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
11 use Data::Query::Constants;
12 use Data::Query::ExprHelpers;
13 # not importing first() as it will clash with our own method
17 # De-duplication in _merge_attr() is disabled, but left in for reference
18 # (the merger is used for other things that ought not to be de-duped)
19 *__HM_DEDUP = sub () { 0 };
29 # this is real - CDBICompat overrides it with insanity
30 # yes, prototype won't matter, but that's for now ;)
33 __PACKAGE__->mk_group_accessors('simple' => qw/_result_class result_source/);
37 DBIx::Class::ResultSet - Represents a query used for fetching a set of results.
41 my $users_rs = $schema->resultset('User');
42 while( $user = $users_rs->next) {
43 print $user->username;
46 my $registered_users_rs = $schema->resultset('User')->search({ registered => 1 });
47 my @cds_in_2005 = $schema->resultset('CD')->search({ year => 2005 })->all();
51 A ResultSet is an object which stores a set of conditions representing
52 a query. It is the backbone of DBIx::Class (i.e. the really
53 important/useful bit).
55 No SQL is executed on the database when a ResultSet is created, it
56 just stores all the conditions needed to create the query.
58 A basic ResultSet representing the data of an entire table is returned
59 by calling C<resultset> on a L<DBIx::Class::Schema> and passing in a
60 L<Source|DBIx::Class::Manual::Glossary/Source> name.
62 my $users_rs = $schema->resultset('User');
64 A new ResultSet is returned from calling L</search> on an existing
65 ResultSet. The new one will contain all the conditions of the
66 original, plus any new conditions added in the C<search> call.
68 A ResultSet also incorporates an implicit iterator. L</next> and L</reset>
69 can be used to walk through all the L<DBIx::Class::Row>s the ResultSet
72 The query that the ResultSet represents is B<only> executed against
73 the database when these methods are called:
74 L</find>, L</next>, L</all>, L</first>, L</single>, L</count>.
76 If a resultset is used in a numeric context it returns the L</count>.
77 However, if it is used in a boolean context it is B<always> true. So if
78 you want to check if a resultset has any results, you must use C<if $rs
81 =head1 CUSTOM ResultSet CLASSES THAT USE Moose
83 If you want to make your custom ResultSet classes with L<Moose>, use a template
86 package MyApp::Schema::ResultSet::User;
89 use namespace::autoclean;
91 extends 'DBIx::Class::ResultSet';
93 sub BUILDARGS { $_[2] }
97 __PACKAGE__->meta->make_immutable;
101 The L<MooseX::NonMoose> is necessary so that the L<Moose> constructor does not
102 clash with the regular ResultSet constructor. Alternatively, you can use:
104 __PACKAGE__->meta->make_immutable(inline_constructor => 0);
106 The L<BUILDARGS|Moose::Manual::Construction/BUILDARGS> is necessary because the
107 signature of the ResultSet C<new> is C<< ->new($source, \%args) >>.
111 =head2 Chaining resultsets
113 Let's say you've got a query that needs to be run to return some data
114 to the user. But, you have an authorization system in place that
115 prevents certain users from seeing certain information. So, you want
116 to construct the basic query in one method, but add constraints to it in
121 my $request = $self->get_request; # Get a request object somehow.
122 my $schema = $self->result_source->schema;
124 my $cd_rs = $schema->resultset('CD')->search({
125 title => $request->param('title'),
126 year => $request->param('year'),
129 $cd_rs = $self->apply_security_policy( $cd_rs );
131 return $cd_rs->all();
134 sub apply_security_policy {
143 =head3 Resolving conditions and attributes
145 When a resultset is chained from another resultset (e.g.:
146 C<< my $new_rs = $old_rs->search(\%extra_cond, \%attrs) >>), conditions
147 and attributes with the same keys need resolving.
149 If any of L</columns>, L</select>, L</as> are present, they reset the
150 original selection, and start the selection "clean".
152 The L</join>, L</prefetch>, L</+columns>, L</+select>, L</+as> attributes
153 are merged into the existing ones from the original resultset.
155 The L</where> and L</having> attributes, and any search conditions, are
156 merged with an SQL C<AND> to the existing condition from the original
159 All other attributes are overridden by any new ones supplied in the
162 =head2 Multiple queries
164 Since a resultset just defines a query, you can do all sorts of
165 things with it with the same object.
167 # Don't hit the DB yet.
168 my $cd_rs = $schema->resultset('CD')->search({
169 title => 'something',
173 # Each of these hits the DB individually.
174 my $count = $cd_rs->count;
175 my $most_recent = $cd_rs->get_column('date_released')->max();
176 my @records = $cd_rs->all;
178 And it's not just limited to SELECT statements.
184 $cd_rs->create({ artist => 'Fred' });
186 Which is the same as:
188 $schema->resultset('CD')->create({
189 title => 'something',
194 See: L</search>, L</count>, L</get_column>, L</all>, L</create>.
202 =item Arguments: L<$source|DBIx::Class::ResultSource>, L<\%attrs?|/ATTRIBUTES>
204 =item Return Value: L<$resultset|/search>
208 The resultset constructor. Takes a source object (usually a
209 L<DBIx::Class::ResultSourceProxy::Table>) and an attribute hash (see
210 L</ATTRIBUTES> below). Does not perform any queries -- these are
211 executed as needed by the other methods.
213 Generally you never construct a resultset manually. Instead you get one
215 C<< $schema->L<resultset|DBIx::Class::Schema/resultset>('$source_name') >>
216 or C<< $another_resultset->L<search|/search>(...) >> (the later called in
219 my $rs = $schema->resultset('CD')->search({ title => '100th Window' });
225 If called on an object, proxies to L</new_result> instead, so
227 my $cd = $schema->resultset('CD')->new({ title => 'Spoon' });
229 will return a CD object, not a ResultSet, and is equivalent to:
231 my $cd = $schema->resultset('CD')->new_result({ title => 'Spoon' });
233 Please also keep in mind that many internals call L</new_result> directly,
234 so overloading this method with the idea of intercepting new result object
235 creation B<will not work>. See also warning pertaining to L</create>.
243 return $class->new_result(@_) if ref $class;
245 my ($source, $attrs) = @_;
246 $source = $source->resolve
247 if $source->isa('DBIx::Class::ResultSourceHandle');
249 $attrs = { %{$attrs||{}} };
250 delete @{$attrs}{qw(_sqlmaker_select_args _related_results_construction)};
252 if ($attrs->{page}) {
253 $attrs->{rows} ||= 10;
256 $attrs->{alias} ||= 'me';
259 result_source => $source,
260 cond => $attrs->{where},
265 # if there is a dark selector, this means we are already in a
266 # chain and the cleanup/sanification was taken care of by
268 $self->_normalize_selection($attrs)
269 unless $attrs->{_dark_selector};
272 $attrs->{result_class} || $source->result_class
282 =item Arguments: L<$cond|DBIx::Class::SQLMaker> | undef, L<\%attrs?|/ATTRIBUTES>
284 =item Return Value: $resultset (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
288 my @cds = $cd_rs->search({ year => 2001 }); # "... WHERE year = 2001"
289 my $new_rs = $cd_rs->search({ year => 2005 });
291 my $new_rs = $cd_rs->search([ { year => 2005 }, { year => 2004 } ]);
292 # year = 2005 OR year = 2004
294 In list context, C<< ->all() >> is called implicitly on the resultset, thus
295 returning a list of L<result|DBIx::Class::Manual::ResultClass> objects instead.
296 To avoid that, use L</search_rs>.
298 If you need to pass in additional attributes but no additional condition,
299 call it as C<search(undef, \%attrs)>.
301 # "SELECT name, artistid FROM $artist_table"
302 my @all_artists = $schema->resultset('Artist')->search(undef, {
303 columns => [qw/name artistid/],
306 For a list of attributes that can be passed to C<search>, see
307 L</ATTRIBUTES>. For more examples of using this function, see
308 L<Searching|DBIx::Class::Manual::Cookbook/Searching>. For a complete
309 documentation for the first argument, see L<SQL::Abstract/"WHERE CLAUSES">
310 and its extension L<DBIx::Class::SQLMaker>.
312 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
316 Note that L</search> does not process/deflate any of the values passed in the
317 L<SQL::Abstract>-compatible search condition structure. This is unlike other
318 condition-bound methods L</new_result>, L</create> and L</find>. The user must ensure
319 manually that any value passed to this method will stringify to something the
320 RDBMS knows how to deal with. A notable example is the handling of L<DateTime>
321 objects, for more info see:
322 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
328 my $rs = $self->search_rs( @_ );
333 elsif (defined wantarray) {
337 # we can be called by a relationship helper, which in
338 # turn may be called in void context due to some braindead
339 # overload or whatever else the user decided to be clever
340 # at this particular day. Thus limit the exception to
341 # external code calls only
342 $self->throw_exception ('->search is *not* a mutator, calling it in void context makes no sense')
343 if (caller)[0] !~ /^\QDBIx::Class::/;
353 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
355 =item Return Value: L<$resultset|/search>
359 This method does the same exact thing as search() except it will
360 always return a resultset, even in list context.
367 my $rsrc = $self->result_source;
368 my ($call_cond, $call_attrs);
370 # Special-case handling for (undef, undef) or (undef)
371 # Note that (foo => undef) is valid deprecated syntax
372 @_ = () if not scalar grep { defined $_ } @_;
378 # fish out attrs in the ($condref, $attr) case
379 elsif (@_ == 2 and ( ! defined $_[0] or (ref $_[0]) ne '') ) {
380 ($call_cond, $call_attrs) = @_;
383 $self->throw_exception('Odd number of arguments to search')
387 carp_unique 'search( %condition ) is deprecated, use search( \%condition ) instead'
388 unless $rsrc->result_class->isa('DBIx::Class::CDBICompat');
390 for my $i (0 .. $#_) {
392 $self->throw_exception ('All keys in condition key/value pairs must be plain scalars')
393 if (! defined $_[$i] or ref $_[$i] ne '');
399 # see if we can keep the cache (no $rs changes)
401 my %safe = (alias => 1, cache => 1);
402 if ( ! List::Util::first { !$safe{$_} } keys %$call_attrs and (
405 ref $call_cond eq 'HASH' && ! keys %$call_cond
407 ref $call_cond eq 'ARRAY' && ! @$call_cond
409 $cache = $self->get_cache;
412 my $old_attrs = { %{$self->{attrs}} };
413 my ($old_having, $old_where) = delete @{$old_attrs}{qw(having where)};
415 my $new_attrs = { %$old_attrs };
417 # take care of call attrs (only if anything is changing)
418 if ($call_attrs and keys %$call_attrs) {
420 # copy for _normalize_selection
421 $call_attrs = { %$call_attrs };
423 my @selector_attrs = qw/select as columns cols +select +as +columns include_columns/;
425 # reset the current selector list if new selectors are supplied
426 if (List::Util::first { exists $call_attrs->{$_} } qw/columns cols select as/) {
427 delete @{$old_attrs}{(@selector_attrs, '_dark_selector')};
430 # Normalize the new selector list (operates on the passed-in attr structure)
431 # Need to do it on every chain instead of only once on _resolved_attrs, in
432 # order to allow detection of empty vs partial 'as'
433 $call_attrs->{_dark_selector} = $old_attrs->{_dark_selector}
434 if $old_attrs->{_dark_selector};
435 $self->_normalize_selection ($call_attrs);
437 # start with blind overwriting merge, exclude selector attrs
438 $new_attrs = { %{$old_attrs}, %{$call_attrs} };
439 delete @{$new_attrs}{@selector_attrs};
441 for (@selector_attrs) {
442 $new_attrs->{$_} = $self->_merge_attr($old_attrs->{$_}, $call_attrs->{$_})
443 if ( exists $old_attrs->{$_} or exists $call_attrs->{$_} );
446 # older deprecated name, use only if {columns} is not there
447 if (my $c = delete $new_attrs->{cols}) {
448 carp_unique( "Resultset attribute 'cols' is deprecated, use 'columns' instead" );
449 if ($new_attrs->{columns}) {
450 carp "Resultset specifies both the 'columns' and the legacy 'cols' attributes - ignoring 'cols'";
453 $new_attrs->{columns} = $c;
458 # join/prefetch use their own crazy merging heuristics
459 foreach my $key (qw/join prefetch/) {
460 $new_attrs->{$key} = $self->_merge_joinpref_attr($old_attrs->{$key}, $call_attrs->{$key})
461 if exists $call_attrs->{$key};
464 # stack binds together
465 $new_attrs->{bind} = [ @{ $old_attrs->{bind} || [] }, @{ $call_attrs->{bind} || [] } ];
469 for ($old_where, $call_cond) {
471 $new_attrs->{where} = $self->_stack_cond (
472 $_, $new_attrs->{where}
477 if (defined $old_having) {
478 $new_attrs->{having} = $self->_stack_cond (
479 $old_having, $new_attrs->{having}
483 my $rs = (ref $self)->new($rsrc, $new_attrs);
485 $rs->set_cache($cache) if ($cache);
491 sub _normalize_selection {
492 my ($self, $attrs) = @_;
495 if ( exists $attrs->{include_columns} ) {
496 carp_unique( "Resultset attribute 'include_columns' is deprecated, use '+columns' instead" );
497 $attrs->{'+columns'} = $self->_merge_attr(
498 $attrs->{'+columns'}, delete $attrs->{include_columns}
502 # columns are always placed first, however
504 # Keep the X vs +X separation until _resolved_attrs time - this allows to
505 # delay the decision on whether to use a default select list ($rsrc->columns)
506 # allowing stuff like the remove_columns helper to work
508 # select/as +select/+as pairs need special handling - the amount of select/as
509 # elements in each pair does *not* have to be equal (think multicolumn
510 # selectors like distinct(foo, bar) ). If the selector is bare (no 'as'
511 # supplied at all) - try to infer the alias, either from the -as parameter
512 # of the selector spec, or use the parameter whole if it looks like a column
513 # name (ugly legacy heuristic). If all fails - leave the selector bare (which
514 # is ok as well), but make sure no more additions to the 'as' chain take place
515 for my $pref ('', '+') {
517 my ($sel, $as) = map {
518 my $key = "${pref}${_}";
520 my $val = [ ref $attrs->{$key} eq 'ARRAY'
522 : $attrs->{$key} || ()
524 delete $attrs->{$key};
528 if (! @$as and ! @$sel ) {
531 elsif (@$as and ! @$sel) {
532 $self->throw_exception(
533 "Unable to handle ${pref}as specification (@$as) without a corresponding ${pref}select"
537 # no as part supplied at all - try to deduce (unless explicit end of named selection is declared)
538 # if any @$as has been supplied we assume the user knows what (s)he is doing
539 # and blindly keep stacking up pieces
540 unless ($attrs->{_dark_selector}) {
543 if ( ref $_ eq 'HASH' and exists $_->{-as} ) {
544 push @$as, $_->{-as};
546 # assume any plain no-space, no-parenthesis string to be a column spec
547 # FIXME - this is retarded but is necessary to support shit like 'count(foo)'
548 elsif ( ! ref $_ and $_ =~ /^ [^\s\(\)]+ $/x) {
551 # if all else fails - raise a flag that no more aliasing will be allowed
553 $attrs->{_dark_selector} = {
555 string => ($dark_sel_dumper ||= do {
556 require Data::Dumper::Concise;
557 Data::Dumper::Concise::DumperObject()->Indent(0);
558 })->Values([$_])->Dump
566 elsif (@$as < @$sel) {
567 $self->throw_exception(
568 "Unable to handle an ${pref}as specification (@$as) with less elements than the corresponding ${pref}select"
571 elsif ($pref and $attrs->{_dark_selector}) {
572 $self->throw_exception(
573 "Unable to process named '+select', resultset contains an unnamed selector $attrs->{_dark_selector}{string}"
579 $attrs->{"${pref}select"} = $self->_merge_attr($attrs->{"${pref}select"}, $sel);
580 $attrs->{"${pref}as"} = $self->_merge_attr($attrs->{"${pref}as"}, $as);
585 my ($self, $left, $right) = @_;
587 # collapse single element top-level conditions
588 # (single pass only, unlikely to need recursion)
589 for ($left, $right) {
590 if (ref $_ eq 'ARRAY') {
598 elsif (ref $_ eq 'HASH') {
599 my ($first, $more) = keys %$_;
602 if (! defined $first) {
606 elsif (! defined $more) {
607 if ($first eq '-and' and ref $_->{'-and'} eq 'HASH') {
610 elsif ($first eq '-or' and ref $_->{'-or'} eq 'ARRAY') {
617 # merge hashes with weeding out of duplicates (simple cases only)
618 if (ref $left eq 'HASH' and ref $right eq 'HASH') {
620 # shallow copy to destroy
621 $right = { %$right };
622 for (grep { exists $right->{$_} } keys %$left) {
623 # the use of eq_deeply here is justified - the rhs of an
624 # expression can contain a lot of twisted weird stuff
625 delete $right->{$_} if Data::Compare::Compare( $left->{$_}, $right->{$_} );
628 $right = undef unless keys %$right;
632 if (defined $left xor defined $right) {
633 return defined $left ? $left : $right;
635 elsif (! defined $left) {
639 return { -and => [ $left, $right ] };
643 =head2 search_literal
645 B<CAVEAT>: C<search_literal> is provided for Class::DBI compatibility and
646 should only be used in that context. C<search_literal> is a convenience
647 method. It is equivalent to calling C<< $schema->search(\[]) >>, but if you
648 want to ensure columns are bound correctly, use L</search>.
