1 package DBIx::Class::ResultSet;
5 use base qw/DBIx::Class/;
7 use DBIx::Class::ResultSetColumn;
8 use Scalar::Util qw/blessed weaken reftype/;
10 use Data::Compare (); # no imports!!! guard against insane architecture
12 # not importing first() as it will clash with our own method
16 # De-duplication in _merge_attr() is disabled, but left in for reference
17 # (the merger is used for other things that ought not to be de-duped)
18 *__HM_DEDUP = sub () { 0 };
28 # this is real - CDBICompat overrides it with insanity
29 # yes, prototype won't matter, but that's for now ;)
32 __PACKAGE__->mk_group_accessors('simple' => qw/_result_class result_source/);
36 DBIx::Class::ResultSet - Represents a query used for fetching a set of results.
40 my $users_rs = $schema->resultset('User');
41 while( $user = $users_rs->next) {
42 print $user->username;
45 my $registered_users_rs = $schema->resultset('User')->search({ registered => 1 });
46 my @cds_in_2005 = $schema->resultset('CD')->search({ year => 2005 })->all();
50 A ResultSet is an object which stores a set of conditions representing
51 a query. It is the backbone of DBIx::Class (i.e. the really
52 important/useful bit).
54 No SQL is executed on the database when a ResultSet is created, it
55 just stores all the conditions needed to create the query.
57 A basic ResultSet representing the data of an entire table is returned
58 by calling C<resultset> on a L<DBIx::Class::Schema> and passing in a
59 L<Source|DBIx::Class::Manual::Glossary/Source> name.
61 my $users_rs = $schema->resultset('User');
63 A new ResultSet is returned from calling L</search> on an existing
64 ResultSet. The new one will contain all the conditions of the
65 original, plus any new conditions added in the C<search> call.
67 A ResultSet also incorporates an implicit iterator. L</next> and L</reset>
68 can be used to walk through all the L<DBIx::Class::Row>s the ResultSet
71 The query that the ResultSet represents is B<only> executed against
72 the database when these methods are called:
73 L</find>, L</next>, L</all>, L</first>, L</single>, L</count>.
75 If a resultset is used in a numeric context it returns the L</count>.
76 However, if it is used in a boolean context it is B<always> true. So if
77 you want to check if a resultset has any results, you must use C<if $rs
80 =head1 CUSTOM ResultSet CLASSES THAT USE Moose
82 If you want to make your custom ResultSet classes with L<Moose>, use a template
85 package MyApp::Schema::ResultSet::User;
88 use namespace::autoclean;
90 extends 'DBIx::Class::ResultSet';
92 sub BUILDARGS { $_[2] }
96 __PACKAGE__->meta->make_immutable;
100 The L<MooseX::NonMoose> is necessary so that the L<Moose> constructor does not
101 clash with the regular ResultSet constructor. Alternatively, you can use:
103 __PACKAGE__->meta->make_immutable(inline_constructor => 0);
105 The L<BUILDARGS|Moose::Manual::Construction/BUILDARGS> is necessary because the
106 signature of the ResultSet C<new> is C<< ->new($source, \%args) >>.
110 =head2 Chaining resultsets
112 Let's say you've got a query that needs to be run to return some data
113 to the user. But, you have an authorization system in place that
114 prevents certain users from seeing certain information. So, you want
115 to construct the basic query in one method, but add constraints to it in
120 my $request = $self->get_request; # Get a request object somehow.
121 my $schema = $self->result_source->schema;
123 my $cd_rs = $schema->resultset('CD')->search({
124 title => $request->param('title'),
125 year => $request->param('year'),
128 $cd_rs = $self->apply_security_policy( $cd_rs );
130 return $cd_rs->all();
133 sub apply_security_policy {
142 =head3 Resolving conditions and attributes
144 When a resultset is chained from another resultset, conditions and
145 attributes with the same keys need resolving.
147 L</join>, L</prefetch>, L</+select>, L</+as> attributes are merged
148 into the existing ones from the original resultset.
150 The L</where> and L</having> attributes, and any search conditions, are
151 merged with an SQL C<AND> to the existing condition from the original
154 All other attributes are overridden by any new ones supplied in the
157 =head2 Multiple queries
159 Since a resultset just defines a query, you can do all sorts of
160 things with it with the same object.
162 # Don't hit the DB yet.
163 my $cd_rs = $schema->resultset('CD')->search({
164 title => 'something',
168 # Each of these hits the DB individually.
169 my $count = $cd_rs->count;
170 my $most_recent = $cd_rs->get_column('date_released')->max();
171 my @records = $cd_rs->all;
173 And it's not just limited to SELECT statements.
179 $cd_rs->create({ artist => 'Fred' });
181 Which is the same as:
183 $schema->resultset('CD')->create({
184 title => 'something',
189 See: L</search>, L</count>, L</get_column>, L</all>, L</create>.
197 =item Arguments: L<$source|DBIx::Class::ResultSource>, L<\%attrs?|/ATTRIBUTES>
199 =item Return Value: L<$resultset|/search>
203 The resultset constructor. Takes a source object (usually a
204 L<DBIx::Class::ResultSourceProxy::Table>) and an attribute hash (see
205 L</ATTRIBUTES> below). Does not perform any queries -- these are
206 executed as needed by the other methods.
208 Generally you never construct a resultset manually. Instead you get one
210 C<< $schema->L<resultset|DBIx::Class::Schema/resultset>('$source_name') >>
211 or C<< $another_resultset->L<search|/search>(...) >> (the later called in
214 my $rs = $schema->resultset('CD')->search({ title => '100th Window' });
220 If called on an object, proxies to L</new_result> instead, so
222 my $cd = $schema->resultset('CD')->new({ title => 'Spoon' });
224 will return a CD object, not a ResultSet, and is equivalent to:
226 my $cd = $schema->resultset('CD')->new_result({ title => 'Spoon' });
228 Please also keep in mind that many internals call L</new_result> directly,
229 so overloading this method with the idea of intercepting new result object
230 creation B<will not work>. See also warning pertaining to L</create>.
238 return $class->new_result(@_) if ref $class;
240 my ($source, $attrs) = @_;
241 $source = $source->resolve
242 if $source->isa('DBIx::Class::ResultSourceHandle');
243 $attrs = { %{$attrs||{}} };
245 if ($attrs->{page}) {
246 $attrs->{rows} ||= 10;
249 $attrs->{alias} ||= 'me';
252 result_source => $source,
253 cond => $attrs->{where},
258 # if there is a dark selector, this means we are already in a
259 # chain and the cleanup/sanification was taken care of by
261 $self->_normalize_selection($attrs)
262 unless $attrs->{_dark_selector};
265 $attrs->{result_class} || $source->result_class
275 =item Arguments: L<$cond|DBIx::Class::SQLMaker> | undef, L<\%attrs?|/ATTRIBUTES>
277 =item Return Value: $resultset (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
281 my @cds = $cd_rs->search({ year => 2001 }); # "... WHERE year = 2001"
282 my $new_rs = $cd_rs->search({ year => 2005 });
284 my $new_rs = $cd_rs->search([ { year => 2005 }, { year => 2004 } ]);
285 # year = 2005 OR year = 2004
287 In list context, C<< ->all() >> is called implicitly on the resultset, thus
288 returning a list of L<result|DBIx::Class::Manual::ResultClass> objects instead.
289 To avoid that, use L</search_rs>.
291 If you need to pass in additional attributes but no additional condition,
292 call it as C<search(undef, \%attrs)>.
294 # "SELECT name, artistid FROM $artist_table"
295 my @all_artists = $schema->resultset('Artist')->search(undef, {
296 columns => [qw/name artistid/],
299 For a list of attributes that can be passed to C<search>, see
300 L</ATTRIBUTES>. For more examples of using this function, see
301 L<Searching|DBIx::Class::Manual::Cookbook/Searching>. For a complete
302 documentation for the first argument, see L<SQL::Abstract>
303 and its extension L<DBIx::Class::SQLMaker>.
305 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
309 Note that L</search> does not process/deflate any of the values passed in the
310 L<SQL::Abstract>-compatible search condition structure. This is unlike other
311 condition-bound methods L</new_result>, L</create> and L</find>. The user must ensure
312 manually that any value passed to this method will stringify to something the
313 RDBMS knows how to deal with. A notable example is the handling of L<DateTime>
314 objects, for more info see:
315 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
321 my $rs = $self->search_rs( @_ );
326 elsif (defined wantarray) {
330 # we can be called by a relationship helper, which in
331 # turn may be called in void context due to some braindead
332 # overload or whatever else the user decided to be clever
333 # at this particular day. Thus limit the exception to
334 # external code calls only
335 $self->throw_exception ('->search is *not* a mutator, calling it in void context makes no sense')
336 if (caller)[0] !~ /^\QDBIx::Class::/;
346 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
348 =item Return Value: L<$resultset|/search>
352 This method does the same exact thing as search() except it will
353 always return a resultset, even in list context.
360 my $rsrc = $self->result_source;
361 my ($call_cond, $call_attrs);
363 # Special-case handling for (undef, undef) or (undef)
364 # Note that (foo => undef) is valid deprecated syntax
365 @_ = () if not scalar grep { defined $_ } @_;
371 # fish out attrs in the ($condref, $attr) case
372 elsif (@_ == 2 and ( ! defined $_[0] or (ref $_[0]) ne '') ) {
373 ($call_cond, $call_attrs) = @_;
376 $self->throw_exception('Odd number of arguments to search')
380 carp_unique 'search( %condition ) is deprecated, use search( \%condition ) instead'
381 unless $rsrc->result_class->isa('DBIx::Class::CDBICompat');
383 for my $i (0 .. $#_) {
385 $self->throw_exception ('All keys in condition key/value pairs must be plain scalars')
386 if (! defined $_[$i] or ref $_[$i] ne '');
392 # see if we can keep the cache (no $rs changes)
394 my %safe = (alias => 1, cache => 1);
395 if ( ! List::Util::first { !$safe{$_} } keys %$call_attrs and (
398 ref $call_cond eq 'HASH' && ! keys %$call_cond
400 ref $call_cond eq 'ARRAY' && ! @$call_cond
402 $cache = $self->get_cache;
405 my $old_attrs = { %{$self->{attrs}} };
406 my $old_having = delete $old_attrs->{having};
407 my $old_where = delete $old_attrs->{where};
409 my $new_attrs = { %$old_attrs };
411 # take care of call attrs (only if anything is changing)
412 if ($call_attrs and keys %$call_attrs) {
414 # copy for _normalize_selection
415 $call_attrs = { %$call_attrs };
417 my @selector_attrs = qw/select as columns cols +select +as +columns include_columns/;
419 # reset the current selector list if new selectors are supplied
420 if (List::Util::first { exists $call_attrs->{$_} } qw/columns cols select as/) {
421 delete @{$old_attrs}{(@selector_attrs, '_dark_selector')};
424 # Normalize the new selector list (operates on the passed-in attr structure)
425 # Need to do it on every chain instead of only once on _resolved_attrs, in
426 # order to allow detection of empty vs partial 'as'
427 $call_attrs->{_dark_selector} = $old_attrs->{_dark_selector}
428 if $old_attrs->{_dark_selector};
429 $self->_normalize_selection ($call_attrs);
431 # start with blind overwriting merge, exclude selector attrs
432 $new_attrs = { %{$old_attrs}, %{$call_attrs} };
433 delete @{$new_attrs}{@selector_attrs};
435 for (@selector_attrs) {
436 $new_attrs->{$_} = $self->_merge_attr($old_attrs->{$_}, $call_attrs->{$_})
437 if ( exists $old_attrs->{$_} or exists $call_attrs->{$_} );
440 # older deprecated name, use only if {columns} is not there
441 if (my $c = delete $new_attrs->{cols}) {
442 carp_unique( "Resultset attribute 'cols' is deprecated, use 'columns' instead" );
443 if ($new_attrs->{columns}) {
444 carp "Resultset specifies both the 'columns' and the legacy 'cols' attributes - ignoring 'cols'";
447 $new_attrs->{columns} = $c;
452 # join/prefetch use their own crazy merging heuristics
453 foreach my $key (qw/join prefetch/) {
454 $new_attrs->{$key} = $self->_merge_joinpref_attr($old_attrs->{$key}, $call_attrs->{$key})
455 if exists $call_attrs->{$key};
458 # stack binds together
459 $new_attrs->{bind} = [ @{ $old_attrs->{bind} || [] }, @{ $call_attrs->{bind} || [] } ];
463 for ($old_where, $call_cond) {
465 $new_attrs->{where} = $self->_stack_cond (
466 $_, $new_attrs->{where}
471 if (defined $old_having) {
472 $new_attrs->{having} = $self->_stack_cond (
473 $old_having, $new_attrs->{having}
477 my $rs = (ref $self)->new($rsrc, $new_attrs);
479 $rs->set_cache($cache) if ($cache);
485 sub _normalize_selection {
486 my ($self, $attrs) = @_;
489 if ( exists $attrs->{include_columns} ) {
490 carp_unique( "Resultset attribute 'include_columns' is deprecated, use '+columns' instead" );
491 $attrs->{'+columns'} = $self->_merge_attr(
492 $attrs->{'+columns'}, delete $attrs->{include_columns}
496 # columns are always placed first, however
498 # Keep the X vs +X separation until _resolved_attrs time - this allows to
499 # delay the decision on whether to use a default select list ($rsrc->columns)
500 # allowing stuff like the remove_columns helper to work
502 # select/as +select/+as pairs need special handling - the amount of select/as
503 # elements in each pair does *not* have to be equal (think multicolumn
504 # selectors like distinct(foo, bar) ). If the selector is bare (no 'as'
505 # supplied at all) - try to infer the alias, either from the -as parameter
506 # of the selector spec, or use the parameter whole if it looks like a column
507 # name (ugly legacy heuristic). If all fails - leave the selector bare (which
508 # is ok as well), but make sure no more additions to the 'as' chain take place
509 for my $pref ('', '+') {
511 my ($sel, $as) = map {
512 my $key = "${pref}${_}";
514 my $val = [ ref $attrs->{$key} eq 'ARRAY'
516 : $attrs->{$key} || ()
518 delete $attrs->{$key};
522 if (! @$as and ! @$sel ) {
525 elsif (@$as and ! @$sel) {
526 $self->throw_exception(
527 "Unable to handle ${pref}as specification (@$as) without a corresponding ${pref}select"
531 # no as part supplied at all - try to deduce (unless explicit end of named selection is declared)
532 # if any @$as has been supplied we assume the user knows what (s)he is doing
533 # and blindly keep stacking up pieces
534 unless ($attrs->{_dark_selector}) {
537 if ( ref $_ eq 'HASH' and exists $_->{-as} ) {
538 push @$as, $_->{-as};
540 # assume any plain no-space, no-parenthesis string to be a column spec
541 # FIXME - this is retarded but is necessary to support shit like 'count(foo)'
542 elsif ( ! ref $_ and $_ =~ /^ [^\s\(\)]+ $/x) {
545 # if all else fails - raise a flag that no more aliasing will be allowed
547 $attrs->{_dark_selector} = {
549 string => ($dark_sel_dumper ||= do {
550 require Data::Dumper::Concise;
551 Data::Dumper::Concise::DumperObject()->Indent(0);
552 })->Values([$_])->Dump
560 elsif (@$as < @$sel) {
561 $self->throw_exception(
562 "Unable to handle an ${pref}as specification (@$as) with less elements than the corresponding ${pref}select"
565 elsif ($pref and $attrs->{_dark_selector}) {
566 $self->throw_exception(
567 "Unable to process named '+select', resultset contains an unnamed selector $attrs->{_dark_selector}{string}"
573 $attrs->{"${pref}select"} = $self->_merge_attr($attrs->{"${pref}select"}, $sel);
574 $attrs->{"${pref}as"} = $self->_merge_attr($attrs->{"${pref}as"}, $as);
579 my ($self, $left, $right) = @_;
581 # collapse single element top-level conditions
582 # (single pass only, unlikely to need recursion)
583 for ($left, $right) {
584 if (ref $_ eq 'ARRAY') {
592 elsif (ref $_ eq 'HASH') {
593 my ($first, $more) = keys %$_;
596 if (! defined $first) {
600 elsif (! defined $more) {
601 if ($first eq '-and' and ref $_->{'-and'} eq 'HASH') {
604 elsif ($first eq '-or' and ref $_->{'-or'} eq 'ARRAY') {
611 # merge hashes with weeding out of duplicates (simple cases only)
612 if (ref $left eq 'HASH' and ref $right eq 'HASH') {
614 # shallow copy to destroy
615 $right = { %$right };
616 for (grep { exists $right->{$_} } keys %$left) {
617 # the use of eq_deeply here is justified - the rhs of an
618 # expression can contain a lot of twisted weird stuff
619 delete $right->{$_} if Data::Compare::Compare( $left->{$_}, $right->{$_} );
622 $right = undef unless keys %$right;
626 if (defined $left xor defined $right) {
627 return defined $left ? $left : $right;
629 elsif (! defined $left) {
633 return { -and => [ $left, $right ] };
637 =head2 search_literal
639 B<CAVEAT>: C<search_literal> is provided for Class::DBI compatibility and
640 should only be used in that context. C<search_literal> is a convenience
641 method. It is equivalent to calling C<< $schema->search(\[]) >>, but if you
642 want to ensure columns are bound correctly, use L</search>.
