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 => $days_ago )
202 sub users_to_warn { shift->active->unverified->created_n_days_ago(7) }
206 See L<DBIx::Class::Schema/load_namespaces> on how DBIC can discover and
207 automatically attach L<Result|DBIx::Class::Manual::ResultClass>-specific
208 L<ResulSet|DBIx::Class::ResultSet> classes.
210 =head3 ResultSet subclassing with Moose and similar constructor-providers
212 Using L<Moose> or L<Moo> in your ResultSet classes is usually overkill, but
213 you may find it useful if your ResultSets contain a lot of business logic
214 (e.g. C<has xml_parser>, C<has json>, etc) or if you just prefer to organize
217 In order to write custom ResultSet classes with L<Moo> you need to use the
218 following template. The L<BUILDARGS|Moo/BUILDARGS> is necessary due to the
219 unusual signature of the L<constructor provided by DBIC
220 |DBIx::Class::ResultSet/new> C<< ->new($source, \%args) >>.
223 extends 'DBIx::Class::ResultSet';
224 sub BUILDARGS { $_[2] } # ::RS::new() expects my ($class, $rsrc, $args) = @_
230 If you want to build your custom ResultSet classes with L<Moose>, you need
231 a similar, though a little more elaborate template in order to interface the
232 inlining of the L<Moose>-provided
233 L<object constructor|Moose::Manual::Construction/WHERE'S THE CONSTRUCTOR?>,
236 package MyApp::Schema::ResultSet::User;
239 use MooseX::NonMoose;
240 extends 'DBIx::Class::ResultSet';
242 sub BUILDARGS { $_[2] } # ::RS::new() expects my ($class, $rsrc, $args) = @_
246 __PACKAGE__->meta->make_immutable;
250 The L<MooseX::NonMoose> is necessary so that the L<Moose> constructor does not
251 entirely overwrite the DBIC one (in contrast L<Moo> does this automatically).
252 Alternatively, you can skip L<MooseX::NonMoose> and get by with just L<Moose>
255 __PACKAGE__->meta->make_immutable(inline_constructor => 0);
263 =item Arguments: L<$source|DBIx::Class::ResultSource>, L<\%attrs?|/ATTRIBUTES>
265 =item Return Value: L<$resultset|/search>
269 The resultset constructor. Takes a source object (usually a
270 L<DBIx::Class::ResultSourceProxy::Table>) and an attribute hash (see
271 L</ATTRIBUTES> below). Does not perform any queries -- these are
272 executed as needed by the other methods.
274 Generally you never construct a resultset manually. Instead you get one
276 C<< $schema->L<resultset|DBIx::Class::Schema/resultset>('$source_name') >>
277 or C<< $another_resultset->L<search|/search>(...) >> (the later called in
280 my $rs = $schema->resultset('CD')->search({ title => '100th Window' });
286 If called on an object, proxies to L</new_result> instead, so
288 my $cd = $schema->resultset('CD')->new({ title => 'Spoon' });
290 will return a CD object, not a ResultSet, and is equivalent to:
292 my $cd = $schema->resultset('CD')->new_result({ title => 'Spoon' });
294 Please also keep in mind that many internals call L</new_result> directly,
295 so overloading this method with the idea of intercepting new result object
296 creation B<will not work>. See also warning pertaining to L</create>.
304 return $class->new_result(@_) if ref $class;
306 my ($source, $attrs) = @_;
307 $source = $source->resolve
308 if $source->isa('DBIx::Class::ResultSourceHandle');
310 $attrs = { %{$attrs||{}} };
311 delete @{$attrs}{qw(_last_sqlmaker_alias_map _related_results_construction)};
313 if ($attrs->{page}) {
314 $attrs->{rows} ||= 10;
317 $attrs->{alias} ||= 'me';
320 result_source => $source,
321 cond => $attrs->{where},
326 # if there is a dark selector, this means we are already in a
327 # chain and the cleanup/sanification was taken care of by
329 $self->_normalize_selection($attrs)
330 unless $attrs->{_dark_selector};
333 $attrs->{result_class} || $source->result_class
343 =item Arguments: L<$cond|DBIx::Class::SQLMaker> | undef, L<\%attrs?|/ATTRIBUTES>
345 =item Return Value: $resultset (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
349 my @cds = $cd_rs->search({ year => 2001 }); # "... WHERE year = 2001"
350 my $new_rs = $cd_rs->search({ year => 2005 });
352 my $new_rs = $cd_rs->search([ { year => 2005 }, { year => 2004 } ]);
353 # year = 2005 OR year = 2004
355 In list context, C<< ->all() >> is called implicitly on the resultset, thus
356 returning a list of L<result|DBIx::Class::Manual::ResultClass> objects instead.
357 To avoid that, use L</search_rs>.
359 If you need to pass in additional attributes but no additional condition,
360 call it as C<search(undef, \%attrs)>.
362 # "SELECT name, artistid FROM $artist_table"
363 my @all_artists = $schema->resultset('Artist')->search(undef, {
364 columns => [qw/name artistid/],
367 For a list of attributes that can be passed to C<search>, see
368 L</ATTRIBUTES>. For more examples of using this function, see
369 L<Searching|DBIx::Class::Manual::Cookbook/SEARCHING>. For a complete
370 documentation for the first argument, see L<SQL::Abstract/"WHERE CLAUSES">
371 and its extension L<DBIx::Class::SQLMaker>.
373 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
377 Note that L</search> does not process/deflate any of the values passed in the
378 L<SQL::Abstract>-compatible search condition structure. This is unlike other
379 condition-bound methods L</new_result>, L</create> and L</find>. The user must ensure
380 manually that any value passed to this method will stringify to something the
381 RDBMS knows how to deal with. A notable example is the handling of L<DateTime>
382 objects, for more info see:
383 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
389 my $rs = $self->search_rs( @_ );
392 DBIx::Class::_ENV_::ASSERT_NO_INTERNAL_WANTARRAY and my $sog = fail_on_internal_wantarray($rs);
395 elsif (defined wantarray) {
399 # we can be called by a relationship helper, which in
400 # turn may be called in void context due to some braindead
401 # overload or whatever else the user decided to be clever
402 # at this particular day. Thus limit the exception to
403 # external code calls only
404 $self->throw_exception ('->search is *not* a mutator, calling it in void context makes no sense')
405 if (caller)[0] !~ /^\QDBIx::Class::/;
415 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
417 =item Return Value: L<$resultset|/search>
421 This method does the same exact thing as search() except it will
422 always return a resultset, even in list context.
429 my $rsrc = $self->result_source;
430 my ($call_cond, $call_attrs);
432 # Special-case handling for (undef, undef) or (undef)
433 # Note that (foo => undef) is valid deprecated syntax
434 @_ = () if not scalar grep { defined $_ } @_;
440 # fish out attrs in the ($condref, $attr) case
441 elsif (@_ == 2 and ( ! defined $_[0] or (ref $_[0]) ne '') ) {
442 ($call_cond, $call_attrs) = @_;
445 $self->throw_exception('Odd number of arguments to search')
449 carp_unique 'search( %condition ) is deprecated, use search( \%condition ) instead'
450 unless $rsrc->result_class->isa('DBIx::Class::CDBICompat');
452 for my $i (0 .. $#_) {
454 $self->throw_exception ('All keys in condition key/value pairs must be plain scalars')
455 if (! defined $_[$i] or ref $_[$i] ne '');
461 # see if we can keep the cache (no $rs changes)
463 my %safe = (alias => 1, cache => 1);
464 if ( ! List::Util::first { !$safe{$_} } keys %$call_attrs and (
467 ref $call_cond eq 'HASH' && ! keys %$call_cond
469 ref $call_cond eq 'ARRAY' && ! @$call_cond
471 $cache = $self->get_cache;
474 my $old_attrs = { %{$self->{attrs}} };
475 my ($old_having, $old_where) = delete @{$old_attrs}{qw(having where)};
477 my $new_attrs = { %$old_attrs };
479 # take care of call attrs (only if anything is changing)
480 if ($call_attrs and keys %$call_attrs) {
482 # copy for _normalize_selection
483 $call_attrs = { %$call_attrs };
485 my @selector_attrs = qw/select as columns cols +select +as +columns include_columns/;
487 # reset the current selector list if new selectors are supplied
488 if (List::Util::first { exists $call_attrs->{$_} } qw/columns cols select as/) {
489 delete @{$old_attrs}{(@selector_attrs, '_dark_selector')};
492 # Normalize the new selector list (operates on the passed-in attr structure)
493 # Need to do it on every chain instead of only once on _resolved_attrs, in
494 # order to allow detection of empty vs partial 'as'
495 $call_attrs->{_dark_selector} = $old_attrs->{_dark_selector}
496 if $old_attrs->{_dark_selector};
497 $self->_normalize_selection ($call_attrs);
499 # start with blind overwriting merge, exclude selector attrs
500 $new_attrs = { %{$old_attrs}, %{$call_attrs} };
501 delete @{$new_attrs}{@selector_attrs};
503 for (@selector_attrs) {
504 $new_attrs->{$_} = $self->_merge_attr($old_attrs->{$_}, $call_attrs->{$_})
505 if ( exists $old_attrs->{$_} or exists $call_attrs->{$_} );
508 # older deprecated name, use only if {columns} is not there
509 if (my $c = delete $new_attrs->{cols}) {
510 carp_unique( "Resultset attribute 'cols' is deprecated, use 'columns' instead" );
511 if ($new_attrs->{columns}) {
512 carp "Resultset specifies both the 'columns' and the legacy 'cols' attributes - ignoring 'cols'";
515 $new_attrs->{columns} = $c;
520 # join/prefetch use their own crazy merging heuristics
521 foreach my $key (qw/join prefetch/) {
522 $new_attrs->{$key} = $self->_merge_joinpref_attr($old_attrs->{$key}, $call_attrs->{$key})
523 if exists $call_attrs->{$key};
526 # stack binds together
527 $new_attrs->{bind} = [ @{ $old_attrs->{bind} || [] }, @{ $call_attrs->{bind} || [] } ];
531 for ($old_where, $call_cond) {
533 $new_attrs->{where} = $self->_stack_cond (
534 $_, $new_attrs->{where}
539 if (defined $old_having) {
540 $new_attrs->{having} = $self->_stack_cond (
541 $old_having, $new_attrs->{having}
545 my $rs = (ref $self)->new($rsrc, $new_attrs);
547 $rs->set_cache($cache) if ($cache);
553 sub _normalize_selection {
554 my ($self, $attrs) = @_;
557 if ( exists $attrs->{include_columns} ) {
558 carp_unique( "Resultset attribute 'include_columns' is deprecated, use '+columns' instead" );
559 $attrs->{'+columns'} = $self->_merge_attr(
560 $attrs->{'+columns'}, delete $attrs->{include_columns}
564 # columns are always placed first, however
566 # Keep the X vs +X separation until _resolved_attrs time - this allows to
567 # delay the decision on whether to use a default select list ($rsrc->columns)
568 # allowing stuff like the remove_columns helper to work
570 # select/as +select/+as pairs need special handling - the amount of select/as
571 # elements in each pair does *not* have to be equal (think multicolumn
572 # selectors like distinct(foo, bar) ). If the selector is bare (no 'as'
573 # supplied at all) - try to infer the alias, either from the -as parameter
574 # of the selector spec, or use the parameter whole if it looks like a column
575 # name (ugly legacy heuristic). If all fails - leave the selector bare (which
576 # is ok as well), but make sure no more additions to the 'as' chain take place
577 for my $pref ('', '+') {
579 my ($sel, $as) = map {
580 my $key = "${pref}${_}";
582 my $val = [ ref $attrs->{$key} eq 'ARRAY'
584 : $attrs->{$key} || ()
586 delete $attrs->{$key};
590 if (! @$as and ! @$sel ) {
593 elsif (@$as and ! @$sel) {
594 $self->throw_exception(
595 "Unable to handle ${pref}as specification (@$as) without a corresponding ${pref}select"
599 # no as part supplied at all - try to deduce (unless explicit end of named selection is declared)
600 # if any @$as has been supplied we assume the user knows what (s)he is doing
601 # and blindly keep stacking up pieces
602 unless ($attrs->{_dark_selector}) {
605 if ( ref $_ eq 'HASH' and exists $_->{-as} ) {
606 push @$as, $_->{-as};
608 # assume any plain no-space, no-parenthesis string to be a column spec
609 # FIXME - this is retarded but is necessary to support shit like 'count(foo)'
610 elsif ( ! ref $_ and $_ =~ /^ [^\s\(\)]+ $/x) {
613 # if all else fails - raise a flag that no more aliasing will be allowed
615 $attrs->{_dark_selector} = {
617 string => ($dark_sel_dumper ||= do {
618 require Data::Dumper::Concise;
619 Data::Dumper::Concise::DumperObject()->Indent(0);
620 })->Values([$_])->Dump
628 elsif (@$as < @$sel) {
629 $self->throw_exception(
630 "Unable to handle an ${pref}as specification (@$as) with less elements than the corresponding ${pref}select"
633 elsif ($pref and $attrs->{_dark_selector}) {
634 $self->throw_exception(
635 "Unable to process named '+select', resultset contains an unnamed selector $attrs->{_dark_selector}{string}"
641 $attrs->{"${pref}select"} = $self->_merge_attr($attrs->{"${pref}select"}, $sel);
642 $attrs->{"${pref}as"} = $self->_merge_attr($attrs->{"${pref}as"}, $as);
647 my ($self, $left, $right) = @_;
650 (ref $_ eq 'ARRAY' and !@$_)
652 (ref $_ eq 'HASH' and ! keys %$_)
653 ) and $_ = undef for ($left, $right);
655 # either on of the two undef or both undef
656 if ( ( (defined $left) xor (defined $right) ) or ! defined $left ) {
657 return defined $left ? $left : $right;
660 my $cond = $self->result_source->schema->storage->_collapse_cond({ -and => [$left, $right] });
662 for my $c (grep { ref $cond->{$_} eq 'ARRAY' and ($cond->{$_}[0]||'') eq '-and' } keys %$cond) {
664 my @vals = sort @{$cond->{$c}}[ 1..$#{$cond->{$c}} ];
665 my @fin = shift @vals;
668 push @fin, $v unless Data::Compare::Compare( $fin[-1], $v );
671 $cond->{$c} = (@fin == 1) ? $fin[0] : [-and => @fin ];
677 =head2 search_literal
679 B<CAVEAT>: C<search_literal> is provided for Class::DBI compatibility and
680 should only be used in that context. C<search_literal> is a convenience
681 method. It is equivalent to calling C<< $schema->search(\[]) >>, but if you
682 want to ensure columns are bound correctly, use L</search>.
684 See L<DBIx::Class::Manual::Cookbook/SEARCHING> and
685 L<DBIx::Class::Manual::FAQ/Searching> for searching techniques that do not
686 require C<search_literal>.
