1 package DBIx::Class::ResultSet;
5 use base qw/DBIx::Class/;
6 use Carp::Clan qw/^DBIx::Class/;
7 use DBIx::Class::Exception;
8 use DBIx::Class::ResultSetColumn;
9 use Scalar::Util qw/blessed weaken/;
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 __PACKAGE__->mk_group_accessors('simple' => qw/_result_class result_source/);
32 DBIx::Class::ResultSet - Represents a query used for fetching a set of results.
36 my $users_rs = $schema->resultset('User');
37 while( $user = $users_rs->next) {
38 print $user->username;
41 my $registered_users_rs = $schema->resultset('User')->search({ registered => 1 });
42 my @cds_in_2005 = $schema->resultset('CD')->search({ year => 2005 })->all();
46 A ResultSet is an object which stores a set of conditions representing
47 a query. It is the backbone of DBIx::Class (i.e. the really
48 important/useful bit).
50 No SQL is executed on the database when a ResultSet is created, it
51 just stores all the conditions needed to create the query.
53 A basic ResultSet representing the data of an entire table is returned
54 by calling C<resultset> on a L<DBIx::Class::Schema> and passing in a
55 L<Source|DBIx::Class::Manual::Glossary/Source> name.
57 my $users_rs = $schema->resultset('User');
59 A new ResultSet is returned from calling L</search> on an existing
60 ResultSet. The new one will contain all the conditions of the
61 original, plus any new conditions added in the C<search> call.
63 A ResultSet also incorporates an implicit iterator. L</next> and L</reset>
64 can be used to walk through all the L<DBIx::Class::Row>s the ResultSet
67 The query that the ResultSet represents is B<only> executed against
68 the database when these methods are called:
69 L</find>, L</next>, L</all>, L</first>, L</single>, L</count>.
71 If a resultset is used in a numeric context it returns the L</count>.
72 However, if it is used in a boolean context it is B<always> true. So if
73 you want to check if a resultset has any results, you must use C<if $rs
78 =head2 Chaining resultsets
80 Let's say you've got a query that needs to be run to return some data
81 to the user. But, you have an authorization system in place that
82 prevents certain users from seeing certain information. So, you want
83 to construct the basic query in one method, but add constraints to it in
88 my $request = $self->get_request; # Get a request object somehow.
89 my $schema = $self->get_schema; # Get the DBIC schema object somehow.
91 my $cd_rs = $schema->resultset('CD')->search({
92 title => $request->param('title'),
93 year => $request->param('year'),
96 $cd_rs = $self->apply_security_policy( $cd_rs );
101 sub apply_security_policy {
110 =head3 Resolving conditions and attributes
112 When a resultset is chained from another resultset, conditions and
113 attributes with the same keys need resolving.
115 L</join>, L</prefetch>, L</+select>, L</+as> attributes are merged
116 into the existing ones from the original resultset.
118 The L</where> and L</having> attributes, and any search conditions, are
119 merged with an SQL C<AND> to the existing condition from the original
122 All other attributes are overridden by any new ones supplied in the
125 =head2 Multiple queries
127 Since a resultset just defines a query, you can do all sorts of
128 things with it with the same object.
130 # Don't hit the DB yet.
131 my $cd_rs = $schema->resultset('CD')->search({
132 title => 'something',
136 # Each of these hits the DB individually.
137 my $count = $cd_rs->count;
138 my $most_recent = $cd_rs->get_column('date_released')->max();
139 my @records = $cd_rs->all;
141 And it's not just limited to SELECT statements.
147 $cd_rs->create({ artist => 'Fred' });
149 Which is the same as:
151 $schema->resultset('CD')->create({
152 title => 'something',
157 See: L</search>, L</count>, L</get_column>, L</all>, L</create>.
165 =item Arguments: $source, \%$attrs
167 =item Return Value: $rs
171 The resultset constructor. Takes a source object (usually a
172 L<DBIx::Class::ResultSourceProxy::Table>) and an attribute hash (see
173 L</ATTRIBUTES> below). Does not perform any queries -- these are
174 executed as needed by the other methods.
176 Generally you won't need to construct a resultset manually. You'll
177 automatically get one from e.g. a L</search> called in scalar context:
179 my $rs = $schema->resultset('CD')->search({ title => '100th Window' });
181 IMPORTANT: If called on an object, proxies to new_result instead so
183 my $cd = $schema->resultset('CD')->new({ title => 'Spoon' });
185 will return a CD object, not a ResultSet.
191 return $class->new_result(@_) if ref $class;
193 my ($source, $attrs) = @_;
194 $source = $source->resolve
195 if $source->isa('DBIx::Class::ResultSourceHandle');
196 $attrs = { %{$attrs||{}} };
198 if ($attrs->{page}) {
199 $attrs->{rows} ||= 10;
202 $attrs->{alias} ||= 'me';
205 result_source => $source,
206 cond => $attrs->{where},
212 $attrs->{result_class} || $source->result_class
222 =item Arguments: $cond, \%attrs?
224 =item Return Value: $resultset (scalar context), @row_objs (list context)
228 my @cds = $cd_rs->search({ year => 2001 }); # "... WHERE year = 2001"
229 my $new_rs = $cd_rs->search({ year => 2005 });
231 my $new_rs = $cd_rs->search([ { year => 2005 }, { year => 2004 } ]);
232 # year = 2005 OR year = 2004
234 If you need to pass in additional attributes but no additional condition,
235 call it as C<search(undef, \%attrs)>.
237 # "SELECT name, artistid FROM $artist_table"
238 my @all_artists = $schema->resultset('Artist')->search(undef, {
239 columns => [qw/name artistid/],
242 For a list of attributes that can be passed to C<search>, see
243 L</ATTRIBUTES>. For more examples of using this function, see
244 L<Searching|DBIx::Class::Manual::Cookbook/Searching>. For a complete
245 documentation for the first argument, see L<SQL::Abstract>.
247 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
251 Note that L</search> does not process/deflate any of the values passed in the
252 L<SQL::Abstract>-compatible search condition structure. This is unlike other
253 condition-bound methods L</new>, L</create> and L</find>. The user must ensure
254 manually that any value passed to this method will stringify to something the
255 RDBMS knows how to deal with. A notable example is the handling of L<DateTime>
256 objects, for more info see:
257 L<DBIx::Class::Manual::Cookbook/Formatting_DateTime_objects_in_queries>.
263 my $rs = $self->search_rs( @_ );
268 elsif (defined wantarray) {
272 # we can be called by a relationship helper, which in
273 # turn may be called in void context due to some braindead
274 # overload or whatever else the user decided to be clever
275 # at this particular day. Thus limit the exception to
276 # external code calls only
277 $self->throw_exception ('->search is *not* a mutator, calling it in void context makes no sense')
278 if (caller)[0] !~ /^\QDBIx::Class::/;
288 =item Arguments: $cond, \%attrs?
290 =item Return Value: $resultset
294 This method does the same exact thing as search() except it will
295 always return a resultset, even in list context.
299 my $callsites_warned;
303 # Special-case handling for (undef, undef).
304 if ( @_ == 2 && !defined $_[1] && !defined $_[0] ) {
310 if (ref $_[-1] eq 'HASH') {
311 # copy for _normalize_selection
312 $call_attrs = { %{ pop @_ } };
314 elsif (! defined $_[-1] ) {
315 pop @_; # search({}, undef)
319 # see if we can keep the cache (no $rs changes)
321 my %safe = (alias => 1, cache => 1);
322 if ( ! List::Util::first { !$safe{$_} } keys %$call_attrs and (
325 ref $_[0] eq 'HASH' && ! keys %{$_[0]}
327 ref $_[0] eq 'ARRAY' && ! @{$_[0]}
329 $cache = $self->get_cache;
332 my $rsrc = $self->result_source;
334 my $old_attrs = { %{$self->{attrs}} };
335 my $old_having = delete $old_attrs->{having};
336 my $old_where = delete $old_attrs->{where};
338 my $new_attrs = { %$old_attrs };
340 # take care of call attrs (only if anything is changing)
341 if (keys %$call_attrs) {
343 $self->throw_exception ('_trailing_select is not a public attribute - do not use it in search()')
344 if ( exists $call_attrs->{_trailing_select} or exists $call_attrs->{'+_trailing_select'} );
346 my @selector_attrs = qw/select as columns cols +select +as +columns include_columns _trailing_select +_trailing_select/;
348 # Normalize the selector list (operates on the passed-in attr structure)
349 # Need to do it on every chain instead of only once on _resolved_attrs, in
350 # order to separate 'as'-ed from blind 'select's
351 $self->_normalize_selection ($call_attrs);
353 # start with blind overwriting merge, exclude selector attrs
354 $new_attrs = { %{$old_attrs}, %{$call_attrs} };
355 delete @{$new_attrs}{@selector_attrs};
357 # reset the current selector list if new selectors are supplied
358 if (List::Util::first { exists $call_attrs->{$_} } qw/columns cols select as/) {
359 delete @{$old_attrs}{@selector_attrs};
362 for (@selector_attrs) {
363 $new_attrs->{$_} = $self->_merge_attr($old_attrs->{$_}, $call_attrs->{$_})
364 if ( exists $old_attrs->{$_} or exists $call_attrs->{$_} );
367 # older deprecated name, use only if {columns} is not there
368 if (my $c = delete $new_attrs->{cols}) {
369 if ($new_attrs->{columns}) {
370 carp "Resultset specifies both the 'columns' and the legacy 'cols' attributes - ignoring 'cols'";
373 $new_attrs->{columns} = $c;
378 # join/prefetch use their own crazy merging heuristics
379 foreach my $key (qw/join prefetch/) {
380 $new_attrs->{$key} = $self->_merge_joinpref_attr($old_attrs->{$key}, $call_attrs->{$key})
381 if exists $call_attrs->{$key};
384 # stack binds together
385 $new_attrs->{bind} = [ @{ $old_attrs->{bind} || [] }, @{ $call_attrs->{bind} || [] } ];
389 # rip apart the rest of @_, parse a condition
392 if (ref $_[0] eq 'HASH') {
393 (keys %{$_[0]}) ? $_[0] : undef
399 $self->throw_exception('Odd number of arguments to search')
407 if( @_ > 1 and ! $rsrc->result_class->isa('DBIx::Class::CDBICompat') ) {
408 # determine callsite obeying Carp::Clan rules (fucking ugly but don't have better ideas)
411 local $SIG{__WARN__} = sub { $w = shift };
415 carp 'search( %condition ) is deprecated, use search( \%condition ) instead'
416 unless $callsites_warned->{$callsite}++;
419 for ($old_where, $call_cond) {
421 $new_attrs->{where} = $self->_stack_cond (
422 $_, $new_attrs->{where}
427 if (defined $old_having) {
428 $new_attrs->{having} = $self->_stack_cond (
429 $old_having, $new_attrs->{having}
433 my $rs = (ref $self)->new($rsrc, $new_attrs);
435 $rs->set_cache($cache) if ($cache);
440 sub _normalize_selection {
441 my ($self, $attrs) = @_;
444 $attrs->{'+columns'} = $self->_merge_attr($attrs->{'+columns'}, delete $attrs->{include_columns})
445 if exists $attrs->{include_columns};
447 # Keep the X vs +X separation until _resolved_attrs time - this allows to
448 # delay the decision on whether to use a default select list ($rsrc->columns)
449 # allowing stuff like the remove_columns helper to work
451 # select/as +select/+as pairs need special handling - the amount of select/as
452 # elements in each pair does *not* have to be equal (think multicolumn
453 # selectors like distinct(foo, bar) ). If the selector is bare (no 'as'
454 # supplied at all) - try to infer the alias, either from the -as parameter
455 # of the selector spec, or use the parameter whole if it looks like a column
456 # name (ugly legacy heuristic). If all fails - leave the selector bare (which
457 # is ok as well), but transport it over a separate attribute to make sure it is
458 # the last thing in the select list, thus unable to throw off the corresponding
460 for my $pref ('', '+') {
462 my ($sel, $as) = map {
463 my $key = "${pref}${_}";
465 my $val = [ ref $attrs->{$key} eq 'ARRAY'
467 : $attrs->{$key} || ()
469 delete $attrs->{$key};
473 if (! @$as and ! @$sel ) {
476 elsif (@$as and ! @$sel) {
477 $self->throw_exception(
478 "Unable to handle ${pref}as specification (@$as) without a corresponding ${pref}select"
482 # no as part supplied at all - try to deduce
483 # if any @$as has been supplied we assume the user knows what (s)he is doing
484 # and blindly keep stacking up pieces
485 my (@new_sel, @new_trailing);
487 if ( ref $_ eq 'HASH' and exists $_->{-as} ) {
488 push @$as, $_->{-as};
491 # assume any plain no-space, no-parenthesis string to be a column spec
492 # FIXME - this is retarded but is necessary to support shit like 'count(foo)'
493 elsif ( ! ref $_ and $_ =~ /^ [^\s\(\)]+ $/x) {
497 # if all else fails - shove the selection to the trailing stack and move on
499 push @new_trailing, $_;
504 $attrs->{"${pref}_trailing_select"} = $self->_merge_attr($attrs->{"${pref}_trailing_select"}, \@new_trailing)
507 elsif (@$as < @$sel) {
508 $self->throw_exception(
509 "Unable to handle an ${pref}as specification (@$as) with less elements than the corresponding ${pref}select"
513 # now see what the result for this pair looks like:
516 # if balanced - treat as a columns entry
517 $attrs->{"${pref}columns"} = $self->_merge_attr(
518 $attrs->{"${pref}columns"},
519 [ map { +{ $as->[$_] => $sel->[$_] } } ( 0 .. $#$as ) ]
523 # unbalanced - shove in select/as, not subject to deduplication in _resolved_attrs
524 $attrs->{"${pref}select"} = $self->_merge_attr($attrs->{"${pref}select"}, $sel);
525 $attrs->{"${pref}as"} = $self->_merge_attr($attrs->{"${pref}as"}, $as);
532 my ($self, $left, $right) = @_;
533 if (defined $left xor defined $right) {
534 return defined $left ? $left : $right;
536 elsif (defined $left) {
537 return { -and => [ map
538 { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ }
546 =head2 search_literal
550 =item Arguments: $sql_fragment, @bind_values
552 =item Return Value: $resultset (scalar context), @row_objs (list context)
556 my @cds = $cd_rs->search_literal('year = ? AND title = ?', qw/2001 Reload/);
557 my $newrs = $artist_rs->search_literal('name = ?', 'Metallica');
559 Pass a literal chunk of SQL to be added to the conditional part of the
562 CAVEAT: C<search_literal> is provided for Class::DBI compatibility and should
563 only be used in that context. C<search_literal> is a convenience method.
