1 package DBIx::Class::ResultSet;
5 use base qw/DBIx::Class/;
7 use DBIx::Class::Exception;
8 use DBIx::Class::ResultSetColumn;
9 use Scalar::Util qw/blessed weaken/;
13 # not importing first() as it will clash with our own method
17 # De-duplication in _merge_attr() is disabled, but left in for reference
18 # (the merger is used for other things that ought not to be de-duped)
19 *__HM_DEDUP = sub () { 0 };
29 __PACKAGE__->mk_group_accessors('simple' => qw/_result_class result_source/);
33 DBIx::Class::ResultSet - Represents a query used for fetching a set of results.
37 my $users_rs = $schema->resultset('User');
38 while( $user = $users_rs->next) {
39 print $user->username;
42 my $registered_users_rs = $schema->resultset('User')->search({ registered => 1 });
43 my @cds_in_2005 = $schema->resultset('CD')->search({ year => 2005 })->all();
47 A ResultSet is an object which stores a set of conditions representing
48 a query. It is the backbone of DBIx::Class (i.e. the really
49 important/useful bit).
51 No SQL is executed on the database when a ResultSet is created, it
52 just stores all the conditions needed to create the query.
54 A basic ResultSet representing the data of an entire table is returned
55 by calling C<resultset> on a L<DBIx::Class::Schema> and passing in a
56 L<Source|DBIx::Class::Manual::Glossary/Source> name.
58 my $users_rs = $schema->resultset('User');
60 A new ResultSet is returned from calling L</search> on an existing
61 ResultSet. The new one will contain all the conditions of the
62 original, plus any new conditions added in the C<search> call.
64 A ResultSet also incorporates an implicit iterator. L</next> and L</reset>
65 can be used to walk through all the L<DBIx::Class::Row>s the ResultSet
68 The query that the ResultSet represents is B<only> executed against
69 the database when these methods are called:
70 L</find>, L</next>, L</all>, L</first>, L</single>, L</count>.
72 If a resultset is used in a numeric context it returns the L</count>.
73 However, if it is used in a boolean context it is B<always> true. So if
74 you want to check if a resultset has any results, you must use C<if $rs
79 =head2 Chaining resultsets
81 Let's say you've got a query that needs to be run to return some data
82 to the user. But, you have an authorization system in place that
83 prevents certain users from seeing certain information. So, you want
84 to construct the basic query in one method, but add constraints to it in
89 my $request = $self->get_request; # Get a request object somehow.
90 my $schema = $self->result_source->schema;
92 my $cd_rs = $schema->resultset('CD')->search({
93 title => $request->param('title'),
94 year => $request->param('year'),
97 $cd_rs = $self->apply_security_policy( $cd_rs );
102 sub apply_security_policy {
111 =head3 Resolving conditions and attributes
113 When a resultset is chained from another resultset, conditions and
114 attributes with the same keys need resolving.
116 L</join>, L</prefetch>, L</+select>, L</+as> attributes are merged
117 into the existing ones from the original resultset.
119 The L</where> and L</having> attributes, and any search conditions, are
120 merged with an SQL C<AND> to the existing condition from the original
123 All other attributes are overridden by any new ones supplied in the
126 =head2 Multiple queries
128 Since a resultset just defines a query, you can do all sorts of
129 things with it with the same object.
131 # Don't hit the DB yet.
132 my $cd_rs = $schema->resultset('CD')->search({
133 title => 'something',
137 # Each of these hits the DB individually.
138 my $count = $cd_rs->count;
139 my $most_recent = $cd_rs->get_column('date_released')->max();
140 my @records = $cd_rs->all;
142 And it's not just limited to SELECT statements.
148 $cd_rs->create({ artist => 'Fred' });
150 Which is the same as:
152 $schema->resultset('CD')->create({
153 title => 'something',
158 See: L</search>, L</count>, L</get_column>, L</all>, L</create>.
166 =item Arguments: $source, \%$attrs
168 =item Return Value: $rs
172 The resultset constructor. Takes a source object (usually a
173 L<DBIx::Class::ResultSourceProxy::Table>) and an attribute hash (see
174 L</ATTRIBUTES> below). Does not perform any queries -- these are
175 executed as needed by the other methods.
177 Generally you won't need to construct a resultset manually. You'll
178 automatically get one from e.g. a L</search> called in scalar context:
180 my $rs = $schema->resultset('CD')->search({ title => '100th Window' });
182 IMPORTANT: If called on an object, proxies to new_result instead so
184 my $cd = $schema->resultset('CD')->new({ title => 'Spoon' });
186 will return a CD object, not a ResultSet.
192 return $class->new_result(@_) if ref $class;
194 my ($source, $attrs) = @_;
195 $source = $source->resolve
196 if $source->isa('DBIx::Class::ResultSourceHandle');
197 $attrs = { %{$attrs||{}} };
199 if ($attrs->{page}) {
200 $attrs->{rows} ||= 10;
203 $attrs->{alias} ||= 'me';
206 result_source => $source,
207 cond => $attrs->{where},
212 # if there is a dark selector, this means we are already in a
213 # chain and the cleanup/sanification was taken care of by
215 $self->_normalize_selection($attrs)
216 unless $attrs->{_dark_selector};
219 $attrs->{result_class} || $source->result_class
229 =item Arguments: $cond, \%attrs?
231 =item Return Value: $resultset (scalar context) || @row_objs (list context)
235 my @cds = $cd_rs->search({ year => 2001 }); # "... WHERE year = 2001"
236 my $new_rs = $cd_rs->search({ year => 2005 });
238 my $new_rs = $cd_rs->search([ { year => 2005 }, { year => 2004 } ]);
239 # year = 2005 OR year = 2004
241 In list context, C<< ->all() >> is called implicitly on the resultset, thus
242 returning a list of row objects instead. To avoid that, use L</search_rs>.
244 If you need to pass in additional attributes but no additional condition,
245 call it as C<search(undef, \%attrs)>.
247 # "SELECT name, artistid FROM $artist_table"
248 my @all_artists = $schema->resultset('Artist')->search(undef, {
249 columns => [qw/name artistid/],
252 For a list of attributes that can be passed to C<search>, see
253 L</ATTRIBUTES>. For more examples of using this function, see
254 L<Searching|DBIx::Class::Manual::Cookbook/Searching>. For a complete
255 documentation for the first argument, see L<SQL::Abstract>
256 and its extension L<DBIx::Class::SQLMaker>.
258 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
262 Note that L</search> does not process/deflate any of the values passed in the
263 L<SQL::Abstract>-compatible search condition structure. This is unlike other
264 condition-bound methods L</new>, L</create> and L</find>. The user must ensure
265 manually that any value passed to this method will stringify to something the
266 RDBMS knows how to deal with. A notable example is the handling of L<DateTime>
267 objects, for more info see:
268 L<DBIx::Class::Manual::Cookbook/Formatting_DateTime_objects_in_queries>.
274 my $rs = $self->search_rs( @_ );
279 elsif (defined wantarray) {
283 # we can be called by a relationship helper, which in
284 # turn may be called in void context due to some braindead
285 # overload or whatever else the user decided to be clever
286 # at this particular day. Thus limit the exception to
287 # external code calls only
288 $self->throw_exception ('->search is *not* a mutator, calling it in void context makes no sense')
289 if (caller)[0] !~ /^\QDBIx::Class::/;
299 =item Arguments: $cond, \%attrs?
301 =item Return Value: $resultset
305 This method does the same exact thing as search() except it will
306 always return a resultset, even in list context.
313 # Special-case handling for (undef, undef).
314 if ( @_ == 2 && !defined $_[1] && !defined $_[0] ) {
320 if (ref $_[-1] eq 'HASH') {
321 # copy for _normalize_selection
322 $call_attrs = { %{ pop @_ } };
324 elsif (! defined $_[-1] ) {
325 pop @_; # search({}, undef)
329 # see if we can keep the cache (no $rs changes)
331 my %safe = (alias => 1, cache => 1);
332 if ( ! List::Util::first { !$safe{$_} } keys %$call_attrs and (
335 ref $_[0] eq 'HASH' && ! keys %{$_[0]}
337 ref $_[0] eq 'ARRAY' && ! @{$_[0]}
339 $cache = $self->get_cache;
342 my $rsrc = $self->result_source;
344 my $old_attrs = { %{$self->{attrs}} };
345 my $old_having = delete $old_attrs->{having};
346 my $old_where = delete $old_attrs->{where};
348 my $new_attrs = { %$old_attrs };
350 # take care of call attrs (only if anything is changing)
351 if (keys %$call_attrs) {
353 my @selector_attrs = qw/select as columns cols +select +as +columns include_columns/;
355 # reset the current selector list if new selectors are supplied
356 if (List::Util::first { exists $call_attrs->{$_} } qw/columns cols select as/) {
357 delete @{$old_attrs}{(@selector_attrs, '_dark_selector')};
360 # Normalize the new selector list (operates on the passed-in attr structure)
361 # Need to do it on every chain instead of only once on _resolved_attrs, in
362 # order to allow detection of empty vs partial 'as'
363 $call_attrs->{_dark_selector} = $old_attrs->{_dark_selector}
364 if $old_attrs->{_dark_selector};
365 $self->_normalize_selection ($call_attrs);
367 # start with blind overwriting merge, exclude selector attrs
368 $new_attrs = { %{$old_attrs}, %{$call_attrs} };
369 delete @{$new_attrs}{@selector_attrs};
371 for (@selector_attrs) {
372 $new_attrs->{$_} = $self->_merge_attr($old_attrs->{$_}, $call_attrs->{$_})
373 if ( exists $old_attrs->{$_} or exists $call_attrs->{$_} );
376 # older deprecated name, use only if {columns} is not there
377 if (my $c = delete $new_attrs->{cols}) {
378 if ($new_attrs->{columns}) {
379 carp "Resultset specifies both the 'columns' and the legacy 'cols' attributes - ignoring 'cols'";
382 $new_attrs->{columns} = $c;
387 # join/prefetch use their own crazy merging heuristics
388 foreach my $key (qw/join prefetch/) {
389 $new_attrs->{$key} = $self->_merge_joinpref_attr($old_attrs->{$key}, $call_attrs->{$key})
390 if exists $call_attrs->{$key};
393 # stack binds together
394 $new_attrs->{bind} = [ @{ $old_attrs->{bind} || [] }, @{ $call_attrs->{bind} || [] } ];
398 # rip apart the rest of @_, parse a condition
401 if (ref $_[0] eq 'HASH') {
402 (keys %{$_[0]}) ? $_[0] : undef
408 $self->throw_exception('Odd number of arguments to search')
416 if( @_ > 1 and ! $rsrc->result_class->isa('DBIx::Class::CDBICompat') ) {
417 carp_unique 'search( %condition ) is deprecated, use search( \%condition ) instead';
420 for ($old_where, $call_cond) {
422 $new_attrs->{where} = $self->_stack_cond (
423 $_, $new_attrs->{where}
428 if (defined $old_having) {
429 $new_attrs->{having} = $self->_stack_cond (
430 $old_having, $new_attrs->{having}
434 my $rs = (ref $self)->new($rsrc, $new_attrs);
436 $rs->set_cache($cache) if ($cache);
442 sub _normalize_selection {
443 my ($self, $attrs) = @_;
446 $attrs->{'+columns'} = $self->_merge_attr($attrs->{'+columns'}, delete $attrs->{include_columns})
447 if exists $attrs->{include_columns};
449 # columns are always placed first, however
451 # Keep the X vs +X separation until _resolved_attrs time - this allows to
452 # delay the decision on whether to use a default select list ($rsrc->columns)
453 # allowing stuff like the remove_columns helper to work
455 # select/as +select/+as pairs need special handling - the amount of select/as
456 # elements in each pair does *not* have to be equal (think multicolumn
457 # selectors like distinct(foo, bar) ). If the selector is bare (no 'as'
458 # supplied at all) - try to infer the alias, either from the -as parameter
459 # of the selector spec, or use the parameter whole if it looks like a column
460 # name (ugly legacy heuristic). If all fails - leave the selector bare (which
461 # is ok as well), but make sure no more additions to the 'as' chain take place
462 for my $pref ('', '+') {
464 my ($sel, $as) = map {
465 my $key = "${pref}${_}";
467 my $val = [ ref $attrs->{$key} eq 'ARRAY'
469 : $attrs->{$key} || ()
471 delete $attrs->{$key};
475 if (! @$as and ! @$sel ) {
478 elsif (@$as and ! @$sel) {
479 $self->throw_exception(
480 "Unable to handle ${pref}as specification (@$as) without a corresponding ${pref}select"
484 # no as part supplied at all - try to deduce (unless explicit end of named selection is declared)
485 # if any @$as has been supplied we assume the user knows what (s)he is doing
486 # and blindly keep stacking up pieces
487 unless ($attrs->{_dark_selector}) {
490 if ( ref $_ eq 'HASH' and exists $_->{-as} ) {
491 push @$as, $_->{-as};
493 # assume any plain no-space, no-parenthesis string to be a column spec
494 # FIXME - this is retarded but is necessary to support shit like 'count(foo)'
495 elsif ( ! ref $_ and $_ =~ /^ [^\s\(\)]+ $/x) {
498 # if all else fails - raise a flag that no more aliasing will be allowed
500 $attrs->{_dark_selector} = {
502 string => ($dark_sel_dumper ||= do {
503 require Data::Dumper::Concise;
504 Data::Dumper::Concise::DumperObject()->Indent(0);
505 })->Values([$_])->Dump
513 elsif (@$as < @$sel) {
514 $self->throw_exception(
515 "Unable to handle an ${pref}as specification (@$as) with less elements than the corresponding ${pref}select"
518 elsif ($pref and $attrs->{_dark_selector}) {
519 $self->throw_exception(
520 "Unable to process named '+select', resultset contains an unnamed selector $attrs->{_dark_selector}{string}"
526 $attrs->{"${pref}select"} = $self->_merge_attr($attrs->{"${pref}select"}, $sel);
527 $attrs->{"${pref}as"} = $self->_merge_attr($attrs->{"${pref}as"}, $as);
532 my ($self, $left, $right) = @_;
534 # collapse single element top-level conditions
535 # (single pass only, unlikely to need recursion)
536 for ($left, $right) {
537 if (ref $_ eq 'ARRAY') {
545 elsif (ref $_ eq 'HASH') {
546 my ($first, $more) = keys %$_;
549 if (! defined $first) {
553 elsif (! defined $more) {
554 if ($first eq '-and' and ref $_->{'-and'} eq 'HASH') {
557 elsif ($first eq '-or' and ref $_->{'-or'} eq 'ARRAY') {
564 # merge hashes with weeding out of duplicates (simple cases only)
565 if (ref $left eq 'HASH' and ref $right eq 'HASH') {
567 # shallow copy to destroy
568 $right = { %$right };
569 for (grep { exists $right->{$_} } keys %$left) {
570 # the use of eq_deeply here is justified - the rhs of an
571 # expression can contain a lot of twisted weird stuff
572 delete $right->{$_} if Compare( $left->{$_}, $right->{$_} );
575 $right = undef unless keys %$right;
579 if (defined $left xor defined $right) {
580 return defined $left ? $left : $right;
582 elsif (! defined $left) {
586 return { -and => [ $left, $right ] };
590 =head2 search_literal
594 =item Arguments: $sql_fragment, @bind_values
596 =item Return Value: $resultset (scalar context) || @row_objs (list context)
600 my @cds = $cd_rs->search_literal('year = ? AND title = ?', qw/2001 Reload/);
601 my $newrs = $artist_rs->search_literal('name = ?', 'Metallica');
603 Pass a literal chunk of SQL to be added to the conditional part of the
606 CAVEAT: C<search_literal> is provided for Class::DBI compatibility and should
607 only be used in that context. C<search_literal> is a convenience method.
