1 package DBIx::Class::ResultSet;
5 use base qw/DBIx::Class/;
7 use DBIx::Class::ResultSetColumn;
8 use Scalar::Util qw/blessed weaken reftype/;
9 use DBIx::Class::_Util qw(
10 fail_on_internal_wantarray is_plain_value is_literal_value
13 use Data::Compare (); # no imports!!! guard against insane architecture
15 # not importing first() as it will clash with our own method
19 # De-duplication in _merge_attr() is disabled, but left in for reference
20 # (the merger is used for other things that ought not to be de-duped)
21 *__HM_DEDUP = sub () { 0 };
31 # this is real - CDBICompat overrides it with insanity
32 # yes, prototype won't matter, but that's for now ;)
35 __PACKAGE__->mk_group_accessors('simple' => qw/_result_class result_source/);
39 DBIx::Class::ResultSet - Represents a query used for fetching a set of results.
43 my $users_rs = $schema->resultset('User');
44 while( $user = $users_rs->next) {
45 print $user->username;
48 my $registered_users_rs = $schema->resultset('User')->search({ registered => 1 });
49 my @cds_in_2005 = $schema->resultset('CD')->search({ year => 2005 })->all();
53 A ResultSet is an object which stores a set of conditions representing
54 a query. It is the backbone of DBIx::Class (i.e. the really
55 important/useful bit).
57 No SQL is executed on the database when a ResultSet is created, it
58 just stores all the conditions needed to create the query.
60 A basic ResultSet representing the data of an entire table is returned
61 by calling C<resultset> on a L<DBIx::Class::Schema> and passing in a
62 L<Source|DBIx::Class::Manual::Glossary/Source> name.
64 my $users_rs = $schema->resultset('User');
66 A new ResultSet is returned from calling L</search> on an existing
67 ResultSet. The new one will contain all the conditions of the
68 original, plus any new conditions added in the C<search> call.
70 A ResultSet also incorporates an implicit iterator. L</next> and L</reset>
71 can be used to walk through all the L<DBIx::Class::Row>s the ResultSet
74 The query that the ResultSet represents is B<only> executed against
75 the database when these methods are called:
76 L</find>, L</next>, L</all>, L</first>, L</single>, L</count>.
78 If a resultset is used in a numeric context it returns the L</count>.
79 However, if it is used in a boolean context it is B<always> true. So if
80 you want to check if a resultset has any results, you must use C<if $rs
85 =head2 Chaining resultsets
87 Let's say you've got a query that needs to be run to return some data
88 to the user. But, you have an authorization system in place that
89 prevents certain users from seeing certain information. So, you want
90 to construct the basic query in one method, but add constraints to it in
95 my $request = $self->get_request; # Get a request object somehow.
96 my $schema = $self->result_source->schema;
98 my $cd_rs = $schema->resultset('CD')->search({
99 title => $request->param('title'),
100 year => $request->param('year'),
103 $cd_rs = $self->apply_security_policy( $cd_rs );
105 return $cd_rs->all();
108 sub apply_security_policy {
117 =head3 Resolving conditions and attributes
119 When a resultset is chained from another resultset (e.g.:
120 C<< my $new_rs = $old_rs->search(\%extra_cond, \%attrs) >>), conditions
121 and attributes with the same keys need resolving.
123 If any of L</columns>, L</select>, L</as> are present, they reset the
124 original selection, and start the selection "clean".
126 The L</join>, L</prefetch>, L</+columns>, L</+select>, L</+as> attributes
127 are merged into the existing ones from the original resultset.
129 The L</where> and L</having> attributes, and any search conditions, are
130 merged with an SQL C<AND> to the existing condition from the original
133 All other attributes are overridden by any new ones supplied in the
136 =head2 Multiple queries
138 Since a resultset just defines a query, you can do all sorts of
139 things with it with the same object.
141 # Don't hit the DB yet.
142 my $cd_rs = $schema->resultset('CD')->search({
143 title => 'something',
147 # Each of these hits the DB individually.
148 my $count = $cd_rs->count;
149 my $most_recent = $cd_rs->get_column('date_released')->max();
150 my @records = $cd_rs->all;
152 And it's not just limited to SELECT statements.
158 $cd_rs->create({ artist => 'Fred' });
160 Which is the same as:
162 $schema->resultset('CD')->create({
163 title => 'something',
168 See: L</search>, L</count>, L</get_column>, L</all>, L</create>.
170 =head2 Custom ResultSet classes
172 To add methods to your resultsets, you can subclass L<DBIx::Class::ResultSet>, similar to:
174 package MyApp::Schema::ResultSet::User;
179 use base 'DBIx::Class::ResultSet';
183 $self->search({ $self->current_source_alias . '.active' => 1 });
188 $self->search({ $self->current_source_alias . '.verified' => 0 });
191 sub created_n_days_ago {
192 my ($self, $days_ago) = @_;
194 $self->current_source_alias . '.create_date' => {
196 $self->result_source->schema->storage->datetime_parser->format_datetime(
197 DateTime->now( time_zone => 'UTC' )->subtract( days => $days_ago )
202 sub users_to_warn { shift->active->unverified->created_n_days_ago(7) }
206 See L<DBIx::Class::Schema/load_namespaces> on how DBIC can discover and
207 automatically attach L<Result|DBIx::Class::Manual::ResultClass>-specific
208 L<ResulSet|DBIx::Class::ResultSet> classes.
210 =head3 ResultSet subclassing with Moose
212 If you want to make your custom ResultSet classes with L<Moose>, use a template
215 package MyApp::Schema::ResultSet::User;
218 use MooseX::NonMoose;
219 extends 'DBIx::Class::ResultSet';
221 sub BUILDARGS { $_[2] }
225 __PACKAGE__->meta->make_immutable;
229 The L<MooseX::NonMoose> is necessary so that the L<Moose> constructor does not
230 clash with the regular ResultSet constructor. Alternatively, you can use:
232 __PACKAGE__->meta->make_immutable(inline_constructor => 0);
234 The L<BUILDARGS|Moose::Manual::Construction/BUILDARGS> is necessary because the
235 signature of the ResultSet C<new> is C<< ->new($source, \%args) >>.
243 =item Arguments: L<$source|DBIx::Class::ResultSource>, L<\%attrs?|/ATTRIBUTES>
245 =item Return Value: L<$resultset|/search>
249 The resultset constructor. Takes a source object (usually a
250 L<DBIx::Class::ResultSourceProxy::Table>) and an attribute hash (see
251 L</ATTRIBUTES> below). Does not perform any queries -- these are
252 executed as needed by the other methods.
254 Generally you never construct a resultset manually. Instead you get one
256 C<< $schema->L<resultset|DBIx::Class::Schema/resultset>('$source_name') >>
257 or C<< $another_resultset->L<search|/search>(...) >> (the later called in
260 my $rs = $schema->resultset('CD')->search({ title => '100th Window' });
266 If called on an object, proxies to L</new_result> instead, so
268 my $cd = $schema->resultset('CD')->new({ title => 'Spoon' });
270 will return a CD object, not a ResultSet, and is equivalent to:
272 my $cd = $schema->resultset('CD')->new_result({ title => 'Spoon' });
274 Please also keep in mind that many internals call L</new_result> directly,
275 so overloading this method with the idea of intercepting new result object
276 creation B<will not work>. See also warning pertaining to L</create>.
284 return $class->new_result(@_) if ref $class;
286 my ($source, $attrs) = @_;
287 $source = $source->resolve
288 if $source->isa('DBIx::Class::ResultSourceHandle');
290 $attrs = { %{$attrs||{}} };
291 delete @{$attrs}{qw(_last_sqlmaker_alias_map _related_results_construction)};
293 if ($attrs->{page}) {
294 $attrs->{rows} ||= 10;
297 $attrs->{alias} ||= 'me';
300 result_source => $source,
301 cond => $attrs->{where},
306 # if there is a dark selector, this means we are already in a
307 # chain and the cleanup/sanification was taken care of by
309 $self->_normalize_selection($attrs)
310 unless $attrs->{_dark_selector};
313 $attrs->{result_class} || $source->result_class
323 =item Arguments: L<$cond|DBIx::Class::SQLMaker> | undef, L<\%attrs?|/ATTRIBUTES>
325 =item Return Value: $resultset (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
329 my @cds = $cd_rs->search({ year => 2001 }); # "... WHERE year = 2001"
330 my $new_rs = $cd_rs->search({ year => 2005 });
332 my $new_rs = $cd_rs->search([ { year => 2005 }, { year => 2004 } ]);
333 # year = 2005 OR year = 2004
335 In list context, C<< ->all() >> is called implicitly on the resultset, thus
336 returning a list of L<result|DBIx::Class::Manual::ResultClass> objects instead.
337 To avoid that, use L</search_rs>.
339 If you need to pass in additional attributes but no additional condition,
340 call it as C<search(undef, \%attrs)>.
342 # "SELECT name, artistid FROM $artist_table"
343 my @all_artists = $schema->resultset('Artist')->search(undef, {
344 columns => [qw/name artistid/],
347 For a list of attributes that can be passed to C<search>, see
348 L</ATTRIBUTES>. For more examples of using this function, see
349 L<Searching|DBIx::Class::Manual::Cookbook/SEARCHING>. For a complete
350 documentation for the first argument, see L<SQL::Abstract/"WHERE CLAUSES">
351 and its extension L<DBIx::Class::SQLMaker>.
353 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
357 Note that L</search> does not process/deflate any of the values passed in the
358 L<SQL::Abstract>-compatible search condition structure. This is unlike other
359 condition-bound methods L</new_result>, L</create> and L</find>. The user must ensure
360 manually that any value passed to this method will stringify to something the
361 RDBMS knows how to deal with. A notable example is the handling of L<DateTime>
362 objects, for more info see:
363 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
369 my $rs = $self->search_rs( @_ );
372 DBIx::Class::_ENV_::ASSERT_NO_INTERNAL_WANTARRAY and my $sog = fail_on_internal_wantarray($rs);
375 elsif (defined wantarray) {
379 # we can be called by a relationship helper, which in
380 # turn may be called in void context due to some braindead
381 # overload or whatever else the user decided to be clever
382 # at this particular day. Thus limit the exception to
383 # external code calls only
384 $self->throw_exception ('->search is *not* a mutator, calling it in void context makes no sense')
385 if (caller)[0] !~ /^\QDBIx::Class::/;
395 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
397 =item Return Value: L<$resultset|/search>
401 This method does the same exact thing as search() except it will
402 always return a resultset, even in list context.
409 my $rsrc = $self->result_source;
410 my ($call_cond, $call_attrs);
412 # Special-case handling for (undef, undef) or (undef)
413 # Note that (foo => undef) is valid deprecated syntax
414 @_ = () if not scalar grep { defined $_ } @_;
420 # fish out attrs in the ($condref, $attr) case
421 elsif (@_ == 2 and ( ! defined $_[0] or (ref $_[0]) ne '') ) {
422 ($call_cond, $call_attrs) = @_;
425 $self->throw_exception('Odd number of arguments to search')
429 carp_unique 'search( %condition ) is deprecated, use search( \%condition ) instead'
430 unless $rsrc->result_class->isa('DBIx::Class::CDBICompat');
432 for my $i (0 .. $#_) {
434 $self->throw_exception ('All keys in condition key/value pairs must be plain scalars')
435 if (! defined $_[$i] or ref $_[$i] ne '');
441 # see if we can keep the cache (no $rs changes)
443 my %safe = (alias => 1, cache => 1);
444 if ( ! List::Util::first { !$safe{$_} } keys %$call_attrs and (
447 ref $call_cond eq 'HASH' && ! keys %$call_cond
449 ref $call_cond eq 'ARRAY' && ! @$call_cond
451 $cache = $self->get_cache;
454 my $old_attrs = { %{$self->{attrs}} };
455 my ($old_having, $old_where) = delete @{$old_attrs}{qw(having where)};
457 my $new_attrs = { %$old_attrs };
459 # take care of call attrs (only if anything is changing)
460 if ($call_attrs and keys %$call_attrs) {
462 # copy for _normalize_selection
463 $call_attrs = { %$call_attrs };
465 my @selector_attrs = qw/select as columns cols +select +as +columns include_columns/;
467 # reset the current selector list if new selectors are supplied
468 if (List::Util::first { exists $call_attrs->{$_} } qw/columns cols select as/) {
469 delete @{$old_attrs}{(@selector_attrs, '_dark_selector')};
472 # Normalize the new selector list (operates on the passed-in attr structure)
473 # Need to do it on every chain instead of only once on _resolved_attrs, in
474 # order to allow detection of empty vs partial 'as'
475 $call_attrs->{_dark_selector} = $old_attrs->{_dark_selector}
476 if $old_attrs->{_dark_selector};
477 $self->_normalize_selection ($call_attrs);
479 # start with blind overwriting merge, exclude selector attrs
480 $new_attrs = { %{$old_attrs}, %{$call_attrs} };
481 delete @{$new_attrs}{@selector_attrs};
483 for (@selector_attrs) {
484 $new_attrs->{$_} = $self->_merge_attr($old_attrs->{$_}, $call_attrs->{$_})
485 if ( exists $old_attrs->{$_} or exists $call_attrs->{$_} );
488 # older deprecated name, use only if {columns} is not there
489 if (my $c = delete $new_attrs->{cols}) {
490 carp_unique( "Resultset attribute 'cols' is deprecated, use 'columns' instead" );
491 if ($new_attrs->{columns}) {
492 carp "Resultset specifies both the 'columns' and the legacy 'cols' attributes - ignoring 'cols'";
495 $new_attrs->{columns} = $c;
500 # join/prefetch use their own crazy merging heuristics
501 foreach my $key (qw/join prefetch/) {
502 $new_attrs->{$key} = $self->_merge_joinpref_attr($old_attrs->{$key}, $call_attrs->{$key})
503 if exists $call_attrs->{$key};
506 # stack binds together
507 $new_attrs->{bind} = [ @{ $old_attrs->{bind} || [] }, @{ $call_attrs->{bind} || [] } ];
511 for ($old_where, $call_cond) {
513 $new_attrs->{where} = $self->_stack_cond (
514 $_, $new_attrs->{where}
519 if (defined $old_having) {
520 $new_attrs->{having} = $self->_stack_cond (
521 $old_having, $new_attrs->{having}
525 my $rs = (ref $self)->new($rsrc, $new_attrs);
527 $rs->set_cache($cache) if ($cache);
533 sub _normalize_selection {
534 my ($self, $attrs) = @_;
537 if ( exists $attrs->{include_columns} ) {
538 carp_unique( "Resultset attribute 'include_columns' is deprecated, use '+columns' instead" );
539 $attrs->{'+columns'} = $self->_merge_attr(
540 $attrs->{'+columns'}, delete $attrs->{include_columns}
544 # columns are always placed first, however
546 # Keep the X vs +X separation until _resolved_attrs time - this allows to
547 # delay the decision on whether to use a default select list ($rsrc->columns)
548 # allowing stuff like the remove_columns helper to work
550 # select/as +select/+as pairs need special handling - the amount of select/as
551 # elements in each pair does *not* have to be equal (think multicolumn
552 # selectors like distinct(foo, bar) ). If the selector is bare (no 'as'
553 # supplied at all) - try to infer the alias, either from the -as parameter
554 # of the selector spec, or use the parameter whole if it looks like a column
555 # name (ugly legacy heuristic). If all fails - leave the selector bare (which
556 # is ok as well), but make sure no more additions to the 'as' chain take place
557 for my $pref ('', '+') {
559 my ($sel, $as) = map {
560 my $key = "${pref}${_}";
562 my $val = [ ref $attrs->{$key} eq 'ARRAY'
564 : $attrs->{$key} || ()
566 delete $attrs->{$key};
570 if (! @$as and ! @$sel ) {
573 elsif (@$as and ! @$sel) {
574 $self->throw_exception(
575 "Unable to handle ${pref}as specification (@$as) without a corresponding ${pref}select"
579 # no as part supplied at all - try to deduce (unless explicit end of named selection is declared)
580 # if any @$as has been supplied we assume the user knows what (s)he is doing
581 # and blindly keep stacking up pieces
582 unless ($attrs->{_dark_selector}) {
585 if ( ref $_ eq 'HASH' and exists $_->{-as} ) {
586 push @$as, $_->{-as};
588 # assume any plain no-space, no-parenthesis string to be a column spec
589 # FIXME - this is retarded but is necessary to support shit like 'count(foo)'
590 elsif ( ! ref $_ and $_ =~ /^ [^\s\(\)]+ $/x) {
593 # if all else fails - raise a flag that no more aliasing will be allowed
595 $attrs->{_dark_selector} = {
597 string => ($dark_sel_dumper ||= do {
598 require Data::Dumper::Concise;
599 Data::Dumper::Concise::DumperObject()->Indent(0);
600 })->Values([$_])->Dump
608 elsif (@$as < @$sel) {
609 $self->throw_exception(
610 "Unable to handle an ${pref}as specification (@$as) with less elements than the corresponding ${pref}select"
613 elsif ($pref and $attrs->{_dark_selector}) {
614 $self->throw_exception(
615 "Unable to process named '+select', resultset contains an unnamed selector $attrs->{_dark_selector}{string}"
621 $attrs->{"${pref}select"} = $self->_merge_attr($attrs->{"${pref}select"}, $sel);
622 $attrs->{"${pref}as"} = $self->_merge_attr($attrs->{"${pref}as"}, $as);
627 my ($self, $left, $right) = @_;
630 (ref $_ eq 'ARRAY' and !@$_)
632 (ref $_ eq 'HASH' and ! keys %$_)
633 ) and $_ = undef for ($left, $right);
635 # either on of the two undef or both undef
636 if ( ( (defined $left) xor (defined $right) ) or ! defined $left ) {
637 return defined $left ? $left : $right;
640 my $cond = $self->result_source->schema->storage->_collapse_cond({ -and => [$left, $right] });
642 for my $c (grep { ref $cond->{$_} eq 'ARRAY' and ($cond->{$_}[0]||'') eq '-and' } keys %$cond) {
644 my @vals = sort @{$cond->{$c}}[ 1..$#{$cond->{$c}} ];
645 my @fin = shift @vals;
648 push @fin, $v unless Data::Compare::Compare( $fin[-1], $v );
651 $cond->{$c} = (@fin == 1) ? $fin[0] : [-and => @fin ];
657 =head2 search_literal
659 B<CAVEAT>: C<search_literal> is provided for Class::DBI compatibility and
660 should only be used in that context. C<search_literal> is a convenience
661 method. It is equivalent to calling C<< $schema->search(\[]) >>, but if you
662 want to ensure columns are bound correctly, use L</search>.