564 It is equivalent to calling $schema->search(\[]), but if you want to ensure
565 columns are bound correctly, use C<search>.
567 Example of how to use C<search> instead of C<search_literal>
569 my @cds = $cd_rs->search_literal('cdid = ? AND (artist = ? OR artist = ?)', (2, 1, 2));
570 my @cds = $cd_rs->search(\[ 'cdid = ? AND (artist = ? OR artist = ?)', [ 'cdid', 2 ], [ 'artist', 1 ], [ 'artist', 2 ] ]);
573 See L<DBIx::Class::Manual::Cookbook/Searching> and
574 L<DBIx::Class::Manual::FAQ/Searching> for searching techniques that do not
575 require C<search_literal>.
580 my ($self, $sql, @bind) = @_;
582 if ( @bind && ref($bind[-1]) eq 'HASH' ) {
585 return $self->search(\[ $sql, map [ __DUMMY__ => $_ ], @bind ], ($attr || () ));
592 =item Arguments: \%columns_values | @pk_values, \%attrs?
594 =item Return Value: $row_object | undef
598 Finds and returns a single row based on supplied criteria. Takes either a
599 hashref with the same format as L</create> (including inference of foreign
600 keys from related objects), or a list of primary key values in the same
601 order as the L<primary columns|DBIx::Class::ResultSource/primary_columns>
602 declaration on the L</result_source>.
604 In either case an attempt is made to combine conditions already existing on
605 the resultset with the condition passed to this method.
607 To aid with preparing the correct query for the storage you may supply the
608 C<key> attribute, which is the name of a
609 L<unique constraint|DBIx::Class::ResultSource/add_unique_constraint> (the
610 unique constraint corresponding to the
611 L<primary columns|DBIx::Class::ResultSource/primary_columns> is always named
612 C<primary>). If the C<key> attribute has been supplied, and DBIC is unable
613 to construct a query that satisfies the named unique constraint fully (
614 non-NULL values for each column member of the constraint) an exception is
617 If no C<key> is specified, the search is carried over all unique constraints
618 which are fully defined by the available condition.
620 If no such constraint is found, C<find> currently defaults to a simple
621 C<< search->(\%column_values) >> which may or may not do what you expect.
622 Note that this fallback behavior may be deprecated in further versions. If
623 you need to search with arbitrary conditions - use L</search>. If the query
624 resulting from this fallback produces more than one row, a warning to the
625 effect is issued, though only the first row is constructed and returned as
628 In addition to C<key>, L</find> recognizes and applies standard
629 L<resultset attributes|/ATTRIBUTES> in the same way as L</search> does.
631 Note that if you have extra concerns about the correctness of the resulting
632 query you need to specify the C<key> attribute and supply the entire condition
633 as an argument to find (since it is not always possible to perform the
634 combination of the resultset condition with the supplied one, especially if
635 the resultset condition contains literal sql).
637 For example, to find a row by its primary key:
639 my $cd = $schema->resultset('CD')->find(5);
641 You can also find a row by a specific unique constraint:
643 my $cd = $schema->resultset('CD')->find(
645 artist => 'Massive Attack',
646 title => 'Mezzanine',
648 { key => 'cd_artist_title' }
651 See also L</find_or_create> and L</update_or_create>.
657 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
659 my $rsrc = $self->result_source;
661 # Parse out the condition from input
663 if (ref $_[0] eq 'HASH') {
664 $call_cond = { %{$_[0]} };
667 my $constraint = exists $attrs->{key} ? $attrs->{key} : 'primary';
668 my @c_cols = $rsrc->unique_constraint_columns($constraint);
670 $self->throw_exception(
671 "No constraint columns, maybe a malformed '$constraint' constraint?"
674 $self->throw_exception (
675 'find() expects either a column/value hashref, or a list of values '
676 . "corresponding to the columns of the specified unique constraint '$constraint'"
677 ) unless @c_cols == @_;
680 @{$call_cond}{@c_cols} = @_;
684 for my $key (keys %$call_cond) {
686 my $keyref = ref($call_cond->{$key})
688 my $relinfo = $rsrc->relationship_info($key)
690 my $val = delete $call_cond->{$key};
692 next if $keyref eq 'ARRAY'; # has_many for multi_create
694 my $rel_q = $rsrc->_resolve_condition(
695 $relinfo->{cond}, $val, $key
697 die "Can't handle complex relationship conditions in find" if ref($rel_q) ne 'HASH';
698 @related{keys %$rel_q} = values %$rel_q;
702 # relationship conditions take precedence (?)
703 @{$call_cond}{keys %related} = values %related;
705 my $alias = exists $attrs->{alias} ? $attrs->{alias} : $self->{attrs}{alias};
707 if (exists $attrs->{key}) {
708 $final_cond = $self->_qualify_cond_columns (
710 $self->_build_unique_cond (
718 elsif ($self->{attrs}{accessor} and $self->{attrs}{accessor} eq 'single') {
719 # This means that we got here after a merger of relationship conditions
720 # in ::Relationship::Base::search_related (the row method), and furthermore
721 # the relationship is of the 'single' type. This means that the condition
722 # provided by the relationship (already attached to $self) is sufficient,
723 # as there can be only one row in the database that would satisfy the
727 # no key was specified - fall down to heuristics mode:
728 # run through all unique queries registered on the resultset, and
729 # 'OR' all qualifying queries together
730 my (@unique_queries, %seen_column_combinations);
731 for my $c_name ($rsrc->unique_constraint_names) {
732 next if $seen_column_combinations{
733 join "\x00", sort $rsrc->unique_constraint_columns($c_name)
736 push @unique_queries, try {
737 $self->_build_unique_cond ($c_name, $call_cond, 'croak_on_nulls')
741 $final_cond = @unique_queries
742 ? [ map { $self->_qualify_cond_columns($_, $alias) } @unique_queries ]
743 : $self->_non_unique_find_fallback ($call_cond, $attrs)
747 # Run the query, passing the result_class since it should propagate for find
748 my $rs = $self->search ($final_cond, {result_class => $self->result_class, %$attrs});
749 if (keys %{$rs->_resolved_attrs->{collapse}}) {
751 carp "Query returned more than one row" if $rs->next;
759 # This is a stop-gap method as agreed during the discussion on find() cleanup:
760 # http://lists.scsys.co.uk/pipermail/dbix-class/2010-October/009535.html
762 # It is invoked when find() is called in legacy-mode with insufficiently-unique
763 # condition. It is provided for overrides until a saner way forward is devised
765 # *NOTE* This is not a public method, and it's *GUARANTEED* to disappear down
766 # the road. Please adjust your tests accordingly to catch this situation early
767 # DBIx::Class::ResultSet->can('_non_unique_find_fallback') is reasonable
769 # The method will not be removed without an adequately complete replacement
770 # for strict-mode enforcement
771 sub _non_unique_find_fallback {
772 my ($self, $cond, $attrs) = @_;
774 return $self->_qualify_cond_columns(
776 exists $attrs->{alias}
778 : $self->{attrs}{alias}
783 sub _qualify_cond_columns {
784 my ($self, $cond, $alias) = @_;
786 my %aliased = %$cond;
787 for (keys %aliased) {
788 $aliased{"$alias.$_"} = delete $aliased{$_}
795 my $callsites_warned_ucond;
796 sub _build_unique_cond {
797 my ($self, $constraint_name, $extra_cond, $croak_on_null) = @_;
799 my @c_cols = $self->result_source->unique_constraint_columns($constraint_name);
801 # combination may fail if $self->{cond} is non-trivial
802 my ($final_cond) = try {
803 $self->_merge_with_rscond ($extra_cond)
808 # trim out everything not in $columns
809 $final_cond = { map {
810 exists $final_cond->{$_}
811 ? ( $_ => $final_cond->{$_} )
815 if (my @missing = grep
816 { ! ($croak_on_null ? defined $final_cond->{$_} : exists $final_cond->{$_}) }
819 $self->throw_exception( sprintf ( "Unable to satisfy requested constraint '%s', no values for column(s): %s",
821 join (', ', map { "'$_'" } @missing),
828 !$ENV{DBIC_NULLABLE_KEY_NOWARN}
830 my @undefs = grep { ! defined $final_cond->{$_} } (keys %$final_cond)
834 local $SIG{__WARN__} = sub { $w = shift };
840 "NULL/undef values supplied for requested unique constraint '%s' (NULL "
841 . 'values in column(s): %s). This is almost certainly not what you wanted, '
842 . 'though you can set DBIC_NULLABLE_KEY_NOWARN to disable this warning.',
844 join (', ', map { "'$_'" } @undefs),
845 )) unless $callsites_warned_ucond->{$callsite}++;
851 =head2 search_related
855 =item Arguments: $rel, $cond, \%attrs?
857 =item Return Value: $new_resultset
861 $new_rs = $cd_rs->search_related('artist', {
865 Searches the specified relationship, optionally specifying a condition and
866 attributes for matching records. See L</ATTRIBUTES> for more information.
871 return shift->related_resultset(shift)->search(@_);
874 =head2 search_related_rs
876 This method works exactly the same as search_related, except that
877 it guarantees a resultset, even in list context.
881 sub search_related_rs {
882 return shift->related_resultset(shift)->search_rs(@_);
889 =item Arguments: none
891 =item Return Value: $cursor
895 Returns a storage-driven cursor to the given resultset. See
896 L<DBIx::Class::Cursor> for more information.
903 my $attrs = $self->_resolved_attrs_copy;
905 return $self->{cursor}
906 ||= $self->result_source->storage->select($attrs->{from}, $attrs->{select},
907 $attrs->{where},$attrs);
914 =item Arguments: $cond?
916 =item Return Value: $row_object | undef
920 my $cd = $schema->resultset('CD')->single({ year => 2001 });
922 Inflates the first result without creating a cursor if the resultset has
923 any records in it; if not returns C<undef>. Used by L</find> as a lean version
926 While this method can take an optional search condition (just like L</search>)
927 being a fast-code-path it does not recognize search attributes. If you need to
928 add extra joins or similar, call L</search> and then chain-call L</single> on the
929 L<DBIx::Class::ResultSet> returned.
935 As of 0.08100, this method enforces the assumption that the preceding
936 query returns only one row. If more than one row is returned, you will receive
939 Query returned more than one row
941 In this case, you should be using L</next> or L</find> instead, or if you really
942 know what you are doing, use the L</rows> attribute to explicitly limit the size
945 This method will also throw an exception if it is called on a resultset prefetching
946 has_many, as such a prefetch implies fetching multiple rows from the database in
947 order to assemble the resulting object.
954 my ($self, $where) = @_;
956 $self->throw_exception('single() only takes search conditions, no attributes. You want ->search( $cond, $attrs )->single()');
959 my $attrs = $self->_resolved_attrs_copy;
961 if (keys %{$attrs->{collapse}}) {
962 $self->throw_exception(
963 'single() can not be used on resultsets prefetching has_many. Use find( \%cond ) or next() instead'
968 if (defined $attrs->{where}) {
971 [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ }
972 $where, delete $attrs->{where} ]
975 $attrs->{where} = $where;
979 my @data = $self->result_source->storage->select_single(
980 $attrs->{from}, $attrs->{select},
981 $attrs->{where}, $attrs
984 return (@data ? ($self->_construct_object(@data))[0] : undef);
990 # Recursively collapse the query, accumulating values for each column.
992 sub _collapse_query {
993 my ($self, $query, $collapsed) = @_;
997 if (ref $query eq 'ARRAY') {
998 foreach my $subquery (@$query) {
999 next unless ref $subquery; # -or
1000 $collapsed = $self->_collapse_query($subquery, $collapsed);
1003 elsif (ref $query eq 'HASH') {
1004 if (keys %$query and (keys %$query)[0] eq '-and') {
1005 foreach my $subquery (@{$query->{-and}}) {
1006 $collapsed = $self->_collapse_query($subquery, $collapsed);
1010 foreach my $col (keys %$query) {
1011 my $value = $query->{$col};
1012 $collapsed->{$col}{$value}++;
1024 =item Arguments: $cond?
1026 =item Return Value: $resultsetcolumn
1030 my $max_length = $rs->get_column('length')->max;
1032 Returns a L<DBIx::Class::ResultSetColumn> instance for a column of the ResultSet.
1037 my ($self, $column) = @_;
1038 my $new = DBIx::Class::ResultSetColumn->new($self, $column);
1046 =item Arguments: $cond, \%attrs?
1048 =item Return Value: $resultset (scalar context), @row_objs (list context)
1052 # WHERE title LIKE '%blue%'
1053 $cd_rs = $rs->search_like({ title => '%blue%'});
1055 Performs a search, but uses C<LIKE> instead of C<=> as the condition. Note
1056 that this is simply a convenience method retained for ex Class::DBI users.