690 =item Arguments: $sql_fragment, @standalone_bind_values
692 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
696 my @cds = $cd_rs->search_literal('year = ? AND title = ?', qw/2001 Reload/);
697 my $newrs = $artist_rs->search_literal('name = ?', 'Metallica');
699 Pass a literal chunk of SQL to be added to the conditional part of the
702 Example of how to use C<search> instead of C<search_literal>
704 my @cds = $cd_rs->search_literal('cdid = ? AND (artist = ? OR artist = ?)', (2, 1, 2));
705 my @cds = $cd_rs->search(\[ 'cdid = ? AND (artist = ? OR artist = ?)', [ 'cdid', 2 ], [ 'artist', 1 ], [ 'artist', 2 ] ]);
710 my ($self, $sql, @bind) = @_;
712 if ( @bind && ref($bind[-1]) eq 'HASH' ) {
715 return $self->search(\[ $sql, map [ {} => $_ ], @bind ], ($attr || () ));
722 =item Arguments: \%columns_values | @pk_values, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
724 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
728 Finds and returns a single row based on supplied criteria. Takes either a
729 hashref with the same format as L</create> (including inference of foreign
730 keys from related objects), or a list of primary key values in the same
731 order as the L<primary columns|DBIx::Class::ResultSource/primary_columns>
732 declaration on the L</result_source>.
734 In either case an attempt is made to combine conditions already existing on
735 the resultset with the condition passed to this method.
737 To aid with preparing the correct query for the storage you may supply the
738 C<key> attribute, which is the name of a
739 L<unique constraint|DBIx::Class::ResultSource/add_unique_constraint> (the
740 unique constraint corresponding to the
741 L<primary columns|DBIx::Class::ResultSource/primary_columns> is always named
742 C<primary>). If the C<key> attribute has been supplied, and DBIC is unable
743 to construct a query that satisfies the named unique constraint fully (
744 non-NULL values for each column member of the constraint) an exception is
747 If no C<key> is specified, the search is carried over all unique constraints
748 which are fully defined by the available condition.
750 If no such constraint is found, C<find> currently defaults to a simple
751 C<< search->(\%column_values) >> which may or may not do what you expect.
752 Note that this fallback behavior may be deprecated in further versions. If
753 you need to search with arbitrary conditions - use L</search>. If the query
754 resulting from this fallback produces more than one row, a warning to the
755 effect is issued, though only the first row is constructed and returned as
758 In addition to C<key>, L</find> recognizes and applies standard
759 L<resultset attributes|/ATTRIBUTES> in the same way as L</search> does.
761 Note that if you have extra concerns about the correctness of the resulting
762 query you need to specify the C<key> attribute and supply the entire condition
763 as an argument to find (since it is not always possible to perform the
764 combination of the resultset condition with the supplied one, especially if
765 the resultset condition contains literal sql).
767 For example, to find a row by its primary key:
769 my $cd = $schema->resultset('CD')->find(5);
771 You can also find a row by a specific unique constraint:
773 my $cd = $schema->resultset('CD')->find(
775 artist => 'Massive Attack',
776 title => 'Mezzanine',
778 { key => 'cd_artist_title' }
781 See also L</find_or_create> and L</update_or_create>.
787 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
789 my $rsrc = $self->result_source;
792 if (exists $attrs->{key}) {
793 $constraint_name = defined $attrs->{key}
795 : $self->throw_exception("An undefined 'key' resultset attribute makes no sense")
799 # Parse out the condition from input
802 if (ref $_[0] eq 'HASH') {
803 $call_cond = { %{$_[0]} };
806 # if only values are supplied we need to default to 'primary'
807 $constraint_name = 'primary' unless defined $constraint_name;
809 my @c_cols = $rsrc->unique_constraint_columns($constraint_name);
811 $self->throw_exception(
812 "No constraint columns, maybe a malformed '$constraint_name' constraint?"
815 $self->throw_exception (
816 'find() expects either a column/value hashref, or a list of values '
817 . "corresponding to the columns of the specified unique constraint '$constraint_name'"
818 ) unless @c_cols == @_;
821 @{$call_cond}{@c_cols} = @_;
825 for my $key (keys %$call_cond) {
827 my $keyref = ref($call_cond->{$key})
829 my $relinfo = $rsrc->relationship_info($key)
831 my $val = delete $call_cond->{$key};
833 next if $keyref eq 'ARRAY'; # has_many for multi_create
835 my ($rel_cond, $crosstable) = $rsrc->_resolve_condition(
836 $relinfo->{cond}, $val, $key, $key
839 $self->throw_exception("Complex condition via relationship '$key' is unsupported in find()")
840 if $crosstable or ref($rel_cond) ne 'HASH';
843 @related{keys %$rel_cond} = values %$rel_cond;
847 # relationship conditions take precedence (?)
848 @{$call_cond}{keys %related} = values %related;
850 my $alias = exists $attrs->{alias} ? $attrs->{alias} : $self->{attrs}{alias};
852 if (defined $constraint_name) {
853 $final_cond = $self->_qualify_cond_columns (
855 $self->_build_unique_cond (
863 elsif ($self->{attrs}{accessor} and $self->{attrs}{accessor} eq 'single') {
864 # This means that we got here after a merger of relationship conditions
865 # in ::Relationship::Base::search_related (the row method), and furthermore
866 # the relationship is of the 'single' type. This means that the condition
867 # provided by the relationship (already attached to $self) is sufficient,
868 # as there can be only one row in the database that would satisfy the
872 # no key was specified - fall down to heuristics mode:
873 # run through all unique queries registered on the resultset, and
874 # 'OR' all qualifying queries together
875 my (@unique_queries, %seen_column_combinations);
876 for my $c_name ($rsrc->unique_constraint_names) {
877 next if $seen_column_combinations{
878 join "\x00", sort $rsrc->unique_constraint_columns($c_name)
881 push @unique_queries, try {
882 $self->_build_unique_cond ($c_name, $call_cond, 'croak_on_nulls')
886 $final_cond = @unique_queries
887 ? [ map { $self->_qualify_cond_columns($_, $alias) } @unique_queries ]
888 : $self->_non_unique_find_fallback ($call_cond, $attrs)
892 # Run the query, passing the result_class since it should propagate for find
893 my $rs = $self->search ($final_cond, {result_class => $self->result_class, %$attrs});
894 if ($rs->_resolved_attrs->{collapse}) {
896 carp "Query returned more than one row" if $rs->next;
904 # This is a stop-gap method as agreed during the discussion on find() cleanup:
905 # http://lists.scsys.co.uk/pipermail/dbix-class/2010-October/009535.html
907 # It is invoked when find() is called in legacy-mode with insufficiently-unique
908 # condition. It is provided for overrides until a saner way forward is devised
910 # *NOTE* This is not a public method, and it's *GUARANTEED* to disappear down
911 # the road. Please adjust your tests accordingly to catch this situation early
912 # DBIx::Class::ResultSet->can('_non_unique_find_fallback') is reasonable
914 # The method will not be removed without an adequately complete replacement
915 # for strict-mode enforcement
916 sub _non_unique_find_fallback {
917 my ($self, $cond, $attrs) = @_;
919 return $self->_qualify_cond_columns(
921 exists $attrs->{alias}
923 : $self->{attrs}{alias}
928 sub _qualify_cond_columns {
929 my ($self, $cond, $alias) = @_;
931 my %aliased = %$cond;
932 for (keys %aliased) {
933 $aliased{"$alias.$_"} = delete $aliased{$_}
940 sub _build_unique_cond {
941 my ($self, $constraint_name, $extra_cond, $croak_on_null) = @_;
943 my @c_cols = $self->result_source->unique_constraint_columns($constraint_name);
945 # combination may fail if $self->{cond} is non-trivial
946 my ($final_cond) = try {
947 $self->_merge_with_rscond ($extra_cond)
952 # trim out everything not in $columns
953 $final_cond = { map {
954 exists $final_cond->{$_}
955 ? ( $_ => $final_cond->{$_} )
959 if (my @missing = grep
960 { ! ($croak_on_null ? defined $final_cond->{$_} : exists $final_cond->{$_}) }
963 $self->throw_exception( sprintf ( "Unable to satisfy requested constraint '%s', no values for column(s): %s",
965 join (', ', map { "'$_'" } @missing),
972 !$ENV{DBIC_NULLABLE_KEY_NOWARN}
974 my @undefs = sort grep { ! defined $final_cond->{$_} } (keys %$final_cond)
976 carp_unique ( sprintf (
977 "NULL/undef values supplied for requested unique constraint '%s' (NULL "
978 . 'values in column(s): %s). This is almost certainly not what you wanted, '
979 . 'though you can set DBIC_NULLABLE_KEY_NOWARN to disable this warning.',
981 join (', ', map { "'$_'" } @undefs),
988 =head2 search_related
992 =item Arguments: $rel_name, $cond?, L<\%attrs?|/ATTRIBUTES>
994 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
998 $new_rs = $cd_rs->search_related('artist', {
1002 Searches the specified relationship, optionally specifying a condition and
1003 attributes for matching records. See L</ATTRIBUTES> for more information.
1005 In list context, C<< ->all() >> is called implicitly on the resultset, thus
1006 returning a list of result objects instead. To avoid that, use L</search_related_rs>.
1008 See also L</search_related_rs>.
1012 sub search_related {
1013 return shift->related_resultset(shift)->search(@_);
1016 =head2 search_related_rs
1018 This method works exactly the same as search_related, except that
1019 it guarantees a resultset, even in list context.
1023 sub search_related_rs {
1024 return shift->related_resultset(shift)->search_rs(@_);
1031 =item Arguments: none
1033 =item Return Value: L<$cursor|DBIx::Class::Cursor>
1037 Returns a storage-driven cursor to the given resultset. See
1038 L<DBIx::Class::Cursor> for more information.
1045 return $self->{cursor} ||= do {
1046 my $attrs = $self->_resolved_attrs;
1047 $self->result_source->storage->select(
1048 $attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs
1057 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1059 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1063 my $cd = $schema->resultset('CD')->single({ year => 2001 });
1065 Inflates the first result without creating a cursor if the resultset has
1066 any records in it; if not returns C<undef>. Used by L</find> as a lean version
1069 While this method can take an optional search condition (just like L</search>)
1070 being a fast-code-path it does not recognize search attributes. If you need to
1071 add extra joins or similar, call L</search> and then chain-call L</single> on the
1072 L<DBIx::Class::ResultSet> returned.
1078 As of 0.08100, this method enforces the assumption that the preceding
1079 query returns only one row. If more than one row is returned, you will receive
1082 Query returned more than one row
1084 In this case, you should be using L</next> or L</find> instead, or if you really
1085 know what you are doing, use the L</rows> attribute to explicitly limit the size
1088 This method will also throw an exception if it is called on a resultset prefetching
1089 has_many, as such a prefetch implies fetching multiple rows from the database in
1090 order to assemble the resulting object.
1097 my ($self, $where) = @_;
1099 $self->throw_exception('single() only takes search conditions, no attributes. You want ->search( $cond, $attrs )->single()');
1102 my $attrs = { %{$self->_resolved_attrs} };
1104 $self->throw_exception(
1105 'single() can not be used on resultsets collapsing a has_many. Use find( \%cond ) or next() instead'
1106 ) if $attrs->{collapse};
1109 if (defined $attrs->{where}) {
1112 [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ }
1113 $where, delete $attrs->{where} ]
1116 $attrs->{where} = $where;
1120 my $data = [ $self->result_source->storage->select_single(
1121 $attrs->{from}, $attrs->{select},
1122 $attrs->{where}, $attrs
1125 return undef unless @$data;
1126 $self->{_stashed_rows} = [ $data ];
1127 $self->_construct_results->[0];
1134 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1136 =item Return Value: L<$resultsetcolumn|DBIx::Class::ResultSetColumn>
1140 my $max_length = $rs->get_column('length')->max;
1142 Returns a L<DBIx::Class::ResultSetColumn> instance for a column of the ResultSet.
1147 my ($self, $column) = @_;
1148 my $new = DBIx::Class::ResultSetColumn->new($self, $column);
1156 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1158 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1162 # WHERE title LIKE '%blue%'
1163 $cd_rs = $rs->search_like({ title => '%blue%'});
1165 Performs a search, but uses C<LIKE> instead of C<=> as the condition. Note
1166 that this is simply a convenience method retained for ex Class::DBI users.
1167 You most likely want to use L</search> with specific operators.
1169 For more information, see L<DBIx::Class::Manual::Cookbook>.
1171 This method is deprecated and will be removed in 0.09. Use L</search()>
1172 instead. An example conversion is:
1174 ->search_like({ foo => 'bar' });
1178 ->search({ foo => { like => 'bar' } });
1185 'search_like() is deprecated and will be removed in DBIC version 0.09.'
1186 .' Instead use ->search({ x => { -like => "y%" } })'
1187 .' (note the outer pair of {}s - they are important!)'
1189 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
1190 my $query = ref $_[0] eq 'HASH' ? { %{shift()} }: {@_};
1191 $query->{$_} = { 'like' => $query->{$_} } for keys %$query;
1192 return $class->search($query, { %$attrs });
1199 =item Arguments: $first, $last
1201 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1205 Returns a resultset or object list representing a subset of elements from the
1206 resultset slice is called on. Indexes are from 0, i.e., to get the first
1207 three records, call:
1209 my ($one, $two, $three) = $rs->slice(0, 2);
1214 my ($self, $min, $max) = @_;
1215 my $attrs = {}; # = { %{ $self->{attrs} || {} } };
1216 $attrs->{offset} = $self->{attrs}{offset} || 0;
1217 $attrs->{offset} += $min;
1218 $attrs->{rows} = ($max ? ($max - $min + 1) : 1);
1219 return $self->search(undef, $attrs);
1226 =item Arguments: none
1228 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1232 Returns the next element in the resultset (C<undef> is there is none).
1234 Can be used to efficiently iterate over records in the resultset:
1236 my $rs = $schema->resultset('CD')->search;
1237 while (my $cd = $rs->next) {
1241 Note that you need to store the resultset object, and call C<next> on it.
1242 Calling C<< resultset('Table')->next >> repeatedly will always return the
1243 first record from the resultset.
1250 if (my $cache = $self->get_cache) {
1251 $self->{all_cache_position} ||= 0;
1252 return $cache->[$self->{all_cache_position}++];
1255 if ($self->{attrs}{cache}) {
1256 delete $self->{pager};
1257 $self->{all_cache_position} = 1;
1258 return ($self->all)[0];
1261 return shift(@{$self->{_stashed_results}}) if @{ $self->{_stashed_results}||[] };
1263 $self->{_stashed_results} = $self->_construct_results
1266 return shift @{$self->{_stashed_results}};
1269 # Constructs as many results as it can in one pass while respecting
1270 # cursor laziness. Several modes of operation:
1272 # * Always builds everything present in @{$self->{_stashed_rows}}
1273 # * If called with $fetch_all true - pulls everything off the cursor and
1274 # builds all result structures (or objects) in one pass
1275 # * If $self->_resolved_attrs->{collapse} is true, checks the order_by
1276 # and if the resultset is ordered properly by the left side:
1277 # * Fetches stuff off the cursor until the "master object" changes,
1278 # and saves the last extra row (if any) in @{$self->{_stashed_rows}}
1280 # * Just fetches, and collapses/constructs everything as if $fetch_all
1281 # was requested (there is no other way to collapse except for an
1283 # * If no collapse is requested - just get the next row, construct and
1285 sub _construct_results {
1286 my ($self, $fetch_all) = @_;
1288 my $rsrc = $self->result_source;
1289 my $attrs = $self->_resolved_attrs;
1294 ! $attrs->{order_by}
1298 my @pcols = $rsrc->primary_columns
1300 # default order for collapsing unless the user asked for something
1301 $attrs->{order_by} = [ map { join '.', $attrs->{alias}, $_} @pcols ];
1302 $attrs->{_ordered_for_collapse} = 1;
1303 $attrs->{_order_is_artificial} = 1;
1306 # this will be used as both initial raw-row collector AND as a RV of
1307 # _construct_results. Not regrowing the array twice matters a lot...
