1 package DBIx::Class::ResultSet;
5 use base qw/DBIx::Class/;
7 use DBIx::Class::ResultSetColumn;
8 use DBIx::Class::ResultClass::HashRefInflator;
9 use Scalar::Util qw/blessed weaken reftype/;
10 use DBIx::Class::_Util qw(
12 fail_on_internal_wantarray fail_on_internal_call UNRESOLVABLE_CONDITION
16 # not importing first() as it will clash with our own method
20 # De-duplication in _merge_attr() is disabled, but left in for reference
21 # (the merger is used for other things that ought not to be de-duped)
22 *__HM_DEDUP = sub () { 0 };
32 # this is real - CDBICompat overrides it with insanity
33 # yes, prototype won't matter, but that's for now ;)
36 __PACKAGE__->mk_group_accessors('simple' => qw/_result_class result_source/);
40 DBIx::Class::ResultSet - Represents a query used for fetching a set of results.
44 my $users_rs = $schema->resultset('User');
45 while( $user = $users_rs->next) {
46 print $user->username;
49 my $registered_users_rs = $schema->resultset('User')->search({ registered => 1 });
50 my @cds_in_2005 = $schema->resultset('CD')->search({ year => 2005 })->all();
54 A ResultSet is an object which stores a set of conditions representing
55 a query. It is the backbone of DBIx::Class (i.e. the really
56 important/useful bit).
58 No SQL is executed on the database when a ResultSet is created, it
59 just stores all the conditions needed to create the query.
61 A basic ResultSet representing the data of an entire table is returned
62 by calling C<resultset> on a L<DBIx::Class::Schema> and passing in a
63 L<Source|DBIx::Class::Manual::Glossary/ResultSource> name.
65 my $users_rs = $schema->resultset('User');
67 A new ResultSet is returned from calling L</search> on an existing
68 ResultSet. The new one will contain all the conditions of the
69 original, plus any new conditions added in the C<search> call.
71 A ResultSet also incorporates an implicit iterator. L</next> and L</reset>
72 can be used to walk through all the L<DBIx::Class::Row>s the ResultSet
75 The query that the ResultSet represents is B<only> executed against
76 the database when these methods are called:
77 L</find>, L</next>, L</all>, L</first>, L</single>, L</count>.
79 If a resultset is used in a numeric context it returns the L</count>.
80 However, if it is used in a boolean context it is B<always> true. So if
81 you want to check if a resultset has any results, you must use C<if $rs
86 =head2 Chaining resultsets
88 Let's say you've got a query that needs to be run to return some data
89 to the user. But, you have an authorization system in place that
90 prevents certain users from seeing certain information. So, you want
91 to construct the basic query in one method, but add constraints to it in
96 my $request = $self->get_request; # Get a request object somehow.
97 my $schema = $self->result_source->schema;
99 my $cd_rs = $schema->resultset('CD')->search({
100 title => $request->param('title'),
101 year => $request->param('year'),
104 $cd_rs = $self->apply_security_policy( $cd_rs );
106 return $cd_rs->all();
109 sub apply_security_policy {
118 =head3 Resolving conditions and attributes
120 When a resultset is chained from another resultset (e.g.:
121 C<< my $new_rs = $old_rs->search(\%extra_cond, \%attrs) >>), conditions
122 and attributes with the same keys need resolving.
124 If any of L</columns>, L</select>, L</as> are present, they reset the
125 original selection, and start the selection "clean".
127 The L</join>, L</prefetch>, L</+columns>, L</+select>, L</+as> attributes
128 are merged into the existing ones from the original resultset.
130 The L</where> and L</having> attributes, and any search conditions, are
131 merged with an SQL C<AND> to the existing condition from the original
134 All other attributes are overridden by any new ones supplied in the
137 =head2 Multiple queries
139 Since a resultset just defines a query, you can do all sorts of
140 things with it with the same object.
142 # Don't hit the DB yet.
143 my $cd_rs = $schema->resultset('CD')->search({
144 title => 'something',
148 # Each of these hits the DB individually.