608 It is equivalent to calling $schema->search(\[]), but if you want to ensure
609 columns are bound correctly, use C<search>.
611 Example of how to use C<search> instead of C<search_literal>
613 my @cds = $cd_rs->search_literal('cdid = ? AND (artist = ? OR artist = ?)', (2, 1, 2));
614 my @cds = $cd_rs->search(\[ 'cdid = ? AND (artist = ? OR artist = ?)', [ 'cdid', 2 ], [ 'artist', 1 ], [ 'artist', 2 ] ]);
617 See L<DBIx::Class::Manual::Cookbook/Searching> and
618 L<DBIx::Class::Manual::FAQ/Searching> for searching techniques that do not
619 require C<search_literal>.
624 my ($self, $sql, @bind) = @_;
626 if ( @bind && ref($bind[-1]) eq 'HASH' ) {
629 return $self->search(\[ $sql, map [ __DUMMY__ => $_ ], @bind ], ($attr || () ));
636 =item Arguments: \%columns_values | @pk_values, \%attrs?
638 =item Return Value: $row_object | undef
642 Finds and returns a single row based on supplied criteria. Takes either a
643 hashref with the same format as L</create> (including inference of foreign
644 keys from related objects), or a list of primary key values in the same
645 order as the L<primary columns|DBIx::Class::ResultSource/primary_columns>
646 declaration on the L</result_source>.
648 In either case an attempt is made to combine conditions already existing on
649 the resultset with the condition passed to this method.
651 To aid with preparing the correct query for the storage you may supply the
652 C<key> attribute, which is the name of a
653 L<unique constraint|DBIx::Class::ResultSource/add_unique_constraint> (the
654 unique constraint corresponding to the
655 L<primary columns|DBIx::Class::ResultSource/primary_columns> is always named
656 C<primary>). If the C<key> attribute has been supplied, and DBIC is unable
657 to construct a query that satisfies the named unique constraint fully (
658 non-NULL values for each column member of the constraint) an exception is
661 If no C<key> is specified, the search is carried over all unique constraints
662 which are fully defined by the available condition.
664 If no such constraint is found, C<find> currently defaults to a simple
665 C<< search->(\%column_values) >> which may or may not do what you expect.
666 Note that this fallback behavior may be deprecated in further versions. If
667 you need to search with arbitrary conditions - use L</search>. If the query
668 resulting from this fallback produces more than one row, a warning to the
669 effect is issued, though only the first row is constructed and returned as
672 In addition to C<key>, L</find> recognizes and applies standard
673 L<resultset attributes|/ATTRIBUTES> in the same way as L</search> does.
675 Note that if you have extra concerns about the correctness of the resulting
676 query you need to specify the C<key> attribute and supply the entire condition
677 as an argument to find (since it is not always possible to perform the
678 combination of the resultset condition with the supplied one, especially if
679 the resultset condition contains literal sql).
681 For example, to find a row by its primary key:
683 my $cd = $schema->resultset('CD')->find(5);
685 You can also find a row by a specific unique constraint:
687 my $cd = $schema->resultset('CD')->find(
689 artist => 'Massive Attack',
690 title => 'Mezzanine',
692 { key => 'cd_artist_title' }
695 See also L</find_or_create> and L</update_or_create>.
701 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
703 my $rsrc = $self->result_source;
705 # Parse out the condition from input
707 if (ref $_[0] eq 'HASH') {
708 $call_cond = { %{$_[0]} };
711 my $constraint = exists $attrs->{key} ? $attrs->{key} : 'primary';
712 my @c_cols = $rsrc->unique_constraint_columns($constraint);
714 $self->throw_exception(
715 "No constraint columns, maybe a malformed '$constraint' constraint?"
718 $self->throw_exception (
719 'find() expects either a column/value hashref, or a list of values '
720 . "corresponding to the columns of the specified unique constraint '$constraint'"
721 ) unless @c_cols == @_;
724 @{$call_cond}{@c_cols} = @_;
728 for my $key (keys %$call_cond) {
730 my $keyref = ref($call_cond->{$key})
732 my $relinfo = $rsrc->relationship_info($key)
734 my $val = delete $call_cond->{$key};
736 next if $keyref eq 'ARRAY'; # has_many for multi_create
738 my $rel_q = $rsrc->_resolve_condition(
739 $relinfo->{cond}, $val, $key, $key
741 die "Can't handle complex relationship conditions in find" if ref($rel_q) ne 'HASH';
742 @related{keys %$rel_q} = values %$rel_q;
746 # relationship conditions take precedence (?)
747 @{$call_cond}{keys %related} = values %related;
749 my $alias = exists $attrs->{alias} ? $attrs->{alias} : $self->{attrs}{alias};
751 if (exists $attrs->{key}) {
752 $final_cond = $self->_qualify_cond_columns (
754 $self->_build_unique_cond (
762 elsif ($self->{attrs}{accessor} and $self->{attrs}{accessor} eq 'single') {
763 # This means that we got here after a merger of relationship conditions
764 # in ::Relationship::Base::search_related (the row method), and furthermore
765 # the relationship is of the 'single' type. This means that the condition
766 # provided by the relationship (already attached to $self) is sufficient,
767 # as there can be only one row in the database that would satisfy the
771 # no key was specified - fall down to heuristics mode:
772 # run through all unique queries registered on the resultset, and
773 # 'OR' all qualifying queries together
774 my (@unique_queries, %seen_column_combinations);
775 for my $c_name ($rsrc->unique_constraint_names) {
776 next if $seen_column_combinations{
777 join "\x00", sort $rsrc->unique_constraint_columns($c_name)
780 push @unique_queries, try {
781 $self->_build_unique_cond ($c_name, $call_cond, 'croak_on_nulls')
785 $final_cond = @unique_queries
786 ? [ map { $self->_qualify_cond_columns($_, $alias) } @unique_queries ]
787 : $self->_non_unique_find_fallback ($call_cond, $attrs)
791 # Run the query, passing the result_class since it should propagate for find
792 my $rs = $self->search ($final_cond, {result_class => $self->result_class, %$attrs});
793 if (keys %{$rs->_resolved_attrs->{collapse}}) {
795 carp "Query returned more than one row" if $rs->next;
803 # This is a stop-gap method as agreed during the discussion on find() cleanup:
804 # http://lists.scsys.co.uk/pipermail/dbix-class/2010-October/009535.html
806 # It is invoked when find() is called in legacy-mode with insufficiently-unique
807 # condition. It is provided for overrides until a saner way forward is devised
809 # *NOTE* This is not a public method, and it's *GUARANTEED* to disappear down
810 # the road. Please adjust your tests accordingly to catch this situation early
811 # DBIx::Class::ResultSet->can('_non_unique_find_fallback') is reasonable
813 # The method will not be removed without an adequately complete replacement
814 # for strict-mode enforcement
815 sub _non_unique_find_fallback {
816 my ($self, $cond, $attrs) = @_;
818 return $self->_qualify_cond_columns(
820 exists $attrs->{alias}
822 : $self->{attrs}{alias}
827 sub _qualify_cond_columns {
828 my ($self, $cond, $alias) = @_;
830 my %aliased = %$cond;
831 for (keys %aliased) {
832 $aliased{"$alias.$_"} = delete $aliased{$_}
839 sub _build_unique_cond {
840 my ($self, $constraint_name, $extra_cond, $croak_on_null) = @_;
842 my @c_cols = $self->result_source->unique_constraint_columns($constraint_name);
844 # combination may fail if $self->{cond} is non-trivial
845 my ($final_cond) = try {
846 $self->_merge_with_rscond ($extra_cond)
851 # trim out everything not in $columns
852 $final_cond = { map {
853 exists $final_cond->{$_}
854 ? ( $_ => $final_cond->{$_} )
858 if (my @missing = grep
859 { ! ($croak_on_null ? defined $final_cond->{$_} : exists $final_cond->{$_}) }
862 $self->throw_exception( sprintf ( "Unable to satisfy requested constraint '%s', no values for column(s): %s",
864 join (', ', map { "'$_'" } @missing),
871 !$ENV{DBIC_NULLABLE_KEY_NOWARN}
873 my @undefs = grep { ! defined $final_cond->{$_} } (keys %$final_cond)
875 carp_unique ( sprintf (
876 "NULL/undef values supplied for requested unique constraint '%s' (NULL "
877 . 'values in column(s): %s). This is almost certainly not what you wanted, '
878 . 'though you can set DBIC_NULLABLE_KEY_NOWARN to disable this warning.',
880 join (', ', map { "'$_'" } @undefs),
887 =head2 search_related
891 =item Arguments: $rel, $cond, \%attrs?
893 =item Return Value: $new_resultset (scalar context) || @row_objs (list context)
897 $new_rs = $cd_rs->search_related('artist', {
901 Searches the specified relationship, optionally specifying a condition and
902 attributes for matching records. See L</ATTRIBUTES> for more information.
904 In list context, C<< ->all() >> is called implicitly on the resultset, thus
905 returning a list of row objects instead. To avoid that, use L</search_related_rs>.
907 See also L</search_related_rs>.
912 return shift->related_resultset(shift)->search(@_);
915 =head2 search_related_rs
917 This method works exactly the same as search_related, except that
918 it guarantees a resultset, even in list context.
922 sub search_related_rs {
923 return shift->related_resultset(shift)->search_rs(@_);
930 =item Arguments: none
932 =item Return Value: $cursor
936 Returns a storage-driven cursor to the given resultset. See
937 L<DBIx::Class::Cursor> for more information.
944 my $attrs = $self->_resolved_attrs_copy;
946 return $self->{cursor}
947 ||= $self->result_source->storage->select($attrs->{from}, $attrs->{select},
948 $attrs->{where},$attrs);
955 =item Arguments: $cond?
957 =item Return Value: $row_object | undef
961 my $cd = $schema->resultset('CD')->single({ year => 2001 });
963 Inflates the first result without creating a cursor if the resultset has
964 any records in it; if not returns C<undef>. Used by L</find> as a lean version
967 While this method can take an optional search condition (just like L</search>)
968 being a fast-code-path it does not recognize search attributes. If you need to
969 add extra joins or similar, call L</search> and then chain-call L</single> on the
970 L<DBIx::Class::ResultSet> returned.
976 As of 0.08100, this method enforces the assumption that the preceding
977 query returns only one row. If more than one row is returned, you will receive
980 Query returned more than one row
982 In this case, you should be using L</next> or L</find> instead, or if you really
983 know what you are doing, use the L</rows> attribute to explicitly limit the size
986 This method will also throw an exception if it is called on a resultset prefetching
987 has_many, as such a prefetch implies fetching multiple rows from the database in
988 order to assemble the resulting object.
995 my ($self, $where) = @_;
997 $self->throw_exception('single() only takes search conditions, no attributes. You want ->search( $cond, $attrs )->single()');
1000 my $attrs = $self->_resolved_attrs_copy;
1002 if (keys %{$attrs->{collapse}}) {
1003 $self->throw_exception(
1004 'single() can not be used on resultsets prefetching has_many. Use find( \%cond ) or next() instead'
1009 if (defined $attrs->{where}) {
1012 [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ }
1013 $where, delete $attrs->{where} ]
1016 $attrs->{where} = $where;
1020 my @data = $self->result_source->storage->select_single(
1021 $attrs->{from}, $attrs->{select},
1022 $attrs->{where}, $attrs
1025 return (@data ? ($self->_construct_object(@data))[0] : undef);
1031 # Recursively collapse the query, accumulating values for each column.
1033 sub _collapse_query {
1034 my ($self, $query, $collapsed) = @_;
1038 if (ref $query eq 'ARRAY') {
1039 foreach my $subquery (@$query) {
1040 next unless ref $subquery; # -or
1041 $collapsed = $self->_collapse_query($subquery, $collapsed);
1044 elsif (ref $query eq 'HASH') {
1045 if (keys %$query and (keys %$query)[0] eq '-and') {
1046 foreach my $subquery (@{$query->{-and}}) {
1047 $collapsed = $self->_collapse_query($subquery, $collapsed);
1051 foreach my $col (keys %$query) {
1052 my $value = $query->{$col};
1053 $collapsed->{$col}{$value}++;
1065 =item Arguments: $cond?
1067 =item Return Value: $resultsetcolumn
1071 my $max_length = $rs->get_column('length')->max;
1073 Returns a L<DBIx::Class::ResultSetColumn> instance for a column of the ResultSet.
1078 my ($self, $column) = @_;
1079 my $new = DBIx::Class::ResultSetColumn->new($self, $column);
1087 =item Arguments: $cond, \%attrs?