1308 # a surprising amount actually
1309 my $rows = delete $self->{_stashed_rows};
1311 my $cursor; # we may not need one at all
1313 my $did_fetch_all = $fetch_all;
1316 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1317 $rows = [ ($rows ? @$rows : ()), $self->cursor->all ];
1319 elsif( $attrs->{collapse} ) {
1321 # a cursor will need to be closed over in case of collapse
1322 $cursor = $self->cursor;
1324 $attrs->{_ordered_for_collapse} = (
1330 ->_main_source_order_by_portion_is_stable($rsrc, $attrs->{order_by}, $attrs->{where})
1332 ) unless defined $attrs->{_ordered_for_collapse};
1334 if (! $attrs->{_ordered_for_collapse}) {
1337 # instead of looping over ->next, use ->all in stealth mode
1338 # *without* calling a ->reset afterwards
1339 # FIXME ENCAPSULATION - encapsulation breach, cursor method additions pending
1340 if (! $cursor->{_done}) {
1341 $rows = [ ($rows ? @$rows : ()), $cursor->all ];
1342 $cursor->{_done} = 1;
1347 if (! $did_fetch_all and ! @{$rows||[]} ) {
1348 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1349 $cursor ||= $self->cursor;
1350 if (scalar (my @r = $cursor->next) ) {
1355 return undef unless @{$rows||[]};
1357 # sanity check - people are too clever for their own good
1358 if ($attrs->{collapse} and my $aliastypes = $attrs->{_last_sqlmaker_alias_map} ) {
1360 my $multiplied_selectors;
1361 for my $sel_alias ( grep { $_ ne $attrs->{alias} } keys %{ $aliastypes->{selecting} } ) {
1363 $aliastypes->{multiplying}{$sel_alias}
1365 $aliastypes->{premultiplied}{$sel_alias}
1367 $multiplied_selectors->{$_} = 1 for values %{$aliastypes->{selecting}{$sel_alias}{-seen_columns}}
1371 for my $i (0 .. $#{$attrs->{as}} ) {
1372 my $sel = $attrs->{select}[$i];
1374 if (ref $sel eq 'SCALAR') {
1377 elsif( ref $sel eq 'REF' and ref $$sel eq 'ARRAY' ) {
1381 $self->throw_exception(
1382 'Result collapse not possible - selection from a has_many source redirected to the main object'
1383 ) if ($multiplied_selectors->{$sel} and $attrs->{as}[$i] !~ /\./);
1387 # hotspot - skip the setter
1388 my $res_class = $self->_result_class;
1390 my $inflator_cref = $self->{_result_inflator}{cref} ||= do {
1391 $res_class->can ('inflate_result')
1392 or $self->throw_exception("Inflator $res_class does not provide an inflate_result() method");
1395 my $infmap = $attrs->{as};
1397 $self->{_result_inflator}{is_core_row} = ( (
1400 ( \&DBIx::Class::Row::inflate_result || die "No ::Row::inflate_result() - can't happen" )
1401 ) ? 1 : 0 ) unless defined $self->{_result_inflator}{is_core_row};
1403 $self->{_result_inflator}{is_hri} = ( (
1404 ! $self->{_result_inflator}{is_core_row}
1407 require DBIx::Class::ResultClass::HashRefInflator
1409 DBIx::Class::ResultClass::HashRefInflator->can('inflate_result')
1411 ) ? 1 : 0 ) unless defined $self->{_result_inflator}{is_hri};
1414 if (! $attrs->{_related_results_construction}) {
1415 # construct a much simpler array->hash folder for the one-table cases right here
1416 if ($self->{_result_inflator}{is_hri}) {
1417 for my $r (@$rows) {
1418 $r = { map { $infmap->[$_] => $r->[$_] } 0..$#$infmap };
1421 # FIXME SUBOPTIMAL this is a very very very hot spot
1422 # while rather optimal we can *still* do much better, by
1423 # building a smarter Row::inflate_result(), and
1424 # switch to feeding it data via a much leaner interface
1426 # crude unscientific benchmarking indicated the shortcut eval is not worth it for
1427 # this particular resultset size
1428 elsif (@$rows < 60) {
1429 for my $r (@$rows) {
1430 $r = $inflator_cref->($res_class, $rsrc, { map { $infmap->[$_] => $r->[$_] } (0..$#$infmap) } );
1435 '$_ = $inflator_cref->($res_class, $rsrc, { %s }) for @$rows',
1436 join (', ', map { "\$infmap->[$_] => \$_->[$_]" } 0..$#$infmap )
1442 $self->{_result_inflator}{is_hri} ? 'hri'
1443 : $self->{_result_inflator}{is_core_row} ? 'classic_pruning'
1444 : 'classic_nonpruning'
1447 # $args and $attrs to _mk_row_parser are separated to delineate what is
1448 # core collapser stuff and what is dbic $rs specific
1449 @{$self->{_row_parser}{$parser_type}}{qw(cref nullcheck)} = $rsrc->_mk_row_parser({
1451 inflate_map => $infmap,
1452 collapse => $attrs->{collapse},
1453 premultiplied => $attrs->{_main_source_premultiplied},
1454 hri_style => $self->{_result_inflator}{is_hri},
1455 prune_null_branches => $self->{_result_inflator}{is_hri} || $self->{_result_inflator}{is_core_row},
1456 }, $attrs) unless $self->{_row_parser}{$parser_type}{cref};
1458 # column_info metadata historically hasn't been too reliable.
1459 # We need to start fixing this somehow (the collapse resolver
1460 # can't work without it). Add an explicit check for the *main*
1461 # result, hopefully this will gradually weed out such errors
1463 # FIXME - this is a temporary kludge that reduces performance
1464 # It is however necessary for the time being
1465 my ($unrolled_non_null_cols_to_check, $err);
1467 if (my $check_non_null_cols = $self->{_row_parser}{$parser_type}{nullcheck} ) {
1470 'Collapse aborted due to invalid ResultSource metadata - the following '
1471 . 'selections are declared non-nullable but NULLs were retrieved: '
1475 COL: for my $i (@$check_non_null_cols) {
1476 ! defined $_->[$i] and push @violating_idx, $i and next COL for @$rows;
1479 $self->throw_exception( $err . join (', ', map { "'$infmap->[$_]'" } @violating_idx ) )
1482 $unrolled_non_null_cols_to_check = join (',', @$check_non_null_cols);
1486 ($did_fetch_all or ! $attrs->{collapse}) ? undef
1487 : defined $unrolled_non_null_cols_to_check ? eval sprintf <<'EOS', $unrolled_non_null_cols_to_check
1489 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1490 my @r = $cursor->next or return;
1491 if (my @violating_idx = grep { ! defined $r[$_] } (%s) ) {
1492 $self->throw_exception( $err . join (', ', map { "'$infmap->[$_]'" } @violating_idx ) )
1498 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1499 my @r = $cursor->next or return;
1504 $self->{_row_parser}{$parser_type}{cref}->(
1506 $next_cref ? ( $next_cref, $self->{_stashed_rows} = [] ) : (),
1509 # Special-case multi-object HRI - there is no $inflator_cref pass
1510 unless ($self->{_result_inflator}{is_hri}) {
1511 $_ = $inflator_cref->($res_class, $rsrc, @$_) for @$rows
1515 # The @$rows check seems odd at first - why wouldn't we want to warn
1516 # regardless? The issue is things like find() etc, where the user
1517 # *knows* only one result will come back. In these cases the ->all
1518 # is not a pessimization, but rather something we actually want
1520 'Unable to properly collapse has_many results in iterator mode due '
1521 . 'to order criteria - performed an eager cursor slurp underneath. '
1522 . 'Consider using ->all() instead'
1523 ) if ( ! $fetch_all and @$rows > 1 );
1528 =head2 result_source
1532 =item Arguments: L<$result_source?|DBIx::Class::ResultSource>
1534 =item Return Value: L<$result_source|DBIx::Class::ResultSource>
1538 An accessor for the primary ResultSource object from which this ResultSet
1545 =item Arguments: $result_class?
1547 =item Return Value: $result_class
1551 An accessor for the class to use when creating result objects. Defaults to
1552 C<< result_source->result_class >> - which in most cases is the name of the
1553 L<"table"|DBIx::Class::Manual::Glossary/"ResultSource"> class.
1555 Note that changing the result_class will also remove any components
1556 that were originally loaded in the source class via
1557 L<DBIx::Class::ResultSource/load_components>. Any overloaded methods
1558 in the original source class will not run.
1563 my ($self, $result_class) = @_;
1564 if ($result_class) {
1566 # don't fire this for an object
1567 $self->ensure_class_loaded($result_class)
1568 unless ref($result_class);
1570 if ($self->get_cache) {
1571 carp_unique('Changing the result_class of a ResultSet instance with cached results is a noop - the cache contents will not be altered');
1573 # FIXME ENCAPSULATION - encapsulation breach, cursor method additions pending
1574 elsif ($self->{cursor} && $self->{cursor}{_pos}) {
1575 $self->throw_exception('Changing the result_class of a ResultSet instance with an active cursor is not supported');
1578 $self->_result_class($result_class);
1580 delete $self->{_result_inflator};
1582 $self->_result_class;
1589 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1591 =item Return Value: $count
1595 Performs an SQL C<COUNT> with the same query as the resultset was built
1596 with to find the number of elements. Passing arguments is equivalent to
1597 C<< $rs->search ($cond, \%attrs)->count >>
1603 return $self->search(@_)->count if @_ and defined $_[0];
1604 return scalar @{ $self->get_cache } if $self->get_cache;
1606 my $attrs = { %{ $self->_resolved_attrs } };
1608 # this is a little optimization - it is faster to do the limit
1609 # adjustments in software, instead of a subquery
1610 my ($rows, $offset) = delete @{$attrs}{qw/rows offset/};
1613 if ($self->_has_resolved_attr (qw/collapse group_by/)) {
1614 $crs = $self->_count_subq_rs ($attrs);
1617 $crs = $self->_count_rs ($attrs);
1619 my $count = $crs->next;
1621 $count -= $offset if $offset;
1622 $count = $rows if $rows and $rows < $count;
1623 $count = 0 if ($count < 0);
1632 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1634 =item Return Value: L<$count_rs|DBIx::Class::ResultSetColumn>
1638 Same as L</count> but returns a L<DBIx::Class::ResultSetColumn> object.
1639 This can be very handy for subqueries:
1641 ->search( { amount => $some_rs->count_rs->as_query } )
1643 As with regular resultsets the SQL query will be executed only after
1644 the resultset is accessed via L</next> or L</all>. That would return
1645 the same single value obtainable via L</count>.
1651 return $self->search(@_)->count_rs if @_;
1653 # this may look like a lack of abstraction (count() does about the same)
1654 # but in fact an _rs *must* use a subquery for the limits, as the
1655 # software based limiting can not be ported if this $rs is to be used
1656 # in a subquery itself (i.e. ->as_query)
1657 if ($self->_has_resolved_attr (qw/collapse group_by offset rows/)) {
1658 return $self->_count_subq_rs($self->{_attrs});
1661 return $self->_count_rs($self->{_attrs});
1666 # returns a ResultSetColumn object tied to the count query
1669 my ($self, $attrs) = @_;
1671 my $rsrc = $self->result_source;
1673 my $tmp_attrs = { %$attrs };
1674 # take off any limits, record_filter is cdbi, and no point of ordering nor locking a count
1675 delete @{$tmp_attrs}{qw/rows offset order_by record_filter for/};
1677 # overwrite the selector (supplied by the storage)
1678 $rsrc->resultset_class->new($rsrc, {
1680 select => $rsrc->storage->_count_select ($rsrc, $attrs),
1682 })->get_column ('count');
1686 # same as above but uses a subquery
1688 sub _count_subq_rs {
1689 my ($self, $attrs) = @_;
1691 my $rsrc = $self->result_source;
1693 my $sub_attrs = { %$attrs };
1694 # extra selectors do not go in the subquery and there is no point of ordering it, nor locking it
1695 delete @{$sub_attrs}{qw/collapse columns as select order_by for/};
1697 # if we multi-prefetch we group_by something unique, as this is what we would
1698 # get out of the rs via ->next/->all. We *DO WANT* to clobber old group_by regardless
1699 if ( $attrs->{collapse} ) {
1700 $sub_attrs->{group_by} = [ map { "$attrs->{alias}.$_" } @{
1701 $rsrc->_identifying_column_set || $self->throw_exception(
1702 'Unable to construct a unique group_by criteria properly collapsing the '
1703 . 'has_many prefetch before count()'
1708 # Calculate subquery selector
1709 if (my $g = $sub_attrs->{group_by}) {
1711 my $sql_maker = $rsrc->storage->sql_maker;
1713 # necessary as the group_by may refer to aliased functions
1715 for my $sel (@{$attrs->{select}}) {
1716 $sel_index->{$sel->{-as}} = $sel
1717 if (ref $sel eq 'HASH' and $sel->{-as});
1720 # anything from the original select mentioned on the group-by needs to make it to the inner selector
1721 # also look for named aggregates referred in the having clause
1722 # having often contains scalarrefs - thus parse it out entirely
1724 if ($attrs->{having}) {
1725 local $sql_maker->{having_bind};
1726 local $sql_maker->{quote_char} = $sql_maker->{quote_char};
1727 local $sql_maker->{name_sep} = $sql_maker->{name_sep};
1728 unless (defined $sql_maker->{quote_char} and length $sql_maker->{quote_char}) {
1729 $sql_maker->{quote_char} = [ "\x00", "\xFF" ];
1730 # if we don't unset it we screw up retarded but unfortunately working
1731 # 'MAX(foo.bar)' => { '>', 3 }
1732 $sql_maker->{name_sep} = '';
1735 my ($lquote, $rquote, $sep) = map { quotemeta $_ } ($sql_maker->_quote_chars, $sql_maker->name_sep);
1737 my $having_sql = $sql_maker->_parse_rs_attrs ({ having => $attrs->{having} });
1740 # search for both a proper quoted qualified string, for a naive unquoted scalarref
1741 # and if all fails for an utterly naive quoted scalar-with-function
1742 while ($having_sql =~ /
1743 $rquote $sep $lquote (.+?) $rquote
1745 [\s,] \w+ \. (\w+) [\s,]
1747 [\s,] $lquote (.+?) $rquote [\s,]
1749 my $part = $1 || $2 || $3; # one of them matched if we got here
1750 unless ($seen_having{$part}++) {
1757 my $colpiece = $sel_index->{$_} || $_;
1759 # unqualify join-based group_by's. Arcane but possible query
1760 # also horrible horrible hack to alias a column (not a func.)
1761 # (probably need to introduce SQLA syntax)
1762 if ($colpiece =~ /\./ && $colpiece !~ /^$attrs->{alias}\./) {
1765 $colpiece = \ sprintf ('%s AS %s', map { $sql_maker->_quote ($_) } ($colpiece, $as) );
1767 push @{$sub_attrs->{select}}, $colpiece;
1771 my @pcols = map { "$attrs->{alias}.$_" } ($rsrc->primary_columns);
1772 $sub_attrs->{select} = @pcols ? \@pcols : [ 1 ];
1775 return $rsrc->resultset_class
1776 ->new ($rsrc, $sub_attrs)
1778 ->search ({}, { columns => { count => $rsrc->storage->_count_select ($rsrc, $attrs) } })
1779 ->get_column ('count');
1783 =head2 count_literal
1785 B<CAVEAT>: C<count_literal> is provided for Class::DBI compatibility and
1786 should only be used in that context. See L</search_literal> for further info.