149 my $count = $cd_rs->count;
150 my $most_recent = $cd_rs->get_column('date_released')->max();
151 my @records = $cd_rs->all;
153 And it's not just limited to SELECT statements.
159 $cd_rs->create({ artist => 'Fred' });
161 Which is the same as:
163 $schema->resultset('CD')->create({
164 title => 'something',
169 See: L</search>, L</count>, L</get_column>, L</all>, L</create>.
171 =head2 Custom ResultSet classes
173 To add methods to your resultsets, you can subclass L<DBIx::Class::ResultSet>, similar to:
175 package MyApp::Schema::ResultSet::User;
180 use base 'DBIx::Class::ResultSet';
184 $self->search({ $self->current_source_alias . '.active' => 1 });
189 $self->search({ $self->current_source_alias . '.verified' => 0 });
192 sub created_n_days_ago {
193 my ($self, $days_ago) = @_;
195 $self->current_source_alias . '.create_date' => {
197 $self->result_source->schema->storage->datetime_parser->format_datetime(
198 DateTime->now( time_zone => 'UTC' )->subtract( days => $days_ago )
203 sub users_to_warn { shift->active->unverified->created_n_days_ago(7) }
207 See L<DBIx::Class::Schema/load_namespaces> on how DBIC can discover and
208 automatically attach L<Result|DBIx::Class::Manual::ResultClass>-specific
209 L<ResulSet|DBIx::Class::ResultSet> classes.
211 =head3 ResultSet subclassing with Moose and similar constructor-providers
213 Using L<Moose> or L<Moo> in your ResultSet classes is usually overkill, but
214 you may find it useful if your ResultSets contain a lot of business logic
215 (e.g. C<has xml_parser>, C<has json>, etc) or if you just prefer to organize
218 In order to write custom ResultSet classes with L<Moo> you need to use the
219 following template. The L<BUILDARGS|Moo/BUILDARGS> is necessary due to the
220 unusual signature of the L<constructor provided by DBIC
221 |DBIx::Class::ResultSet/new> C<< ->new($source, \%args) >>.
224 extends 'DBIx::Class::ResultSet';
225 sub BUILDARGS { $_[2] } # ::RS::new() expects my ($class, $rsrc, $args) = @_
231 If you want to build your custom ResultSet classes with L<Moose>, you need
232 a similar, though a little more elaborate template in order to interface the
233 inlining of the L<Moose>-provided
234 L<object constructor|Moose::Manual::Construction/WHERE'S THE CONSTRUCTOR?>,
237 package MyApp::Schema::ResultSet::User;
240 use MooseX::NonMoose;
241 extends 'DBIx::Class::ResultSet';
243 sub BUILDARGS { $_[2] } # ::RS::new() expects my ($class, $rsrc, $args) = @_
247 __PACKAGE__->meta->make_immutable;
251 The L<MooseX::NonMoose> is necessary so that the L<Moose> constructor does not
252 entirely overwrite the DBIC one (in contrast L<Moo> does this automatically).
253 Alternatively, you can skip L<MooseX::NonMoose> and get by with just L<Moose>
256 __PACKAGE__->meta->make_immutable(inline_constructor => 0);
264 =item Arguments: L<$source|DBIx::Class::ResultSource>, L<\%attrs?|/ATTRIBUTES>
266 =item Return Value: L<$resultset|/search>
270 The resultset constructor. Takes a source object (usually a
271 L<DBIx::Class::ResultSourceProxy::Table>) and an attribute hash (see
272 L</ATTRIBUTES> below). Does not perform any queries -- these are
273 executed as needed by the other methods.
275 Generally you never construct a resultset manually. Instead you get one
277 C<< $schema->L<resultset|DBIx::Class::Schema/resultset>('$source_name') >>
278 or C<< $another_resultset->L<search|/search>(...) >> (the later called in
281 my $rs = $schema->resultset('CD')->search({ title => '100th Window' });
287 If called on an object, proxies to L</new_result> instead, so
289 my $cd = $schema->resultset('CD')->new({ title => 'Spoon' });
291 will return a CD object, not a ResultSet, and is equivalent to:
293 my $cd = $schema->resultset('CD')->new_result({ title => 'Spoon' });
295 Please also keep in mind that many internals call L</new_result> directly,
296 so overloading this method with the idea of intercepting new result object
297 creation B<will not work>. See also warning pertaining to L</create>.
307 DBIx::Class::_ENV_::ASSERT_NO_INTERNAL_INDIRECT_CALLS and fail_on_internal_call;
308 return $class->new_result(@_);
311 my ($source, $attrs) = @_;
312 $source = $source->resolve
313 if $source->isa('DBIx::Class::ResultSourceHandle');
315 $attrs = { %{$attrs||{}} };
316 delete @{$attrs}{qw(_last_sqlmaker_alias_map _simple_passthrough_construction)};
318 if ($attrs->{page}) {
319 $attrs->{rows} ||= 10;
322 $attrs->{alias} ||= 'me';
325 result_source => $source,
326 cond => $attrs->{where},
331 # if there is a dark selector, this means we are already in a
332 # chain and the cleanup/sanification was taken care of by
334 $self->_normalize_selection($attrs)
335 unless $attrs->{_dark_selector};
338 $attrs->{result_class} || $source->result_class
348 =item Arguments: L<$cond|DBIx::Class::SQLMaker> | undef, L<\%attrs?|/ATTRIBUTES>
350 =item Return Value: $resultset (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
354 my @cds = $cd_rs->search({ year => 2001 }); # "... WHERE year = 2001"
355 my $new_rs = $cd_rs->search({ year => 2005 });
357 my $new_rs = $cd_rs->search([ { year => 2005 }, { year => 2004 } ]);
358 # year = 2005 OR year = 2004
360 In list context, C<< ->all() >> is called implicitly on the resultset, thus
361 returning a list of L<result|DBIx::Class::Manual::ResultClass> objects instead.