1089 =item Return Value: $resultset (scalar context) || @row_objs (list context)
1093 # WHERE title LIKE '%blue%'
1094 $cd_rs = $rs->search_like({ title => '%blue%'});
1096 Performs a search, but uses C<LIKE> instead of C<=> as the condition. Note
1097 that this is simply a convenience method retained for ex Class::DBI users.
1098 You most likely want to use L</search> with specific operators.
1100 For more information, see L<DBIx::Class::Manual::Cookbook>.
1102 This method is deprecated and will be removed in 0.09. Use L</search()>
1103 instead. An example conversion is:
1105 ->search_like({ foo => 'bar' });
1109 ->search({ foo => { like => 'bar' } });
1116 'search_like() is deprecated and will be removed in DBIC version 0.09.'
1117 .' Instead use ->search({ x => { -like => "y%" } })'
1118 .' (note the outer pair of {}s - they are important!)'
1120 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
1121 my $query = ref $_[0] eq 'HASH' ? { %{shift()} }: {@_};
1122 $query->{$_} = { 'like' => $query->{$_} } for keys %$query;
1123 return $class->search($query, { %$attrs });
1130 =item Arguments: $first, $last
1132 =item Return Value: $resultset (scalar context) || @row_objs (list context)
1136 Returns a resultset or object list representing a subset of elements from the
1137 resultset slice is called on. Indexes are from 0, i.e., to get the first
1138 three records, call:
1140 my ($one, $two, $three) = $rs->slice(0, 2);
1145 my ($self, $min, $max) = @_;
1146 my $attrs = {}; # = { %{ $self->{attrs} || {} } };
1147 $attrs->{offset} = $self->{attrs}{offset} || 0;
1148 $attrs->{offset} += $min;
1149 $attrs->{rows} = ($max ? ($max - $min + 1) : 1);
1150 return $self->search(undef, $attrs);
1151 #my $slice = (ref $self)->new($self->result_source, $attrs);
1152 #return (wantarray ? $slice->all : $slice);
1159 =item Arguments: none
1161 =item Return Value: $result | undef
1165 Returns the next element in the resultset (C<undef> is there is none).
1167 Can be used to efficiently iterate over records in the resultset:
1169 my $rs = $schema->resultset('CD')->search;
1170 while (my $cd = $rs->next) {
1174 Note that you need to store the resultset object, and call C<next> on it.
1175 Calling C<< resultset('Table')->next >> repeatedly will always return the
1176 first record from the resultset.
1182 if (my $cache = $self->get_cache) {
1183 $self->{all_cache_position} ||= 0;
1184 return $cache->[$self->{all_cache_position}++];
1186 if ($self->{attrs}{cache}) {
1187 delete $self->{pager};
1188 $self->{all_cache_position} = 1;
1189 return ($self->all)[0];
1191 if ($self->{stashed_objects}) {
1192 my $obj = shift(@{$self->{stashed_objects}});
1193 delete $self->{stashed_objects} unless @{$self->{stashed_objects}};
1197 exists $self->{stashed_row}
1198 ? @{delete $self->{stashed_row}}
1199 : $self->cursor->next
1201 return undef unless (@row);
1202 my ($row, @more) = $self->_construct_object(@row);
1203 $self->{stashed_objects} = \@more if @more;
1207 sub _construct_object {
1208 my ($self, @row) = @_;
1210 my $info = $self->_collapse_result($self->{_attrs}{as}, \@row)
1212 my @new = $self->result_class->inflate_result($self->result_source, @$info);
1213 @new = $self->{_attrs}{record_filter}->(@new)
1214 if exists $self->{_attrs}{record_filter};
1218 sub _collapse_result {
1219 my ($self, $as_proto, $row) = @_;
1223 # 'foo' => [ undef, 'foo' ]
1224 # 'foo.bar' => [ 'foo', 'bar' ]
1225 # 'foo.bar.baz' => [ 'foo.bar', 'baz' ]
1227 my @construct_as = map { [ (/^(?:(.*)\.)?([^.]+)$/) ] } @$as_proto;
1229 my %collapse = %{$self->{_attrs}{collapse}||{}};
1233 # if we're doing collapsing (has_many prefetch) we need to grab records
1234 # until the PK changes, so fill @pri_index. if not, we leave it empty so
1235 # we know we don't have to bother.
1237 # the reason for not using the collapse stuff directly is because if you
1238 # had for e.g. two artists in a row with no cds, the collapse info for
1239 # both would be NULL (undef) so you'd lose the second artist
1241 # store just the index so we can check the array positions from the row
1242 # without having to contruct the full hash
1244 if (keys %collapse) {
1245 my %pri = map { ($_ => 1) } $self->result_source->_pri_cols;
1246 foreach my $i (0 .. $#construct_as) {
1247 next if defined($construct_as[$i][0]); # only self table
1248 if (delete $pri{$construct_as[$i][1]}) {
1249 push(@pri_index, $i);
1251 last unless keys %pri; # short circuit (Johnny Five Is Alive!)
1255 # no need to do an if, it'll be empty if @pri_index is empty anyway
1257 my %pri_vals = map { ($_ => $copy[$_]) } @pri_index;
1261 do { # no need to check anything at the front, we always want the first row
1265 foreach my $this_as (@construct_as) {
1266 $const{$this_as->[0]||''}{$this_as->[1]} = shift(@copy);
1269 push(@const_rows, \%const);
1271 } until ( # no pri_index => no collapse => drop straight out
1274 do { # get another row, stash it, drop out if different PK
1276 @copy = $self->cursor->next;
1277 $self->{stashed_row} = \@copy;
1279 # last thing in do block, counts as true if anything doesn't match
1281 # check xor defined first for NULL vs. NOT NULL then if one is
1282 # defined the other must be so check string equality
1285 (defined $pri_vals{$_} ^ defined $copy[$_])
1286 || (defined $pri_vals{$_} && ($pri_vals{$_} ne $copy[$_]))
1291 my $alias = $self->{attrs}{alias};
1298 foreach my $const (@const_rows) {
1299 scalar @const_keys or do {
1300 @const_keys = sort { length($a) <=> length($b) } keys %$const;
1302 foreach my $key (@const_keys) {
1305 my @parts = split(/\./, $key);
1307 my $data = $const->{$key};
1308 foreach my $p (@parts) {
1309 $target = $target->[1]->{$p} ||= [];
1311 if ($cur eq ".${key}" && (my @ckey = @{$collapse{$cur}||[]})) {
1312 # collapsing at this point and on final part
1313 my $pos = $collapse_pos{$cur};
1314 CK: foreach my $ck (@ckey) {
1315 if (!defined $pos->{$ck} || $pos->{$ck} ne $data->{$ck}) {
1316 $collapse_pos{$cur} = $data;
1317 delete @collapse_pos{ # clear all positioning for sub-entries
1318 grep { m/^\Q${cur}.\E/ } keys %collapse_pos
1325 if (exists $collapse{$cur}) {
1326 $target = $target->[-1];
1329 $target->[0] = $data;
1331 $info->[0] = $const->{$key};
1339 =head2 result_source
1343 =item Arguments: $result_source?
1345 =item Return Value: $result_source
1349 An accessor for the primary ResultSource object from which this ResultSet
1356 =item Arguments: $result_class?
1358 =item Return Value: $result_class
1362 An accessor for the class to use when creating row objects. Defaults to
1363 C<< result_source->result_class >> - which in most cases is the name of the
1364 L<"table"|DBIx::Class::Manual::Glossary/"ResultSource"> class.
1366 Note that changing the result_class will also remove any components
1367 that were originally loaded in the source class via
1368 L<DBIx::Class::ResultSource/load_components>. Any overloaded methods
1369 in the original source class will not run.
1374 my ($self, $result_class) = @_;
1375 if ($result_class) {
1376 unless (ref $result_class) { # don't fire this for an object
1377 $self->ensure_class_loaded($result_class);
1379 $self->_result_class($result_class);
1380 # THIS LINE WOULD BE A BUG - this accessor specifically exists to
1381 # permit the user to set result class on one result set only; it only
1382 # chains if provided to search()
1383 #$self->{attrs}{result_class} = $result_class if ref $self;
1385 $self->_result_class;
1392 =item Arguments: $cond, \%attrs??
1394 =item Return Value: $count
1398 Performs an SQL C<COUNT> with the same query as the resultset was built
1399 with to find the number of elements. Passing arguments is equivalent to
1400 C<< $rs->search ($cond, \%attrs)->count >>
1406 return $self->search(@_)->count if @_ and defined $_[0];
1407 return scalar @{ $self->get_cache } if $self->get_cache;
1409 my $attrs = $self->_resolved_attrs_copy;
1411 # this is a little optimization - it is faster to do the limit
1412 # adjustments in software, instead of a subquery
1413 my $rows = delete $attrs->{rows};
1414 my $offset = delete $attrs->{offset};
1417 if ($self->_has_resolved_attr (qw/collapse group_by/)) {
1418 $crs = $self->_count_subq_rs ($attrs);
1421 $crs = $self->_count_rs ($attrs);
1423 my $count = $crs->next;
1425 $count -= $offset if $offset;
1426 $count = $rows if $rows and $rows < $count;
1427 $count = 0 if ($count < 0);
1436 =item Arguments: $cond, \%attrs??
1438 =item Return Value: $count_rs
1442 Same as L</count> but returns a L<DBIx::Class::ResultSetColumn> object.
1443 This can be very handy for subqueries:
1445 ->search( { amount => $some_rs->count_rs->as_query } )
1447 As with regular resultsets the SQL query will be executed only after
1448 the resultset is accessed via L</next> or L</all>. That would return
1449 the same single value obtainable via L</count>.
1455 return $self->search(@_)->count_rs if @_;
1457 # this may look like a lack of abstraction (count() does about the same)
1458 # but in fact an _rs *must* use a subquery for the limits, as the
1459 # software based limiting can not be ported if this $rs is to be used
1460 # in a subquery itself (i.e. ->as_query)
1461 if ($self->_has_resolved_attr (qw/collapse group_by offset rows/)) {
1462 return $self->_count_subq_rs;
1465 return $self->_count_rs;
1470 # returns a ResultSetColumn object tied to the count query
1473 my ($self, $attrs) = @_;
1475 my $rsrc = $self->result_source;
1476 $attrs ||= $self->_resolved_attrs;
1478 my $tmp_attrs = { %$attrs };
1479 # take off any limits, record_filter is cdbi, and no point of ordering nor locking a count
1480 delete @{$tmp_attrs}{qw/rows offset order_by record_filter for/};
1482 # overwrite the selector (supplied by the storage)
1483 $tmp_attrs->{select} = $rsrc->storage->_count_select ($rsrc, $attrs);
1484 $tmp_attrs->{as} = 'count';
1485 delete @{$tmp_attrs}{qw/columns/};
1487 my $tmp_rs = $rsrc->resultset_class->new($rsrc, $tmp_attrs)->get_column ('count');
1493 # same as above but uses a subquery
1495 sub _count_subq_rs {
1496 my ($self, $attrs) = @_;
1498 my $rsrc = $self->result_source;
1499 $attrs ||= $self->_resolved_attrs;
1501 my $sub_attrs = { %$attrs };
1502 # extra selectors do not go in the subquery and there is no point of ordering it, nor locking it
1503 delete @{$sub_attrs}{qw/collapse columns as select _prefetch_selector_range order_by for/};
1505 # if we multi-prefetch we group_by primary keys only as this is what we would
1506 # get out of the rs via ->next/->all. We *DO WANT* to clobber old group_by regardless
1507 if ( keys %{$attrs->{collapse}} ) {
1508 $sub_attrs->{group_by} = [ map { "$attrs->{alias}.$_" } ($rsrc->_pri_cols) ]
1511 # Calculate subquery selector
1512 if (my $g = $sub_attrs->{group_by}) {
1514 my $sql_maker = $rsrc->storage->sql_maker;
1516 # necessary as the group_by may refer to aliased functions
1518 for my $sel (@{$attrs->{select}}) {
1519 $sel_index->{$sel->{-as}} = $sel
1520 if (ref $sel eq 'HASH' and $sel->{-as});
1523 # anything from the original select mentioned on the group-by needs to make it to the inner selector
1524 # also look for named aggregates referred in the having clause
1525 # having often contains scalarrefs - thus parse it out entirely
1527 if ($attrs->{having}) {
1528 local $sql_maker->{having_bind};
1529 local $sql_maker->{quote_char} = $sql_maker->{quote_char};
1530 local $sql_maker->{name_sep} = $sql_maker->{name_sep};
1531 unless (defined $sql_maker->{quote_char} and length $sql_maker->{quote_char}) {
1532 $sql_maker->{quote_char} = [ "\x00", "\xFF" ];
1533 # if we don't unset it we screw up retarded but unfortunately working
1534 # 'MAX(foo.bar)' => { '>', 3 }
1535 $sql_maker->{name_sep} = '';
1538 my ($lquote, $rquote, $sep) = map { quotemeta $_ } ($sql_maker->_quote_chars, $sql_maker->name_sep);
1540 my $sql = $sql_maker->_parse_rs_attrs ({ having => $attrs->{having} });
1542 # search for both a proper quoted qualified string, for a naive unquoted scalarref
1543 # and if all fails for an utterly naive quoted scalar-with-function
1545 $rquote $sep $lquote (.+?) $rquote
1547 [\s,] \w+ \. (\w+) [\s,]
1549 [\s,] $lquote (.+?) $rquote [\s,]
1551 push @parts, ($1 || $2 || $3); # one of them matched if we got here
1556 my $colpiece = $sel_index->{$_} || $_;
1558 # unqualify join-based group_by's. Arcane but possible query
1559 # also horrible horrible hack to alias a column (not a func.)