664 See L<DBIx::Class::Manual::Cookbook/SEARCHING> and
665 L<DBIx::Class::Manual::FAQ/Searching> for searching techniques that do not
666 require C<search_literal>.
670 =item Arguments: $sql_fragment, @standalone_bind_values
672 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
676 my @cds = $cd_rs->search_literal('year = ? AND title = ?', qw/2001 Reload/);
677 my $newrs = $artist_rs->search_literal('name = ?', 'Metallica');
679 Pass a literal chunk of SQL to be added to the conditional part of the
682 Example of how to use C<search> instead of C<search_literal>
684 my @cds = $cd_rs->search_literal('cdid = ? AND (artist = ? OR artist = ?)', (2, 1, 2));
685 my @cds = $cd_rs->search(\[ 'cdid = ? AND (artist = ? OR artist = ?)', [ 'cdid', 2 ], [ 'artist', 1 ], [ 'artist', 2 ] ]);
690 my ($self, $sql, @bind) = @_;
692 if ( @bind && ref($bind[-1]) eq 'HASH' ) {
695 return $self->search(\[ $sql, map [ {} => $_ ], @bind ], ($attr || () ));
702 =item Arguments: \%columns_values | @pk_values, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
704 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
708 Finds and returns a single row based on supplied criteria. Takes either a
709 hashref with the same format as L</create> (including inference of foreign
710 keys from related objects), or a list of primary key values in the same
711 order as the L<primary columns|DBIx::Class::ResultSource/primary_columns>
712 declaration on the L</result_source>.
714 In either case an attempt is made to combine conditions already existing on
715 the resultset with the condition passed to this method.
717 To aid with preparing the correct query for the storage you may supply the
718 C<key> attribute, which is the name of a
719 L<unique constraint|DBIx::Class::ResultSource/add_unique_constraint> (the
720 unique constraint corresponding to the
721 L<primary columns|DBIx::Class::ResultSource/primary_columns> is always named
722 C<primary>). If the C<key> attribute has been supplied, and DBIC is unable
723 to construct a query that satisfies the named unique constraint fully (
724 non-NULL values for each column member of the constraint) an exception is
727 If no C<key> is specified, the search is carried over all unique constraints
728 which are fully defined by the available condition.
730 If no such constraint is found, C<find> currently defaults to a simple
731 C<< search->(\%column_values) >> which may or may not do what you expect.
732 Note that this fallback behavior may be deprecated in further versions. If
733 you need to search with arbitrary conditions - use L</search>. If the query
734 resulting from this fallback produces more than one row, a warning to the
735 effect is issued, though only the first row is constructed and returned as
738 In addition to C<key>, L</find> recognizes and applies standard
739 L<resultset attributes|/ATTRIBUTES> in the same way as L</search> does.
741 Note that if you have extra concerns about the correctness of the resulting
742 query you need to specify the C<key> attribute and supply the entire condition
743 as an argument to find (since it is not always possible to perform the
744 combination of the resultset condition with the supplied one, especially if
745 the resultset condition contains literal sql).
747 For example, to find a row by its primary key:
749 my $cd = $schema->resultset('CD')->find(5);
751 You can also find a row by a specific unique constraint:
753 my $cd = $schema->resultset('CD')->find(
755 artist => 'Massive Attack',
756 title => 'Mezzanine',
758 { key => 'cd_artist_title' }
761 See also L</find_or_create> and L</update_or_create>.
767 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
769 my $rsrc = $self->result_source;
772 if (exists $attrs->{key}) {
773 $constraint_name = defined $attrs->{key}
775 : $self->throw_exception("An undefined 'key' resultset attribute makes no sense")
779 # Parse out the condition from input
782 if (ref $_[0] eq 'HASH') {
783 $call_cond = { %{$_[0]} };
786 # if only values are supplied we need to default to 'primary'
787 $constraint_name = 'primary' unless defined $constraint_name;
789 my @c_cols = $rsrc->unique_constraint_columns($constraint_name);
791 $self->throw_exception(
792 "No constraint columns, maybe a malformed '$constraint_name' constraint?"
795 $self->throw_exception (
796 'find() expects either a column/value hashref, or a list of values '
797 . "corresponding to the columns of the specified unique constraint '$constraint_name'"
798 ) unless @c_cols == @_;
801 @{$call_cond}{@c_cols} = @_;
805 for my $key (keys %$call_cond) {
807 my $keyref = ref($call_cond->{$key})
809 my $relinfo = $rsrc->relationship_info($key)
811 my $val = delete $call_cond->{$key};
813 next if $keyref eq 'ARRAY'; # has_many for multi_create
815 my ($rel_cond, $crosstable) = $rsrc->_resolve_condition(
816 $relinfo->{cond}, $val, $key, $key
819 $self->throw_exception("Complex condition via relationship '$key' is unsupported in find()")
820 if $crosstable or ref($rel_cond) ne 'HASH';
823 @related{keys %$rel_cond} = values %$rel_cond;
827 # relationship conditions take precedence (?)
828 @{$call_cond}{keys %related} = values %related;
830 my $alias = exists $attrs->{alias} ? $attrs->{alias} : $self->{attrs}{alias};
832 if (defined $constraint_name) {
833 $final_cond = $self->_qualify_cond_columns (
835 $self->_build_unique_cond (
843 elsif ($self->{attrs}{accessor} and $self->{attrs}{accessor} eq 'single') {
844 # This means that we got here after a merger of relationship conditions
845 # in ::Relationship::Base::search_related (the row method), and furthermore
846 # the relationship is of the 'single' type. This means that the condition
847 # provided by the relationship (already attached to $self) is sufficient,
848 # as there can be only one row in the database that would satisfy the
852 # no key was specified - fall down to heuristics mode:
853 # run through all unique queries registered on the resultset, and
854 # 'OR' all qualifying queries together
855 my (@unique_queries, %seen_column_combinations);
856 for my $c_name ($rsrc->unique_constraint_names) {
857 next if $seen_column_combinations{
858 join "\x00", sort $rsrc->unique_constraint_columns($c_name)
861 push @unique_queries, try {
862 $self->_build_unique_cond ($c_name, $call_cond, 'croak_on_nulls')
866 $final_cond = @unique_queries
867 ? [ map { $self->_qualify_cond_columns($_, $alias) } @unique_queries ]
868 : $self->_non_unique_find_fallback ($call_cond, $attrs)
872 # Run the query, passing the result_class since it should propagate for find
873 my $rs = $self->search ($final_cond, {result_class => $self->result_class, %$attrs});
874 if ($rs->_resolved_attrs->{collapse}) {
876 carp "Query returned more than one row" if $rs->next;
884 # This is a stop-gap method as agreed during the discussion on find() cleanup:
885 # http://lists.scsys.co.uk/pipermail/dbix-class/2010-October/009535.html
887 # It is invoked when find() is called in legacy-mode with insufficiently-unique
888 # condition. It is provided for overrides until a saner way forward is devised
890 # *NOTE* This is not a public method, and it's *GUARANTEED* to disappear down
891 # the road. Please adjust your tests accordingly to catch this situation early
892 # DBIx::Class::ResultSet->can('_non_unique_find_fallback') is reasonable
894 # The method will not be removed without an adequately complete replacement
895 # for strict-mode enforcement
896 sub _non_unique_find_fallback {
897 my ($self, $cond, $attrs) = @_;
899 return $self->_qualify_cond_columns(
901 exists $attrs->{alias}
903 : $self->{attrs}{alias}
908 sub _qualify_cond_columns {
909 my ($self, $cond, $alias) = @_;
911 my %aliased = %$cond;
912 for (keys %aliased) {
913 $aliased{"$alias.$_"} = delete $aliased{$_}
920 sub _build_unique_cond {
921 my ($self, $constraint_name, $extra_cond, $croak_on_null) = @_;
923 my @c_cols = $self->result_source->unique_constraint_columns($constraint_name);
925 # combination may fail if $self->{cond} is non-trivial
926 my ($final_cond) = try {
927 $self->_merge_with_rscond ($extra_cond)
932 # trim out everything not in $columns
933 $final_cond = { map {
934 exists $final_cond->{$_}
935 ? ( $_ => $final_cond->{$_} )
939 if (my @missing = grep
940 { ! ($croak_on_null ? defined $final_cond->{$_} : exists $final_cond->{$_}) }
943 $self->throw_exception( sprintf ( "Unable to satisfy requested constraint '%s', no values for column(s): %s",
945 join (', ', map { "'$_'" } @missing),
952 !$ENV{DBIC_NULLABLE_KEY_NOWARN}
954 my @undefs = sort grep { ! defined $final_cond->{$_} } (keys %$final_cond)
956 carp_unique ( sprintf (
957 "NULL/undef values supplied for requested unique constraint '%s' (NULL "
958 . 'values in column(s): %s). This is almost certainly not what you wanted, '
959 . 'though you can set DBIC_NULLABLE_KEY_NOWARN to disable this warning.',
961 join (', ', map { "'$_'" } @undefs),
968 =head2 search_related
972 =item Arguments: $rel_name, $cond?, L<\%attrs?|/ATTRIBUTES>
974 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
978 $new_rs = $cd_rs->search_related('artist', {
982 Searches the specified relationship, optionally specifying a condition and
983 attributes for matching records. See L</ATTRIBUTES> for more information.
985 In list context, C<< ->all() >> is called implicitly on the resultset, thus
986 returning a list of result objects instead. To avoid that, use L</search_related_rs>.
988 See also L</search_related_rs>.
993 return shift->related_resultset(shift)->search(@_);
996 =head2 search_related_rs
998 This method works exactly the same as search_related, except that
999 it guarantees a resultset, even in list context.
1003 sub search_related_rs {
1004 return shift->related_resultset(shift)->search_rs(@_);
1011 =item Arguments: none
1013 =item Return Value: L<$cursor|DBIx::Class::Cursor>
1017 Returns a storage-driven cursor to the given resultset. See
1018 L<DBIx::Class::Cursor> for more information.
1025 return $self->{cursor} ||= do {
1026 my $attrs = $self->_resolved_attrs;
1027 $self->result_source->storage->select(
1028 $attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs
1037 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1039 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1043 my $cd = $schema->resultset('CD')->single({ year => 2001 });
1045 Inflates the first result without creating a cursor if the resultset has
1046 any records in it; if not returns C<undef>. Used by L</find> as a lean version
1049 While this method can take an optional search condition (just like L</search>)
1050 being a fast-code-path it does not recognize search attributes. If you need to
1051 add extra joins or similar, call L</search> and then chain-call L</single> on the
1052 L<DBIx::Class::ResultSet> returned.
1058 As of 0.08100, this method enforces the assumption that the preceding
1059 query returns only one row. If more than one row is returned, you will receive
1062 Query returned more than one row
1064 In this case, you should be using L</next> or L</find> instead, or if you really
1065 know what you are doing, use the L</rows> attribute to explicitly limit the size
1068 This method will also throw an exception if it is called on a resultset prefetching
1069 has_many, as such a prefetch implies fetching multiple rows from the database in
1070 order to assemble the resulting object.
1077 my ($self, $where) = @_;
1079 $self->throw_exception('single() only takes search conditions, no attributes. You want ->search( $cond, $attrs )->single()');
1082 my $attrs = { %{$self->_resolved_attrs} };
1084 $self->throw_exception(
1085 'single() can not be used on resultsets collapsing a has_many. Use find( \%cond ) or next() instead'
1086 ) if $attrs->{collapse};
1089 if (defined $attrs->{where}) {
1092 [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ }
1093 $where, delete $attrs->{where} ]
1096 $attrs->{where} = $where;
1100 my $data = [ $self->result_source->storage->select_single(
1101 $attrs->{from}, $attrs->{select},
1102 $attrs->{where}, $attrs
1105 return undef unless @$data;
1106 $self->{_stashed_rows} = [ $data ];
1107 $self->_construct_results->[0];
1114 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1116 =item Return Value: L<$resultsetcolumn|DBIx::Class::ResultSetColumn>
1120 my $max_length = $rs->get_column('length')->max;
1122 Returns a L<DBIx::Class::ResultSetColumn> instance for a column of the ResultSet.
1127 my ($self, $column) = @_;
1128 my $new = DBIx::Class::ResultSetColumn->new($self, $column);
1136 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1138 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1142 # WHERE title LIKE '%blue%'
1143 $cd_rs = $rs->search_like({ title => '%blue%'});
1145 Performs a search, but uses C<LIKE> instead of C<=> as the condition. Note
1146 that this is simply a convenience method retained for ex Class::DBI users.
1147 You most likely want to use L</search> with specific operators.
1149 For more information, see L<DBIx::Class::Manual::Cookbook>.
1151 This method is deprecated and will be removed in 0.09. Use L</search()>
1152 instead. An example conversion is:
1154 ->search_like({ foo => 'bar' });
1158 ->search({ foo => { like => 'bar' } });
1165 'search_like() is deprecated and will be removed in DBIC version 0.09.'
1166 .' Instead use ->search({ x => { -like => "y%" } })'
1167 .' (note the outer pair of {}s - they are important!)'