1790 =item Arguments: $sql_fragment, @standalone_bind_values
1792 =item Return Value: $count
1796 Counts the results in a literal query. Equivalent to calling L</search_literal>
1797 with the passed arguments, then L</count>.
1801 sub count_literal { shift->search_literal(@_)->count; }
1807 =item Arguments: none
1809 =item Return Value: L<@result_objs|DBIx::Class::Manual::ResultClass>
1813 Returns all elements in the resultset.
1820 $self->throw_exception("all() doesn't take any arguments, you probably wanted ->search(...)->all()");
1823 delete @{$self}{qw/_stashed_rows _stashed_results/};
1825 if (my $c = $self->get_cache) {
1829 $self->cursor->reset;
1831 my $objs = $self->_construct_results('fetch_all') || [];
1833 $self->set_cache($objs) if $self->{attrs}{cache};
1842 =item Arguments: none
1844 =item Return Value: $self
1848 Resets the resultset's cursor, so you can iterate through the elements again.
1849 Implicitly resets the storage cursor, so a subsequent L</next> will trigger
1857 delete @{$self}{qw/_stashed_rows _stashed_results/};
1858 $self->{all_cache_position} = 0;
1859 $self->cursor->reset;
1867 =item Arguments: none
1869 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1873 L<Resets|/reset> the resultset (causing a fresh query to storage) and returns
1874 an object for the first result (or C<undef> if the resultset is empty).
1879 return $_[0]->reset->next;
1885 # Determines whether and what type of subquery is required for the $rs operation.
1886 # If grouping is necessary either supplies its own, or verifies the current one
1887 # After all is done delegates to the proper storage method.
1889 sub _rs_update_delete {
1890 my ($self, $op, $values) = @_;
1892 my $rsrc = $self->result_source;
1893 my $storage = $rsrc->schema->storage;
1895 my $attrs = { %{$self->_resolved_attrs} };
1897 my $join_classifications;
1898 my ($existing_group_by) = delete @{$attrs}{qw(group_by _grouped_by_distinct)};
1900 # do we need a subquery for any reason?
1902 defined $existing_group_by
1904 # if {from} is unparseable wrap a subq
1905 ref($attrs->{from}) ne 'ARRAY'
1907 # limits call for a subq
1908 $self->_has_resolved_attr(qw/rows offset/)
1911 # simplify the joinmap, so we can further decide if a subq is necessary
1912 if (!$needs_subq and @{$attrs->{from}} > 1) {
1914 ($attrs->{from}, $join_classifications) =
1915 $storage->_prune_unused_joins ($attrs);
1917 # any non-pruneable non-local restricting joins imply subq
1918 $needs_subq = defined List::Util::first { $_ ne $attrs->{alias} } keys %{ $join_classifications->{restricting} || {} };
1921 # check if the head is composite (by now all joins are thrown out unless $needs_subq)
1923 (ref $attrs->{from}[0]) ne 'HASH'
1925 ref $attrs->{from}[0]{ $attrs->{from}[0]{-alias} }
1929 # do we need anything like a subquery?
1930 if (! $needs_subq) {
1931 # Most databases do not allow aliasing of tables in UPDATE/DELETE. Thus
1932 # a condition containing 'me' or other table prefixes will not work
1933 # at all. Tell SQLMaker to dequalify idents via a gross hack.
1935 my $sqla = $rsrc->storage->sql_maker;
1936 local $sqla->{_dequalify_idents} = 1;
1937 \[ $sqla->_recurse_where($self->{cond}) ];
1941 # we got this far - means it is time to wrap a subquery
1942 my $idcols = $rsrc->_identifying_column_set || $self->throw_exception(
1944 "Unable to perform complex resultset %s() without an identifying set of columns on source '%s'",
1950 # make a new $rs selecting only the PKs (that's all we really need for the subq)
1951 delete $attrs->{$_} for qw/select as collapse/;
1952 $attrs->{columns} = [ map { "$attrs->{alias}.$_" } @$idcols ];
1954 # this will be consumed by the pruner waaaaay down the stack
1955 $attrs->{_force_prune_multiplying_joins} = 1;
1957 my $subrs = (ref $self)->new($rsrc, $attrs);
1959 if (@$idcols == 1) {
1960 $cond = { $idcols->[0] => { -in => $subrs->as_query } };
1962 elsif ($storage->_use_multicolumn_in) {
1963 # no syntax for calling this properly yet
1964 # !!! EXPERIMENTAL API !!! WILL CHANGE !!!
1965 $cond = $storage->sql_maker->_where_op_multicolumn_in (
1966 $idcols, # how do I convey a list of idents...? can binds reside on lhs?
1971 # if all else fails - get all primary keys and operate over a ORed set
1972 # wrap in a transaction for consistency
1973 # this is where the group_by/multiplication starts to matter
1977 # we do not need to check pre-multipliers, since if the premulti is there, its
1978 # parent (who is multi) will be there too
1979 keys %{ $join_classifications->{multiplying} || {} }
1981 # make sure if there is a supplied group_by it matches the columns compiled above
1982 # perfectly. Anything else can not be sanely executed on most databases so croak
1983 # right then and there
1984 if ($existing_group_by) {
1985 my @current_group_by = map
1986 { $_ =~ /\./ ? $_ : "$attrs->{alias}.$_" }
1991 join ("\x00", sort @current_group_by)
1993 join ("\x00", sort @{$attrs->{columns}} )
1995 $self->throw_exception (
1996 "You have just attempted a $op operation on a resultset which does group_by"
1997 . ' on columns other than the primary keys, while DBIC internally needs to retrieve'
1998 . ' the primary keys in a subselect. All sane RDBMS engines do not support this'
1999 . ' kind of queries. Please retry the operation with a modified group_by or'
2000 . ' without using one at all.'
2005 $subrs = $subrs->search({}, { group_by => $attrs->{columns} });
2008 $guard = $storage->txn_scope_guard;
2010 for my $row ($subrs->cursor->all) {
2012 { $idcols->[$_] => $row->[$_] }
2019 my $res = $cond ? $storage->$op (
2021 $op eq 'update' ? $values : (),
2025 $guard->commit if $guard;
2034 =item Arguments: \%values
2036 =item Return Value: $underlying_storage_rv
2040 Sets the specified columns in the resultset to the supplied values in a
2041 single query. Note that this will not run any accessor/set_column/update
2042 triggers, nor will it update any result object instances derived from this
2043 resultset (this includes the contents of the L<resultset cache|/set_cache>
2044 if any). See L</update_all> if you need to execute any on-update
2045 triggers or cascades defined either by you or a
2046 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
2048 The return value is a pass through of what the underlying
2049 storage backend returned, and may vary. See L<DBI/execute> for the most
2054 Note that L</update> does not process/deflate any of the values passed in.
2055 This is unlike the corresponding L<DBIx::Class::Row/update>. The user must
2056 ensure manually that any value passed to this method will stringify to
2057 something the RDBMS knows how to deal with. A notable example is the
2058 handling of L<DateTime> objects, for more info see:
2059 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
2064 my ($self, $values) = @_;
2065 $self->throw_exception('Values for update must be a hash')
2066 unless ref $values eq 'HASH';
2068 return $self->_rs_update_delete ('update', $values);
2075 =item Arguments: \%values
2077 =item Return Value: 1
2081 Fetches all objects and updates them one at a time via
2082 L<DBIx::Class::Row/update>. Note that C<update_all> will run DBIC defined
2083 triggers, while L</update> will not.
2088 my ($self, $values) = @_;
2089 $self->throw_exception('Values for update_all must be a hash')
2090 unless ref $values eq 'HASH';
2092 my $guard = $self->result_source->schema->txn_scope_guard;
2093 $_->update({%$values}) for $self->all; # shallow copy - update will mangle it
2102 =item Arguments: none
2104 =item Return Value: $underlying_storage_rv
2108 Deletes the rows matching this resultset in a single query. Note that this
2109 will not run any delete triggers, nor will it alter the
2110 L<in_storage|DBIx::Class::Row/in_storage> status of any result object instances
2111 derived from this resultset (this includes the contents of the
2112 L<resultset cache|/set_cache> if any). See L</delete_all> if you need to
2113 execute any on-delete triggers or cascades defined either by you or a
2114 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
2116 The return value is a pass through of what the underlying storage backend
2117 returned, and may vary. See L<DBI/execute> for the most common case.
2123 $self->throw_exception('delete does not accept any arguments')
2126 return $self->_rs_update_delete ('delete');
2133 =item Arguments: none
2135 =item Return Value: 1
2139 Fetches all objects and deletes them one at a time via
2140 L<DBIx::Class::Row/delete>. Note that C<delete_all> will run DBIC defined
2141 triggers, while L</delete> will not.
2147 $self->throw_exception('delete_all does not accept any arguments')
2150 my $guard = $self->result_source->schema->txn_scope_guard;
2151 $_->delete for $self->all;
2160 =item Arguments: [ \@column_list, \@row_values+ ] | [ \%col_data+ ]
2162 =item Return Value: L<\@result_objects|DBIx::Class::Manual::ResultClass> (scalar context) | L<@result_objects|DBIx::Class::Manual::ResultClass> (list context)
2166 Accepts either an arrayref of hashrefs or alternatively an arrayref of
2173 The context of this method call has an important effect on what is
2174 submitted to storage. In void context data is fed directly to fastpath
2175 insertion routines provided by the underlying storage (most often
2176 L<DBI/execute_for_fetch>), bypassing the L<new|DBIx::Class::Row/new> and
2177 L<insert|DBIx::Class::Row/insert> calls on the
2178 L<Result|DBIx::Class::Manual::ResultClass> class, including any
2179 augmentation of these methods provided by components. For example if you
2180 are using something like L<DBIx::Class::UUIDColumns> to create primary
2181 keys for you, you will find that your PKs are empty. In this case you
2182 will have to explicitly force scalar or list context in order to create
2187 In non-void (scalar or list) context, this method is simply a wrapper
2188 for L</create>. Depending on list or scalar context either a list of
2189 L<Result|DBIx::Class::Manual::ResultClass> objects or an arrayref
2190 containing these objects is returned.
2192 When supplying data in "arrayref of arrayrefs" invocation style, the
2193 first element should be a list of column names and each subsequent
2194 element should be a data value in the earlier specified column order.
2197 $schema->resultset("Artist")->populate([
2198 [ qw( artistid name ) ],
2199 [ 100, 'A Formally Unknown Singer' ],
2200 [ 101, 'A singer that jumped the shark two albums ago' ],
2201 [ 102, 'An actually cool singer' ],
2204 For the arrayref of hashrefs style each hashref should be a structure
2205 suitable for passing to L</create>. Multi-create is also permitted with
2208 $schema->resultset("Artist")->populate([
2209 { artistid => 4, name => 'Manufactured Crap', cds => [
2210 { title => 'My First CD', year => 2006 },
2211 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2214 { artistid => 5, name => 'Angsty-Whiny Girl', cds => [
2215 { title => 'My parents sold me to a record company', year => 2005 },
2216 { title => 'Why Am I So Ugly?', year => 2006 },
2217 { title => 'I Got Surgery and am now Popular', year => 2007 }
2222 If you attempt a void-context multi-create as in the example above (each
2223 Artist also has the related list of CDs), and B<do not> supply the
2224 necessary autoinc foreign key information, this method will proxy to the
2225 less efficient L</create>, and then throw the Result objects away. In this
2226 case there are obviously no benefits to using this method over L</create>.
2233 # cruft placed in standalone method
2234 my $data = $self->_normalize_populate_args(@_);
2236 return unless @$data;
2238 if(defined wantarray) {
2239 my @created = map { $self->create($_) } @$data;
2240 return wantarray ? @created : \@created;
2243 my $first = $data->[0];
2245 # if a column is a registered relationship, and is a non-blessed hash/array, consider
2246 # it relationship data
2247 my (@rels, @columns);
2248 my $rsrc = $self->result_source;
2249 my $rels = { map { $_ => $rsrc->relationship_info($_) } $rsrc->relationships };
2250 for (keys %$first) {
2251 my $ref = ref $first->{$_};
2252 $rels->{$_} && ($ref eq 'ARRAY' or $ref eq 'HASH')
2258 my @pks = $rsrc->primary_columns;
2260 ## do the belongs_to relationships
2261 foreach my $index (0..$#$data) {
2263 # delegate to create() for any dataset without primary keys with specified relationships
2264 if (grep { !defined $data->[$index]->{$_} } @pks ) {
2266 if (grep { ref $data->[$index]{$r} eq $_ } qw/HASH ARRAY/) { # a related set must be a HASH or AoH
2267 my @ret = $self->populate($data);
2273 foreach my $rel (@rels) {
2274 next unless ref $data->[$index]->{$rel} eq "HASH";
2275 my $result = $self->related_resultset($rel)->create($data->[$index]->{$rel});
2276 my (undef, $reverse_relinfo) = %{$rsrc->reverse_relationship_info($rel)};
2277 my $related = $result->result_source->_resolve_condition(
2278 $reverse_relinfo->{cond},
2284 delete $data->[$index]->{$rel};
2285 $data->[$index] = {%{$data->[$index]}, %$related};
2287 push @columns, keys %$related if $index == 0;
2291 ## inherit the data locked in the conditions of the resultset
2292 my ($rs_data) = $self->_merge_with_rscond({});
2293 delete @{$rs_data}{@columns};
2295 ## do bulk insert on current row
2296 $rsrc->storage->insert_bulk(
2298 [@columns, keys %$rs_data],
2299 [ map { [ @$_{@columns}, values %$rs_data ] } @$data ],
2302 ## do the has_many relationships
2303 foreach my $item (@$data) {
2307 foreach my $rel (@rels) {
2308 next unless ref $item->{$rel} eq "ARRAY" && @{ $item->{$rel} };
2310 $main_row ||= $self->new_result({map { $_ => $item->{$_} } @pks});
2312 my $child = $main_row->$rel;
2314 my $related = $child->result_source->_resolve_condition(
2315 $rels->{$rel}{cond},
2321 my @rows_to_add = ref $item->{$rel} eq 'ARRAY' ? @{$item->{$rel}} : ($item->{$rel});
2322 my @populate = map { {%$_, %$related} } @rows_to_add;
2324 $child->populate( \@populate );
2331 # populate() arguments went over several incarnations
2332 # What we ultimately support is AoH
2333 sub _normalize_populate_args {
2334 my ($self, $arg) = @_;
2336 if (ref $arg eq 'ARRAY') {
2340 elsif (ref $arg->[0] eq 'HASH') {
2343 elsif (ref $arg->[0] eq 'ARRAY') {
2345 my @colnames = @{$arg->[0]};
2346 foreach my $values (@{$arg}[1 .. $#$arg]) {
2347 push @ret, { map { $colnames[$_] => $values->[$_] } (0 .. $#colnames) };
2353 $self->throw_exception('Populate expects an arrayref of hashrefs or arrayref of arrayrefs');
2360 =item Arguments: none
2362 =item Return Value: L<$pager|Data::Page>
2366 Returns a L<Data::Page> object for the current resultset. Only makes
2367 sense for queries with a C<page> attribute.
