual::Cookbook> for details.
4105 Indicates additional column names for those added via L</+select>. See L</as>.
4113 =item Value: ($rel_name | \@rel_names | \%rel_names)
4117 Contains a list of relationships that should be joined for this query. For
4120 # Get CDs by Nine Inch Nails
4121 my $rs = $schema->resultset('CD')->search(
4122 { 'artist.name' => 'Nine Inch Nails' },
4123 { join => 'artist' }
4126 Can also contain a hash reference to refer to the other relation's relations.
4129 package MyApp::Schema::Track;
4130 use base qw/DBIx::Class/;
4131 __PACKAGE__->table('track');
4132 __PACKAGE__->add_columns(qw/trackid cd position title/);
4133 __PACKAGE__->set_primary_key('trackid');
4134 __PACKAGE__->belongs_to(cd => 'MyApp::Schema::CD');
4137 # In your application
4138 my $rs = $schema->resultset('Artist')->search(
4139 { 'track.title' => 'Teardrop' },
4141 join => { cd => 'track' },
4142 order_by => 'artist.name',
4146 You need to use the relationship (not the table) name in conditions,
4147 because they are aliased as such. The current table is aliased as "me", so
4148 you need to use me.column_name in order to avoid ambiguity. For example:
4150 # Get CDs from 1984 with a 'Foo' track
4151 my $rs = $schema->resultset('CD')->search(
4154 'tracks.name' => 'Foo'
4156 { join => 'tracks' }
4159 If the same join is supplied twice, it will be aliased to <rel>_2 (and
4160 similarly for a third time). For e.g.
4162 my $rs = $schema->resultset('Artist')->search({
4163 'cds.title' => 'Down to Earth',
4164 'cds_2.title' => 'Popular',
4166 join => [ qw/cds cds/ ],
4169 will return a set of all artists that have both a cd with title 'Down
4170 to Earth' and a cd with title 'Popular'.
4172 If you want to fetch related objects from other tables as well, see L</prefetch>
4175 NOTE: An internal join-chain pruner will discard certain joins while
4176 constructing the actual SQL query, as long as the joins in question do not
4177 affect the retrieved result. This for example includes 1:1 left joins
4178 that are not part of the restriction specification (WHERE/HAVING) nor are
4179 a part of the query selection.
4181 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
4187 =item Value: (0 | 1)
4191 When set to a true value, indicates that any rows fetched from joined has_many
4192 relationships are to be aggregated into the corresponding "parent" object. For
4193 example, the resultset:
4195 my $rs = $schema->resultset('CD')->search({}, {
4196 '+columns' => [ qw/ tracks.title tracks.position / ],
4201 While executing the following query:
4203 SELECT me.*, tracks.title, tracks.position
4205 LEFT JOIN track tracks
4206 ON tracks.cdid = me.cdid
4208 Will return only as many objects as there are rows in the CD source, even
4209 though the result of the query may span many rows. Each of these CD objects
4210 will in turn have multiple "Track" objects hidden behind the has_many
4211 generated accessor C<tracks>. Without C<< collapse => 1 >>, the return values
4212 of this resultset would be as many CD objects as there are tracks (a "Cartesian
4213 product"), with each CD object containing exactly one of all fetched Track data.
4215 When a collapse is requested on a non-ordered resultset, an order by some
4216 unique part of the main source (the left-most table) is inserted automatically.
4217 This is done so that the resultset is allowed to be "lazy" - calling
4218 L<< $rs->next|/next >> will fetch only as many rows as it needs to build the next
4219 object with all of its related data.
4221 If an L</order_by> is already declared, and orders the resultset in a way that
4222 makes collapsing as described above impossible (e.g. C<< ORDER BY
4223 has_many_rel.column >> or C<ORDER BY RANDOM()>), DBIC will automatically
4224 switch to "eager" mode and slurp the entire resultset before constructing the
4225 first object returned by L</next>.
4227 Setting this attribute on a resultset that does not join any has_many
4228 relations is a no-op.