1057 You most likely want to use L</search> with specific operators.
1059 For more information, see L<DBIx::Class::Manual::Cookbook>.
1061 This method is deprecated and will be removed in 0.09. Use L</search()>
1062 instead. An example conversion is:
1064 ->search_like({ foo => 'bar' });
1068 ->search({ foo => { like => 'bar' } });
1075 'search_like() is deprecated and will be removed in DBIC version 0.09.'
1076 .' Instead use ->search({ x => { -like => "y%" } })'
1077 .' (note the outer pair of {}s - they are important!)'
1079 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
1080 my $query = ref $_[0] eq 'HASH' ? { %{shift()} }: {@_};
1081 $query->{$_} = { 'like' => $query->{$_} } for keys %$query;
1082 return $class->search($query, { %$attrs });
1089 =item Arguments: $first, $last
1091 =item Return Value: $resultset (scalar context), @row_objs (list context)
1095 Returns a resultset or object list representing a subset of elements from the
1096 resultset slice is called on. Indexes are from 0, i.e., to get the first
1097 three records, call:
1099 my ($one, $two, $three) = $rs->slice(0, 2);
1104 my ($self, $min, $max) = @_;
1105 my $attrs = {}; # = { %{ $self->{attrs} || {} } };
1106 $attrs->{offset} = $self->{attrs}{offset} || 0;
1107 $attrs->{offset} += $min;
1108 $attrs->{rows} = ($max ? ($max - $min + 1) : 1);
1109 return $self->search(undef, $attrs);
1110 #my $slice = (ref $self)->new($self->result_source, $attrs);
1111 #return (wantarray ? $slice->all : $slice);
1118 =item Arguments: none
1120 =item Return Value: $result | undef
1124 Returns the next element in the resultset (C<undef> is there is none).
1126 Can be used to efficiently iterate over records in the resultset:
1128 my $rs = $schema->resultset('CD')->search;
1129 while (my $cd = $rs->next) {
1133 Note that you need to store the resultset object, and call C<next> on it.
1134 Calling C<< resultset('Table')->next >> repeatedly will always return the
1135 first record from the resultset.
1141 if (my $cache = $self->get_cache) {
1142 $self->{all_cache_position} ||= 0;
1143 return $cache->[$self->{all_cache_position}++];
1145 if ($self->{attrs}{cache}) {
1146 delete $self->{pager};
1147 $self->{all_cache_position} = 1;
1148 return ($self->all)[0];
1150 if ($self->{stashed_objects}) {
1151 my $obj = shift(@{$self->{stashed_objects}});
1152 delete $self->{stashed_objects} unless @{$self->{stashed_objects}};
1156 exists $self->{stashed_row}
1157 ? @{delete $self->{stashed_row}}
1158 : $self->cursor->next
1160 return undef unless (@row);
1161 my ($row, @more) = $self->_construct_object(@row);
1162 $self->{stashed_objects} = \@more if @more;
1166 sub _construct_object {
1167 my ($self, @row) = @_;
1169 my $info = $self->_collapse_result($self->{_attrs}{as}, \@row)
1171 my @new = $self->result_class->inflate_result($self->result_source, @$info);
1172 @new = $self->{_attrs}{record_filter}->(@new)
1173 if exists $self->{_attrs}{record_filter};
1177 sub _collapse_result {
1178 my ($self, $as_proto, $row) = @_;
1182 # 'foo' => [ undef, 'foo' ]
1183 # 'foo.bar' => [ 'foo', 'bar' ]
1184 # 'foo.bar.baz' => [ 'foo.bar', 'baz' ]
1186 my @construct_as = map { [ (/^(?:(.*)\.)?([^.]+)$/) ] } @$as_proto;
1188 my %collapse = %{$self->{_attrs}{collapse}||{}};
1192 # if we're doing collapsing (has_many prefetch) we need to grab records
1193 # until the PK changes, so fill @pri_index. if not, we leave it empty so
1194 # we know we don't have to bother.
1196 # the reason for not using the collapse stuff directly is because if you
1197 # had for e.g. two artists in a row with no cds, the collapse info for
1198 # both would be NULL (undef) so you'd lose the second artist
1200 # store just the index so we can check the array positions from the row
1201 # without having to contruct the full hash
1203 if (keys %collapse) {
1204 my %pri = map { ($_ => 1) } $self->result_source->_pri_cols;
1205 foreach my $i (0 .. $#construct_as) {
1206 next if defined($construct_as[$i][0]); # only self table
1207 if (delete $pri{$construct_as[$i][1]}) {
1208 push(@pri_index, $i);
1210 last unless keys %pri; # short circuit (Johnny Five Is Alive!)
1214 # no need to do an if, it'll be empty if @pri_index is empty anyway
1216 my %pri_vals = map { ($_ => $copy[$_]) } @pri_index;
1220 do { # no need to check anything at the front, we always want the first row
1224 foreach my $this_as (@construct_as) {
1225 $const{$this_as->[0]||''}{$this_as->[1]} = shift(@copy);
1228 push(@const_rows, \%const);
1230 } until ( # no pri_index => no collapse => drop straight out
1233 do { # get another row, stash it, drop out if different PK
1235 @copy = $self->cursor->next;
1236 $self->{stashed_row} = \@copy;
1238 # last thing in do block, counts as true if anything doesn't match
1240 # check xor defined first for NULL vs. NOT NULL then if one is
1241 # defined the other must be so check string equality
1244 (defined $pri_vals{$_} ^ defined $copy[$_])
1245 || (defined $pri_vals{$_} && ($pri_vals{$_} ne $copy[$_]))
1250 my $alias = $self->{attrs}{alias};
1257 foreach my $const (@const_rows) {
1258 scalar @const_keys or do {
1259 @const_keys = sort { length($a) <=> length($b) } keys %$const;
1261 foreach my $key (@const_keys) {
1264 my @parts = split(/\./, $key);
1266 my $data = $const->{$key};
1267 foreach my $p (@parts) {
1268 $target = $target->[1]->{$p} ||= [];
1270 if ($cur eq ".${key}" && (my @ckey = @{$collapse{$cur}||[]})) {
1271 # collapsing at this point and on final part
1272 my $pos = $collapse_pos{$cur};
1273 CK: foreach my $ck (@ckey) {
1274 if (!defined $pos->{$ck} || $pos->{$ck} ne $data->{$ck}) {
1275 $collapse_pos{$cur} = $data;
1276 delete @collapse_pos{ # clear all positioning for sub-entries
1277 grep { m/^\Q${cur}.\E/ } keys %collapse_pos
1284 if (exists $collapse{$cur}) {
1285 $target = $target->[-1];
1288 $target->[0] = $data;
1290 $info->[0] = $const->{$key};
1298 =head2 result_source
1302 =item Arguments: $result_source?
1304 =item Return Value: $result_source
1308 An accessor for the primary ResultSource object from which this ResultSet
1315 =item Arguments: $result_class?
1317 =item Return Value: $result_class
1321 An accessor for the class to use when creating row objects. Defaults to
1322 C<< result_source->result_class >> - which in most cases is the name of the
1323 L<"table"|DBIx::Class::Manual::Glossary/"ResultSource"> class.
1325 Note that changing the result_class will also remove any components
1326 that were originally loaded in the source class via
1327 L<DBIx::Class::ResultSource/load_components>. Any overloaded methods
1328 in the original source class will not run.
1333 my ($self, $result_class) = @_;
1334 if ($result_class) {
1335 unless (ref $result_class) { # don't fire this for an object
1336 $self->ensure_class_loaded($result_class);
1338 $self->_result_class($result_class);
1339 # THIS LINE WOULD BE A BUG - this accessor specifically exists to
1340 # permit the user to set result class on one result set only; it only
1341 # chains if provided to search()
1342 #$self->{attrs}{result_class} = $result_class if ref $self;
1344 $self->_result_class;
1351 =item Arguments: $cond, \%attrs??
1353 =item Return Value: $count
1357 Performs an SQL C<COUNT> with the same query as the resultset was built
1358 with to find the number of elements. Passing arguments is equivalent to
1359 C<< $rs->search ($cond, \%attrs)->count >>
1365 return $self->search(@_)->count if @_ and defined $_[0];
1366 return scalar @{ $self->get_cache } if $self->get_cache;
1368 my $attrs = $self->_resolved_attrs_copy;
1370 # this is a little optimization - it is faster to do the limit
1371 # adjustments in software, instead of a subquery
1372 my $rows = delete $attrs->{rows};
1373 my $offset = delete $attrs->{offset};
1376 if ($self->_has_resolved_attr (qw/collapse group_by/)) {
1377 $crs = $self->_count_subq_rs ($attrs);
1380 $crs = $self->_count_rs ($attrs);
1382 my $count = $crs->next;
1384 $count -= $offset if $offset;
1385 $count = $rows if $rows and $rows < $count;
1386 $count = 0 if ($count < 0);
1395 =item Arguments: $cond, \%attrs??
1397 =item Return Value: $count_rs
1401 Same as L</count> but returns a L<DBIx::Class::ResultSetColumn> object.
1402 This can be very handy for subqueries:
1404 ->search( { amount => $some_rs->count_rs->as_query } )
1406 As with regular resultsets the SQL query will be executed only after
1407 the resultset is accessed via L</next> or L</all>. That would return
1408 the same single value obtainable via L</count>.
1414 return $self->search(@_)->count_rs if @_;
1416 # this may look like a lack of abstraction (count() does about the same)
1417 # but in fact an _rs *must* use a subquery for the limits, as the
1418 # software based limiting can not be ported if this $rs is to be used
1419 # in a subquery itself (i.e. ->as_query)
1420 if ($self->_has_resolved_attr (qw/collapse group_by offset rows/)) {
1421 return $self->_count_subq_rs;
1424 return $self->_count_rs;
1429 # returns a ResultSetColumn object tied to the count query
1432 my ($self, $attrs) = @_;
1434 my $rsrc = $self->result_source;
1435 $attrs ||= $self->_resolved_attrs;
1437 my $tmp_attrs = { %$attrs };
1438 # take off any limits, record_filter is cdbi, and no point of ordering nor locking a count
1439 delete @{$tmp_attrs}{qw/rows offset order_by record_filter for/};
1441 # overwrite the selector (supplied by the storage)
1442 $tmp_attrs->{select} = $rsrc->storage->_count_select ($rsrc, $attrs);
1443 $tmp_attrs->{as} = 'count';
1444 delete @{$tmp_attrs}{qw/columns _trailing_select/};
1446 my $tmp_rs = $rsrc->resultset_class->new($rsrc, $tmp_attrs)->get_column ('count');
1452 # same as above but uses a subquery
1454 sub _count_subq_rs {
1455 my ($self, $attrs) = @_;
1457 my $rsrc = $self->result_source;
1458 $attrs ||= $self->_resolved_attrs;
1460 my $sub_attrs = { %$attrs };
1461 # extra selectors do not go in the subquery and there is no point of ordering it, nor locking it
1462 delete @{$sub_attrs}{qw/collapse columns as select _prefetch_selector_range _trailing_select order_by for/};
1464 # if we multi-prefetch we group_by primary keys only as this is what we would
1465 # get out of the rs via ->next/->all. We *DO WANT* to clobber old group_by regardless
1466 if ( keys %{$attrs->{collapse}} ) {
1467 $sub_attrs->{group_by} = [ map { "$attrs->{alias}.$_" } ($rsrc->_pri_cols) ]
1470 # Calculate subquery selector
1471 if (my $g = $sub_attrs->{group_by}) {
1473 my $sql_maker = $rsrc->storage->sql_maker;
1475 # necessary as the group_by may refer to aliased functions
1477 for my $sel (@{$attrs->{select}}) {
1478 $sel_index->{$sel->{-as}} = $sel
1479 if (ref $sel eq 'HASH' and $sel->{-as});
1482 # anything from the original select mentioned on the group-by needs to make it to the inner selector
1483 # also look for named aggregates referred in the having clause
1484 # having often contains scalarrefs - thus parse it out entirely
1486 if ($attrs->{having}) {
1487 local $sql_maker->{having_bind};
1488 local $sql_maker->{quote_char} = $sql_maker->{quote_char};
1489 local $sql_maker->{name_sep} = $sql_maker->{name_sep};
1490 unless (defined $sql_maker->{quote_char} and length $sql_maker->{quote_char}) {
1491 $sql_maker->{quote_char} = [ "\x00", "\xFF" ];
1492 # if we don't unset it we screw up retarded but unfortunately working
1493 # 'MAX(foo.bar)' => { '>', 3 }
1494 $sql_maker->{name_sep} = '';
1497 my ($lquote, $rquote, $sep) = map { quotemeta $_ } ($sql_maker->_quote_chars, $sql_maker->name_sep);
1499 my $sql = $sql_maker->_parse_rs_attrs ({ having => $attrs->{having} });
1501 # search for both a proper quoted qualified string, for a naive unquoted scalarref
1502 # and if all fails for an utterly naive quoted scalar-with-function
1504 $rquote $sep $lquote (.+?) $rquote
1506 [\s,] \w+ \. (\w+) [\s,]
1508 [\s,] $lquote (.+?) $rquote [\s,]
1510 push @parts, ($1 || $2 || $3); # one of them matched if we got here
1515 my $colpiece = $sel_index->{$_} || $_;
1517 # unqualify join-based group_by's. Arcane but possible query
1518 # also horrible horrible hack to alias a column (not a func.)