1560 # (probably need to introduce SQLA syntax)
1561 if ($colpiece =~ /\./ && $colpiece !~ /^$attrs->{alias}\./) {
1564 $colpiece = \ sprintf ('%s AS %s', map { $sql_maker->_quote ($_) } ($colpiece, $as) );
1566 push @{$sub_attrs->{select}}, $colpiece;
1570 my @pcols = map { "$attrs->{alias}.$_" } ($rsrc->primary_columns);
1571 $sub_attrs->{select} = @pcols ? \@pcols : [ 1 ];
1574 return $rsrc->resultset_class
1575 ->new ($rsrc, $sub_attrs)
1577 ->search ({}, { columns => { count => $rsrc->storage->_count_select ($rsrc, $attrs) } })
1578 ->get_column ('count');
1585 =head2 count_literal
1589 =item Arguments: $sql_fragment, @bind_values
1591 =item Return Value: $count
1595 Counts the results in a literal query. Equivalent to calling L</search_literal>
1596 with the passed arguments, then L</count>.
1600 sub count_literal { shift->search_literal(@_)->count; }
1606 =item Arguments: none
1608 =item Return Value: @objects
1612 Returns all elements in the resultset.
1619 $self->throw_exception("all() doesn't take any arguments, you probably wanted ->search(...)->all()");
1622 return @{ $self->get_cache } if $self->get_cache;
1626 if (keys %{$self->_resolved_attrs->{collapse}}) {
1627 # Using $self->cursor->all is really just an optimisation.
1628 # If we're collapsing has_many prefetches it probably makes
1629 # very little difference, and this is cleaner than hacking
1630 # _construct_object to survive the approach
1631 $self->cursor->reset;
1632 my @row = $self->cursor->next;
1634 push(@obj, $self->_construct_object(@row));
1635 @row = (exists $self->{stashed_row}
1636 ? @{delete $self->{stashed_row}}
1637 : $self->cursor->next);
1640 @obj = map { $self->_construct_object(@$_) } $self->cursor->all;
1643 $self->set_cache(\@obj) if $self->{attrs}{cache};
1652 =item Arguments: none
1654 =item Return Value: $self
1658 Resets the resultset's cursor, so you can iterate through the elements again.
1659 Implicitly resets the storage cursor, so a subsequent L</next> will trigger
1666 delete $self->{_attrs} if exists $self->{_attrs};
1667 $self->{all_cache_position} = 0;
1668 $self->cursor->reset;
1676 =item Arguments: none
1678 =item Return Value: $object | undef
1682 Resets the resultset and returns an object for the first result (or C<undef>
1683 if the resultset is empty).
1688 return $_[0]->reset->next;
1694 # Determines whether and what type of subquery is required for the $rs operation.
1695 # If grouping is necessary either supplies its own, or verifies the current one
1696 # After all is done delegates to the proper storage method.
1698 sub _rs_update_delete {
1699 my ($self, $op, $values) = @_;
1701 my $rsrc = $self->result_source;
1703 my $needs_group_by_subq = $self->_has_resolved_attr (qw/collapse group_by -join/);
1704 my $needs_subq = $needs_group_by_subq || $self->_has_resolved_attr(qw/rows offset/);
1706 if ($needs_group_by_subq or $needs_subq) {
1708 # make a new $rs selecting only the PKs (that's all we really need)
1709 my $attrs = $self->_resolved_attrs_copy;
1712 delete $attrs->{$_} for qw/collapse _collapse_order_by select _prefetch_selector_range as/;
1713 $attrs->{columns} = [ map { "$attrs->{alias}.$_" } ($self->result_source->_pri_cols) ];
1715 if ($needs_group_by_subq) {
1716 # make sure no group_by was supplied, or if there is one - make sure it matches
1717 # the columns compiled above perfectly. Anything else can not be sanely executed
1718 # on most databases so croak right then and there
1720 if (my $g = $attrs->{group_by}) {
1721 my @current_group_by = map
1722 { $_ =~ /\./ ? $_ : "$attrs->{alias}.$_" }
1727 join ("\x00", sort @current_group_by)
1729 join ("\x00", sort @{$attrs->{columns}} )
1731 $self->throw_exception (
1732 "You have just attempted a $op operation on a resultset which does group_by"
1733 . ' on columns other than the primary keys, while DBIC internally needs to retrieve'
1734 . ' the primary keys in a subselect. All sane RDBMS engines do not support this'
1735 . ' kind of queries. Please retry the operation with a modified group_by or'
1736 . ' without using one at all.'
1741 $attrs->{group_by} = $attrs->{columns};
1745 my $subrs = (ref $self)->new($rsrc, $attrs);
1746 return $self->result_source->storage->_subq_update_delete($subrs, $op, $values);
1749 # Most databases do not allow aliasing of tables in UPDATE/DELETE. Thus
1750 # a condition containing 'me' or other table prefixes will not work
1751 # at all. What this code tries to do (badly) is to generate a condition
1752 # with the qualifiers removed, by exploiting the quote mechanism of sqla
1754 # this is atrocious and should be replaced by normal sqla introspection
1756 my ($sql, @bind) = do {
1757 my $sqla = $rsrc->storage->sql_maker;
1758 local $sqla->{_dequalify_idents} = 1;
1759 $sqla->_recurse_where($self->{cond});
1762 return $rsrc->storage->$op(
1764 $op eq 'update' ? $values : (),
1765 $self->{cond} ? \[$sql, @bind] : (),
1774 =item Arguments: \%values
1776 =item Return Value: $storage_rv
1780 Sets the specified columns in the resultset to the supplied values in a
1781 single query. Note that this will not run any accessor/set_column/update
1782 triggers, nor will it update any row object instances derived from this
1783 resultset (this includes the contents of the L<resultset cache|/set_cache>
1784 if any). See L</update_all> if you need to execute any on-update
1785 triggers or cascades defined either by you or a
1786 L<result component|DBIx::Class::Manual::Component/WHAT_IS_A_COMPONENT>.
1788 The return value is a pass through of what the underlying
1789 storage backend returned, and may vary. See L<DBI/execute> for the most
1794 Note that L</update> does not process/deflate any of the values passed in.
1795 This is unlike the corresponding L<DBIx::Class::Row/update>. The user must
1796 ensure manually that any value passed to this method will stringify to
1797 something the RDBMS knows how to deal with. A notable example is the
1798 handling of L<DateTime> objects, for more info see:
1799 L<DBIx::Class::Manual::Cookbook/Formatting_DateTime_objects_in_queries>.
1804 my ($self, $values) = @_;
1805 $self->throw_exception('Values for update must be a hash')
1806 unless ref $values eq 'HASH';
1808 return $self->_rs_update_delete ('update', $values);
1815 =item Arguments: \%values
1817 =item Return Value: 1
1821 Fetches all objects and updates them one at a time via
1822 L<DBIx::Class::Row/update>. Note that C<update_all> will run DBIC defined
1823 triggers, while L</update> will not.
1828 my ($self, $values) = @_;
1829 $self->throw_exception('Values for update_all must be a hash')
1830 unless ref $values eq 'HASH';
1832 my $guard = $self->result_source->schema->txn_scope_guard;
1833 $_->update({%$values}) for $self->all; # shallow copy - update will mangle it
1842 =item Arguments: none
1844 =item Return Value: $storage_rv
1848 Deletes the rows matching this resultset in a single query. Note that this
1849 will not run any delete triggers, nor will it alter the
1850 L<in_storage|DBIx::Class::Row/in_storage> status of any row object instances
1851 derived from this resultset (this includes the contents of the
1852 L<resultset cache|/set_cache> if any). See L</delete_all> if you need to
1853 execute any on-delete triggers or cascades defined either by you or a
1854 L<result component|DBIx::Class::Manual::Component/WHAT_IS_A_COMPONENT>.
1856 The return value is a pass through of what the underlying storage backend
1857 returned, and may vary. See L<DBI/execute> for the most common case.
1863 $self->throw_exception('delete does not accept any arguments')
1866 return $self->_rs_update_delete ('delete');
1873 =item Arguments: none
1875 =item Return Value: 1
1879 Fetches all objects and deletes them one at a time via
1880 L<DBIx::Class::Row/delete>. Note that C<delete_all> will run DBIC defined
1881 triggers, while L</delete> will not.
1887 $self->throw_exception('delete_all does not accept any arguments')
1890 my $guard = $self->result_source->schema->txn_scope_guard;
1891 $_->delete for $self->all;
1900 =item Arguments: \@data;
1904 Accepts either an arrayref of hashrefs or alternatively an arrayref of arrayrefs.
1905 For the arrayref of hashrefs style each hashref should be a structure suitable
1906 for submitting to a $resultset->create(...) method.
1908 In void context, C<insert_bulk> in L<DBIx::Class::Storage::DBI> is used
1909 to insert the data, as this is a faster method.
1911 Otherwise, each set of data is inserted into the database using
1912 L<DBIx::Class::ResultSet/create>, and the resulting objects are
1913 accumulated into an array. The array itself, or an array reference
1914 is returned depending on scalar or list context.
1916 Example: Assuming an Artist Class that has many CDs Classes relating:
1918 my $Artist_rs = $schema->resultset("Artist");
1920 ## Void Context Example
1921 $Artist_rs->populate([
1922 { artistid => 4, name => 'Manufactured Crap', cds => [
1923 { title => 'My First CD', year => 2006 },
1924 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
1927 { artistid => 5, name => 'Angsty-Whiny Girl', cds => [
1928 { title => 'My parents sold me to a record company', year => 2005 },
1929 { title => 'Why Am I So Ugly?', year => 2006 },
1930 { title => 'I Got Surgery and am now Popular', year => 2007 }
1935 ## Array Context Example
1936 my ($ArtistOne, $ArtistTwo, $ArtistThree) = $Artist_rs->populate([
1937 { name => "Artist One"},
1938 { name => "Artist Two"},
1939 { name => "Artist Three", cds=> [
1940 { title => "First CD", year => 2007},
1941 { title => "Second CD", year => 2008},
1945 print $ArtistOne->name; ## response is 'Artist One'
1946 print $ArtistThree->cds->count ## reponse is '2'
1948 For the arrayref of arrayrefs style, the first element should be a list of the
1949 fieldsnames to which the remaining elements are rows being inserted. For
1952 $Arstist_rs->populate([
1953 [qw/artistid name/],
1954 [100, 'A Formally Unknown Singer'],
1955 [101, 'A singer that jumped the shark two albums ago'],
1956 [102, 'An actually cool singer'],
1959 Please note an important effect on your data when choosing between void and
1960 wantarray context. Since void context goes straight to C<insert_bulk> in
1961 L<DBIx::Class::Storage::DBI> this will skip any component that is overriding
1962 C<insert>. So if you are using something like L<DBIx-Class-UUIDColumns> to
1963 create primary keys for you, you will find that your PKs are empty. In this
1964 case you will have to use the wantarray context in order to create those
1972 # cruft placed in standalone method
1973 my $data = $self->_normalize_populate_args(@_);
1975 return unless @$data;
1977 if(defined wantarray) {
1979 foreach my $item (@$data) {
1980 push(@created, $self->create($item));
1982 return wantarray ? @created : \@created;
1985 my $first = $data->[0];
1987 # if a column is a registered relationship, and is a non-blessed hash/array, consider
1988 # it relationship data
1989 my (@rels, @columns);
1990 my $rsrc = $self->result_source;
1991 my $rels = { map { $_ => $rsrc->relationship_info($_) } $rsrc->relationships };
1992 for (keys %$first) {
1993 my $ref = ref $first->{$_};
1994 $rels->{$_} && ($ref eq 'ARRAY' or $ref eq 'HASH')
2000 my @pks = $rsrc->primary_columns;
2002 ## do the belongs_to relationships
2003 foreach my $index (0..$#$data) {
2005 # delegate to create() for any dataset without primary keys with specified relationships
2006 if (grep { !defined $data->[$index]->{$_} } @pks ) {
2008 if (grep { ref $data->[$index]{$r} eq $_ } qw/HASH ARRAY/) { # a related set must be a HASH or AoH
2009 my @ret = $self->populate($data);
2015 foreach my $rel (@rels) {
2016 next unless ref $data->[$index]->{$rel} eq "HASH";
2017 my $result = $self->related_resultset($rel)->create($data->[$index]->{$rel});
2018 my ($reverse_relname, $reverse_relinfo) = %{$rsrc->reverse_relationship_info($rel)};
2019 my $related = $result->result_source->_resolve_condition(
2020 $reverse_relinfo->{cond},
2026 delete $data->[$index]->{$rel};
2027 $data->[$index] = {%{$data->[$index]}, %$related};
2029 push @columns, keys %$related if $index == 0;
2033 ## inherit the data locked in the conditions of the resultset
2034 my ($rs_data) = $self->_merge_with_rscond({});
2035 delete @{$rs_data}{@columns};
2036 my @inherit_cols = keys %$rs_data;
2037 my @inherit_data = values %$rs_data;
2039 ## do bulk insert on current row
2040 $rsrc->storage->insert_bulk(
2042 [@columns, @inherit_cols],
2043 [ map { [ @$_{@columns}, @inherit_data ] } @$data ],
2046 ## do the has_many relationships
2047 foreach my $item (@$data) {
2051 foreach my $rel (@rels) {
2052 next unless ref $item->{$rel} eq "ARRAY" && @{ $item->{$rel} };
2054 $main_row ||= $self->new_result({map { $_ => $item->{$_} } @pks});
2056 my $child = $main_row->$rel;
2058 my $related = $child->result_source->_resolve_condition(
2059 $rels->{$rel}{cond},
2065 my @rows_to_add = ref $item->{$rel} eq 'ARRAY' ? @{$item->{$rel}} : ($item->{$rel});
2066 my @populate = map { {%$_, %$related} } @rows_to_add;
2068 $child->populate( \@populate );
2075 # populate() argumnets went over several incarnations
2076 # What we ultimately support is AoH
2077 sub _normalize_populate_args {
2078 my ($self, $arg) = @_;
2080 if (ref $arg eq 'ARRAY') {
2084 elsif (ref $arg->[0] eq 'HASH') {
2087 elsif (ref $arg->[0] eq 'ARRAY') {
2089 my @colnames = @{$arg->[0]};
2090 foreach my $values (@{$arg}[1 .. $#$arg]) {
2091 push @ret, { map { $colnames[$_] => $values->[$_] } (0 .. $#colnames) };
2097 $self->throw_exception('Populate expects an arrayref of hashrefs or arrayref of arrayrefs');
2104 =item Arguments: none
2106 =item Return Value: $pager
2110 Return Value a L<Data::Page> object for the current resultset. Only makes
2111 sense for queries with a C<page> attribute.