4230 For a more in-depth discussion, see L</PREFETCHING>.
4236 =item Value: ($rel_name | \@rel_names | \%rel_names)
4240 This attribute is a shorthand for specifying a L</join> spec, adding all
4241 columns from the joined related sources as L</+columns> and setting
4242 L</collapse> to a true value. For example, the following two queries are
4245 my $rs = $schema->resultset('Artist')->search({}, {
4246 prefetch => { cds => ['genre', 'tracks' ] },
4251 my $rs = $schema->resultset('Artist')->search({}, {
4252 join => { cds => ['genre', 'tracks' ] },
4256 { +{ "cds.$_" => "cds.$_" } }
4257 $schema->source('Artist')->related_source('cds')->columns
4260 { +{ "cds.genre.$_" => "genre.$_" } }
4261 $schema->source('Artist')->related_source('cds')->related_source('genre')->columns
4264 { +{ "cds.tracks.$_" => "tracks.$_" } }
4265 $schema->source('Artist')->related_source('cds')->related_source('tracks')->columns
4270 Both producing the following SQL:
4272 SELECT me.artistid, me.name, me.rank, me.charfield,
4273 cds.cdid, cds.artist, cds.title, cds.year, cds.genreid, cds.single_track,
4274 genre.genreid, genre.name,
4275 tracks.trackid, tracks.cd, tracks.position, tracks.title, tracks.last_updated_on, tracks.last_updated_at
4278 ON cds.artist = me.artistid
4279 LEFT JOIN genre genre
4280 ON genre.genreid = cds.genreid
4281 LEFT JOIN track tracks
4282 ON tracks.cd = cds.cdid
4283 ORDER BY me.artistid
4285 While L</prefetch> implies a L</join>, it is ok to mix the two together, as
4286 the arguments are properly merged and generally do the right thing. For
4287 example, you may want to do the following:
4289 my $artists_and_cds_without_genre = $schema->resultset('Artist')->search(
4290 { 'genre.genreid' => undef },
4292 join => { cds => 'genre' },
4297 Which generates the following SQL:
4299 SELECT me.artistid, me.name, me.rank, me.charfield,
4300 cds.cdid, cds.artist, cds.title, cds.year, cds.genreid, cds.single_track
4303 ON cds.artist = me.artistid
4304 LEFT JOIN genre genre
4305 ON genre.genreid = cds.genreid
4306 WHERE genre.genreid IS NULL
4307 ORDER BY me.artistid
4309 For a more in-depth discussion, see L</PREFETCHING>.
4315 =item Value: $source_alias
4319 Sets the source alias for the query. Normally, this defaults to C<me>, but
4320 nested search queries (sub-SELECTs) might need specific aliases set to
4321 reference inner queries. For example:
4324 ->related_resultset('CDs')
4325 ->related_resultset('Tracks')
4327 'track.id' => { -ident => 'none_search.id' },
4331 my $ids = $self->search({
4334 alias => 'none_search',
4335 group_by => 'none_search.id',
4336 })->get_column('id')->as_query;
4338 $self->search({ id => { -in => $ids } })
4340 This attribute is directly tied to L</current_source_alias>.
4350 Makes the resultset paged and specifies the page to retrieve. Effectively
4351 identical to creating a non-pages resultset and then calling ->page($page)
4354 If L</rows> attribute is not specified it defaults to 10 rows per page.
4356 When you have a paged resultset, L</count> will only return the number
4357 of rows in the page. To get the total, use the L</pager> and call
4358 C<total_entries> on it.
4368 Specifies the maximum number of rows for direct retrieval or the number of
4369 rows per page if the page attribute or method is used.
4375 =item Value: $offset
4379 Specifies the (zero-based) row number for the first row to be returned, or the
4380 of the first row of the first page if paging is used.
4382 =head2 software_limit
4386 =item Value: (0 | 1)
4390 When combined with L</rows> and/or L</offset> the generated SQL will not
4391 include any limit dialect stanzas. Instead the entire result will be selected
4392 as if no limits were specified, and DBIC will perform the limit locally, by
4393 artificially advancing and finishing the resulting L</cursor>.