1169 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
1170 my $query = ref $_[0] eq 'HASH' ? { %{shift()} }: {@_};
1171 $query->{$_} = { 'like' => $query->{$_} } for keys %$query;
1172 return $class->search($query, { %$attrs });
1179 =item Arguments: $first, $last
1181 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1185 Returns a resultset or object list representing a subset of elements from the
1186 resultset slice is called on. Indexes are from 0, i.e., to get the first
1187 three records, call:
1189 my ($one, $two, $three) = $rs->slice(0, 2);
1194 my ($self, $min, $max) = @_;
1195 my $attrs = {}; # = { %{ $self->{attrs} || {} } };
1196 $attrs->{offset} = $self->{attrs}{offset} || 0;
1197 $attrs->{offset} += $min;
1198 $attrs->{rows} = ($max ? ($max - $min + 1) : 1);
1199 return $self->search(undef, $attrs);
1206 =item Arguments: none
1208 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1212 Returns the next element in the resultset (C<undef> is there is none).
1214 Can be used to efficiently iterate over records in the resultset:
1216 my $rs = $schema->resultset('CD')->search;
1217 while (my $cd = $rs->next) {
1221 Note that you need to store the resultset object, and call C<next> on it.
1222 Calling C<< resultset('Table')->next >> repeatedly will always return the
1223 first record from the resultset.
1230 if (my $cache = $self->get_cache) {
1231 $self->{all_cache_position} ||= 0;
1232 return $cache->[$self->{all_cache_position}++];
1235 if ($self->{attrs}{cache}) {
1236 delete $self->{pager};
1237 $self->{all_cache_position} = 1;
1238 return ($self->all)[0];
1241 return shift(@{$self->{_stashed_results}}) if @{ $self->{_stashed_results}||[] };
1243 $self->{_stashed_results} = $self->_construct_results
1246 return shift @{$self->{_stashed_results}};
1249 # Constructs as many results as it can in one pass while respecting
1250 # cursor laziness. Several modes of operation:
1252 # * Always builds everything present in @{$self->{_stashed_rows}}
1253 # * If called with $fetch_all true - pulls everything off the cursor and
1254 # builds all result structures (or objects) in one pass
1255 # * If $self->_resolved_attrs->{collapse} is true, checks the order_by
1256 # and if the resultset is ordered properly by the left side:
1257 # * Fetches stuff off the cursor until the "master object" changes,
1258 # and saves the last extra row (if any) in @{$self->{_stashed_rows}}
1260 # * Just fetches, and collapses/constructs everything as if $fetch_all
1261 # was requested (there is no other way to collapse except for an
1263 # * If no collapse is requested - just get the next row, construct and
1265 sub _construct_results {
1266 my ($self, $fetch_all) = @_;
1268 my $rsrc = $self->result_source;
1269 my $attrs = $self->_resolved_attrs;
1274 ! $attrs->{order_by}
1278 my @pcols = $rsrc->primary_columns
1280 # default order for collapsing unless the user asked for something
1281 $attrs->{order_by} = [ map { join '.', $attrs->{alias}, $_} @pcols ];
1282 $attrs->{_ordered_for_collapse} = 1;
1283 $attrs->{_order_is_artificial} = 1;
1286 # this will be used as both initial raw-row collector AND as a RV of
1287 # _construct_results. Not regrowing the array twice matters a lot...
1288 # a surprising amount actually
1289 my $rows = delete $self->{_stashed_rows};
1291 my $cursor; # we may not need one at all
1293 my $did_fetch_all = $fetch_all;
1296 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1297 $rows = [ ($rows ? @$rows : ()), $self->cursor->all ];
1299 elsif( $attrs->{collapse} ) {
1301 # a cursor will need to be closed over in case of collapse
1302 $cursor = $self->cursor;
1304 $attrs->{_ordered_for_collapse} = (
1310 ->_main_source_order_by_portion_is_stable($rsrc, $attrs->{order_by}, $attrs->{where})
1312 ) unless defined $attrs->{_ordered_for_collapse};
1314 if (! $attrs->{_ordered_for_collapse}) {
1317 # instead of looping over ->next, use ->all in stealth mode
1318 # *without* calling a ->reset afterwards
1319 # FIXME ENCAPSULATION - encapsulation breach, cursor method additions pending
1320 if (! $cursor->{_done}) {
1321 $rows = [ ($rows ? @$rows : ()), $cursor->all ];
1322 $cursor->{_done} = 1;
1327 if (! $did_fetch_all and ! @{$rows||[]} ) {
1328 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1329 $cursor ||= $self->cursor;
1330 if (scalar (my @r = $cursor->next) ) {
1335 return undef unless @{$rows||[]};
1337 # sanity check - people are too clever for their own good
1338 if ($attrs->{collapse} and my $aliastypes = $attrs->{_last_sqlmaker_alias_map} ) {
1340 my $multiplied_selectors;
1341 for my $sel_alias ( grep { $_ ne $attrs->{alias} } keys %{ $aliastypes->{selecting} } ) {
1343 $aliastypes->{multiplying}{$sel_alias}
1345 $aliastypes->{premultiplied}{$sel_alias}
1347 $multiplied_selectors->{$_} = 1 for values %{$aliastypes->{selecting}{$sel_alias}{-seen_columns}}
1351 for my $i (0 .. $#{$attrs->{as}} ) {
1352 my $sel = $attrs->{select}[$i];
1354 if (ref $sel eq 'SCALAR') {
1357 elsif( ref $sel eq 'REF' and ref $$sel eq 'ARRAY' ) {
1361 $self->throw_exception(
1362 'Result collapse not possible - selection from a has_many source redirected to the main object'
1363 ) if ($multiplied_selectors->{$sel} and $attrs->{as}[$i] !~ /\./);
1367 # hotspot - skip the setter
1368 my $res_class = $self->_result_class;
1370 my $inflator_cref = $self->{_result_inflator}{cref} ||= do {
1371 $res_class->can ('inflate_result')
1372 or $self->throw_exception("Inflator $res_class does not provide an inflate_result() method");
1375 my $infmap = $attrs->{as};
1377 $self->{_result_inflator}{is_core_row} = ( (
1380 ( \&DBIx::Class::Row::inflate_result || die "No ::Row::inflate_result() - can't happen" )
1381 ) ? 1 : 0 ) unless defined $self->{_result_inflator}{is_core_row};
1383 $self->{_result_inflator}{is_hri} = ( (
1384 ! $self->{_result_inflator}{is_core_row}
1387 require DBIx::Class::ResultClass::HashRefInflator
1389 DBIx::Class::ResultClass::HashRefInflator->can('inflate_result')
1391 ) ? 1 : 0 ) unless defined $self->{_result_inflator}{is_hri};
1394 if (! $attrs->{_related_results_construction}) {
1395 # construct a much simpler array->hash folder for the one-table cases right here
1396 if ($self->{_result_inflator}{is_hri}) {
1397 for my $r (@$rows) {
1398 $r = { map { $infmap->[$_] => $r->[$_] } 0..$#$infmap };
1401 # FIXME SUBOPTIMAL this is a very very very hot spot
1402 # while rather optimal we can *still* do much better, by
1403 # building a smarter Row::inflate_result(), and
1404 # switch to feeding it data via a much leaner interface
1406 # crude unscientific benchmarking indicated the shortcut eval is not worth it for
1407 # this particular resultset size
1408 elsif (@$rows < 60) {
1409 for my $r (@$rows) {
1410 $r = $inflator_cref->($res_class, $rsrc, { map { $infmap->[$_] => $r->[$_] } (0..$#$infmap) } );
1415 '$_ = $inflator_cref->($res_class, $rsrc, { %s }) for @$rows',
1416 join (', ', map { "\$infmap->[$_] => \$_->[$_]" } 0..$#$infmap )
1422 $self->{_result_inflator}{is_hri} ? 'hri'
1423 : $self->{_result_inflator}{is_core_row} ? 'classic_pruning'
1424 : 'classic_nonpruning'
1427 # $args and $attrs to _mk_row_parser are separated to delineate what is
1428 # core collapser stuff and what is dbic $rs specific
1429 @{$self->{_row_parser}{$parser_type}}{qw(cref nullcheck)} = $rsrc->_mk_row_parser({
1431 inflate_map => $infmap,
1432 collapse => $attrs->{collapse},
1433 premultiplied => $attrs->{_main_source_premultiplied},
1434 hri_style => $self->{_result_inflator}{is_hri},
1435 prune_null_branches => $self->{_result_inflator}{is_hri} || $self->{_result_inflator}{is_core_row},
1436 }, $attrs) unless $self->{_row_parser}{$parser_type}{cref};
1438 # column_info metadata historically hasn't been too reliable.
1439 # We need to start fixing this somehow (the collapse resolver
1440 # can't work without it). Add an explicit check for the *main*
1441 # result, hopefully this will gradually weed out such errors
1443 # FIXME - this is a temporary kludge that reduces performance
1444 # It is however necessary for the time being
1445 my ($unrolled_non_null_cols_to_check, $err);
1447 if (my $check_non_null_cols = $self->{_row_parser}{$parser_type}{nullcheck} ) {
1450 'Collapse aborted due to invalid ResultSource metadata - the following '
1451 . 'selections are declared non-nullable but NULLs were retrieved: '
1455 COL: for my $i (@$check_non_null_cols) {
1456 ! defined $_->[$i] and push @violating_idx, $i and next COL for @$rows;
1459 $self->throw_exception( $err . join (', ', map { "'$infmap->[$_]'" } @violating_idx ) )
1462 $unrolled_non_null_cols_to_check = join (',', @$check_non_null_cols);
1466 ($did_fetch_all or ! $attrs->{collapse}) ? undef
1467 : defined $unrolled_non_null_cols_to_check ? eval sprintf <<'EOS', $unrolled_non_null_cols_to_check
1469 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1470 my @r = $cursor->next or return;
1471 if (my @violating_idx = grep { ! defined $r[$_] } (%s) ) {
1472 $self->throw_exception( $err . join (', ', map { "'$infmap->[$_]'" } @violating_idx ) )
1478 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1479 my @r = $cursor->next or return;
1484 $self->{_row_parser}{$parser_type}{cref}->(
1486 $next_cref ? ( $next_cref, $self->{_stashed_rows} = [] ) : (),
1489 # Special-case multi-object HRI - there is no $inflator_cref pass
1490 unless ($self->{_result_inflator}{is_hri}) {
1491 $_ = $inflator_cref->($res_class, $rsrc, @$_) for @$rows
1495 # The @$rows check seems odd at first - why wouldn't we want to warn
1496 # regardless? The issue is things like find() etc, where the user
1497 # *knows* only one result will come back. In these cases the ->all
1498 # is not a pessimization, but rather something we actually want
1500 'Unable to properly collapse has_many results in iterator mode due '
1501 . 'to order criteria - performed an eager cursor slurp underneath. '
1502 . 'Consider using ->all() instead'
1503 ) if ( ! $fetch_all and @$rows > 1 );
1508 =head2 result_source
1512 =item Arguments: L<$result_source?|DBIx::Class::ResultSource>
1514 =item Return Value: L<$result_source|DBIx::Class::ResultSource>
1518 An accessor for the primary ResultSource object from which this ResultSet
1525 =item Arguments: $result_class?
1527 =item Return Value: $result_class
1531 An accessor for the class to use when creating result objects. Defaults to
1532 C<< result_source->result_class >> - which in most cases is the name of the
1533 L<"table"|DBIx::Class::Manual::Glossary/"ResultSource"> class.
1535 Note that changing the result_class will also remove any components
1536 that were originally loaded in the source class via
1537 L<DBIx::Class::ResultSource/load_components>. Any overloaded methods
1538 in the original source class will not run.
1543 my ($self, $result_class) = @_;
1544 if ($result_class) {
1546 # don't fire this for an object
1547 $self->ensure_class_loaded($result_class)
1548 unless ref($result_class);
1550 if ($self->get_cache) {
1551 carp_unique('Changing the result_class of a ResultSet instance with cached results is a noop - the cache contents will not be altered');
1553 # FIXME ENCAPSULATION - encapsulation breach, cursor method additions pending
1554 elsif ($self->{cursor} && $self->{cursor}{_pos}) {
1555 $self->throw_exception('Changing the result_class of a ResultSet instance with an active cursor is not supported');
1558 $self->_result_class($result_class);
1560 delete $self->{_result_inflator};
1562 $self->_result_class;
1569 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1571 =item Return Value: $count
1575 Performs an SQL C<COUNT> with the same query as the resultset was built
1576 with to find the number of elements. Passing arguments is equivalent to
1577 C<< $rs->search ($cond, \%attrs)->count >>
1583 return $self->search(@_)->count if @_ and defined $_[0];
1584 return scalar @{ $self->get_cache } if $self->get_cache;
1586 my $attrs = { %{ $self->_resolved_attrs } };
1588 # this is a little optimization - it is faster to do the limit
1589 # adjustments in software, instead of a subquery
1590 my ($rows, $offset) = delete @{$attrs}{qw/rows offset/};
1593 if ($self->_has_resolved_attr (qw/collapse group_by/)) {
1594 $crs = $self->_count_subq_rs ($attrs);
1597 $crs = $self->_count_rs ($attrs);
1599 my $count = $crs->next;
1601 $count -= $offset if $offset;
1602 $count = $rows if $rows and $rows < $count;
1603 $count = 0 if ($count < 0);
1612 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1614 =item Return Value: L<$count_rs|DBIx::Class::ResultSetColumn>
1618 Same as L</count> but returns a L<DBIx::Class::ResultSetColumn> object.
1619 This can be very handy for subqueries:
1621 ->search( { amount => $some_rs->count_rs->as_query } )
1623 As with regular resultsets the SQL query will be executed only after
1624 the resultset is accessed via L</next> or L</all>. That would return
1625 the same single value obtainable via L</count>.
1631 return $self->search(@_)->count_rs if @_;
1633 # this may look like a lack of abstraction (count() does about the same)
1634 # but in fact an _rs *must* use a subquery for the limits, as the
1635 # software based limiting can not be ported if this $rs is to be used
1636 # in a subquery itself (i.e. ->as_query)
1637 if ($self->_has_resolved_attr (qw/collapse group_by offset rows/)) {
1638 return $self->_count_subq_rs($self->{_attrs});
1641 return $self->_count_rs($self->{_attrs});
1646 # returns a ResultSetColumn object tied to the count query
1649 my ($self, $attrs) = @_;
1651 my $rsrc = $self->result_source;
1653 my $tmp_attrs = { %$attrs };
1654 # take off any limits, record_filter is cdbi, and no point of ordering nor locking a count
1655 delete @{$tmp_attrs}{qw/rows offset order_by record_filter for/};
1657 # overwrite the selector (supplied by the storage)
1658 $rsrc->resultset_class->new($rsrc, {
1660 select => $rsrc->storage->_count_select ($rsrc, $attrs),
1662 })->get_column ('count');
1666 # same as above but uses a subquery
1668 sub _count_subq_rs {
1669 my ($self, $attrs) = @_;
1671 my $rsrc = $self->result_source;
1673 my $sub_attrs = { %$attrs };
1674 # extra selectors do not go in the subquery and there is no point of ordering it, nor locking it
1675 delete @{$sub_attrs}{qw/collapse columns as select order_by for/};
1677 # if we multi-prefetch we group_by something unique, as this is what we would
1678 # get out of the rs via ->next/->all. We *DO WANT* to clobber old group_by regardless
1679 if ( $attrs->{collapse} ) {
1680 $sub_attrs->{group_by} = [ map { "$attrs->{alias}.$_" } @{
1681 $rsrc->_identifying_column_set || $self->throw_exception(
1682 'Unable to construct a unique group_by criteria properly collapsing the '
1683 . 'has_many prefetch before count()'
1688 # Calculate subquery selector
1689 if (my $g = $sub_attrs->{group_by}) {
1691 my $sql_maker = $rsrc->storage->sql_maker;
1693 # necessary as the group_by may refer to aliased functions
1695 for my $sel (@{$attrs->{select}}) {
1696 $sel_index->{$sel->{-as}} = $sel
1697 if (ref $sel eq 'HASH' and $sel->{-as});
1700 # anything from the original select mentioned on the group-by needs to make it to the inner selector
1701 # also look for named aggregates referred in the having clause
1702 # having often contains scalarrefs - thus parse it out entirely
1704 if ($attrs->{having}) {
1705 local $sql_maker->{having_bind};
1706 local $sql_maker->{quote_char} = $sql_maker->{quote_char};
1707 local $sql_maker->{name_sep} = $sql_maker->{name_sep};
1708 unless (defined $sql_maker->{quote_char} and length $sql_maker->{quote_char}) {
1709 $sql_maker->{quote_char} = [ "\x00", "\xFF" ];
1710 # if we don't unset it we screw up retarded but unfortunately working
1711 # 'MAX(foo.bar)' => { '>', 3 }
1712 $sql_maker->{name_sep} = '';
1715 my ($lquote, $rquote, $sep) = map { quotemeta $_ } ($sql_maker->_quote_chars, $sql_maker->name_sep);
1717 my $having_sql = $sql_maker->_parse_rs_attrs ({ having => $attrs->{having} });
1720 # search for both a proper quoted qualified string, for a naive unquoted scalarref
1721 # and if all fails for an utterly naive quoted scalar-with-function
1722 while ($having_sql =~ /
1723 $rquote $sep $lquote (.+?) $rquote
1725 [\s,] \w+ \. (\w+) [\s,]
1727 [\s,] $lquote (.+?) $rquote [\s,]
1729 my $part = $1 || $2 || $3; # one of them matched if we got here
1730 unless ($seen_having{$part}++) {
1737 my $colpiece = $sel_index->{$_} || $_;
1739 # unqualify join-based group_by's. Arcane but possible query
1740 # also horrible horrible hack to alias a column (not a func.)