650 See L<DBIx::Class::Manual::Cookbook/Searching> and
651 L<DBIx::Class::Manual::FAQ/Searching> for searching techniques that do not
652 require C<search_literal>.
656 =item Arguments: $sql_fragment, @standalone_bind_values
658 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
662 my @cds = $cd_rs->search_literal('year = ? AND title = ?', qw/2001 Reload/);
663 my $newrs = $artist_rs->search_literal('name = ?', 'Metallica');
665 Pass a literal chunk of SQL to be added to the conditional part of the
668 Example of how to use C<search> instead of C<search_literal>
670 my @cds = $cd_rs->search_literal('cdid = ? AND (artist = ? OR artist = ?)', (2, 1, 2));
671 my @cds = $cd_rs->search(\[ 'cdid = ? AND (artist = ? OR artist = ?)', [ 'cdid', 2 ], [ 'artist', 1 ], [ 'artist', 2 ] ]);
676 my ($self, $sql, @bind) = @_;
678 if ( @bind && ref($bind[-1]) eq 'HASH' ) {
681 return $self->search(\[ $sql, map [ {} => $_ ], @bind ], ($attr || () ));
688 =item Arguments: \%columns_values | @pk_values, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
690 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
694 Finds and returns a single row based on supplied criteria. Takes either a
695 hashref with the same format as L</create> (including inference of foreign
696 keys from related objects), or a list of primary key values in the same
697 order as the L<primary columns|DBIx::Class::ResultSource/primary_columns>
698 declaration on the L</result_source>.
700 In either case an attempt is made to combine conditions already existing on
701 the resultset with the condition passed to this method.
703 To aid with preparing the correct query for the storage you may supply the
704 C<key> attribute, which is the name of a
705 L<unique constraint|DBIx::Class::ResultSource/add_unique_constraint> (the
706 unique constraint corresponding to the
707 L<primary columns|DBIx::Class::ResultSource/primary_columns> is always named
708 C<primary>). If the C<key> attribute has been supplied, and DBIC is unable
709 to construct a query that satisfies the named unique constraint fully (
710 non-NULL values for each column member of the constraint) an exception is
713 If no C<key> is specified, the search is carried over all unique constraints
714 which are fully defined by the available condition.
716 If no such constraint is found, C<find> currently defaults to a simple
717 C<< search->(\%column_values) >> which may or may not do what you expect.
718 Note that this fallback behavior may be deprecated in further versions. If
719 you need to search with arbitrary conditions - use L</search>. If the query
720 resulting from this fallback produces more than one row, a warning to the
721 effect is issued, though only the first row is constructed and returned as
724 In addition to C<key>, L</find> recognizes and applies standard
725 L<resultset attributes|/ATTRIBUTES> in the same way as L</search> does.
727 Note that if you have extra concerns about the correctness of the resulting
728 query you need to specify the C<key> attribute and supply the entire condition
729 as an argument to find (since it is not always possible to perform the
730 combination of the resultset condition with the supplied one, especially if
731 the resultset condition contains literal sql).
733 For example, to find a row by its primary key:
735 my $cd = $schema->resultset('CD')->find(5);
737 You can also find a row by a specific unique constraint:
739 my $cd = $schema->resultset('CD')->find(
741 artist => 'Massive Attack',
742 title => 'Mezzanine',
744 { key => 'cd_artist_title' }
747 See also L</find_or_create> and L</update_or_create>.
753 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
755 my $rsrc = $self->result_source;
758 if (exists $attrs->{key}) {
759 $constraint_name = defined $attrs->{key}
761 : $self->throw_exception("An undefined 'key' resultset attribute makes no sense")
765 # Parse out the condition from input
768 if (ref $_[0] eq 'HASH') {
769 $call_cond = { %{$_[0]} };
772 # if only values are supplied we need to default to 'primary'
773 $constraint_name = 'primary' unless defined $constraint_name;
775 my @c_cols = $rsrc->unique_constraint_columns($constraint_name);
777 $self->throw_exception(
778 "No constraint columns, maybe a malformed '$constraint_name' constraint?"
781 $self->throw_exception (
782 'find() expects either a column/value hashref, or a list of values '
783 . "corresponding to the columns of the specified unique constraint '$constraint_name'"
784 ) unless @c_cols == @_;
787 @{$call_cond}{@c_cols} = @_;
791 for my $key (keys %$call_cond) {
793 my $keyref = ref($call_cond->{$key})
795 my $relinfo = $rsrc->relationship_info($key)
797 my $val = delete $call_cond->{$key};
799 next if $keyref eq 'ARRAY'; # has_many for multi_create
801 my $rel_q = $rsrc->_resolve_condition(
802 $relinfo->{cond}, $val, $key, $key
804 die "Can't handle complex relationship conditions in find" if ref($rel_q) ne 'HASH';
805 @related{keys %$rel_q} = values %$rel_q;
809 # relationship conditions take precedence (?)
810 @{$call_cond}{keys %related} = values %related;
812 my $alias = exists $attrs->{alias} ? $attrs->{alias} : $self->{attrs}{alias};
814 if (defined $constraint_name) {
815 $final_cond = $self->_qualify_cond_columns (
817 $self->_build_unique_cond (
825 elsif ($self->{attrs}{accessor} and $self->{attrs}{accessor} eq 'single') {
826 # This means that we got here after a merger of relationship conditions
827 # in ::Relationship::Base::search_related (the row method), and furthermore
828 # the relationship is of the 'single' type. This means that the condition
829 # provided by the relationship (already attached to $self) is sufficient,
830 # as there can be only one row in the database that would satisfy the
834 # no key was specified - fall down to heuristics mode:
835 # run through all unique queries registered on the resultset, and
836 # 'OR' all qualifying queries together
837 my (@unique_queries, %seen_column_combinations);
838 for my $c_name ($rsrc->unique_constraint_names) {
839 next if $seen_column_combinations{
840 join "\x00", sort $rsrc->unique_constraint_columns($c_name)
843 push @unique_queries, try {
844 $self->_build_unique_cond ($c_name, $call_cond, 'croak_on_nulls')
848 $final_cond = @unique_queries
849 ? [ map { $self->_qualify_cond_columns($_, $alias) } @unique_queries ]
850 : $self->_non_unique_find_fallback ($call_cond, $attrs)
854 # Run the query, passing the result_class since it should propagate for find
855 my $rs = $self->search ($final_cond, {result_class => $self->result_class, %$attrs});
856 if ($rs->_resolved_attrs->{collapse}) {
858 carp "Query returned more than one row" if $rs->next;
866 # This is a stop-gap method as agreed during the discussion on find() cleanup:
867 # http://lists.scsys.co.uk/pipermail/dbix-class/2010-October/009535.html
869 # It is invoked when find() is called in legacy-mode with insufficiently-unique
870 # condition. It is provided for overrides until a saner way forward is devised
872 # *NOTE* This is not a public method, and it's *GUARANTEED* to disappear down
873 # the road. Please adjust your tests accordingly to catch this situation early
874 # DBIx::Class::ResultSet->can('_non_unique_find_fallback') is reasonable
876 # The method will not be removed without an adequately complete replacement
877 # for strict-mode enforcement
878 sub _non_unique_find_fallback {
879 my ($self, $cond, $attrs) = @_;
881 return $self->_qualify_cond_columns(
883 exists $attrs->{alias}
885 : $self->{attrs}{alias}
890 sub _qualify_cond_columns {
891 my ($self, $cond, $alias) = @_;
893 my %aliased = %$cond;
894 for (keys %aliased) {
895 $aliased{"$alias.$_"} = delete $aliased{$_}
902 sub _build_unique_cond {
903 my ($self, $constraint_name, $extra_cond, $croak_on_null) = @_;
905 my @c_cols = $self->result_source->unique_constraint_columns($constraint_name);
907 # combination may fail if $self->{cond} is non-trivial
908 my ($final_cond) = try {
909 $self->_merge_with_rscond ($extra_cond)
914 # trim out everything not in $columns
915 $final_cond = { map {
916 exists $final_cond->{$_}
917 ? ( $_ => $final_cond->{$_} )
921 if (my @missing = grep
922 { ! ($croak_on_null ? defined $final_cond->{$_} : exists $final_cond->{$_}) }
925 $self->throw_exception( sprintf ( "Unable to satisfy requested constraint '%s', no values for column(s): %s",
927 join (', ', map { "'$_'" } @missing),
934 !$ENV{DBIC_NULLABLE_KEY_NOWARN}
936 my @undefs = sort grep { ! defined $final_cond->{$_} } (keys %$final_cond)
938 carp_unique ( sprintf (
939 "NULL/undef values supplied for requested unique constraint '%s' (NULL "
940 . 'values in column(s): %s). This is almost certainly not what you wanted, '
941 . 'though you can set DBIC_NULLABLE_KEY_NOWARN to disable this warning.',
943 join (', ', map { "'$_'" } @undefs),
950 =head2 search_related
954 =item Arguments: $rel_name, $cond?, L<\%attrs?|/ATTRIBUTES>
956 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
960 $new_rs = $cd_rs->search_related('artist', {
964 Searches the specified relationship, optionally specifying a condition and
965 attributes for matching records. See L</ATTRIBUTES> for more information.
967 In list context, C<< ->all() >> is called implicitly on the resultset, thus
968 returning a list of result objects instead. To avoid that, use L</search_related_rs>.
970 See also L</search_related_rs>.
975 return shift->related_resultset(shift)->search(@_);
978 =head2 search_related_rs
980 This method works exactly the same as search_related, except that
981 it guarantees a resultset, even in list context.
985 sub search_related_rs {
986 return shift->related_resultset(shift)->search_rs(@_);
993 =item Arguments: none
995 =item Return Value: L<$cursor|DBIx::Class::Cursor>
999 Returns a storage-driven cursor to the given resultset. See
1000 L<DBIx::Class::Cursor> for more information.
1007 return $self->{cursor} ||= do {
1008 my $attrs = $self->_resolved_attrs;
1009 $self->result_source->storage->select(
1010 $attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs
1019 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1021 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1025 my $cd = $schema->resultset('CD')->single({ year => 2001 });
1027 Inflates the first result without creating a cursor if the resultset has
1028 any records in it; if not returns C<undef>. Used by L</find> as a lean version
1031 While this method can take an optional search condition (just like L</search>)
1032 being a fast-code-path it does not recognize search attributes. If you need to
1033 add extra joins or similar, call L</search> and then chain-call L</single> on the
1034 L<DBIx::Class::ResultSet> returned.
1040 As of 0.08100, this method enforces the assumption that the preceding
1041 query returns only one row. If more than one row is returned, you will receive
1044 Query returned more than one row
1046 In this case, you should be using L</next> or L</find> instead, or if you really
1047 know what you are doing, use the L</rows> attribute to explicitly limit the size
1050 This method will also throw an exception if it is called on a resultset prefetching
1051 has_many, as such a prefetch implies fetching multiple rows from the database in
1052 order to assemble the resulting object.
1059 my ($self, $where) = @_;
1061 $self->throw_exception('single() only takes search conditions, no attributes. You want ->search( $cond, $attrs )->single()');
1064 my $attrs = { %{$self->_resolved_attrs} };
1066 $self->throw_exception(
1067 'single() can not be used on resultsets collapsing a has_many. Use find( \%cond ) or next() instead'
1068 ) if $attrs->{collapse};
1071 if (defined $attrs->{where}) {
1074 [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ }
1075 $where, delete $attrs->{where} ]
1078 $attrs->{where} = $where;
1082 my $data = [ $self->result_source->storage->select_single(
1083 $attrs->{from}, $attrs->{select},
1084 $attrs->{where}, $attrs
1087 return undef unless @$data;
1088 $self->{_stashed_rows} = [ $data ];
1089 $self->_construct_results->[0];
1095 # Recursively collapse the query, accumulating values for each column.
1097 sub _collapse_query {
1098 my ($self, $query, $collapsed) = @_;
1102 if (ref $query eq 'ARRAY') {
1103 foreach my $subquery (@$query) {
1104 next unless ref $subquery; # -or
1105 $collapsed = $self->_collapse_query($subquery, $collapsed);
1108 elsif (ref $query eq 'HASH') {
1109 if (keys %$query and (keys %$query)[0] eq '-and') {
1110 foreach my $subquery (@{$query->{-and}}) {
1111 $collapsed = $self->_collapse_query($subquery, $collapsed);
1115 foreach my $col (keys %$query) {
1116 my $value = $query->{$col};
1117 $collapsed->{$col}{$value}++;
1129 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1131 =item Return Value: L<$resultsetcolumn|DBIx::Class::ResultSetColumn>
1135 my $max_length = $rs->get_column('length')->max;
1137 Returns a L<DBIx::Class::ResultSetColumn> instance for a column of the ResultSet.
1142 my ($self, $column) = @_;
1143 my $new = DBIx::Class::ResultSetColumn->new($self, $column);
1151 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1153 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1157 # WHERE title LIKE '%blue%'
1158 $cd_rs = $rs->search_like({ title => '%blue%'});
1160 Performs a search, but uses C<LIKE> instead of C<=> as the condition. Note
1161 that this is simply a convenience method retained for ex Class::DBI users.
1162 You most likely want to use L</search> with specific operators.
1164 For more information, see L<DBIx::Class::Manual::Cookbook>.
1166 This method is deprecated and will be removed in 0.09. Use L</search()>
1167 instead. An example conversion is:
1169 ->search_like({ foo => 'bar' });
1173 ->search({ foo => { like => 'bar' } });
1180 'search_like() is deprecated and will be removed in DBIC version 0.09.'
1181 .' Instead use ->search({ x => { -like => "y%" } })'
1182 .' (note the outer pair of {}s - they are important!)'