2369 To get the full count of entries for a paged resultset, call
2370 C<total_entries> on the L<Data::Page> object.
2377 return $self->{pager} if $self->{pager};
2379 my $attrs = $self->{attrs};
2380 if (!defined $attrs->{page}) {
2381 $self->throw_exception("Can't create pager for non-paged rs");
2383 elsif ($attrs->{page} <= 0) {
2384 $self->throw_exception('Invalid page number (page-numbers are 1-based)');
2386 $attrs->{rows} ||= 10;
2388 # throw away the paging flags and re-run the count (possibly
2389 # with a subselect) to get the real total count
2390 my $count_attrs = { %$attrs };
2391 delete @{$count_attrs}{qw/rows offset page pager/};
2393 my $total_rs = (ref $self)->new($self->result_source, $count_attrs);
2395 require DBIx::Class::ResultSet::Pager;
2396 return $self->{pager} = DBIx::Class::ResultSet::Pager->new(
2397 sub { $total_rs->count }, #lazy-get the total
2399 $self->{attrs}{page},
2407 =item Arguments: $page_number
2409 =item Return Value: L<$resultset|/search>
2413 Returns a resultset for the $page_number page of the resultset on which page
2414 is called, where each page contains a number of rows equal to the 'rows'
2415 attribute set on the resultset (10 by default).
2420 my ($self, $page) = @_;
2421 return (ref $self)->new($self->result_source, { %{$self->{attrs}}, page => $page });
2428 =item Arguments: \%col_data
2430 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2434 Creates a new result object in the resultset's result class and returns
2435 it. The row is not inserted into the database at this point, call
2436 L<DBIx::Class::Row/insert> to do that. Calling L<DBIx::Class::Row/in_storage>
2437 will tell you whether the result object has been inserted or not.
2439 Passes the hashref of input on to L<DBIx::Class::Row/new>.
2444 my ($self, $values) = @_;
2446 $self->throw_exception( "new_result takes only one argument - a hashref of values" )
2449 $self->throw_exception( "new_result expects a hashref" )
2450 unless (ref $values eq 'HASH');
2452 my ($merged_cond, $cols_from_relations) = $self->_merge_with_rscond($values);
2454 my $new = $self->result_class->new({
2456 ( @$cols_from_relations
2457 ? (-cols_from_relations => $cols_from_relations)
2460 -result_source => $self->result_source, # DO NOT REMOVE THIS, REQUIRED
2464 reftype($new) eq 'HASH'
2470 carp_unique (sprintf (
2471 "%s->new returned a blessed empty hashref - a strong indicator something is wrong with its inheritance chain",
2472 $self->result_class,
2479 # _merge_with_rscond
2481 # Takes a simple hash of K/V data and returns its copy merged with the
2482 # condition already present on the resultset. Additionally returns an
2483 # arrayref of value/condition names, which were inferred from related
2484 # objects (this is needed for in-memory related objects)
2485 sub _merge_with_rscond {
2486 my ($self, $data) = @_;
2488 my (%new_data, @cols_from_relations);
2490 my $alias = $self->{attrs}{alias};
2492 if (! defined $self->{cond}) {
2493 # just massage $data below
2495 elsif ($self->{cond} eq $DBIx::Class::ResultSource::UNRESOLVABLE_CONDITION) {
2496 %new_data = %{ $self->{attrs}{related_objects} || {} }; # nothing might have been inserted yet
2497 @cols_from_relations = keys %new_data;
2499 elsif (ref $self->{cond} ne 'HASH') {
2500 $self->throw_exception(
2501 "Can't abstract implicit construct, resultset condition not a hash"
2505 if ($self->{cond}) {
2506 my $implied = $self->_remove_alias(
2507 $self->result_source->schema->storage->_collapse_cond($self->{cond}),
2511 for my $c (keys %$implied) {
2512 my $v = $implied->{$c};
2513 if ( ! length ref $v or is_plain_value($v) ) {
2517 ref $v eq 'HASH' and keys %$v == 1 and exists $v->{'='} and is_literal_value($v->{'='})
2519 $new_data{$c} = $v->{'='};
2525 # precedence must be given to passed values over values inherited from
2526 # the cond, so the order here is important.
2529 %{ $self->_remove_alias($data, $alias) },
2532 return (\%new_data, \@cols_from_relations);
2535 # _has_resolved_attr
2537 # determines if the resultset defines at least one
2538 # of the attributes supplied
2540 # used to determine if a subquery is necessary
2542 # supports some virtual attributes:
2544 # This will scan for any joins being present on the resultset.
2545 # It is not a mere key-search but a deep inspection of {from}
2548 sub _has_resolved_attr {
2549 my ($self, @attr_names) = @_;
2551 my $attrs = $self->_resolved_attrs;
2555 for my $n (@attr_names) {
2556 if (grep { $n eq $_ } (qw/-join/) ) {
2557 $extra_checks{$n}++;
2561 my $attr = $attrs->{$n};
2563 next if not defined $attr;
2565 if (ref $attr eq 'HASH') {
2566 return 1 if keys %$attr;
2568 elsif (ref $attr eq 'ARRAY') {
2576 # a resolved join is expressed as a multi-level from
2578 $extra_checks{-join}
2580 ref $attrs->{from} eq 'ARRAY'
2582 @{$attrs->{from}} > 1
2590 # Remove the specified alias from the specified query hash. A copy is made so
2591 # the original query is not modified.
2594 my ($self, $query, $alias) = @_;
2596 my %orig = %{ $query || {} };
2599 foreach my $key (keys %orig) {
2601 $unaliased{$key} = $orig{$key};
2604 $unaliased{$1} = $orig{$key}
2605 if $key =~ m/^(?:\Q$alias\E\.)?([^.]+)$/;
2615 =item Arguments: none
2617 =item Return Value: \[ $sql, L<@bind_values|/DBIC BIND VALUES> ]
2621 Returns the SQL query and bind vars associated with the invocant.
2623 This is generally used as the RHS for a subquery.
2630 my $attrs = { %{ $self->_resolved_attrs } };
2632 my $aq = $self->result_source->storage->_select_args_to_query (
2633 $attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs
2643 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2645 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2649 my $artist = $schema->resultset('Artist')->find_or_new(
2650 { artist => 'fred' }, { key => 'artists' });
2652 $cd->cd_to_producer->find_or_new({ producer => $producer },
2653 { key => 'primary' });
2655 Find an existing record from this resultset using L</find>. if none exists,
2656 instantiate a new result object and return it. The object will not be saved
2657 into your storage until you call L<DBIx::Class::Row/insert> on it.
2659 You most likely want this method when looking for existing rows using a unique
2660 constraint that is not the primary key, or looking for related rows.
2662 If you want objects to be saved immediately, use L</find_or_create> instead.
2664 B<Note>: Make sure to read the documentation of L</find> and understand the
2665 significance of the C<key> attribute, as its lack may skew your search, and
2666 subsequently result in spurious new objects.
2668 B<Note>: Take care when using C<find_or_new> with a table having
2669 columns with default values that you intend to be automatically
2670 supplied by the database (e.g. an auto_increment primary key column).
2671 In normal usage, the value of such columns should NOT be included at
2672 all in the call to C<find_or_new>, even when set to C<undef>.
2678 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2679 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2680 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2683 return $self->new_result($hash);
2690 =item Arguments: \%col_data
2692 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2696 Attempt to create a single new row or a row with multiple related rows
2697 in the table represented by the resultset (and related tables). This
2698 will not check for duplicate rows before inserting, use
2699 L</find_or_create> to do that.
2701 To create one row for this resultset, pass a hashref of key/value
2702 pairs representing the columns of the table and the values you wish to
2703 store. If the appropriate relationships are set up, foreign key fields
2704 can also be passed an object representing the foreign row, and the
2705 value will be set to its primary key.
2707 To create related objects, pass a hashref of related-object column values
2708 B<keyed on the relationship name>. If the relationship is of type C<multi>
2709 (L<DBIx::Class::Relationship/has_many>) - pass an arrayref of hashrefs.
2710 The process will correctly identify columns holding foreign keys, and will
2711 transparently populate them from the keys of the corresponding relation.
2712 This can be applied recursively, and will work correctly for a structure
2713 with an arbitrary depth and width, as long as the relationships actually
2714 exists and the correct column data has been supplied.
2716 Instead of hashrefs of plain related data (key/value pairs), you may
2717 also pass new or inserted objects. New objects (not inserted yet, see
2718 L</new_result>), will be inserted into their appropriate tables.
2720 Effectively a shortcut for C<< ->new_result(\%col_data)->insert >>.
2722 Example of creating a new row.
2724 $person_rs->create({
2725 name=>"Some Person",
2726 email=>"somebody@someplace.com"
2729 Example of creating a new row and also creating rows in a related C<has_many>
2730 or C<has_one> resultset. Note Arrayref.
2733 { artistid => 4, name => 'Manufactured Crap', cds => [
2734 { title => 'My First CD', year => 2006 },
2735 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2740 Example of creating a new row and also creating a row in a related
2741 C<belongs_to> resultset. Note Hashref.
2744 title=>"Music for Silly Walks",
2747 name=>"Silly Musician",
2755 When subclassing ResultSet never attempt to override this method. Since
2756 it is a simple shortcut for C<< $self->new_result($attrs)->insert >>, a
2757 lot of the internals simply never call it, so your override will be
2758 bypassed more often than not. Override either L<DBIx::Class::Row/new>
2759 or L<DBIx::Class::Row/insert> depending on how early in the
2760 L</create> process you need to intervene. See also warning pertaining to
2768 my ($self, $col_data) = @_;
2769 $self->throw_exception( "create needs a hashref" )
2770 unless ref $col_data eq 'HASH';
2771 return $self->new_result($col_data)->insert;
2774 =head2 find_or_create
2778 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2780 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2784 $cd->cd_to_producer->find_or_create({ producer => $producer },
2785 { key => 'primary' });
2787 Tries to find a record based on its primary key or unique constraints; if none
2788 is found, creates one and returns that instead.
2790 my $cd = $schema->resultset('CD')->find_or_create({
2792 artist => 'Massive Attack',
2793 title => 'Mezzanine',
2797 Also takes an optional C<key> attribute, to search by a specific key or unique
2798 constraint. For example:
2800 my $cd = $schema->resultset('CD')->find_or_create(
2802 artist => 'Massive Attack',
2803 title => 'Mezzanine',
2805 { key => 'cd_artist_title' }
2808 B<Note>: Make sure to read the documentation of L</find> and understand the
2809 significance of the C<key> attribute, as its lack may skew your search, and
2810 subsequently result in spurious row creation.
2812 B<Note>: Because find_or_create() reads from the database and then
2813 possibly inserts based on the result, this method is subject to a race
2814 condition. Another process could create a record in the table after
2815 the find has completed and before the create has started. To avoid
2816 this problem, use find_or_create() inside a transaction.
2818 B<Note>: Take care when using C<find_or_create> with a table having
2819 columns with default values that you intend to be automatically
2820 supplied by the database (e.g. an auto_increment primary key column).
2821 In normal usage, the value of such columns should NOT be included at
2822 all in the call to C<find_or_create>, even when set to C<undef>.
2824 See also L</find> and L</update_or_create>. For information on how to declare
2825 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2827 If you need to know if an existing row was found or a new one created use
2828 L</find_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
2829 to call L<DBIx::Class::Row/insert> to save the newly created row to the
2832 my $cd = $schema->resultset('CD')->find_or_new({
2834 artist => 'Massive Attack',
2835 title => 'Mezzanine',
2839 if( !$cd->in_storage ) {
2846 sub find_or_create {
2848 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2849 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2850 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2853 return $self->create($hash);
2856 =head2 update_or_create
2860 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2862 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2866 $resultset->update_or_create({ col => $val, ... });
2868 Like L</find_or_create>, but if a row is found it is immediately updated via
2869 C<< $found_row->update (\%col_data) >>.
2872 Takes an optional C<key> attribute to search on a specific unique constraint.
2875 # In your application
2876 my $cd = $schema->resultset('CD')->update_or_create(
2878 artist => 'Massive Attack',
2879 title => 'Mezzanine',
2882 { key => 'cd_artist_title' }
2885 $cd->cd_to_producer->update_or_create({
2886 producer => $producer,
2892 B<Note>: Make sure to read the documentation of L</find> and understand the
2893 significance of the C<key> attribute, as its lack may skew your search, and
2894 subsequently result in spurious row creation.
2896 B<Note>: Take care when using C<update_or_create> with a table having
2897 columns with default values that you intend to be automatically
2898 supplied by the database (e.g. an auto_increment primary key column).
2899 In normal usage, the value of such columns should NOT be included at
2900 all in the call to C<update_or_create>, even when set to C<undef>.
2902 See also L</find> and L</find_or_create>. For information on how to declare
2903 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2905 If you need to know if an existing row was updated or a new one created use
2906 L</update_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
2907 to call L<DBIx::Class::Row/insert> to save the newly created row to the
2912 sub update_or_create {
2914 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2915 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2917 my $row = $self->find($cond, $attrs);
2919 $row->update($cond);
2923 return $self->create($cond);
2926 =head2 update_or_new
2930 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2932 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2936 $resultset->update_or_new({ col => $val, ... });
2938 Like L</find_or_new> but if a row is found it is immediately updated via
2939 C<< $found_row->update (\%col_data) >>.
2943 # In your application
2944 my $cd = $schema->resultset('CD')->update_or_new(
2946 artist => 'Massive Attack',
2947 title => 'Mezzanine',
2950 { key => 'cd_artist_title' }
2953 if ($cd->in_storage) {
2954 # the cd was updated
2957 # the cd is not yet in the database, let's insert it
2961 B<Note>: Make sure to read the documentation of L</find> and understand the
2962 significance of the C<key> attribute, as its lack may skew your search, and
2963 subsequently result in spurious new objects.
2965 B<Note>: Take care when using C<update_or_new> with a table having
2966 columns with default values that you intend to be automatically
2967 supplied by the database (e.g. an auto_increment primary key column).
2968 In normal usage, the value of such columns should NOT be included at
2969 all in the call to C<update_or_new>, even when set to C<undef>.
2971 See also L</find>, L</find_or_create> and L</find_or_new>.
2977 my $attrs = ( @_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {} );
2978 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2980 my $row = $self->find( $cond, $attrs );
2981 if ( defined $row ) {
2982 $row->update($cond);
2986 return $self->new_result($cond);
2993 =item Arguments: none
2995 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass> | undef
2999 Gets the contents of the cache for the resultset, if the cache is set.
3001 The cache is populated either by using the L</prefetch> attribute to
3002 L</search> or by calling L</set_cache>.
3014 =item Arguments: L<\@result_objs|DBIx::Class::Manual::ResultClass>
3016 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass>
3020 Sets the contents of the cache for the resultset. Expects an arrayref
3021 of objects of the same class as those produced by the resultset. Note that
3022 if the cache is set, the resultset will return the cached objects rather
3023 than re-querying the database even if the cache attr is not set.
3025 The contents of the cache can also be populated by using the
3026 L</prefetch> attribute to L</search>.
3031 my ( $self, $data ) = @_;
3032 $self->throw_exception("set_cache requires an arrayref")
3033 if defined($data) && (ref $data ne 'ARRAY');
3034 $self->{all_cache} = $data;
3041 =item Arguments: none
3043 =item Return Value: undef
3047 Clears the cache for the resultset.
3052 shift->set_cache(undef);
3059 =item Arguments: none
3061 =item Return Value: true, if the resultset has been paginated
3069 return !!$self->{attrs}{page};
3076 =item Arguments: none
3078 =item Return Value: true, if the resultset has been ordered with C<order_by>.