1741 # (probably need to introduce SQLA syntax)
1742 if ($colpiece =~ /\./ && $colpiece !~ /^$attrs->{alias}\./) {
1745 $colpiece = \ sprintf ('%s AS %s', map { $sql_maker->_quote ($_) } ($colpiece, $as) );
1747 push @{$sub_attrs->{select}}, $colpiece;
1751 my @pcols = map { "$attrs->{alias}.$_" } ($rsrc->primary_columns);
1752 $sub_attrs->{select} = @pcols ? \@pcols : [ 1 ];
1755 return $rsrc->resultset_class
1756 ->new ($rsrc, $sub_attrs)
1758 ->search ({}, { columns => { count => $rsrc->storage->_count_select ($rsrc, $attrs) } })
1759 ->get_column ('count');
1763 =head2 count_literal
1765 B<CAVEAT>: C<count_literal> is provided for Class::DBI compatibility and
1766 should only be used in that context. See L</search_literal> for further info.
1770 =item Arguments: $sql_fragment, @standalone_bind_values
1772 =item Return Value: $count
1776 Counts the results in a literal query. Equivalent to calling L</search_literal>
1777 with the passed arguments, then L</count>.
1781 sub count_literal { shift->search_literal(@_)->count; }
1787 =item Arguments: none
1789 =item Return Value: L<@result_objs|DBIx::Class::Manual::ResultClass>
1793 Returns all elements in the resultset.
1800 $self->throw_exception("all() doesn't take any arguments, you probably wanted ->search(...)->all()");
1803 delete @{$self}{qw/_stashed_rows _stashed_results/};
1805 if (my $c = $self->get_cache) {
1809 $self->cursor->reset;
1811 my $objs = $self->_construct_results('fetch_all') || [];
1813 $self->set_cache($objs) if $self->{attrs}{cache};
1822 =item Arguments: none
1824 =item Return Value: $self
1828 Resets the resultset's cursor, so you can iterate through the elements again.
1829 Implicitly resets the storage cursor, so a subsequent L</next> will trigger
1837 delete @{$self}{qw/_stashed_rows _stashed_results/};
1838 $self->{all_cache_position} = 0;
1839 $self->cursor->reset;
1847 =item Arguments: none
1849 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1853 L<Resets|/reset> the resultset (causing a fresh query to storage) and returns
1854 an object for the first result (or C<undef> if the resultset is empty).
1859 return $_[0]->reset->next;
1865 # Determines whether and what type of subquery is required for the $rs operation.
1866 # If grouping is necessary either supplies its own, or verifies the current one
1867 # After all is done delegates to the proper storage method.
1869 sub _rs_update_delete {
1870 my ($self, $op, $values) = @_;
1872 my $rsrc = $self->result_source;
1873 my $storage = $rsrc->schema->storage;
1875 my $attrs = { %{$self->_resolved_attrs} };
1877 my $join_classifications;
1878 my ($existing_group_by) = delete @{$attrs}{qw(group_by _grouped_by_distinct)};
1880 # do we need a subquery for any reason?
1882 defined $existing_group_by
1884 # if {from} is unparseable wrap a subq
1885 ref($attrs->{from}) ne 'ARRAY'
1887 # limits call for a subq
1888 $self->_has_resolved_attr(qw/rows offset/)
1891 # simplify the joinmap, so we can further decide if a subq is necessary
1892 if (!$needs_subq and @{$attrs->{from}} > 1) {
1894 ($attrs->{from}, $join_classifications) =
1895 $storage->_prune_unused_joins ($attrs);
1897 # any non-pruneable non-local restricting joins imply subq
1898 $needs_subq = defined List::Util::first { $_ ne $attrs->{alias} } keys %{ $join_classifications->{restricting} || {} };
1901 # check if the head is composite (by now all joins are thrown out unless $needs_subq)
1903 (ref $attrs->{from}[0]) ne 'HASH'
1905 ref $attrs->{from}[0]{ $attrs->{from}[0]{-alias} }
1909 # do we need anything like a subquery?
1910 if (! $needs_subq) {
1911 # Most databases do not allow aliasing of tables in UPDATE/DELETE. Thus
1912 # a condition containing 'me' or other table prefixes will not work
1913 # at all. Tell SQLMaker to dequalify idents via a gross hack.
1915 my $sqla = $rsrc->storage->sql_maker;
1916 local $sqla->{_dequalify_idents} = 1;
1917 \[ $sqla->_recurse_where($self->{cond}) ];
1921 # we got this far - means it is time to wrap a subquery
1922 my $idcols = $rsrc->_identifying_column_set || $self->throw_exception(
1924 "Unable to perform complex resultset %s() without an identifying set of columns on source '%s'",
1930 # make a new $rs selecting only the PKs (that's all we really need for the subq)
1931 delete $attrs->{$_} for qw/select as collapse/;
1932 $attrs->{columns} = [ map { "$attrs->{alias}.$_" } @$idcols ];
1934 # this will be consumed by the pruner waaaaay down the stack
1935 $attrs->{_force_prune_multiplying_joins} = 1;
1937 my $subrs = (ref $self)->new($rsrc, $attrs);
1939 if (@$idcols == 1) {
1940 $cond = { $idcols->[0] => { -in => $subrs->as_query } };
1942 elsif ($storage->_use_multicolumn_in) {
1943 # no syntax for calling this properly yet
1944 # !!! EXPERIMENTAL API !!! WILL CHANGE !!!
1945 $cond = $storage->sql_maker->_where_op_multicolumn_in (
1946 $idcols, # how do I convey a list of idents...? can binds reside on lhs?
1951 # if all else fails - get all primary keys and operate over a ORed set
1952 # wrap in a transaction for consistency
1953 # this is where the group_by/multiplication starts to matter
1957 # we do not need to check pre-multipliers, since if the premulti is there, its
1958 # parent (who is multi) will be there too
1959 keys %{ $join_classifications->{multiplying} || {} }
1961 # make sure if there is a supplied group_by it matches the columns compiled above
1962 # perfectly. Anything else can not be sanely executed on most databases so croak
1963 # right then and there
1964 if ($existing_group_by) {
1965 my @current_group_by = map
1966 { $_ =~ /\./ ? $_ : "$attrs->{alias}.$_" }
1971 join ("\x00", sort @current_group_by)
1973 join ("\x00", sort @{$attrs->{columns}} )
1975 $self->throw_exception (
1976 "You have just attempted a $op operation on a resultset which does group_by"
1977 . ' on columns other than the primary keys, while DBIC internally needs to retrieve'
1978 . ' the primary keys in a subselect. All sane RDBMS engines do not support this'
1979 . ' kind of queries. Please retry the operation with a modified group_by or'
1980 . ' without using one at all.'
1985 $subrs = $subrs->search({}, { group_by => $attrs->{columns} });
1988 $guard = $storage->txn_scope_guard;
1990 for my $row ($subrs->cursor->all) {
1992 { $idcols->[$_] => $row->[$_] }
1999 my $res = $cond ? $storage->$op (
2001 $op eq 'update' ? $values : (),
2005 $guard->commit if $guard;
2014 =item Arguments: \%values
2016 =item Return Value: $underlying_storage_rv
2020 Sets the specified columns in the resultset to the supplied values in a
2021 single query. Note that this will not run any accessor/set_column/update
2022 triggers, nor will it update any result object instances derived from this
2023 resultset (this includes the contents of the L<resultset cache|/set_cache>
2024 if any). See L</update_all> if you need to execute any on-update
2025 triggers or cascades defined either by you or a
2026 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
2028 The return value is a pass through of what the underlying
2029 storage backend returned, and may vary. See L<DBI/execute> for the most
2034 Note that L</update> does not process/deflate any of the values passed in.
2035 This is unlike the corresponding L<DBIx::Class::Row/update>. The user must
2036 ensure manually that any value passed to this method will stringify to
2037 something the RDBMS knows how to deal with. A notable example is the
2038 handling of L<DateTime> objects, for more info see:
2039 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
2044 my ($self, $values) = @_;
2045 $self->throw_exception('Values for update must be a hash')
2046 unless ref $values eq 'HASH';
2048 return $self->_rs_update_delete ('update', $values);
2055 =item Arguments: \%values
2057 =item Return Value: 1
2061 Fetches all objects and updates them one at a time via
2062 L<DBIx::Class::Row/update>. Note that C<update_all> will run DBIC defined
2063 triggers, while L</update> will not.
2068 my ($self, $values) = @_;
2069 $self->throw_exception('Values for update_all must be a hash')
2070 unless ref $values eq 'HASH';
2072 my $guard = $self->result_source->schema->txn_scope_guard;
2073 $_->update({%$values}) for $self->all; # shallow copy - update will mangle it
2082 =item Arguments: none
2084 =item Return Value: $underlying_storage_rv
2088 Deletes the rows matching this resultset in a single query. Note that this
2089 will not run any delete triggers, nor will it alter the
2090 L<in_storage|DBIx::Class::Row/in_storage> status of any result object instances
2091 derived from this resultset (this includes the contents of the
2092 L<resultset cache|/set_cache> if any). See L</delete_all> if you need to
2093 execute any on-delete triggers or cascades defined either by you or a
2094 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
2096 The return value is a pass through of what the underlying storage backend
2097 returned, and may vary. See L<DBI/execute> for the most common case.
2103 $self->throw_exception('delete does not accept any arguments')
2106 return $self->_rs_update_delete ('delete');
2113 =item Arguments: none
2115 =item Return Value: 1
2119 Fetches all objects and deletes them one at a time via
2120 L<DBIx::Class::Row/delete>. Note that C<delete_all> will run DBIC defined
2121 triggers, while L</delete> will not.
2127 $self->throw_exception('delete_all does not accept any arguments')
2130 my $guard = $self->result_source->schema->txn_scope_guard;
2131 $_->delete for $self->all;
2140 =item Arguments: [ \@column_list, \@row_values+ ] | [ \%col_data+ ]
2142 =item Return Value: L<\@result_objects|DBIx::Class::Manual::ResultClass> (scalar context) | L<@result_objects|DBIx::Class::Manual::ResultClass> (list context)
2146 Accepts either an arrayref of hashrefs or alternatively an arrayref of
2153 The context of this method call has an important effect on what is
2154 submitted to storage. In void context data is fed directly to fastpath
2155 insertion routines provided by the underlying storage (most often
2156 L<DBI/execute_for_fetch>), bypassing the L<new|DBIx::Class::Row/new> and
2157 L<insert|DBIx::Class::Row/insert> calls on the
2158 L<Result|DBIx::Class::Manual::ResultClass> class, including any
2159 augmentation of these methods provided by components. For example if you
2160 are using something like L<DBIx::Class::UUIDColumns> to create primary
2161 keys for you, you will find that your PKs are empty. In this case you
2162 will have to explicitly force scalar or list context in order to create
2167 In non-void (scalar or list) context, this method is simply a wrapper
2168 for L</create>. Depending on list or scalar context either a list of
2169 L<Result|DBIx::Class::Manual::ResultClass> objects or an arrayref
2170 containing these objects is returned.
2172 When supplying data in "arrayref of arrayrefs" invocation style, the
2173 first element should be a list of column names and each subsequent
2174 element should be a data value in the earlier specified column order.
2177 $schema->resultset("Artist")->populate([
2178 [ qw( artistid name ) ],
2179 [ 100, 'A Formally Unknown Singer' ],
2180 [ 101, 'A singer that jumped the shark two albums ago' ],
2181 [ 102, 'An actually cool singer' ],
2184 For the arrayref of hashrefs style each hashref should be a structure
2185 suitable for passing to L</create>. Multi-create is also permitted with
2188 $schema->resultset("Artist")->populate([
2189 { artistid => 4, name => 'Manufactured Crap', cds => [
2190 { title => 'My First CD', year => 2006 },
2191 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2194 { artistid => 5, name => 'Angsty-Whiny Girl', cds => [
2195 { title => 'My parents sold me to a record company', year => 2005 },
2196 { title => 'Why Am I So Ugly?', year => 2006 },
2197 { title => 'I Got Surgery and am now Popular', year => 2007 }
2202 If you attempt a void-context multi-create as in the example above (each
2203 Artist also has the related list of CDs), and B<do not> supply the
2204 necessary autoinc foreign key information, this method will proxy to the
2205 less efficient L</create>, and then throw the Result objects away. In this
2206 case there are obviously no benefits to using this method over L</create>.
2213 # cruft placed in standalone method
2214 my $data = $self->_normalize_populate_args(@_);
2216 return unless @$data;
2218 if(defined wantarray) {
2219 my @created = map { $self->create($_) } @$data;
2220 return wantarray ? @created : \@created;
2223 my $first = $data->[0];
2225 # if a column is a registered relationship, and is a non-blessed hash/array, consider
2226 # it relationship data
2227 my (@rels, @columns);
2228 my $rsrc = $self->result_source;
2229 my $rels = { map { $_ => $rsrc->relationship_info($_) } $rsrc->relationships };
2230 for (keys %$first) {
2231 my $ref = ref $first->{$_};
2232 $rels->{$_} && ($ref eq 'ARRAY' or $ref eq 'HASH')
2238 my @pks = $rsrc->primary_columns;
2240 ## do the belongs_to relationships
2241 foreach my $index (0..$#$data) {
2243 # delegate to create() for any dataset without primary keys with specified relationships
2244 if (grep { !defined $data->[$index]->{$_} } @pks ) {
2246 if (grep { ref $data->[$index]{$r} eq $_ } qw/HASH ARRAY/) { # a related set must be a HASH or AoH
2247 my @ret = $self->populate($data);
2253 foreach my $rel (@rels) {
2254 next unless ref $data->[$index]->{$rel} eq "HASH";
2255 my $result = $self->related_resultset($rel)->create($data->[$index]->{$rel});
2256 my (undef, $reverse_relinfo) = %{$rsrc->reverse_relationship_info($rel)};
2257 my $related = $result->result_source->_resolve_condition(
2258 $reverse_relinfo->{cond},
2264 delete $data->[$index]->{$rel};
2265 $data->[$index] = {%{$data->[$index]}, %$related};
2267 push @columns, keys %$related if $index == 0;
2271 ## inherit the data locked in the conditions of the resultset
2272 my ($rs_data) = $self->_merge_with_rscond({});
2273 delete @{$rs_data}{@columns};
2275 ## do bulk insert on current row
2276 $rsrc->storage->insert_bulk(
2278 [@columns, keys %$rs_data],
2279 [ map { [ @$_{@columns}, values %$rs_data ] } @$data ],
2282 ## do the has_many relationships
2283 foreach my $item (@$data) {
2287 foreach my $rel (@rels) {
2288 next unless ref $item->{$rel} eq "ARRAY" && @{ $item->{$rel} };
2290 $main_row ||= $self->new_result({map { $_ => $item->{$_} } @pks});
2292 my $child = $main_row->$rel;
2294 my $related = $child->result_source->_resolve_condition(
2295 $rels->{$rel}{cond},
2301 my @rows_to_add = ref $item->{$rel} eq 'ARRAY' ? @{$item->{$rel}} : ($item->{$rel});
2302 my @populate = map { {%$_, %$related} } @rows_to_add;
2304 $child->populate( \@populate );
2311 # populate() arguments went over several incarnations
2312 # What we ultimately support is AoH
2313 sub _normalize_populate_args {
2314 my ($self, $arg) = @_;
2316 if (ref $arg eq 'ARRAY') {
2320 elsif (ref $arg->[0] eq 'HASH') {
2323 elsif (ref $arg->[0] eq 'ARRAY') {
2325 my @colnames = @{$arg->[0]};
2326 foreach my $values (@{$arg}[1 .. $#$arg]) {
2327 push @ret, { map { $colnames[$_] => $values->[$_] } (0 .. $#colnames) };
2333 $self->throw_exception('Populate expects an arrayref of hashrefs or arrayref of arrayrefs');
2340 =item Arguments: none
2342 =item Return Value: L<$pager|Data::Page>
2346 Returns a L<Data::Page> object for the current resultset. Only makes
2347 sense for queries with a C<page> attribute.