1184 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
1185 my $query = ref $_[0] eq 'HASH' ? { %{shift()} }: {@_};
1186 $query->{$_} = { 'like' => $query->{$_} } for keys %$query;
1187 return $class->search($query, { %$attrs });
1194 =item Arguments: $first, $last
1196 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1200 Returns a resultset or object list representing a subset of elements from the
1201 resultset slice is called on. Indexes are from 0, i.e., to get the first
1202 three records, call:
1204 my ($one, $two, $three) = $rs->slice(0, 2);
1209 my ($self, $min, $max) = @_;
1210 my $attrs = {}; # = { %{ $self->{attrs} || {} } };
1211 $attrs->{offset} = $self->{attrs}{offset} || 0;
1212 $attrs->{offset} += $min;
1213 $attrs->{rows} = ($max ? ($max - $min + 1) : 1);
1214 return $self->search(undef, $attrs);
1215 #my $slice = (ref $self)->new($self->result_source, $attrs);
1216 #return (wantarray ? $slice->all : $slice);
1223 =item Arguments: none
1225 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1229 Returns the next element in the resultset (C<undef> is there is none).
1231 Can be used to efficiently iterate over records in the resultset:
1233 my $rs = $schema->resultset('CD')->search;
1234 while (my $cd = $rs->next) {
1238 Note that you need to store the resultset object, and call C<next> on it.
1239 Calling C<< resultset('Table')->next >> repeatedly will always return the
1240 first record from the resultset.
1247 if (my $cache = $self->get_cache) {
1248 $self->{all_cache_position} ||= 0;
1249 return $cache->[$self->{all_cache_position}++];
1252 if ($self->{attrs}{cache}) {
1253 delete $self->{pager};
1254 $self->{all_cache_position} = 1;
1255 return ($self->all)[0];
1258 return shift(@{$self->{_stashed_results}}) if @{ $self->{_stashed_results}||[] };
1260 $self->{_stashed_results} = $self->_construct_results
1263 return shift @{$self->{_stashed_results}};
1266 # Constructs as many results as it can in one pass while respecting
1267 # cursor laziness. Several modes of operation:
1269 # * Always builds everything present in @{$self->{_stashed_rows}}
1270 # * If called with $fetch_all true - pulls everything off the cursor and
1271 # builds all result structures (or objects) in one pass
1272 # * If $self->_resolved_attrs->{collapse} is true, checks the order_by
1273 # and if the resultset is ordered properly by the left side:
1274 # * Fetches stuff off the cursor until the "master object" changes,
1275 # and saves the last extra row (if any) in @{$self->{_stashed_rows}}
1277 # * Just fetches, and collapses/constructs everything as if $fetch_all
1278 # was requested (there is no other way to collapse except for an
1280 # * If no collapse is requested - just get the next row, construct and
1282 sub _construct_results {
1283 my ($self, $fetch_all) = @_;
1285 my $rsrc = $self->result_source;
1286 my $attrs = $self->_resolved_attrs;
1291 ! $attrs->{order_by}
1295 my @pcols = $rsrc->primary_columns
1297 # default order for collapsing unless the user asked for something
1298 $attrs->{order_by} = [ map { join '.', $attrs->{alias}, $_} @pcols ];
1299 $attrs->{_ordered_for_collapse} = 1;
1300 $attrs->{_order_is_artificial} = 1;
1303 # this will be used as both initial raw-row collector AND as a RV of
1304 # _construct_results. Not regrowing the array twice matters a lot...
1305 # a surprising amount actually
1306 my $rows = delete $self->{_stashed_rows};
1308 my $cursor; # we may not need one at all
1310 my $did_fetch_all = $fetch_all;
1313 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1314 $rows = [ ($rows ? @$rows : ()), $self->cursor->all ];
1316 elsif( $attrs->{collapse} ) {
1318 # a cursor will need to be closed over in case of collapse
1319 $cursor = $self->cursor;
1321 $attrs->{_ordered_for_collapse} = (
1327 ->_main_source_order_by_portion_is_stable($rsrc, $attrs->{order_by}, $attrs->{where})
1329 ) unless defined $attrs->{_ordered_for_collapse};
1331 if (! $attrs->{_ordered_for_collapse}) {
1334 # instead of looping over ->next, use ->all in stealth mode
1335 # *without* calling a ->reset afterwards
1336 # FIXME ENCAPSULATION - encapsulation breach, cursor method additions pending
1337 if (! $cursor->{_done}) {
1338 $rows = [ ($rows ? @$rows : ()), $cursor->all ];
1339 $cursor->{_done} = 1;
1344 if (! $did_fetch_all and ! @{$rows||[]} ) {
1345 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1346 $cursor ||= $self->cursor;
1347 if (scalar (my @r = $cursor->next) ) {
1352 return undef unless @{$rows||[]};
1354 # sanity check - people are too clever for their own good
1355 if ($attrs->{collapse} and my $aliastypes = $attrs->{_sqlmaker_select_args}[3]{_aliastypes} ) {
1357 my $multiplied_selectors;
1358 for my $sel_alias ( grep { $_ ne $attrs->{alias} } keys %{ $aliastypes->{selecting} } ) {
1360 $aliastypes->{multiplying}{$sel_alias}
1362 scalar grep { $aliastypes->{multiplying}{(values %$_)[0]} } @{ $aliastypes->{selecting}{$sel_alias}{-parents} }
1364 $multiplied_selectors->{$_} = 1 for values %{$aliastypes->{selecting}{$sel_alias}{-seen_columns}}
1368 for my $i (0 .. $#{$attrs->{as}} ) {
1369 my $sel = $attrs->{select}[$i];
1371 if (ref $sel eq 'SCALAR') {
1374 elsif( ref $sel eq 'REF' and ref $$sel eq 'ARRAY' ) {
1378 $self->throw_exception(
1379 'Result collapse not possible - selection from a has_many source redirected to the main object'
1380 ) if ($multiplied_selectors->{$sel} and $attrs->{as}[$i] !~ /\./);
1384 # hotspot - skip the setter
1385 my $res_class = $self->_result_class;
1387 my $inflator_cref = $self->{_result_inflator}{cref} ||= do {
1388 $res_class->can ('inflate_result')
1389 or $self->throw_exception("Inflator $res_class does not provide an inflate_result() method");
1392 my $infmap = $attrs->{as};
1394 $self->{_result_inflator}{is_core_row} = ( (
1397 ( \&DBIx::Class::Row::inflate_result || die "No ::Row::inflate_result() - can't happen" )
1398 ) ? 1 : 0 ) unless defined $self->{_result_inflator}{is_core_row};
1400 $self->{_result_inflator}{is_hri} = ( (
1401 ! $self->{_result_inflator}{is_core_row}
1404 require DBIx::Class::ResultClass::HashRefInflator
1406 DBIx::Class::ResultClass::HashRefInflator->can('inflate_result')
1408 ) ? 1 : 0 ) unless defined $self->{_result_inflator}{is_hri};
1411 if (! $attrs->{_related_results_construction}) {
1412 # construct a much simpler array->hash folder for the one-table cases right here
1413 if ($self->{_result_inflator}{is_hri}) {
1414 for my $r (@$rows) {
1415 $r = { map { $infmap->[$_] => $r->[$_] } 0..$#$infmap };
1418 # FIXME SUBOPTIMAL this is a very very very hot spot
1419 # while rather optimal we can *still* do much better, by
1420 # building a smarter Row::inflate_result(), and
1421 # switch to feeding it data via a much leaner interface
1423 # crude unscientific benchmarking indicated the shortcut eval is not worth it for
1424 # this particular resultset size
1425 elsif (@$rows < 60) {
1426 for my $r (@$rows) {
1427 $r = $inflator_cref->($res_class, $rsrc, { map { $infmap->[$_] => $r->[$_] } (0..$#$infmap) } );
1432 '$_ = $inflator_cref->($res_class, $rsrc, { %s }) for @$rows',
1433 join (', ', map { "\$infmap->[$_] => \$_->[$_]" } 0..$#$infmap )
1439 $self->{_result_inflator}{is_hri} ? 'hri'
1440 : $self->{_result_inflator}{is_core_row} ? 'classic_pruning'
1441 : 'classic_nonpruning'
1444 # $args and $attrs to _mk_row_parser are separated to delineate what is
1445 # core collapser stuff and what is dbic $rs specific
1446 @{$self->{_row_parser}{$parser_type}}{qw(cref nullcheck)} = $rsrc->_mk_row_parser({
1448 inflate_map => $infmap,
1449 collapse => $attrs->{collapse},
1450 premultiplied => $attrs->{_main_source_premultiplied},
1451 hri_style => $self->{_result_inflator}{is_hri},
1452 prune_null_branches => $self->{_result_inflator}{is_hri} || $self->{_result_inflator}{is_core_row},
1453 }, $attrs) unless $self->{_row_parser}{$parser_type}{cref};
1455 # column_info metadata historically hasn't been too reliable.
1456 # We need to start fixing this somehow (the collapse resolver
1457 # can't work without it). Add an explicit check for the *main*
1458 # result, hopefully this will gradually weed out such errors
1460 # FIXME - this is a temporary kludge that reduces performance
1461 # It is however necessary for the time being
1462 my ($unrolled_non_null_cols_to_check, $err);
1464 if (my $check_non_null_cols = $self->{_row_parser}{$parser_type}{nullcheck} ) {
1467 'Collapse aborted due to invalid ResultSource metadata - the following '
1468 . 'selections are declared non-nullable but NULLs were retrieved: '
1472 COL: for my $i (@$check_non_null_cols) {
1473 ! defined $_->[$i] and push @violating_idx, $i and next COL for @$rows;
1476 $self->throw_exception( $err . join (', ', map { "'$infmap->[$_]'" } @violating_idx ) )
1479 $unrolled_non_null_cols_to_check = join (',', @$check_non_null_cols);
1483 ($did_fetch_all or ! $attrs->{collapse}) ? undef
1484 : defined $unrolled_non_null_cols_to_check ? eval sprintf <<'EOS', $unrolled_non_null_cols_to_check
1486 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1487 my @r = $cursor->next or return;
1488 if (my @violating_idx = grep { ! defined $r[$_] } (%s) ) {
1489 $self->throw_exception( $err . join (', ', map { "'$infmap->[$_]'" } @violating_idx ) )
1495 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1496 my @r = $cursor->next or return;
1501 $self->{_row_parser}{$parser_type}{cref}->(
1503 $next_cref ? ( $next_cref, $self->{_stashed_rows} = [] ) : (),
1506 # Special-case multi-object HRI - there is no $inflator_cref pass
1507 unless ($self->{_result_inflator}{is_hri}) {
1508 $_ = $inflator_cref->($res_class, $rsrc, @$_) for @$rows
1512 # The @$rows check seems odd at first - why wouldn't we want to warn
1513 # regardless? The issue is things like find() etc, where the user
1514 # *knows* only one result will come back. In these cases the ->all
1515 # is not a pessimization, but rather something we actually want
1517 'Unable to properly collapse has_many results in iterator mode due '
1518 . 'to order criteria - performed an eager cursor slurp underneath. '
1519 . 'Consider using ->all() instead'
1520 ) if ( ! $fetch_all and @$rows > 1 );
1525 =head2 result_source
1529 =item Arguments: L<$result_source?|DBIx::Class::ResultSource>
1531 =item Return Value: L<$result_source|DBIx::Class::ResultSource>
1535 An accessor for the primary ResultSource object from which this ResultSet
1542 =item Arguments: $result_class?
1544 =item Return Value: $result_class
1548 An accessor for the class to use when creating result objects. Defaults to
1549 C<< result_source->result_class >> - which in most cases is the name of the
1550 L<"table"|DBIx::Class::Manual::Glossary/"ResultSource"> class.
1552 Note that changing the result_class will also remove any components
1553 that were originally loaded in the source class via
1554 L<DBIx::Class::ResultSource/load_components>. Any overloaded methods
1555 in the original source class will not run.
1560 my ($self, $result_class) = @_;
1561 if ($result_class) {
1563 # don't fire this for an object
1564 $self->ensure_class_loaded($result_class)
1565 unless ref($result_class);
1567 if ($self->get_cache) {
1568 carp_unique('Changing the result_class of a ResultSet instance with cached results is a noop - the cache contents will not be altered');
1570 # FIXME ENCAPSULATION - encapsulation breach, cursor method additions pending
1571 elsif ($self->{cursor} && $self->{cursor}{_pos}) {
1572 $self->throw_exception('Changing the result_class of a ResultSet instance with an active cursor is not supported');
1575 $self->_result_class($result_class);
1577 delete $self->{_result_inflator};
1579 $self->_result_class;
1586 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1588 =item Return Value: $count
1592 Performs an SQL C<COUNT> with the same query as the resultset was built
1593 with to find the number of elements. Passing arguments is equivalent to
1594 C<< $rs->search ($cond, \%attrs)->count >>
1600 return $self->search(@_)->count if @_ and defined $_[0];
1601 return scalar @{ $self->get_cache } if $self->get_cache;
1603 my $attrs = { %{ $self->_resolved_attrs } };
1605 # this is a little optimization - it is faster to do the limit
1606 # adjustments in software, instead of a subquery
1607 my ($rows, $offset) = delete @{$attrs}{qw/rows offset/};
1610 if ($self->_has_resolved_attr (qw/collapse group_by/)) {
1611 $crs = $self->_count_subq_rs ($attrs);
1614 $crs = $self->_count_rs ($attrs);
1616 my $count = $crs->next;
1618 $count -= $offset if $offset;
1619 $count = $rows if $rows and $rows < $count;
1620 $count = 0 if ($count < 0);
1629 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1631 =item Return Value: L<$count_rs|DBIx::Class::ResultSetColumn>
1635 Same as L</count> but returns a L<DBIx::Class::ResultSetColumn> object.
1636 This can be very handy for subqueries:
1638 ->search( { amount => $some_rs->count_rs->as_query } )
1640 As with regular resultsets the SQL query will be executed only after
1641 the resultset is accessed via L</next> or L</all>. That would return
1642 the same single value obtainable via L</count>.