2113 To get the full count of entries for a paged resultset, call
2114 C<total_entries> on the L<Data::Page> object.
2118 # make a wizard good for both a scalar and a hashref
2119 my $mk_lazy_count_wizard = sub {
2120 require Variable::Magic;
2122 my $stash = { total_rs => shift };
2123 my $slot = shift; # only used by the hashref magic
2125 my $magic = Variable::Magic::wizard (
2126 data => sub { $stash },
2132 # set value lazily, and dispell for good
2133 ${$_[0]} = $_[1]{total_rs}->count;
2134 Variable::Magic::dispell (${$_[0]}, $_[1]{magic_selfref});
2138 # an explicit set implies dispell as well
2139 # the unless() is to work around "fun and giggles" below
2140 Variable::Magic::dispell (${$_[0]}, $_[1]{magic_selfref})
2141 unless (caller(2))[3] eq 'DBIx::Class::ResultSet::pager';
2148 if ($_[2] eq $slot and !$_[1]{inactive}) {
2149 my $cnt = $_[1]{total_rs}->count;
2150 $_[0]->{$slot} = $cnt;
2152 # attempting to dispell in a fetch handle (works in store), seems
2153 # to invariable segfault on 5.10, 5.12, 5.13 :(
2154 # so use an inactivator instead
2155 #Variable::Magic::dispell (%{$_[0]}, $_[1]{magic_selfref});
2161 if (! $_[1]{inactive} and $_[2] eq $slot) {
2162 #Variable::Magic::dispell (%{$_[0]}, $_[1]{magic_selfref});
2164 unless (caller(2))[3] eq 'DBIx::Class::ResultSet::pager';
2171 $stash->{magic_selfref} = $magic;
2172 weaken ($stash->{magic_selfref}); # this fails on 5.8.1
2177 # the tie class for 5.8.1
2179 package # hide from pause
2180 DBIx::Class::__DBIC_LAZY_RS_COUNT__;
2181 use base qw/Tie::Hash/;
2183 sub FIRSTKEY { my $dummy = scalar keys %{$_[0]{data}}; each %{$_[0]{data}} }
2184 sub NEXTKEY { each %{$_[0]{data}} }
2185 sub EXISTS { exists $_[0]{data}{$_[1]} }
2186 sub DELETE { delete $_[0]{data}{$_[1]} }
2187 sub CLEAR { %{$_[0]{data}} = () }
2188 sub SCALAR { scalar %{$_[0]{data}} }
2191 $_[1]{data} = {%{$_[1]{selfref}}};
2192 %{$_[1]{selfref}} = ();
2193 Scalar::Util::weaken ($_[1]{selfref});
2194 return bless ($_[1], $_[0]);
2198 if ($_[1] eq $_[0]{slot}) {
2199 my $cnt = $_[0]{data}{$_[1]} = $_[0]{total_rs}->count;
2200 untie %{$_[0]{selfref}};
2201 %{$_[0]{selfref}} = %{$_[0]{data}};
2210 $_[0]{data}{$_[1]} = $_[2];
2211 if ($_[1] eq $_[0]{slot}) {
2212 untie %{$_[0]{selfref}};
2213 %{$_[0]{selfref}} = %{$_[0]{data}};
2222 return $self->{pager} if $self->{pager};
2224 my $attrs = $self->{attrs};
2225 if (!defined $attrs->{page}) {
2226 $self->throw_exception("Can't create pager for non-paged rs");
2228 elsif ($attrs->{page} <= 0) {
2229 $self->throw_exception('Invalid page number (page-numbers are 1-based)');
2231 $attrs->{rows} ||= 10;
2233 # throw away the paging flags and re-run the count (possibly
2234 # with a subselect) to get the real total count
2235 my $count_attrs = { %$attrs };
2236 delete $count_attrs->{$_} for qw/rows offset page pager/;
2237 my $total_rs = (ref $self)->new($self->result_source, $count_attrs);
2240 ### the following may seem awkward and dirty, but it's a thought-experiment
2241 ### necessary for future development of DBIx::DS. Do *NOT* change this code
2242 ### before talking to ribasushi/mst
2245 my $pager = Data::Page->new(
2246 0, #start with an empty set
2248 $self->{attrs}{page},
2251 my $data_slot = 'total_entries';
2253 # Since we are interested in a cached value (once it's set - it's set), every
2254 # technique will detach from the magic-host once the time comes to fire the
2255 # ->count (or in the segfaulting case of >= 5.10 it will deactivate itself)
2257 if ($] < 5.008003) {
2258 # 5.8.1 throws 'Modification of a read-only value attempted' when one tries
2259 # to weakref the magic container :(
2261 tie (%$pager, 'DBIx::Class::__DBIC_LAZY_RS_COUNT__',
2262 { slot => $data_slot, total_rs => $total_rs, selfref => $pager }
2265 elsif ($] < 5.010) {
2266 # We can use magic on the hash value slot. It's interesting that the magic is
2267 # attached to the hash-slot, and does *not* stop working once I do the dummy
2268 # assignments after the cast()
2269 # tested on 5.8.3 and 5.8.9
2270 my $magic = $mk_lazy_count_wizard->($total_rs);
2271 Variable::Magic::cast ( $pager->{$data_slot}, $magic );
2273 # this is for fun and giggles
2274 $pager->{$data_slot} = -1;
2275 $pager->{$data_slot} = 0;
2277 # this does not work for scalars, but works with
2279 #my %vals = %$pager;
2284 # And the uvar magic
2285 # works on 5.10.1, 5.12.1 and 5.13.4 in its current form,
2286 # however see the wizard maker for more notes
2287 my $magic = $mk_lazy_count_wizard->($total_rs, $data_slot);
2288 Variable::Magic::cast ( %$pager, $magic );
2291 $pager->{$data_slot} = -1;
2292 $pager->{$data_slot} = 0;
2300 return $self->{pager} = $pager;
2307 =item Arguments: $page_number
2309 =item Return Value: $rs
2313 Returns a resultset for the $page_number page of the resultset on which page
2314 is called, where each page contains a number of rows equal to the 'rows'
2315 attribute set on the resultset (10 by default).
2320 my ($self, $page) = @_;
2321 return (ref $self)->new($self->result_source, { %{$self->{attrs}}, page => $page });
2328 =item Arguments: \%vals
2330 =item Return Value: $rowobject
2334 Creates a new row object in the resultset's result class and returns
2335 it. The row is not inserted into the database at this point, call
2336 L<DBIx::Class::Row/insert> to do that. Calling L<DBIx::Class::Row/in_storage>
2337 will tell you whether the row object has been inserted or not.
2339 Passes the hashref of input on to L<DBIx::Class::Row/new>.
2344 my ($self, $values) = @_;
2345 $self->throw_exception( "new_result needs a hash" )
2346 unless (ref $values eq 'HASH');
2348 my ($merged_cond, $cols_from_relations) = $self->_merge_with_rscond($values);
2352 @$cols_from_relations
2353 ? (-cols_from_relations => $cols_from_relations)
2355 -result_source => $self->result_source, # DO NOT REMOVE THIS, REQUIRED
2358 return $self->result_class->new(\%new);
2361 # _merge_with_rscond
2363 # Takes a simple hash of K/V data and returns its copy merged with the
2364 # condition already present on the resultset. Additionally returns an
2365 # arrayref of value/condition names, which were inferred from related
2366 # objects (this is needed for in-memory related objects)
2367 sub _merge_with_rscond {
2368 my ($self, $data) = @_;
2370 my (%new_data, @cols_from_relations);
2372 my $alias = $self->{attrs}{alias};
2374 if (! defined $self->{cond}) {
2375 # just massage $data below
2377 elsif ($self->{cond} eq $DBIx::Class::ResultSource::UNRESOLVABLE_CONDITION) {
2378 %new_data = %{ $self->{attrs}{related_objects} || {} }; # nothing might have been inserted yet
2379 @cols_from_relations = keys %new_data;
2381 elsif (ref $self->{cond} ne 'HASH') {
2382 $self->throw_exception(
2383 "Can't abstract implicit construct, resultset condition not a hash"
2387 # precendence must be given to passed values over values inherited from
2388 # the cond, so the order here is important.
2389 my $collapsed_cond = $self->_collapse_cond($self->{cond});
2390 my %implied = %{$self->_remove_alias($collapsed_cond, $alias)};
2392 while ( my($col, $value) = each %implied ) {
2393 my $vref = ref $value;
2399 (keys %$value)[0] eq '='
2401 $new_data{$col} = $value->{'='};
2403 elsif( !$vref or $vref eq 'SCALAR' or blessed($value) ) {
2404 $new_data{$col} = $value;
2411 %{ $self->_remove_alias($data, $alias) },
2414 return (\%new_data, \@cols_from_relations);
2417 # _has_resolved_attr
2419 # determines if the resultset defines at least one
2420 # of the attributes supplied
2422 # used to determine if a subquery is neccessary
2424 # supports some virtual attributes:
2426 # This will scan for any joins being present on the resultset.
2427 # It is not a mere key-search but a deep inspection of {from}
2430 sub _has_resolved_attr {
2431 my ($self, @attr_names) = @_;
2433 my $attrs = $self->_resolved_attrs;
2437 for my $n (@attr_names) {
2438 if (grep { $n eq $_ } (qw/-join/) ) {
2439 $extra_checks{$n}++;
2443 my $attr = $attrs->{$n};
2445 next if not defined $attr;
2447 if (ref $attr eq 'HASH') {
2448 return 1 if keys %$attr;
2450 elsif (ref $attr eq 'ARRAY') {
2458 # a resolved join is expressed as a multi-level from
2460 $extra_checks{-join}
2462 ref $attrs->{from} eq 'ARRAY'
2464 @{$attrs->{from}} > 1
2472 # Recursively collapse the condition.
2474 sub _collapse_cond {
2475 my ($self, $cond, $collapsed) = @_;
2479 if (ref $cond eq 'ARRAY') {
2480 foreach my $subcond (@$cond) {
2481 next unless ref $subcond; # -or
2482 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2485 elsif (ref $cond eq 'HASH') {
2486 if (keys %$cond and (keys %$cond)[0] eq '-and') {
2487 foreach my $subcond (@{$cond->{-and}}) {
2488 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2492 foreach my $col (keys %$cond) {
2493 my $value = $cond->{$col};
2494 $collapsed->{$col} = $value;
2504 # Remove the specified alias from the specified query hash. A copy is made so
2505 # the original query is not modified.
2508 my ($self, $query, $alias) = @_;
2510 my %orig = %{ $query || {} };
2513 foreach my $key (keys %orig) {
2515 $unaliased{$key} = $orig{$key};
2518 $unaliased{$1} = $orig{$key}
2519 if $key =~ m/^(?:\Q$alias\E\.)?([^.]+)$/;
2529 =item Arguments: none
2531 =item Return Value: \[ $sql, @bind ]
2535 Returns the SQL query and bind vars associated with the invocant.
2537 This is generally used as the RHS for a subquery.
2544 my $attrs = $self->_resolved_attrs_copy;
2549 # my ($sql, \@bind, \%dbi_bind_attrs) = _select_args_to_query (...)
2550 # $sql also has no wrapping parenthesis in list ctx
2552 my $sqlbind = $self->result_source->storage
2553 ->_select_args_to_query ($attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs);
2562 =item Arguments: \%vals, \%attrs?
2564 =item Return Value: $rowobject
2568 my $artist = $schema->resultset('Artist')->find_or_new(
2569 { artist => 'fred' }, { key => 'artists' });
2571 $cd->cd_to_producer->find_or_new({ producer => $producer },
2572 { key => 'primary });
2574 Find an existing record from this resultset using L</find>. if none exists,
2575 instantiate a new result object and return it. The object will not be saved
2576 into your storage until you call L<DBIx::Class::Row/insert> on it.
2578 You most likely want this method when looking for existing rows using a unique
2579 constraint that is not the primary key, or looking for related rows.
2581 If you want objects to be saved immediately, use L</find_or_create> instead.
2583 B<Note>: Make sure to read the documentation of L</find> and understand the
2584 significance of the C<key> attribute, as its lack may skew your search, and
2585 subsequently result in spurious new objects.
2587 B<Note>: Take care when using C<find_or_new> with a table having
2588 columns with default values that you intend to be automatically
2589 supplied by the database (e.g. an auto_increment primary key column).
2590 In normal usage, the value of such columns should NOT be included at
2591 all in the call to C<find_or_new>, even when set to C<undef>.
2597 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2598 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2599 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2602 return $self->new_result($hash);
2609 =item Arguments: \%vals
2611 =item Return Value: a L<DBIx::Class::Row> $object
2615 Attempt to create a single new row or a row with multiple related rows
2616 in the table represented by the resultset (and related tables). This
2617 will not check for duplicate rows before inserting, use
2618 L</find_or_create> to do that.
2620 To create one row for this resultset, pass a hashref of key/value
2621 pairs representing the columns of the table and the values you wish to
2622 store. If the appropriate relationships are set up, foreign key fields
2623 can also be passed an object representing the foreign row, and the
2624 value will be set to its primary key.
2626 To create related objects, pass a hashref of related-object column values
2627 B<keyed on the relationship name>. If the relationship is of type C<multi>
2628 (L<DBIx::Class::Relationship/has_many>) - pass an arrayref of hashrefs.
2629 The process will correctly identify columns holding foreign keys, and will
2630 transparently populate them from the keys of the corresponding relation.
2631 This can be applied recursively, and will work correctly for a structure
2632 with an arbitrary depth and width, as long as the relationships actually
2633 exists and the correct column data has been supplied.
2636 Instead of hashrefs of plain related data (key/value pairs), you may
2637 also pass new or inserted objects. New objects (not inserted yet, see
2638 L</new>), will be inserted into their appropriate tables.
2640 Effectively a shortcut for C<< ->new_result(\%vals)->insert >>.
2642 Example of creating a new row.
2644 $person_rs->create({
2645 name=>"Some Person",
2646 email=>"somebody@someplace.com"
2649 Example of creating a new row and also creating rows in a related C<has_many>
2650 or C<has_one> resultset. Note Arrayref.