2349 To get the full count of entries for a paged resultset, call
2350 C<total_entries> on the L<Data::Page> object.
2357 return $self->{pager} if $self->{pager};
2359 my $attrs = $self->{attrs};
2360 if (!defined $attrs->{page}) {
2361 $self->throw_exception("Can't create pager for non-paged rs");
2363 elsif ($attrs->{page} <= 0) {
2364 $self->throw_exception('Invalid page number (page-numbers are 1-based)');
2366 $attrs->{rows} ||= 10;
2368 # throw away the paging flags and re-run the count (possibly
2369 # with a subselect) to get the real total count
2370 my $count_attrs = { %$attrs };
2371 delete @{$count_attrs}{qw/rows offset page pager/};
2373 my $total_rs = (ref $self)->new($self->result_source, $count_attrs);
2375 require DBIx::Class::ResultSet::Pager;
2376 return $self->{pager} = DBIx::Class::ResultSet::Pager->new(
2377 sub { $total_rs->count }, #lazy-get the total
2379 $self->{attrs}{page},
2387 =item Arguments: $page_number
2389 =item Return Value: L<$resultset|/search>
2393 Returns a resultset for the $page_number page of the resultset on which page
2394 is called, where each page contains a number of rows equal to the 'rows'
2395 attribute set on the resultset (10 by default).
2400 my ($self, $page) = @_;
2401 return (ref $self)->new($self->result_source, { %{$self->{attrs}}, page => $page });
2408 =item Arguments: \%col_data
2410 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2414 Creates a new result object in the resultset's result class and returns
2415 it. The row is not inserted into the database at this point, call
2416 L<DBIx::Class::Row/insert> to do that. Calling L<DBIx::Class::Row/in_storage>
2417 will tell you whether the result object has been inserted or not.
2419 Passes the hashref of input on to L<DBIx::Class::Row/new>.
2424 my ($self, $values) = @_;
2426 $self->throw_exception( "new_result takes only one argument - a hashref of values" )
2429 $self->throw_exception( "new_result expects a hashref" )
2430 unless (ref $values eq 'HASH');
2432 my ($merged_cond, $cols_from_relations) = $self->_merge_with_rscond($values);
2434 my $new = $self->result_class->new({
2436 ( @$cols_from_relations
2437 ? (-cols_from_relations => $cols_from_relations)
2440 -result_source => $self->result_source, # DO NOT REMOVE THIS, REQUIRED
2444 reftype($new) eq 'HASH'
2450 carp_unique (sprintf (
2451 "%s->new returned a blessed empty hashref - a strong indicator something is wrong with its inheritance chain",
2452 $self->result_class,
2459 # _merge_with_rscond
2461 # Takes a simple hash of K/V data and returns its copy merged with the
2462 # condition already present on the resultset. Additionally returns an
2463 # arrayref of value/condition names, which were inferred from related
2464 # objects (this is needed for in-memory related objects)
2465 sub _merge_with_rscond {
2466 my ($self, $data) = @_;
2468 my (%new_data, @cols_from_relations);
2470 my $alias = $self->{attrs}{alias};
2472 if (! defined $self->{cond}) {
2473 # just massage $data below
2475 elsif ($self->{cond} eq $DBIx::Class::ResultSource::UNRESOLVABLE_CONDITION) {
2476 %new_data = %{ $self->{attrs}{related_objects} || {} }; # nothing might have been inserted yet
2477 @cols_from_relations = keys %new_data;
2479 elsif (ref $self->{cond} ne 'HASH') {
2480 $self->throw_exception(
2481 "Can't abstract implicit construct, resultset condition not a hash"
2485 if ($self->{cond}) {
2486 my $implied = $self->_remove_alias(
2487 $self->result_source->schema->storage->_collapse_cond($self->{cond}),
2491 for my $c (keys %$implied) {
2492 my $v = $implied->{$c};
2493 if ( ! length ref $v or is_plain_value($v) ) {
2497 ref $v eq 'HASH' and keys %$v == 1 and exists $v->{'='} and is_literal_value($v->{'='})
2499 $new_data{$c} = $v->{'='};
2505 # precedence must be given to passed values over values inherited from
2506 # the cond, so the order here is important.
2509 %{ $self->_remove_alias($data, $alias) },
2512 return (\%new_data, \@cols_from_relations);
2515 # _has_resolved_attr
2517 # determines if the resultset defines at least one
2518 # of the attributes supplied
2520 # used to determine if a subquery is necessary
2522 # supports some virtual attributes:
2524 # This will scan for any joins being present on the resultset.
2525 # It is not a mere key-search but a deep inspection of {from}
2528 sub _has_resolved_attr {
2529 my ($self, @attr_names) = @_;
2531 my $attrs = $self->_resolved_attrs;
2535 for my $n (@attr_names) {
2536 if (grep { $n eq $_ } (qw/-join/) ) {
2537 $extra_checks{$n}++;
2541 my $attr = $attrs->{$n};
2543 next if not defined $attr;
2545 if (ref $attr eq 'HASH') {
2546 return 1 if keys %$attr;
2548 elsif (ref $attr eq 'ARRAY') {
2556 # a resolved join is expressed as a multi-level from
2558 $extra_checks{-join}
2560 ref $attrs->{from} eq 'ARRAY'
2562 @{$attrs->{from}} > 1
2570 # Remove the specified alias from the specified query hash. A copy is made so
2571 # the original query is not modified.
2574 my ($self, $query, $alias) = @_;
2576 my %orig = %{ $query || {} };
2579 foreach my $key (keys %orig) {
2581 $unaliased{$key} = $orig{$key};
2584 $unaliased{$1} = $orig{$key}
2585 if $key =~ m/^(?:\Q$alias\E\.)?([^.]+)$/;
2595 =item Arguments: none
2597 =item Return Value: \[ $sql, L<@bind_values|/DBIC BIND VALUES> ]
2601 Returns the SQL query and bind vars associated with the invocant.
2603 This is generally used as the RHS for a subquery.
2610 my $attrs = { %{ $self->_resolved_attrs } };
2612 my $aq = $self->result_source->storage->_select_args_to_query (
2613 $attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs
2623 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2625 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2629 my $artist = $schema->resultset('Artist')->find_or_new(
2630 { artist => 'fred' }, { key => 'artists' });
2632 $cd->cd_to_producer->find_or_new({ producer => $producer },
2633 { key => 'primary' });
2635 Find an existing record from this resultset using L</find>. if none exists,
2636 instantiate a new result object and return it. The object will not be saved
2637 into your storage until you call L<DBIx::Class::Row/insert> on it.
2639 You most likely want this method when looking for existing rows using a unique
2640 constraint that is not the primary key, or looking for related rows.
2642 If you want objects to be saved immediately, use L</find_or_create> instead.
2644 B<Note>: Make sure to read the documentation of L</find> and understand the
2645 significance of the C<key> attribute, as its lack may skew your search, and
2646 subsequently result in spurious new objects.
2648 B<Note>: Take care when using C<find_or_new> with a table having
2649 columns with default values that you intend to be automatically
2650 supplied by the database (e.g. an auto_increment primary key column).
2651 In normal usage, the value of such columns should NOT be included at
2652 all in the call to C<find_or_new>, even when set to C<undef>.
2658 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2659 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2660 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2663 return $self->new_result($hash);
2670 =item Arguments: \%col_data
2672 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2676 Attempt to create a single new row or a row with multiple related rows
2677 in the table represented by the resultset (and related tables). This
2678 will not check for duplicate rows before inserting, use
2679 L</find_or_create> to do that.
2681 To create one row for this resultset, pass a hashref of key/value
2682 pairs representing the columns of the table and the values you wish to
2683 store. If the appropriate relationships are set up, foreign key fields
2684 can also be passed an object representing the foreign row, and the
2685 value will be set to its primary key.
2687 To create related objects, pass a hashref of related-object column values
2688 B<keyed on the relationship name>. If the relationship is of type C<multi>
2689 (L<DBIx::Class::Relationship/has_many>) - pass an arrayref of hashrefs.
2690 The process will correctly identify columns holding foreign keys, and will
2691 transparently populate them from the keys of the corresponding relation.
2692 This can be applied recursively, and will work correctly for a structure
2693 with an arbitrary depth and width, as long as the relationships actually
2694 exists and the correct column data has been supplied.
2696 Instead of hashrefs of plain related data (key/value pairs), you may
2697 also pass new or inserted objects. New objects (not inserted yet, see
2698 L</new_result>), will be inserted into their appropriate tables.
2700 Effectively a shortcut for C<< ->new_result(\%col_data)->insert >>.
2702 Example of creating a new row.
2704 $person_rs->create({
2705 name=>"Some Person",
2706 email=>"somebody@someplace.com"
2709 Example of creating a new row and also creating rows in a related C<has_many>
2710 or C<has_one> resultset. Note Arrayref.
2713 { artistid => 4, name => 'Manufactured Crap', cds => [
2714 { title => 'My First CD', year => 2006 },
2715 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2720 Example of creating a new row and also creating a row in a related
2721 C<belongs_to> resultset. Note Hashref.
2724 title=>"Music for Silly Walks",
2727 name=>"Silly Musician",
2735 When subclassing ResultSet never attempt to override this method. Since
2736 it is a simple shortcut for C<< $self->new_result($attrs)->insert >>, a
2737 lot of the internals simply never call it, so your override will be
2738 bypassed more often than not. Override either L<DBIx::Class::Row/new>
2739 or L<DBIx::Class::Row/insert> depending on how early in the
2740 L</create> process you need to intervene. See also warning pertaining to
2748 my ($self, $col_data) = @_;
2749 $self->throw_exception( "create needs a hashref" )
2750 unless ref $col_data eq 'HASH';
2751 return $self->new_result($col_data)->insert;
2754 =head2 find_or_create
2758 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2760 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2764 $cd->cd_to_producer->find_or_create({ producer => $producer },
2765 { key => 'primary' });
2767 Tries to find a record based on its primary key or unique constraints; if none
2768 is found, creates one and returns that instead.
2770 my $cd = $schema->resultset('CD')->find_or_create({
2772 artist => 'Massive Attack',
2773 title => 'Mezzanine',
2777 Also takes an optional C<key> attribute, to search by a specific key or unique
2778 constraint. For example:
2780 my $cd = $schema->resultset('CD')->find_or_create(
2782 artist => 'Massive Attack',
2783 title => 'Mezzanine',
2785 { key => 'cd_artist_title' }
2788 B<Note>: Make sure to read the documentation of L</find> and understand the
2789 significance of the C<key> attribute, as its lack may skew your search, and
2790 subsequently result in spurious row creation.
2792 B<Note>: Because find_or_create() reads from the database and then
2793 possibly inserts based on the result, this method is subject to a race
2794 condition. Another process could create a record in the table after
2795 the find has completed and before the create has started. To avoid
2796 this problem, use find_or_create() inside a transaction.
2798 B<Note>: Take care when using C<find_or_create> with a table having
2799 columns with default values that you intend to be automatically
2800 supplied by the database (e.g. an auto_increment primary key column).
2801 In normal usage, the value of such columns should NOT be included at
2802 all in the call to C<find_or_create>, even when set to C<undef>.
2804 See also L</find> and L</update_or_create>. For information on how to declare
2805 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2807 If you need to know if an existing row was found or a new one created use
2808 L</find_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
2809 to call L<DBIx::Class::Row/insert> to save the newly created row to the
2812 my $cd = $schema->resultset('CD')->find_or_new({
2814 artist => 'Massive Attack',
2815 title => 'Mezzanine',
2819 if( !$cd->in_storage ) {
2826 sub find_or_create {
2828 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2829 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2830 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2833 return $self->create($hash);
2836 =head2 update_or_create
2840 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2842 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2846 $resultset->update_or_create({ col => $val, ... });
2848 Like L</find_or_create>, but if a row is found it is immediately updated via
2849 C<< $found_row->update (\%col_data) >>.
2852 Takes an optional C<key> attribute to search on a specific unique constraint.
2855 # In your application
2856 my $cd = $schema->resultset('CD')->update_or_create(
2858 artist => 'Massive Attack',
2859 title => 'Mezzanine',
2862 { key => 'cd_artist_title' }
2865 $cd->cd_to_producer->update_or_create({
2866 producer => $producer,
2872 B<Note>: Make sure to read the documentation of L</find> and understand the
2873 significance of the C<key> attribute, as its lack may skew your search, and
2874 subsequently result in spurious row creation.
2876 B<Note>: Take care when using C<update_or_create> with a table having
2877 columns with default values that you intend to be automatically
2878 supplied by the database (e.g. an auto_increment primary key column).
2879 In normal usage, the value of such columns should NOT be included at
2880 all in the call to C<update_or_create>, even when set to C<undef>.
2882 See also L</find> and L</find_or_create>. For information on how to declare
2883 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2885 If you need to know if an existing row was updated or a new one created use
2886 L</update_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
2887 to call L<DBIx::Class::Row/insert> to save the newly created row to the
2892 sub update_or_create {
2894 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2895 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2897 my $row = $self->find($cond, $attrs);
2899 $row->update($cond);
2903 return $self->create($cond);
2906 =head2 update_or_new
2910 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2912 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2916 $resultset->update_or_new({ col => $val, ... });
2918 Like L</find_or_new> but if a row is found it is immediately updated via
2919 C<< $found_row->update (\%col_data) >>.
2923 # In your application
2924 my $cd = $schema->resultset('CD')->update_or_new(
2926 artist => 'Massive Attack',
2927 title => 'Mezzanine',
2930 { key => 'cd_artist_title' }
2933 if ($cd->in_storage) {
2934 # the cd was updated
2937 # the cd is not yet in the database, let's insert it
2941 B<Note>: Make sure to read the documentation of L</find> and understand the
2942 significance of the C<key> attribute, as its lack may skew your search, and
2943 subsequently result in spurious new objects.
2945 B<Note>: Take care when using C<update_or_new> with a table having
2946 columns with default values that you intend to be automatically
2947 supplied by the database (e.g. an auto_increment primary key column).
2948 In normal usage, the value of such columns should NOT be included at
2949 all in the call to C<update_or_new>, even when set to C<undef>.