1648 return $self->search(@_)->count_rs if @_;
1650 # this may look like a lack of abstraction (count() does about the same)
1651 # but in fact an _rs *must* use a subquery for the limits, as the
1652 # software based limiting can not be ported if this $rs is to be used
1653 # in a subquery itself (i.e. ->as_query)
1654 if ($self->_has_resolved_attr (qw/collapse group_by offset rows/)) {
1655 return $self->_count_subq_rs($self->{_attrs});
1658 return $self->_count_rs($self->{_attrs});
1663 # returns a ResultSetColumn object tied to the count query
1666 my ($self, $attrs) = @_;
1668 my $rsrc = $self->result_source;
1670 my $tmp_attrs = { %$attrs };
1671 # take off any limits, record_filter is cdbi, and no point of ordering nor locking a count
1672 delete @{$tmp_attrs}{qw/rows offset order_by record_filter for/};
1674 # overwrite the selector (supplied by the storage)
1675 $rsrc->resultset_class->new($rsrc, {
1677 select => $rsrc->storage->_count_select ($rsrc, $attrs),
1679 })->get_column ('count');
1683 # same as above but uses a subquery
1685 sub _count_subq_rs {
1686 my ($self, $attrs) = @_;
1688 my $rsrc = $self->result_source;
1690 my $sub_attrs = { %$attrs };
1691 # extra selectors do not go in the subquery and there is no point of ordering it, nor locking it
1692 delete @{$sub_attrs}{qw/collapse columns as select order_by for/};
1694 # if we multi-prefetch we group_by something unique, as this is what we would
1695 # get out of the rs via ->next/->all. We *DO WANT* to clobber old group_by regardless
1696 if ( $attrs->{collapse} ) {
1697 $sub_attrs->{group_by} = [ map { "$attrs->{alias}.$_" } @{
1698 $rsrc->_identifying_column_set || $self->throw_exception(
1699 'Unable to construct a unique group_by criteria properly collapsing the '
1700 . 'has_many prefetch before count()'
1705 # Calculate subquery selector
1706 if (my $g = $sub_attrs->{group_by}) {
1708 my $sql_maker = $rsrc->storage->sql_maker;
1710 # necessary as the group_by may refer to aliased functions
1712 for my $sel (@{$attrs->{select}}) {
1713 $sel_index->{$sel->{-as}} = $sel
1714 if (ref $sel eq 'HASH' and $sel->{-as});
1717 # anything from the original select mentioned on the group-by needs to make it to the inner selector
1718 # also look for named aggregates referred in the having clause
1719 # having often contains scalarrefs - thus parse it out entirely
1721 if ($attrs->{having}) {
1722 local $sql_maker->{having_bind};
1723 local $sql_maker->{quote_char} = $sql_maker->{quote_char};
1724 local $sql_maker->{name_sep} = $sql_maker->{name_sep};
1725 unless (defined $sql_maker->{quote_char} and length $sql_maker->{quote_char}) {
1726 $sql_maker->{quote_char} = [ "\x00", "\xFF" ];
1727 # if we don't unset it we screw up retarded but unfortunately working
1728 # 'MAX(foo.bar)' => { '>', 3 }
1729 $sql_maker->{name_sep} = '';
1732 $sql_maker->clear_renderer;
1733 $sql_maker->clear_converter;
1735 my ($lquote, $rquote, $sep) = map { quotemeta $_ } ($sql_maker->_quote_chars, $sql_maker->name_sep);
1737 my $having_sql = $sql_maker->_render_sqla(where => $attrs->{having});
1739 $sql_maker->clear_renderer;
1740 $sql_maker->clear_converter;
1744 # search for both a proper quoted qualified string, for a naive unquoted scalarref
1745 # and if all fails for an utterly naive quoted scalar-with-function
1746 while ($having_sql =~ /
1747 (?: $rquote $sep)? $lquote (.+?) $rquote
1749 [\s,] \w+ \. (\w+) [\s,]
1751 [\s,] $lquote (.+?) $rquote [\s,]
1753 my $part = $1 || $2 || $3; # one of them matched if we got here
1754 unless ($seen_having{$part}++) {
1761 my $colpiece = $sel_index->{$_} || $_;
1763 # unqualify join-based group_by's. Arcane but possible query
1764 # also horrible horrible hack to alias a column (not a func.)
1765 # (probably need to introduce SQLA syntax)
1766 if ($colpiece =~ /\./ && $colpiece !~ /^$attrs->{alias}\./) {
1769 $colpiece = \ sprintf ('%s AS %s', map { $sql_maker->_quote ($_) } ($colpiece, $as) );
1771 push @{$sub_attrs->{select}}, $colpiece;
1775 my @pcols = map { "$attrs->{alias}.$_" } ($rsrc->primary_columns);
1776 $sub_attrs->{select} = @pcols ? \@pcols : [ 1 ];
1779 return $rsrc->resultset_class
1780 ->new ($rsrc, $sub_attrs)
1782 ->search ({}, { columns => { count => $rsrc->storage->_count_select ($rsrc, $attrs) } })
1783 ->get_column ('count');
1787 =head2 count_literal
1789 B<CAVEAT>: C<count_literal> is provided for Class::DBI compatibility and
1790 should only be used in that context. See L</search_literal> for further info.
1794 =item Arguments: $sql_fragment, @standalone_bind_values
1796 =item Return Value: $count
1800 Counts the results in a literal query. Equivalent to calling L</search_literal>
1801 with the passed arguments, then L</count>.
1805 sub count_literal { shift->search_literal(@_)->count; }
1811 =item Arguments: none
1813 =item Return Value: L<@result_objs|DBIx::Class::Manual::ResultClass>
1817 Returns all elements in the resultset.
1824 $self->throw_exception("all() doesn't take any arguments, you probably wanted ->search(...)->all()");
1827 delete @{$self}{qw/_stashed_rows _stashed_results/};
1829 if (my $c = $self->get_cache) {
1833 $self->cursor->reset;
1835 my $objs = $self->_construct_results('fetch_all') || [];
1837 $self->set_cache($objs) if $self->{attrs}{cache};
1846 =item Arguments: none
1848 =item Return Value: $self
1852 Resets the resultset's cursor, so you can iterate through the elements again.
1853 Implicitly resets the storage cursor, so a subsequent L</next> will trigger
1861 delete @{$self}{qw/_stashed_rows _stashed_results/};
1862 $self->{all_cache_position} = 0;
1863 $self->cursor->reset;
1871 =item Arguments: none
1873 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1877 L<Resets|/reset> the resultset (causing a fresh query to storage) and returns
1878 an object for the first result (or C<undef> if the resultset is empty).
1883 return $_[0]->reset->next;
1889 # Determines whether and what type of subquery is required for the $rs operation.
1890 # If grouping is necessary either supplies its own, or verifies the current one
1891 # After all is done delegates to the proper storage method.
1893 sub _rs_update_delete {
1894 my ($self, $op, $values) = @_;
1896 my $rsrc = $self->result_source;
1897 my $storage = $rsrc->schema->storage;
1899 my $attrs = { %{$self->_resolved_attrs} };
1901 my $join_classifications;
1902 my ($existing_group_by) = delete @{$attrs}{qw(group_by _grouped_by_distinct)};
1904 # do we need a subquery for any reason?
1906 defined $existing_group_by
1908 # if {from} is unparseable wrap a subq
1909 ref($attrs->{from}) ne 'ARRAY'
1911 # limits call for a subq
1912 $self->_has_resolved_attr(qw/rows offset/)
1915 # simplify the joinmap, so we can further decide if a subq is necessary
1916 if (!$needs_subq and @{$attrs->{from}} > 1) {
1918 ($attrs->{from}, $join_classifications) =
1919 $storage->_prune_unused_joins ($attrs->{from}, $attrs->{select}, $self->{cond}, $attrs);
1921 # any non-pruneable non-local restricting joins imply subq
1922 $needs_subq = defined List::Util::first { $_ ne $attrs->{alias} } keys %{ $join_classifications->{restricting} || {} };
1925 # check if the head is composite (by now all joins are thrown out unless $needs_subq)
1927 (ref $attrs->{from}[0]) ne 'HASH'
1929 ref $attrs->{from}[0]{ $attrs->{from}[0]{-alias} }
1933 # do we need anything like a subquery?
1934 if (! $needs_subq) {
1935 # Most databases do not allow aliasing of tables in UPDATE/DELETE. Thus
1936 # a condition containing 'me' or other table prefixes will not work
1937 # at all - so we convert the WHERE to a dq tree now, dequalify all
1938 # identifiers found therein via a scan across the tree, and then use
1939 # \{} style to pass the result onwards for use in the final query
1940 if ($self->{cond}) {
1942 my $converter = $rsrc->storage->sql_maker->converter;
1944 DQ_IDENTIFIER ,=> sub { $_ = [ $_->[-1] ] for $_[0]->{elements} }
1945 }, my $where_dq = $converter->_where_to_dq($self->{cond}));
1951 # we got this far - means it is time to wrap a subquery
1952 my $idcols = $rsrc->_identifying_column_set || $self->throw_exception(
1954 "Unable to perform complex resultset %s() without an identifying set of columns on source '%s'",
1960 # make a new $rs selecting only the PKs (that's all we really need for the subq)
1961 delete $attrs->{$_} for qw/select as collapse/;
1962 $attrs->{columns} = [ map { "$attrs->{alias}.$_" } @$idcols ];
1964 # this will be consumed by the pruner waaaaay down the stack
1965 $attrs->{_force_prune_multiplying_joins} = 1;
1967 my $subrs = (ref $self)->new($rsrc, $attrs);
1969 if (@$idcols == 1) {
1970 $cond = { $idcols->[0] => { -in => $subrs->as_query } };
1972 elsif ($storage->_use_multicolumn_in) {
1973 # no syntax for calling this properly yet
1974 # !!! EXPERIMENTAL API !!! WILL CHANGE !!!
1975 my $left = $storage->sql_maker->_render_sqla(select_select => $idcols);
1976 $left =~ s/^SELECT //i;
1977 my $right = $storage->sql_maker
1979 ->_literal_to_dq(${$subrs->as_query});
1981 { 'SQL.Naive' => 'in' },
1982 [ Literal(SQL => "( $left )"), $right ],
1986 # if all else fails - get all primary keys and operate over a ORed set
1987 # wrap in a transaction for consistency
1988 # this is where the group_by/multiplication starts to matter
1992 keys %{ $join_classifications->{multiplying} || {} }
1994 # make sure if there is a supplied group_by it matches the columns compiled above
1995 # perfectly. Anything else can not be sanely executed on most databases so croak
1996 # right then and there
1997 if ($existing_group_by) {
1998 my @current_group_by = map
1999 { $_ =~ /\./ ? $_ : "$attrs->{alias}.$_" }
2004 join ("\x00", sort @current_group_by)
2006 join ("\x00", sort @{$attrs->{columns}} )
2008 $self->throw_exception (
2009 "You have just attempted a $op operation on a resultset which does group_by"
2010 . ' on columns other than the primary keys, while DBIC internally needs to retrieve'
2011 . ' the primary keys in a subselect. All sane RDBMS engines do not support this'
2012 . ' kind of queries. Please retry the operation with a modified group_by or'
2013 . ' without using one at all.'
2018 $subrs = $subrs->search({}, { group_by => $attrs->{columns} });
2021 $guard = $storage->txn_scope_guard;
2024 for my $row ($subrs->cursor->all) {
2026 { $idcols->[$_] => $row->[$_] }
2033 my $res = $storage->$op (
2035 $op eq 'update' ? $values : (),
2039 $guard->commit if $guard;
2048 =item Arguments: \%values
2050 =item Return Value: $underlying_storage_rv
2054 Sets the specified columns in the resultset to the supplied values in a
2055 single query. Note that this will not run any accessor/set_column/update
2056 triggers, nor will it update any result object instances derived from this
2057 resultset (this includes the contents of the L<resultset cache|/set_cache>
2058 if any). See L</update_all> if you need to execute any on-update
2059 triggers or cascades defined either by you or a
2060 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
2062 The return value is a pass through of what the underlying
2063 storage backend returned, and may vary. See L<DBI/execute> for the most
2068 Note that L</update> does not process/deflate any of the values passed in.
2069 This is unlike the corresponding L<DBIx::Class::Row/update>. The user must
2070 ensure manually that any value passed to this method will stringify to
2071 something the RDBMS knows how to deal with. A notable example is the
2072 handling of L<DateTime> objects, for more info see:
2073 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
2078 my ($self, $values) = @_;
2079 $self->throw_exception('Values for update must be a hash')
2080 unless ref $values eq 'HASH';
2082 return $self->_rs_update_delete ('update', $values);
2089 =item Arguments: \%values
2091 =item Return Value: 1
2095 Fetches all objects and updates them one at a time via
2096 L<DBIx::Class::Row/update>. Note that C<update_all> will run DBIC defined
2097 triggers, while L</update> will not.
2102 my ($self, $values) = @_;
2103 $self->throw_exception('Values for update_all must be a hash')
2104 unless ref $values eq 'HASH';
2106 my $guard = $self->result_source->schema->txn_scope_guard;
2107 $_->update({%$values}) for $self->all; # shallow copy - update will mangle it
2116 =item Arguments: none
2118 =item Return Value: $underlying_storage_rv
2122 Deletes the rows matching this resultset in a single query. Note that this
2123 will not run any delete triggers, nor will it alter the
2124 L<in_storage|DBIx::Class::Row/in_storage> status of any result object instances
2125 derived from this resultset (this includes the contents of the
2126 L<resultset cache|/set_cache> if any). See L</delete_all> if you need to
2127 execute any on-delete triggers or cascades defined either by you or a
2128 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
2130 The return value is a pass through of what the underlying storage backend
2131 returned, and may vary. See L<DBI/execute> for the most common case.
2137 $self->throw_exception('delete does not accept any arguments')
2140 return $self->_rs_update_delete ('delete');
2147 =item Arguments: none
2149 =item Return Value: 1
2153 Fetches all objects and deletes them one at a time via
2154 L<DBIx::Class::Row/delete>. Note that C<delete_all> will run DBIC defined
2155 triggers, while L</delete> will not.
2161 $self->throw_exception('delete_all does not accept any arguments')
2164 my $guard = $self->result_source->schema->txn_scope_guard;
2165 $_->delete for $self->all;
2174 =item Arguments: [ \@column_list, \@row_values+ ] | [ \%col_data+ ]
2176 =item Return Value: L<\@result_objects|DBIx::Class::Manual::ResultClass> (scalar context) | L<@result_objects|DBIx::Class::Manual::ResultClass> (list context)
2180 Accepts either an arrayref of hashrefs or alternatively an arrayref of
2187 The context of this method call has an important effect on what is
2188 submitted to storage. In void context data is fed directly to fastpath
2189 insertion routines provided by the underlying storage (most often
2190 L<DBI/execute_for_fetch>), bypassing the L<new|DBIx::Class::Row/new> and
2191 L<insert|DBIx::Class::Row/insert> calls on the
2192 L<Result|DBIx::Class::Manual::ResultClass> class, including any
2193 augmentation of these methods provided by components. For example if you
2194 are using something like L<DBIx::Class::UUIDColumns> to create primary
2195 keys for you, you will find that your PKs are empty. In this case you
2196 will have to explicitly force scalar or list context in order to create
2201 In non-void (scalar or list) context, this method is simply a wrapper
2202 for L</create>. Depending on list or scalar context either a list of
2203 L<Result|DBIx::Class::Manual::ResultClass> objects or an arrayref
2204 containing these objects is returned.
2206 When supplying data in "arrayref of arrayrefs" invocation style, the
2207 first element should be a list of column names and each subsequent
2208 element should be a data value in the earlier specified column order.
2211 $schema->resultset("Artist")->populate([
2212 [ qw( artistid name ) ],
2213 [ 100, 'A Formally Unknown Singer' ],
2214 [ 101, 'A singer that jumped the shark two albums ago' ],
2215 [ 102, 'An actually cool singer' ],
2218 For the arrayref of hashrefs style each hashref should be a structure
2219 suitable for passing to L</create>. Multi-create is also permitted with
2222 $schema->resultset("Artist")->populate([
2223 { artistid => 4, name => 'Manufactured Crap', cds => [
2224 { title => 'My First CD', year => 2006 },
2225 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2228 { artistid => 5, name => 'Angsty-Whiny Girl', cds => [
2229 { title => 'My parents sold me to a record company', year => 2005 },
2230 { title => 'Why Am I So Ugly?', year => 2006 },
2231 { title => 'I Got Surgery and am now Popular', year => 2007 }
2236 If you attempt a void-context multi-create as in the example above (each
2237 Artist also has the related list of CDs), and B<do not> supply the
2238 necessary autoinc foreign key information, this method will proxy to the
2239 less efficient L</create>, and then throw the Result objects away. In this
2240 case there are obviously no benefits to using this method over L</create>.