1519 # (probably need to introduce SQLA syntax)
1520 if ($colpiece =~ /\./ && $colpiece !~ /^$attrs->{alias}\./) {
1523 $colpiece = \ sprintf ('%s AS %s', map { $sql_maker->_quote ($_) } ($colpiece, $as) );
1525 push @{$sub_attrs->{select}}, $colpiece;
1529 my @pcols = map { "$attrs->{alias}.$_" } ($rsrc->primary_columns);
1530 $sub_attrs->{select} = @pcols ? \@pcols : [ 1 ];
1533 return $rsrc->resultset_class
1534 ->new ($rsrc, $sub_attrs)
1536 ->search ({}, { columns => { count => $rsrc->storage->_count_select ($rsrc, $attrs) } })
1537 ->get_column ('count');
1544 =head2 count_literal
1548 =item Arguments: $sql_fragment, @bind_values
1550 =item Return Value: $count
1554 Counts the results in a literal query. Equivalent to calling L</search_literal>
1555 with the passed arguments, then L</count>.
1559 sub count_literal { shift->search_literal(@_)->count; }
1565 =item Arguments: none
1567 =item Return Value: @objects
1571 Returns all elements in the resultset. Called implicitly if the resultset
1572 is returned in list context.
1579 $self->throw_exception("all() doesn't take any arguments, you probably wanted ->search(...)->all()");
1582 return @{ $self->get_cache } if $self->get_cache;
1586 if (keys %{$self->_resolved_attrs->{collapse}}) {
1587 # Using $self->cursor->all is really just an optimisation.
1588 # If we're collapsing has_many prefetches it probably makes
1589 # very little difference, and this is cleaner than hacking
1590 # _construct_object to survive the approach
1591 $self->cursor->reset;
1592 my @row = $self->cursor->next;
1594 push(@obj, $self->_construct_object(@row));
1595 @row = (exists $self->{stashed_row}
1596 ? @{delete $self->{stashed_row}}
1597 : $self->cursor->next);
1600 @obj = map { $self->_construct_object(@$_) } $self->cursor->all;
1603 $self->set_cache(\@obj) if $self->{attrs}{cache};
1612 =item Arguments: none
1614 =item Return Value: $self
1618 Resets the resultset's cursor, so you can iterate through the elements again.
1619 Implicitly resets the storage cursor, so a subsequent L</next> will trigger
1626 delete $self->{_attrs} if exists $self->{_attrs};
1627 $self->{all_cache_position} = 0;
1628 $self->cursor->reset;
1636 =item Arguments: none
1638 =item Return Value: $object | undef
1642 Resets the resultset and returns an object for the first result (or C<undef>
1643 if the resultset is empty).
1648 return $_[0]->reset->next;
1654 # Determines whether and what type of subquery is required for the $rs operation.
1655 # If grouping is necessary either supplies its own, or verifies the current one
1656 # After all is done delegates to the proper storage method.
1658 sub _rs_update_delete {
1659 my ($self, $op, $values) = @_;
1661 my $rsrc = $self->result_source;
1663 # if a condition exists we need to strip all table qualifiers
1664 # if this is not possible we'll force a subquery below
1665 my $cond = $rsrc->schema->storage->_strip_cond_qualifiers ($self->{cond});
1667 my $needs_group_by_subq = $self->_has_resolved_attr (qw/collapse group_by -join/);
1668 my $needs_subq = $needs_group_by_subq || (not defined $cond) || $self->_has_resolved_attr(qw/rows offset/);
1670 if ($needs_group_by_subq or $needs_subq) {
1672 # make a new $rs selecting only the PKs (that's all we really need)
1673 my $attrs = $self->_resolved_attrs_copy;
1676 delete $attrs->{$_} for qw/collapse _collapse_order_by select _prefetch_selector_range as/;
1677 $attrs->{columns} = [ map { "$attrs->{alias}.$_" } ($self->result_source->_pri_cols) ];
1679 if ($needs_group_by_subq) {
1680 # make sure no group_by was supplied, or if there is one - make sure it matches
1681 # the columns compiled above perfectly. Anything else can not be sanely executed
1682 # on most databases so croak right then and there
1684 if (my $g = $attrs->{group_by}) {
1685 my @current_group_by = map
1686 { $_ =~ /\./ ? $_ : "$attrs->{alias}.$_" }
1691 join ("\x00", sort @current_group_by)
1693 join ("\x00", sort @{$attrs->{columns}} )
1695 $self->throw_exception (
1696 "You have just attempted a $op operation on a resultset which does group_by"
1697 . ' on columns other than the primary keys, while DBIC internally needs to retrieve'
1698 . ' the primary keys in a subselect. All sane RDBMS engines do not support this'
1699 . ' kind of queries. Please retry the operation with a modified group_by or'
1700 . ' without using one at all.'
1705 $attrs->{group_by} = $attrs->{columns};
1709 my $subrs = (ref $self)->new($rsrc, $attrs);
1710 return $self->result_source->storage->_subq_update_delete($subrs, $op, $values);
1713 return $rsrc->storage->$op(
1715 $op eq 'update' ? $values : (),
1725 =item Arguments: \%values
1727 =item Return Value: $storage_rv
1731 Sets the specified columns in the resultset to the supplied values in a
1732 single query. Note that this will not run any accessor/set_column/update
1733 triggers, nor will it update any row object instances derived from this
1734 resultset (this includes the contents of the L<resultset cache|/set_cache>
1735 if any). See L</update_all> if you need to execute any on-update
1736 triggers or cascades defined either by you or a
1737 L<result component|DBIx::Class::Manual::Component/WHAT_IS_A_COMPONENT>.
1739 The return value is a pass through of what the underlying
1740 storage backend returned, and may vary. See L<DBI/execute> for the most
1745 Note that L</update> does not process/deflate any of the values passed in.
1746 This is unlike the corresponding L<DBIx::Class::Row/update>. The user must
1747 ensure manually that any value passed to this method will stringify to
1748 something the RDBMS knows how to deal with. A notable example is the
1749 handling of L<DateTime> objects, for more info see:
1750 L<DBIx::Class::Manual::Cookbook/Formatting_DateTime_objects_in_queries>.
1755 my ($self, $values) = @_;
1756 $self->throw_exception('Values for update must be a hash')
1757 unless ref $values eq 'HASH';
1759 return $self->_rs_update_delete ('update', $values);
1766 =item Arguments: \%values
1768 =item Return Value: 1
1772 Fetches all objects and updates them one at a time via
1773 L<DBIx::Class::Row/update>. Note that C<update_all> will run DBIC defined
1774 triggers, while L</update> will not.
1779 my ($self, $values) = @_;
1780 $self->throw_exception('Values for update_all must be a hash')
1781 unless ref $values eq 'HASH';
1783 my $guard = $self->result_source->schema->txn_scope_guard;
1784 $_->update($values) for $self->all;
1793 =item Arguments: none
1795 =item Return Value: $storage_rv
1799 Deletes the rows matching this resultset in a single query. Note that this
1800 will not run any delete triggers, nor will it alter the
1801 L<in_storage|DBIx::Class::Row/in_storage> status of any row object instances
1802 derived from this resultset (this includes the contents of the
1803 L<resultset cache|/set_cache> if any). See L</delete_all> if you need to
1804 execute any on-delete triggers or cascades defined either by you or a
1805 L<result component|DBIx::Class::Manual::Component/WHAT_IS_A_COMPONENT>.
1807 The return value is a pass through of what the underlying storage backend
1808 returned, and may vary. See L<DBI/execute> for the most common case.
1814 $self->throw_exception('delete does not accept any arguments')
1817 return $self->_rs_update_delete ('delete');
1824 =item Arguments: none
1826 =item Return Value: 1
1830 Fetches all objects and deletes them one at a time via
1831 L<DBIx::Class::Row/delete>. Note that C<delete_all> will run DBIC defined
1832 triggers, while L</delete> will not.
1838 $self->throw_exception('delete_all does not accept any arguments')
1841 my $guard = $self->result_source->schema->txn_scope_guard;
1842 $_->delete for $self->all;
1851 =item Arguments: \@data;
1855 Accepts either an arrayref of hashrefs or alternatively an arrayref of arrayrefs.
1856 For the arrayref of hashrefs style each hashref should be a structure suitable
1857 forsubmitting to a $resultset->create(...) method.
1859 In void context, C<insert_bulk> in L<DBIx::Class::Storage::DBI> is used
1860 to insert the data, as this is a faster method.
1862 Otherwise, each set of data is inserted into the database using
1863 L<DBIx::Class::ResultSet/create>, and the resulting objects are
1864 accumulated into an array. The array itself, or an array reference
1865 is returned depending on scalar or list context.
1867 Example: Assuming an Artist Class that has many CDs Classes relating:
1869 my $Artist_rs = $schema->resultset("Artist");
1871 ## Void Context Example
1872 $Artist_rs->populate([
1873 { artistid => 4, name => 'Manufactured Crap', cds => [
1874 { title => 'My First CD', year => 2006 },
1875 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
1878 { artistid => 5, name => 'Angsty-Whiny Girl', cds => [
1879 { title => 'My parents sold me to a record company', year => 2005 },
1880 { title => 'Why Am I So Ugly?', year => 2006 },
1881 { title => 'I Got Surgery and am now Popular', year => 2007 }
1886 ## Array Context Example
1887 my ($ArtistOne, $ArtistTwo, $ArtistThree) = $Artist_rs->populate([
1888 { name => "Artist One"},
1889 { name => "Artist Two"},
1890 { name => "Artist Three", cds=> [
1891 { title => "First CD", year => 2007},
1892 { title => "Second CD", year => 2008},
1896 print $ArtistOne->name; ## response is 'Artist One'
1897 print $ArtistThree->cds->count ## reponse is '2'
1899 For the arrayref of arrayrefs style, the first element should be a list of the
1900 fieldsnames to which the remaining elements are rows being inserted. For
1903 $Arstist_rs->populate([
1904 [qw/artistid name/],
1905 [100, 'A Formally Unknown Singer'],
1906 [101, 'A singer that jumped the shark two albums ago'],
1907 [102, 'An actually cool singer'],
1910 Please note an important effect on your data when choosing between void and
1911 wantarray context. Since void context goes straight to C<insert_bulk> in
1912 L<DBIx::Class::Storage::DBI> this will skip any component that is overriding
1913 C<insert>. So if you are using something like L<DBIx-Class-UUIDColumns> to
1914 create primary keys for you, you will find that your PKs are empty. In this
1915 case you will have to use the wantarray context in order to create those
1923 # cruft placed in standalone method
1924 my $data = $self->_normalize_populate_args(@_);
1926 if(defined wantarray) {
1928 foreach my $item (@$data) {
1929 push(@created, $self->create($item));
1931 return wantarray ? @created : \@created;
1934 my $first = $data->[0];
1936 # if a column is a registered relationship, and is a non-blessed hash/array, consider
1937 # it relationship data
1938 my (@rels, @columns);
1939 my $rsrc = $self->result_source;
1940 my $rels = { map { $_ => $rsrc->relationship_info($_) } $rsrc->relationships };
1941 for (keys %$first) {
1942 my $ref = ref $first->{$_};
1943 $rels->{$_} && ($ref eq 'ARRAY' or $ref eq 'HASH')
1949 my @pks = $rsrc->primary_columns;
1951 ## do the belongs_to relationships
1952 foreach my $index (0..$#$data) {
1954 # delegate to create() for any dataset without primary keys with specified relationships
1955 if (grep { !defined $data->[$index]->{$_} } @pks ) {
1957 if (grep { ref $data->[$index]{$r} eq $_ } qw/HASH ARRAY/) { # a related set must be a HASH or AoH
1958 my @ret = $self->populate($data);
1964 foreach my $rel (@rels) {
1965 next unless ref $data->[$index]->{$rel} eq "HASH";
1966 my $result = $self->related_resultset($rel)->create($data->[$index]->{$rel});
1967 my ($reverse_relname, $reverse_relinfo) = %{$rsrc->reverse_relationship_info($rel)};
1968 my $related = $result->result_source->_resolve_condition(
1969 $reverse_relinfo->{cond},
1974 delete $data->[$index]->{$rel};
1975 $data->[$index] = {%{$data->[$index]}, %$related};
1977 push @columns, keys %$related if $index == 0;
1981 ## inherit the data locked in the conditions of the resultset
1982 my ($rs_data) = $self->_merge_with_rscond({});
1983 delete @{$rs_data}{@columns};
1984 my @inherit_cols = keys %$rs_data;
1985 my @inherit_data = values %$rs_data;
1987 ## do bulk insert on current row
1988 $rsrc->storage->insert_bulk(
1990 [@columns, @inherit_cols],
1991 [ map { [ @$_{@columns}, @inherit_data ] } @$data ],
1994 ## do the has_many relationships
1995 foreach my $item (@$data) {
1999 foreach my $rel (@rels) {
2000 next unless ref $item->{$rel} eq "ARRAY" && @{ $item->{$rel} };
2002 $main_row ||= $self->new_result({map { $_ => $item->{$_} } @pks});
2004 my $child = $main_row->$rel;
2006 my $related = $child->result_source->_resolve_condition(
2007 $rels->{$rel}{cond},
2012 my @rows_to_add = ref $item->{$rel} eq 'ARRAY' ? @{$item->{$rel}} : ($item->{$rel});
2013 my @populate = map { {%$_, %$related} } @rows_to_add;
2015 $child->populate( \@populate );
2022 # populate() argumnets went over several incarnations
2023 # What we ultimately support is AoH
2024 sub _normalize_populate_args {
2025 my ($self, $arg) = @_;
2027 if (ref $arg eq 'ARRAY') {
2028 if (ref $arg->[0] eq 'HASH') {
2031 elsif (ref $arg->[0] eq 'ARRAY') {
2033 my @colnames = @{$arg->[0]};
2034 foreach my $values (@{$arg}[1 .. $#$arg]) {
2035 push @ret, { map { $colnames[$_] => $values->[$_] } (0 .. $#colnames) };
2041 $self->throw_exception('Populate expects an arrayref of hashrefs or arrayref of arrayrefs');
2048 =item Arguments: none
2050 =item Return Value: $pager
2054 Return Value a L<Data::Page> object for the current resultset. Only makes
2055 sense for queries with a C<page> attribute.