644 See L<DBIx::Class::Manual::Cookbook/Searching> and
645 L<DBIx::Class::Manual::FAQ/Searching> for searching techniques that do not
646 require C<search_literal>.
650 =item Arguments: $sql_fragment, @standalone_bind_values
652 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
656 my @cds = $cd_rs->search_literal('year = ? AND title = ?', qw/2001 Reload/);
657 my $newrs = $artist_rs->search_literal('name = ?', 'Metallica');
659 Pass a literal chunk of SQL to be added to the conditional part of the
662 Example of how to use C<search> instead of C<search_literal>
664 my @cds = $cd_rs->search_literal('cdid = ? AND (artist = ? OR artist = ?)', (2, 1, 2));
665 my @cds = $cd_rs->search(\[ 'cdid = ? AND (artist = ? OR artist = ?)', [ 'cdid', 2 ], [ 'artist', 1 ], [ 'artist', 2 ] ]);
670 my ($self, $sql, @bind) = @_;
672 if ( @bind && ref($bind[-1]) eq 'HASH' ) {
675 return $self->search(\[ $sql, map [ {} => $_ ], @bind ], ($attr || () ));
682 =item Arguments: \%columns_values | @pk_values, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
684 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
688 Finds and returns a single row based on supplied criteria. Takes either a
689 hashref with the same format as L</create> (including inference of foreign
690 keys from related objects), or a list of primary key values in the same
691 order as the L<primary columns|DBIx::Class::ResultSource/primary_columns>
692 declaration on the L</result_source>.
694 In either case an attempt is made to combine conditions already existing on
695 the resultset with the condition passed to this method.
697 To aid with preparing the correct query for the storage you may supply the
698 C<key> attribute, which is the name of a
699 L<unique constraint|DBIx::Class::ResultSource/add_unique_constraint> (the
700 unique constraint corresponding to the
701 L<primary columns|DBIx::Class::ResultSource/primary_columns> is always named
702 C<primary>). If the C<key> attribute has been supplied, and DBIC is unable
703 to construct a query that satisfies the named unique constraint fully (
704 non-NULL values for each column member of the constraint) an exception is
707 If no C<key> is specified, the search is carried over all unique constraints
708 which are fully defined by the available condition.
710 If no such constraint is found, C<find> currently defaults to a simple
711 C<< search->(\%column_values) >> which may or may not do what you expect.
712 Note that this fallback behavior may be deprecated in further versions. If
713 you need to search with arbitrary conditions - use L</search>. If the query
714 resulting from this fallback produces more than one row, a warning to the
715 effect is issued, though only the first row is constructed and returned as
718 In addition to C<key>, L</find> recognizes and applies standard
719 L<resultset attributes|/ATTRIBUTES> in the same way as L</search> does.
721 Note that if you have extra concerns about the correctness of the resulting
722 query you need to specify the C<key> attribute and supply the entire condition
723 as an argument to find (since it is not always possible to perform the
724 combination of the resultset condition with the supplied one, especially if
725 the resultset condition contains literal sql).
727 For example, to find a row by its primary key:
729 my $cd = $schema->resultset('CD')->find(5);
731 You can also find a row by a specific unique constraint:
733 my $cd = $schema->resultset('CD')->find(
735 artist => 'Massive Attack',
736 title => 'Mezzanine',
738 { key => 'cd_artist_title' }
741 See also L</find_or_create> and L</update_or_create>.
747 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
749 my $rsrc = $self->result_source;
752 if (exists $attrs->{key}) {
753 $constraint_name = defined $attrs->{key}
755 : $self->throw_exception("An undefined 'key' resultset attribute makes no sense")
759 # Parse out the condition from input
762 if (ref $_[0] eq 'HASH') {
763 $call_cond = { %{$_[0]} };
766 # if only values are supplied we need to default to 'primary'
767 $constraint_name = 'primary' unless defined $constraint_name;
769 my @c_cols = $rsrc->unique_constraint_columns($constraint_name);
771 $self->throw_exception(
772 "No constraint columns, maybe a malformed '$constraint_name' constraint?"
775 $self->throw_exception (
776 'find() expects either a column/value hashref, or a list of values '
777 . "corresponding to the columns of the specified unique constraint '$constraint_name'"
778 ) unless @c_cols == @_;
781 @{$call_cond}{@c_cols} = @_;
785 for my $key (keys %$call_cond) {
787 my $keyref = ref($call_cond->{$key})
789 my $relinfo = $rsrc->relationship_info($key)
791 my $val = delete $call_cond->{$key};
793 next if $keyref eq 'ARRAY'; # has_many for multi_create
795 my $rel_q = $rsrc->_resolve_condition(
796 $relinfo->{cond}, $val, $key, $key
798 die "Can't handle complex relationship conditions in find" if ref($rel_q) ne 'HASH';
799 @related{keys %$rel_q} = values %$rel_q;
803 # relationship conditions take precedence (?)
804 @{$call_cond}{keys %related} = values %related;
806 my $alias = exists $attrs->{alias} ? $attrs->{alias} : $self->{attrs}{alias};
808 if (defined $constraint_name) {
809 $final_cond = $self->_qualify_cond_columns (
811 $self->_build_unique_cond (
819 elsif ($self->{attrs}{accessor} and $self->{attrs}{accessor} eq 'single') {
820 # This means that we got here after a merger of relationship conditions
821 # in ::Relationship::Base::search_related (the row method), and furthermore
822 # the relationship is of the 'single' type. This means that the condition
823 # provided by the relationship (already attached to $self) is sufficient,
824 # as there can be only one row in the database that would satisfy the
828 # no key was specified - fall down to heuristics mode:
829 # run through all unique queries registered on the resultset, and
830 # 'OR' all qualifying queries together
831 my (@unique_queries, %seen_column_combinations);
832 for my $c_name ($rsrc->unique_constraint_names) {
833 next if $seen_column_combinations{
834 join "\x00", sort $rsrc->unique_constraint_columns($c_name)
837 push @unique_queries, try {
838 $self->_build_unique_cond ($c_name, $call_cond, 'croak_on_nulls')
842 $final_cond = @unique_queries
843 ? [ map { $self->_qualify_cond_columns($_, $alias) } @unique_queries ]
844 : $self->_non_unique_find_fallback ($call_cond, $attrs)
848 # Run the query, passing the result_class since it should propagate for find
849 my $rs = $self->search ($final_cond, {result_class => $self->result_class, %$attrs});
850 if (keys %{$rs->_resolved_attrs->{collapse}}) {
852 carp "Query returned more than one row" if $rs->next;
860 # This is a stop-gap method as agreed during the discussion on find() cleanup:
861 # http://lists.scsys.co.uk/pipermail/dbix-class/2010-October/009535.html
863 # It is invoked when find() is called in legacy-mode with insufficiently-unique
864 # condition. It is provided for overrides until a saner way forward is devised
866 # *NOTE* This is not a public method, and it's *GUARANTEED* to disappear down
867 # the road. Please adjust your tests accordingly to catch this situation early
868 # DBIx::Class::ResultSet->can('_non_unique_find_fallback') is reasonable
870 # The method will not be removed without an adequately complete replacement
871 # for strict-mode enforcement
872 sub _non_unique_find_fallback {
873 my ($self, $cond, $attrs) = @_;
875 return $self->_qualify_cond_columns(
877 exists $attrs->{alias}
879 : $self->{attrs}{alias}
884 sub _qualify_cond_columns {
885 my ($self, $cond, $alias) = @_;
887 my %aliased = %$cond;
888 for (keys %aliased) {
889 $aliased{"$alias.$_"} = delete $aliased{$_}
896 sub _build_unique_cond {
897 my ($self, $constraint_name, $extra_cond, $croak_on_null) = @_;
899 my @c_cols = $self->result_source->unique_constraint_columns($constraint_name);
901 # combination may fail if $self->{cond} is non-trivial
902 my ($final_cond) = try {
903 $self->_merge_with_rscond ($extra_cond)
908 # trim out everything not in $columns
909 $final_cond = { map {
910 exists $final_cond->{$_}
911 ? ( $_ => $final_cond->{$_} )
915 if (my @missing = grep
916 { ! ($croak_on_null ? defined $final_cond->{$_} : exists $final_cond->{$_}) }
919 $self->throw_exception( sprintf ( "Unable to satisfy requested constraint '%s', no values for column(s): %s",
921 join (', ', map { "'$_'" } @missing),
928 !$ENV{DBIC_NULLABLE_KEY_NOWARN}
930 my @undefs = sort grep { ! defined $final_cond->{$_} } (keys %$final_cond)
932 carp_unique ( sprintf (
933 "NULL/undef values supplied for requested unique constraint '%s' (NULL "
934 . 'values in column(s): %s). This is almost certainly not what you wanted, '
935 . 'though you can set DBIC_NULLABLE_KEY_NOWARN to disable this warning.',
937 join (', ', map { "'$_'" } @undefs),
944 =head2 search_related
948 =item Arguments: $rel_name, $cond?, L<\%attrs?|/ATTRIBUTES>
950 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
954 $new_rs = $cd_rs->search_related('artist', {
958 Searches the specified relationship, optionally specifying a condition and
959 attributes for matching records. See L</ATTRIBUTES> for more information.
961 In list context, C<< ->all() >> is called implicitly on the resultset, thus
962 returning a list of result objects instead. To avoid that, use L</search_related_rs>.
964 See also L</search_related_rs>.
969 return shift->related_resultset(shift)->search(@_);
972 =head2 search_related_rs
974 This method works exactly the same as search_related, except that
975 it guarantees a resultset, even in list context.
979 sub search_related_rs {
980 return shift->related_resultset(shift)->search_rs(@_);
987 =item Arguments: none
989 =item Return Value: L<$cursor|DBIx::Class::Cursor>
993 Returns a storage-driven cursor to the given resultset. See
994 L<DBIx::Class::Cursor> for more information.
1001 return $self->{cursor} ||= do {
1002 my $attrs = { %{$self->_resolved_attrs } };
1003 $self->result_source->storage->select(
1004 $attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs
1013 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1015 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1019 my $cd = $schema->resultset('CD')->single({ year => 2001 });
1021 Inflates the first result without creating a cursor if the resultset has
1022 any records in it; if not returns C<undef>. Used by L</find> as a lean version
1025 While this method can take an optional search condition (just like L</search>)
1026 being a fast-code-path it does not recognize search attributes. If you need to
1027 add extra joins or similar, call L</search> and then chain-call L</single> on the
1028 L<DBIx::Class::ResultSet> returned.
1034 As of 0.08100, this method enforces the assumption that the preceding
1035 query returns only one row. If more than one row is returned, you will receive
1038 Query returned more than one row
1040 In this case, you should be using L</next> or L</find> instead, or if you really
1041 know what you are doing, use the L</rows> attribute to explicitly limit the size
1044 This method will also throw an exception if it is called on a resultset prefetching
1045 has_many, as such a prefetch implies fetching multiple rows from the database in
1046 order to assemble the resulting object.
1053 my ($self, $where) = @_;
1055 $self->throw_exception('single() only takes search conditions, no attributes. You want ->search( $cond, $attrs )->single()');
1058 my $attrs = { %{$self->_resolved_attrs} };
1060 if (keys %{$attrs->{collapse}}) {
1061 $self->throw_exception(
1062 'single() can not be used on resultsets prefetching has_many. Use find( \%cond ) or next() instead'
1067 if (defined $attrs->{where}) {
1070 [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ }
1071 $where, delete $attrs->{where} ]
1074 $attrs->{where} = $where;
1078 my @data = $self->result_source->storage->select_single(
1079 $attrs->{from}, $attrs->{select},
1080 $attrs->{where}, $attrs
1083 return (@data ? ($self->_construct_object(@data))[0] : undef);
1089 # Recursively collapse the query, accumulating values for each column.