2247 # cruft placed in standalone method
2248 my $data = $self->_normalize_populate_args(@_);
2250 return unless @$data;
2252 if(defined wantarray) {
2253 my @created = map { $self->create($_) } @$data;
2254 return wantarray ? @created : \@created;
2257 my $first = $data->[0];
2259 # if a column is a registered relationship, and is a non-blessed hash/array, consider
2260 # it relationship data
2261 my (@rels, @columns);
2262 my $rsrc = $self->result_source;
2263 my $rels = { map { $_ => $rsrc->relationship_info($_) } $rsrc->relationships };
2264 for (keys %$first) {
2265 my $ref = ref $first->{$_};
2266 $rels->{$_} && ($ref eq 'ARRAY' or $ref eq 'HASH')
2272 my @pks = $rsrc->primary_columns;
2274 ## do the belongs_to relationships
2275 foreach my $index (0..$#$data) {
2277 # delegate to create() for any dataset without primary keys with specified relationships
2278 if (grep { !defined $data->[$index]->{$_} } @pks ) {
2280 if (grep { ref $data->[$index]{$r} eq $_ } qw/HASH ARRAY/) { # a related set must be a HASH or AoH
2281 my @ret = $self->populate($data);
2287 foreach my $rel (@rels) {
2288 next unless ref $data->[$index]->{$rel} eq "HASH";
2289 my $result = $self->related_resultset($rel)->create($data->[$index]->{$rel});
2290 my ($reverse_relname, $reverse_relinfo) = %{$rsrc->reverse_relationship_info($rel)};
2291 my $related = $result->result_source->_resolve_condition(
2292 $reverse_relinfo->{cond},
2298 delete $data->[$index]->{$rel};
2299 $data->[$index] = {%{$data->[$index]}, %$related};
2301 push @columns, keys %$related if $index == 0;
2305 ## inherit the data locked in the conditions of the resultset
2306 my ($rs_data) = $self->_merge_with_rscond({});
2307 delete @{$rs_data}{@columns};
2309 ## do bulk insert on current row
2310 $rsrc->storage->insert_bulk(
2312 [@columns, keys %$rs_data],
2313 [ map { [ @$_{@columns}, values %$rs_data ] } @$data ],
2316 ## do the has_many relationships
2317 foreach my $item (@$data) {
2321 foreach my $rel (@rels) {
2322 next unless ref $item->{$rel} eq "ARRAY" && @{ $item->{$rel} };
2324 $main_row ||= $self->new_result({map { $_ => $item->{$_} } @pks});
2326 my $child = $main_row->$rel;
2328 my $related = $child->result_source->_resolve_condition(
2329 $rels->{$rel}{cond},
2335 my @rows_to_add = ref $item->{$rel} eq 'ARRAY' ? @{$item->{$rel}} : ($item->{$rel});
2336 my @populate = map { {%$_, %$related} } @rows_to_add;
2338 $child->populate( \@populate );
2345 # populate() arguments went over several incarnations
2346 # What we ultimately support is AoH
2347 sub _normalize_populate_args {
2348 my ($self, $arg) = @_;
2350 if (ref $arg eq 'ARRAY') {
2354 elsif (ref $arg->[0] eq 'HASH') {
2357 elsif (ref $arg->[0] eq 'ARRAY') {
2359 my @colnames = @{$arg->[0]};
2360 foreach my $values (@{$arg}[1 .. $#$arg]) {
2361 push @ret, { map { $colnames[$_] => $values->[$_] } (0 .. $#colnames) };
2367 $self->throw_exception('Populate expects an arrayref of hashrefs or arrayref of arrayrefs');
2374 =item Arguments: none
2376 =item Return Value: L<$pager|Data::Page>
2380 Returns a L<Data::Page> object for the current resultset. Only makes
2381 sense for queries with a C<page> attribute.
2383 To get the full count of entries for a paged resultset, call
2384 C<total_entries> on the L<Data::Page> object.
2391 return $self->{pager} if $self->{pager};
2393 my $attrs = $self->{attrs};
2394 if (!defined $attrs->{page}) {
2395 $self->throw_exception("Can't create pager for non-paged rs");
2397 elsif ($attrs->{page} <= 0) {
2398 $self->throw_exception('Invalid page number (page-numbers are 1-based)');
2400 $attrs->{rows} ||= 10;
2402 # throw away the paging flags and re-run the count (possibly
2403 # with a subselect) to get the real total count
2404 my $count_attrs = { %$attrs };
2405 delete @{$count_attrs}{qw/rows offset page pager/};
2407 my $total_rs = (ref $self)->new($self->result_source, $count_attrs);
2409 require DBIx::Class::ResultSet::Pager;
2410 return $self->{pager} = DBIx::Class::ResultSet::Pager->new(
2411 sub { $total_rs->count }, #lazy-get the total
2413 $self->{attrs}{page},
2421 =item Arguments: $page_number
2423 =item Return Value: L<$resultset|/search>
2427 Returns a resultset for the $page_number page of the resultset on which page
2428 is called, where each page contains a number of rows equal to the 'rows'
2429 attribute set on the resultset (10 by default).
2434 my ($self, $page) = @_;
2435 return (ref $self)->new($self->result_source, { %{$self->{attrs}}, page => $page });
2442 =item Arguments: \%col_data
2444 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2448 Creates a new result object in the resultset's result class and returns
2449 it. The row is not inserted into the database at this point, call
2450 L<DBIx::Class::Row/insert> to do that. Calling L<DBIx::Class::Row/in_storage>
2451 will tell you whether the result object has been inserted or not.
2453 Passes the hashref of input on to L<DBIx::Class::Row/new>.
2458 my ($self, $values) = @_;
2460 $self->throw_exception( "new_result takes only one argument - a hashref of values" )
2463 $self->throw_exception( "new_result expects a hashref" )
2464 unless (ref $values eq 'HASH');
2466 my ($merged_cond, $cols_from_relations) = $self->_merge_with_rscond($values);
2468 my $new = $self->result_class->new({
2470 ( @$cols_from_relations
2471 ? (-cols_from_relations => $cols_from_relations)
2474 -result_source => $self->result_source, # DO NOT REMOVE THIS, REQUIRED
2478 reftype($new) eq 'HASH'
2484 carp_unique (sprintf (
2485 "%s->new returned a blessed empty hashref - a strong indicator something is wrong with its inheritance chain",
2486 $self->result_class,
2493 # _merge_with_rscond
2495 # Takes a simple hash of K/V data and returns its copy merged with the
2496 # condition already present on the resultset. Additionally returns an
2497 # arrayref of value/condition names, which were inferred from related
2498 # objects (this is needed for in-memory related objects)
2499 sub _merge_with_rscond {
2500 my ($self, $data) = @_;
2502 my (%new_data, @cols_from_relations);
2504 my $alias = $self->{attrs}{alias};
2506 if (! defined $self->{cond}) {
2507 # just massage $data below
2509 elsif ($self->{cond} eq $DBIx::Class::ResultSource::UNRESOLVABLE_CONDITION) {
2510 %new_data = %{ $self->{attrs}{related_objects} || {} }; # nothing might have been inserted yet
2511 @cols_from_relations = keys %new_data;
2513 elsif (ref $self->{cond} ne 'HASH') {
2514 $self->throw_exception(
2515 "Can't abstract implicit construct, resultset condition not a hash"
2519 # precedence must be given to passed values over values inherited from
2520 # the cond, so the order here is important.
2521 my $collapsed_cond = $self->_collapse_cond($self->{cond});
2522 my %implied = %{$self->_remove_alias($collapsed_cond, $alias)};
2524 while ( my($col, $value) = each %implied ) {
2525 my $vref = ref $value;
2531 (keys %$value)[0] eq '='
2533 $new_data{$col} = $value->{'='};
2535 elsif( !$vref or $vref eq 'SCALAR' or blessed($value) ) {
2536 $new_data{$col} = $value;
2543 %{ $self->_remove_alias($data, $alias) },
2546 return (\%new_data, \@cols_from_relations);
2549 # _has_resolved_attr
2551 # determines if the resultset defines at least one
2552 # of the attributes supplied
2554 # used to determine if a subquery is necessary
2556 # supports some virtual attributes:
2558 # This will scan for any joins being present on the resultset.
2559 # It is not a mere key-search but a deep inspection of {from}
2562 sub _has_resolved_attr {
2563 my ($self, @attr_names) = @_;
2565 my $attrs = $self->_resolved_attrs;
2569 for my $n (@attr_names) {
2570 if (grep { $n eq $_ } (qw/-join/) ) {
2571 $extra_checks{$n}++;
2575 my $attr = $attrs->{$n};
2577 next if not defined $attr;
2579 if (ref $attr eq 'HASH') {
2580 return 1 if keys %$attr;
2582 elsif (ref $attr eq 'ARRAY') {
2590 # a resolved join is expressed as a multi-level from
2592 $extra_checks{-join}
2594 ref $attrs->{from} eq 'ARRAY'
2596 @{$attrs->{from}} > 1
2604 # Recursively collapse the condition.
2606 sub _collapse_cond {
2607 my ($self, $cond, $collapsed) = @_;
2611 if (ref $cond eq 'ARRAY') {
2612 foreach my $subcond (@$cond) {
2613 next unless ref $subcond; # -or
2614 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2617 elsif (ref $cond eq 'HASH') {
2618 if (keys %$cond and (keys %$cond)[0] eq '-and') {
2619 foreach my $subcond (@{$cond->{-and}}) {
2620 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2624 foreach my $col (keys %$cond) {
2625 my $value = $cond->{$col};
2626 $collapsed->{$col} = $value;
2636 # Remove the specified alias from the specified query hash. A copy is made so
2637 # the original query is not modified.
2640 my ($self, $query, $alias) = @_;
2642 my %orig = %{ $query || {} };
2645 foreach my $key (keys %orig) {
2647 $unaliased{$key} = $orig{$key};
2650 $unaliased{$1} = $orig{$key}
2651 if $key =~ m/^(?:\Q$alias\E\.)?([^.]+)$/;
2661 =item Arguments: none
2663 =item Return Value: \[ $sql, L<@bind_values|/DBIC BIND VALUES> ]
2667 Returns the SQL query and bind vars associated with the invocant.
2669 This is generally used as the RHS for a subquery.
2676 my $attrs = { %{ $self->_resolved_attrs } };
2678 my $aq = $self->result_source->storage->_select_args_to_query (
2679 $attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs
2682 $self->{_attrs}{_sqlmaker_select_args} = $attrs->{_sqlmaker_select_args};
2691 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2693 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2697 my $artist = $schema->resultset('Artist')->find_or_new(
2698 { artist => 'fred' }, { key => 'artists' });
2700 $cd->cd_to_producer->find_or_new({ producer => $producer },
2701 { key => 'primary' });
2703 Find an existing record from this resultset using L</find>. if none exists,
2704 instantiate a new result object and return it. The object will not be saved
2705 into your storage until you call L<DBIx::Class::Row/insert> on it.
2707 You most likely want this method when looking for existing rows using a unique
2708 constraint that is not the primary key, or looking for related rows.
2710 If you want objects to be saved immediately, use L</find_or_create> instead.
2712 B<Note>: Make sure to read the documentation of L</find> and understand the
2713 significance of the C<key> attribute, as its lack may skew your search, and
2714 subsequently result in spurious new objects.
2716 B<Note>: Take care when using C<find_or_new> with a table having
2717 columns with default values that you intend to be automatically
2718 supplied by the database (e.g. an auto_increment primary key column).
2719 In normal usage, the value of such columns should NOT be included at
2720 all in the call to C<find_or_new>, even when set to C<undef>.
2726 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2727 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2728 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2731 return $self->new_result($hash);
2738 =item Arguments: \%col_data
2740 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2744 Attempt to create a single new row or a row with multiple related rows
2745 in the table represented by the resultset (and related tables). This
2746 will not check for duplicate rows before inserting, use
2747 L</find_or_create> to do that.
2749 To create one row for this resultset, pass a hashref of key/value
2750 pairs representing the columns of the table and the values you wish to
2751 store. If the appropriate relationships are set up, foreign key fields
2752 can also be passed an object representing the foreign row, and the
2753 value will be set to its primary key.
2755 To create related objects, pass a hashref of related-object column values
2756 B<keyed on the relationship name>. If the relationship is of type C<multi>
2757 (L<DBIx::Class::Relationship/has_many>) - pass an arrayref of hashrefs.
2758 The process will correctly identify columns holding foreign keys, and will
2759 transparently populate them from the keys of the corresponding relation.
2760 This can be applied recursively, and will work correctly for a structure
2761 with an arbitrary depth and width, as long as the relationships actually
2762 exists and the correct column data has been supplied.
2764 Instead of hashrefs of plain related data (key/value pairs), you may
2765 also pass new or inserted objects. New objects (not inserted yet, see
2766 L</new_result>), will be inserted into their appropriate tables.
2768 Effectively a shortcut for C<< ->new_result(\%col_data)->insert >>.
2770 Example of creating a new row.
2772 $person_rs->create({
2773 name=>"Some Person",
2774 email=>"somebody@someplace.com"
2777 Example of creating a new row and also creating rows in a related C<has_many>
2778 or C<has_one> resultset. Note Arrayref.
2781 { artistid => 4, name => 'Manufactured Crap', cds => [
2782 { title => 'My First CD', year => 2006 },
2783 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2788 Example of creating a new row and also creating a row in a related
2789 C<belongs_to> resultset. Note Hashref.
2792 title=>"Music for Silly Walks",
2795 name=>"Silly Musician",
2803 When subclassing ResultSet never attempt to override this method. Since
2804 it is a simple shortcut for C<< $self->new_result($attrs)->insert >>, a
2805 lot of the internals simply never call it, so your override will be
2806 bypassed more often than not. Override either L<DBIx::Class::Row/new>
2807 or L<DBIx::Class::Row/insert> depending on how early in the
2808 L</create> process you need to intervene. See also warning pertaining to
2816 my ($self, $col_data) = @_;
2817 $self->throw_exception( "create needs a hashref" )
2818 unless ref $col_data eq 'HASH';
2819 return $self->new_result($col_data)->insert;
2822 =head2 find_or_create
2826 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2828 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2832 $cd->cd_to_producer->find_or_create({ producer => $producer },
2833 { key => 'primary' });
2835 Tries to find a record based on its primary key or unique constraints; if none
2836 is found, creates one and returns that instead.
2838 my $cd = $schema->resultset('CD')->find_or_create({
2840 artist => 'Massive Attack',
2841 title => 'Mezzanine',
2845 Also takes an optional C<key> attribute, to search by a specific key or unique
2846 constraint. For example:
2848 my $cd = $schema->resultset('CD')->find_or_create(
2850 artist => 'Massive Attack',
2851 title => 'Mezzanine',
2853 { key => 'cd_artist_title' }
2856 B<Note>: Make sure to read the documentation of L</find> and understand the
2857 significance of the C<key> attribute, as its lack may skew your search, and
2858 subsequently result in spurious row creation.
2860 B<Note>: Because find_or_create() reads from the database and then
2861 possibly inserts based on the result, this method is subject to a race
2862 condition. Another process could create a record in the table after
2863 the find has completed and before the create has started. To avoid
2864 this problem, use find_or_create() inside a transaction.
2866 B<Note>: Take care when using C<find_or_create> with a table having
2867 columns with default values that you intend to be automatically
2868 supplied by the database (e.g. an auto_increment primary key column).
2869 In normal usage, the value of such columns should NOT be included at
2870 all in the call to C<find_or_create>, even when set to C<undef>.
2872 See also L</find> and L</update_or_create>. For information on how to declare
2873 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2875 If you need to know if an existing row was found or a new one created use
2876 L</find_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
2877 to call L<DBIx::Class::Row/insert> to save the newly created row to the
2880 my $cd = $schema->resultset('CD')->find_or_new({
2882 artist => 'Massive Attack',
2883 title => 'Mezzanine',
2887 if( !$cd->in_storage ) {
2894 sub find_or_create {
2896 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2897 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2898 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2901 return $self->create($hash);
2904 =head2 update_or_create
2908 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2910 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2914 $resultset->update_or_create({ col => $val, ... });
2916 Like L</find_or_create>, but if a row is found it is immediately updated via
2917 C<< $found_row->update (\%col_data) >>.
2920 Takes an optional C<key> attribute to search on a specific unique constraint.
2923 # In your application
2924 my $cd = $schema->resultset('CD')->update_or_create(
2926 artist => 'Massive Attack',
2927 title => 'Mezzanine',
2930 { key => 'cd_artist_title' }
2933 $cd->cd_to_producer->update_or_create({
2934 producer => $producer,
2940 B<Note>: Make sure to read the documentation of L</find> and understand the
2941 significance of the C<key> attribute, as its lack may skew your search, and
2942 subsequently result in spurious row creation.