2057 To get the full count of entries for a paged resultset, call
2058 C<total_entries> on the L<Data::Page> object.
2062 # make a wizard good for both a scalar and a hashref
2063 my $mk_lazy_count_wizard = sub {
2064 require Variable::Magic;
2066 my $stash = { total_rs => shift };
2067 my $slot = shift; # only used by the hashref magic
2069 my $magic = Variable::Magic::wizard (
2070 data => sub { $stash },
2076 # set value lazily, and dispell for good
2077 ${$_[0]} = $_[1]{total_rs}->count;
2078 Variable::Magic::dispell (${$_[0]}, $_[1]{magic_selfref});
2082 # an explicit set implies dispell as well
2083 # the unless() is to work around "fun and giggles" below
2084 Variable::Magic::dispell (${$_[0]}, $_[1]{magic_selfref})
2085 unless (caller(2))[3] eq 'DBIx::Class::ResultSet::pager';
2092 if ($_[2] eq $slot and !$_[1]{inactive}) {
2093 my $cnt = $_[1]{total_rs}->count;
2094 $_[0]->{$slot} = $cnt;
2096 # attempting to dispell in a fetch handle (works in store), seems
2097 # to invariable segfault on 5.10, 5.12, 5.13 :(
2098 # so use an inactivator instead
2099 #Variable::Magic::dispell (%{$_[0]}, $_[1]{magic_selfref});
2105 if (! $_[1]{inactive} and $_[2] eq $slot) {
2106 #Variable::Magic::dispell (%{$_[0]}, $_[1]{magic_selfref});
2108 unless (caller(2))[3] eq 'DBIx::Class::ResultSet::pager';
2115 $stash->{magic_selfref} = $magic;
2116 weaken ($stash->{magic_selfref}); # this fails on 5.8.1
2121 # the tie class for 5.8.1
2123 package # hide from pause
2124 DBIx::Class::__DBIC_LAZY_RS_COUNT__;
2125 use base qw/Tie::Hash/;
2127 sub FIRSTKEY { my $dummy = scalar keys %{$_[0]{data}}; each %{$_[0]{data}} }
2128 sub NEXTKEY { each %{$_[0]{data}} }
2129 sub EXISTS { exists $_[0]{data}{$_[1]} }
2130 sub DELETE { delete $_[0]{data}{$_[1]} }
2131 sub CLEAR { %{$_[0]{data}} = () }
2132 sub SCALAR { scalar %{$_[0]{data}} }
2135 $_[1]{data} = {%{$_[1]{selfref}}};
2136 %{$_[1]{selfref}} = ();
2137 Scalar::Util::weaken ($_[1]{selfref});
2138 return bless ($_[1], $_[0]);
2142 if ($_[1] eq $_[0]{slot}) {
2143 my $cnt = $_[0]{data}{$_[1]} = $_[0]{total_rs}->count;
2144 untie %{$_[0]{selfref}};
2145 %{$_[0]{selfref}} = %{$_[0]{data}};
2154 $_[0]{data}{$_[1]} = $_[2];
2155 if ($_[1] eq $_[0]{slot}) {
2156 untie %{$_[0]{selfref}};
2157 %{$_[0]{selfref}} = %{$_[0]{data}};
2166 return $self->{pager} if $self->{pager};
2168 if ($self->get_cache) {
2169 $self->throw_exception ('Pagers on cached resultsets are not supported');
2172 my $attrs = $self->{attrs};
2173 if (!defined $attrs->{page}) {
2174 $self->throw_exception("Can't create pager for non-paged rs");
2176 elsif ($attrs->{page} <= 0) {
2177 $self->throw_exception('Invalid page number (page-numbers are 1-based)');
2179 $attrs->{rows} ||= 10;
2181 # throw away the paging flags and re-run the count (possibly
2182 # with a subselect) to get the real total count
2183 my $count_attrs = { %$attrs };
2184 delete $count_attrs->{$_} for qw/rows offset page pager/;
2185 my $total_rs = (ref $self)->new($self->result_source, $count_attrs);
2188 ### the following may seem awkward and dirty, but it's a thought-experiment
2189 ### necessary for future development of DBIx::DS. Do *NOT* change this code
2190 ### before talking to ribasushi/mst
2193 my $pager = Data::Page->new(
2194 0, #start with an empty set
2196 $self->{attrs}{page},
2199 my $data_slot = 'total_entries';
2201 # Since we are interested in a cached value (once it's set - it's set), every
2202 # technique will detach from the magic-host once the time comes to fire the
2203 # ->count (or in the segfaulting case of >= 5.10 it will deactivate itself)
2205 if ($] < 5.008003) {
2206 # 5.8.1 throws 'Modification of a read-only value attempted' when one tries
2207 # to weakref the magic container :(
2209 tie (%$pager, 'DBIx::Class::__DBIC_LAZY_RS_COUNT__',
2210 { slot => $data_slot, total_rs => $total_rs, selfref => $pager }
2213 elsif ($] < 5.010) {
2214 # We can use magic on the hash value slot. It's interesting that the magic is
2215 # attached to the hash-slot, and does *not* stop working once I do the dummy
2216 # assignments after the cast()
2217 # tested on 5.8.3 and 5.8.9
2218 my $magic = $mk_lazy_count_wizard->($total_rs);
2219 Variable::Magic::cast ( $pager->{$data_slot}, $magic );
2221 # this is for fun and giggles
2222 $pager->{$data_slot} = -1;
2223 $pager->{$data_slot} = 0;
2225 # this does not work for scalars, but works with
2227 #my %vals = %$pager;
2232 # And the uvar magic
2233 # works on 5.10.1, 5.12.1 and 5.13.4 in its current form,
2234 # however see the wizard maker for more notes
2235 my $magic = $mk_lazy_count_wizard->($total_rs, $data_slot);
2236 Variable::Magic::cast ( %$pager, $magic );
2239 $pager->{$data_slot} = -1;
2240 $pager->{$data_slot} = 0;
2248 return $self->{pager} = $pager;
2255 =item Arguments: $page_number
2257 =item Return Value: $rs
2261 Returns a resultset for the $page_number page of the resultset on which page
2262 is called, where each page contains a number of rows equal to the 'rows'
2263 attribute set on the resultset (10 by default).
2268 my ($self, $page) = @_;
2269 return (ref $self)->new($self->result_source, { %{$self->{attrs}}, page => $page });
2276 =item Arguments: \%vals
2278 =item Return Value: $rowobject
2282 Creates a new row object in the resultset's result class and returns
2283 it. The row is not inserted into the database at this point, call
2284 L<DBIx::Class::Row/insert> to do that. Calling L<DBIx::Class::Row/in_storage>
2285 will tell you whether the row object has been inserted or not.
2287 Passes the hashref of input on to L<DBIx::Class::Row/new>.
2292 my ($self, $values) = @_;
2293 $self->throw_exception( "new_result needs a hash" )
2294 unless (ref $values eq 'HASH');
2296 my ($merged_cond, $cols_from_relations) = $self->_merge_with_rscond($values);
2300 @$cols_from_relations
2301 ? (-cols_from_relations => $cols_from_relations)
2303 -result_source => $self->result_source, # DO NOT REMOVE THIS, REQUIRED
2306 return $self->result_class->new(\%new);
2309 # _merge_with_rscond
2311 # Takes a simple hash of K/V data and returns its copy merged with the
2312 # condition already present on the resultset. Additionally returns an
2313 # arrayref of value/condition names, which were inferred from related
2314 # objects (this is needed for in-memory related objects)
2315 sub _merge_with_rscond {
2316 my ($self, $data) = @_;
2318 my (%new_data, @cols_from_relations);
2320 my $alias = $self->{attrs}{alias};
2322 if (! defined $self->{cond}) {
2323 # just massage $data below
2325 elsif ($self->{cond} eq $DBIx::Class::ResultSource::UNRESOLVABLE_CONDITION) {
2326 %new_data = %{ $self->{attrs}{related_objects} || {} }; # nothing might have been inserted yet
2327 @cols_from_relations = keys %new_data;
2329 elsif (ref $self->{cond} ne 'HASH') {
2330 $self->throw_exception(
2331 "Can't abstract implicit construct, resultset condition not a hash"
2335 # precendence must be given to passed values over values inherited from
2336 # the cond, so the order here is important.
2337 my $collapsed_cond = $self->_collapse_cond($self->{cond});
2338 my %implied = %{$self->_remove_alias($collapsed_cond, $alias)};
2340 while ( my($col, $value) = each %implied ) {
2341 my $vref = ref $value;
2342 if ($vref eq 'HASH' && keys(%$value) && (keys %$value)[0] eq '=') {
2343 $new_data{$col} = $value->{'='};
2345 elsif( !$vref or $vref eq 'SCALAR' or blessed($value) ) {
2346 $new_data{$col} = $value;
2353 %{ $self->_remove_alias($data, $alias) },
2356 return (\%new_data, \@cols_from_relations);
2359 # _has_resolved_attr
2361 # determines if the resultset defines at least one
2362 # of the attributes supplied
2364 # used to determine if a subquery is neccessary
2366 # supports some virtual attributes:
2368 # This will scan for any joins being present on the resultset.
2369 # It is not a mere key-search but a deep inspection of {from}
2372 sub _has_resolved_attr {
2373 my ($self, @attr_names) = @_;
2375 my $attrs = $self->_resolved_attrs;
2379 for my $n (@attr_names) {
2380 if (grep { $n eq $_ } (qw/-join/) ) {
2381 $extra_checks{$n}++;
2385 my $attr = $attrs->{$n};
2387 next if not defined $attr;
2389 if (ref $attr eq 'HASH') {
2390 return 1 if keys %$attr;
2392 elsif (ref $attr eq 'ARRAY') {
2400 # a resolved join is expressed as a multi-level from
2402 $extra_checks{-join}
2404 ref $attrs->{from} eq 'ARRAY'
2406 @{$attrs->{from}} > 1
2414 # Recursively collapse the condition.
2416 sub _collapse_cond {
2417 my ($self, $cond, $collapsed) = @_;
2421 if (ref $cond eq 'ARRAY') {
2422 foreach my $subcond (@$cond) {
2423 next unless ref $subcond; # -or
2424 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2427 elsif (ref $cond eq 'HASH') {
2428 if (keys %$cond and (keys %$cond)[0] eq '-and') {
2429 foreach my $subcond (@{$cond->{-and}}) {
2430 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2434 foreach my $col (keys %$cond) {
2435 my $value = $cond->{$col};
2436 $collapsed->{$col} = $value;
2446 # Remove the specified alias from the specified query hash. A copy is made so
2447 # the original query is not modified.
2450 my ($self, $query, $alias) = @_;
2452 my %orig = %{ $query || {} };
2455 foreach my $key (keys %orig) {
2457 $unaliased{$key} = $orig{$key};
2460 $unaliased{$1} = $orig{$key}
2461 if $key =~ m/^(?:\Q$alias\E\.)?([^.]+)$/;
2471 =item Arguments: none
2473 =item Return Value: \[ $sql, @bind ]
2477 Returns the SQL query and bind vars associated with the invocant.
2479 This is generally used as the RHS for a subquery.
2486 my $attrs = $self->_resolved_attrs_copy;
2491 # my ($sql, \@bind, \%dbi_bind_attrs) = _select_args_to_query (...)
2492 # $sql also has no wrapping parenthesis in list ctx
2494 my $sqlbind = $self->result_source->storage
2495 ->_select_args_to_query ($attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs);
2504 =item Arguments: \%vals, \%attrs?
2506 =item Return Value: $rowobject
2510 my $artist = $schema->resultset('Artist')->find_or_new(
2511 { artist => 'fred' }, { key => 'artists' });
2513 $cd->cd_to_producer->find_or_new({ producer => $producer },
2514 { key => 'primary });
2516 Find an existing record from this resultset using L</find>. if none exists,
2517 instantiate a new result object and return it. The object will not be saved
2518 into your storage until you call L<DBIx::Class::Row/insert> on it.
2520 You most likely want this method when looking for existing rows using a unique
2521 constraint that is not the primary key, or looking for related rows.
2523 If you want objects to be saved immediately, use L</find_or_create> instead.
2525 B<Note>: Make sure to read the documentation of L</find> and understand the
2526 significance of the C<key> attribute, as its lack may skew your search, and
2527 subsequently result in spurious new objects.
2529 B<Note>: Take care when using C<find_or_new> with a table having
2530 columns with default values that you intend to be automatically
2531 supplied by the database (e.g. an auto_increment primary key column).
2532 In normal usage, the value of such columns should NOT be included at
2533 all in the call to C<find_or_new>, even when set to C<undef>.
2539 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2540 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2541 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2544 return $self->new_result($hash);
2551 =item Arguments: \%vals
2553 =item Return Value: a L<DBIx::Class::Row> $object
2557 Attempt to create a single new row or a row with multiple related rows
2558 in the table represented by the resultset (and related tables). This
2559 will not check for duplicate rows before inserting, use
2560 L</find_or_create> to do that.
2562 To create one row for this resultset, pass a hashref of key/value
2563 pairs representing the columns of the table and the values you wish to
2564 store. If the appropriate relationships are set up, foreign key fields
2565 can also be passed an object representing the foreign row, and the
2566 value will be set to its primary key.