2951 See also L</find>, L</find_or_create> and L</find_or_new>.
2957 my $attrs = ( @_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {} );
2958 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2960 my $row = $self->find( $cond, $attrs );
2961 if ( defined $row ) {
2962 $row->update($cond);
2966 return $self->new_result($cond);
2973 =item Arguments: none
2975 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass> | undef
2979 Gets the contents of the cache for the resultset, if the cache is set.
2981 The cache is populated either by using the L</prefetch> attribute to
2982 L</search> or by calling L</set_cache>.
2994 =item Arguments: L<\@result_objs|DBIx::Class::Manual::ResultClass>
2996 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass>
3000 Sets the contents of the cache for the resultset. Expects an arrayref
3001 of objects of the same class as those produced by the resultset. Note that
3002 if the cache is set, the resultset will return the cached objects rather
3003 than re-querying the database even if the cache attr is not set.
3005 The contents of the cache can also be populated by using the
3006 L</prefetch> attribute to L</search>.
3011 my ( $self, $data ) = @_;
3012 $self->throw_exception("set_cache requires an arrayref")
3013 if defined($data) && (ref $data ne 'ARRAY');
3014 $self->{all_cache} = $data;
3021 =item Arguments: none
3023 =item Return Value: undef
3027 Clears the cache for the resultset.
3032 shift->set_cache(undef);
3039 =item Arguments: none
3041 =item Return Value: true, if the resultset has been paginated
3049 return !!$self->{attrs}{page};
3056 =item Arguments: none
3058 =item Return Value: true, if the resultset has been ordered with C<order_by>.
3066 return scalar $self->result_source->storage->_extract_order_criteria($self->{attrs}{order_by});
3069 =head2 related_resultset
3073 =item Arguments: $rel_name
3075 =item Return Value: L<$resultset|/search>
3079 Returns a related resultset for the supplied relationship name.
3081 $artist_rs = $schema->resultset('CD')->related_resultset('Artist');
3085 sub related_resultset {
3086 my ($self, $rel) = @_;
3088 return $self->{related_resultsets}{$rel}
3089 if defined $self->{related_resultsets}{$rel};
3091 return $self->{related_resultsets}{$rel} = do {
3092 my $rsrc = $self->result_source;
3093 my $rel_info = $rsrc->relationship_info($rel);
3095 $self->throw_exception(
3096 "search_related: result source '" . $rsrc->source_name .
3097 "' has no such relationship $rel")
3100 my $attrs = $self->_chain_relationship($rel);
3102 my $join_count = $attrs->{seen_join}{$rel};
3104 my $alias = $self->result_source->storage
3105 ->relname_to_table_alias($rel, $join_count);
3107 # since this is search_related, and we already slid the select window inwards
3108 # (the select/as attrs were deleted in the beginning), we need to flip all
3109 # left joins to inner, so we get the expected results
3110 # read the comment on top of the actual function to see what this does
3111 $attrs->{from} = $rsrc->schema->storage->_inner_join_to_node ($attrs->{from}, $alias);
3114 #XXX - temp fix for result_class bug. There likely is a more elegant fix -groditi
3115 delete @{$attrs}{qw(result_class alias)};
3117 my $rel_source = $rsrc->related_source($rel);
3121 # The reason we do this now instead of passing the alias to the
3122 # search_rs below is that if you wrap/overload resultset on the
3123 # source you need to know what alias it's -going- to have for things
3124 # to work sanely (e.g. RestrictWithObject wants to be able to add
3125 # extra query restrictions, and these may need to be $alias.)
3127 my $rel_attrs = $rel_source->resultset_attributes;
3128 local $rel_attrs->{alias} = $alias;
3130 $rel_source->resultset
3134 where => $attrs->{where},
3138 if (my $cache = $self->get_cache) {
3139 my @related_cache = map
3140 { $_->related_resultset($rel)->get_cache || () }
3144 $new->set_cache([ map @$_, @related_cache ]) if @related_cache == @$cache;
3151 =head2 current_source_alias
3155 =item Arguments: none
3157 =item Return Value: $source_alias
3161 Returns the current table alias for the result source this resultset is built
3162 on, that will be used in the SQL query. Usually it is C<me>.
3164 Currently the source alias that refers to the result set returned by a
3165 L</search>/L</find> family method depends on how you got to the resultset: it's
3166 C<me> by default, but eg. L</search_related> aliases it to the related result
3167 source name (and keeps C<me> referring to the original result set). The long
3168 term goal is to make L<DBIx::Class> always alias the current resultset as C<me>
3169 (and make this method unnecessary).
3171 Thus it's currently necessary to use this method in predefined queries (see
3172 L<DBIx::Class::Manual::Cookbook/Predefined searches>) when referring to the
3173 source alias of the current result set:
3175 # in a result set class
3177 my ($self, $user) = @_;
3179 my $me = $self->current_source_alias;
3181 return $self->search({
3182 "$me.modified" => $user->id,
3188 sub current_source_alias {
3189 return (shift->{attrs} || {})->{alias} || 'me';
3192 =head2 as_subselect_rs
3196 =item Arguments: none
3198 =item Return Value: L<$resultset|/search>
3202 Act as a barrier to SQL symbols. The resultset provided will be made into a
3203 "virtual view" by including it as a subquery within the from clause. From this
3204 point on, any joined tables are inaccessible to ->search on the resultset (as if
3205 it were simply where-filtered without joins). For example:
3207 my $rs = $schema->resultset('Bar')->search({'x.name' => 'abc'},{ join => 'x' });
3209 # 'x' now pollutes the query namespace
3211 # So the following works as expected
3212 my $ok_rs = $rs->search({'x.other' => 1});
3214 # But this doesn't: instead of finding a 'Bar' related to two x rows (abc and
3215 # def) we look for one row with contradictory terms and join in another table
3216 # (aliased 'x_2') which we never use
3217 my $broken_rs = $rs->search({'x.name' => 'def'});
3219 my $rs2 = $rs->as_subselect_rs;
3221 # doesn't work - 'x' is no longer accessible in $rs2, having been sealed away
3222 my $not_joined_rs = $rs2->search({'x.other' => 1});
3224 # works as expected: finds a 'table' row related to two x rows (abc and def)
3225 my $correctly_joined_rs = $rs2->search({'x.name' => 'def'});
3227 Another example of when one might use this would be to select a subset of
3228 columns in a group by clause:
3230 my $rs = $schema->resultset('Bar')->search(undef, {
3231 group_by => [qw{ id foo_id baz_id }],
3232 })->as_subselect_rs->search(undef, {
3233 columns => [qw{ id foo_id }]
3236 In the above example normally columns would have to be equal to the group by,
3237 but because we isolated the group by into a subselect the above works.
3241 sub as_subselect_rs {
3244 my $attrs = $self->_resolved_attrs;
3246 my $fresh_rs = (ref $self)->new (
3247 $self->result_source
3250 # these pieces will be locked in the subquery
3251 delete $fresh_rs->{cond};
3252 delete @{$fresh_rs->{attrs}}{qw/where bind/};
3254 return $fresh_rs->search( {}, {
3256 $attrs->{alias} => $self->as_query,
3257 -alias => $attrs->{alias},
3258 -rsrc => $self->result_source,
3260 alias => $attrs->{alias},
3264 # This code is called by search_related, and makes sure there
3265 # is clear separation between the joins before, during, and
3266 # after the relationship. This information is needed later
3267 # in order to properly resolve prefetch aliases (any alias
3268 # with a relation_chain_depth less than the depth of the
3269 # current prefetch is not considered)
3271 # The increments happen twice per join. An even number means a
3272 # relationship specified via a search_related, whereas an odd
3273 # number indicates a join/prefetch added via attributes
3275 # Also this code will wrap the current resultset (the one we
3276 # chain to) in a subselect IFF it contains limiting attributes
3277 sub _chain_relationship {
3278 my ($self, $rel) = @_;
3279 my $source = $self->result_source;
3280 my $attrs = { %{$self->{attrs}||{}} };
3282 # we need to take the prefetch the attrs into account before we
3283 # ->_resolve_join as otherwise they get lost - captainL
3284 my $join = $self->_merge_joinpref_attr( $attrs->{join}, $attrs->{prefetch} );
3286 delete @{$attrs}{qw/join prefetch collapse group_by distinct _grouped_by_distinct select as columns +select +as +columns/};
3288 my $seen = { %{ (delete $attrs->{seen_join}) || {} } };
3291 my @force_subq_attrs = qw/offset rows group_by having/;
3294 ($attrs->{from} && ref $attrs->{from} ne 'ARRAY')
3296 $self->_has_resolved_attr (@force_subq_attrs)
3298 # Nuke the prefetch (if any) before the new $rs attrs
3299 # are resolved (prefetch is useless - we are wrapping
3300 # a subquery anyway).
3301 my $rs_copy = $self->search;
3302 $rs_copy->{attrs}{join} = $self->_merge_joinpref_attr (
3303 $rs_copy->{attrs}{join},
3304 delete $rs_copy->{attrs}{prefetch},
3309 -alias => $attrs->{alias},
3310 $attrs->{alias} => $rs_copy->as_query,
3312 delete @{$attrs}{@force_subq_attrs, qw/where bind/};
3313 $seen->{-relation_chain_depth} = 0;
3315 elsif ($attrs->{from}) { #shallow copy suffices
3316 $from = [ @{$attrs->{from}} ];
3321 -alias => $attrs->{alias},
3322 $attrs->{alias} => $source->from,
3326 my $jpath = ($seen->{-relation_chain_depth})
3327 ? $from->[-1][0]{-join_path}
3330 my @requested_joins = $source->_resolve_join(
3337 push @$from, @requested_joins;
3339 $seen->{-relation_chain_depth}++;
3341 # if $self already had a join/prefetch specified on it, the requested
3342 # $rel might very well be already included. What we do in this case
3343 # is effectively a no-op (except that we bump up the chain_depth on
3344 # the join in question so we could tell it *is* the search_related)
3347 # we consider the last one thus reverse
3348 for my $j (reverse @requested_joins) {
3349 my ($last_j) = keys %{$j->[0]{-join_path}[-1]};
3350 if ($rel eq $last_j) {
3351 $j->[0]{-relation_chain_depth}++;
3357 unless ($already_joined) {
3358 push @$from, $source->_resolve_join(
3366 $seen->{-relation_chain_depth}++;
3368 return {%$attrs, from => $from, seen_join => $seen};
3371 sub _resolved_attrs {
3373 return $self->{_attrs} if $self->{_attrs};
3375 my $attrs = { %{ $self->{attrs} || {} } };
3376 my $source = $self->result_source;
3377 my $alias = $attrs->{alias};
3379 $self->throw_exception("Specifying distinct => 1 in conjunction with collapse => 1 is unsupported")
3380 if $attrs->{collapse} and $attrs->{distinct};
3382 # default selection list
3383 $attrs->{columns} = [ $source->columns ]
3384 unless List::Util::first { exists $attrs->{$_} } qw/columns cols select as/;
3386 # merge selectors together
3387 for (qw/columns select as/) {
3388 $attrs->{$_} = $self->_merge_attr($attrs->{$_}, delete $attrs->{"+$_"})
3389 if $attrs->{$_} or $attrs->{"+$_"};
3392 # disassemble columns
3394 if (my $cols = delete $attrs->{columns}) {
3395 for my $c (ref $cols eq 'ARRAY' ? @$cols : $cols) {
3396 if (ref $c eq 'HASH') {
3397 for my $as (sort keys %$c) {
3398 push @sel, $c->{$as};
3409 # when trying to weed off duplicates later do not go past this point -
3410 # everything added from here on is unbalanced "anyone's guess" stuff
3411 my $dedup_stop_idx = $#as;
3413 push @as, @{ ref $attrs->{as} eq 'ARRAY' ? $attrs->{as} : [ $attrs->{as} ] }
3415 push @sel, @{ ref $attrs->{select} eq 'ARRAY' ? $attrs->{select} : [ $attrs->{select} ] }
3416 if $attrs->{select};
3418 # assume all unqualified selectors to apply to the current alias (legacy stuff)
3419 $_ = (ref $_ or $_ =~ /\./) ? $_ : "$alias.$_" for @sel;
3421 # disqualify all $alias.col as-bits (inflate-map mandated)
3422 $_ = ($_ =~ /^\Q$alias.\E(.+)$/) ? $1 : $_ for @as;
3424 # de-duplicate the result (remove *identical* select/as pairs)
3425 # and also die on duplicate {as} pointing to different {select}s
3426 # not using a c-style for as the condition is prone to shrinkage
3429 while ($i <= $dedup_stop_idx) {
3430 if ($seen->{"$sel[$i] \x00\x00 $as[$i]"}++) {
3435 elsif ($seen->{$as[$i]}++) {
3436 $self->throw_exception(
3437 "inflate_result() alias '$as[$i]' specified twice with different SQL-side {select}-ors"
3445 $attrs->{select} = \@sel;
3446 $attrs->{as} = \@as;
3448 $attrs->{from} ||= [{
3450 -alias => $self->{attrs}{alias},
3451 $self->{attrs}{alias} => $source->from,
3454 if ( $attrs->{join} || $attrs->{prefetch} ) {
3456 $self->throw_exception ('join/prefetch can not be used with a custom {from}')
3457 if ref $attrs->{from} ne 'ARRAY';
3459 my $join = (delete $attrs->{join}) || {};
3461 if ( defined $attrs->{prefetch} ) {
3462 $join = $self->_merge_joinpref_attr( $join, $attrs->{prefetch} );
3465 $attrs->{from} = # have to copy here to avoid corrupting the original
3467 @{ $attrs->{from} },
3468 $source->_resolve_join(
3471 { %{ $attrs->{seen_join} || {} } },
3472 ( $attrs->{seen_join} && keys %{$attrs->{seen_join}})
3473 ? $attrs->{from}[-1][0]{-join_path}
3480 if ( defined $attrs->{order_by} ) {
3481 $attrs->{order_by} = (
3482 ref( $attrs->{order_by} ) eq 'ARRAY'
3483 ? [ @{ $attrs->{order_by} } ]
3484 : [ $attrs->{order_by} || () ]
3488 if ($attrs->{group_by} and ref $attrs->{group_by} ne 'ARRAY') {
3489 $attrs->{group_by} = [ $attrs->{group_by} ];
3493 # generate selections based on the prefetch helper
3494 my ($prefetch, @prefetch_select, @prefetch_as);
3495 $prefetch = $self->_merge_joinpref_attr( {}, delete $attrs->{prefetch} )
3496 if defined $attrs->{prefetch};
3500 $self->throw_exception("Unable to prefetch, resultset contains an unnamed selector $attrs->{_dark_selector}{string}")
3501 if $attrs->{_dark_selector};
3503 $self->throw_exception("Specifying prefetch in conjunction with an explicit collapse => 0 is unsupported")
3504 if defined $attrs->{collapse} and ! $attrs->{collapse};
3506 $attrs->{collapse} = 1;
3508 # this is a separate structure (we don't look in {from} directly)
3509 # as the resolver needs to shift things off the lists to work
3510 # properly (identical-prefetches on different branches)
3512 if (ref $attrs->{from} eq 'ARRAY') {
3514 my $start_depth = $attrs->{seen_join}{-relation_chain_depth} || 0;
3516 for my $j ( @{$attrs->{from}}[1 .. $#{$attrs->{from}} ] ) {
3517 next unless $j->[0]{-alias};
3518 next unless $j->[0]{-join_path};
3519 next if ($j->[0]{-relation_chain_depth} || 0) < $start_depth;
3521 my @jpath = map { keys %$_ } @{$j->[0]{-join_path}};
3524 $p = $p->{$_} ||= {} for @jpath[ ($start_depth/2) .. $#jpath]; #only even depths are actual jpath boundaries