2944 B<Note>: Take care when using C<update_or_create> with a table having
2945 columns with default values that you intend to be automatically
2946 supplied by the database (e.g. an auto_increment primary key column).
2947 In normal usage, the value of such columns should NOT be included at
2948 all in the call to C<update_or_create>, even when set to C<undef>.
2950 See also L</find> and L</find_or_create>. For information on how to declare
2951 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2953 If you need to know if an existing row was updated or a new one created use
2954 L</update_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
2955 to call L<DBIx::Class::Row/insert> to save the newly created row to the
2960 sub update_or_create {
2962 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2963 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2965 my $row = $self->find($cond, $attrs);
2967 $row->update($cond);
2971 return $self->create($cond);
2974 =head2 update_or_new
2978 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2980 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2984 $resultset->update_or_new({ col => $val, ... });
2986 Like L</find_or_new> but if a row is found it is immediately updated via
2987 C<< $found_row->update (\%col_data) >>.
2991 # In your application
2992 my $cd = $schema->resultset('CD')->update_or_new(
2994 artist => 'Massive Attack',
2995 title => 'Mezzanine',
2998 { key => 'cd_artist_title' }
3001 if ($cd->in_storage) {
3002 # the cd was updated
3005 # the cd is not yet in the database, let's insert it
3009 B<Note>: Make sure to read the documentation of L</find> and understand the
3010 significance of the C<key> attribute, as its lack may skew your search, and
3011 subsequently result in spurious new objects.
3013 B<Note>: Take care when using C<update_or_new> with a table having
3014 columns with default values that you intend to be automatically
3015 supplied by the database (e.g. an auto_increment primary key column).
3016 In normal usage, the value of such columns should NOT be included at
3017 all in the call to C<update_or_new>, even when set to C<undef>.
3019 See also L</find>, L</find_or_create> and L</find_or_new>.
3025 my $attrs = ( @_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {} );
3026 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
3028 my $row = $self->find( $cond, $attrs );
3029 if ( defined $row ) {
3030 $row->update($cond);
3034 return $self->new_result($cond);
3041 =item Arguments: none
3043 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass> | undef
3047 Gets the contents of the cache for the resultset, if the cache is set.
3049 The cache is populated either by using the L</prefetch> attribute to
3050 L</search> or by calling L</set_cache>.
3062 =item Arguments: L<\@result_objs|DBIx::Class::Manual::ResultClass>
3064 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass>
3068 Sets the contents of the cache for the resultset. Expects an arrayref
3069 of objects of the same class as those produced by the resultset. Note that
3070 if the cache is set, the resultset will return the cached objects rather
3071 than re-querying the database even if the cache attr is not set.
3073 The contents of the cache can also be populated by using the
3074 L</prefetch> attribute to L</search>.
3079 my ( $self, $data ) = @_;
3080 $self->throw_exception("set_cache requires an arrayref")
3081 if defined($data) && (ref $data ne 'ARRAY');
3082 $self->{all_cache} = $data;
3089 =item Arguments: none
3091 =item Return Value: undef
3095 Clears the cache for the resultset.
3100 shift->set_cache(undef);
3107 =item Arguments: none
3109 =item Return Value: true, if the resultset has been paginated
3117 return !!$self->{attrs}{page};
3124 =item Arguments: none
3126 =item Return Value: true, if the resultset has been ordered with C<order_by>.
3134 return scalar $self->result_source->storage->_extract_order_criteria($self->{attrs}{order_by});
3137 =head2 related_resultset
3141 =item Arguments: $rel_name
3143 =item Return Value: L<$resultset|/search>
3147 Returns a related resultset for the supplied relationship name.
3149 $artist_rs = $schema->resultset('CD')->related_resultset('Artist');
3153 sub related_resultset {
3154 my ($self, $rel) = @_;
3156 return $self->{related_resultsets}{$rel}
3157 if defined $self->{related_resultsets}{$rel};
3159 return $self->{related_resultsets}{$rel} = do {
3160 my $rsrc = $self->result_source;
3161 my $rel_info = $rsrc->relationship_info($rel);
3163 $self->throw_exception(
3164 "search_related: result source '" . $rsrc->source_name .
3165 "' has no such relationship $rel")
3168 my $attrs = $self->_chain_relationship($rel);
3170 my $join_count = $attrs->{seen_join}{$rel};
3172 my $alias = $self->result_source->storage
3173 ->relname_to_table_alias($rel, $join_count);
3175 # since this is search_related, and we already slid the select window inwards
3176 # (the select/as attrs were deleted in the beginning), we need to flip all
3177 # left joins to inner, so we get the expected results
3178 # read the comment on top of the actual function to see what this does
3179 $attrs->{from} = $rsrc->schema->storage->_inner_join_to_node ($attrs->{from}, $alias);
3182 #XXX - temp fix for result_class bug. There likely is a more elegant fix -groditi
3183 delete @{$attrs}{qw(result_class alias)};
3185 my $rel_source = $rsrc->related_source($rel);
3189 # The reason we do this now instead of passing the alias to the
3190 # search_rs below is that if you wrap/overload resultset on the
3191 # source you need to know what alias it's -going- to have for things
3192 # to work sanely (e.g. RestrictWithObject wants to be able to add
3193 # extra query restrictions, and these may need to be $alias.)
3195 my $rel_attrs = $rel_source->resultset_attributes;
3196 local $rel_attrs->{alias} = $alias;
3198 $rel_source->resultset
3202 where => $attrs->{where},
3206 if (my $cache = $self->get_cache) {
3207 my @related_cache = map
3208 { @{$_->related_resultset($rel)->get_cache||[]} }
3212 $new->set_cache(\@related_cache) if @related_cache;
3219 =head2 current_source_alias
3223 =item Arguments: none
3225 =item Return Value: $source_alias
3229 Returns the current table alias for the result source this resultset is built
3230 on, that will be used in the SQL query. Usually it is C<me>.
3232 Currently the source alias that refers to the result set returned by a
3233 L</search>/L</find> family method depends on how you got to the resultset: it's
3234 C<me> by default, but eg. L</search_related> aliases it to the related result
3235 source name (and keeps C<me> referring to the original result set). The long
3236 term goal is to make L<DBIx::Class> always alias the current resultset as C<me>
3237 (and make this method unnecessary).
3239 Thus it's currently necessary to use this method in predefined queries (see
3240 L<DBIx::Class::Manual::Cookbook/Predefined searches>) when referring to the
3241 source alias of the current result set:
3243 # in a result set class
3245 my ($self, $user) = @_;
3247 my $me = $self->current_source_alias;
3249 return $self->search({
3250 "$me.modified" => $user->id,
3256 sub current_source_alias {
3257 return (shift->{attrs} || {})->{alias} || 'me';
3260 =head2 as_subselect_rs
3264 =item Arguments: none
3266 =item Return Value: L<$resultset|/search>
3270 Act as a barrier to SQL symbols. The resultset provided will be made into a
3271 "virtual view" by including it as a subquery within the from clause. From this
3272 point on, any joined tables are inaccessible to ->search on the resultset (as if
3273 it were simply where-filtered without joins). For example:
3275 my $rs = $schema->resultset('Bar')->search({'x.name' => 'abc'},{ join => 'x' });
3277 # 'x' now pollutes the query namespace
3279 # So the following works as expected
3280 my $ok_rs = $rs->search({'x.other' => 1});
3282 # But this doesn't: instead of finding a 'Bar' related to two x rows (abc and
3283 # def) we look for one row with contradictory terms and join in another table
3284 # (aliased 'x_2') which we never use
3285 my $broken_rs = $rs->search({'x.name' => 'def'});
3287 my $rs2 = $rs->as_subselect_rs;
3289 # doesn't work - 'x' is no longer accessible in $rs2, having been sealed away
3290 my $not_joined_rs = $rs2->search({'x.other' => 1});
3292 # works as expected: finds a 'table' row related to two x rows (abc and def)
3293 my $correctly_joined_rs = $rs2->search({'x.name' => 'def'});
3295 Another example of when one might use this would be to select a subset of
3296 columns in a group by clause:
3298 my $rs = $schema->resultset('Bar')->search(undef, {
3299 group_by => [qw{ id foo_id baz_id }],
3300 })->as_subselect_rs->search(undef, {
3301 columns => [qw{ id foo_id }]
3304 In the above example normally columns would have to be equal to the group by,
3305 but because we isolated the group by into a subselect the above works.
3309 sub as_subselect_rs {
3312 my $attrs = $self->_resolved_attrs;
3314 my $fresh_rs = (ref $self)->new (
3315 $self->result_source
3318 # these pieces will be locked in the subquery
3319 delete $fresh_rs->{cond};
3320 delete @{$fresh_rs->{attrs}}{qw/where bind/};
3322 return $fresh_rs->search( {}, {
3324 $attrs->{alias} => $self->as_query,
3325 -alias => $attrs->{alias},
3326 -rsrc => $self->result_source,
3328 alias => $attrs->{alias},
3332 # This code is called by search_related, and makes sure there
3333 # is clear separation between the joins before, during, and
3334 # after the relationship. This information is needed later
3335 # in order to properly resolve prefetch aliases (any alias
3336 # with a relation_chain_depth less than the depth of the
3337 # current prefetch is not considered)
3339 # The increments happen twice per join. An even number means a
3340 # relationship specified via a search_related, whereas an odd
3341 # number indicates a join/prefetch added via attributes
3343 # Also this code will wrap the current resultset (the one we
3344 # chain to) in a subselect IFF it contains limiting attributes
3345 sub _chain_relationship {
3346 my ($self, $rel) = @_;
3347 my $source = $self->result_source;
3348 my $attrs = { %{$self->{attrs}||{}} };
3350 # we need to take the prefetch the attrs into account before we
3351 # ->_resolve_join as otherwise they get lost - captainL
3352 my $join = $self->_merge_joinpref_attr( $attrs->{join}, $attrs->{prefetch} );
3354 delete @{$attrs}{qw/join prefetch collapse group_by distinct _grouped_by_distinct select as columns +select +as +columns/};
3356 my $seen = { %{ (delete $attrs->{seen_join}) || {} } };
3359 my @force_subq_attrs = qw/offset rows group_by having/;
3362 ($attrs->{from} && ref $attrs->{from} ne 'ARRAY')
3364 $self->_has_resolved_attr (@force_subq_attrs)
3366 # Nuke the prefetch (if any) before the new $rs attrs
3367 # are resolved (prefetch is useless - we are wrapping
3368 # a subquery anyway).
3369 my $rs_copy = $self->search;
3370 $rs_copy->{attrs}{join} = $self->_merge_joinpref_attr (
3371 $rs_copy->{attrs}{join},
3372 delete $rs_copy->{attrs}{prefetch},
3377 -alias => $attrs->{alias},
3378 $attrs->{alias} => $rs_copy->as_query,
3380 delete @{$attrs}{@force_subq_attrs, qw/where bind/};
3381 $seen->{-relation_chain_depth} = 0;
3383 elsif ($attrs->{from}) { #shallow copy suffices
3384 $from = [ @{$attrs->{from}} ];
3389 -alias => $attrs->{alias},
3390 $attrs->{alias} => $source->from,
3394 my $jpath = ($seen->{-relation_chain_depth})
3395 ? $from->[-1][0]{-join_path}
3398 my @requested_joins = $source->_resolve_join(
3405 push @$from, @requested_joins;
3407 $seen->{-relation_chain_depth}++;
3409 # if $self already had a join/prefetch specified on it, the requested
3410 # $rel might very well be already included. What we do in this case
3411 # is effectively a no-op (except that we bump up the chain_depth on
3412 # the join in question so we could tell it *is* the search_related)
3415 # we consider the last one thus reverse
3416 for my $j (reverse @requested_joins) {
3417 my ($last_j) = keys %{$j->[0]{-join_path}[-1]};
3418 if ($rel eq $last_j) {
3419 $j->[0]{-relation_chain_depth}++;
3425 unless ($already_joined) {
3426 push @$from, $source->_resolve_join(
3434 $seen->{-relation_chain_depth}++;
3436 return {%$attrs, from => $from, seen_join => $seen};
3439 sub _resolved_attrs {
3441 return $self->{_attrs} if $self->{_attrs};
3443 my $attrs = { %{ $self->{attrs} || {} } };
3444 my $source = $self->result_source;
3445 my $alias = $attrs->{alias};
3447 # default selection list
3448 $attrs->{columns} = [ $source->columns ]
3449 unless List::Util::first { exists $attrs->{$_} } qw/columns cols select as/;
3451 # merge selectors together
3452 for (qw/columns select as/) {
3453 $attrs->{$_} = $self->_merge_attr($attrs->{$_}, delete $attrs->{"+$_"})
3454 if $attrs->{$_} or $attrs->{"+$_"};
3457 # disassemble columns
3459 if (my $cols = delete $attrs->{columns}) {
3460 for my $c (ref $cols eq 'ARRAY' ? @$cols : $cols) {
3461 if (ref $c eq 'HASH') {
3462 for my $as (sort keys %$c) {
3463 push @sel, $c->{$as};
3474 # when trying to weed off duplicates later do not go past this point -
3475 # everything added from here on is unbalanced "anyone's guess" stuff
3476 my $dedup_stop_idx = $#as;
3478 push @as, @{ ref $attrs->{as} eq 'ARRAY' ? $attrs->{as} : [ $attrs->{as} ] }
3480 push @sel, @{ ref $attrs->{select} eq 'ARRAY' ? $attrs->{select} : [ $attrs->{select} ] }
3481 if $attrs->{select};
3483 # assume all unqualified selectors to apply to the current alias (legacy stuff)
3484 $_ = (ref $_ or $_ =~ /\./) ? $_ : "$alias.$_" for @sel;
3486 # disqualify all $alias.col as-bits (inflate-map mandated)
3487 $_ = ($_ =~ /^\Q$alias.\E(.+)$/) ? $1 : $_ for @as;
3489 # de-duplicate the result (remove *identical* select/as pairs)
3490 # and also die on duplicate {as} pointing to different {select}s
3491 # not using a c-style for as the condition is prone to shrinkage
3494 while ($i <= $dedup_stop_idx) {
3495 if ($seen->{"$sel[$i] \x00\x00 $as[$i]"}++) {
3500 elsif ($seen->{$as[$i]}++) {