2568 To create related objects, pass a hashref of related-object column values
2569 B<keyed on the relationship name>. If the relationship is of type C<multi>
2570 (L<DBIx::Class::Relationship/has_many>) - pass an arrayref of hashrefs.
2571 The process will correctly identify columns holding foreign keys, and will
2572 transparently populate them from the keys of the corresponding relation.
2573 This can be applied recursively, and will work correctly for a structure
2574 with an arbitrary depth and width, as long as the relationships actually
2575 exists and the correct column data has been supplied.
2578 Instead of hashrefs of plain related data (key/value pairs), you may
2579 also pass new or inserted objects. New objects (not inserted yet, see
2580 L</new>), will be inserted into their appropriate tables.
2582 Effectively a shortcut for C<< ->new_result(\%vals)->insert >>.
2584 Example of creating a new row.
2586 $person_rs->create({
2587 name=>"Some Person",
2588 email=>"somebody@someplace.com"
2591 Example of creating a new row and also creating rows in a related C<has_many>
2592 or C<has_one> resultset. Note Arrayref.
2595 { artistid => 4, name => 'Manufactured Crap', cds => [
2596 { title => 'My First CD', year => 2006 },
2597 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2602 Example of creating a new row and also creating a row in a related
2603 C<belongs_to> resultset. Note Hashref.
2606 title=>"Music for Silly Walks",
2609 name=>"Silly Musician",
2617 When subclassing ResultSet never attempt to override this method. Since
2618 it is a simple shortcut for C<< $self->new_result($attrs)->insert >>, a
2619 lot of the internals simply never call it, so your override will be
2620 bypassed more often than not. Override either L<new|DBIx::Class::Row/new>
2621 or L<insert|DBIx::Class::Row/insert> depending on how early in the
2622 L</create> process you need to intervene.
2629 my ($self, $attrs) = @_;
2630 $self->throw_exception( "create needs a hashref" )
2631 unless ref $attrs eq 'HASH';
2632 return $self->new_result($attrs)->insert;
2635 =head2 find_or_create
2639 =item Arguments: \%vals, \%attrs?
2641 =item Return Value: $rowobject
2645 $cd->cd_to_producer->find_or_create({ producer => $producer },
2646 { key => 'primary' });
2648 Tries to find a record based on its primary key or unique constraints; if none
2649 is found, creates one and returns that instead.
2651 my $cd = $schema->resultset('CD')->find_or_create({
2653 artist => 'Massive Attack',
2654 title => 'Mezzanine',
2658 Also takes an optional C<key> attribute, to search by a specific key or unique
2659 constraint. For example:
2661 my $cd = $schema->resultset('CD')->find_or_create(
2663 artist => 'Massive Attack',
2664 title => 'Mezzanine',
2666 { key => 'cd_artist_title' }
2669 B<Note>: Make sure to read the documentation of L</find> and understand the
2670 significance of the C<key> attribute, as its lack may skew your search, and
2671 subsequently result in spurious row creation.
2673 B<Note>: Because find_or_create() reads from the database and then
2674 possibly inserts based on the result, this method is subject to a race
2675 condition. Another process could create a record in the table after
2676 the find has completed and before the create has started. To avoid
2677 this problem, use find_or_create() inside a transaction.
2679 B<Note>: Take care when using C<find_or_create> with a table having
2680 columns with default values that you intend to be automatically
2681 supplied by the database (e.g. an auto_increment primary key column).
2682 In normal usage, the value of such columns should NOT be included at
2683 all in the call to C<find_or_create>, even when set to C<undef>.
2685 See also L</find> and L</update_or_create>. For information on how to declare
2686 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2690 sub find_or_create {
2692 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2693 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2694 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2697 return $self->create($hash);
2700 =head2 update_or_create
2704 =item Arguments: \%col_values, { key => $unique_constraint }?
2706 =item Return Value: $row_object
2710 $resultset->update_or_create({ col => $val, ... });
2712 Like L</find_or_create>, but if a row is found it is immediately updated via
2713 C<< $found_row->update (\%col_values) >>.
2716 Takes an optional C<key> attribute to search on a specific unique constraint.
2719 # In your application
2720 my $cd = $schema->resultset('CD')->update_or_create(
2722 artist => 'Massive Attack',
2723 title => 'Mezzanine',
2726 { key => 'cd_artist_title' }
2729 $cd->cd_to_producer->update_or_create({
2730 producer => $producer,
2736 B<Note>: Make sure to read the documentation of L</find> and understand the
2737 significance of the C<key> attribute, as its lack may skew your search, and
2738 subsequently result in spurious row creation.
2740 B<Note>: Take care when using C<update_or_create> with a table having
2741 columns with default values that you intend to be automatically
2742 supplied by the database (e.g. an auto_increment primary key column).
2743 In normal usage, the value of such columns should NOT be included at
2744 all in the call to C<update_or_create>, even when set to C<undef>.
2746 See also L</find> and L</find_or_create>. For information on how to declare
2747 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2751 sub update_or_create {
2753 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2754 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2756 my $row = $self->find($cond, $attrs);
2758 $row->update($cond);
2762 return $self->create($cond);
2765 =head2 update_or_new
2769 =item Arguments: \%col_values, { key => $unique_constraint }?
2771 =item Return Value: $rowobject
2775 $resultset->update_or_new({ col => $val, ... });
2777 Like L</find_or_new> but if a row is found it is immediately updated via
2778 C<< $found_row->update (\%col_values) >>.
2782 # In your application
2783 my $cd = $schema->resultset('CD')->update_or_new(
2785 artist => 'Massive Attack',
2786 title => 'Mezzanine',
2789 { key => 'cd_artist_title' }
2792 if ($cd->in_storage) {
2793 # the cd was updated
2796 # the cd is not yet in the database, let's insert it
2800 B<Note>: Make sure to read the documentation of L</find> and understand the
2801 significance of the C<key> attribute, as its lack may skew your search, and
2802 subsequently result in spurious new objects.
2804 B<Note>: Take care when using C<update_or_new> with a table having
2805 columns with default values that you intend to be automatically
2806 supplied by the database (e.g. an auto_increment primary key column).
2807 In normal usage, the value of such columns should NOT be included at
2808 all in the call to C<update_or_new>, even when set to C<undef>.
2810 See also L</find>, L</find_or_create> and L</find_or_new>.
2816 my $attrs = ( @_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {} );
2817 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2819 my $row = $self->find( $cond, $attrs );
2820 if ( defined $row ) {
2821 $row->update($cond);
2825 return $self->new_result($cond);
2832 =item Arguments: none
2834 =item Return Value: \@cache_objects | undef
2838 Gets the contents of the cache for the resultset, if the cache is set.
2840 The cache is populated either by using the L</prefetch> attribute to
2841 L</search> or by calling L</set_cache>.
2853 =item Arguments: \@cache_objects
2855 =item Return Value: \@cache_objects
2859 Sets the contents of the cache for the resultset. Expects an arrayref
2860 of objects of the same class as those produced by the resultset. Note that
2861 if the cache is set the resultset will return the cached objects rather
2862 than re-querying the database even if the cache attr is not set.
2864 The contents of the cache can also be populated by using the
2865 L</prefetch> attribute to L</search>.
2870 my ( $self, $data ) = @_;
2871 $self->throw_exception("set_cache requires an arrayref")
2872 if defined($data) && (ref $data ne 'ARRAY');
2873 $self->{all_cache} = $data;
2880 =item Arguments: none
2882 =item Return Value: undef
2886 Clears the cache for the resultset.
2891 shift->set_cache(undef);
2898 =item Arguments: none
2900 =item Return Value: true, if the resultset has been paginated
2908 return !!$self->{attrs}{page};
2915 =item Arguments: none
2917 =item Return Value: true, if the resultset has been ordered with C<order_by>.
2925 return scalar $self->result_source->storage->_extract_order_criteria($self->{attrs}{order_by});
2928 =head2 related_resultset
2932 =item Arguments: $relationship_name
2934 =item Return Value: $resultset
2938 Returns a related resultset for the supplied relationship name.
2940 $artist_rs = $schema->resultset('CD')->related_resultset('Artist');
2944 sub related_resultset {
2945 my ($self, $rel) = @_;
2947 $self->{related_resultsets} ||= {};
2948 return $self->{related_resultsets}{$rel} ||= do {
2949 my $rsrc = $self->result_source;
2950 my $rel_info = $rsrc->relationship_info($rel);
2952 $self->throw_exception(
2953 "search_related: result source '" . $rsrc->source_name .
2954 "' has no such relationship $rel")
2957 my $attrs = $self->_chain_relationship($rel);
2959 my $join_count = $attrs->{seen_join}{$rel};
2961 my $alias = $self->result_source->storage
2962 ->relname_to_table_alias($rel, $join_count);
2964 # since this is search_related, and we already slid the select window inwards
2965 # (the select/as attrs were deleted in the beginning), we need to flip all
2966 # left joins to inner, so we get the expected results
2967 # read the comment on top of the actual function to see what this does
2968 $attrs->{from} = $rsrc->schema->storage->_inner_join_to_node ($attrs->{from}, $alias);
2971 #XXX - temp fix for result_class bug. There likely is a more elegant fix -groditi
2972 delete @{$attrs}{qw(result_class alias)};
2976 if (my $cache = $self->get_cache) {
2977 if ($cache->[0] && $cache->[0]->related_resultset($rel)->get_cache) {
2978 $new_cache = [ map { @{$_->related_resultset($rel)->get_cache} }
2983 my $rel_source = $rsrc->related_source($rel);
2987 # The reason we do this now instead of passing the alias to the
2988 # search_rs below is that if you wrap/overload resultset on the
2989 # source you need to know what alias it's -going- to have for things
2990 # to work sanely (e.g. RestrictWithObject wants to be able to add
2991 # extra query restrictions, and these may need to be $alias.)
2993 my $rel_attrs = $rel_source->resultset_attributes;
2994 local $rel_attrs->{alias} = $alias;
2996 $rel_source->resultset
3000 where => $attrs->{where},
3003 $new->set_cache($new_cache) if $new_cache;
3008 =head2 current_source_alias
3012 =item Arguments: none
3014 =item Return Value: $source_alias
3018 Returns the current table alias for the result source this resultset is built
3019 on, that will be used in the SQL query. Usually it is C<me>.
3021 Currently the source alias that refers to the result set returned by a
3022 L</search>/L</find> family method depends on how you got to the resultset: it's
3023 C<me> by default, but eg. L</search_related> aliases it to the related result
3024 source name (and keeps C<me> referring to the original result set). The long
3025 term goal is to make L<DBIx::Class> always alias the current resultset as C<me>
3026 (and make this method unnecessary).
3028 Thus it's currently necessary to use this method in predefined queries (see
3029 L<DBIx::Class::Manual::Cookbook/Predefined searches>) when referring to the
3030 source alias of the current result set:
3032 # in a result set class
3034 my ($self, $user) = @_;
3036 my $me = $self->current_source_alias;
3038 return $self->search(
3039 "$me.modified" => $user->id,
3045 sub current_source_alias {
3048 return ($self->{attrs} || {})->{alias} || 'me';
3051 =head2 as_subselect_rs
3055 =item Arguments: none
3057 =item Return Value: $resultset
3061 Act as a barrier to SQL symbols. The resultset provided will be made into a
3062 "virtual view" by including it as a subquery within the from clause. From this
3063 point on, any joined tables are inaccessible to ->search on the resultset (as if
3064 it were simply where-filtered without joins). For example:
3066 my $rs = $schema->resultset('Bar')->search({'x.name' => 'abc'},{ join => 'x' });
3068 # 'x' now pollutes the query namespace
3070 # So the following works as expected
3071 my $ok_rs = $rs->search({'x.other' => 1});
3073 # But this doesn't: instead of finding a 'Bar' related to two x rows (abc and
3074 # def) we look for one row with contradictory terms and join in another table
3075 # (aliased 'x_2') which we never use
3076 my $broken_rs = $rs->search({'x.name' => 'def'});
3078 my $rs2 = $rs->as_subselect_rs;
3080 # doesn't work - 'x' is no longer accessible in $rs2, having been sealed away
3081 my $not_joined_rs = $rs2->search({'x.other' => 1});
3083 # works as expected: finds a 'table' row related to two x rows (abc and def)
3084 my $correctly_joined_rs = $rs2->search({'x.name' => 'def'});
3086 Another example of when one might use this would be to select a subset of
3087 columns in a group by clause:
3089 my $rs = $schema->resultset('Bar')->search(undef, {
3090 group_by => [qw{ id foo_id baz_id }],
3091 })->as_subselect_rs->search(undef, {
3092 columns => [qw{ id foo_id }]
3095 In the above example normally columns would have to be equal to the group by,
3096 but because we isolated the group by into a subselect the above works.