2653 { artistid => 4, name => 'Manufactured Crap', cds => [
2654 { title => 'My First CD', year => 2006 },
2655 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2660 Example of creating a new row and also creating a row in a related
2661 C<belongs_to> resultset. Note Hashref.
2664 title=>"Music for Silly Walks",
2667 name=>"Silly Musician",
2675 When subclassing ResultSet never attempt to override this method. Since
2676 it is a simple shortcut for C<< $self->new_result($attrs)->insert >>, a
2677 lot of the internals simply never call it, so your override will be
2678 bypassed more often than not. Override either L<new|DBIx::Class::Row/new>
2679 or L<insert|DBIx::Class::Row/insert> depending on how early in the
2680 L</create> process you need to intervene.
2687 my ($self, $attrs) = @_;
2688 $self->throw_exception( "create needs a hashref" )
2689 unless ref $attrs eq 'HASH';
2690 return $self->new_result($attrs)->insert;
2693 =head2 find_or_create
2697 =item Arguments: \%vals, \%attrs?
2699 =item Return Value: $rowobject
2703 $cd->cd_to_producer->find_or_create({ producer => $producer },
2704 { key => 'primary' });
2706 Tries to find a record based on its primary key or unique constraints; if none
2707 is found, creates one and returns that instead.
2709 my $cd = $schema->resultset('CD')->find_or_create({
2711 artist => 'Massive Attack',
2712 title => 'Mezzanine',
2716 Also takes an optional C<key> attribute, to search by a specific key or unique
2717 constraint. For example:
2719 my $cd = $schema->resultset('CD')->find_or_create(
2721 artist => 'Massive Attack',
2722 title => 'Mezzanine',
2724 { key => 'cd_artist_title' }
2727 B<Note>: Make sure to read the documentation of L</find> and understand the
2728 significance of the C<key> attribute, as its lack may skew your search, and
2729 subsequently result in spurious row creation.
2731 B<Note>: Because find_or_create() reads from the database and then
2732 possibly inserts based on the result, this method is subject to a race
2733 condition. Another process could create a record in the table after
2734 the find has completed and before the create has started. To avoid
2735 this problem, use find_or_create() inside a transaction.
2737 B<Note>: Take care when using C<find_or_create> with a table having
2738 columns with default values that you intend to be automatically
2739 supplied by the database (e.g. an auto_increment primary key column).
2740 In normal usage, the value of such columns should NOT be included at
2741 all in the call to C<find_or_create>, even when set to C<undef>.
2743 See also L</find> and L</update_or_create>. For information on how to declare
2744 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2748 sub find_or_create {
2750 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2751 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2752 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2755 return $self->create($hash);
2758 =head2 update_or_create
2762 =item Arguments: \%col_values, { key => $unique_constraint }?
2764 =item Return Value: $row_object
2768 $resultset->update_or_create({ col => $val, ... });
2770 Like L</find_or_create>, but if a row is found it is immediately updated via
2771 C<< $found_row->update (\%col_values) >>.
2774 Takes an optional C<key> attribute to search on a specific unique constraint.
2777 # In your application
2778 my $cd = $schema->resultset('CD')->update_or_create(
2780 artist => 'Massive Attack',
2781 title => 'Mezzanine',
2784 { key => 'cd_artist_title' }
2787 $cd->cd_to_producer->update_or_create({
2788 producer => $producer,
2794 B<Note>: Make sure to read the documentation of L</find> and understand the
2795 significance of the C<key> attribute, as its lack may skew your search, and
2796 subsequently result in spurious row creation.
2798 B<Note>: Take care when using C<update_or_create> with a table having
2799 columns with default values that you intend to be automatically
2800 supplied by the database (e.g. an auto_increment primary key column).
2801 In normal usage, the value of such columns should NOT be included at
2802 all in the call to C<update_or_create>, even when set to C<undef>.
2804 See also L</find> and L</find_or_create>. For information on how to declare
2805 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2809 sub update_or_create {
2811 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2812 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2814 my $row = $self->find($cond, $attrs);
2816 $row->update($cond);
2820 return $self->create($cond);
2823 =head2 update_or_new
2827 =item Arguments: \%col_values, { key => $unique_constraint }?
2829 =item Return Value: $rowobject
2833 $resultset->update_or_new({ col => $val, ... });
2835 Like L</find_or_new> but if a row is found it is immediately updated via
2836 C<< $found_row->update (\%col_values) >>.
2840 # In your application
2841 my $cd = $schema->resultset('CD')->update_or_new(
2843 artist => 'Massive Attack',
2844 title => 'Mezzanine',
2847 { key => 'cd_artist_title' }
2850 if ($cd->in_storage) {
2851 # the cd was updated
2854 # the cd is not yet in the database, let's insert it
2858 B<Note>: Make sure to read the documentation of L</find> and understand the
2859 significance of the C<key> attribute, as its lack may skew your search, and
2860 subsequently result in spurious new objects.
2862 B<Note>: Take care when using C<update_or_new> with a table having
2863 columns with default values that you intend to be automatically
2864 supplied by the database (e.g. an auto_increment primary key column).
2865 In normal usage, the value of such columns should NOT be included at
2866 all in the call to C<update_or_new>, even when set to C<undef>.
2868 See also L</find>, L</find_or_create> and L</find_or_new>.
2874 my $attrs = ( @_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {} );
2875 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2877 my $row = $self->find( $cond, $attrs );
2878 if ( defined $row ) {
2879 $row->update($cond);
2883 return $self->new_result($cond);
2890 =item Arguments: none
2892 =item Return Value: \@cache_objects | undef
2896 Gets the contents of the cache for the resultset, if the cache is set.
2898 The cache is populated either by using the L</prefetch> attribute to
2899 L</search> or by calling L</set_cache>.
2911 =item Arguments: \@cache_objects
2913 =item Return Value: \@cache_objects
2917 Sets the contents of the cache for the resultset. Expects an arrayref
2918 of objects of the same class as those produced by the resultset. Note that
2919 if the cache is set the resultset will return the cached objects rather
2920 than re-querying the database even if the cache attr is not set.
2922 The contents of the cache can also be populated by using the
2923 L</prefetch> attribute to L</search>.
2928 my ( $self, $data ) = @_;
2929 $self->throw_exception("set_cache requires an arrayref")
2930 if defined($data) && (ref $data ne 'ARRAY');
2931 $self->{all_cache} = $data;
2938 =item Arguments: none
2940 =item Return Value: undef
2944 Clears the cache for the resultset.
2949 shift->set_cache(undef);
2956 =item Arguments: none
2958 =item Return Value: true, if the resultset has been paginated
2966 return !!$self->{attrs}{page};
2973 =item Arguments: none
2975 =item Return Value: true, if the resultset has been ordered with C<order_by>.
2983 return scalar $self->result_source->storage->_extract_order_criteria($self->{attrs}{order_by});
2986 =head2 related_resultset
2990 =item Arguments: $relationship_name
2992 =item Return Value: $resultset
2996 Returns a related resultset for the supplied relationship name.
2998 $artist_rs = $schema->resultset('CD')->related_resultset('Artist');
3002 sub related_resultset {
3003 my ($self, $rel) = @_;
3005 $self->{related_resultsets} ||= {};
3006 return $self->{related_resultsets}{$rel} ||= do {
3007 my $rsrc = $self->result_source;
3008 my $rel_info = $rsrc->relationship_info($rel);
3010 $self->throw_exception(
3011 "search_related: result source '" . $rsrc->source_name .
3012 "' has no such relationship $rel")
3015 my $attrs = $self->_chain_relationship($rel);
3017 my $join_count = $attrs->{seen_join}{$rel};
3019 my $alias = $self->result_source->storage
3020 ->relname_to_table_alias($rel, $join_count);
3022 # since this is search_related, and we already slid the select window inwards
3023 # (the select/as attrs were deleted in the beginning), we need to flip all
3024 # left joins to inner, so we get the expected results
3025 # read the comment on top of the actual function to see what this does
3026 $attrs->{from} = $rsrc->schema->storage->_inner_join_to_node ($attrs->{from}, $alias);
3029 #XXX - temp fix for result_class bug. There likely is a more elegant fix -groditi
3030 delete @{$attrs}{qw(result_class alias)};
3034 if (my $cache = $self->get_cache) {
3035 if ($cache->[0] && $cache->[0]->related_resultset($rel)->get_cache) {
3036 $new_cache = [ map { @{$_->related_resultset($rel)->get_cache} }
3041 my $rel_source = $rsrc->related_source($rel);
3045 # The reason we do this now instead of passing the alias to the
3046 # search_rs below is that if you wrap/overload resultset on the
3047 # source you need to know what alias it's -going- to have for things
3048 # to work sanely (e.g. RestrictWithObject wants to be able to add
3049 # extra query restrictions, and these may need to be $alias.)
3051 my $rel_attrs = $rel_source->resultset_attributes;
3052 local $rel_attrs->{alias} = $alias;
3054 $rel_source->resultset
3058 where => $attrs->{where},
3061 $new->set_cache($new_cache) if $new_cache;
3066 =head2 current_source_alias
3070 =item Arguments: none
3072 =item Return Value: $source_alias
3076 Returns the current table alias for the result source this resultset is built
3077 on, that will be used in the SQL query. Usually it is C<me>.
3079 Currently the source alias that refers to the result set returned by a
3080 L</search>/L</find> family method depends on how you got to the resultset: it's
3081 C<me> by default, but eg. L</search_related> aliases it to the related result
3082 source name (and keeps C<me> referring to the original result set). The long
3083 term goal is to make L<DBIx::Class> always alias the current resultset as C<me>
3084 (and make this method unnecessary).
3086 Thus it's currently necessary to use this method in predefined queries (see
3087 L<DBIx::Class::Manual::Cookbook/Predefined searches>) when referring to the
3088 source alias of the current result set:
3090 # in a result set class
3092 my ($self, $user) = @_;
3094 my $me = $self->current_source_alias;
3096 return $self->search(
3097 "$me.modified" => $user->id,
3103 sub current_source_alias {
3106 return ($self->{attrs} || {})->{alias} || 'me';
3109 =head2 as_subselect_rs
3113 =item Arguments: none
3115 =item Return Value: $resultset
3119 Act as a barrier to SQL symbols. The resultset provided will be made into a
3120 "virtual view" by including it as a subquery within the from clause. From this
3121 point on, any joined tables are inaccessible to ->search on the resultset (as if
3122 it were simply where-filtered without joins). For example:
3124 my $rs = $schema->resultset('Bar')->search({'x.name' => 'abc'},{ join => 'x' });
3126 # 'x' now pollutes the query namespace
3128 # So the following works as expected
3129 my $ok_rs = $rs->search({'x.other' => 1});
3131 # But this doesn't: instead of finding a 'Bar' related to two x rows (abc and
3132 # def) we look for one row with contradictory terms and join in another table
3133 # (aliased 'x_2') which we never use
3134 my $broken_rs = $rs->search({'x.name' => 'def'});
3136 my $rs2 = $rs->as_subselect_rs;
3138 # doesn't work - 'x' is no longer accessible in $rs2, having been sealed away
3139 my $not_joined_rs = $rs2->search({'x.other' => 1});
3141 # works as expected: finds a 'table' row related to two x rows (abc and def)
3142 my $correctly_joined_rs = $rs2->search({'x.name' => 'def'});
3144 Another example of when one might use this would be to select a subset of
3145 columns in a group by clause:
3147 my $rs = $schema->resultset('Bar')->search(undef, {
3148 group_by => [qw{ id foo_id baz_id }],
3149 })->as_subselect_rs->search(undef, {
3150 columns => [qw{ id foo_id }]
3153 In the above example normally columns would have to be equal to the group by,
3154 but because we isolated the group by into a subselect the above works.
3158 sub as_subselect_rs {
3161 my $attrs = $self->_resolved_attrs;
3163 my $fresh_rs = (ref $self)->new (
3164 $self->result_source
3167 # these pieces will be locked in the subquery
3168 delete $fresh_rs->{cond};
3169 delete @{$fresh_rs->{attrs}}{qw/where bind/};
3171 return $fresh_rs->search( {}, {
3173 $attrs->{alias} => $self->as_query,
3174 -alias => $attrs->{alias},
3175 -rsrc => $self->result_source,
3177 alias => $attrs->{alias},
3181 # This code is called by search_related, and makes sure there
3182 # is clear separation between the joins before, during, and
3183 # after the relationship. This information is needed later
3184 # in order to properly resolve prefetch aliases (any alias
3185 # with a relation_chain_depth less than the depth of the
3186 # current prefetch is not considered)
3188 # The increments happen twice per join. An even number means a
3189 # relationship specified via a search_related, whereas an odd
3190 # number indicates a join/prefetch added via attributes
3192 # Also this code will wrap the current resultset (the one we
3193 # chain to) in a subselect IFF it contains limiting attributes
3194 sub _chain_relationship {
3195 my ($self, $rel) = @_;
3196 my $source = $self->result_source;
3197 my $attrs = { %{$self->{attrs}||{}} };
3199 # we need to take the prefetch the attrs into account before we
3200 # ->_resolve_join as otherwise they get lost - captainL
3201 my $join = $self->_merge_joinpref_attr( $attrs->{join}, $attrs->{prefetch} );
3203 delete @{$attrs}{qw/join prefetch collapse group_by distinct select as columns +select +as +columns/};
3205 my $seen = { %{ (delete $attrs->{seen_join}) || {} } };
3208 my @force_subq_attrs = qw/offset rows group_by having/;
3211 ($attrs->{from} && ref $attrs->{from} ne 'ARRAY')
3213 $self->_has_resolved_attr (@force_subq_attrs)
3215 # Nuke the prefetch (if any) before the new $rs attrs
3216 # are resolved (prefetch is useless - we are wrapping
3217 # a subquery anyway).