3501 $self->throw_exception(
3502 "inflate_result() alias '$as[$i]' specified twice with different SQL-side {select}-ors"
3510 $attrs->{select} = \@sel;
3511 $attrs->{as} = \@as;
3513 $attrs->{from} ||= [{
3515 -alias => $self->{attrs}{alias},
3516 $self->{attrs}{alias} => $source->from,
3519 if ( $attrs->{join} || $attrs->{prefetch} ) {
3521 $self->throw_exception ('join/prefetch can not be used with a custom {from}')
3522 if ref $attrs->{from} ne 'ARRAY';
3524 my $join = (delete $attrs->{join}) || {};
3526 if ( defined $attrs->{prefetch} ) {
3527 $join = $self->_merge_joinpref_attr( $join, $attrs->{prefetch} );
3530 $attrs->{from} = # have to copy here to avoid corrupting the original
3532 @{ $attrs->{from} },
3533 $source->_resolve_join(
3536 { %{ $attrs->{seen_join} || {} } },
3537 ( $attrs->{seen_join} && keys %{$attrs->{seen_join}})
3538 ? $attrs->{from}[-1][0]{-join_path}
3545 if ( defined $attrs->{order_by} ) {
3546 $attrs->{order_by} = (
3547 ref( $attrs->{order_by} ) eq 'ARRAY'
3548 ? [ @{ $attrs->{order_by} } ]
3549 : [ $attrs->{order_by} || () ]
3553 if ($attrs->{group_by} and ref $attrs->{group_by} ne 'ARRAY') {
3554 $attrs->{group_by} = [ $attrs->{group_by} ];
3557 # generate the distinct induced group_by early, as prefetch will be carried via a
3558 # subquery (since a group_by is present)
3559 if (delete $attrs->{distinct}) {
3560 if ($attrs->{group_by}) {
3561 carp_unique ("Useless use of distinct on a grouped resultset ('distinct' is ignored when a 'group_by' is present)");
3564 $attrs->{_grouped_by_distinct} = 1;
3565 # distinct affects only the main selection part, not what prefetch may
3567 $attrs->{group_by} = $source->storage->_group_over_selection($attrs);
3571 # generate selections based on the prefetch helper
3573 $prefetch = $self->_merge_joinpref_attr( {}, delete $attrs->{prefetch} )
3574 if defined $attrs->{prefetch};
3578 $self->throw_exception("Unable to prefetch, resultset contains an unnamed selector $attrs->{_dark_selector}{string}")
3579 if $attrs->{_dark_selector};
3581 $attrs->{collapse} = 1;
3583 # this is a separate structure (we don't look in {from} directly)
3584 # as the resolver needs to shift things off the lists to work
3585 # properly (identical-prefetches on different branches)
3587 if (ref $attrs->{from} eq 'ARRAY') {
3589 my $start_depth = $attrs->{seen_join}{-relation_chain_depth} || 0;
3591 for my $j ( @{$attrs->{from}}[1 .. $#{$attrs->{from}} ] ) {
3592 next unless $j->[0]{-alias};
3593 next unless $j->[0]{-join_path};
3594 next if ($j->[0]{-relation_chain_depth} || 0) < $start_depth;
3596 my @jpath = map { keys %$_ } @{$j->[0]{-join_path}};
3599 $p = $p->{$_} ||= {} for @jpath[ ($start_depth/2) .. $#jpath]; #only even depths are actual jpath boundaries
3600 push @{$p->{-join_aliases} }, $j->[0]{-alias};
3604 my @prefetch = $source->_resolve_prefetch( $prefetch, $alias, $join_map );
3606 push @{ $attrs->{select} }, (map { $_->[0] } @prefetch);
3607 push @{ $attrs->{as} }, (map { $_->[1] } @prefetch);
3610 if ( List::Util::first { $_ =~ /\./ } @{$attrs->{as}} ) {
3611 $attrs->{_related_results_construction} = 1;
3614 # run through the resulting joinstructure (starting from our current slot)
3615 # and unset collapse if proven unnecessary
3617 # also while we are at it find out if the current root source has
3618 # been premultiplied by previous related_source chaining
3620 # this allows to predict whether a root object with all other relation
3621 # data set to NULL is in fact unique
3622 if ($attrs->{collapse}) {
3624 if (ref $attrs->{from} eq 'ARRAY') {
3626 if (@{$attrs->{from}} == 1) {
3627 # no joins - no collapse
3628 $attrs->{collapse} = 0;
3631 # find where our table-spec starts
3632 my @fromlist = @{$attrs->{from}};
3634 my $t = shift @fromlist;
3637 # me vs join from-spec distinction - a ref means non-root
3638 if (ref $t eq 'ARRAY') {
3640 $is_multi ||= ! $t->{-is_single};
3642 last if ($t->{-alias} && $t->{-alias} eq $alias);
3643 $attrs->{_main_source_premultiplied} ||= $is_multi;
3646 # no non-singles remaining, nor any premultiplication - nothing to collapse
3648 ! $attrs->{_main_source_premultiplied}
3650 ! List::Util::first { ! $_->[0]{-is_single} } @fromlist
3652 $attrs->{collapse} = 0;
3658 # if we can not analyze the from - err on the side of safety
3659 $attrs->{_main_source_premultiplied} = 1;
3663 # if both page and offset are specified, produce a combined offset
3664 # even though it doesn't make much sense, this is what pre 081xx has
3666 if (my $page = delete $attrs->{page}) {
3668 ($attrs->{rows} * ($page - 1))
3670 ($attrs->{offset} || 0)
3674 return $self->{_attrs} = $attrs;
3678 my ($self, $attr) = @_;
3680 if (ref $attr eq 'HASH') {
3681 return $self->_rollout_hash($attr);
3682 } elsif (ref $attr eq 'ARRAY') {
3683 return $self->_rollout_array($attr);
3689 sub _rollout_array {
3690 my ($self, $attr) = @_;
3693 foreach my $element (@{$attr}) {
3694 if (ref $element eq 'HASH') {
3695 push( @rolled_array, @{ $self->_rollout_hash( $element ) } );
3696 } elsif (ref $element eq 'ARRAY') {
3697 # XXX - should probably recurse here
3698 push( @rolled_array, @{$self->_rollout_array($element)} );
3700 push( @rolled_array, $element );
3703 return \@rolled_array;
3707 my ($self, $attr) = @_;
3710 foreach my $key (keys %{$attr}) {
3711 push( @rolled_array, { $key => $attr->{$key} } );
3713 return \@rolled_array;
3716 sub _calculate_score {
3717 my ($self, $a, $b) = @_;
3719 if (defined $a xor defined $b) {
3722 elsif (not defined $a) {
3726 if (ref $b eq 'HASH') {
3727 my ($b_key) = keys %{$b};
3728 if (ref $a eq 'HASH') {
3729 my ($a_key) = keys %{$a};
3730 if ($a_key eq $b_key) {
3731 return (1 + $self->_calculate_score( $a->{$a_key}, $b->{$b_key} ));
3736 return ($a eq $b_key) ? 1 : 0;
3739 if (ref $a eq 'HASH') {
3740 my ($a_key) = keys %{$a};
3741 return ($b eq $a_key) ? 1 : 0;
3743 return ($b eq $a) ? 1 : 0;
3748 sub _merge_joinpref_attr {
3749 my ($self, $orig, $import) = @_;
3751 return $import unless defined($orig);
3752 return $orig unless defined($import);
3754 $orig = $self->_rollout_attr($orig);
3755 $import = $self->_rollout_attr($import);
3758 foreach my $import_element ( @{$import} ) {
3759 # find best candidate from $orig to merge $b_element into
3760 my $best_candidate = { position => undef, score => 0 }; my $position = 0;
3761 foreach my $orig_element ( @{$orig} ) {
3762 my $score = $self->_calculate_score( $orig_element, $import_element );
3763 if ($score > $best_candidate->{score}) {
3764 $best_candidate->{position} = $position;
3765 $best_candidate->{score} = $score;
3769 my ($import_key) = ( ref $import_element eq 'HASH' ) ? keys %{$import_element} : ($import_element);
3770 $import_key = '' if not defined $import_key;
3772 if ($best_candidate->{score} == 0 || exists $seen_keys->{$import_key}) {
3773 push( @{$orig}, $import_element );
3775 my $orig_best = $orig->[$best_candidate->{position}];
3776 # merge orig_best and b_element together and replace original with merged
3777 if (ref $orig_best ne 'HASH') {
3778 $orig->[$best_candidate->{position}] = $import_element;
3779 } elsif (ref $import_element eq 'HASH') {
3780 my ($key) = keys %{$orig_best};
3781 $orig->[$best_candidate->{position}] = { $key => $self->_merge_joinpref_attr($orig_best->{$key}, $import_element->{$key}) };
3784 $seen_keys->{$import_key} = 1; # don't merge the same key twice
3787 return @$orig ? $orig : ();
3795 require Hash::Merge;
3796 my $hm = Hash::Merge->new;
3798 $hm->specify_behavior({
3801 my ($defl, $defr) = map { defined $_ } (@_[0,1]);
3803 if ($defl xor $defr) {
3804 return [ $defl ? $_[0] : $_[1] ];
3809 elsif (__HM_DEDUP and $_[0] eq $_[1]) {
3813 return [$_[0], $_[1]];
3817 return $_[1] if !defined $_[0];
3818 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3819 return [$_[0], @{$_[1]}]
3822 return [] if !defined $_[0] and !keys %{$_[1]};
3823 return [ $_[1] ] if !defined $_[0];
3824 return [ $_[0] ] if !keys %{$_[1]};
3825 return [$_[0], $_[1]]
3830 return $_[0] if !defined $_[1];
3831 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3832 return [@{$_[0]}, $_[1]]
3835 my @ret = @{$_[0]} or return $_[1];
3836 return [ @ret, @{$_[1]} ] unless __HM_DEDUP;
3837 my %idx = map { $_ => 1 } @ret;
3838 push @ret, grep { ! defined $idx{$_} } (@{$_[1]});
3842 return [ $_[1] ] if ! @{$_[0]};
3843 return $_[0] if !keys %{$_[1]};
3844 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3845 return [ @{$_[0]}, $_[1] ];
3850 return [] if !keys %{$_[0]} and !defined $_[1];
3851 return [ $_[0] ] if !defined $_[1];
3852 return [ $_[1] ] if !keys %{$_[0]};
3853 return [$_[0], $_[1]]
3856 return [] if !keys %{$_[0]} and !@{$_[1]};
3857 return [ $_[0] ] if !@{$_[1]};
3858 return $_[1] if !keys %{$_[0]};
3859 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3860 return [ $_[0], @{$_[1]} ];
3863 return [] if !keys %{$_[0]} and !keys %{$_[1]};
3864 return [ $_[0] ] if !keys %{$_[1]};
3865 return [ $_[1] ] if !keys %{$_[0]};
3866 return [ $_[0] ] if $_[0] eq $_[1];
3867 return [ $_[0], $_[1] ];
3870 } => 'DBIC_RS_ATTR_MERGER');
3874 return $hm->merge ($_[1], $_[2]);
3878 sub STORABLE_freeze {
3879 my ($self, $cloning) = @_;
3880 my $to_serialize = { %$self };
3882 # A cursor in progress can't be serialized (and would make little sense anyway)
3883 # the parser can be regenerated (and can't be serialized)
3884 delete @{$to_serialize}{qw/cursor _row_parser _result_inflator/};
3886 # nor is it sensical to store a not-yet-fired-count pager
3887 if ($to_serialize->{pager} and ref $to_serialize->{pager}{total_entries} eq 'CODE') {
3888 delete $to_serialize->{pager};
3891 Storable::nfreeze($to_serialize);
3894 # need this hook for symmetry
3896 my ($self, $cloning, $serialized) = @_;
3898 %$self = %{ Storable::thaw($serialized) };
3904 =head2 throw_exception
3906 See L<DBIx::Class::Schema/throw_exception> for details.
3910 sub throw_exception {
3913 if (ref $self and my $rsrc = $self->result_source) {
3914 $rsrc->throw_exception(@_)
3917 DBIx::Class::Exception->throw(@_);
3925 # XXX: FIXME: Attributes docs need clearing up
3929 Attributes are used to refine a ResultSet in various ways when
3930 searching for data. They can be passed to any method which takes an
3931 C<\%attrs> argument. See L</search>, L</search_rs>, L</find>,
3934 Default attributes can be set on the result class using
3935 L<DBIx::Class::ResultSource/resultset_attributes>. (Please read
3936 the CAVEATS on that feature before using it!)
3938 These are in no particular order:
3944 =item Value: ( $order_by | \@order_by | \%order_by )
3948 Which column(s) to order the results by.
3950 [The full list of suitable values is documented in
3951 L<SQL::Abstract/"ORDER BY CLAUSES">; the following is a summary of
3954 If a single column name, or an arrayref of names is supplied, the
3955 argument is passed through directly to SQL. The hashref syntax allows
3956 for connection-agnostic specification of ordering direction:
3958 For descending order:
3960 order_by => { -desc => [qw/col1 col2 col3/] }
3962 For explicit ascending order:
3964 order_by => { -asc => 'col' }
3966 The old scalarref syntax (i.e. order_by => \'year DESC') is still
3967 supported, although you are strongly encouraged to use the hashref
3968 syntax as outlined above.
3974 =item Value: \@columns | \%columns | $column
3978 Shortcut to request a particular set of columns to be retrieved. Each
3979 column spec may be a string (a table column name), or a hash (in which
3980 case the key is the C<as> value, and the value is used as the C<select>
3981 expression). Adds C<me.> onto the start of any column without a C<.> in
3982 it and sets C<select> from that, then auto-populates C<as> from
3983 C<select> as normal. (You may also use the C<cols> attribute, as in
3984 earlier versions of DBIC, but this is deprecated.)
3986 Essentially C<columns> does the same as L</select> and L</as>.
3988 columns => [ 'foo', { bar => 'baz' } ]
3992 select => [qw/foo baz/],
3999 =item Value: \@columns
4003 Indicates additional columns to be selected from storage. Works the same as
4004 L</columns> but adds columns to the selection. (You may also use the
4005 C<include_columns> attribute, as in earlier versions of DBIC, but this is
4006 deprecated). For example:-
4008 $schema->resultset('CD')->search(undef, {
4009 '+columns' => ['artist.name'],
4013 would return all CDs and include a 'name' column to the information
4014 passed to object inflation. Note that the 'artist' is the name of the
4015 column (or relationship) accessor, and 'name' is the name of the column
4016 accessor in the related table.
4018 B<NOTE:> You need to explicitly quote '+columns' when defining the attribute.
4019 Not doing so causes Perl to incorrectly interpret +columns as a bareword with a
4020 unary plus operator before it.
4022 =head2 include_columns
4026 =item Value: \@columns
4030 Deprecated. Acts as a synonym for L</+columns> for backward compatibility.
4036 =item Value: \@select_columns
4040 Indicates which columns should be selected from the storage. You can use
4041 column names, or in the case of RDBMS back ends, function or stored procedure
4044 $rs = $schema->resultset('Employee')->search(undef, {
4047 { count => 'employeeid' },
4048 { max => { length => 'name' }, -as => 'longest_name' }
4053 SELECT name, COUNT( employeeid ), MAX( LENGTH( name ) ) AS longest_name FROM employee
4055 B<NOTE:> You will almost always need a corresponding L</as> attribute when you
4056 use L</select>, to instruct DBIx::Class how to store the result of the column.
4057 Also note that the L</as> attribute has nothing to do with the SQL-side 'AS'
4058 identifier aliasing. You can however alias a function, so you can use it in
4059 e.g. an C<ORDER BY> clause. This is done via the C<-as> B<select function
4060 attribute> supplied as shown in the example above.
4062 B<NOTE:> You need to explicitly quote '+select'/'+as' when defining the attributes.
4063 Not doing so causes Perl to incorrectly interpret them as a bareword with a
4064 unary plus operator before it.