3525 push @{$p->{-join_aliases} }, $j->[0]{-alias};
3529 my @prefetch = $source->_resolve_prefetch( $prefetch, $alias, $join_map );
3531 # save these for after distinct resolution
3532 @prefetch_select = map { $_->[0] } @prefetch;
3533 @prefetch_as = map { $_->[1] } @prefetch;
3536 # run through the resulting joinstructure (starting from our current slot)
3537 # and unset collapse if proven unnecessary
3539 # also while we are at it find out if the current root source has
3540 # been premultiplied by previous related_source chaining
3542 # this allows to predict whether a root object with all other relation
3543 # data set to NULL is in fact unique
3544 if ($attrs->{collapse}) {
3546 if (ref $attrs->{from} eq 'ARRAY') {
3548 if (@{$attrs->{from}} == 1) {
3549 # no joins - no collapse
3550 $attrs->{collapse} = 0;
3553 # find where our table-spec starts
3554 my @fromlist = @{$attrs->{from}};
3556 my $t = shift @fromlist;
3559 # me vs join from-spec distinction - a ref means non-root
3560 if (ref $t eq 'ARRAY') {
3562 $is_multi ||= ! $t->{-is_single};
3564 last if ($t->{-alias} && $t->{-alias} eq $alias);
3565 $attrs->{_main_source_premultiplied} ||= $is_multi;
3568 # no non-singles remaining, nor any premultiplication - nothing to collapse
3570 ! $attrs->{_main_source_premultiplied}
3572 ! List::Util::first { ! $_->[0]{-is_single} } @fromlist
3574 $attrs->{collapse} = 0;
3580 # if we can not analyze the from - err on the side of safety
3581 $attrs->{_main_source_premultiplied} = 1;
3585 # generate the distinct induced group_by before injecting the prefetched select/as parts
3586 if (delete $attrs->{distinct}) {
3587 if ($attrs->{group_by}) {
3588 carp_unique ("Useless use of distinct on a grouped resultset ('distinct' is ignored when a 'group_by' is present)");
3591 $attrs->{_grouped_by_distinct} = 1;
3592 # distinct affects only the main selection part, not what prefetch may add below
3593 ($attrs->{group_by}, my $new_order) = $source->storage->_group_over_selection($attrs);
3595 # FIXME possibly ignore a rewritten order_by (may turn out to be an issue)
3596 # The thinking is: if we are collapsing the subquerying prefetch engine will
3597 # rip stuff apart for us anyway, and we do not want to have a potentially
3598 # function-converted external order_by
3599 # ( there is an explicit if ( collapse && _grouped_by_distinct ) check in DBIHacks )
3600 $attrs->{order_by} = $new_order unless $attrs->{collapse};
3604 # inject prefetch-bound selection (if any)
3605 push @{$attrs->{select}}, @prefetch_select;
3606 push @{$attrs->{as}}, @prefetch_as;
3608 # whether we can get away with the dumbest (possibly DBI-internal) collapser
3609 if ( List::Util::first { $_ =~ /\./ } @{$attrs->{as}} ) {
3610 $attrs->{_related_results_construction} = 1;
3613 # if both page and offset are specified, produce a combined offset
3614 # even though it doesn't make much sense, this is what pre 081xx has
3616 if (my $page = delete $attrs->{page}) {
3618 ($attrs->{rows} * ($page - 1))
3620 ($attrs->{offset} || 0)
3624 return $self->{_attrs} = $attrs;
3628 my ($self, $attr) = @_;
3630 if (ref $attr eq 'HASH') {
3631 return $self->_rollout_hash($attr);
3632 } elsif (ref $attr eq 'ARRAY') {
3633 return $self->_rollout_array($attr);
3639 sub _rollout_array {
3640 my ($self, $attr) = @_;
3643 foreach my $element (@{$attr}) {
3644 if (ref $element eq 'HASH') {
3645 push( @rolled_array, @{ $self->_rollout_hash( $element ) } );
3646 } elsif (ref $element eq 'ARRAY') {
3647 # XXX - should probably recurse here
3648 push( @rolled_array, @{$self->_rollout_array($element)} );
3650 push( @rolled_array, $element );
3653 return \@rolled_array;
3657 my ($self, $attr) = @_;
3660 foreach my $key (keys %{$attr}) {
3661 push( @rolled_array, { $key => $attr->{$key} } );
3663 return \@rolled_array;
3666 sub _calculate_score {
3667 my ($self, $a, $b) = @_;
3669 if (defined $a xor defined $b) {
3672 elsif (not defined $a) {
3676 if (ref $b eq 'HASH') {
3677 my ($b_key) = keys %{$b};
3678 if (ref $a eq 'HASH') {
3679 my ($a_key) = keys %{$a};
3680 if ($a_key eq $b_key) {
3681 return (1 + $self->_calculate_score( $a->{$a_key}, $b->{$b_key} ));
3686 return ($a eq $b_key) ? 1 : 0;
3689 if (ref $a eq 'HASH') {
3690 my ($a_key) = keys %{$a};
3691 return ($b eq $a_key) ? 1 : 0;
3693 return ($b eq $a) ? 1 : 0;
3698 sub _merge_joinpref_attr {
3699 my ($self, $orig, $import) = @_;
3701 return $import unless defined($orig);
3702 return $orig unless defined($import);
3704 $orig = $self->_rollout_attr($orig);
3705 $import = $self->_rollout_attr($import);
3708 foreach my $import_element ( @{$import} ) {
3709 # find best candidate from $orig to merge $b_element into
3710 my $best_candidate = { position => undef, score => 0 }; my $position = 0;
3711 foreach my $orig_element ( @{$orig} ) {
3712 my $score = $self->_calculate_score( $orig_element, $import_element );
3713 if ($score > $best_candidate->{score}) {
3714 $best_candidate->{position} = $position;
3715 $best_candidate->{score} = $score;
3719 my ($import_key) = ( ref $import_element eq 'HASH' ) ? keys %{$import_element} : ($import_element);
3720 $import_key = '' if not defined $import_key;
3722 if ($best_candidate->{score} == 0 || exists $seen_keys->{$import_key}) {
3723 push( @{$orig}, $import_element );
3725 my $orig_best = $orig->[$best_candidate->{position}];
3726 # merge orig_best and b_element together and replace original with merged
3727 if (ref $orig_best ne 'HASH') {
3728 $orig->[$best_candidate->{position}] = $import_element;
3729 } elsif (ref $import_element eq 'HASH') {
3730 my ($key) = keys %{$orig_best};
3731 $orig->[$best_candidate->{position}] = { $key => $self->_merge_joinpref_attr($orig_best->{$key}, $import_element->{$key}) };
3734 $seen_keys->{$import_key} = 1; # don't merge the same key twice
3737 return @$orig ? $orig : ();
3745 require Hash::Merge;
3746 my $hm = Hash::Merge->new;
3748 $hm->specify_behavior({
3751 my ($defl, $defr) = map { defined $_ } (@_[0,1]);
3753 if ($defl xor $defr) {
3754 return [ $defl ? $_[0] : $_[1] ];
3759 elsif (__HM_DEDUP and $_[0] eq $_[1]) {
3763 return [$_[0], $_[1]];
3767 return $_[1] if !defined $_[0];
3768 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3769 return [$_[0], @{$_[1]}]
3772 return [] if !defined $_[0] and !keys %{$_[1]};
3773 return [ $_[1] ] if !defined $_[0];
3774 return [ $_[0] ] if !keys %{$_[1]};
3775 return [$_[0], $_[1]]
3780 return $_[0] if !defined $_[1];
3781 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3782 return [@{$_[0]}, $_[1]]
3785 my @ret = @{$_[0]} or return $_[1];
3786 return [ @ret, @{$_[1]} ] unless __HM_DEDUP;
3787 my %idx = map { $_ => 1 } @ret;
3788 push @ret, grep { ! defined $idx{$_} } (@{$_[1]});
3792 return [ $_[1] ] if ! @{$_[0]};
3793 return $_[0] if !keys %{$_[1]};
3794 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3795 return [ @{$_[0]}, $_[1] ];
3800 return [] if !keys %{$_[0]} and !defined $_[1];
3801 return [ $_[0] ] if !defined $_[1];
3802 return [ $_[1] ] if !keys %{$_[0]};
3803 return [$_[0], $_[1]]
3806 return [] if !keys %{$_[0]} and !@{$_[1]};
3807 return [ $_[0] ] if !@{$_[1]};
3808 return $_[1] if !keys %{$_[0]};
3809 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3810 return [ $_[0], @{$_[1]} ];
3813 return [] if !keys %{$_[0]} and !keys %{$_[1]};
3814 return [ $_[0] ] if !keys %{$_[1]};
3815 return [ $_[1] ] if !keys %{$_[0]};
3816 return [ $_[0] ] if $_[0] eq $_[1];
3817 return [ $_[0], $_[1] ];
3820 } => 'DBIC_RS_ATTR_MERGER');
3824 return $hm->merge ($_[1], $_[2]);
3828 sub STORABLE_freeze {
3829 my ($self, $cloning) = @_;
3830 my $to_serialize = { %$self };
3832 # A cursor in progress can't be serialized (and would make little sense anyway)
3833 # the parser can be regenerated (and can't be serialized)
3834 delete @{$to_serialize}{qw/cursor _row_parser _result_inflator/};
3836 # nor is it sensical to store a not-yet-fired-count pager
3837 if ($to_serialize->{pager} and ref $to_serialize->{pager}{total_entries} eq 'CODE') {
3838 delete $to_serialize->{pager};
3841 Storable::nfreeze($to_serialize);
3844 # need this hook for symmetry
3846 my ($self, $cloning, $serialized) = @_;
3848 %$self = %{ Storable::thaw($serialized) };
3854 =head2 throw_exception
3856 See L<DBIx::Class::Schema/throw_exception> for details.
3860 sub throw_exception {
3863 if (ref $self and my $rsrc = $self->result_source) {
3864 $rsrc->throw_exception(@_)
3867 DBIx::Class::Exception->throw(@_);
3875 # XXX: FIXME: Attributes docs need clearing up
3879 Attributes are used to refine a ResultSet in various ways when
3880 searching for data. They can be passed to any method which takes an
3881 C<\%attrs> argument. See L</search>, L</search_rs>, L</find>,
3884 Default attributes can be set on the result class using
3885 L<DBIx::Class::ResultSource/resultset_attributes>. (Please read
3886 the CAVEATS on that feature before using it!)
3888 These are in no particular order:
3894 =item Value: ( $order_by | \@order_by | \%order_by )
3898 Which column(s) to order the results by.
3900 [The full list of suitable values is documented in
3901 L<SQL::Abstract/"ORDER BY CLAUSES">; the following is a summary of
3904 If a single column name, or an arrayref of names is supplied, the
3905 argument is passed through directly to SQL. The hashref syntax allows
3906 for connection-agnostic specification of ordering direction:
3908 For descending order:
3910 order_by => { -desc => [qw/col1 col2 col3/] }
3912 For explicit ascending order:
3914 order_by => { -asc => 'col' }
3916 The old scalarref syntax (i.e. order_by => \'year DESC') is still
3917 supported, although you are strongly encouraged to use the hashref
3918 syntax as outlined above.
3924 =item Value: \@columns | \%columns | $column
3928 Shortcut to request a particular set of columns to be retrieved. Each
3929 column spec may be a string (a table column name), or a hash (in which
3930 case the key is the C<as> value, and the value is used as the C<select>
3931 expression). Adds C<me.> onto the start of any column without a C<.> in
3932 it and sets C<select> from that, then auto-populates C<as> from
3933 C<select> as normal. (You may also use the C<cols> attribute, as in
3934 earlier versions of DBIC, but this is deprecated)
3936 Essentially C<columns> does the same as L</select> and L</as>.
3938 columns => [ 'some_column', { dbic_slot => 'another_column' } ]
3942 select => [qw(some_column another_column)],
3943 as => [qw(some_column dbic_slot)]
3945 If you want to individually retrieve related columns (in essence perform
3946 manual prefetch) you have to make sure to specify the correct inflation slot
3947 chain such that it matches existing relationships:
3949 my $rs = $schema->resultset('Artist')->search({}, {
3950 # required to tell DBIC to collapse has_many relationships
3952 join => { cds => 'tracks'},
3954 'cds.cdid' => 'cds.cdid',
3955 'cds.tracks.title' => 'tracks.title',
3961 B<NOTE:> You B<MUST> explicitly quote C<'+columns'> when using this attribute.
3962 Not doing so causes Perl to incorrectly interpret C<+columns> as a bareword
3963 with a unary plus operator before it, which is the same as simply C<columns>.
3967 =item Value: \@extra_columns
3971 Indicates additional columns to be selected from storage. Works the same as
3972 L</columns> but adds columns to the current selection. (You may also use the
3973 C<include_columns> attribute, as in earlier versions of DBIC, but this is
3976 $schema->resultset('CD')->search(undef, {
3977 '+columns' => ['artist.name'],
3981 would return all CDs and include a 'name' column to the information
3982 passed to object inflation. Note that the 'artist' is the name of the
3983 column (or relationship) accessor, and 'name' is the name of the column
3984 accessor in the related table.
3990 =item Value: \@select_columns
3994 Indicates which columns should be selected from the storage. You can use
3995 column names, or in the case of RDBMS back ends, function or stored procedure
3998 $rs = $schema->resultset('Employee')->search(undef, {
4001 { count => 'employeeid' },
4002 { max => { length => 'name' }, -as => 'longest_name' }
4007 SELECT name, COUNT( employeeid ), MAX( LENGTH( name ) ) AS longest_name FROM employee
4009 B<NOTE:> You will almost always need a corresponding L</as> attribute when you
4010 use L</select>, to instruct DBIx::Class how to store the result of the column.
4011 Also note that the L</as> attribute has nothing to do with the SQL-side 'AS'
4012 identifier aliasing. You can however alias a function, so you can use it in
4013 e.g. an C<ORDER BY> clause. This is done via the C<-as> B<select function
4014 attribute> supplied as shown in the example above.
4018 B<NOTE:> You B<MUST> explicitly quote C<'+select'> when using this attribute.
4019 Not doing so causes Perl to incorrectly interpret C<+select> as a bareword
4020 with a unary plus operator before it, which is the same as simply C<select>.
4024 =item Value: \@extra_select_columns
4028 Indicates additional columns to be selected from storage. Works the same as
4029 L</select> but adds columns to the current selection, instead of specifying
4030 a new explicit list.
4036 =item Value: \@inflation_names
4040 Indicates DBIC-side names for object inflation. That is L</as> indicates the
4041 slot name in which the column value will be stored within the
4042 L<Row|DBIx::Class::Row> object. The value will then be accessible via this
4043 identifier by the C<get_column> method (or via the object accessor B<if one
4044 with the same name already exists>) as shown below. The L</as> attribute has
4045 B<nothing to do> with the SQL-side C<AS>. See L</select> for details.
4047 $rs = $schema->resultset('Employee')->search(undef, {
4050 { count => 'employeeid' },
4051 { max => { length => 'name' }, -as => 'longest_name' }
4060 If the object against which the search is performed already has an accessor
4061 matching a column name specified in C<as>, the value can be retrieved using
4062 the accessor as normal:
4064 my $name = $employee->name();
4066 If on the other hand an accessor does not exist in the object, you need to
4067 use C<get_column> instead:
4069 my $employee_count = $employee->get_column('employee_count');
4071 You can create your own accessors if required - see
4072 L<DBIx::Class::Manual::Cookbook> for details.
4076 B<NOTE:> You B<MUST> explicitly quote C<'+as'> when using this attribute.
4077 Not doing so causes Perl to incorrectly interpret C<+as> as a bareword
4078 with a unary plus operator before it, which is the same as simply C<as>.