4395 This is the recommended way of performing resultset limiting when no sane RDBMS
4396 implementation is available (e.g.
4397 L<Sybase ASE|DBIx::Class::Storage::DBI::Sybase::ASE> using the
4398 L<Generic Sub Query|DBIx::Class::SQLMaker::LimitDialects/GenericSubQ> hack)
4404 =item Value: \@columns
4408 A arrayref of columns to group by. Can include columns of joined tables.
4410 group_by => [qw/ column1 column2 ... /]
4416 =item Value: $condition
4420 HAVING is a select statement attribute that is applied between GROUP BY and
4421 ORDER BY. It is applied to the after the grouping calculations have been
4424 having => { 'count_employee' => { '>=', 100 } }
4426 or with an in-place function in which case literal SQL is required:
4428 having => \[ 'count(employee) >= ?', [ count => 100 ] ]
4434 =item Value: (0 | 1)
4438 Set to 1 to group by all columns. If the resultset already has a group_by
4439 attribute, this setting is ignored and an appropriate warning is issued.
4445 Adds to the WHERE clause.
4447 # only return rows WHERE deleted IS NULL for all searches
4448 __PACKAGE__->resultset_attributes({ where => { deleted => undef } });
4450 Can be overridden by passing C<< { where => undef } >> as an attribute
4453 For more complicated where clauses see L<SQL::Abstract/WHERE CLAUSES>.
4459 Set to 1 to cache search results. This prevents extra SQL queries if you
4460 revisit rows in your ResultSet:
4462 my $resultset = $schema->resultset('Artist')->search( undef, { cache => 1 } );
4464 while( my $artist = $resultset->next ) {
4468 $rs->first; # without cache, this would issue a query
4470 By default, searches are not cached.
4472 For more examples of using these attributes, see
4473 L<DBIx::Class::Manual::Cookbook>.
4479 =item Value: ( 'update' | 'shared' | \$scalar )
4483 Set to 'update' for a SELECT ... FOR UPDATE or 'shared' for a SELECT
4484 ... FOR SHARED. If \$scalar is passed, this is taken directly and embedded in the
4489 DBIx::Class supports arbitrary related data prefetching from multiple related
4490 sources. Any combination of relationship types and column sets are supported.
4491 If L<collapsing|/collapse> is requested, there is an additional requirement of
4492 selecting enough data to make every individual object uniquely identifiable.
4494 Here are some more involved examples, based on the following relationship map:
4497 My::Schema::CD->belongs_to( artist => 'My::Schema::Artist' );
4498 My::Schema::CD->might_have( liner_note => 'My::Schema::LinerNotes' );
4499 My::Schema::CD->has_many( tracks => 'My::Schema::Track' );
4501 My::Schema::Artist->belongs_to( record_label => 'My::Schema::RecordLabel' );
4503 My::Schema::Track->has_many( guests => 'My::Schema::Guest' );
4507 my $rs = $schema->resultset('Tag')->search(
4516 The initial search results in SQL like the following:
4518 SELECT tag.*, cd.*, artist.* FROM tag
4519 JOIN cd ON tag.cd = cd.cdid
4520 JOIN artist ON cd.artist = artist.artistid
4522 L<DBIx::Class> has no need to go back to the database when we access the
4523 C<cd> or C<artist> relationships, which saves us two SQL statements in this
4526 Simple prefetches will be joined automatically, so there is no need
4527 for a C<join> attribute in the above search.