4070 Indicates additional columns to be selected from storage. Works the same as
4071 L</select> but adds columns to the default selection, instead of specifying
4080 =item Value: \@inflation_names
4084 Indicates column names for object inflation. That is L</as> indicates the
4085 slot name in which the column value will be stored within the
4086 L<Row|DBIx::Class::Row> object. The value will then be accessible via this
4087 identifier by the C<get_column> method (or via the object accessor B<if one
4088 with the same name already exists>) as shown below. The L</as> attribute has
4089 B<nothing to do> with the SQL-side C<AS>. See L</select> for details.
4091 $rs = $schema->resultset('Employee')->search(undef, {
4094 { count => 'employeeid' },
4095 { max => { length => 'name' }, -as => 'longest_name' }
4104 If the object against which the search is performed already has an accessor
4105 matching a column name specified in C<as>, the value can be retrieved using
4106 the accessor as normal:
4108 my $name = $employee->name();
4110 If on the other hand an accessor does not exist in the object, you need to
4111 use C<get_column> instead:
4113 my $employee_count = $employee->get_column('employee_count');
4115 You can create your own accessors if required - see
4116 L<DBIx::Class::Manual::Cookbook> for details.
4122 Indicates additional column names for those added via L</+select>. See L</as>.
4130 =item Value: ($rel_name | \@rel_names | \%rel_names)
4134 Contains a list of relationships that should be joined for this query. For
4137 # Get CDs by Nine Inch Nails
4138 my $rs = $schema->resultset('CD')->search(
4139 { 'artist.name' => 'Nine Inch Nails' },
4140 { join => 'artist' }
4143 Can also contain a hash reference to refer to the other relation's relations.
4146 package MyApp::Schema::Track;
4147 use base qw/DBIx::Class/;
4148 __PACKAGE__->table('track');
4149 __PACKAGE__->add_columns(qw/trackid cd position title/);
4150 __PACKAGE__->set_primary_key('trackid');
4151 __PACKAGE__->belongs_to(cd => 'MyApp::Schema::CD');
4154 # In your application
4155 my $rs = $schema->resultset('Artist')->search(
4156 { 'track.title' => 'Teardrop' },
4158 join => { cd => 'track' },
4159 order_by => 'artist.name',
4163 You need to use the relationship (not the table) name in conditions,
4164 because they are aliased as such. The current table is aliased as "me", so
4165 you need to use me.column_name in order to avoid ambiguity. For example:
4167 # Get CDs from 1984 with a 'Foo' track
4168 my $rs = $schema->resultset('CD')->search(
4171 'tracks.name' => 'Foo'
4173 { join => 'tracks' }
4176 If the same join is supplied twice, it will be aliased to <rel>_2 (and
4177 similarly for a third time). For e.g.
4179 my $rs = $schema->resultset('Artist')->search({
4180 'cds.title' => 'Down to Earth',
4181 'cds_2.title' => 'Popular',
4183 join => [ qw/cds cds/ ],
4186 will return a set of all artists that have both a cd with title 'Down
4187 to Earth' and a cd with title 'Popular'.
4189 If you want to fetch related objects from other tables as well, see L</prefetch>
4192 NOTE: An internal join-chain pruner will discard certain joins while
4193 constructing the actual SQL query, as long as the joins in question do not
4194 affect the retrieved result. This for example includes 1:1 left joins
4195 that are not part of the restriction specification (WHERE/HAVING) nor are
4196 a part of the query selection.
4198 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
4204 =item Value: (0 | 1)
4208 When set to a true value, indicates that any rows fetched from joined has_many
4209 relationships are to be aggregated into the corresponding "parent" object. For
4210 example, the resultset:
4212 my $rs = $schema->resultset('CD')->search({}, {
4213 '+columns' => [ qw/ tracks.title tracks.position / ],
4218 While executing the following query:
4220 SELECT me.*, tracks.title, tracks.position
4222 LEFT JOIN track tracks
4223 ON tracks.cdid = me.cdid
4225 Will return only as many objects as there are rows in the CD source, even
4226 though the result of the query may span many rows. Each of these CD objects
4227 will in turn have multiple "Track" objects hidden behind the has_many
4228 generated accessor C<tracks>. Without C<< collapse => 1 >>, the return values
4229 of this resultset would be as many CD objects as there are tracks (a "Cartesian
4230 product"), with each CD object containing exactly one of all fetched Track data.
4232 When a collapse is requested on a non-ordered resultset, an order by some
4233 unique part of the main source (the left-most table) is inserted automatically.
4234 This is done so that the resultset is allowed to be "lazy" - calling
4235 L<< $rs->next|/next >> will fetch only as many rows as it needs to build the next
4236 object with all of its related data.
4238 If an L</order_by> is already declared, and orders the resultset in a way that
4239 makes collapsing as described above impossible (e.g. C<< ORDER BY
4240 has_many_rel.column >> or C<ORDER BY RANDOM()>), DBIC will automatically
4241 switch to "eager" mode and slurp the entire resultset before constructing the
4242 first object returned by L</next>.
4244 Setting this attribute on a resultset that does not join any has_many
4245 relations is a no-op.
4247 For a more in-depth discussion, see L</PREFETCHING>.
4253 =item Value: ($rel_name | \@rel_names | \%rel_names)
4257 This attribute is a shorthand for specifying a L</join> spec, adding all
4258 columns from the joined related sources as L</+columns> and setting
4259 L</collapse> to a true value. For example, the following two queries are
4262 my $rs = $schema->resultset('Artist')->search({}, {
4263 prefetch => { cds => ['genre', 'tracks' ] },
4268 my $rs = $schema->resultset('Artist')->search({}, {
4269 join => { cds => ['genre', 'tracks' ] },
4273 { +{ "cds.$_" => "cds.$_" } }
4274 $schema->source('Artist')->related_source('cds')->columns
4277 { +{ "cds.genre.$_" => "genre.$_" } }
4278 $schema->source('Artist')->related_source('cds')->related_source('genre')->columns
4281 { +{ "cds.tracks.$_" => "tracks.$_" } }
4282 $schema->source('Artist')->related_source('cds')->related_source('tracks')->columns
4287 Both producing the following SQL:
4289 SELECT me.artistid, me.name, me.rank, me.charfield,
4290 cds.cdid, cds.artist, cds.title, cds.year, cds.genreid, cds.single_track,
4291 genre.genreid, genre.name,
4292 tracks.trackid, tracks.cd, tracks.position, tracks.title, tracks.last_updated_on, tracks.last_updated_at
4295 ON cds.artist = me.artistid
4296 LEFT JOIN genre genre
4297 ON genre.genreid = cds.genreid
4298 LEFT JOIN track tracks
4299 ON tracks.cd = cds.cdid
4300 ORDER BY me.artistid
4302 While L</prefetch> implies a L</join>, it is ok to mix the two together, as
4303 the arguments are properly merged and generally do the right thing. For
4304 example, you may want to do the following:
4306 my $artists_and_cds_without_genre = $schema->resultset('Artist')->search(
4307 { 'genre.genreid' => undef },
4309 join => { cds => 'genre' },
4314 Which generates the following SQL:
4316 SELECT me.artistid, me.name, me.rank, me.charfield,
4317 cds.cdid, cds.artist, cds.title, cds.year, cds.genreid, cds.single_track
4320 ON cds.artist = me.artistid
4321 LEFT JOIN genre genre
4322 ON genre.genreid = cds.genreid
4323 WHERE genre.genreid IS NULL
4324 ORDER BY me.artistid
4326 For a more in-depth discussion, see L</PREFETCHING>.
4332 =item Value: $source_alias
4336 Sets the source alias for the query. Normally, this defaults to C<me>, but
4337 nested search queries (sub-SELECTs) might need specific aliases set to
4338 reference inner queries. For example:
4341 ->related_resultset('CDs')
4342 ->related_resultset('Tracks')
4344 'track.id' => { -ident => 'none_search.id' },
4348 my $ids = $self->search({
4351 alias => 'none_search',
4352 group_by => 'none_search.id',
4353 })->get_column('id')->as_query;
4355 $self->search({ id => { -in => $ids } })
4357 This attribute is directly tied to L</current_source_alias>.
4367 Makes the resultset paged and specifies the page to retrieve. Effectively
4368 identical to creating a non-pages resultset and then calling ->page($page)
4371 If L</rows> attribute is not specified it defaults to 10 rows per page.
4373 When you have a paged resultset, L</count> will only return the number
4374 of rows in the page. To get the total, use the L</pager> and call
4375 C<total_entries> on it.
4385 Specifies the maximum number of rows for direct retrieval or the number of
4386 rows per page if the page attribute or method is used.
4392 =item Value: $offset
4396 Specifies the (zero-based) row number for the first row to be returned, or the
4397 of the first row of the first page if paging is used.
4399 =head2 software_limit
4403 =item Value: (0 | 1)
4407 When combined with L</rows> and/or L</offset> the generated SQL will not
4408 include any limit dialect stanzas. Instead the entire result will be selected
4409 as if no limits were specified, and DBIC will perform the limit locally, by
4410 artificially advancing and finishing the resulting L</cursor>.
4412 This is the recommended way of performing resultset limiting when no sane RDBMS
4413 implementation is available (e.g.
4414 L<Sybase ASE|DBIx::Class::Storage::DBI::Sybase::ASE> using the
4415 L<Generic Sub Query|DBIx::Class::SQLMaker::LimitDialects/GenericSubQ> hack)
4421 =item Value: \@columns
4425 A arrayref of columns to group by. Can include columns of joined tables.
4427 group_by => [qw/ column1 column2 ... /]
4433 =item Value: $condition
4437 HAVING is a select statement attribute that is applied between GROUP BY and
4438 ORDER BY. It is applied to the after the grouping calculations have been
4441 having => { 'count_employee' => { '>=', 100 } }
4443 or with an in-place function in which case literal SQL is required:
4445 having => \[ 'count(employee) >= ?', [ count => 100 ] ]
4451 =item Value: (0 | 1)
4455 Set to 1 to group by all columns. If the resultset already has a group_by
4456 attribute, this setting is ignored and an appropriate warning is issued.
4462 Adds to the WHERE clause.
4464 # only return rows WHERE deleted IS NULL for all searches
4465 __PACKAGE__->resultset_attributes({ where => { deleted => undef } });
4467 Can be overridden by passing C<< { where => undef } >> as an attribute
4470 For more complicated where clauses see L<SQL::Abstract/WHERE CLAUSES>.
4476 Set to 1 to cache search results. This prevents extra SQL queries if you
4477 revisit rows in your ResultSet:
4479 my $resultset = $schema->resultset('Artist')->search( undef, { cache => 1 } );
4481 while( my $artist = $resultset->next ) {
4485 $rs->first; # without cache, this would issue a query
4487 By default, searches are not cached.
4489 For more examples of using these attributes, see
4490 L<DBIx::Class::Manual::Cookbook>.
4496 =item Value: ( 'update' | 'shared' | \$scalar )
4500 Set to 'update' for a SELECT ... FOR UPDATE or 'shared' for a SELECT
4501 ... FOR SHARED. If \$scalar is passed, this is taken directly and embedded in the
4506 DBIx::Class supports arbitrary related data prefetching from multiple related
4507 sources. Any combination of relationship types and column sets are supported.
4508 If L<collapsing|/collapse> is requested, there is an additional requirement of
4509 selecting enough data to make every individual object uniquely identifiable.
4511 Here are some more involved examples, based on the following relationship map:
4514 My::Schema::CD->belongs_to( artist => 'My::Schema::Artist' );
4515 My::Schema::CD->might_have( liner_note => 'My::Schema::LinerNotes' );
4516 My::Schema::CD->has_many( tracks => 'My::Schema::Track' );
4518 My::Schema::Artist->belongs_to( record_label => 'My::Schema::RecordLabel' );
4520 My::Schema::Track->has_many( guests => 'My::Schema::Guest' );
4524 my $rs = $schema->resultset('Tag')->search(
4533 The initial search results in SQL like the following:
4535 SELECT tag.*, cd.*, artist.* FROM tag
4536 JOIN cd ON tag.cd = cd.cdid
4537 JOIN artist ON cd.artist = artist.artistid
4539 L<DBIx::Class> has no need to go back to the database when we access the
4540 C<cd> or C<artist> relationships, which saves us two SQL statements in this
4543 Simple prefetches will be joined automatically, so there is no need
4544 for a C<join> attribute in the above search.
4546 The L</prefetch> attribute can be used with any of the relationship types
4547 and multiple prefetches can be specified together. Below is a more complex
4548 example that prefetches a CD's artist, its liner notes (if present),
4549 the cover image, the tracks on that CD, and the guests on those
4552 my $rs = $schema->resultset('CD')->search(
4556 { artist => 'record_label'}, # belongs_to => belongs_to
4557 'liner_note', # might_have
4558 'cover_image', # has_one
4559 { tracks => 'guests' }, # has_many => has_many
4564 This will produce SQL like the following:
4566 SELECT cd.*, artist.*, record_label.*, liner_note.*, cover_image.*,
4570 ON artist.artistid = me.artistid
4571 JOIN record_label record_label
4572 ON record_label.labelid = artist.labelid
4573 LEFT JOIN track tracks
4574 ON tracks.cdid = me.cdid
4575 LEFT JOIN guest guests
4576 ON guests.trackid = track.trackid
4577 LEFT JOIN liner_notes liner_note
4578 ON liner_note.cdid = me.cdid
4579 JOIN cd_artwork cover_image
4580 ON cover_image.cdid = me.cdid
4583 Now the C<artist>, C<record_label>, C<liner_note>, C<cover_image>,
4584 C<tracks>, and C<guests> of the CD will all be available through the
4585 relationship accessors without the need for additional queries to the
4590 Prefetch does a lot of deep magic. As such, it may not behave exactly
4591 as you might expect.
4597 Prefetch uses the L</cache> to populate the prefetched relationships. This
4598 may or may not be what you want.
4602 If you specify a condition on a prefetched relationship, ONLY those
4603 rows that match the prefetched condition will be fetched into that relationship.
4604 This means that adding prefetch to a search() B<may alter> what is returned by
4605 traversing a relationship. So, if you have C<< Artist->has_many(CDs) >> and you do
4607 my $artist_rs = $schema->resultset('Artist')->search({
4613 my $count = $artist_rs->first->cds->count;
4615 my $artist_rs_prefetch = $artist_rs->search( {}, { prefetch => 'cds' } );
4617 my $prefetch_count = $artist_rs_prefetch->first->cds->count;
4619 cmp_ok( $count, '==', $prefetch_count, "Counts should be the same" );
4621 That cmp_ok() may or may not pass depending on the datasets involved. In other
4622 words the C<WHERE> condition would apply to the entire dataset, just like
4623 it would in regular SQL. If you want to add a condition only to the "right side"
4624 of a C<LEFT JOIN> - consider declaring and using a L<relationship with a custom
4625 condition|DBIx::Class::Relationship::Base/condition>
4629 =head1 DBIC BIND VALUES
4631 Because DBIC may need more information to bind values than just the column name
4632 and value itself, it uses a special format for both passing and receiving bind
4633 values. Each bind value should be composed of an arrayref of
4634 C<< [ \%args => $val ] >>. The format of C<< \%args >> is currently:
4640 If present (in any form), this is what is being passed directly to bind_param.
4641 Note that different DBD's expect different bind args. (e.g. DBD::SQLite takes
4642 a single numerical type, while DBD::Pg takes a hashref if bind options.)
4644 If this is specified, all other bind options described below are ignored.
4648 If present, this is used to infer the actual bind attribute by passing to
4649 C<< $resolved_storage->bind_attribute_by_data_type() >>. Defaults to the
4650 "data_type" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>.
4652 Note that the data type is somewhat freeform (hence the sqlt_ prefix);
4653 currently drivers are expected to "Do the Right Thing" when given a common
4654 datatype name. (Not ideal, but that's what we got at this point.)
4658 Currently used to correctly allocate buffers for bind_param_inout().
4659 Defaults to "size" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>,
4660 or to a sensible value based on the "data_type".
4664 Used to fill in missing sqlt_datatype and sqlt_size attributes (if they are
4665 explicitly specified they are never overridden). Also used by some weird DBDs,
4666 where the column name should be available at bind_param time (e.g. Oracle).
4670 For backwards compatibility and convenience, the following shortcuts are
4673 [ $name => $val ] === [ { dbic_colname => $name }, $val ]
4674 [ \$dt => $val ] === [ { sqlt_datatype => $dt }, $val ]
4675 [ undef, $val ] === [ {}, $val ]
4676 $val === [ {}, $val ]
4678 =head1 AUTHOR AND CONTRIBUTORS
4680 See L<AUTHOR|DBIx::Class/AUTHOR> and L<CONTRIBUTORS|DBIx::Class/CONTRIBUTORS> in DBIx::Class
4684 You may distribute this code under the same terms as Perl itself.