3086 return scalar $self->result_source->storage->_extract_order_criteria($self->{attrs}{order_by});
3089 =head2 related_resultset
3093 =item Arguments: $rel_name
3095 =item Return Value: L<$resultset|/search>
3099 Returns a related resultset for the supplied relationship name.
3101 $artist_rs = $schema->resultset('CD')->related_resultset('Artist');
3105 sub related_resultset {
3106 my ($self, $rel) = @_;
3108 return $self->{related_resultsets}{$rel}
3109 if defined $self->{related_resultsets}{$rel};
3111 return $self->{related_resultsets}{$rel} = do {
3112 my $rsrc = $self->result_source;
3113 my $rel_info = $rsrc->relationship_info($rel);
3115 $self->throw_exception(
3116 "search_related: result source '" . $rsrc->source_name .
3117 "' has no such relationship $rel")
3120 my $attrs = $self->_chain_relationship($rel);
3122 my $join_count = $attrs->{seen_join}{$rel};
3124 my $alias = $self->result_source->storage
3125 ->relname_to_table_alias($rel, $join_count);
3127 # since this is search_related, and we already slid the select window inwards
3128 # (the select/as attrs were deleted in the beginning), we need to flip all
3129 # left joins to inner, so we get the expected results
3130 # read the comment on top of the actual function to see what this does
3131 $attrs->{from} = $rsrc->schema->storage->_inner_join_to_node ($attrs->{from}, $alias);
3134 #XXX - temp fix for result_class bug. There likely is a more elegant fix -groditi
3135 delete @{$attrs}{qw(result_class alias)};
3137 my $rel_source = $rsrc->related_source($rel);
3141 # The reason we do this now instead of passing the alias to the
3142 # search_rs below is that if you wrap/overload resultset on the
3143 # source you need to know what alias it's -going- to have for things
3144 # to work sanely (e.g. RestrictWithObject wants to be able to add
3145 # extra query restrictions, and these may need to be $alias.)
3147 my $rel_attrs = $rel_source->resultset_attributes;
3148 local $rel_attrs->{alias} = $alias;
3150 $rel_source->resultset
3154 where => $attrs->{where},
3158 if (my $cache = $self->get_cache) {
3159 my @related_cache = map
3160 { $_->related_resultset($rel)->get_cache || () }
3164 $new->set_cache([ map @$_, @related_cache ]) if @related_cache == @$cache;
3171 =head2 current_source_alias
3175 =item Arguments: none
3177 =item Return Value: $source_alias
3181 Returns the current table alias for the result source this resultset is built
3182 on, that will be used in the SQL query. Usually it is C<me>.
3184 Currently the source alias that refers to the result set returned by a
3185 L</search>/L</find> family method depends on how you got to the resultset: it's
3186 C<me> by default, but eg. L</search_related> aliases it to the related result
3187 source name (and keeps C<me> referring to the original result set). The long
3188 term goal is to make L<DBIx::Class> always alias the current resultset as C<me>
3189 (and make this method unnecessary).
3191 Thus it's currently necessary to use this method in predefined queries (see
3192 L<DBIx::Class::Manual::Cookbook/Predefined searches>) when referring to the
3193 source alias of the current result set:
3195 # in a result set class
3197 my ($self, $user) = @_;
3199 my $me = $self->current_source_alias;
3201 return $self->search({
3202 "$me.modified" => $user->id,
3208 sub current_source_alias {
3209 return (shift->{attrs} || {})->{alias} || 'me';
3212 =head2 as_subselect_rs
3216 =item Arguments: none
3218 =item Return Value: L<$resultset|/search>
3222 Act as a barrier to SQL symbols. The resultset provided will be made into a
3223 "virtual view" by including it as a subquery within the from clause. From this
3224 point on, any joined tables are inaccessible to ->search on the resultset (as if
3225 it were simply where-filtered without joins). For example:
3227 my $rs = $schema->resultset('Bar')->search({'x.name' => 'abc'},{ join => 'x' });
3229 # 'x' now pollutes the query namespace
3231 # So the following works as expected
3232 my $ok_rs = $rs->search({'x.other' => 1});
3234 # But this doesn't: instead of finding a 'Bar' related to two x rows (abc and
3235 # def) we look for one row with contradictory terms and join in another table
3236 # (aliased 'x_2') which we never use
3237 my $broken_rs = $rs->search({'x.name' => 'def'});
3239 my $rs2 = $rs->as_subselect_rs;
3241 # doesn't work - 'x' is no longer accessible in $rs2, having been sealed away
3242 my $not_joined_rs = $rs2->search({'x.other' => 1});
3244 # works as expected: finds a 'table' row related to two x rows (abc and def)
3245 my $correctly_joined_rs = $rs2->search({'x.name' => 'def'});
3247 Another example of when one might use this would be to select a subset of
3248 columns in a group by clause:
3250 my $rs = $schema->resultset('Bar')->search(undef, {
3251 group_by => [qw{ id foo_id baz_id }],
3252 })->as_subselect_rs->search(undef, {
3253 columns => [qw{ id foo_id }]
3256 In the above example normally columns would have to be equal to the group by,
3257 but because we isolated the group by into a subselect the above works.
3261 sub as_subselect_rs {
3264 my $attrs = $self->_resolved_attrs;
3266 my $fresh_rs = (ref $self)->new (
3267 $self->result_source
3270 # these pieces will be locked in the subquery
3271 delete $fresh_rs->{cond};
3272 delete @{$fresh_rs->{attrs}}{qw/where bind/};
3274 return $fresh_rs->search( {}, {
3276 $attrs->{alias} => $self->as_query,
3277 -alias => $attrs->{alias},
3278 -rsrc => $self->result_source,
3280 alias => $attrs->{alias},
3284 # This code is called by search_related, and makes sure there
3285 # is clear separation between the joins before, during, and
3286 # after the relationship. This information is needed later
3287 # in order to properly resolve prefetch aliases (any alias
3288 # with a relation_chain_depth less than the depth of the
3289 # current prefetch is not considered)
3291 # The increments happen twice per join. An even number means a
3292 # relationship specified via a search_related, whereas an odd
3293 # number indicates a join/prefetch added via attributes
3295 # Also this code will wrap the current resultset (the one we
3296 # chain to) in a subselect IFF it contains limiting attributes
3297 sub _chain_relationship {
3298 my ($self, $rel) = @_;
3299 my $source = $self->result_source;
3300 my $attrs = { %{$self->{attrs}||{}} };
3302 # we need to take the prefetch the attrs into account before we
3303 # ->_resolve_join as otherwise they get lost - captainL
3304 my $join = $self->_merge_joinpref_attr( $attrs->{join}, $attrs->{prefetch} );
3306 delete @{$attrs}{qw/join prefetch collapse group_by distinct _grouped_by_distinct select as columns +select +as +columns/};
3308 my $seen = { %{ (delete $attrs->{seen_join}) || {} } };
3311 my @force_subq_attrs = qw/offset rows group_by having/;
3314 ($attrs->{from} && ref $attrs->{from} ne 'ARRAY')
3316 $self->_has_resolved_attr (@force_subq_attrs)
3318 # Nuke the prefetch (if any) before the new $rs attrs
3319 # are resolved (prefetch is useless - we are wrapping
3320 # a subquery anyway).
3321 my $rs_copy = $self->search;
3322 $rs_copy->{attrs}{join} = $self->_merge_joinpref_attr (
3323 $rs_copy->{attrs}{join},
3324 delete $rs_copy->{attrs}{prefetch},
3329 -alias => $attrs->{alias},
3330 $attrs->{alias} => $rs_copy->as_query,
3332 delete @{$attrs}{@force_subq_attrs, qw/where bind/};
3333 $seen->{-relation_chain_depth} = 0;
3335 elsif ($attrs->{from}) { #shallow copy suffices
3336 $from = [ @{$attrs->{from}} ];
3341 -alias => $attrs->{alias},
3342 $attrs->{alias} => $source->from,
3346 my $jpath = ($seen->{-relation_chain_depth})
3347 ? $from->[-1][0]{-join_path}
3350 my @requested_joins = $source->_resolve_join(
3357 push @$from, @requested_joins;
3359 $seen->{-relation_chain_depth}++;
3361 # if $self already had a join/prefetch specified on it, the requested
3362 # $rel might very well be already included. What we do in this case
3363 # is effectively a no-op (except that we bump up the chain_depth on
3364 # the join in question so we could tell it *is* the search_related)
3367 # we consider the last one thus reverse
3368 for my $j (reverse @requested_joins) {
3369 my ($last_j) = keys %{$j->[0]{-join_path}[-1]};
3370 if ($rel eq $last_j) {
3371 $j->[0]{-relation_chain_depth}++;
3377 unless ($already_joined) {
3378 push @$from, $source->_resolve_join(
3386 $seen->{-relation_chain_depth}++;
3388 return {%$attrs, from => $from, seen_join => $seen};
3391 sub _resolved_attrs {
3393 return $self->{_attrs} if $self->{_attrs};
3395 my $attrs = { %{ $self->{attrs} || {} } };
3396 my $source = $self->result_source;
3397 my $alias = $attrs->{alias};
3399 $self->throw_exception("Specifying distinct => 1 in conjunction with collapse => 1 is unsupported")
3400 if $attrs->{collapse} and $attrs->{distinct};
3402 # default selection list
3403 $attrs->{columns} = [ $source->columns ]
3404 unless List::Util::first { exists $attrs->{$_} } qw/columns cols select as/;
3406 # merge selectors together
3407 for (qw/columns select as/) {
3408 $attrs->{$_} = $self->_merge_attr($attrs->{$_}, delete $attrs->{"+$_"})
3409 if $attrs->{$_} or $attrs->{"+$_"};
3412 # disassemble columns
3414 if (my $cols = delete $attrs->{columns}) {
3415 for my $c (ref $cols eq 'ARRAY' ? @$cols : $cols) {
3416 if (ref $c eq 'HASH') {
3417 for my $as (sort keys %$c) {
3418 push @sel, $c->{$as};
3429 # when trying to weed off duplicates later do not go past this point -
3430 # everything added from here on is unbalanced "anyone's guess" stuff
3431 my $dedup_stop_idx = $#as;
3433 push @as, @{ ref $attrs->{as} eq 'ARRAY' ? $attrs->{as} : [ $attrs->{as} ] }
3435 push @sel, @{ ref $attrs->{select} eq 'ARRAY' ? $attrs->{select} : [ $attrs->{select} ] }
3436 if $attrs->{select};
3438 # assume all unqualified selectors to apply to the current alias (legacy stuff)
3439 $_ = (ref $_ or $_ =~ /\./) ? $_ : "$alias.$_" for @sel;
3441 # disqualify all $alias.col as-bits (inflate-map mandated)
3442 $_ = ($_ =~ /^\Q$alias.\E(.+)$/) ? $1 : $_ for @as;
3444 # de-duplicate the result (remove *identical* select/as pairs)
3445 # and also die on duplicate {as} pointing to different {select}s
3446 # not using a c-style for as the condition is prone to shrinkage
3449 while ($i <= $dedup_stop_idx) {
3450 if ($seen->{"$sel[$i] \x00\x00 $as[$i]"}++) {
3455 elsif ($seen->{$as[$i]}++) {
3456 $self->throw_exception(
3457 "inflate_result() alias '$as[$i]' specified twice with different SQL-side {select}-ors"
3465 $attrs->{select} = \@sel;
3466 $attrs->{as} = \@as;
3468 $attrs->{from} ||= [{
3470 -alias => $self->{attrs}{alias},
3471 $self->{attrs}{alias} => $source->from,
3474 if ( $attrs->{join} || $attrs->{prefetch} ) {
3476 $self->throw_exception ('join/prefetch can not be used with a custom {from}')
3477 if ref $attrs->{from} ne 'ARRAY';
3479 my $join = (delete $attrs->{join}) || {};
3481 if ( defined $attrs->{prefetch} ) {
3482 $join = $self->_merge_joinpref_attr( $join, $attrs->{prefetch} );
3485 $attrs->{from} = # have to copy here to avoid corrupting the original
3487 @{ $attrs->{from} },
3488 $source->_resolve_join(
3491 { %{ $attrs->{seen_join} || {} } },
3492 ( $attrs->{seen_join} && keys %{$attrs->{seen_join}})
3493 ? $attrs->{from}[-1][0]{-join_path}
3500 if ( defined $attrs->{order_by} ) {
3501 $attrs->{order_by} = (
3502 ref( $attrs->{order_by} ) eq 'ARRAY'
3503 ? [ @{ $attrs->{order_by} } ]
3504 : [ $attrs->{order_by} || () ]
3508 if ($attrs->{group_by} and ref $attrs->{group_by} ne 'ARRAY') {
3509 $attrs->{group_by} = [ $attrs->{group_by} ];
3513 # generate selections based on the prefetch helper
3514 my ($prefetch, @prefetch_select, @prefetch_as);
3515 $prefetch = $self->_merge_joinpref_attr( {}, delete $attrs->{prefetch} )
3516 if defined $attrs->{prefetch};
3520 $self->throw_exception("Unable to prefetch, resultset contains an unnamed selector $attrs->{_dark_selector}{string}")
3521 if $attrs->{_dark_selector};
3523 $self->throw_exception("Specifying prefetch in conjunction with an explicit collapse => 0 is unsupported")
3524 if defined $attrs->{collapse} and ! $attrs->{collapse};
3526 $attrs->{collapse} = 1;
3528 # this is a separate structure (we don't look in {from} directly)
3529 # as the resolver needs to shift things off the lists to work
3530 # properly (identical-prefetches on different branches)
3532 if (ref $attrs->{from} eq 'ARRAY') {
3534 my $start_depth = $attrs->{seen_join}{-relation_chain_depth} || 0;
3536 for my $j ( @{$attrs->{from}}[1 .. $#{$attrs->{from}} ] ) {
3537 next unless $j->[0]{-alias};
3538 next unless $j->[0]{-join_path};
3539 next if ($j->[0]{-relation_chain_depth} || 0) < $start_depth;
3541 my @jpath = map { keys %$_ } @{$j->[0]{-join_path}};
3544 $p = $p->{$_} ||= {} for @jpath[ ($start_depth/2) .. $#jpath]; #only even depths are actual jpath boundaries
3545 push @{$p->{-join_aliases} }, $j->[0]{-alias};
3549 my @prefetch = $source->_resolve_prefetch( $prefetch, $alias, $join_map );
3551 # save these for after distinct resolution
3552 @prefetch_select = map { $_->[0] } @prefetch;
3553 @prefetch_as = map { $_->[1] } @prefetch;
3556 # run through the resulting joinstructure (starting from our current slot)
3557 # and unset collapse if proven unnecessary
3559 # also while we are at it find out if the current root source has
3560 # been premultiplied by previous related_source chaining
3562 # this allows to predict whether a root object with all other relation
3563 # data set to NULL is in fact unique
3564 if ($attrs->{collapse}) {
3566 if (ref $attrs->{from} eq 'ARRAY') {
3568 if (@{$attrs->{from}} == 1) {
3569 # no joins - no collapse
3570 $attrs->{collapse} = 0;
3573 # find where our table-spec starts
3574 my @fromlist = @{$attrs->{from}};
3576 my $t = shift @fromlist;
3579 # me vs join from-spec distinction - a ref means non-root
3580 if (ref $t eq 'ARRAY') {
3582 $is_multi ||= ! $t->{-is_single};
3584 last if ($t->{-alias} && $t->{-alias} eq $alias);
3585 $attrs->{_main_source_premultiplied} ||= $is_multi;
3588 # no non-singles remaining, nor any premultiplication - nothing to collapse
3590 ! $attrs->{_main_source_premultiplied}
3592 ! List::Util::first { ! $_->[0]{-is_single} } @fromlist
3594 $attrs->{collapse} = 0;
3600 # if we can not analyze the from - err on the side of safety
3601 $attrs->{_main_source_premultiplied} = 1;
3605 # generate the distinct induced group_by before injecting the prefetched select/as parts