1091 sub _collapse_query {
1092 my ($self, $query, $collapsed) = @_;
1096 if (ref $query eq 'ARRAY') {
1097 foreach my $subquery (@$query) {
1098 next unless ref $subquery; # -or
1099 $collapsed = $self->_collapse_query($subquery, $collapsed);
1102 elsif (ref $query eq 'HASH') {
1103 if (keys %$query and (keys %$query)[0] eq '-and') {
1104 foreach my $subquery (@{$query->{-and}}) {
1105 $collapsed = $self->_collapse_query($subquery, $collapsed);
1109 foreach my $col (keys %$query) {
1110 my $value = $query->{$col};
1111 $collapsed->{$col}{$value}++;
1123 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1125 =item Return Value: L<$resultsetcolumn|DBIx::Class::ResultSetColumn>
1129 my $max_length = $rs->get_column('length')->max;
1131 Returns a L<DBIx::Class::ResultSetColumn> instance for a column of the ResultSet.
1136 my ($self, $column) = @_;
1137 my $new = DBIx::Class::ResultSetColumn->new($self, $column);
1145 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1147 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1151 # WHERE title LIKE '%blue%'
1152 $cd_rs = $rs->search_like({ title => '%blue%'});
1154 Performs a search, but uses C<LIKE> instead of C<=> as the condition. Note
1155 that this is simply a convenience method retained for ex Class::DBI users.
1156 You most likely want to use L</search> with specific operators.
1158 For more information, see L<DBIx::Class::Manual::Cookbook>.
1160 This method is deprecated and will be removed in 0.09. Use L</search()>
1161 instead. An example conversion is:
1163 ->search_like({ foo => 'bar' });
1167 ->search({ foo => { like => 'bar' } });
1174 'search_like() is deprecated and will be removed in DBIC version 0.09.'
1175 .' Instead use ->search({ x => { -like => "y%" } })'
1176 .' (note the outer pair of {}s - they are important!)'
1178 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
1179 my $query = ref $_[0] eq 'HASH' ? { %{shift()} }: {@_};
1180 $query->{$_} = { 'like' => $query->{$_} } for keys %$query;
1181 return $class->search($query, { %$attrs });
1188 =item Arguments: $first, $last
1190 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1194 Returns a resultset or object list representing a subset of elements from the
1195 resultset slice is called on. Indexes are from 0, i.e., to get the first
1196 three records, call:
1198 my ($one, $two, $three) = $rs->slice(0, 2);
1203 my ($self, $min, $max) = @_;
1204 my $attrs = {}; # = { %{ $self->{attrs} || {} } };
1205 $attrs->{offset} = $self->{attrs}{offset} || 0;
1206 $attrs->{offset} += $min;
1207 $attrs->{rows} = ($max ? ($max - $min + 1) : 1);
1208 return $self->search(undef, $attrs);
1209 #my $slice = (ref $self)->new($self->result_source, $attrs);
1210 #return (wantarray ? $slice->all : $slice);
1217 =item Arguments: none
1219 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1223 Returns the next element in the resultset (C<undef> is there is none).
1225 Can be used to efficiently iterate over records in the resultset:
1227 my $rs = $schema->resultset('CD')->search;
1228 while (my $cd = $rs->next) {
1232 Note that you need to store the resultset object, and call C<next> on it.
1233 Calling C<< resultset('Table')->next >> repeatedly will always return the
1234 first record from the resultset.
1240 if (my $cache = $self->get_cache) {
1241 $self->{all_cache_position} ||= 0;
1242 return $cache->[$self->{all_cache_position}++];
1244 if ($self->{attrs}{cache}) {
1245 delete $self->{pager};
1246 $self->{all_cache_position} = 1;
1247 return ($self->all)[0];
1249 if ($self->{stashed_objects}) {
1250 my $obj = shift(@{$self->{stashed_objects}});
1251 delete $self->{stashed_objects} unless @{$self->{stashed_objects}};
1255 exists $self->{stashed_row}
1256 ? @{delete $self->{stashed_row}}
1257 : $self->cursor->next
1259 return undef unless (@row);
1260 my ($row, @more) = $self->_construct_object(@row);
1261 $self->{stashed_objects} = \@more if @more;
1265 sub _construct_object {
1266 my ($self, @row) = @_;
1268 my $info = $self->_collapse_result($self->{_attrs}{as}, \@row)
1270 my @new = $self->result_class->inflate_result($self->result_source, @$info);
1271 @new = $self->{_attrs}{record_filter}->(@new)
1272 if exists $self->{_attrs}{record_filter};
1276 sub _collapse_result {
1277 my ($self, $as_proto, $row) = @_;
1281 # 'foo' => [ undef, 'foo' ]
1282 # 'foo.bar' => [ 'foo', 'bar' ]
1283 # 'foo.bar.baz' => [ 'foo.bar', 'baz' ]
1285 my @construct_as = map { [ (/^(?:(.*)\.)?([^.]+)$/) ] } @$as_proto;
1287 my %collapse = %{$self->{_attrs}{collapse}||{}};
1291 # if we're doing collapsing (has_many prefetch) we need to grab records
1292 # until the PK changes, so fill @pri_index. if not, we leave it empty so
1293 # we know we don't have to bother.
1295 # the reason for not using the collapse stuff directly is because if you
1296 # had for e.g. two artists in a row with no cds, the collapse info for
1297 # both would be NULL (undef) so you'd lose the second artist
1299 # store just the index so we can check the array positions from the row
1300 # without having to contruct the full hash
1302 if (keys %collapse) {
1303 my %pri = map { ($_ => 1) } $self->result_source->_pri_cols;
1304 foreach my $i (0 .. $#construct_as) {
1305 next if defined($construct_as[$i][0]); # only self table
1306 if (delete $pri{$construct_as[$i][1]}) {
1307 push(@pri_index, $i);
1309 last unless keys %pri; # short circuit (Johnny Five Is Alive!)
1313 # no need to do an if, it'll be empty if @pri_index is empty anyway
1315 my %pri_vals = map { ($_ => $copy[$_]) } @pri_index;
1319 do { # no need to check anything at the front, we always want the first row
1323 foreach my $this_as (@construct_as) {
1324 $const{$this_as->[0]||''}{$this_as->[1]} = shift(@copy);
1327 push(@const_rows, \%const);
1329 } until ( # no pri_index => no collapse => drop straight out
1332 do { # get another row, stash it, drop out if different PK
1334 @copy = $self->cursor->next;
1335 $self->{stashed_row} = \@copy;
1337 # last thing in do block, counts as true if anything doesn't match
1339 # check xor defined first for NULL vs. NOT NULL then if one is
1340 # defined the other must be so check string equality
1343 (defined $pri_vals{$_} ^ defined $copy[$_])
1344 || (defined $pri_vals{$_} && ($pri_vals{$_} ne $copy[$_]))
1349 my $alias = $self->{attrs}{alias};
1356 foreach my $const (@const_rows) {
1357 scalar @const_keys or do {
1358 @const_keys = sort { length($a) <=> length($b) } keys %$const;
1360 foreach my $key (@const_keys) {
1363 my @parts = split(/\./, $key);
1365 my $data = $const->{$key};
1366 foreach my $p (@parts) {
1367 $target = $target->[1]->{$p} ||= [];
1369 if ($cur eq ".${key}" && (my @ckey = @{$collapse{$cur}||[]})) {
1370 # collapsing at this point and on final part
1371 my $pos = $collapse_pos{$cur};
1372 CK: foreach my $ck (@ckey) {
1373 if (!defined $pos->{$ck} || $pos->{$ck} ne $data->{$ck}) {
1374 $collapse_pos{$cur} = $data;
1375 delete @collapse_pos{ # clear all positioning for sub-entries
1376 grep { m/^\Q${cur}.\E/ } keys %collapse_pos
1383 if (exists $collapse{$cur}) {
1384 $target = $target->[-1];
1387 $target->[0] = $data;
1389 $info->[0] = $const->{$key};
1397 =head2 result_source
1401 =item Arguments: L<$result_source?|DBIx::Class::ResultSource>
1403 =item Return Value: L<$result_source|DBIx::Class::ResultSource>
1407 An accessor for the primary ResultSource object from which this ResultSet
1414 =item Arguments: $result_class?
1416 =item Return Value: $result_class
1420 An accessor for the class to use when creating result objects. Defaults to
1421 C<< result_source->result_class >> - which in most cases is the name of the
1422 L<"table"|DBIx::Class::Manual::Glossary/"ResultSource"> class.
1424 Note that changing the result_class will also remove any components
1425 that were originally loaded in the source class via
1426 L<DBIx::Class::ResultSource/load_components>. Any overloaded methods
1427 in the original source class will not run.
1432 my ($self, $result_class) = @_;
1433 if ($result_class) {
1434 unless (ref $result_class) { # don't fire this for an object
1435 $self->ensure_class_loaded($result_class);
1437 $self->_result_class($result_class);
1438 # THIS LINE WOULD BE A BUG - this accessor specifically exists to
1439 # permit the user to set result class on one result set only; it only
1440 # chains if provided to search()
1441 #$self->{attrs}{result_class} = $result_class if ref $self;
1443 $self->_result_class;
1450 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1452 =item Return Value: $count
1456 Performs an SQL C<COUNT> with the same query as the resultset was built
1457 with to find the number of elements. Passing arguments is equivalent to
1458 C<< $rs->search ($cond, \%attrs)->count >>
1464 return $self->search(@_)->count if @_ and defined $_[0];
1465 return scalar @{ $self->get_cache } if $self->get_cache;
1467 my $attrs = { %{ $self->_resolved_attrs } };
1469 # this is a little optimization - it is faster to do the limit
1470 # adjustments in software, instead of a subquery
1471 my $rows = delete $attrs->{rows};
1472 my $offset = delete $attrs->{offset};
1475 if ($self->_has_resolved_attr (qw/collapse group_by/)) {
1476 $crs = $self->_count_subq_rs ($attrs);
1479 $crs = $self->_count_rs ($attrs);
1481 my $count = $crs->next;
1483 $count -= $offset if $offset;
1484 $count = $rows if $rows and $rows < $count;
1485 $count = 0 if ($count < 0);
1494 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1496 =item Return Value: L<$count_rs|DBIx::Class::ResultSetColumn>
1500 Same as L</count> but returns a L<DBIx::Class::ResultSetColumn> object.
1501 This can be very handy for subqueries:
1503 ->search( { amount => $some_rs->count_rs->as_query } )
1505 As with regular resultsets the SQL query will be executed only after
1506 the resultset is accessed via L</next> or L</all>. That would return
1507 the same single value obtainable via L</count>.
1513 return $self->search(@_)->count_rs if @_;
1515 # this may look like a lack of abstraction (count() does about the same)
1516 # but in fact an _rs *must* use a subquery for the limits, as the
1517 # software based limiting can not be ported if this $rs is to be used
1518 # in a subquery itself (i.e. ->as_query)
1519 if ($self->_has_resolved_attr (qw/collapse group_by offset rows/)) {
1520 return $self->_count_subq_rs;
1523 return $self->_count_rs;
1528 # returns a ResultSetColumn object tied to the count query
1531 my ($self, $attrs) = @_;
1533 my $rsrc = $self->result_source;
1534 $attrs ||= $self->_resolved_attrs;
1536 my $tmp_attrs = { %$attrs };
1537 # take off any limits, record_filter is cdbi, and no point of ordering nor locking a count
1538 delete @{$tmp_attrs}{qw/rows offset order_by record_filter for/};
1540 # overwrite the selector (supplied by the storage)
1541 $tmp_attrs->{select} = $rsrc->storage->_count_select ($rsrc, $attrs);
1542 $tmp_attrs->{as} = 'count';
1543 delete @{$tmp_attrs}{qw/columns/};
1545 my $tmp_rs = $rsrc->resultset_class->new($rsrc, $tmp_attrs)->get_column ('count');
1551 # same as above but uses a subquery
1553 sub _count_subq_rs {
1554 my ($self, $attrs) = @_;
1556 my $rsrc = $self->result_source;
1557 $attrs ||= $self->_resolved_attrs;
1559 my $sub_attrs = { %$attrs };
1560 # extra selectors do not go in the subquery and there is no point of ordering it, nor locking it
1561 delete @{$sub_attrs}{qw/collapse columns as select _prefetch_selector_range order_by for/};
1563 # if we multi-prefetch we group_by something unique, as this is what we would
1564 # get out of the rs via ->next/->all. We *DO WANT* to clobber old group_by regardless
1565 if ( keys %{$attrs->{collapse}} ) {
1566 $sub_attrs->{group_by} = [ map { "$attrs->{alias}.$_" } @{
1567 $rsrc->_identifying_column_set || $self->throw_exception(
1568 'Unable to construct a unique group_by criteria properly collapsing the '
1569 . 'has_many prefetch before count()'
1574 # Calculate subquery selector
1575 if (my $g = $sub_attrs->{group_by}) {
1577 my $sql_maker = $rsrc->storage->sql_maker;
1579 # necessary as the group_by may refer to aliased functions
1581 for my $sel (@{$attrs->{select}}) {
1582 $sel_index->{$sel->{-as}} = $sel
1583 if (ref $sel eq 'HASH' and $sel->{-as});
1586 # anything from the original select mentioned on the group-by needs to make it to the inner selector
1587 # also look for named aggregates referred in the having clause
1588 # having often contains scalarrefs - thus parse it out entirely
1590 if ($attrs->{having}) {
1591 local $sql_maker->{having_bind};
1592 local $sql_maker->{quote_char} = $sql_maker->{quote_char};
1593 local $sql_maker->{name_sep} = $sql_maker->{name_sep};
1594 unless (defined $sql_maker->{quote_char} and length $sql_maker->{quote_char}) {
1595 $sql_maker->{quote_char} = [ "\x00", "\xFF" ];
1596 # if we don't unset it we screw up retarded but unfortunately working
1597 # 'MAX(foo.bar)' => { '>', 3 }
1598 $sql_maker->{name_sep} = '';
1601 my ($lquote, $rquote, $sep) = map { quotemeta $_ } ($sql_maker->_quote_chars, $sql_maker->name_sep);
1603 my $having_sql = $sql_maker->_parse_rs_attrs ({ having => $attrs->{having} });
1606 # search for both a proper quoted qualified string, for a naive unquoted scalarref
1607 # and if all fails for an utterly naive quoted scalar-with-function
1608 while ($having_sql =~ /
1609 $rquote $sep $lquote (.+?) $rquote
1611 [\s,] \w+ \. (\w+) [\s,]
1613 [\s,] $lquote (.+?) $rquote [\s,]
1615 my $part = $1 || $2 || $3; # one of them matched if we got here
1616 unless ($seen_having{$part}++) {
1623 my $colpiece = $sel_index->{$_} || $_;
1625 # unqualify join-based group_by's. Arcane but possible query
1626 # also horrible horrible hack to alias a column (not a func.)