4082 =item Value: \@extra_inflation_names
4086 Indicates additional inflation names for selectors added via L</+select>. See L</as>.
4092 =item Value: ($rel_name | \@rel_names | \%rel_names)
4096 Contains a list of relationships that should be joined for this query. For
4099 # Get CDs by Nine Inch Nails
4100 my $rs = $schema->resultset('CD')->search(
4101 { 'artist.name' => 'Nine Inch Nails' },
4102 { join => 'artist' }
4105 Can also contain a hash reference to refer to the other relation's relations.
4108 package MyApp::Schema::Track;
4109 use base qw/DBIx::Class/;
4110 __PACKAGE__->table('track');
4111 __PACKAGE__->add_columns(qw/trackid cd position title/);
4112 __PACKAGE__->set_primary_key('trackid');
4113 __PACKAGE__->belongs_to(cd => 'MyApp::Schema::CD');
4116 # In your application
4117 my $rs = $schema->resultset('Artist')->search(
4118 { 'track.title' => 'Teardrop' },
4120 join => { cd => 'track' },
4121 order_by => 'artist.name',
4125 You need to use the relationship (not the table) name in conditions,
4126 because they are aliased as such. The current table is aliased as "me", so
4127 you need to use me.column_name in order to avoid ambiguity. For example:
4129 # Get CDs from 1984 with a 'Foo' track
4130 my $rs = $schema->resultset('CD')->search(
4133 'tracks.name' => 'Foo'
4135 { join => 'tracks' }
4138 If the same join is supplied twice, it will be aliased to <rel>_2 (and
4139 similarly for a third time). For e.g.
4141 my $rs = $schema->resultset('Artist')->search({
4142 'cds.title' => 'Down to Earth',
4143 'cds_2.title' => 'Popular',
4145 join => [ qw/cds cds/ ],
4148 will return a set of all artists that have both a cd with title 'Down
4149 to Earth' and a cd with title 'Popular'.
4151 If you want to fetch related objects from other tables as well, see L</prefetch>
4154 NOTE: An internal join-chain pruner will discard certain joins while
4155 constructing the actual SQL query, as long as the joins in question do not
4156 affect the retrieved result. This for example includes 1:1 left joins
4157 that are not part of the restriction specification (WHERE/HAVING) nor are
4158 a part of the query selection.
4160 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
4166 =item Value: (0 | 1)
4170 When set to a true value, indicates that any rows fetched from joined has_many
4171 relationships are to be aggregated into the corresponding "parent" object. For
4172 example, the resultset:
4174 my $rs = $schema->resultset('CD')->search({}, {
4175 '+columns' => [ qw/ tracks.title tracks.position / ],
4180 While executing the following query:
4182 SELECT me.*, tracks.title, tracks.position
4184 LEFT JOIN track tracks
4185 ON tracks.cdid = me.cdid
4187 Will return only as many objects as there are rows in the CD source, even
4188 though the result of the query may span many rows. Each of these CD objects
4189 will in turn have multiple "Track" objects hidden behind the has_many
4190 generated accessor C<tracks>. Without C<< collapse => 1 >>, the return values
4191 of this resultset would be as many CD objects as there are tracks (a "Cartesian
4192 product"), with each CD object containing exactly one of all fetched Track data.
4194 When a collapse is requested on a non-ordered resultset, an order by some
4195 unique part of the main source (the left-most table) is inserted automatically.
4196 This is done so that the resultset is allowed to be "lazy" - calling
4197 L<< $rs->next|/next >> will fetch only as many rows as it needs to build the next
4198 object with all of its related data.
4200 If an L</order_by> is already declared, and orders the resultset in a way that
4201 makes collapsing as described above impossible (e.g. C<< ORDER BY
4202 has_many_rel.column >> or C<ORDER BY RANDOM()>), DBIC will automatically
4203 switch to "eager" mode and slurp the entire resultset before constructing the
4204 first object returned by L</next>.
4206 Setting this attribute on a resultset that does not join any has_many
4207 relations is a no-op.
4209 For a more in-depth discussion, see L</PREFETCHING>.
4215 =item Value: ($rel_name | \@rel_names | \%rel_names)
4219 This attribute is a shorthand for specifying a L</join> spec, adding all
4220 columns from the joined related sources as L</+columns> and setting
4221 L</collapse> to a true value. For example, the following two queries are
4224 my $rs = $schema->resultset('Artist')->search({}, {
4225 prefetch => { cds => ['genre', 'tracks' ] },
4230 my $rs = $schema->resultset('Artist')->search({}, {
4231 join => { cds => ['genre', 'tracks' ] },
4235 { +{ "cds.$_" => "cds.$_" } }
4236 $schema->source('Artist')->related_source('cds')->columns
4239 { +{ "cds.genre.$_" => "genre.$_" } }
4240 $schema->source('Artist')->related_source('cds')->related_source('genre')->columns
4243 { +{ "cds.tracks.$_" => "tracks.$_" } }
4244 $schema->source('Artist')->related_source('cds')->related_source('tracks')->columns
4249 Both producing the following SQL:
4251 SELECT me.artistid, me.name, me.rank, me.charfield,
4252 cds.cdid, cds.artist, cds.title, cds.year, cds.genreid, cds.single_track,
4253 genre.genreid, genre.name,
4254 tracks.trackid, tracks.cd, tracks.position, tracks.title, tracks.last_updated_on, tracks.last_updated_at
4257 ON cds.artist = me.artistid
4258 LEFT JOIN genre genre
4259 ON genre.genreid = cds.genreid
4260 LEFT JOIN track tracks
4261 ON tracks.cd = cds.cdid
4262 ORDER BY me.artistid
4264 While L</prefetch> implies a L</join>, it is ok to mix the two together, as
4265 the arguments are properly merged and generally do the right thing. For
4266 example, you may want to do the following:
4268 my $artists_and_cds_without_genre = $schema->resultset('Artist')->search(
4269 { 'genre.genreid' => undef },
4271 join => { cds => 'genre' },
4276 Which generates the following SQL:
4278 SELECT me.artistid, me.name, me.rank, me.charfield,
4279 cds.cdid, cds.artist, cds.title, cds.year, cds.genreid, cds.single_track
4282 ON cds.artist = me.artistid
4283 LEFT JOIN genre genre
4284 ON genre.genreid = cds.genreid
4285 WHERE genre.genreid IS NULL
4286 ORDER BY me.artistid
4288 For a more in-depth discussion, see L</PREFETCHING>.
4294 =item Value: $source_alias
4298 Sets the source alias for the query. Normally, this defaults to C<me>, but
4299 nested search queries (sub-SELECTs) might need specific aliases set to
4300 reference inner queries. For example:
4303 ->related_resultset('CDs')
4304 ->related_resultset('Tracks')
4306 'track.id' => { -ident => 'none_search.id' },
4310 my $ids = $self->search({
4313 alias => 'none_search',
4314 group_by => 'none_search.id',
4315 })->get_column('id')->as_query;
4317 $self->search({ id => { -in => $ids } })
4319 This attribute is directly tied to L</current_source_alias>.
4329 Makes the resultset paged and specifies the page to retrieve. Effectively
4330 identical to creating a non-pages resultset and then calling ->page($page)
4333 If L</rows> attribute is not specified it defaults to 10 rows per page.
4335 When you have a paged resultset, L</count> will only return the number
4336 of rows in the page. To get the total, use the L</pager> and call
4337 C<total_entries> on it.
4347 Specifies the maximum number of rows for direct retrieval or the number of
4348 rows per page if the page attribute or method is used.
4354 =item Value: $offset
4358 Specifies the (zero-based) row number for the first row to be returned, or the
4359 of the first row of the first page if paging is used.
4361 =head2 software_limit
4365 =item Value: (0 | 1)
4369 When combined with L</rows> and/or L</offset> the generated SQL will not
4370 include any limit dialect stanzas. Instead the entire result will be selected
4371 as if no limits were specified, and DBIC will perform the limit locally, by
4372 artificially advancing and finishing the resulting L</cursor>.
4374 This is the recommended way of performing resultset limiting when no sane RDBMS
4375 implementation is available (e.g.
4376 L<Sybase ASE|DBIx::Class::Storage::DBI::Sybase::ASE> using the
4377 L<Generic Sub Query|DBIx::Class::SQLMaker::LimitDialects/GenericSubQ> hack)
4383 =item Value: \@columns
4387 A arrayref of columns to group by. Can include columns of joined tables.
4389 group_by => [qw/ column1 column2 ... /]
4395 =item Value: $condition
4399 HAVING is a select statement attribute that is applied between GROUP BY and
4400 ORDER BY. It is applied to the after the grouping calculations have been
4403 having => { 'count_employee' => { '>=', 100 } }
4405 or with an in-place function in which case literal SQL is required:
4407 having => \[ 'count(employee) >= ?', [ count => 100 ] ]
4413 =item Value: (0 | 1)
4417 Set to 1 to automatically generate a L</group_by> clause based on the selection
4418 (including intelligent handling of L</order_by> contents). Note that the group
4419 criteria calculation takes place over the B<final> selection. This includes
4420 any L</+columns>, L</+select> or L</order_by> additions in subsequent
4421 L</search> calls, and standalone columns selected via
4422 L<DBIx::Class::ResultSetColumn> (L</get_column>). A notable exception are the
4423 extra selections specified via L</prefetch> - such selections are explicitly
4424 excluded from group criteria calculations.
4426 If the final ResultSet also explicitly defines a L</group_by> attribute, this
4427 setting is ignored and an appropriate warning is issued.
4433 Adds to the WHERE clause.
4435 # only return rows WHERE deleted IS NULL for all searches
4436 __PACKAGE__->resultset_attributes({ where => { deleted => undef } });
4438 Can be overridden by passing C<< { where => undef } >> as an attribute
4441 For more complicated where clauses see L<SQL::Abstract/WHERE CLAUSES>.
4447 Set to 1 to cache search results. This prevents extra SQL queries if you
4448 revisit rows in your ResultSet:
4450 my $resultset = $schema->resultset('Artist')->search( undef, { cache => 1 } );
4452 while( my $artist = $resultset->next ) {
4456 $rs->first; # without cache, this would issue a query
4458 By default, searches are not cached.
4460 For more examples of using these attributes, see
4461 L<DBIx::Class::Manual::Cookbook>.
4467 =item Value: ( 'update' | 'shared' | \$scalar )
4471 Set to 'update' for a SELECT ... FOR UPDATE or 'shared' for a SELECT
4472 ... FOR SHARED. If \$scalar is passed, this is taken directly and embedded in the
4477 DBIx::Class supports arbitrary related data prefetching from multiple related
4478 sources. Any combination of relationship types and column sets are supported.
4479 If L<collapsing|/collapse> is requested, there is an additional requirement of
4480 selecting enough data to make every individual object uniquely identifiable.
4482 Here are some more involved examples, based on the following relationship map:
4485 My::Schema::CD->belongs_to( artist => 'My::Schema::Artist' );
4486 My::Schema::CD->might_have( liner_note => 'My::Schema::LinerNotes' );
4487 My::Schema::CD->has_many( tracks => 'My::Schema::Track' );
4489 My::Schema::Artist->belongs_to( record_label => 'My::Schema::RecordLabel' );
4491 My::Schema::Track->has_many( guests => 'My::Schema::Guest' );
4495 my $rs = $schema->resultset('Tag')->search(
4504 The initial search results in SQL like the following:
4506 SELECT tag.*, cd.*, artist.* FROM tag
4507 JOIN cd ON tag.cd = cd.cdid
4508 JOIN artist ON cd.artist = artist.artistid
4510 L<DBIx::Class> has no need to go back to the database when we access the
4511 C<cd> or C<artist> relationships, which saves us two SQL statements in this
4514 Simple prefetches will be joined automatically, so there is no need
4515 for a C<join> attribute in the above search.
4517 The L</prefetch> attribute can be used with any of the relationship types
4518 and multiple prefetches can be specified together. Below is a more complex
4519 example that prefetches a CD's artist, its liner notes (if present),
4520 the cover image, the tracks on that CD, and the guests on those
4523 my $rs = $schema->resultset('CD')->search(
4527 { artist => 'record_label'}, # belongs_to => belongs_to
4528 'liner_note', # might_have
4529 'cover_image', # has_one
4530 { tracks => 'guests' }, # has_many => has_many
4535 This will produce SQL like the following:
4537 SELECT cd.*, artist.*, record_label.*, liner_note.*, cover_image.*,
4541 ON artist.artistid = me.artistid
4542 JOIN record_label record_label
4543 ON record_label.labelid = artist.labelid
4544 LEFT JOIN track tracks
4545 ON tracks.cdid = me.cdid
4546 LEFT JOIN guest guests
4547 ON guests.trackid = track.trackid
4548 LEFT JOIN liner_notes liner_note
4549 ON liner_note.cdid = me.cdid
4550 JOIN cd_artwork cover_image
4551 ON cover_image.cdid = me.cdid
4554 Now the C<artist>, C<record_label>, C<liner_note>, C<cover_image>,
4555 C<tracks>, and C<guests> of the CD will all be available through the
4556 relationship accessors without the need for additional queries to the
4561 Prefetch does a lot of deep magic. As such, it may not behave exactly
4562 as you might expect.
4568 Prefetch uses the L</cache> to populate the prefetched relationships. This
4569 may or may not be what you want.
4573 If you specify a condition on a prefetched relationship, ONLY those
4574 rows that match the prefetched condition will be fetched into that relationship.
4575 This means that adding prefetch to a search() B<may alter> what is returned by
4576 traversing a relationship. So, if you have C<< Artist->has_many(CDs) >> and you do
4578 my $artist_rs = $schema->resultset('Artist')->search({
4584 my $count = $artist_rs->first->cds->count;
4586 my $artist_rs_prefetch = $artist_rs->search( {}, { prefetch => 'cds' } );
4588 my $prefetch_count = $artist_rs_prefetch->first->cds->count;
4590 cmp_ok( $count, '==', $prefetch_count, "Counts should be the same" );
4592 That cmp_ok() may or may not pass depending on the datasets involved. In other
4593 words the C<WHERE> condition would apply to the entire dataset, just like
4594 it would in regular SQL. If you want to add a condition only to the "right side"
4595 of a C<LEFT JOIN> - consider declaring and using a L<relationship with a custom
4596 condition|DBIx::Class::Relationship::Base/condition>
4600 =head1 DBIC BIND VALUES
4602 Because DBIC may need more information to bind values than just the column name
4603 and value itself, it uses a special format for both passing and receiving bind
4604 values. Each bind value should be composed of an arrayref of
4605 C<< [ \%args => $val ] >>. The format of C<< \%args >> is currently:
4611 If present (in any form), this is what is being passed directly to bind_param.
4612 Note that different DBD's expect different bind args. (e.g. DBD::SQLite takes
4613 a single numerical type, while DBD::Pg takes a hashref if bind options.)
4615 If this is specified, all other bind options described below are ignored.
4619 If present, this is used to infer the actual bind attribute by passing to
4620 C<< $resolved_storage->bind_attribute_by_data_type() >>. Defaults to the
4621 "data_type" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>.
4623 Note that the data type is somewhat freeform (hence the sqlt_ prefix);
4624 currently drivers are expected to "Do the Right Thing" when given a common
4625 datatype name. (Not ideal, but that's what we got at this point.)
4629 Currently used to correctly allocate buffers for bind_param_inout().
4630 Defaults to "size" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>,
4631 or to a sensible value based on the "data_type".
4635 Used to fill in missing sqlt_datatype and sqlt_size attributes (if they are
4636 explicitly specified they are never overridden). Also used by some weird DBDs,
4637 where the column name should be available at bind_param time (e.g. Oracle).
4641 For backwards compatibility and convenience, the following shortcuts are
4644 [ $name => $val ] === [ { dbic_colname => $name }, $val ]
4645 [ \$dt => $val ] === [ { sqlt_datatype => $dt }, $val ]
4646 [ undef, $val ] === [ {}, $val ]
4647 $val === [ {}, $val ]
4649 =head1 AUTHOR AND CONTRIBUTORS
4651 See L<AUTHOR|DBIx::Class/AUTHOR> and L<CONTRIBUTORS|DBIx::Class/CONTRIBUTORS> in DBIx::Class
4655 You may distribute this code under the same terms as Perl itself.