362 To avoid that, use L</search_rs>.
364 If you need to pass in additional attributes but no additional condition,
365 call it as C<search(undef, \%attrs)>.
367 # "SELECT name, artistid FROM $artist_table"
368 my @all_artists = $schema->resultset('Artist')->search(undef, {
369 columns => [qw/name artistid/],
372 For a list of attributes that can be passed to C<search>, see
373 L</ATTRIBUTES>. For more examples of using this function, see
374 L<Searching|DBIx::Class::Manual::Cookbook/SEARCHING>. For a complete
375 documentation for the first argument, see L<SQL::Abstract/"WHERE CLAUSES">
376 and its extension L<DBIx::Class::SQLMaker>.
378 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
382 Note that L</search> does not process/deflate any of the values passed in the
383 L<SQL::Abstract>-compatible search condition structure. This is unlike other
384 condition-bound methods L</new_result>, L</create> and L</find>. The user must ensure
385 manually that any value passed to this method will stringify to something the
386 RDBMS knows how to deal with. A notable example is the handling of L<DateTime>
387 objects, for more info see:
388 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
394 my $rs = $self->search_rs( @_ );
397 DBIx::Class::_ENV_::ASSERT_NO_INTERNAL_WANTARRAY and my $sog = fail_on_internal_wantarray;
400 elsif (defined wantarray) {
404 # we can be called by a relationship helper, which in
405 # turn may be called in void context due to some braindead
406 # overload or whatever else the user decided to be clever
407 # at this particular day. Thus limit the exception to
408 # external code calls only
409 $self->throw_exception ('->search is *not* a mutator, calling it in void context makes no sense')
410 if (caller)[0] !~ /^\QDBIx::Class::/;
420 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
422 =item Return Value: L<$resultset|/search>
426 This method does the same exact thing as search() except it will
427 always return a resultset, even in list context.
434 my $rsrc = $self->result_source;
435 my ($call_cond, $call_attrs);
437 # Special-case handling for (undef, undef) or (undef)
438 # Note that (foo => undef) is valid deprecated syntax
439 @_ = () if not scalar grep { defined $_ } @_;
445 # fish out attrs in the ($condref, $attr) case
446 elsif (@_ == 2 and ( ! defined $_[0] or length ref $_[0] ) ) {
447 ($call_cond, $call_attrs) = @_;
450 $self->throw_exception('Odd number of arguments to search')
454 carp_unique 'search( %condition ) is deprecated, use search( \%condition ) instead'
455 unless $rsrc->result_class->isa('DBIx::Class::CDBICompat');
457 for my $i (0 .. $#_) {
459 $self->throw_exception ('All keys in condition key/value pairs must be plain scalars')
460 if (! defined $_[$i] or length ref $_[$i] );
466 # see if we can keep the cache (no $rs changes)
468 my %safe = (alias => 1, cache => 1);
469 if ( ! List::Util::first { !$safe{$_} } keys %$call_attrs and (
472 ref $call_cond eq 'HASH' && ! keys %$call_cond
474 ref $call_cond eq 'ARRAY' && ! @$call_cond
476 $cache = $self->get_cache;
479 my $old_attrs = { %{$self->{attrs}} };
480 my ($old_having, $old_where) = delete @{$old_attrs}{qw(having where)};
482 my $new_attrs = { %$old_attrs };
484 # take care of call attrs (only if anything is changing)
485 if ($call_attrs and keys %$call_attrs) {
487 # copy for _normalize_selection
488 $call_attrs = { %$call_attrs };
490 my @selector_attrs = qw/select as columns cols +select +as +columns include_columns/;
492 # reset the current selector list if new selectors are supplied
493 if (List::Util::first { exists $call_attrs->{$_} } qw/columns cols select as/) {
494 delete @{$old_attrs}{(@selector_attrs, '_dark_selector')};
497 # Normalize the new selector list (operates on the passed-in attr structure)
498 # Need to do it on every chain instead of only once on _resolved_attrs, in
499 # order to allow detection of empty vs partial 'as'
500 $call_attrs->{_dark_selector} = $old_attrs->{_dark_selector}
501 if $old_attrs->{_dark_selector};
502 $self->_normalize_selection ($call_attrs);