3100 sub as_subselect_rs {
3103 my $attrs = $self->_resolved_attrs;
3105 my $fresh_rs = (ref $self)->new (
3106 $self->result_source
3109 # these pieces will be locked in the subquery
3110 delete $fresh_rs->{cond};
3111 delete @{$fresh_rs->{attrs}}{qw/where bind/};
3113 return $fresh_rs->search( {}, {
3115 $attrs->{alias} => $self->as_query,
3116 -alias => $attrs->{alias},
3117 -rsrc => $self->result_source,
3119 alias => $attrs->{alias},
3123 # This code is called by search_related, and makes sure there
3124 # is clear separation between the joins before, during, and
3125 # after the relationship. This information is needed later
3126 # in order to properly resolve prefetch aliases (any alias
3127 # with a relation_chain_depth less than the depth of the
3128 # current prefetch is not considered)
3130 # The increments happen twice per join. An even number means a
3131 # relationship specified via a search_related, whereas an odd
3132 # number indicates a join/prefetch added via attributes
3134 # Also this code will wrap the current resultset (the one we
3135 # chain to) in a subselect IFF it contains limiting attributes
3136 sub _chain_relationship {
3137 my ($self, $rel) = @_;
3138 my $source = $self->result_source;
3139 my $attrs = { %{$self->{attrs}||{}} };
3141 # we need to take the prefetch the attrs into account before we
3142 # ->_resolve_join as otherwise they get lost - captainL
3143 my $join = $self->_merge_joinpref_attr( $attrs->{join}, $attrs->{prefetch} );
3145 delete @{$attrs}{qw/join prefetch collapse group_by distinct select as columns +select +as +columns/};
3147 my $seen = { %{ (delete $attrs->{seen_join}) || {} } };
3150 my @force_subq_attrs = qw/offset rows group_by having/;
3153 ($attrs->{from} && ref $attrs->{from} ne 'ARRAY')
3155 $self->_has_resolved_attr (@force_subq_attrs)
3157 # Nuke the prefetch (if any) before the new $rs attrs
3158 # are resolved (prefetch is useless - we are wrapping
3159 # a subquery anyway).
3160 my $rs_copy = $self->search;
3161 $rs_copy->{attrs}{join} = $self->_merge_joinpref_attr (
3162 $rs_copy->{attrs}{join},
3163 delete $rs_copy->{attrs}{prefetch},
3168 -alias => $attrs->{alias},
3169 $attrs->{alias} => $rs_copy->as_query,
3171 delete @{$attrs}{@force_subq_attrs, qw/where bind/};
3172 $seen->{-relation_chain_depth} = 0;
3174 elsif ($attrs->{from}) { #shallow copy suffices
3175 $from = [ @{$attrs->{from}} ];
3180 -alias => $attrs->{alias},
3181 $attrs->{alias} => $source->from,
3185 my $jpath = ($seen->{-relation_chain_depth})
3186 ? $from->[-1][0]{-join_path}
3189 my @requested_joins = $source->_resolve_join(
3196 push @$from, @requested_joins;
3198 $seen->{-relation_chain_depth}++;
3200 # if $self already had a join/prefetch specified on it, the requested
3201 # $rel might very well be already included. What we do in this case
3202 # is effectively a no-op (except that we bump up the chain_depth on
3203 # the join in question so we could tell it *is* the search_related)
3206 # we consider the last one thus reverse
3207 for my $j (reverse @requested_joins) {
3208 my ($last_j) = keys %{$j->[0]{-join_path}[-1]};
3209 if ($rel eq $last_j) {
3210 $j->[0]{-relation_chain_depth}++;
3216 unless ($already_joined) {
3217 push @$from, $source->_resolve_join(
3225 $seen->{-relation_chain_depth}++;
3227 return {%$attrs, from => $from, seen_join => $seen};
3230 # too many times we have to do $attrs = { %{$self->_resolved_attrs} }
3231 sub _resolved_attrs_copy {
3233 return { %{$self->_resolved_attrs (@_)} };
3236 sub _resolved_attrs {
3238 return $self->{_attrs} if $self->{_attrs};
3240 my $attrs = { %{ $self->{attrs} || {} } };
3241 my $source = $self->result_source;
3242 my $alias = $attrs->{alias};
3244 # one last pass of normalization
3245 $self->_normalize_selection($attrs);
3247 # default selection list
3248 $attrs->{columns} = [ $source->columns ]
3249 unless List::Util::first { exists $attrs->{$_} } qw/columns cols select as _trailing_select/;
3251 # merge selectors together
3252 for (qw/columns select as _trailing_select/) {
3253 $attrs->{$_} = $self->_merge_attr($attrs->{$_}, $attrs->{"+$_"})
3254 if $attrs->{$_} or $attrs->{"+$_"};
3257 # disassemble columns
3259 if (my $cols = delete $attrs->{columns}) {
3260 for my $c (ref $cols eq 'ARRAY' ? @$cols : $cols) {
3261 if (ref $c eq 'HASH') {
3262 for my $as (keys %$c) {
3263 push @sel, $c->{$as};
3274 # when trying to weed off duplicates later do not go past this point -
3275 # everything added from here on is unbalanced "anyone's guess" stuff
3276 my $dedup_stop_idx = $#as;
3278 push @as, @{ ref $attrs->{as} eq 'ARRAY' ? $attrs->{as} : [ $attrs->{as} ] }
3280 push @sel, @{ ref $attrs->{select} eq 'ARRAY' ? $attrs->{select} : [ $attrs->{select} ] }
3281 if $attrs->{select};
3283 # assume all unqualified selectors to apply to the current alias (legacy stuff)
3285 $_ = (ref $_ or $_ =~ /\./) ? $_ : "$alias.$_";
3288 # disqualify all $alias.col as-bits (collapser mandated)
3290 $_ = ($_ =~ /^\Q$alias.\E(.+)$/) ? $1 : $_;
3293 # de-duplicate the result (remove *identical* select/as pairs)
3294 # and also die on duplicate {as} pointing to different {select}s
3295 # not using a c-style for as the condition is prone to shrinkage
3298 while ($i <= $dedup_stop_idx) {
3299 if ($seen->{"$sel[$i] \x00\x00 $as[$i]"}++) {
3304 elsif ($seen->{$as[$i]}++) {
3305 $self->throw_exception(
3306 "inflate_result() alias '$as[$i]' specified twice with different SQL-side {select}-ors"
3314 $attrs->{select} = \@sel;
3315 $attrs->{as} = \@as;
3317 $attrs->{from} ||= [{
3319 -alias => $self->{attrs}{alias},
3320 $self->{attrs}{alias} => $source->from,
3323 if ( $attrs->{join} || $attrs->{prefetch} ) {
3325 $self->throw_exception ('join/prefetch can not be used with a custom {from}')
3326 if ref $attrs->{from} ne 'ARRAY';
3328 my $join = (delete $attrs->{join}) || {};
3330 if ( defined $attrs->{prefetch} ) {
3331 $join = $self->_merge_joinpref_attr( $join, $attrs->{prefetch} );
3334 $attrs->{from} = # have to copy here to avoid corrupting the original
3336 @{ $attrs->{from} },
3337 $source->_resolve_join(
3340 { %{ $attrs->{seen_join} || {} } },
3341 ( $attrs->{seen_join} && keys %{$attrs->{seen_join}})
3342 ? $attrs->{from}[-1][0]{-join_path}
3349 if ( defined $attrs->{order_by} ) {
3350 $attrs->{order_by} = (
3351 ref( $attrs->{order_by} ) eq 'ARRAY'
3352 ? [ @{ $attrs->{order_by} } ]
3353 : [ $attrs->{order_by} || () ]
3357 if ($attrs->{group_by} and ref $attrs->{group_by} ne 'ARRAY') {
3358 $attrs->{group_by} = [ $attrs->{group_by} ];
3361 # generate the distinct induced group_by early, as prefetch will be carried via a
3362 # subquery (since a group_by is present)
3363 if (delete $attrs->{distinct}) {
3364 if ($attrs->{group_by}) {
3365 carp ("Useless use of distinct on a grouped resultset ('distinct' is ignored when a 'group_by' is present)");
3368 # distinct affects only the main selection part, not what prefetch may
3369 # add below. However trailing is not yet a part of the selection as
3370 # prefetch must insert before it
3371 $attrs->{group_by} = $source->storage->_group_over_selection (
3373 [ @{$attrs->{select}||[]}, @{$attrs->{_trailing_select}||[]} ],
3379 $attrs->{collapse} ||= {};
3380 if ($attrs->{prefetch}) {
3381 my $prefetch = $self->_merge_joinpref_attr( {}, delete $attrs->{prefetch} );
3383 my $prefetch_ordering = [];
3385 # this is a separate structure (we don't look in {from} directly)
3386 # as the resolver needs to shift things off the lists to work
3387 # properly (identical-prefetches on different branches)
3389 if (ref $attrs->{from} eq 'ARRAY') {
3391 my $start_depth = $attrs->{seen_join}{-relation_chain_depth} || 0;
3393 for my $j ( @{$attrs->{from}}[1 .. $#{$attrs->{from}} ] ) {
3394 next unless $j->[0]{-alias};
3395 next unless $j->[0]{-join_path};
3396 next if ($j->[0]{-relation_chain_depth} || 0) < $start_depth;
3398 my @jpath = map { keys %$_ } @{$j->[0]{-join_path}};
3401 $p = $p->{$_} ||= {} for @jpath[ ($start_depth/2) .. $#jpath]; #only even depths are actual jpath boundaries
3402 push @{$p->{-join_aliases} }, $j->[0]{-alias};
3407 $source->_resolve_prefetch( $prefetch, $alias, $join_map, $prefetch_ordering, $attrs->{collapse} );
3409 # we need to somehow mark which columns came from prefetch
3411 my $sel_end = $#{$attrs->{select}};
3412 $attrs->{_prefetch_selector_range} = [ $sel_end + 1, $sel_end + @prefetch ];
3415 push @{ $attrs->{select} }, (map { $_->[0] } @prefetch);
3416 push @{ $attrs->{as} }, (map { $_->[1] } @prefetch);
3418 push( @{$attrs->{order_by}}, @$prefetch_ordering );
3419 $attrs->{_collapse_order_by} = \@$prefetch_ordering;
3423 push @{ $attrs->{select} }, @{$attrs->{_trailing_select}}
3424 if $attrs->{_trailing_select};
3426 # if both page and offset are specified, produce a combined offset
3427 # even though it doesn't make much sense, this is what pre 081xx has
3429 if (my $page = delete $attrs->{page}) {
3431 ($attrs->{rows} * ($page - 1))
3433 ($attrs->{offset} || 0)
3437 return $self->{_attrs} = $attrs;
3441 my ($self, $attr) = @_;
3443 if (ref $attr eq 'HASH') {
3444 return $self->_rollout_hash($attr);
3445 } elsif (ref $attr eq 'ARRAY') {
3446 return $self->_rollout_array($attr);
3452 sub _rollout_array {
3453 my ($self, $attr) = @_;
3456 foreach my $element (@{$attr}) {
3457 if (ref $element eq 'HASH') {
3458 push( @rolled_array, @{ $self->_rollout_hash( $element ) } );
3459 } elsif (ref $element eq 'ARRAY') {
3460 # XXX - should probably recurse here
3461 push( @rolled_array, @{$self->_rollout_array($element)} );
3463 push( @rolled_array, $element );
3466 return \@rolled_array;
3470 my ($self, $attr) = @_;
3473 foreach my $key (keys %{$attr}) {
3474 push( @rolled_array, { $key => $attr->{$key} } );
3476 return \@rolled_array;
3479 sub _calculate_score {
3480 my ($self, $a, $b) = @_;
3482 if (defined $a xor defined $b) {
3485 elsif (not defined $a) {
3489 if (ref $b eq 'HASH') {
3490 my ($b_key) = keys %{$b};
3491 if (ref $a eq 'HASH') {
3492 my ($a_key) = keys %{$a};
3493 if ($a_key eq $b_key) {
3494 return (1 + $self->_calculate_score( $a->{$a_key}, $b->{$b_key} ));
3499 return ($a eq $b_key) ? 1 : 0;
3502 if (ref $a eq 'HASH') {
3503 my ($a_key) = keys %{$a};
3504 return ($b eq $a_key) ? 1 : 0;
3506 return ($b eq $a) ? 1 : 0;
3511 sub _merge_joinpref_attr {
3512 my ($self, $orig, $import) = @_;
3514 return $import unless defined($orig);
3515 return $orig unless defined($import);
3517 $orig = $self->_rollout_attr($orig);
3518 $import = $self->_rollout_attr($import);
3521 foreach my $import_element ( @{$import} ) {
3522 # find best candidate from $orig to merge $b_element into
3523 my $best_candidate = { position => undef, score => 0 }; my $position = 0;
3524 foreach my $orig_element ( @{$orig} ) {
3525 my $score = $self->_calculate_score( $orig_element, $import_element );
3526 if ($score > $best_candidate->{score}) {
3527 $best_candidate->{position} = $position;
3528 $best_candidate->{score} = $score;
3532 my ($import_key) = ( ref $import_element eq 'HASH' ) ? keys %{$import_element} : ($import_element);
3534 if ($best_candidate->{score} == 0 || exists $seen_keys->{$import_key}) {
3535 push( @{$orig}, $import_element );
3537 my $orig_best = $orig->[$best_candidate->{position}];
3538 # merge orig_best and b_element together and replace original with merged
3539 if (ref $orig_best ne 'HASH') {
3540 $orig->[$best_candidate->{position}] = $import_element;
3541 } elsif (ref $import_element eq 'HASH') {
3542 my ($key) = keys %{$orig_best};
3543 $orig->[$best_candidate->{position}] = { $key => $self->_merge_joinpref_attr($orig_best->{$key}, $import_element->{$key}) };
3546 $seen_keys->{$import_key} = 1; # don't merge the same key twice
3557 require Hash::Merge;
3558 my $hm = Hash::Merge->new;
3560 $hm->specify_behavior({
3563 my ($defl, $defr) = map { defined $_ } (@_[0,1]);
3565 if ($defl xor $defr) {
3566 return [ $defl ? $_[0] : $_[1] ];
3571 elsif (__HM_DEDUP and $_[0] eq $_[1]) {
3575 return [$_[0], $_[1]];
3579 return $_[1] if !defined $_[0];
3580 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3581 return [$_[0], @{$_[1]}]
3584 return [] if !defined $_[0] and !keys %{$_[1]};
3585 return [ $_[1] ] if !defined $_[0];
3586 return [ $_[0] ] if !keys %{$_[1]};
3587 return [$_[0], $_[1]]
3592 return $_[0] if !defined $_[1];
3593 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3594 return [@{$_[0]}, $_[1]]
3597 my @ret = @{$_[0]} or return $_[1];
3598 return [ @ret, @{$_[1]} ] unless __HM_DEDUP;
3599 my %idx = map { $_ => 1 } @ret;
3600 push @ret, grep { ! defined $idx{$_} } (@{$_[1]});
3604 return [ $_[1] ] if ! @{$_[0]};
3605 return $_[0] if !keys %{$_[1]};
3606 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3607 return [ @{$_[0]}, $_[1] ];
3612 return [] if !keys %{$_[0]} and !defined $_[1];
3613 return [ $_[0] ] if !defined $_[1];
3614 return [ $_[1] ] if !keys %{$_[0]};
3615 return [$_[0], $_[1]]
3618 return [] if !keys %{$_[0]} and !@{$_[1]};
3619 return [ $_[0] ] if !@{$_[1]};
3620 return $_[1] if !keys %{$_[0]};
3621 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3622 return [ $_[0], @{$_[1]} ];
3625 return [] if !keys %{$_[0]} and !keys %{$_[1]};
3626 return [ $_[0] ] if !keys %{$_[1]};
3627 return [ $_[1] ] if !keys %{$_[0]};
3628 return [ $_[0] ] if $_[0] eq $_[1];
3629 return [ $_[0], $_[1] ];
3632 } => 'DBIC_RS_ATTR_MERGER');
3636 return $hm->merge ($_[1], $_[2]);
3640 sub STORABLE_freeze {
3641 my ($self, $cloning) = @_;
3642 my $to_serialize = { %$self };
3644 # A cursor in progress can't be serialized (and would make little sense anyway)
3645 delete $to_serialize->{cursor};
3647 Storable::nfreeze($to_serialize);
3650 # need this hook for symmetry
3652 my ($self, $cloning, $serialized) = @_;
3654 %$self = %{ Storable::thaw($serialized) };