3218 my $rs_copy = $self->search;
3219 $rs_copy->{attrs}{join} = $self->_merge_joinpref_attr (
3220 $rs_copy->{attrs}{join},
3221 delete $rs_copy->{attrs}{prefetch},
3226 -alias => $attrs->{alias},
3227 $attrs->{alias} => $rs_copy->as_query,
3229 delete @{$attrs}{@force_subq_attrs, qw/where bind/};
3230 $seen->{-relation_chain_depth} = 0;
3232 elsif ($attrs->{from}) { #shallow copy suffices
3233 $from = [ @{$attrs->{from}} ];
3238 -alias => $attrs->{alias},
3239 $attrs->{alias} => $source->from,
3243 my $jpath = ($seen->{-relation_chain_depth})
3244 ? $from->[-1][0]{-join_path}
3247 my @requested_joins = $source->_resolve_join(
3254 push @$from, @requested_joins;
3256 $seen->{-relation_chain_depth}++;
3258 # if $self already had a join/prefetch specified on it, the requested
3259 # $rel might very well be already included. What we do in this case
3260 # is effectively a no-op (except that we bump up the chain_depth on
3261 # the join in question so we could tell it *is* the search_related)
3264 # we consider the last one thus reverse
3265 for my $j (reverse @requested_joins) {
3266 my ($last_j) = keys %{$j->[0]{-join_path}[-1]};
3267 if ($rel eq $last_j) {
3268 $j->[0]{-relation_chain_depth}++;
3274 unless ($already_joined) {
3275 push @$from, $source->_resolve_join(
3283 $seen->{-relation_chain_depth}++;
3285 return {%$attrs, from => $from, seen_join => $seen};
3288 # too many times we have to do $attrs = { %{$self->_resolved_attrs} }
3289 sub _resolved_attrs_copy {
3291 return { %{$self->_resolved_attrs (@_)} };
3294 sub _resolved_attrs {
3296 return $self->{_attrs} if $self->{_attrs};
3298 my $attrs = { %{ $self->{attrs} || {} } };
3299 my $source = $self->result_source;
3300 my $alias = $attrs->{alias};
3302 # default selection list
3303 $attrs->{columns} = [ $source->columns ]
3304 unless List::Util::first { exists $attrs->{$_} } qw/columns cols select as/;
3306 # merge selectors together
3307 for (qw/columns select as/) {
3308 $attrs->{$_} = $self->_merge_attr($attrs->{$_}, delete $attrs->{"+$_"})
3309 if $attrs->{$_} or $attrs->{"+$_"};
3312 # disassemble columns
3314 if (my $cols = delete $attrs->{columns}) {
3315 for my $c (ref $cols eq 'ARRAY' ? @$cols : $cols) {
3316 if (ref $c eq 'HASH') {
3317 for my $as (keys %$c) {
3318 push @sel, $c->{$as};
3329 # when trying to weed off duplicates later do not go past this point -
3330 # everything added from here on is unbalanced "anyone's guess" stuff
3331 my $dedup_stop_idx = $#as;
3333 push @as, @{ ref $attrs->{as} eq 'ARRAY' ? $attrs->{as} : [ $attrs->{as} ] }
3335 push @sel, @{ ref $attrs->{select} eq 'ARRAY' ? $attrs->{select} : [ $attrs->{select} ] }
3336 if $attrs->{select};
3338 # assume all unqualified selectors to apply to the current alias (legacy stuff)
3340 $_ = (ref $_ or $_ =~ /\./) ? $_ : "$alias.$_";
3343 # disqualify all $alias.col as-bits (collapser mandated)
3345 $_ = ($_ =~ /^\Q$alias.\E(.+)$/) ? $1 : $_;
3348 # de-duplicate the result (remove *identical* select/as pairs)
3349 # and also die on duplicate {as} pointing to different {select}s
3350 # not using a c-style for as the condition is prone to shrinkage
3353 while ($i <= $dedup_stop_idx) {
3354 if ($seen->{"$sel[$i] \x00\x00 $as[$i]"}++) {
3359 elsif ($seen->{$as[$i]}++) {
3360 $self->throw_exception(
3361 "inflate_result() alias '$as[$i]' specified twice with different SQL-side {select}-ors"
3369 $attrs->{select} = \@sel;
3370 $attrs->{as} = \@as;
3372 $attrs->{from} ||= [{
3374 -alias => $self->{attrs}{alias},
3375 $self->{attrs}{alias} => $source->from,
3378 if ( $attrs->{join} || $attrs->{prefetch} ) {
3380 $self->throw_exception ('join/prefetch can not be used with a custom {from}')
3381 if ref $attrs->{from} ne 'ARRAY';
3383 my $join = (delete $attrs->{join}) || {};
3385 if ( defined $attrs->{prefetch} ) {
3386 $join = $self->_merge_joinpref_attr( $join, $attrs->{prefetch} );
3389 $attrs->{from} = # have to copy here to avoid corrupting the original
3391 @{ $attrs->{from} },
3392 $source->_resolve_join(
3395 { %{ $attrs->{seen_join} || {} } },
3396 ( $attrs->{seen_join} && keys %{$attrs->{seen_join}})
3397 ? $attrs->{from}[-1][0]{-join_path}
3404 if ( defined $attrs->{order_by} ) {
3405 $attrs->{order_by} = (
3406 ref( $attrs->{order_by} ) eq 'ARRAY'
3407 ? [ @{ $attrs->{order_by} } ]
3408 : [ $attrs->{order_by} || () ]
3412 if ($attrs->{group_by} and ref $attrs->{group_by} ne 'ARRAY') {
3413 $attrs->{group_by} = [ $attrs->{group_by} ];
3416 # generate the distinct induced group_by early, as prefetch will be carried via a
3417 # subquery (since a group_by is present)
3418 if (delete $attrs->{distinct}) {
3419 if ($attrs->{group_by}) {
3420 carp_unique ("Useless use of distinct on a grouped resultset ('distinct' is ignored when a 'group_by' is present)");
3423 # distinct affects only the main selection part, not what prefetch may
3425 $attrs->{group_by} = $source->storage->_group_over_selection (
3433 $attrs->{collapse} ||= {};
3434 if ($attrs->{prefetch}) {
3436 $self->throw_exception("Unable to prefetch, resultset contains an unnamed selector $attrs->{_dark_selector}{string}")
3437 if $attrs->{_dark_selector};
3439 my $prefetch = $self->_merge_joinpref_attr( {}, delete $attrs->{prefetch} );
3441 my $prefetch_ordering = [];
3443 # this is a separate structure (we don't look in {from} directly)
3444 # as the resolver needs to shift things off the lists to work
3445 # properly (identical-prefetches on different branches)
3447 if (ref $attrs->{from} eq 'ARRAY') {
3449 my $start_depth = $attrs->{seen_join}{-relation_chain_depth} || 0;
3451 for my $j ( @{$attrs->{from}}[1 .. $#{$attrs->{from}} ] ) {
3452 next unless $j->[0]{-alias};
3453 next unless $j->[0]{-join_path};
3454 next if ($j->[0]{-relation_chain_depth} || 0) < $start_depth;
3456 my @jpath = map { keys %$_ } @{$j->[0]{-join_path}};
3459 $p = $p->{$_} ||= {} for @jpath[ ($start_depth/2) .. $#jpath]; #only even depths are actual jpath boundaries
3460 push @{$p->{-join_aliases} }, $j->[0]{-alias};
3465 $source->_resolve_prefetch( $prefetch, $alias, $join_map, $prefetch_ordering, $attrs->{collapse} );
3467 # we need to somehow mark which columns came from prefetch
3469 my $sel_end = $#{$attrs->{select}};
3470 $attrs->{_prefetch_selector_range} = [ $sel_end + 1, $sel_end + @prefetch ];
3473 push @{ $attrs->{select} }, (map { $_->[0] } @prefetch);
3474 push @{ $attrs->{as} }, (map { $_->[1] } @prefetch);
3476 push( @{$attrs->{order_by}}, @$prefetch_ordering );
3477 $attrs->{_collapse_order_by} = \@$prefetch_ordering;
3481 # if both page and offset are specified, produce a combined offset
3482 # even though it doesn't make much sense, this is what pre 081xx has
3484 if (my $page = delete $attrs->{page}) {
3486 ($attrs->{rows} * ($page - 1))
3488 ($attrs->{offset} || 0)
3492 return $self->{_attrs} = $attrs;
3496 my ($self, $attr) = @_;
3498 if (ref $attr eq 'HASH') {
3499 return $self->_rollout_hash($attr);
3500 } elsif (ref $attr eq 'ARRAY') {
3501 return $self->_rollout_array($attr);
3507 sub _rollout_array {
3508 my ($self, $attr) = @_;
3511 foreach my $element (@{$attr}) {
3512 if (ref $element eq 'HASH') {
3513 push( @rolled_array, @{ $self->_rollout_hash( $element ) } );
3514 } elsif (ref $element eq 'ARRAY') {
3515 # XXX - should probably recurse here
3516 push( @rolled_array, @{$self->_rollout_array($element)} );
3518 push( @rolled_array, $element );
3521 return \@rolled_array;
3525 my ($self, $attr) = @_;
3528 foreach my $key (keys %{$attr}) {
3529 push( @rolled_array, { $key => $attr->{$key} } );
3531 return \@rolled_array;
3534 sub _calculate_score {
3535 my ($self, $a, $b) = @_;
3537 if (defined $a xor defined $b) {
3540 elsif (not defined $a) {
3544 if (ref $b eq 'HASH') {
3545 my ($b_key) = keys %{$b};
3546 if (ref $a eq 'HASH') {
3547 my ($a_key) = keys %{$a};
3548 if ($a_key eq $b_key) {
3549 return (1 + $self->_calculate_score( $a->{$a_key}, $b->{$b_key} ));
3554 return ($a eq $b_key) ? 1 : 0;
3557 if (ref $a eq 'HASH') {
3558 my ($a_key) = keys %{$a};
3559 return ($b eq $a_key) ? 1 : 0;
3561 return ($b eq $a) ? 1 : 0;
3566 sub _merge_joinpref_attr {
3567 my ($self, $orig, $import) = @_;
3569 return $import unless defined($orig);
3570 return $orig unless defined($import);
3572 $orig = $self->_rollout_attr($orig);
3573 $import = $self->_rollout_attr($import);
3576 foreach my $import_element ( @{$import} ) {
3577 # find best candidate from $orig to merge $b_element into
3578 my $best_candidate = { position => undef, score => 0 }; my $position = 0;
3579 foreach my $orig_element ( @{$orig} ) {
3580 my $score = $self->_calculate_score( $orig_element, $import_element );
3581 if ($score > $best_candidate->{score}) {
3582 $best_candidate->{position} = $position;
3583 $best_candidate->{score} = $score;
3587 my ($import_key) = ( ref $import_element eq 'HASH' ) ? keys %{$import_element} : ($import_element);
3589 if ($best_candidate->{score} == 0 || exists $seen_keys->{$import_key}) {
3590 push( @{$orig}, $import_element );
3592 my $orig_best = $orig->[$best_candidate->{position}];
3593 # merge orig_best and b_element together and replace original with merged
3594 if (ref $orig_best ne 'HASH') {
3595 $orig->[$best_candidate->{position}] = $import_element;
3596 } elsif (ref $import_element eq 'HASH') {
3597 my ($key) = keys %{$orig_best};
3598 $orig->[$best_candidate->{position}] = { $key => $self->_merge_joinpref_attr($orig_best->{$key}, $import_element->{$key}) };
3601 $seen_keys->{$import_key} = 1; # don't merge the same key twice
3612 require Hash::Merge;
3613 my $hm = Hash::Merge->new;
3615 $hm->specify_behavior({
3618 my ($defl, $defr) = map { defined $_ } (@_[0,1]);
3620 if ($defl xor $defr) {
3621 return [ $defl ? $_[0] : $_[1] ];
3626 elsif (__HM_DEDUP and $_[0] eq $_[1]) {
3630 return [$_[0], $_[1]];
3634 return $_[1] if !defined $_[0];
3635 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3636 return [$_[0], @{$_[1]}]
3639 return [] if !defined $_[0] and !keys %{$_[1]};
3640 return [ $_[1] ] if !defined $_[0];
3641 return [ $_[0] ] if !keys %{$_[1]};
3642 return [$_[0], $_[1]]
3647 return $_[0] if !defined $_[1];
3648 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3649 return [@{$_[0]}, $_[1]]
3652 my @ret = @{$_[0]} or return $_[1];
3653 return [ @ret, @{$_[1]} ] unless __HM_DEDUP;
3654 my %idx = map { $_ => 1 } @ret;
3655 push @ret, grep { ! defined $idx{$_} } (@{$_[1]});
3659 return [ $_[1] ] if ! @{$_[0]};
3660 return $_[0] if !keys %{$_[1]};
3661 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3662 return [ @{$_[0]}, $_[1] ];
3667 return [] if !keys %{$_[0]} and !defined $_[1];
3668 return [ $_[0] ] if !defined $_[1];
3669 return [ $_[1] ] if !keys %{$_[0]};
3670 return [$_[0], $_[1]]
3673 return [] if !keys %{$_[0]} and !@{$_[1]};
3674 return [ $_[0] ] if !@{$_[1]};
3675 return $_[1] if !keys %{$_[0]};
3676 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3677 return [ $_[0], @{$_[1]} ];
3680 return [] if !keys %{$_[0]} and !keys %{$_[1]};
3681 return [ $_[0] ] if !keys %{$_[1]};
3682 return [ $_[1] ] if !keys %{$_[0]};
3683 return [ $_[0] ] if $_[0] eq $_[1];
3684 return [ $_[0], $_[1] ];
3687 } => 'DBIC_RS_ATTR_MERGER');
3691 return $hm->merge ($_[1], $_[2]);
3695 sub STORABLE_freeze {
3696 my ($self, $cloning) = @_;
3697 my $to_serialize = { %$self };
3699 # A cursor in progress can't be serialized (and would make little sense anyway)
3700 delete $to_serialize->{cursor};
3702 Storable::nfreeze($to_serialize);
3705 # need this hook for symmetry
3707 my ($self, $cloning, $serialized) = @_;
3709 %$self = %{ Storable::thaw($serialized) };
3715 =head2 throw_exception
3717 See L<DBIx::Class::Schema/throw_exception> for details.
3721 sub throw_exception {
3724 if (ref $self and my $rsrc = $self->result_source) {
3725 $rsrc->throw_exception(@_)
3728 DBIx::Class::Exception->throw(@_);
3732 # XXX: FIXME: Attributes docs need clearing up
3736 Attributes are used to refine a ResultSet in various ways when
3737 searching for data. They can be passed to any method which takes an
3738 C<\%attrs> argument. See L</search>, L</search_rs>, L</find>,
3741 These are in no particular order:
3747 =item Value: ( $order_by | \@order_by | \%order_by )
3751 Which column(s) to order the results by.
3753 [The full list of suitable values is documented in
3754 L<SQL::Abstract/"ORDER BY CLAUSES">; the following is a summary of
3757 If a single column name, or an arrayref of names is supplied, the
3758 argument is passed through directly to SQL. The hashref syntax allows
3759 for connection-agnostic specification of ordering direction:
3761 For descending order:
3763 order_by => { -desc => [qw/col1 col2 col3/] }
3765 For explicit ascending order:
3767 order_by => { -asc => 'col' }
3769 The old scalarref syntax (i.e. order_by => \'year DESC') is still
3770 supported, although you are strongly encouraged to use the hashref
3771 syntax as outlined above.