1627 # (probably need to introduce SQLA syntax)
1628 if ($colpiece =~ /\./ && $colpiece !~ /^$attrs->{alias}\./) {
1631 $colpiece = \ sprintf ('%s AS %s', map { $sql_maker->_quote ($_) } ($colpiece, $as) );
1633 push @{$sub_attrs->{select}}, $colpiece;
1637 my @pcols = map { "$attrs->{alias}.$_" } ($rsrc->primary_columns);
1638 $sub_attrs->{select} = @pcols ? \@pcols : [ 1 ];
1641 return $rsrc->resultset_class
1642 ->new ($rsrc, $sub_attrs)
1644 ->search ({}, { columns => { count => $rsrc->storage->_count_select ($rsrc, $attrs) } })
1645 ->get_column ('count');
1649 =head2 count_literal
1651 B<CAVEAT>: C<count_literal> is provided for Class::DBI compatibility and
1652 should only be used in that context. See L</search_literal> for further info.
1656 =item Arguments: $sql_fragment, @standalone_bind_values
1658 =item Return Value: $count
1662 Counts the results in a literal query. Equivalent to calling L</search_literal>
1663 with the passed arguments, then L</count>.
1667 sub count_literal { shift->search_literal(@_)->count; }
1673 =item Arguments: none
1675 =item Return Value: L<@result_objs|DBIx::Class::Manual::ResultClass>
1679 Returns all elements in the resultset.
1686 $self->throw_exception("all() doesn't take any arguments, you probably wanted ->search(...)->all()");
1689 return @{ $self->get_cache } if $self->get_cache;
1693 if (keys %{$self->_resolved_attrs->{collapse}}) {
1694 # Using $self->cursor->all is really just an optimisation.
1695 # If we're collapsing has_many prefetches it probably makes
1696 # very little difference, and this is cleaner than hacking
1697 # _construct_object to survive the approach
1698 $self->cursor->reset;
1699 my @row = $self->cursor->next;
1701 push(@obj, $self->_construct_object(@row));
1702 @row = (exists $self->{stashed_row}
1703 ? @{delete $self->{stashed_row}}
1704 : $self->cursor->next);
1707 @obj = map { $self->_construct_object(@$_) } $self->cursor->all;
1710 $self->set_cache(\@obj) if $self->{attrs}{cache};
1719 =item Arguments: none
1721 =item Return Value: $self
1725 Resets the resultset's cursor, so you can iterate through the elements again.
1726 Implicitly resets the storage cursor, so a subsequent L</next> will trigger
1733 $self->{all_cache_position} = 0;
1734 $self->cursor->reset;
1742 =item Arguments: none
1744 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1748 L<Resets|/reset> the resultset (causing a fresh query to storage) and returns
1749 an object for the first result (or C<undef> if the resultset is empty).
1754 return $_[0]->reset->next;
1760 # Determines whether and what type of subquery is required for the $rs operation.
1761 # If grouping is necessary either supplies its own, or verifies the current one
1762 # After all is done delegates to the proper storage method.
1764 sub _rs_update_delete {
1765 my ($self, $op, $values) = @_;
1767 my $rsrc = $self->result_source;
1768 my $storage = $rsrc->schema->storage;
1770 my $attrs = { %{$self->_resolved_attrs} };
1772 my $join_classifications;
1773 my $existing_group_by = delete $attrs->{group_by};
1775 # do we need a subquery for any reason?
1777 defined $existing_group_by
1779 # if {from} is unparseable wrap a subq
1780 ref($attrs->{from}) ne 'ARRAY'
1782 # limits call for a subq
1783 $self->_has_resolved_attr(qw/rows offset/)
1786 # simplify the joinmap, so we can further decide if a subq is necessary
1787 if (!$needs_subq and @{$attrs->{from}} > 1) {
1788 $attrs->{from} = $storage->_prune_unused_joins ($attrs->{from}, $attrs->{select}, $self->{cond}, $attrs);
1790 # check if there are any joins left after the prune
1791 if ( @{$attrs->{from}} > 1 ) {
1792 $join_classifications = $storage->_resolve_aliastypes_from_select_args (
1793 [ @{$attrs->{from}}[1 .. $#{$attrs->{from}}] ],
1799 # any non-pruneable joins imply subq
1800 $needs_subq = scalar keys %{ $join_classifications->{restricting} || {} };
1804 # check if the head is composite (by now all joins are thrown out unless $needs_subq)
1806 (ref $attrs->{from}[0]) ne 'HASH'
1808 ref $attrs->{from}[0]{ $attrs->{from}[0]{-alias} }
1812 # do we need anything like a subquery?
1813 if (! $needs_subq) {
1814 # Most databases do not allow aliasing of tables in UPDATE/DELETE. Thus
1815 # a condition containing 'me' or other table prefixes will not work
1816 # at all. Tell SQLMaker to dequalify idents via a gross hack.
1818 my $sqla = $rsrc->storage->sql_maker;
1819 local $sqla->{_dequalify_idents} = 1;
1820 \[ $sqla->_recurse_where($self->{cond}) ];
1824 # we got this far - means it is time to wrap a subquery
1825 my $idcols = $rsrc->_identifying_column_set || $self->throw_exception(
1827 "Unable to perform complex resultset %s() without an identifying set of columns on source '%s'",
1833 # make a new $rs selecting only the PKs (that's all we really need for the subq)
1834 delete $attrs->{$_} for qw/collapse _collapse_order_by select _prefetch_selector_range as/;
1835 $attrs->{columns} = [ map { "$attrs->{alias}.$_" } @$idcols ];
1836 $attrs->{group_by} = \ ''; # FIXME - this is an evil hack, it causes the optimiser to kick in and throw away the LEFT joins
1837 my $subrs = (ref $self)->new($rsrc, $attrs);
1839 if (@$idcols == 1) {
1840 $cond = { $idcols->[0] => { -in => $subrs->as_query } };
1842 elsif ($storage->_use_multicolumn_in) {
1843 # no syntax for calling this properly yet
1844 # !!! EXPERIMENTAL API !!! WILL CHANGE !!!
1845 $cond = $storage->sql_maker->_where_op_multicolumn_in (
1846 $idcols, # how do I convey a list of idents...? can binds reside on lhs?
1851 # if all else fails - get all primary keys and operate over a ORed set
1852 # wrap in a transaction for consistency
1853 # this is where the group_by/multiplication starts to matter
1857 keys %{ $join_classifications->{multiplying} || {} }
1859 # make sure if there is a supplied group_by it matches the columns compiled above
1860 # perfectly. Anything else can not be sanely executed on most databases so croak
1861 # right then and there
1862 if ($existing_group_by) {
1863 my @current_group_by = map
1864 { $_ =~ /\./ ? $_ : "$attrs->{alias}.$_" }
1869 join ("\x00", sort @current_group_by)
1871 join ("\x00", sort @{$attrs->{columns}} )
1873 $self->throw_exception (
1874 "You have just attempted a $op operation on a resultset which does group_by"
1875 . ' on columns other than the primary keys, while DBIC internally needs to retrieve'
1876 . ' the primary keys in a subselect. All sane RDBMS engines do not support this'
1877 . ' kind of queries. Please retry the operation with a modified group_by or'
1878 . ' without using one at all.'
1883 $subrs = $subrs->search({}, { group_by => $attrs->{columns} });
1886 $guard = $storage->txn_scope_guard;
1889 for my $row ($subrs->cursor->all) {
1891 { $idcols->[$_] => $row->[$_] }
1898 my $res = $storage->$op (
1900 $op eq 'update' ? $values : (),
1904 $guard->commit if $guard;
1913 =item Arguments: \%values
1915 =item Return Value: $underlying_storage_rv
1919 Sets the specified columns in the resultset to the supplied values in a
1920 single query. Note that this will not run any accessor/set_column/update
1921 triggers, nor will it update any result object instances derived from this
1922 resultset (this includes the contents of the L<resultset cache|/set_cache>
1923 if any). See L</update_all> if you need to execute any on-update
1924 triggers or cascades defined either by you or a
1925 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
1927 The return value is a pass through of what the underlying
1928 storage backend returned, and may vary. See L<DBI/execute> for the most
1933 Note that L</update> does not process/deflate any of the values passed in.
1934 This is unlike the corresponding L<DBIx::Class::Row/update>. The user must
1935 ensure manually that any value passed to this method will stringify to
1936 something the RDBMS knows how to deal with. A notable example is the
1937 handling of L<DateTime> objects, for more info see:
1938 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
1943 my ($self, $values) = @_;
1944 $self->throw_exception('Values for update must be a hash')
1945 unless ref $values eq 'HASH';
1947 return $self->_rs_update_delete ('update', $values);
1954 =item Arguments: \%values
1956 =item Return Value: 1
1960 Fetches all objects and updates them one at a time via
1961 L<DBIx::Class::Row/update>. Note that C<update_all> will run DBIC defined
1962 triggers, while L</update> will not.
1967 my ($self, $values) = @_;
1968 $self->throw_exception('Values for update_all must be a hash')
1969 unless ref $values eq 'HASH';
1971 my $guard = $self->result_source->schema->txn_scope_guard;
1972 $_->update({%$values}) for $self->all; # shallow copy - update will mangle it
1981 =item Arguments: none
1983 =item Return Value: $underlying_storage_rv
1987 Deletes the rows matching this resultset in a single query. Note that this
1988 will not run any delete triggers, nor will it alter the
1989 L<in_storage|DBIx::Class::Row/in_storage> status of any result object instances
1990 derived from this resultset (this includes the contents of the
1991 L<resultset cache|/set_cache> if any). See L</delete_all> if you need to
1992 execute any on-delete triggers or cascades defined either by you or a
1993 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
1995 The return value is a pass through of what the underlying storage backend
1996 returned, and may vary. See L<DBI/execute> for the most common case.
2002 $self->throw_exception('delete does not accept any arguments')
2005 return $self->_rs_update_delete ('delete');
2012 =item Arguments: none
2014 =item Return Value: 1
2018 Fetches all objects and deletes them one at a time via
2019 L<DBIx::Class::Row/delete>. Note that C<delete_all> will run DBIC defined
2020 triggers, while L</delete> will not.
2026 $self->throw_exception('delete_all does not accept any arguments')
2029 my $guard = $self->result_source->schema->txn_scope_guard;
2030 $_->delete for $self->all;
2039 =item Arguments: [ \@column_list, \@row_values+ ] | [ \%col_data+ ]
2041 =item Return Value: L<\@result_objects|DBIx::Class::Manual::ResultClass> (scalar context) | L<@result_objects|DBIx::Class::Manual::ResultClass> (list context)
2045 Accepts either an arrayref of hashrefs or alternatively an arrayref of
2052 The context of this method call has an important effect on what is
2053 submitted to storage. In void context data is fed directly to fastpath
2054 insertion routines provided by the underlying storage (most often
2055 L<DBI/execute_for_fetch>), bypassing the L<new|DBIx::Class::Row/new> and
2056 L<insert|DBIx::Class::Row/insert> calls on the
2057 L<Result|DBIx::Class::Manual::ResultClass> class, including any
2058 augmentation of these methods provided by components. For example if you
2059 are using something like L<DBIx::Class::UUIDColumns> to create primary
2060 keys for you, you will find that your PKs are empty. In this case you
2061 will have to explicitly force scalar or list context in order to create
2066 In non-void (scalar or list) context, this method is simply a wrapper
2067 for L</create>. Depending on list or scalar context either a list of
2068 L<Result|DBIx::Class::Manual::ResultClass> objects or an arrayref
2069 containing these objects is returned.
2071 When supplying data in "arrayref of arrayrefs" invocation style, the
2072 first element should be a list of column names and each subsequent
2073 element should be a data value in the earlier specified column order.
2076 $Arstist_rs->populate([
2077 [ qw( artistid name ) ],
2078 [ 100, 'A Formally Unknown Singer' ],
2079 [ 101, 'A singer that jumped the shark two albums ago' ],
2080 [ 102, 'An actually cool singer' ],
2083 For the arrayref of hashrefs style each hashref should be a structure
2084 suitable for passing to L</create>. Multi-create is also permitted with
2087 $schema->resultset("Artist")->populate([
2088 { artistid => 4, name => 'Manufactured Crap', cds => [
2089 { title => 'My First CD', year => 2006 },
2090 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2093 { artistid => 5, name => 'Angsty-Whiny Girl', cds => [
2094 { title => 'My parents sold me to a record company', year => 2005 },
2095 { title => 'Why Am I So Ugly?', year => 2006 },
2096 { title => 'I Got Surgery and am now Popular', year => 2007 }
2101 If you attempt a void-context multi-create as in the example above (each
2102 Artist also has the related list of CDs), and B<do not> supply the
2103 necessary autoinc foreign key information, this method will proxy to the
2104 less efficient L</create>, and then throw the Result objects away. In this
2105 case there are obviously no benefits to using this method over L</create>.