3660 =head2 throw_exception
3662 See L<DBIx::Class::Schema/throw_exception> for details.
3666 sub throw_exception {
3669 if (ref $self and my $rsrc = $self->result_source) {
3670 $rsrc->throw_exception(@_)
3673 DBIx::Class::Exception->throw(@_);
3677 # XXX: FIXME: Attributes docs need clearing up
3681 Attributes are used to refine a ResultSet in various ways when
3682 searching for data. They can be passed to any method which takes an
3683 C<\%attrs> argument. See L</search>, L</search_rs>, L</find>,
3686 These are in no particular order:
3692 =item Value: ( $order_by | \@order_by | \%order_by )
3696 Which column(s) to order the results by.
3698 [The full list of suitable values is documented in
3699 L<SQL::Abstract/"ORDER BY CLAUSES">; the following is a summary of
3702 If a single column name, or an arrayref of names is supplied, the
3703 argument is passed through directly to SQL. The hashref syntax allows
3704 for connection-agnostic specification of ordering direction:
3706 For descending order:
3708 order_by => { -desc => [qw/col1 col2 col3/] }
3710 For explicit ascending order:
3712 order_by => { -asc => 'col' }
3714 The old scalarref syntax (i.e. order_by => \'year DESC') is still
3715 supported, although you are strongly encouraged to use the hashref
3716 syntax as outlined above.
3722 =item Value: \@columns
3726 Shortcut to request a particular set of columns to be retrieved. Each
3727 column spec may be a string (a table column name), or a hash (in which
3728 case the key is the C<as> value, and the value is used as the C<select>
3729 expression). Adds C<me.> onto the start of any column without a C<.> in
3730 it and sets C<select> from that, then auto-populates C<as> from
3731 C<select> as normal. (You may also use the C<cols> attribute, as in
3732 earlier versions of DBIC.)
3734 Essentially C<columns> does the same as L</select> and L</as>.
3736 columns => [ 'foo', { bar => 'baz' } ]
3740 select => [qw/foo baz/],
3747 =item Value: \@columns
3751 Indicates additional columns to be selected from storage. Works the same
3752 as L</columns> but adds columns to the selection. (You may also use the
3753 C<include_columns> attribute, as in earlier versions of DBIC). For
3756 $schema->resultset('CD')->search(undef, {
3757 '+columns' => ['artist.name'],
3761 would return all CDs and include a 'name' column to the information
3762 passed to object inflation. Note that the 'artist' is the name of the
3763 column (or relationship) accessor, and 'name' is the name of the column
3764 accessor in the related table.
3766 B<NOTE:> You need to explicitly quote '+columns' when defining the attribute.
3767 Not doing so causes Perl to incorrectly interpret +columns as a bareword with a
3768 unary plus operator before it.
3770 =head2 include_columns
3774 =item Value: \@columns
3778 Deprecated. Acts as a synonym for L</+columns> for backward compatibility.
3784 =item Value: \@select_columns
3788 Indicates which columns should be selected from the storage. You can use
3789 column names, or in the case of RDBMS back ends, function or stored procedure
3792 $rs = $schema->resultset('Employee')->search(undef, {
3795 { count => 'employeeid' },
3796 { max => { length => 'name' }, -as => 'longest_name' }
3801 SELECT name, COUNT( employeeid ), MAX( LENGTH( name ) ) AS longest_name FROM employee
3803 B<NOTE:> You will almost always need a corresponding L</as> attribute when you
3804 use L</select>, to instruct DBIx::Class how to store the result of the column.
3805 Also note that the L</as> attribute has nothing to do with the SQL-side 'AS'
3806 identifier aliasing. You can however alias a function, so you can use it in
3807 e.g. an C<ORDER BY> clause. This is done via the C<-as> B<select function
3808 attribute> supplied as shown in the example above.
3810 B<NOTE:> You need to explicitly quote '+select'/'+as' when defining the attributes.
3811 Not doing so causes Perl to incorrectly interpret them as a bareword with a
3812 unary plus operator before it.
3818 Indicates additional columns to be selected from storage. Works the same as
3819 L</select> but adds columns to the default selection, instead of specifying
3828 Indicates additional column names for those added via L</+select>. See L</as>.
3836 =item Value: \@inflation_names
3840 Indicates column names for object inflation. That is L</as> indicates the
3841 slot name in which the column value will be stored within the
3842 L<Row|DBIx::Class::Row> object. The value will then be accessible via this
3843 identifier by the C<get_column> method (or via the object accessor B<if one
3844 with the same name already exists>) as shown below. The L</as> attribute has
3845 B<nothing to do> with the SQL-side C<AS>. See L</select> for details.
3847 $rs = $schema->resultset('Employee')->search(undef, {
3850 { count => 'employeeid' },
3851 { max => { length => 'name' }, -as => 'longest_name' }
3860 If the object against which the search is performed already has an accessor
3861 matching a column name specified in C<as>, the value can be retrieved using
3862 the accessor as normal:
3864 my $name = $employee->name();
3866 If on the other hand an accessor does not exist in the object, you need to
3867 use C<get_column> instead:
3869 my $employee_count = $employee->get_column('employee_count');
3871 You can create your own accessors if required - see
3872 L<DBIx::Class::Manual::Cookbook> for details.
3878 =item Value: ($rel_name | \@rel_names | \%rel_names)
3882 Contains a list of relationships that should be joined for this query. For
3885 # Get CDs by Nine Inch Nails
3886 my $rs = $schema->resultset('CD')->search(
3887 { 'artist.name' => 'Nine Inch Nails' },
3888 { join => 'artist' }
3891 Can also contain a hash reference to refer to the other relation's relations.
3894 package MyApp::Schema::Track;
3895 use base qw/DBIx::Class/;
3896 __PACKAGE__->table('track');
3897 __PACKAGE__->add_columns(qw/trackid cd position title/);
3898 __PACKAGE__->set_primary_key('trackid');
3899 __PACKAGE__->belongs_to(cd => 'MyApp::Schema::CD');
3902 # In your application
3903 my $rs = $schema->resultset('Artist')->search(
3904 { 'track.title' => 'Teardrop' },
3906 join => { cd => 'track' },
3907 order_by => 'artist.name',
3911 You need to use the relationship (not the table) name in conditions,
3912 because they are aliased as such. The current table is aliased as "me", so
3913 you need to use me.column_name in order to avoid ambiguity. For example:
3915 # Get CDs from 1984 with a 'Foo' track
3916 my $rs = $schema->resultset('CD')->search(
3919 'tracks.name' => 'Foo'
3921 { join => 'tracks' }
3924 If the same join is supplied twice, it will be aliased to <rel>_2 (and
3925 similarly for a third time). For e.g.
3927 my $rs = $schema->resultset('Artist')->search({
3928 'cds.title' => 'Down to Earth',
3929 'cds_2.title' => 'Popular',
3931 join => [ qw/cds cds/ ],
3934 will return a set of all artists that have both a cd with title 'Down
3935 to Earth' and a cd with title 'Popular'.
3937 If you want to fetch related objects from other tables as well, see C<prefetch>
3940 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
3946 =item Value: ($rel_name | \@rel_names | \%rel_names)
3950 Contains one or more relationships that should be fetched along with
3951 the main query (when they are accessed afterwards the data will
3952 already be available, without extra queries to the database). This is
3953 useful for when you know you will need the related objects, because it
3954 saves at least one query:
3956 my $rs = $schema->resultset('Tag')->search(
3965 The initial search results in SQL like the following:
3967 SELECT tag.*, cd.*, artist.* FROM tag
3968 JOIN cd ON tag.cd = cd.cdid
3969 JOIN artist ON cd.artist = artist.artistid
3971 L<DBIx::Class> has no need to go back to the database when we access the
3972 C<cd> or C<artist> relationships, which saves us two SQL statements in this
3975 Simple prefetches will be joined automatically, so there is no need
3976 for a C<join> attribute in the above search.
3978 C<prefetch> can be used with the following relationship types: C<belongs_to>,
3979 C<has_one> (or if you're using C<add_relationship>, any relationship declared
3980 with an accessor type of 'single' or 'filter'). A more complex example that
3981 prefetches an artists cds, the tracks on those cds, and the tags associated
3982 with that artist is given below (assuming many-to-many from artists to tags):
3984 my $rs = $schema->resultset('Artist')->search(
3988 { cds => 'tracks' },
3989 { artist_tags => 'tags' }
3995 B<NOTE:> If you specify a C<prefetch> attribute, the C<join> and C<select>
3996 attributes will be ignored.
3998 B<CAVEATs>: Prefetch does a lot of deep magic. As such, it may not behave
3999 exactly as you might expect.
4005 Prefetch uses the L</cache> to populate the prefetched relationships. This
4006 may or may not be what you want.
4010 If you specify a condition on a prefetched relationship, ONLY those
4011 rows that match the prefetched condition will be fetched into that relationship.
4012 This means that adding prefetch to a search() B<may alter> what is returned by
4013 traversing a relationship. So, if you have C<< Artist->has_many(CDs) >> and you do
4015 my $artist_rs = $schema->resultset('Artist')->search({
4021 my $count = $artist_rs->first->cds->count;
4023 my $artist_rs_prefetch = $artist_rs->search( {}, { prefetch => 'cds' } );
4025 my $prefetch_count = $artist_rs_prefetch->first->cds->count;
4027 cmp_ok( $count, '==', $prefetch_count, "Counts should be the same" );
4029 that cmp_ok() may or may not pass depending on the datasets involved. This
4030 behavior may or may not survive the 0.09 transition.
4042 Makes the resultset paged and specifies the page to retrieve. Effectively
4043 identical to creating a non-pages resultset and then calling ->page($page)
4046 If L</rows> attribute is not specified it defaults to 10 rows per page.
4048 When you have a paged resultset, L</count> will only return the number
4049 of rows in the page. To get the total, use the L</pager> and call
4050 C<total_entries> on it.
4060 Specifies the maximum number of rows for direct retrieval or the number of
4061 rows per page if the page attribute or method is used.
4067 =item Value: $offset
4071 Specifies the (zero-based) row number for the first row to be returned, or the
4072 of the first row of the first page if paging is used.
4078 =item Value: \@columns
4082 A arrayref of columns to group by. Can include columns of joined tables.
4084 group_by => [qw/ column1 column2 ... /]
4090 =item Value: $condition
4094 HAVING is a select statement attribute that is applied between GROUP BY and
4095 ORDER BY. It is applied to the after the grouping calculations have been
4098 having => { 'count_employee' => { '>=', 100 } }
4100 or with an in-place function in which case literal SQL is required:
4102 having => \[ 'count(employee) >= ?', [ count => 100 ] ]
4108 =item Value: (0 | 1)
4112 Set to 1 to group by all columns. If the resultset already has a group_by
4113 attribute, this setting is ignored and an appropriate warning is issued.
4119 Adds to the WHERE clause.
4121 # only return rows WHERE deleted IS NULL for all searches
4122 __PACKAGE__->resultset_attributes({ where => { deleted => undef } }); )
4124 Can be overridden by passing C<< { where => undef } >> as an attribute
4131 Set to 1 to cache search results. This prevents extra SQL queries if you
4132 revisit rows in your ResultSet:
4134 my $resultset = $schema->resultset('Artist')->search( undef, { cache => 1 } );
4136 while( my $artist = $resultset->next ) {
4140 $rs->first; # without cache, this would issue a query
4142 By default, searches are not cached.
4144 For more examples of using these attributes, see
4145 L<DBIx::Class::Manual::Cookbook>.
4151 =item Value: ( 'update' | 'shared' )
4155 Set to 'update' for a SELECT ... FOR UPDATE or 'shared' for a SELECT