3777 =item Value: \@columns
3781 Shortcut to request a particular set of columns to be retrieved. Each
3782 column spec may be a string (a table column name), or a hash (in which
3783 case the key is the C<as> value, and the value is used as the C<select>
3784 expression). Adds C<me.> onto the start of any column without a C<.> in
3785 it and sets C<select> from that, then auto-populates C<as> from
3786 C<select> as normal. (You may also use the C<cols> attribute, as in
3787 earlier versions of DBIC.)
3789 Essentially C<columns> does the same as L</select> and L</as>.
3791 columns => [ 'foo', { bar => 'baz' } ]
3795 select => [qw/foo baz/],
3802 =item Value: \@columns
3806 Indicates additional columns to be selected from storage. Works the same
3807 as L</columns> but adds columns to the selection. (You may also use the
3808 C<include_columns> attribute, as in earlier versions of DBIC). For
3811 $schema->resultset('CD')->search(undef, {
3812 '+columns' => ['artist.name'],
3816 would return all CDs and include a 'name' column to the information
3817 passed to object inflation. Note that the 'artist' is the name of the
3818 column (or relationship) accessor, and 'name' is the name of the column
3819 accessor in the related table.
3821 B<NOTE:> You need to explicitly quote '+columns' when defining the attribute.
3822 Not doing so causes Perl to incorrectly interpret +columns as a bareword with a
3823 unary plus operator before it.
3825 =head2 include_columns
3829 =item Value: \@columns
3833 Deprecated. Acts as a synonym for L</+columns> for backward compatibility.
3839 =item Value: \@select_columns
3843 Indicates which columns should be selected from the storage. You can use
3844 column names, or in the case of RDBMS back ends, function or stored procedure
3847 $rs = $schema->resultset('Employee')->search(undef, {
3850 { count => 'employeeid' },
3851 { max => { length => 'name' }, -as => 'longest_name' }
3856 SELECT name, COUNT( employeeid ), MAX( LENGTH( name ) ) AS longest_name FROM employee
3858 B<NOTE:> You will almost always need a corresponding L</as> attribute when you
3859 use L</select>, to instruct DBIx::Class how to store the result of the column.
3860 Also note that the L</as> attribute has nothing to do with the SQL-side 'AS'
3861 identifier aliasing. You can however alias a function, so you can use it in
3862 e.g. an C<ORDER BY> clause. This is done via the C<-as> B<select function
3863 attribute> supplied as shown in the example above.
3865 B<NOTE:> You need to explicitly quote '+select'/'+as' when defining the attributes.
3866 Not doing so causes Perl to incorrectly interpret them as a bareword with a
3867 unary plus operator before it.
3873 Indicates additional columns to be selected from storage. Works the same as
3874 L</select> but adds columns to the default selection, instead of specifying
3883 Indicates additional column names for those added via L</+select>. See L</as>.
3891 =item Value: \@inflation_names
3895 Indicates column names for object inflation. That is L</as> indicates the
3896 slot name in which the column value will be stored within the
3897 L<Row|DBIx::Class::Row> object. The value will then be accessible via this
3898 identifier by the C<get_column> method (or via the object accessor B<if one
3899 with the same name already exists>) as shown below. The L</as> attribute has
3900 B<nothing to do> with the SQL-side C<AS>. See L</select> for details.
3902 $rs = $schema->resultset('Employee')->search(undef, {
3905 { count => 'employeeid' },
3906 { max => { length => 'name' }, -as => 'longest_name' }
3915 If the object against which the search is performed already has an accessor
3916 matching a column name specified in C<as>, the value can be retrieved using
3917 the accessor as normal:
3919 my $name = $employee->name();
3921 If on the other hand an accessor does not exist in the object, you need to
3922 use C<get_column> instead:
3924 my $employee_count = $employee->get_column('employee_count');
3926 You can create your own accessors if required - see
3927 L<DBIx::Class::Manual::Cookbook> for details.
3933 =item Value: ($rel_name | \@rel_names | \%rel_names)
3937 Contains a list of relationships that should be joined for this query. For
3940 # Get CDs by Nine Inch Nails
3941 my $rs = $schema->resultset('CD')->search(
3942 { 'artist.name' => 'Nine Inch Nails' },
3943 { join => 'artist' }
3946 Can also contain a hash reference to refer to the other relation's relations.
3949 package MyApp::Schema::Track;
3950 use base qw/DBIx::Class/;
3951 __PACKAGE__->table('track');
3952 __PACKAGE__->add_columns(qw/trackid cd position title/);
3953 __PACKAGE__->set_primary_key('trackid');
3954 __PACKAGE__->belongs_to(cd => 'MyApp::Schema::CD');
3957 # In your application
3958 my $rs = $schema->resultset('Artist')->search(
3959 { 'track.title' => 'Teardrop' },
3961 join => { cd => 'track' },
3962 order_by => 'artist.name',
3966 You need to use the relationship (not the table) name in conditions,
3967 because they are aliased as such. The current table is aliased as "me", so
3968 you need to use me.column_name in order to avoid ambiguity. For example:
3970 # Get CDs from 1984 with a 'Foo' track
3971 my $rs = $schema->resultset('CD')->search(
3974 'tracks.name' => 'Foo'
3976 { join => 'tracks' }
3979 If the same join is supplied twice, it will be aliased to <rel>_2 (and
3980 similarly for a third time). For e.g.
3982 my $rs = $schema->resultset('Artist')->search({
3983 'cds.title' => 'Down to Earth',
3984 'cds_2.title' => 'Popular',
3986 join => [ qw/cds cds/ ],
3989 will return a set of all artists that have both a cd with title 'Down
3990 to Earth' and a cd with title 'Popular'.
3992 If you want to fetch related objects from other tables as well, see C<prefetch>
3995 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
4001 =item Value: ($rel_name | \@rel_names | \%rel_names)
4005 Contains one or more relationships that should be fetched along with
4006 the main query (when they are accessed afterwards the data will
4007 already be available, without extra queries to the database). This is
4008 useful for when you know you will need the related objects, because it
4009 saves at least one query:
4011 my $rs = $schema->resultset('Tag')->search(
4020 The initial search results in SQL like the following:
4022 SELECT tag.*, cd.*, artist.* FROM tag
4023 JOIN cd ON tag.cd = cd.cdid
4024 JOIN artist ON cd.artist = artist.artistid
4026 L<DBIx::Class> has no need to go back to the database when we access the
4027 C<cd> or C<artist> relationships, which saves us two SQL statements in this
4030 Simple prefetches will be joined automatically, so there is no need
4031 for a C<join> attribute in the above search.
4033 L</prefetch> can be used with the any of the relationship types and
4034 multiple prefetches can be specified together. Below is a more complex
4035 example that prefetches a CD's artist, its liner notes (if present),
4036 the cover image, the tracks on that cd, and the guests on those
4040 My::Schema::CD->belongs_to( artist => 'My::Schema::Artist' );
4041 My::Schema::CD->might_have( liner_note => 'My::Schema::LinerNotes' );
4042 My::Schema::CD->has_one( cover_image => 'My::Schema::Artwork' );
4043 My::Schema::CD->has_many( tracks => 'My::Schema::Track' );
4045 My::Schema::Artist->belongs_to( record_label => 'My::Schema::RecordLabel' );
4047 My::Schema::Track->has_many( guests => 'My::Schema::Guest' );
4050 my $rs = $schema->resultset('CD')->search(
4054 { artist => 'record_label'}, # belongs_to => belongs_to
4055 'liner_note', # might_have
4056 'cover_image', # has_one
4057 { tracks => 'guests' }, # has_many => has_many
4062 This will produce SQL like the following:
4064 SELECT cd.*, artist.*, record_label.*, liner_note.*, cover_image.*,
4068 ON artist.artistid = me.artistid
4069 JOIN record_label record_label
4070 ON record_label.labelid = artist.labelid
4071 LEFT JOIN track tracks
4072 ON tracks.cdid = me.cdid
4073 LEFT JOIN guest guests
4074 ON guests.trackid = track.trackid
4075 LEFT JOIN liner_notes liner_note
4076 ON liner_note.cdid = me.cdid
4077 JOIN cd_artwork cover_image
4078 ON cover_image.cdid = me.cdid
4081 Now the C<artist>, C<record_label>, C<liner_note>, C<cover_image>,
4082 C<tracks>, and C<guests> of the CD will all be available through the
4083 relationship accessors without the need for additional queries to the
4086 However, there is one caveat to be observed: it can be dangerous to
4087 prefetch more than one L<has_many|DBIx::Class::Relationship/has_many>
4088 relationship on a given level. e.g.:
4090 my $rs = $schema->resultset('CD')->search(
4094 'tracks', # has_many
4095 { cd_to_producer => 'producer' }, # has_many => belongs_to (i.e. m2m)
4100 In fact, C<DBIx::Class> will emit the following warning:
4102 Prefetching multiple has_many rels tracks and cd_to_producer at top
4103 level will explode the number of row objects retrievable via ->next
4104 or ->all. Use at your own risk.
4106 The collapser currently can't identify duplicate tuples for multiple
4107 L<has_many|DBIx::Class::Relationship/has_many> relationships and as a
4108 result the second L<has_many|DBIx::Class::Relationship/has_many>
4109 relation could contain redundant objects.
4111 =head3 Using L</prefetch> with L</join>
4113 L</prefetch> implies a L</join> with the equivalent argument, and is
4114 properly merged with any existing L</join> specification. So the
4117 my $rs = $schema->resultset('CD')->search(
4118 {'record_label.name' => 'Music Product Ltd.'},
4120 join => {artist => 'record_label'},
4121 prefetch => 'artist',
4125 ... will work, searching on the record label's name, but only
4126 prefetching the C<artist>.
4128 =head3 Using L</prefetch> with L</select> / L</+select> / L</as> / L</+as>
4130 L</prefetch> implies a L</+select>/L</+as> with the fields of the
4131 prefetched relations. So given:
4133 my $rs = $schema->resultset('CD')->search(
4136 select => ['cd.title'],
4138 prefetch => 'artist',
4142 The L</select> becomes: C<'cd.title', 'artist.*'> and the L</as>
4143 becomes: C<'cd_title', 'artist.*'>.
4147 Prefetch does a lot of deep magic. As such, it may not behave exactly
4148 as you might expect.
4154 Prefetch uses the L</cache> to populate the prefetched relationships. This
4155 may or may not be what you want.
4159 If you specify a condition on a prefetched relationship, ONLY those
4160 rows that match the prefetched condition will be fetched into that relationship.
4161 This means that adding prefetch to a search() B<may alter> what is returned by
4162 traversing a relationship. So, if you have C<< Artist->has_many(CDs) >> and you do
4164 my $artist_rs = $schema->resultset('Artist')->search({
4170 my $count = $artist_rs->first->cds->count;
4172 my $artist_rs_prefetch = $artist_rs->search( {}, { prefetch => 'cds' } );
4174 my $prefetch_count = $artist_rs_prefetch->first->cds->count;
4176 cmp_ok( $count, '==', $prefetch_count, "Counts should be the same" );
4178 that cmp_ok() may or may not pass depending on the datasets involved. This
4179 behavior may or may not survive the 0.09 transition.
4191 Makes the resultset paged and specifies the page to retrieve. Effectively
4192 identical to creating a non-pages resultset and then calling ->page($page)
4195 If L</rows> attribute is not specified it defaults to 10 rows per page.
4197 When you have a paged resultset, L</count> will only return the number
4198 of rows in the page. To get the total, use the L</pager> and call
4199 C<total_entries> on it.
4209 Specifies the maximum number of rows for direct retrieval or the number of
4210 rows per page if the page attribute or method is used.
4216 =item Value: $offset
4220 Specifies the (zero-based) row number for the first row to be returned, or the
4221 of the first row of the first page if paging is used.
4227 =item Value: \@columns
4231 A arrayref of columns to group by. Can include columns of joined tables.
4233 group_by => [qw/ column1 column2 ... /]
4239 =item Value: $condition
4243 HAVING is a select statement attribute that is applied between GROUP BY and
4244 ORDER BY. It is applied to the after the grouping calculations have been
4247 having => { 'count_employee' => { '>=', 100 } }
4249 or with an in-place function in which case literal SQL is required:
4251 having => \[ 'count(employee) >= ?', [ count => 100 ] ]
4257 =item Value: (0 | 1)
4261 Set to 1 to group by all columns. If the resultset already has a group_by
4262 attribute, this setting is ignored and an appropriate warning is issued.
4268 Adds to the WHERE clause.
4270 # only return rows WHERE deleted IS NULL for all searches
4271 __PACKAGE__->resultset_attributes({ where => { deleted => undef } }); )
4273 Can be overridden by passing C<< { where => undef } >> as an attribute
4280 Set to 1 to cache search results. This prevents extra SQL queries if you
4281 revisit rows in your ResultSet:
4283 my $resultset = $schema->resultset('Artist')->search( undef, { cache => 1 } );
4285 while( my $artist = $resultset->next ) {
4289 $rs->first; # without cache, this would issue a query
4291 By default, searches are not cached.
4293 For more examples of using these attributes, see
4294 L<DBIx::Class::Manual::Cookbook>.
4300 =item Value: ( 'update' | 'shared' )
4304 Set to 'update' for a SELECT ... FOR UPDATE or 'shared' for a SELECT