2112 # cruft placed in standalone method
2113 my $data = $self->_normalize_populate_args(@_);
2115 return unless @$data;
2117 if(defined wantarray) {
2118 my @created = map { $self->create($_) } @$data;
2119 return wantarray ? @created : \@created;
2122 my $first = $data->[0];
2124 # if a column is a registered relationship, and is a non-blessed hash/array, consider
2125 # it relationship data
2126 my (@rels, @columns);
2127 my $rsrc = $self->result_source;
2128 my $rels = { map { $_ => $rsrc->relationship_info($_) } $rsrc->relationships };
2129 for (keys %$first) {
2130 my $ref = ref $first->{$_};
2131 $rels->{$_} && ($ref eq 'ARRAY' or $ref eq 'HASH')
2137 my @pks = $rsrc->primary_columns;
2139 ## do the belongs_to relationships
2140 foreach my $index (0..$#$data) {
2142 # delegate to create() for any dataset without primary keys with specified relationships
2143 if (grep { !defined $data->[$index]->{$_} } @pks ) {
2145 if (grep { ref $data->[$index]{$r} eq $_ } qw/HASH ARRAY/) { # a related set must be a HASH or AoH
2146 my @ret = $self->populate($data);
2152 foreach my $rel (@rels) {
2153 next unless ref $data->[$index]->{$rel} eq "HASH";
2154 my $result = $self->related_resultset($rel)->create($data->[$index]->{$rel});
2155 my ($reverse_relname, $reverse_relinfo) = %{$rsrc->reverse_relationship_info($rel)};
2156 my $related = $result->result_source->_resolve_condition(
2157 $reverse_relinfo->{cond},
2163 delete $data->[$index]->{$rel};
2164 $data->[$index] = {%{$data->[$index]}, %$related};
2166 push @columns, keys %$related if $index == 0;
2170 ## inherit the data locked in the conditions of the resultset
2171 my ($rs_data) = $self->_merge_with_rscond({});
2172 delete @{$rs_data}{@columns};
2174 ## do bulk insert on current row
2175 $rsrc->storage->insert_bulk(
2177 [@columns, keys %$rs_data],
2178 [ map { [ @$_{@columns}, values %$rs_data ] } @$data ],
2181 ## do the has_many relationships
2182 foreach my $item (@$data) {
2186 foreach my $rel (@rels) {
2187 next unless ref $item->{$rel} eq "ARRAY" && @{ $item->{$rel} };
2189 $main_row ||= $self->new_result({map { $_ => $item->{$_} } @pks});
2191 my $child = $main_row->$rel;
2193 my $related = $child->result_source->_resolve_condition(
2194 $rels->{$rel}{cond},
2200 my @rows_to_add = ref $item->{$rel} eq 'ARRAY' ? @{$item->{$rel}} : ($item->{$rel});
2201 my @populate = map { {%$_, %$related} } @rows_to_add;
2203 $child->populate( \@populate );
2210 # populate() argumnets went over several incarnations
2211 # What we ultimately support is AoH
2212 sub _normalize_populate_args {
2213 my ($self, $arg) = @_;
2215 if (ref $arg eq 'ARRAY') {
2219 elsif (ref $arg->[0] eq 'HASH') {
2222 elsif (ref $arg->[0] eq 'ARRAY') {
2224 my @colnames = @{$arg->[0]};
2225 foreach my $values (@{$arg}[1 .. $#$arg]) {
2226 push @ret, { map { $colnames[$_] => $values->[$_] } (0 .. $#colnames) };
2232 $self->throw_exception('Populate expects an arrayref of hashrefs or arrayref of arrayrefs');
2239 =item Arguments: none
2241 =item Return Value: L<$pager|Data::Page>
2245 Returns a L<Data::Page> object for the current resultset. Only makes
2246 sense for queries with a C<page> attribute.
2248 To get the full count of entries for a paged resultset, call
2249 C<total_entries> on the L<Data::Page> object.
2256 return $self->{pager} if $self->{pager};
2258 my $attrs = $self->{attrs};
2259 if (!defined $attrs->{page}) {
2260 $self->throw_exception("Can't create pager for non-paged rs");
2262 elsif ($attrs->{page} <= 0) {
2263 $self->throw_exception('Invalid page number (page-numbers are 1-based)');
2265 $attrs->{rows} ||= 10;
2267 # throw away the paging flags and re-run the count (possibly
2268 # with a subselect) to get the real total count
2269 my $count_attrs = { %$attrs };
2270 delete $count_attrs->{$_} for qw/rows offset page pager/;
2272 my $total_rs = (ref $self)->new($self->result_source, $count_attrs);
2274 require DBIx::Class::ResultSet::Pager;
2275 return $self->{pager} = DBIx::Class::ResultSet::Pager->new(
2276 sub { $total_rs->count }, #lazy-get the total
2278 $self->{attrs}{page},
2286 =item Arguments: $page_number
2288 =item Return Value: L<$resultset|/search>
2292 Returns a resultset for the $page_number page of the resultset on which page
2293 is called, where each page contains a number of rows equal to the 'rows'
2294 attribute set on the resultset (10 by default).
2299 my ($self, $page) = @_;
2300 return (ref $self)->new($self->result_source, { %{$self->{attrs}}, page => $page });
2307 =item Arguments: \%col_data
2309 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2313 Creates a new result object in the resultset's result class and returns
2314 it. The row is not inserted into the database at this point, call
2315 L<DBIx::Class::Row/insert> to do that. Calling L<DBIx::Class::Row/in_storage>
2316 will tell you whether the result object has been inserted or not.
2318 Passes the hashref of input on to L<DBIx::Class::Row/new>.
2323 my ($self, $values) = @_;
2325 $self->throw_exception( "new_result takes only one argument - a hashref of values" )
2328 $self->throw_exception( "new_result expects a hashref" )
2329 unless (ref $values eq 'HASH');
2331 my ($merged_cond, $cols_from_relations) = $self->_merge_with_rscond($values);
2333 my $new = $self->result_class->new({
2335 ( @$cols_from_relations
2336 ? (-cols_from_relations => $cols_from_relations)
2339 -result_source => $self->result_source, # DO NOT REMOVE THIS, REQUIRED
2343 reftype($new) eq 'HASH'
2349 carp_unique (sprintf (
2350 "%s->new returned a blessed empty hashref - a strong indicator something is wrong with its inheritance chain",
2351 $self->result_class,
2358 # _merge_with_rscond
2360 # Takes a simple hash of K/V data and returns its copy merged with the
2361 # condition already present on the resultset. Additionally returns an
2362 # arrayref of value/condition names, which were inferred from related
2363 # objects (this is needed for in-memory related objects)
2364 sub _merge_with_rscond {
2365 my ($self, $data) = @_;
2367 my (%new_data, @cols_from_relations);
2369 my $alias = $self->{attrs}{alias};
2371 if (! defined $self->{cond}) {
2372 # just massage $data below
2374 elsif ($self->{cond} eq $DBIx::Class::ResultSource::UNRESOLVABLE_CONDITION) {
2375 %new_data = %{ $self->{attrs}{related_objects} || {} }; # nothing might have been inserted yet
2376 @cols_from_relations = keys %new_data;
2378 elsif (ref $self->{cond} ne 'HASH') {
2379 $self->throw_exception(
2380 "Can't abstract implicit construct, resultset condition not a hash"
2384 # precendence must be given to passed values over values inherited from
2385 # the cond, so the order here is important.
2386 my $collapsed_cond = $self->_collapse_cond($self->{cond});
2387 my %implied = %{$self->_remove_alias($collapsed_cond, $alias)};
2389 while ( my($col, $value) = each %implied ) {
2390 my $vref = ref $value;
2396 (keys %$value)[0] eq '='
2398 $new_data{$col} = $value->{'='};
2400 elsif( !$vref or $vref eq 'SCALAR' or blessed($value) ) {
2401 $new_data{$col} = $value;
2408 %{ $self->_remove_alias($data, $alias) },
2411 return (\%new_data, \@cols_from_relations);
2414 # _has_resolved_attr
2416 # determines if the resultset defines at least one
2417 # of the attributes supplied
2419 # used to determine if a subquery is neccessary
2421 # supports some virtual attributes:
2423 # This will scan for any joins being present on the resultset.
2424 # It is not a mere key-search but a deep inspection of {from}
2427 sub _has_resolved_attr {
2428 my ($self, @attr_names) = @_;
2430 my $attrs = $self->_resolved_attrs;
2434 for my $n (@attr_names) {
2435 if (grep { $n eq $_ } (qw/-join/) ) {
2436 $extra_checks{$n}++;
2440 my $attr = $attrs->{$n};
2442 next if not defined $attr;
2444 if (ref $attr eq 'HASH') {
2445 return 1 if keys %$attr;
2447 elsif (ref $attr eq 'ARRAY') {
2455 # a resolved join is expressed as a multi-level from
2457 $extra_checks{-join}
2459 ref $attrs->{from} eq 'ARRAY'
2461 @{$attrs->{from}} > 1
2469 # Recursively collapse the condition.
2471 sub _collapse_cond {
2472 my ($self, $cond, $collapsed) = @_;
2476 if (ref $cond eq 'ARRAY') {
2477 foreach my $subcond (@$cond) {
2478 next unless ref $subcond; # -or
2479 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2482 elsif (ref $cond eq 'HASH') {
2483 if (keys %$cond and (keys %$cond)[0] eq '-and') {
2484 foreach my $subcond (@{$cond->{-and}}) {
2485 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2489 foreach my $col (keys %$cond) {
2490 my $value = $cond->{$col};
2491 $collapsed->{$col} = $value;
2501 # Remove the specified alias from the specified query hash. A copy is made so
2502 # the original query is not modified.
2505 my ($self, $query, $alias) = @_;
2507 my %orig = %{ $query || {} };
2510 foreach my $key (keys %orig) {
2512 $unaliased{$key} = $orig{$key};
2515 $unaliased{$1} = $orig{$key}
2516 if $key =~ m/^(?:\Q$alias\E\.)?([^.]+)$/;
2526 =item Arguments: none
2528 =item Return Value: \[ $sql, L<@bind_values|/DBIC BIND VALUES> ]
2532 Returns the SQL query and bind vars associated with the invocant.
2534 This is generally used as the RHS for a subquery.
2541 my $attrs = { %{ $self->_resolved_attrs } };
2546 # my ($sql, \@bind, \%dbi_bind_attrs) = _select_args_to_query (...)
2547 # $sql also has no wrapping parenthesis in list ctx
2549 my $sqlbind = $self->result_source->storage
2550 ->_select_args_to_query ($attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs);
2559 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2561 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2565 my $artist = $schema->resultset('Artist')->find_or_new(
2566 { artist => 'fred' }, { key => 'artists' });
2568 $cd->cd_to_producer->find_or_new({ producer => $producer },
2569 { key => 'primary });
2571 Find an existing record from this resultset using L</find>. if none exists,
2572 instantiate a new result object and return it. The object will not be saved
2573 into your storage until you call L<DBIx::Class::Row/insert> on it.
2575 You most likely want this method when looking for existing rows using a unique
2576 constraint that is not the primary key, or looking for related rows.
2578 If you want objects to be saved immediately, use L</find_or_create> instead.
2580 B<Note>: Make sure to read the documentation of L</find> and understand the
2581 significance of the C<key> attribute, as its lack may skew your search, and
2582 subsequently result in spurious new objects.
2584 B<Note>: Take care when using C<find_or_new> with a table having
2585 columns with default values that you intend to be automatically
2586 supplied by the database (e.g. an auto_increment primary key column).
2587 In normal usage, the value of such columns should NOT be included at
2588 all in the call to C<find_or_new>, even when set to C<undef>.
2594 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2595 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2596 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2599 return $self->new_result($hash);
2606 =item Arguments: \%col_data
2608 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2612 Attempt to create a single new row or a row with multiple related rows
2613 in the table represented by the resultset (and related tables). This
2614 will not check for duplicate rows before inserting, use
2615 L</find_or_create> to do that.
2617 To create one row for this resultset, pass a hashref of key/value
2618 pairs representing the columns of the table and the values you wish to
2619 store. If the appropriate relationships are set up, foreign key fields
2620 can also be passed an object representing the foreign row, and the
2621 value will be set to its primary key.
2623 To create related objects, pass a hashref of related-object column values
2624 B<keyed on the relationship name>. If the relationship is of type C<multi>
2625 (L<DBIx::Class::Relationship/has_many>) - pass an arrayref of hashrefs.
2626 The process will correctly identify columns holding foreign keys, and will
2627 transparently populate them from the keys of the corresponding relation.
2628 This can be applied recursively, and will work correctly for a structure
2629 with an arbitrary depth and width, as long as the relationships actually
2630 exists and the correct column data has been supplied.
2632 Instead of hashrefs of plain related data (key/value pairs), you may
2633 also pass new or inserted objects. New objects (not inserted yet, see
2634 L</new_result>), will be inserted into their appropriate tables.
2636 Effectively a shortcut for C<< ->new_result(\%col_data)->insert >>.
2638 Example of creating a new row.
2640 $person_rs->create({
2641 name=>"Some Person",
2642 email=>"somebody@someplace.com"
2645 Example of creating a new row and also creating rows in a related C<has_many>
2646 or C<has_one> resultset. Note Arrayref.
2649 { artistid => 4, name => 'Manufactured Crap', cds => [
2650 { title => 'My First CD', year => 2006 },
2651 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2656 Example of creating a new row and also creating a row in a related
2657 C<belongs_to> resultset. Note Hashref.
2660 title=>"Music for Silly Walks",
2663 name=>"Silly Musician",
2671 When subclassing ResultSet never attempt to override this method. Since
2672 it is a simple shortcut for C<< $self->new_result($attrs)->insert >>, a
2673 lot of the internals simply never call it, so your override will be
2674 bypassed more often than not. Override either L<DBIx::Class::Row/new>
2675 or L<DBIx::Class::Row/insert> depending on how early in the
2676 L</create> process you need to intervene. See also warning pertaining to
2684 my ($self, $attrs) = @_;
2685 $self->throw_exception( "create needs a hashref" )
2686 unless ref $attrs eq 'HASH';
2687 return $self->new_result($attrs)->insert;
2690 =head2 find_or_create
2694 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2696 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2700 $cd->cd_to_producer->find_or_create({ producer => $producer },
2701 { key => 'primary' });
2703 Tries to find a record based on its primary key or unique constraints; if none
2704 is found, creates one and returns that instead.
2706 my $cd = $schema->resultset('CD')->find_or_create({
2708 artist => 'Massive Attack',
2709 title => 'Mezzanine',
2713 Also takes an optional C<key> attribute, to search by a specific key or unique
2714 constraint. For example:
2716 my $cd = $schema->resultset('CD')->find_or_create(
2718 artist => 'Massive Attack',
2719 title => 'Mezzanine',
2721 { key => 'cd_artist_title' }
2724 B<Note>: Make sure to read the documentation of L</find> and understand the
2725 significance of the C<key> attribute, as its lack may skew your search, and
2726 subsequently result in spurious row creation.
2728 B<Note>: Because find_or_create() reads from the database and then
2729 possibly inserts based on the result, this method is subject to a race
2730 condition. Another process could create a record in the table after
2731 the find has completed and before the create has started. To avoid
2732 this problem, use find_or_create() inside a transaction.
2734 B<Note>: Take care when using C<find_or_create> with a table having
2735 columns with default values that you intend to be automatically
2736 supplied by the database (e.g. an auto_increment primary key column).
2737 In normal usage, the value of such columns should NOT be included at
2738 all in the call to C<find_or_create>, even when set to C<undef>.