3606 if (delete $attrs->{distinct}) {
3607 if ($attrs->{group_by}) {
3608 carp_unique ("Useless use of distinct on a grouped resultset ('distinct' is ignored when a 'group_by' is present)");
3611 $attrs->{_grouped_by_distinct} = 1;
3612 # distinct affects only the main selection part, not what prefetch may add below
3613 ($attrs->{group_by}, my $new_order) = $source->storage->_group_over_selection($attrs);
3615 # FIXME possibly ignore a rewritten order_by (may turn out to be an issue)
3616 # The thinking is: if we are collapsing the subquerying prefetch engine will
3617 # rip stuff apart for us anyway, and we do not want to have a potentially
3618 # function-converted external order_by
3619 # ( there is an explicit if ( collapse && _grouped_by_distinct ) check in DBIHacks )
3620 $attrs->{order_by} = $new_order unless $attrs->{collapse};
3624 # inject prefetch-bound selection (if any)
3625 push @{$attrs->{select}}, @prefetch_select;
3626 push @{$attrs->{as}}, @prefetch_as;
3628 # whether we can get away with the dumbest (possibly DBI-internal) collapser
3629 if ( List::Util::first { $_ =~ /\./ } @{$attrs->{as}} ) {
3630 $attrs->{_related_results_construction} = 1;
3633 # if both page and offset are specified, produce a combined offset
3634 # even though it doesn't make much sense, this is what pre 081xx has
3636 if (my $page = delete $attrs->{page}) {
3638 ($attrs->{rows} * ($page - 1))
3640 ($attrs->{offset} || 0)
3644 return $self->{_attrs} = $attrs;
3648 my ($self, $attr) = @_;
3650 if (ref $attr eq 'HASH') {
3651 return $self->_rollout_hash($attr);
3652 } elsif (ref $attr eq 'ARRAY') {
3653 return $self->_rollout_array($attr);
3659 sub _rollout_array {
3660 my ($self, $attr) = @_;
3663 foreach my $element (@{$attr}) {
3664 if (ref $element eq 'HASH') {
3665 push( @rolled_array, @{ $self->_rollout_hash( $element ) } );
3666 } elsif (ref $element eq 'ARRAY') {
3667 # XXX - should probably recurse here
3668 push( @rolled_array, @{$self->_rollout_array($element)} );
3670 push( @rolled_array, $element );
3673 return \@rolled_array;
3677 my ($self, $attr) = @_;
3680 foreach my $key (keys %{$attr}) {
3681 push( @rolled_array, { $key => $attr->{$key} } );
3683 return \@rolled_array;
3686 sub _calculate_score {
3687 my ($self, $a, $b) = @_;
3689 if (defined $a xor defined $b) {
3692 elsif (not defined $a) {
3696 if (ref $b eq 'HASH') {
3697 my ($b_key) = keys %{$b};
3698 if (ref $a eq 'HASH') {
3699 my ($a_key) = keys %{$a};
3700 if ($a_key eq $b_key) {
3701 return (1 + $self->_calculate_score( $a->{$a_key}, $b->{$b_key} ));
3706 return ($a eq $b_key) ? 1 : 0;
3709 if (ref $a eq 'HASH') {
3710 my ($a_key) = keys %{$a};
3711 return ($b eq $a_key) ? 1 : 0;
3713 return ($b eq $a) ? 1 : 0;
3718 sub _merge_joinpref_attr {
3719 my ($self, $orig, $import) = @_;
3721 return $import unless defined($orig);
3722 return $orig unless defined($import);
3724 $orig = $self->_rollout_attr($orig);
3725 $import = $self->_rollout_attr($import);
3728 foreach my $import_element ( @{$import} ) {
3729 # find best candidate from $orig to merge $b_element into
3730 my $best_candidate = { position => undef, score => 0 }; my $position = 0;
3731 foreach my $orig_element ( @{$orig} ) {
3732 my $score = $self->_calculate_score( $orig_element, $import_element );
3733 if ($score > $best_candidate->{score}) {
3734 $best_candidate->{position} = $position;
3735 $best_candidate->{score} = $score;
3739 my ($import_key) = ( ref $import_element eq 'HASH' ) ? keys %{$import_element} : ($import_element);
3740 $import_key = '' if not defined $import_key;
3742 if ($best_candidate->{score} == 0 || exists $seen_keys->{$import_key}) {
3743 push( @{$orig}, $import_element );
3745 my $orig_best = $orig->[$best_candidate->{position}];
3746 # merge orig_best and b_element together and replace original with merged
3747 if (ref $orig_best ne 'HASH') {
3748 $orig->[$best_candidate->{position}] = $import_element;
3749 } elsif (ref $import_element eq 'HASH') {
3750 my ($key) = keys %{$orig_best};
3751 $orig->[$best_candidate->{position}] = { $key => $self->_merge_joinpref_attr($orig_best->{$key}, $import_element->{$key}) };
3754 $seen_keys->{$import_key} = 1; # don't merge the same key twice
3757 return @$orig ? $orig : ();
3765 require Hash::Merge;
3766 my $hm = Hash::Merge->new;
3768 $hm->specify_behavior({
3771 my ($defl, $defr) = map { defined $_ } (@_[0,1]);
3773 if ($defl xor $defr) {
3774 return [ $defl ? $_[0] : $_[1] ];
3779 elsif (__HM_DEDUP and $_[0] eq $_[1]) {
3783 return [$_[0], $_[1]];
3787 return $_[1] if !defined $_[0];
3788 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3789 return [$_[0], @{$_[1]}]
3792 return [] if !defined $_[0] and !keys %{$_[1]};
3793 return [ $_[1] ] if !defined $_[0];
3794 return [ $_[0] ] if !keys %{$_[1]};
3795 return [$_[0], $_[1]]
3800 return $_[0] if !defined $_[1];
3801 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3802 return [@{$_[0]}, $_[1]]
3805 my @ret = @{$_[0]} or return $_[1];
3806 return [ @ret, @{$_[1]} ] unless __HM_DEDUP;
3807 my %idx = map { $_ => 1 } @ret;
3808 push @ret, grep { ! defined $idx{$_} } (@{$_[1]});
3812 return [ $_[1] ] if ! @{$_[0]};
3813 return $_[0] if !keys %{$_[1]};
3814 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3815 return [ @{$_[0]}, $_[1] ];
3820 return [] if !keys %{$_[0]} and !defined $_[1];
3821 return [ $_[0] ] if !defined $_[1];
3822 return [ $_[1] ] if !keys %{$_[0]};
3823 return [$_[0], $_[1]]
3826 return [] if !keys %{$_[0]} and !@{$_[1]};
3827 return [ $_[0] ] if !@{$_[1]};
3828 return $_[1] if !keys %{$_[0]};
3829 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3830 return [ $_[0], @{$_[1]} ];
3833 return [] if !keys %{$_[0]} and !keys %{$_[1]};
3834 return [ $_[0] ] if !keys %{$_[1]};
3835 return [ $_[1] ] if !keys %{$_[0]};
3836 return [ $_[0] ] if $_[0] eq $_[1];
3837 return [ $_[0], $_[1] ];
3840 } => 'DBIC_RS_ATTR_MERGER');
3844 return $hm->merge ($_[1], $_[2]);
3848 sub STORABLE_freeze {
3849 my ($self, $cloning) = @_;
3850 my $to_serialize = { %$self };
3852 # A cursor in progress can't be serialized (and would make little sense anyway)
3853 # the parser can be regenerated (and can't be serialized)
3854 delete @{$to_serialize}{qw/cursor _row_parser _result_inflator/};
3856 # nor is it sensical to store a not-yet-fired-count pager
3857 if ($to_serialize->{pager} and ref $to_serialize->{pager}{total_entries} eq 'CODE') {
3858 delete $to_serialize->{pager};
3861 Storable::nfreeze($to_serialize);
3864 # need this hook for symmetry
3866 my ($self, $cloning, $serialized) = @_;
3868 %$self = %{ Storable::thaw($serialized) };
3874 =head2 throw_exception
3876 See L<DBIx::Class::Schema/throw_exception> for details.
3880 sub throw_exception {
3883 if (ref $self and my $rsrc = $self->result_source) {
3884 $rsrc->throw_exception(@_)
3887 DBIx::Class::Exception->throw(@_);
3895 # XXX: FIXME: Attributes docs need clearing up
3899 Attributes are used to refine a ResultSet in various ways when
3900 searching for data. They can be passed to any method which takes an
3901 C<\%attrs> argument. See L</search>, L</search_rs>, L</find>,
3904 Default attributes can be set on the result class using
3905 L<DBIx::Class::ResultSource/resultset_attributes>. (Please read
3906 the CAVEATS on that feature before using it!)
3908 These are in no particular order:
3914 =item Value: ( $order_by | \@order_by | \%order_by )
3918 Which column(s) to order the results by.
3920 [The full list of suitable values is documented in
3921 L<SQL::Abstract/"ORDER BY CLAUSES">; the following is a summary of
3924 If a single column name, or an arrayref of names is supplied, the
3925 argument is passed through directly to SQL. The hashref syntax allows
3926 for connection-agnostic specification of ordering direction:
3928 For descending order:
3930 order_by => { -desc => [qw/col1 col2 col3/] }
3932 For explicit ascending order:
3934 order_by => { -asc => 'col' }
3936 The old scalarref syntax (i.e. order_by => \'year DESC') is still
3937 supported, although you are strongly encouraged to use the hashref
3938 syntax as outlined above.
3944 =item Value: \@columns | \%columns | $column
3948 Shortcut to request a particular set of columns to be retrieved. Each
3949 column spec may be a string (a table column name), or a hash (in which
3950 case the key is the C<as> value, and the value is used as the C<select>
3951 expression). Adds C<me.> onto the start of any column without a C<.> in
3952 it and sets C<select> from that, then auto-populates C<as> from
3953 C<select> as normal. (You may also use the C<cols> attribute, as in
3954 earlier versions of DBIC, but this is deprecated)
3956 Essentially C<columns> does the same as L</select> and L</as>.
3958 columns => [ 'some_column', { dbic_slot => 'another_column' } ]
3962 select => [qw(some_column another_column)],
3963 as => [qw(some_column dbic_slot)]
3965 If you want to individually retrieve related columns (in essence perform
3966 manual prefetch) you have to make sure to specify the correct inflation slot
3967 chain such that it matches existing relationships:
3969 my $rs = $schema->resultset('Artist')->search({}, {
3970 # required to tell DBIC to collapse has_many relationships
3972 join => { cds => 'tracks'},
3974 'cds.cdid' => 'cds.cdid',
3975 'cds.tracks.title' => 'tracks.title',
3981 B<NOTE:> You B<MUST> explicitly quote C<'+columns'> when using this attribute.
3982 Not doing so causes Perl to incorrectly interpret C<+columns> as a bareword
3983 with a unary plus operator before it, which is the same as simply C<columns>.
3987 =item Value: \@extra_columns
3991 Indicates additional columns to be selected from storage. Works the same as
3992 L</columns> but adds columns to the current selection. (You may also use the
3993 C<include_columns> attribute, as in earlier versions of DBIC, but this is
3996 $schema->resultset('CD')->search(undef, {
3997 '+columns' => ['artist.name'],
4001 would return all CDs and include a 'name' column to the information
4002 passed to object inflation. Note that the 'artist' is the name of the
4003 column (or relationship) accessor, and 'name' is the name of the column
4004 accessor in the related table.
4010 =item Value: \@select_columns
4014 Indicates which columns should be selected from the storage. You can use
4015 column names, or in the case of RDBMS back ends, function or stored procedure
4018 $rs = $schema->resultset('Employee')->search(undef, {
4021 { count => 'employeeid' },
4022 { max => { length => 'name' }, -as => 'longest_name' }
4027 SELECT name, COUNT( employeeid ), MAX( LENGTH( name ) ) AS longest_name FROM employee
4029 B<NOTE:> You will almost always need a corresponding L</as> attribute when you
4030 use L</select>, to instruct DBIx::Class how to store the result of the column.
4031 Also note that the L</as> attribute has nothing to do with the SQL-side 'AS'
4032 identifier aliasing. You can however alias a function, so you can use it in
4033 e.g. an C<ORDER BY> clause. This is done via the C<-as> B<select function
4034 attribute> supplied as shown in the example above.
4038 B<NOTE:> You B<MUST> explicitly quote C<'+select'> when using this attribute.
4039 Not doing so causes Perl to incorrectly interpret C<+select> as a bareword
4040 with a unary plus operator before it, which is the same as simply C<select>.
4044 =item Value: \@extra_select_columns
4048 Indicates additional columns to be selected from storage. Works the same as
4049 L</select> but adds columns to the current selection, instead of specifying
4050 a new explicit list.
4056 =item Value: \@inflation_names
4060 Indicates DBIC-side names for object inflation. That is L</as> indicates the
4061 slot name in which the column value will be stored within the
4062 L<Row|DBIx::Class::Row> object. The value will then be accessible via this
4063 identifier by the C<get_column> method (or via the object accessor B<if one
4064 with the same name already exists>) as shown below. The L</as> attribute has
4065 B<nothing to do> with the SQL-side C<AS>. See L</select> for details.
4067 $rs = $schema->resultset('Employee')->search(undef, {
4070 { count => 'employeeid' },
4071 { max => { length => 'name' }, -as => 'longest_name' }
4080 If the object against which the search is performed already has an accessor
4081 matching a column name specified in C<as>, the value can be retrieved using
4082 the accessor as normal:
4084 my $name = $employee->name();
4086 If on the other hand an accessor does not exist in the object, you need to
4087 use C<get_column> instead:
4089 my $employee_count = $employee->get_column('employee_count');
4091 You can create your own accessors if required - see
4092 L<DBIx::Class::Manual::Cookbook> for details.
4096 B<NOTE:> You B<MUST> explicitly quote C<'+as'> when using this attribute.
4097 Not doing so causes Perl to incorrectly interpret C<+as> as a bareword
4098 with a unary plus operator before it, which is the same as simply C<as>.
4102 =item Value: \@extra_inflation_names
4106 Indicates additional inflation names for selectors added via L</+select>. See L</as>.
4112 =item Value: ($rel_name | \@rel_names | \%rel_names)
4116 Contains a list of relationships that should be joined for this query. For
4119 # Get CDs by Nine Inch Nails
4120 my $rs = $schema->resultset('CD')->search(
4121 { 'artist.name' => 'Nine Inch Nails' },
4122 { join => 'artist' }
4125 Can also contain a hash reference to refer to the other relation's relations.
4128 package MyApp::Schema::Track;
4129 use base qw/DBIx::Class/;
4130 __PACKAGE__->table('track');
4131 __PACKAGE__->add_columns(qw/trackid cd position title/);
4132 __PACKAGE__->set_primary_key('trackid');
4133 __PACKAGE__->belongs_to(cd => 'MyApp::Schema::CD');
4136 # In your application
4137 my $rs = $schema->resultset('Artist')->search(
4138 { 'track.title' => 'Teardrop' },
4140 join => { cd => 'track' },
4141 order_by => 'artist.name',
4145 You need to use the relationship (not the table) name in conditions,
4146 because they are aliased as such. The current table is aliased as "me", so
4147 you need to use me.column_name in order to avoid ambiguity. For example:
4149 # Get CDs from 1984 with a 'Foo' track
4150 my $rs = $schema->resultset('CD')->search(
4153 'tracks.name' => 'Foo'
4155 { join => 'tracks' }
4158 If the same join is supplied twice, it will be aliased to <rel>_2 (and
4159 similarly for a third time). For e.g.