4529 The L</prefetch> attribute can be used with any of the relationship types
4530 and multiple prefetches can be specified together. Below is a more complex
4531 example that prefetches a CD's artist, its liner notes (if present),
4532 the cover image, the tracks on that CD, and the guests on those
4535 my $rs = $schema->resultset('CD')->search(
4539 { artist => 'record_label'}, # belongs_to => belongs_to
4540 'liner_note', # might_have
4541 'cover_image', # has_one
4542 { tracks => 'guests' }, # has_many => has_many
4547 This will produce SQL like the following:
4549 SELECT cd.*, artist.*, record_label.*, liner_note.*, cover_image.*,
4553 ON artist.artistid = me.artistid
4554 JOIN record_label record_label
4555 ON record_label.labelid = artist.labelid
4556 LEFT JOIN track tracks
4557 ON tracks.cdid = me.cdid
4558 LEFT JOIN guest guests
4559 ON guests.trackid = track.trackid
4560 LEFT JOIN liner_notes liner_note
4561 ON liner_note.cdid = me.cdid
4562 JOIN cd_artwork cover_image
4563 ON cover_image.cdid = me.cdid
4566 Now the C<artist>, C<record_label>, C<liner_note>, C<cover_image>,
4567 C<tracks>, and C<guests> of the CD will all be available through the
4568 relationship accessors without the need for additional queries to the
4573 Prefetch does a lot of deep magic. As such, it may not behave exactly
4574 as you might expect.
4580 Prefetch uses the L</cache> to populate the prefetched relationships. This
4581 may or may not be what you want.
4585 If you specify a condition on a prefetched relationship, ONLY those
4586 rows that match the prefetched condition will be fetched into that relationship.
4587 This means that adding prefetch to a search() B<may alter> what is returned by
4588 traversing a relationship. So, if you have C<< Artist->has_many(CDs) >> and you do
4590 my $artist_rs = $schema->resultset('Artist')->search({
4596 my $count = $artist_rs->first->cds->count;
4598 my $artist_rs_prefetch = $artist_rs->search( {}, { prefetch => 'cds' } );
4600 my $prefetch_count = $artist_rs_prefetch->first->cds->count;
4602 cmp_ok( $count, '==', $prefetch_count, "Counts should be the same" );
4604 That cmp_ok() may or may not pass depending on the datasets involved. In other
4605 words the C<WHERE> condition would apply to the entire dataset, just like
4606 it would in regular SQL. If you want to add a condition only to the "right side"
4607 of a C<LEFT JOIN> - consider declaring and using a L<relationship with a custom
4608 condition|DBIx::Class::Relationship::Base/condition>
4612 =head1 DBIC BIND VALUES
4614 Because DBIC may need more information to bind values than just the column name
4615 and value itself, it uses a special format for both passing and receiving bind
4616 values. Each bind value should be composed of an arrayref of
4617 C<< [ \%args => $val ] >>. The format of C<< \%args >> is currently:
4623 If present (in any form), this is what is being passed directly to bind_param.
4624 Note that different DBD's expect different bind args. (e.g. DBD::SQLite takes
4625 a single numerical type, while DBD::Pg takes a hashref if bind options.)
4627 If this is specified, all other bind options described below are ignored.
4631 If present, this is used to infer the actual bind attribute by passing to
4632 C<< $resolved_storage->bind_attribute_by_data_type() >>. Defaults to the
4633 "data_type" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>.
4635 Note that the data type is somewhat freeform (hence the sqlt_ prefix);
4636 currently drivers are expected to "Do the Right Thing" when given a common
4637 datatype name. (Not ideal, but that's what we got at this point.)
4641 Currently used to correctly allocate buffers for bind_param_inout().
4642 Defaults to "size" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>,
4643 or to a sensible value based on the "data_type".
4647 Used to fill in missing sqlt_datatype and sqlt_size attributes (if they are
4648 explicitly specified they are never overridden). Also used by some weird DBDs,
4649 where the column name should be available at bind_param time (e.g. Oracle).
4653 For backwards compatibility and convenience, the following shortcuts are
4656 [ $name => $val ] === [ { dbic_colname => $name }, $val ]
4657 [ \$dt => $val ] === [ { sqlt_datatype => $dt }, $val ]
4658 [ undef, $val ] === [ {}, $val ]
4659 $val === [ {}, $val ]
4661 =head1 AUTHOR AND CONTRIBUTORS
4663 See L<AUTHOR|DBIx::Class/AUTHOR> and L<CONTRIBUTORS|DBIx::Class/CONTRIBUTORS> in DBIx::Class
4667 You may distribute this code under the same terms as Perl itself.