2740 See also L</find> and L</update_or_create>. For information on how to declare
2741 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2743 If you need to know if an existing row was found or a new one created use
2744 L</find_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
2745 to call L<DBIx::Class::Row/insert> to save the newly created row to the
2748 my $cd = $schema->resultset('CD')->find_or_new({
2750 artist => 'Massive Attack',
2751 title => 'Mezzanine',
2755 if( !$cd->in_storage ) {
2762 sub find_or_create {
2764 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2765 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2766 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2769 return $self->create($hash);
2772 =head2 update_or_create
2776 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2778 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2782 $resultset->update_or_create({ col => $val, ... });
2784 Like L</find_or_create>, but if a row is found it is immediately updated via
2785 C<< $found_row->update (\%col_data) >>.
2788 Takes an optional C<key> attribute to search on a specific unique constraint.
2791 # In your application
2792 my $cd = $schema->resultset('CD')->update_or_create(
2794 artist => 'Massive Attack',
2795 title => 'Mezzanine',
2798 { key => 'cd_artist_title' }
2801 $cd->cd_to_producer->update_or_create({
2802 producer => $producer,
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>: Take care when using C<update_or_create> with a table having
2813 columns with default values that you intend to be automatically
2814 supplied by the database (e.g. an auto_increment primary key column).
2815 In normal usage, the value of such columns should NOT be included at
2816 all in the call to C<update_or_create>, even when set to C<undef>.
2818 See also L</find> and L</find_or_create>. For information on how to declare
2819 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2821 If you need to know if an existing row was updated or a new one created use
2822 L</update_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
2823 to call L<DBIx::Class::Row/insert> to save the newly created row to the
2828 sub update_or_create {
2830 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2831 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2833 my $row = $self->find($cond, $attrs);
2835 $row->update($cond);
2839 return $self->create($cond);
2842 =head2 update_or_new
2846 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2848 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2852 $resultset->update_or_new({ col => $val, ... });
2854 Like L</find_or_new> but if a row is found it is immediately updated via
2855 C<< $found_row->update (\%col_data) >>.
2859 # In your application
2860 my $cd = $schema->resultset('CD')->update_or_new(
2862 artist => 'Massive Attack',
2863 title => 'Mezzanine',
2866 { key => 'cd_artist_title' }
2869 if ($cd->in_storage) {
2870 # the cd was updated
2873 # the cd is not yet in the database, let's insert it
2877 B<Note>: Make sure to read the documentation of L</find> and understand the
2878 significance of the C<key> attribute, as its lack may skew your search, and
2879 subsequently result in spurious new objects.
2881 B<Note>: Take care when using C<update_or_new> with a table having
2882 columns with default values that you intend to be automatically
2883 supplied by the database (e.g. an auto_increment primary key column).
2884 In normal usage, the value of such columns should NOT be included at
2885 all in the call to C<update_or_new>, even when set to C<undef>.
2887 See also L</find>, L</find_or_create> and L</find_or_new>.
2893 my $attrs = ( @_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {} );
2894 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2896 my $row = $self->find( $cond, $attrs );
2897 if ( defined $row ) {
2898 $row->update($cond);
2902 return $self->new_result($cond);
2909 =item Arguments: none
2911 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass> | undef
2915 Gets the contents of the cache for the resultset, if the cache is set.
2917 The cache is populated either by using the L</prefetch> attribute to
2918 L</search> or by calling L</set_cache>.
2930 =item Arguments: L<\@result_objs|DBIx::Class::Manual::ResultClass>
2932 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass>
2936 Sets the contents of the cache for the resultset. Expects an arrayref
2937 of objects of the same class as those produced by the resultset. Note that
2938 if the cache is set, the resultset will return the cached objects rather
2939 than re-querying the database even if the cache attr is not set.
2941 The contents of the cache can also be populated by using the
2942 L</prefetch> attribute to L</search>.
2947 my ( $self, $data ) = @_;
2948 $self->throw_exception("set_cache requires an arrayref")
2949 if defined($data) && (ref $data ne 'ARRAY');
2950 $self->{all_cache} = $data;
2957 =item Arguments: none
2959 =item Return Value: undef
2963 Clears the cache for the resultset.
2968 shift->set_cache(undef);
2975 =item Arguments: none
2977 =item Return Value: true, if the resultset has been paginated
2985 return !!$self->{attrs}{page};
2992 =item Arguments: none
2994 =item Return Value: true, if the resultset has been ordered with C<order_by>.
3002 return scalar $self->result_source->storage->_extract_order_criteria($self->{attrs}{order_by});
3005 =head2 related_resultset
3009 =item Arguments: $rel_name
3011 =item Return Value: L<$resultset|/search>
3015 Returns a related resultset for the supplied relationship name.
3017 $artist_rs = $schema->resultset('CD')->related_resultset('Artist');
3021 sub related_resultset {
3022 my ($self, $rel) = @_;
3024 $self->{related_resultsets} ||= {};
3025 return $self->{related_resultsets}{$rel} ||= do {
3026 my $rsrc = $self->result_source;
3027 my $rel_info = $rsrc->relationship_info($rel);
3029 $self->throw_exception(
3030 "search_related: result source '" . $rsrc->source_name .
3031 "' has no such relationship $rel")
3034 my $attrs = $self->_chain_relationship($rel);
3036 my $join_count = $attrs->{seen_join}{$rel};
3038 my $alias = $self->result_source->storage
3039 ->relname_to_table_alias($rel, $join_count);
3041 # since this is search_related, and we already slid the select window inwards
3042 # (the select/as attrs were deleted in the beginning), we need to flip all
3043 # left joins to inner, so we get the expected results
3044 # read the comment on top of the actual function to see what this does
3045 $attrs->{from} = $rsrc->schema->storage->_inner_join_to_node ($attrs->{from}, $alias);
3048 #XXX - temp fix for result_class bug. There likely is a more elegant fix -groditi
3049 delete @{$attrs}{qw(result_class alias)};
3053 if (my $cache = $self->get_cache) {
3054 if ($cache->[0] && $cache->[0]->related_resultset($rel)->get_cache) {
3055 $new_cache = [ map { @{$_->related_resultset($rel)->get_cache} }
3060 my $rel_source = $rsrc->related_source($rel);
3064 # The reason we do this now instead of passing the alias to the
3065 # search_rs below is that if you wrap/overload resultset on the
3066 # source you need to know what alias it's -going- to have for things
3067 # to work sanely (e.g. RestrictWithObject wants to be able to add
3068 # extra query restrictions, and these may need to be $alias.)
3070 my $rel_attrs = $rel_source->resultset_attributes;
3071 local $rel_attrs->{alias} = $alias;
3073 $rel_source->resultset
3077 where => $attrs->{where},
3080 $new->set_cache($new_cache) if $new_cache;
3085 =head2 current_source_alias
3089 =item Arguments: none
3091 =item Return Value: $source_alias
3095 Returns the current table alias for the result source this resultset is built
3096 on, that will be used in the SQL query. Usually it is C<me>.
3098 Currently the source alias that refers to the result set returned by a
3099 L</search>/L</find> family method depends on how you got to the resultset: it's
3100 C<me> by default, but eg. L</search_related> aliases it to the related result
3101 source name (and keeps C<me> referring to the original result set). The long
3102 term goal is to make L<DBIx::Class> always alias the current resultset as C<me>
3103 (and make this method unnecessary).
3105 Thus it's currently necessary to use this method in predefined queries (see
3106 L<DBIx::Class::Manual::Cookbook/Predefined searches>) when referring to the
3107 source alias of the current result set:
3109 # in a result set class
3111 my ($self, $user) = @_;
3113 my $me = $self->current_source_alias;
3115 return $self->search({
3116 "$me.modified" => $user->id,
3122 sub current_source_alias {
3123 return (shift->{attrs} || {})->{alias} || 'me';
3126 =head2 as_subselect_rs
3130 =item Arguments: none
3132 =item Return Value: L<$resultset|/search>
3136 Act as a barrier to SQL symbols. The resultset provided will be made into a
3137 "virtual view" by including it as a subquery within the from clause. From this
3138 point on, any joined tables are inaccessible to ->search on the resultset (as if
3139 it were simply where-filtered without joins). For example:
3141 my $rs = $schema->resultset('Bar')->search({'x.name' => 'abc'},{ join => 'x' });
3143 # 'x' now pollutes the query namespace
3145 # So the following works as expected
3146 my $ok_rs = $rs->search({'x.other' => 1});
3148 # But this doesn't: instead of finding a 'Bar' related to two x rows (abc and
3149 # def) we look for one row with contradictory terms and join in another table
3150 # (aliased 'x_2') which we never use
3151 my $broken_rs = $rs->search({'x.name' => 'def'});
3153 my $rs2 = $rs->as_subselect_rs;
3155 # doesn't work - 'x' is no longer accessible in $rs2, having been sealed away
3156 my $not_joined_rs = $rs2->search({'x.other' => 1});
3158 # works as expected: finds a 'table' row related to two x rows (abc and def)
3159 my $correctly_joined_rs = $rs2->search({'x.name' => 'def'});
3161 Another example of when one might use this would be to select a subset of
3162 columns in a group by clause:
3164 my $rs = $schema->resultset('Bar')->search(undef, {
3165 group_by => [qw{ id foo_id baz_id }],
3166 })->as_subselect_rs->search(undef, {
3167 columns => [qw{ id foo_id }]
3170 In the above example normally columns would have to be equal to the group by,
3171 but because we isolated the group by into a subselect the above works.
3175 sub as_subselect_rs {
3178 my $attrs = $self->_resolved_attrs;
3180 my $fresh_rs = (ref $self)->new (
3181 $self->result_source
3184 # these pieces will be locked in the subquery
3185 delete $fresh_rs->{cond};
3186 delete @{$fresh_rs->{attrs}}{qw/where bind/};
3188 return $fresh_rs->search( {}, {
3190 $attrs->{alias} => $self->as_query,
3191 -alias => $attrs->{alias},
3192 -rsrc => $self->result_source,
3194 alias => $attrs->{alias},
3198 # This code is called by search_related, and makes sure there
3199 # is clear separation between the joins before, during, and
3200 # after the relationship. This information is needed later
3201 # in order to properly resolve prefetch aliases (any alias
3202 # with a relation_chain_depth less than the depth of the
3203 # current prefetch is not considered)
3205 # The increments happen twice per join. An even number means a
3206 # relationship specified via a search_related, whereas an odd
3207 # number indicates a join/prefetch added via attributes
3209 # Also this code will wrap the current resultset (the one we
3210 # chain to) in a subselect IFF it contains limiting attributes
3211 sub _chain_relationship {
3212 my ($self, $rel) = @_;
3213 my $source = $self->result_source;
3214 my $attrs = { %{$self->{attrs}||{}} };
3216 # we need to take the prefetch the attrs into account before we
3217 # ->_resolve_join as otherwise they get lost - captainL
3218 my $join = $self->_merge_joinpref_attr( $attrs->{join}, $attrs->{prefetch} );
3220 delete @{$attrs}{qw/join prefetch collapse group_by distinct select as columns +select +as +columns/};
3222 my $seen = { %{ (delete $attrs->{seen_join}) || {} } };
3225 my @force_subq_attrs = qw/offset rows group_by having/;
3228 ($attrs->{from} && ref $attrs->{from} ne 'ARRAY')
3230 $self->_has_resolved_attr (@force_subq_attrs)
3232 # Nuke the prefetch (if any) before the new $rs attrs
3233 # are resolved (prefetch is useless - we are wrapping
3234 # a subquery anyway).