4161 my $rs = $schema->resultset('Artist')->search({
4162 'cds.title' => 'Down to Earth',
4163 'cds_2.title' => 'Popular',
4165 join => [ qw/cds cds/ ],
4168 will return a set of all artists that have both a cd with title 'Down
4169 to Earth' and a cd with title 'Popular'.
4171 If you want to fetch related objects from other tables as well, see L</prefetch>
4174 NOTE: An internal join-chain pruner will discard certain joins while
4175 constructing the actual SQL query, as long as the joins in question do not
4176 affect the retrieved result. This for example includes 1:1 left joins
4177 that are not part of the restriction specification (WHERE/HAVING) nor are
4178 a part of the query selection.
4180 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
4186 =item Value: (0 | 1)
4190 When set to a true value, indicates that any rows fetched from joined has_many
4191 relationships are to be aggregated into the corresponding "parent" object. For
4192 example, the resultset:
4194 my $rs = $schema->resultset('CD')->search({}, {
4195 '+columns' => [ qw/ tracks.title tracks.position / ],
4200 While executing the following query:
4202 SELECT me.*, tracks.title, tracks.position
4204 LEFT JOIN track tracks
4205 ON tracks.cdid = me.cdid
4207 Will return only as many objects as there are rows in the CD source, even
4208 though the result of the query may span many rows. Each of these CD objects
4209 will in turn have multiple "Track" objects hidden behind the has_many
4210 generated accessor C<tracks>. Without C<< collapse => 1 >>, the return values
4211 of this resultset would be as many CD objects as there are tracks (a "Cartesian
4212 product"), with each CD object containing exactly one of all fetched Track data.
4214 When a collapse is requested on a non-ordered resultset, an order by some
4215 unique part of the main source (the left-most table) is inserted automatically.
4216 This is done so that the resultset is allowed to be "lazy" - calling
4217 L<< $rs->next|/next >> will fetch only as many rows as it needs to build the next
4218 object with all of its related data.
4220 If an L</order_by> is already declared, and orders the resultset in a way that
4221 makes collapsing as described above impossible (e.g. C<< ORDER BY
4222 has_many_rel.column >> or C<ORDER BY RANDOM()>), DBIC will automatically
4223 switch to "eager" mode and slurp the entire resultset before constructing the
4224 first object returned by L</next>.
4226 Setting this attribute on a resultset that does not join any has_many
4227 relations is a no-op.
4229 For a more in-depth discussion, see L</PREFETCHING>.
4235 =item Value: ($rel_name | \@rel_names | \%rel_names)
4239 This attribute is a shorthand for specifying a L</join> spec, adding all
4240 columns from the joined related sources as L</+columns> and setting
4241 L</collapse> to a true value. For example, the following two queries are
4244 my $rs = $schema->resultset('Artist')->search({}, {
4245 prefetch => { cds => ['genre', 'tracks' ] },
4250 my $rs = $schema->resultset('Artist')->search({}, {
4251 join => { cds => ['genre', 'tracks' ] },
4255 { +{ "cds.$_" => "cds.$_" } }
4256 $schema->source('Artist')->related_source('cds')->columns
4259 { +{ "cds.genre.$_" => "genre.$_" } }
4260 $schema->source('Artist')->related_source('cds')->related_source('genre')->columns
4263 { +{ "cds.tracks.$_" => "tracks.$_" } }
4264 $schema->source('Artist')->related_source('cds')->related_source('tracks')->columns
4269 Both producing the following SQL:
4271 SELECT me.artistid, me.name, me.rank, me.charfield,
4272 cds.cdid, cds.artist, cds.title, cds.year, cds.genreid, cds.single_track,
4273 genre.genreid, genre.name,
4274 tracks.trackid, tracks.cd, tracks.position, tracks.title, tracks.last_updated_on, tracks.last_updated_at
4277 ON cds.artist = me.artistid
4278 LEFT JOIN genre genre
4279 ON genre.genreid = cds.genreid
4280 LEFT JOIN track tracks
4281 ON tracks.cd = cds.cdid
4282 ORDER BY me.artistid
4284 While L</prefetch> implies a L</join>, it is ok to mix the two together, as
4285 the arguments are properly merged and generally do the right thing. For
4286 example, you may want to do the following:
4288 my $artists_and_cds_without_genre = $schema->resultset('Artist')->search(
4289 { 'genre.genreid' => undef },
4291 join => { cds => 'genre' },
4296 Which generates the following SQL:
4298 SELECT me.artistid, me.name, me.rank, me.charfield,
4299 cds.cdid, cds.artist, cds.title, cds.year, cds.genreid, cds.single_track
4302 ON cds.artist = me.artistid
4303 LEFT JOIN genre genre
4304 ON genre.genreid = cds.genreid
4305 WHERE genre.genreid IS NULL
4306 ORDER BY me.artistid
4308 For a more in-depth discussion, see L</PREFETCHING>.
4314 =item Value: $source_alias
4318 Sets the source alias for the query. Normally, this defaults to C<me>, but
4319 nested search queries (sub-SELECTs) might need specific aliases set to
4320 reference inner queries. For example:
4323 ->related_resultset('CDs')
4324 ->related_resultset('Tracks')
4326 'track.id' => { -ident => 'none_search.id' },
4330 my $ids = $self->search({
4333 alias => 'none_search',
4334 group_by => 'none_search.id',
4335 })->get_column('id')->as_query;
4337 $self->search({ id => { -in => $ids } })
4339 This attribute is directly tied to L</current_source_alias>.
4349 Makes the resultset paged and specifies the page to retrieve. Effectively
4350 identical to creating a non-pages resultset and then calling ->page($page)
4353 If L</rows> attribute is not specified it defaults to 10 rows per page.
4355 When you have a paged resultset, L</count> will only return the number
4356 of rows in the page. To get the total, use the L</pager> and call
4357 C<total_entries> on it.
4367 Specifies the maximum number of rows for direct retrieval or the number of
4368 rows per page if the page attribute or method is used.
4374 =item Value: $offset
4378 Specifies the (zero-based) row number for the first row to be returned, or the
4379 of the first row of the first page if paging is used.
4381 =head2 software_limit
4385 =item Value: (0 | 1)
4389 When combined with L</rows> and/or L</offset> the generated SQL will not
4390 include any limit dialect stanzas. Instead the entire result will be selected
4391 as if no limits were specified, and DBIC will perform the limit locally, by
4392 artificially advancing and finishing the resulting L</cursor>.
4394 This is the recommended way of performing resultset limiting when no sane RDBMS
4395 implementation is available (e.g.
4396 L<Sybase ASE|DBIx::Class::Storage::DBI::Sybase::ASE> using the
4397 L<Generic Sub Query|DBIx::Class::SQLMaker::LimitDialects/GenericSubQ> hack)
4403 =item Value: \@columns
4407 A arrayref of columns to group by. Can include columns of joined tables.
4409 group_by => [qw/ column1 column2 ... /]
4415 =item Value: $condition
4419 HAVING is a select statement attribute that is applied between GROUP BY and
4420 ORDER BY. It is applied to the after the grouping calculations have been
4423 having => { 'count_employee' => { '>=', 100 } }
4425 or with an in-place function in which case literal SQL is required:
4427 having => \[ 'count(employee) >= ?', [ count => 100 ] ]
4433 =item Value: (0 | 1)
4437 Set to 1 to automatically generate a L</group_by> clause based on the selection
4438 (including intelligent handling of L</order_by> contents). Note that the group
4439 criteria calculation takes place over the B<final> selection. This includes
4440 any L</+columns>, L</+select> or L</order_by> additions in subsequent
4441 L</search> calls, and standalone columns selected via
4442 L<DBIx::Class::ResultSetColumn> (L</get_column>). A notable exception are the
4443 extra selections specified via L</prefetch> - such selections are explicitly
4444 excluded from group criteria calculations.
4446 If the final ResultSet also explicitly defines a L</group_by> attribute, this
4447 setting is ignored and an appropriate warning is issued.
4453 Adds to the WHERE clause.
4455 # only return rows WHERE deleted IS NULL for all searches
4456 __PACKAGE__->resultset_attributes({ where => { deleted => undef } });
4458 Can be overridden by passing C<< { where => undef } >> as an attribute
4461 For more complicated where clauses see L<SQL::Abstract/WHERE CLAUSES>.
4467 Set to 1 to cache search results. This prevents extra SQL queries if you
4468 revisit rows in your ResultSet:
4470 my $resultset = $schema->resultset('Artist')->search( undef, { cache => 1 } );
4472 while( my $artist = $resultset->next ) {
4476 $rs->first; # without cache, this would issue a query
4478 By default, searches are not cached.
4480 For more examples of using these attributes, see
4481 L<DBIx::Class::Manual::Cookbook>.
4487 =item Value: ( 'update' | 'shared' | \$scalar )
4491 Set to 'update' for a SELECT ... FOR UPDATE or 'shared' for a SELECT
4492 ... FOR SHARED. If \$scalar is passed, this is taken directly and embedded in the
4497 DBIx::Class supports arbitrary related data prefetching from multiple related
4498 sources. Any combination of relationship types and column sets are supported.
4499 If L<collapsing|/collapse> is requested, there is an additional requirement of
4500 selecting enough data to make every individual object uniquely identifiable.
4502 Here are some more involved examples, based on the following relationship map:
4505 My::Schema::CD->belongs_to( artist => 'My::Schema::Artist' );
4506 My::Schema::CD->might_have( liner_note => 'My::Schema::LinerNotes' );
4507 My::Schema::CD->has_many( tracks => 'My::Schema::Track' );
4509 My::Schema::Artist->belongs_to( record_label => 'My::Schema::RecordLabel' );
4511 My::Schema::Track->has_many( guests => 'My::Schema::Guest' );
4515 my $rs = $schema->resultset('Tag')->search(
4524 The initial search results in SQL like the following:
4526 SELECT tag.*, cd.*, artist.* FROM tag
4527 JOIN cd ON tag.cd = cd.cdid
4528 JOIN artist ON cd.artist = artist.artistid
4530 L<DBIx::Class> has no need to go back to the database when we access the
4531 C<cd> or C<artist> relationships, which saves us two SQL statements in this
4534 Simple prefetches will be joined automatically, so there is no need
4535 for a C<join> attribute in the above search.
4537 The L</prefetch> attribute can be used with any of the relationship types
4538 and multiple prefetches can be specified together. Below is a more complex
4539 example that prefetches a CD's artist, its liner notes (if present),
4540 the cover image, the tracks on that CD, and the guests on those
4543 my $rs = $schema->resultset('CD')->search(
4547 { artist => 'record_label'}, # belongs_to => belongs_to
4548 'liner_note', # might_have
4549 'cover_image', # has_one
4550 { tracks => 'guests' }, # has_many => has_many
4555 This will produce SQL like the following:
4557 SELECT cd.*, artist.*, record_label.*, liner_note.*, cover_image.*,
4561 ON artist.artistid = me.artistid
4562 JOIN record_label record_label
4563 ON record_label.labelid = artist.labelid
4564 LEFT JOIN track tracks
4565 ON tracks.cdid = me.cdid
4566 LEFT JOIN guest guests
4567 ON guests.trackid = track.trackid
4568 LEFT JOIN liner_notes liner_note
4569 ON liner_note.cdid = me.cdid
4570 JOIN cd_artwork cover_image
4571 ON cover_image.cdid = me.cdid
4574 Now the C<artist>, C<record_label>, C<liner_note>, C<cover_image>,
4575 C<tracks>, and C<guests> of the CD will all be available through the
4576 relationship accessors without the need for additional queries to the
4581 Prefetch does a lot of deep magic. As such, it may not behave exactly
4582 as you might expect.
4588 Prefetch uses the L</cache> to populate the prefetched relationships. This
4589 may or may not be what you want.
4593 If you specify a condition on a prefetched relationship, ONLY those
4594 rows that match the prefetched condition will be fetched into that relationship.
4595 This means that adding prefetch to a search() B<may alter> what is returned by
4596 traversing a relationship. So, if you have C<< Artist->has_many(CDs) >> and you do
4598 my $artist_rs = $schema->resultset('Artist')->search({
4604 my $count = $artist_rs->first->cds->count;
4606 my $artist_rs_prefetch = $artist_rs->search( {}, { prefetch => 'cds' } );
4608 my $prefetch_count = $artist_rs_prefetch->first->cds->count;
4610 cmp_ok( $count, '==', $prefetch_count, "Counts should be the same" );
4612 That cmp_ok() may or may not pass depending on the datasets involved. In other
4613 words the C<WHERE> condition would apply to the entire dataset, just like
4614 it would in regular SQL. If you want to add a condition only to the "right side"
4615 of a C<LEFT JOIN> - consider declaring and using a L<relationship with a custom
4616 condition|DBIx::Class::Relationship::Base/condition>
4620 =head1 DBIC BIND VALUES
4622 Because DBIC may need more information to bind values than just the column name
4623 and value itself, it uses a special format for both passing and receiving bind
4624 values. Each bind value should be composed of an arrayref of
4625 C<< [ \%args => $val ] >>. The format of C<< \%args >> is currently:
4631 If present (in any form), this is what is being passed directly to bind_param.
4632 Note that different DBD's expect different bind args. (e.g. DBD::SQLite takes
4633 a single numerical type, while DBD::Pg takes a hashref if bind options.)
4635 If this is specified, all other bind options described below are ignored.
4639 If present, this is used to infer the actual bind attribute by passing to
4640 C<< $resolved_storage->bind_attribute_by_data_type() >>. Defaults to the
4641 "data_type" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>.
4643 Note that the data type is somewhat freeform (hence the sqlt_ prefix);
4644 currently drivers are expected to "Do the Right Thing" when given a common
4645 datatype name. (Not ideal, but that's what we got at this point.)
4649 Currently used to correctly allocate buffers for bind_param_inout().
4650 Defaults to "size" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>,
4651 or to a sensible value based on the "data_type".
4655 Used to fill in missing sqlt_datatype and sqlt_size attributes (if they are
4656 explicitly specified they are never overridden). Also used by some weird DBDs,
4657 where the column name should be available at bind_param time (e.g. Oracle).
4661 For backwards compatibility and convenience, the following shortcuts are
4664 [ $name => $val ] === [ { dbic_colname => $name }, $val ]
4665 [ \$dt => $val ] === [ { sqlt_datatype => $dt }, $val ]
4666 [ undef, $val ] === [ {}, $val ]
4667 $val === [ {}, $val ]
4669 =head1 AUTHOR AND CONTRIBUTORS
4671 See L<AUTHOR|DBIx::Class/AUTHOR> and L<CONTRIBUTORS|DBIx::Class/CONTRIBUTORS> in DBIx::Class
4675 You may distribute this code under the same terms as Perl itself.