3235 my $rs_copy = $self->search;
3236 $rs_copy->{attrs}{join} = $self->_merge_joinpref_attr (
3237 $rs_copy->{attrs}{join},
3238 delete $rs_copy->{attrs}{prefetch},
3243 -alias => $attrs->{alias},
3244 $attrs->{alias} => $rs_copy->as_query,
3246 delete @{$attrs}{@force_subq_attrs, qw/where bind/};
3247 $seen->{-relation_chain_depth} = 0;
3249 elsif ($attrs->{from}) { #shallow copy suffices
3250 $from = [ @{$attrs->{from}} ];
3255 -alias => $attrs->{alias},
3256 $attrs->{alias} => $source->from,
3260 my $jpath = ($seen->{-relation_chain_depth})
3261 ? $from->[-1][0]{-join_path}
3264 my @requested_joins = $source->_resolve_join(
3271 push @$from, @requested_joins;
3273 $seen->{-relation_chain_depth}++;
3275 # if $self already had a join/prefetch specified on it, the requested
3276 # $rel might very well be already included. What we do in this case
3277 # is effectively a no-op (except that we bump up the chain_depth on
3278 # the join in question so we could tell it *is* the search_related)
3281 # we consider the last one thus reverse
3282 for my $j (reverse @requested_joins) {
3283 my ($last_j) = keys %{$j->[0]{-join_path}[-1]};
3284 if ($rel eq $last_j) {
3285 $j->[0]{-relation_chain_depth}++;
3291 unless ($already_joined) {
3292 push @$from, $source->_resolve_join(
3300 $seen->{-relation_chain_depth}++;
3302 return {%$attrs, from => $from, seen_join => $seen};
3305 sub _resolved_attrs {
3307 return $self->{_attrs} if $self->{_attrs};
3309 my $attrs = { %{ $self->{attrs} || {} } };
3310 my $source = $self->result_source;
3311 my $alias = $attrs->{alias};
3313 # default selection list
3314 $attrs->{columns} = [ $source->columns ]
3315 unless List::Util::first { exists $attrs->{$_} } qw/columns cols select as/;
3317 # merge selectors together
3318 for (qw/columns select as/) {
3319 $attrs->{$_} = $self->_merge_attr($attrs->{$_}, delete $attrs->{"+$_"})
3320 if $attrs->{$_} or $attrs->{"+$_"};
3323 # disassemble columns
3325 if (my $cols = delete $attrs->{columns}) {
3326 for my $c (ref $cols eq 'ARRAY' ? @$cols : $cols) {
3327 if (ref $c eq 'HASH') {
3328 for my $as (sort keys %$c) {
3329 push @sel, $c->{$as};
3340 # when trying to weed off duplicates later do not go past this point -
3341 # everything added from here on is unbalanced "anyone's guess" stuff
3342 my $dedup_stop_idx = $#as;
3344 push @as, @{ ref $attrs->{as} eq 'ARRAY' ? $attrs->{as} : [ $attrs->{as} ] }
3346 push @sel, @{ ref $attrs->{select} eq 'ARRAY' ? $attrs->{select} : [ $attrs->{select} ] }
3347 if $attrs->{select};
3349 # assume all unqualified selectors to apply to the current alias (legacy stuff)
3351 $_ = (ref $_ or $_ =~ /\./) ? $_ : "$alias.$_";
3354 # disqualify all $alias.col as-bits (collapser mandated)
3356 $_ = ($_ =~ /^\Q$alias.\E(.+)$/) ? $1 : $_;
3359 # de-duplicate the result (remove *identical* select/as pairs)
3360 # and also die on duplicate {as} pointing to different {select}s
3361 # not using a c-style for as the condition is prone to shrinkage
3364 while ($i <= $dedup_stop_idx) {
3365 if ($seen->{"$sel[$i] \x00\x00 $as[$i]"}++) {
3370 elsif ($seen->{$as[$i]}++) {
3371 $self->throw_exception(
3372 "inflate_result() alias '$as[$i]' specified twice with different SQL-side {select}-ors"
3380 $attrs->{select} = \@sel;
3381 $attrs->{as} = \@as;
3383 $attrs->{from} ||= [{
3385 -alias => $self->{attrs}{alias},
3386 $self->{attrs}{alias} => $source->from,
3389 if ( $attrs->{join} || $attrs->{prefetch} ) {
3391 $self->throw_exception ('join/prefetch can not be used with a custom {from}')
3392 if ref $attrs->{from} ne 'ARRAY';
3394 my $join = (delete $attrs->{join}) || {};
3396 if ( defined $attrs->{prefetch} ) {
3397 $join = $self->_merge_joinpref_attr( $join, $attrs->{prefetch} );
3400 $attrs->{from} = # have to copy here to avoid corrupting the original
3402 @{ $attrs->{from} },
3403 $source->_resolve_join(
3406 { %{ $attrs->{seen_join} || {} } },
3407 ( $attrs->{seen_join} && keys %{$attrs->{seen_join}})
3408 ? $attrs->{from}[-1][0]{-join_path}
3415 if ( defined $attrs->{order_by} ) {
3416 $attrs->{order_by} = (
3417 ref( $attrs->{order_by} ) eq 'ARRAY'
3418 ? [ @{ $attrs->{order_by} } ]
3419 : [ $attrs->{order_by} || () ]
3423 if ($attrs->{group_by} and ref $attrs->{group_by} ne 'ARRAY') {
3424 $attrs->{group_by} = [ $attrs->{group_by} ];
3427 # generate the distinct induced group_by early, as prefetch will be carried via a
3428 # subquery (since a group_by is present)
3429 if (delete $attrs->{distinct}) {
3430 if ($attrs->{group_by}) {
3431 carp_unique ("Useless use of distinct on a grouped resultset ('distinct' is ignored when a 'group_by' is present)");
3434 # distinct affects only the main selection part, not what prefetch may
3436 $attrs->{group_by} = $source->storage->_group_over_selection (
3444 $attrs->{collapse} ||= {};
3445 if ($attrs->{prefetch}) {
3447 $self->throw_exception("Unable to prefetch, resultset contains an unnamed selector $attrs->{_dark_selector}{string}")
3448 if $attrs->{_dark_selector};
3450 my $prefetch = $self->_merge_joinpref_attr( {}, delete $attrs->{prefetch} );
3452 my $prefetch_ordering = [];
3454 # this is a separate structure (we don't look in {from} directly)
3455 # as the resolver needs to shift things off the lists to work
3456 # properly (identical-prefetches on different branches)
3458 if (ref $attrs->{from} eq 'ARRAY') {
3460 my $start_depth = $attrs->{seen_join}{-relation_chain_depth} || 0;
3462 for my $j ( @{$attrs->{from}}[1 .. $#{$attrs->{from}} ] ) {
3463 next unless $j->[0]{-alias};
3464 next unless $j->[0]{-join_path};
3465 next if ($j->[0]{-relation_chain_depth} || 0) < $start_depth;
3467 my @jpath = map { keys %$_ } @{$j->[0]{-join_path}};
3470 $p = $p->{$_} ||= {} for @jpath[ ($start_depth/2) .. $#jpath]; #only even depths are actual jpath boundaries
3471 push @{$p->{-join_aliases} }, $j->[0]{-alias};
3476 $source->_resolve_prefetch( $prefetch, $alias, $join_map, $prefetch_ordering, $attrs->{collapse} );
3478 # we need to somehow mark which columns came from prefetch
3480 my $sel_end = $#{$attrs->{select}};
3481 $attrs->{_prefetch_selector_range} = [ $sel_end + 1, $sel_end + @prefetch ];
3484 push @{ $attrs->{select} }, (map { $_->[0] } @prefetch);
3485 push @{ $attrs->{as} }, (map { $_->[1] } @prefetch);
3487 push( @{$attrs->{order_by}}, @$prefetch_ordering );
3488 $attrs->{_collapse_order_by} = \@$prefetch_ordering;
3491 # if both page and offset are specified, produce a combined offset
3492 # even though it doesn't make much sense, this is what pre 081xx has
3494 if (my $page = delete $attrs->{page}) {
3496 ($attrs->{rows} * ($page - 1))
3498 ($attrs->{offset} || 0)
3502 return $self->{_attrs} = $attrs;
3506 my ($self, $attr) = @_;
3508 if (ref $attr eq 'HASH') {
3509 return $self->_rollout_hash($attr);
3510 } elsif (ref $attr eq 'ARRAY') {
3511 return $self->_rollout_array($attr);
3517 sub _rollout_array {
3518 my ($self, $attr) = @_;
3521 foreach my $element (@{$attr}) {
3522 if (ref $element eq 'HASH') {
3523 push( @rolled_array, @{ $self->_rollout_hash( $element ) } );
3524 } elsif (ref $element eq 'ARRAY') {
3525 # XXX - should probably recurse here
3526 push( @rolled_array, @{$self->_rollout_array($element)} );
3528 push( @rolled_array, $element );
3531 return \@rolled_array;
3535 my ($self, $attr) = @_;
3538 foreach my $key (keys %{$attr}) {
3539 push( @rolled_array, { $key => $attr->{$key} } );
3541 return \@rolled_array;
3544 sub _calculate_score {
3545 my ($self, $a, $b) = @_;
3547 if (defined $a xor defined $b) {
3550 elsif (not defined $a) {
3554 if (ref $b eq 'HASH') {
3555 my ($b_key) = keys %{$b};
3556 if (ref $a eq 'HASH') {
3557 my ($a_key) = keys %{$a};
3558 if ($a_key eq $b_key) {
3559 return (1 + $self->_calculate_score( $a->{$a_key}, $b->{$b_key} ));
3564 return ($a eq $b_key) ? 1 : 0;
3567 if (ref $a eq 'HASH') {
3568 my ($a_key) = keys %{$a};
3569 return ($b eq $a_key) ? 1 : 0;
3571 return ($b eq $a) ? 1 : 0;
3576 sub _merge_joinpref_attr {
3577 my ($self, $orig, $import) = @_;
3579 return $import unless defined($orig);
3580 return $orig unless defined($import);
3582 $orig = $self->_rollout_attr($orig);
3583 $import = $self->_rollout_attr($import);
3586 foreach my $import_element ( @{$import} ) {
3587 # find best candidate from $orig to merge $b_element into
3588 my $best_candidate = { position => undef, score => 0 }; my $position = 0;
3589 foreach my $orig_element ( @{$orig} ) {
3590 my $score = $self->_calculate_score( $orig_element, $import_element );
3591 if ($score > $best_candidate->{score}) {
3592 $best_candidate->{position} = $position;
3593 $best_candidate->{score} = $score;
3597 my ($import_key) = ( ref $import_element eq 'HASH' ) ? keys %{$import_element} : ($import_element);
3598 $import_key = '' if not defined $import_key;
3600 if ($best_candidate->{score} == 0 || exists $seen_keys->{$import_key}) {
3601 push( @{$orig}, $import_element );
3603 my $orig_best = $orig->[$best_candidate->{position}];
3604 # merge orig_best and b_element together and replace original with merged
3605 if (ref $orig_best ne 'HASH') {
3606 $orig->[$best_candidate->{position}] = $import_element;
3607 } elsif (ref $import_element eq 'HASH') {
3608 my ($key) = keys %{$orig_best};
3609 $orig->[$best_candidate->{position}] = { $key => $self->_merge_joinpref_attr($orig_best->{$key}, $import_element->{$key}) };
3612 $seen_keys->{$import_key} = 1; # don't merge the same key twice
3623 require Hash::Merge;
3624 my $hm = Hash::Merge->new;
3626 $hm->specify_behavior({
3629 my ($defl, $defr) = map { defined $_ } (@_[0,1]);
3631 if ($defl xor $defr) {
3632 return [ $defl ? $_[0] : $_[1] ];
3637 elsif (__HM_DEDUP and $_[0] eq $_[1]) {
3641 return [$_[0], $_[1]];
3645 return $_[1] if !defined $_[0];
3646 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3647 return [$_[0], @{$_[1]}]
3650 return [] if !defined $_[0] and !keys %{$_[1]};
3651 return [ $_[1] ] if !defined $_[0];
3652 return [ $_[0] ] if !keys %{$_[1]};
3653 return [$_[0], $_[1]]
3658 return $_[0] if !defined $_[1];
3659 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3660 return [@{$_[0]}, $_[1]]
3663 my @ret = @{$_[0]} or return $_[1];
3664 return [ @ret, @{$_[1]} ] unless __HM_DEDUP;
3665 my %idx = map { $_ => 1 } @ret;
3666 push @ret, grep { ! defined $idx{$_} } (@{$_[1]});
3670 return [ $_[1] ] if ! @{$_[0]};
3671 return $_[0] if !keys %{$_[1]};
3672 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3673 return [ @{$_[0]}, $_[1] ];
3678 return [] if !keys %{$_[0]} and !defined $_[1];
3679 return [ $_[0] ] if !defined $_[1];
3680 return [ $_[1] ] if !keys %{$_[0]};
3681 return [$_[0], $_[1]]
3684 return [] if !keys %{$_[0]} and !@{$_[1]};
3685 return [ $_[0] ] if !@{$_[1]};
3686 return $_[1] if !keys %{$_[0]};
3687 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3688 return [ $_[0], @{$_[1]} ];
3691 return [] if !keys %{$_[0]} and !keys %{$_[1]};
3692 return [ $_[0] ] if !keys %{$_[1]};
3693 return [ $_[1] ] if !keys %{$_[0]};
3694 return [ $_[0] ] if $_[0] eq $_[1];
3695 return [ $_[0], $_[1] ];
3698 } => 'DBIC_RS_ATTR_MERGER');
3702 return $hm->merge ($_[1], $_[2]);
3706 sub STORABLE_freeze {
3707 my ($self, $cloning) = @_;
3708 my $to_serialize = { %$self };
3710 # A cursor in progress can't be serialized (and would make little sense anyway)
3711 delete $to_serialize->{cursor};
3713 # nor is it sensical to store a not-yet-fired-count pager
3714 if ($to_serialize->{pager} and ref $to_serialize->{pager}{total_entries} eq 'CODE') {
3715 delete $to_serialize->{pager};
3718 Storable::nfreeze($to_serialize);
3721 # need this hook for symmetry
3723 my ($self, $cloning, $serialized) = @_;
3725 %$self = %{ Storable::thaw($serialized) };
3731 =head2 throw_exception
3733 See L<DBIx::Class::Schema/throw_exception> for details.
3737 sub throw_exception {
3740 if (ref $self and my $rsrc = $self->result_source) {
3741 $rsrc->throw_exception(@_)
3744 DBIx::Class::Exception->throw(@_);
3748 # XXX: FIXME: Attributes docs need clearing up
3752 Attributes are used to refine a ResultSet in various ways when
3753 searching for data. They can be passed to any method which takes an
3754 C<\%attrs> argument. See L</search>, L</search_rs>, L</find>,
3757 Default attributes can be set on the result class using
3758 L<DBIx::Class::ResultSource/resultset_attributes>. (Please read
3759 the CAVEATS on that feature before using it!)
3761 These are in no particular order:
3767 =item Value: ( $order_by | \@order_by | \%order_by )
3771 Which column(s) to order the results by.
3773 [The full list of suitable values is documented in
3774 L<SQL::Abstract/"ORDER BY CLAUSES">; the following is a summary of
3777 If a single column name, or an arrayref of names is supplied, the
3778 argument is passed through directly to SQL. The hashref syntax allows
3779 for connection-agnostic specification of ordering direction:
3781 For descending order:
3783 order_by => { -desc => [qw/col1 col2 col3/] }
3785 For explicit ascending order:
3787 order_by => { -asc => 'col' }
3789 The old scalarref syntax (i.e. order_by => \'year DESC') is still
3790 supported, although you are strongly encouraged to use the hashref
3791 syntax as outlined above.
3797 =item Value: \@columns
3801 Shortcut to request a particular set of columns to be retrieved. Each
3802 column spec may be a string (a table column name), or a hash (in which
3803 case the key is the C<as> value, and the value is used as the C<select>
3804 expression). Adds C<me.> onto the start of any column without a C<.> in
3805 it and sets C<select> from that, then auto-populates C<as> from
3806 C<select> as normal. (You may also use the C<cols> attribute, as in
3807 earlier versions of DBIC, but this is deprecated.)
3809 Essentially C<columns> does the same as L</select> and L</as>.
3811 columns => [ 'foo', { bar => 'baz' } ]
3815 select => [qw/foo baz/],
3822 =item Value: \@columns
3826 Indicates additional columns to be selected from storage. Works the same as
3827 L</columns> but adds columns to the selection. (You may also use the
3828 C<include_columns> attribute, as in earlier versions of DBIC, but this is
3829 deprecated). For example:-
3831 $schema->resultset('CD')->search(undef, {
3832 '+columns' => ['artist.name'],
3836 would return all CDs and include a 'name' column to the information
3837 passed to object inflation. Note that the 'artist' is the name of the
3838 column (or relationship) accessor, and 'name' is the name of the column
3839 accessor in the related table.
3841 B<NOTE:> You need to explicitly quote '+columns' when defining the attribute.
3842 Not doing so causes Perl to incorrectly interpret +columns as a bareword with a
3843 unary plus operator before it.
3845 =head2 include_columns
3849 =item Value: \@columns
3853 Deprecated. Acts as a synonym for L</+columns> for backward compatibility.
3859 =item Value: \@select_columns
3863 Indicates which columns should be selected from the storage. You can use
3864 column names, or in the case of RDBMS back ends, function or stored procedure
3867 $rs = $schema->resultset('Employee')->search(undef, {
3870 { count => 'employeeid' },
3871 { max => { length => 'name' }, -as => 'longest_name' }
3876 SELECT name, COUNT( employeeid ), MAX( LENGTH( name ) ) AS longest_name FROM employee
3878 B<NOTE:> You will almost always need a corresponding L</as> attribute when you
3879 use L</select>, to instruct DBIx::Class how to store the result of the column.
3880 Also note that the L</as> attribute has nothing to do with the SQL-side 'AS'
3881 identifier aliasing. You can however alias a function, so you can use it in
3882 e.g. an C<ORDER BY> clause. This is done via the C<-as> B<select function
3883 attribute> supplied as shown in the example above.
3885 B<NOTE:> You need to explicitly quote '+select'/'+as' when defining the attributes.
3886 Not doing so causes Perl to incorrectly interpret them as a bareword with a
3887 unary plus operator before it.
3893 Indicates additional columns to be selected from storage. Works the same as
3894 L</select> but adds columns to the default selection, instead of specifying
3903 Indicates additional column names for those added via L</+select>. See L</as>.
3911 =item Value: \@inflation_names
3915 Indicates column names for object inflation. That is L</as> indicates the
3916 slot name in which the column value will be stored within the
3917 L<Row|DBIx::Class::Row> object. The value will then be accessible via this
3918 identifier by the C<get_column> method (or via the object accessor B<if one
3919 with the same name already exists>) as shown below. The L</as> attribute has
3920 B<nothing to do> with the SQL-side C<AS>. See L</select> for details.
3922 $rs = $schema->resultset('Employee')->search(undef, {
3925 { count => 'employeeid' },
3926 { max => { length => 'name' }, -as => 'longest_name' }
3935 If the object against which the search is performed already has an accessor
3936 matching a column name specified in C<as>, the value can be retrieved using
3937 the accessor as normal:
3939 my $name = $employee->name();
3941 If on the other hand an accessor does not exist in the object, you need to
3942 use C<get_column> instead:
3944 my $employee_count = $employee->get_column('employee_count');
3946 You can create your own accessors if required - see
3947 L<DBIx::Class::Manual::Cookbook> for details.
3953 =item Value: ($rel_name | \@rel_names | \%rel_names)
3957 Contains a list of relationships that should be joined for this query. For
3960 # Get CDs by Nine Inch Nails
3961 my $rs = $schema->resultset('CD')->search(
3962 { 'artist.name' => 'Nine Inch Nails' },
3963 { join => 'artist' }
3966 Can also contain a hash reference to refer to the other relation's relations.
3969 package MyApp::Schema::Track;
3970 use base qw/DBIx::Class/;
3971 __PACKAGE__->table('track');
3972 __PACKAGE__->add_columns(qw/trackid cd position title/);
3973 __PACKAGE__->set_primary_key('trackid');
3974 __PACKAGE__->belongs_to(cd => 'MyApp::Schema::CD');
3977 # In your application
3978 my $rs = $schema->resultset('Artist')->search(
3979 { 'track.title' => 'Teardrop' },
3981 join => { cd => 'track' },
3982 order_by => 'artist.name',
3986 You need to use the relationship (not the table) name in conditions,
3987 because they are aliased as such. The current table is aliased as "me", so
3988 you need to use me.column_name in order to avoid ambiguity. For example:
3990 # Get CDs from 1984 with a 'Foo' track
3991 my $rs = $schema->resultset('CD')->search(
3994 'tracks.name' => 'Foo'
3996 { join => 'tracks' }
3999 If the same join is supplied twice, it will be aliased to <rel>_2 (and
4000 similarly for a third time). For e.g.
4002 my $rs = $schema->resultset('Artist')->search({
4003 'cds.title' => 'Down to Earth',
4004 'cds_2.title' => 'Popular',
4006 join => [ qw/cds cds/ ],
4009 will return a set of all artists that have both a cd with title 'Down
4010 to Earth' and a cd with title 'Popular'.
4012 If you want to fetch related objects from other tables as well, see C<prefetch>
4015 NOTE: An internal join-chain pruner will discard certain joins while
4016 constructing the actual SQL query, as long as the joins in question do not
4017 affect the retrieved result. This for example includes 1:1 left joins
4018 that are not part of the restriction specification (WHERE/HAVING) nor are
4019 a part of the query selection.
4021 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
4027 =item Value: ($rel_name | \@rel_names | \%rel_names)
4031 Contains one or more relationships that should be fetched along with
4032 the main query (when they are accessed afterwards the data will
4033 already be available, without extra queries to the database). This is
4034 useful for when you know you will need the related objects, because it
4035 saves at least one query:
4037 my $rs = $schema->resultset('Tag')->search(
4046 The initial search results in SQL like the following:
4048 SELECT tag.*, cd.*, artist.* FROM tag
4049 JOIN cd ON tag.cd = cd.cdid
4050 JOIN artist ON cd.artist = artist.artistid
4052 L<DBIx::Class> has no need to go back to the database when we access the
4053 C<cd> or C<artist> relationships, which saves us two SQL statements in this
4056 Simple prefetches will be joined automatically, so there is no need
4057 for a C<join> attribute in the above search.
4059 L</prefetch> can be used with the any of the relationship types and
4060 multiple prefetches can be specified together. Below is a more complex
4061 example that prefetches a CD's artist, its liner notes (if present),
4062 the cover image, the tracks on that cd, and the guests on those
4066 My::Schema::CD->belongs_to( artist => 'My::Schema::Artist' );
4067 My::Schema::CD->might_have( liner_note => 'My::Schema::LinerNotes' );
4068 My::Schema::CD->has_one( cover_image => 'My::Schema::Artwork' );
4069 My::Schema::CD->has_many( tracks => 'My::Schema::Track' );
4071 My::Schema::Artist->belongs_to( record_label => 'My::Schema::RecordLabel' );
4073 My::Schema::Track->has_many( guests => 'My::Schema::Guest' );
4076 my $rs = $schema->resultset('CD')->search(
4080 { artist => 'record_label'}, # belongs_to => belongs_to
4081 'liner_note', # might_have
4082 'cover_image', # has_one
4083 { tracks => 'guests' }, # has_many => has_many
4088 This will produce SQL like the following:
4090 SELECT cd.*, artist.*, record_label.*, liner_note.*, cover_image.*,
4094 ON artist.artistid = me.artistid
4095 JOIN record_label record_label
4096 ON record_label.labelid = artist.labelid
4097 LEFT JOIN track tracks
4098 ON tracks.cdid = me.cdid
4099 LEFT JOIN guest guests
4100 ON guests.trackid = track.trackid
4101 LEFT JOIN liner_notes liner_note
4102 ON liner_note.cdid = me.cdid
4103 JOIN cd_artwork cover_image
4104 ON cover_image.cdid = me.cdid
4107 Now the C<artist>, C<record_label>, C<liner_note>, C<cover_image>,
4108 C<tracks>, and C<guests> of the CD will all be available through the
4109 relationship accessors without the need for additional queries to the
4112 However, there is one caveat to be observed: it can be dangerous to
4113 prefetch more than one L<has_many|DBIx::Class::Relationship/has_many>
4114 relationship on a given level. e.g.:
4116 my $rs = $schema->resultset('CD')->search(
4120 'tracks', # has_many
4121 { cd_to_producer => 'producer' }, # has_many => belongs_to (i.e. m2m)
4126 The collapser currently can't identify duplicate tuples for multiple
4127 L<has_many|DBIx::Class::Relationship/has_many> relationships and as a
4128 result the second L<has_many|DBIx::Class::Relationship/has_many>
4129 relation could contain redundant objects.
4131 =head3 Using L</prefetch> with L</join>
4133 L</prefetch> implies a L</join> with the equivalent argument, and is
4134 properly merged with any existing L</join> specification. So the
4137 my $rs = $schema->resultset('CD')->search(
4138 {'record_label.name' => 'Music Product Ltd.'},
4140 join => {artist => 'record_label'},
4141 prefetch => 'artist',
4145 ... will work, searching on the record label's name, but only
4146 prefetching the C<artist>.
4148 =head3 Using L</prefetch> with L</select> / L</+select> / L</as> / L</+as>
4150 L</prefetch> implies a L</+select>/L</+as> with the fields of the
4151 prefetched relations. So given:
4153 my $rs = $schema->resultset('CD')->search(
4156 select => ['cd.title'],
4158 prefetch => 'artist',
4162 The L</select> becomes: C<'cd.title', 'artist.*'> and the L</as>
4163 becomes: C<'cd_title', 'artist.*'>.
4167 Prefetch does a lot of deep magic. As such, it may not behave exactly
4168 as you might expect.
4174 Prefetch uses the L</cache> to populate the prefetched relationships. This
4175 may or may not be what you want.
4179 If you specify a condition on a prefetched relationship, ONLY those
4180 rows that match the prefetched condition will be fetched into that relationship.
4181 This means that adding prefetch to a search() B<may alter> what is returned by
4182 traversing a relationship. So, if you have C<< Artist->has_many(CDs) >> and you do
4184 my $artist_rs = $schema->resultset('Artist')->search({
4190 my $count = $artist_rs->first->cds->count;
4192 my $artist_rs_prefetch = $artist_rs->search( {}, { prefetch => 'cds' } );
4194 my $prefetch_count = $artist_rs_prefetch->first->cds->count;
4196 cmp_ok( $count, '==', $prefetch_count, "Counts should be the same" );
4198 that cmp_ok() may or may not pass depending on the datasets involved. This
4199 behavior may or may not survive the 0.09 transition.
4207 =item Value: $source_alias
4211 Sets the source alias for the query. Normally, this defaults to C<me>, but
4212 nested search queries (sub-SELECTs) might need specific aliases set to
4213 reference inner queries. For example:
4216 ->related_resultset('CDs')
4217 ->related_resultset('Tracks')
4219 'track.id' => { -ident => 'none_search.id' },
4223 my $ids = $self->search({
4226 alias => 'none_search',
4227 group_by => 'none_search.id',
4228 })->get_column('id')->as_query;
4230 $self->search({ id => { -in => $ids } })
4232 This attribute is directly tied to L</current_source_alias>.
4242 Makes the resultset paged and specifies the page to retrieve. Effectively
4243 identical to creating a non-pages resultset and then calling ->page($page)
4246 If L</rows> attribute is not specified it defaults to 10 rows per page.
4248 When you have a paged resultset, L</count> will only return the number
4249 of rows in the page. To get the total, use the L</pager> and call
4250 C<total_entries> on it.
4260 Specifies the maximum number of rows for direct retrieval or the number of
4261 rows per page if the page attribute or method is used.
4267 =item Value: $offset
4271 Specifies the (zero-based) row number for the first row to be returned, or the
4272 of the first row of the first page if paging is used.
4274 =head2 software_limit
4278 =item Value: (0 | 1)
4282 When combined with L</rows> and/or L</offset> the generated SQL will not
4283 include any limit dialect stanzas. Instead the entire result will be selected
4284 as if no limits were specified, and DBIC will perform the limit locally, by
4285 artificially advancing and finishing the resulting L</cursor>.
4287 This is the recommended way of performing resultset limiting when no sane RDBMS
4288 implementation is available (e.g.
4289 L<Sybase ASE|DBIx::Class::Storage::DBI::Sybase::ASE> using the
4290 L<Generic Sub Query|DBIx::Class::SQLMaker::LimitDialects/GenericSubQ> hack)
4296 =item Value: \@columns
4300 A arrayref of columns to group by. Can include columns of joined tables.
4302 group_by => [qw/ column1 column2 ... /]
4308 =item Value: $condition
4312 HAVING is a select statement attribute that is applied between GROUP BY and
4313 ORDER BY. It is applied to the after the grouping calculations have been
4316 having => { 'count_employee' => { '>=', 100 } }
4318 or with an in-place function in which case literal SQL is required:
4320 having => \[ 'count(employee) >= ?', [ count => 100 ] ]
4326 =item Value: (0 | 1)
4330 Set to 1 to group by all columns. If the resultset already has a group_by
4331 attribute, this setting is ignored and an appropriate warning is issued.
4337 Adds to the WHERE clause.
4339 # only return rows WHERE deleted IS NULL for all searches
4340 __PACKAGE__->resultset_attributes({ where => { deleted => undef } });
4342 Can be overridden by passing C<< { where => undef } >> as an attribute
4345 For more complicated where clauses see L<SQL::Abstract/WHERE CLAUSES>.
4351 Set to 1 to cache search results. This prevents extra SQL queries if you
4352 revisit rows in your ResultSet:
4354 my $resultset = $schema->resultset('Artist')->search( undef, { cache => 1 } );
4356 while( my $artist = $resultset->next ) {
4360 $rs->first; # without cache, this would issue a query
4362 By default, searches are not cached.
4364 For more examples of using these attributes, see
4365 L<DBIx::Class::Manual::Cookbook>.
4371 =item Value: ( 'update' | 'shared' | \$scalar )
4375 Set to 'update' for a SELECT ... FOR UPDATE or 'shared' for a SELECT
4376 ... FOR SHARED. If \$scalar is passed, this is taken directly and embedded in the
4379 =head1 DBIC BIND VALUES
4381 Because DBIC may need more information to bind values than just the column name
4382 and value itself, it uses a special format for both passing and receiving bind
4383 values. Each bind value should be composed of an arrayref of
4384 C<< [ \%args => $val ] >>. The format of C<< \%args >> is currently:
4390 If present (in any form), this is what is being passed directly to bind_param.
4391 Note that different DBD's expect different bind args. (e.g. DBD::SQLite takes
4392 a single numerical type, while DBD::Pg takes a hashref if bind options.)
4394 If this is specified, all other bind options described below are ignored.
4398 If present, this is used to infer the actual bind attribute by passing to
4399 C<< $resolved_storage->bind_attribute_by_data_type() >>. Defaults to the
4400 "data_type" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>.
4402 Note that the data type is somewhat freeform (hence the sqlt_ prefix);
4403 currently drivers are expected to "Do the Right Thing" when given a common
4404 datatype name. (Not ideal, but that's what we got at this point.)
4408 Currently used to correctly allocate buffers for bind_param_inout().
4409 Defaults to "size" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>,
4410 or to a sensible value based on the "data_type".
4414 Used to fill in missing sqlt_datatype and sqlt_size attributes (if they are
4415 explicitly specified they are never overriden). Also used by some weird DBDs,
4416 where the column name should be available at bind_param time (e.g. Oracle).
4420 For backwards compatibility and convenience, the following shortcuts are
4423 [ $name => $val ] === [ { dbic_colname => $name }, $val ]
4424 [ \$dt => $val ] === [ { sqlt_datatype => $dt }, $val ]
4425 [ undef, $val ] === [ {}, $val ]
4427 =head1 AUTHOR AND CONTRIBUTORS
4429 See L<AUTHOR|DBIx::Class/AUTHOR> and L<CONTRIBUTORS|DBIx::Class/CONTRIBUTORS> in DBIx::Class
4433 You may distribute this code under the same terms as Perl itself.