504 # start with blind overwriting merge, exclude selector attrs
505 $new_attrs = { %{$old_attrs}, %{$call_attrs} };
506 delete @{$new_attrs}{@selector_attrs};
508 for (@selector_attrs) {
509 $new_attrs->{$_} = $self->_merge_attr($old_attrs->{$_}, $call_attrs->{$_})
510 if ( exists $old_attrs->{$_} or exists $call_attrs->{$_} );
513 # older deprecated name, use only if {columns} is not there
514 if (my $c = delete $new_attrs->{cols}) {
515 carp_unique( "Resultset attribute 'cols' is deprecated, use 'columns' instead" );
516 if ($new_attrs->{columns}) {
517 carp "Resultset specifies both the 'columns' and the legacy 'cols' attributes - ignoring 'cols'";
520 $new_attrs->{columns} = $c;
525 # join/prefetch use their own crazy merging heuristics
526 foreach my $key (qw/join prefetch/) {
527 $new_attrs->{$key} = $self->_merge_joinpref_attr($old_attrs->{$key}, $call_attrs->{$key})
528 if exists $call_attrs->{$key};
531 # stack binds together
532 $new_attrs->{bind} = [ @{ $old_attrs->{bind} || [] }, @{ $call_attrs->{bind} || [] } ];
536 for ($old_where, $call_cond) {
538 $new_attrs->{where} = $self->_stack_cond (
539 $_, $new_attrs->{where}
544 if (defined $old_having) {
545 $new_attrs->{having} = $self->_stack_cond (
546 $old_having, $new_attrs->{having}
550 my $rs = (ref $self)->new($rsrc, $new_attrs);
552 $rs->set_cache($cache) if ($cache);
558 sub _normalize_selection {
559 my ($self, $attrs) = @_;
562 if ( exists $attrs->{include_columns} ) {
563 carp_unique( "Resultset attribute 'include_columns' is deprecated, use '+columns' instead" );
564 $attrs->{'+columns'} = $self->_merge_attr(
565 $attrs->{'+columns'}, delete $attrs->{include_columns}
569 # columns are always placed first, however
571 # Keep the X vs +X separation until _resolved_attrs time - this allows to
572 # delay the decision on whether to use a default select list ($rsrc->columns)
573 # allowing stuff like the remove_columns helper to work
575 # select/as +select/+as pairs need special handling - the amount of select/as
576 # elements in each pair does *not* have to be equal (think multicolumn
577 # selectors like distinct(foo, bar) ). If the selector is bare (no 'as'
578 # supplied at all) - try to infer the alias, either from the -as parameter
579 # of the selector spec, or use the parameter whole if it looks like a column
580 # name (ugly legacy heuristic). If all fails - leave the selector bare (which
581 # is ok as well), but make sure no more additions to the 'as' chain take place
582 for my $pref ('', '+') {
584 my ($sel, $as) = map {
585 my $key = "${pref}${_}";
587 my $val = [ ref $attrs->{$key} eq 'ARRAY'
589 : $attrs->{$key} || ()
591 delete $attrs->{$key};
595 if (! @$as and ! @$sel ) {
598 elsif (@$as and ! @$sel) {
599 $self->throw_exception(
600 "Unable to handle ${pref}as specification (@$as) without a corresponding ${pref}select"
604 # no as part supplied at all - try to deduce (unless explicit end of named selection is declared)
605 # if any @$as has been supplied we assume the user knows what (s)he is doing
606 # and blindly keep stacking up pieces
607 unless ($attrs->{_dark_selector}) {
610 if ( ref $_ eq 'HASH' and exists $_->{-as} ) {
611 push @$as, $_->{-as};
613 # assume any plain no-space, no-parenthesis string to be a column spec
614 # FIXME - this is retarded but is necessary to support shit like 'count(foo)'
615 elsif ( ! ref $_ and $_ =~ /^ [^\s\(\)]+ $/x) {
618 # if all else fails - raise a flag that no more aliasing will be allowed
620 $attrs->{_dark_selector} = {
622 string => ($dark_sel_dumper ||= do {
623 require Data::Dumper::Concise;
624 Data::Dumper::Concise::DumperObject()->Indent(0);
625 })->Values([$_])->Dump
633 elsif (@$as < @$sel) {
634 $self->throw_exception(
635 "Unable to handle an ${pref}as specification (@$as) with less elements than the corresponding ${pref}select"
638 elsif ($pref and $attrs->{_dark_selector}) {
639 $self->throw_exception(
640 "Unable to process named '+select', resultset contains an unnamed selector $attrs->{_dark_selector}{string}"
646 $attrs->{"${pref}select"} = $self->_merge_attr($attrs->{"${pref}select"}, $sel);
647 $attrs->{"${pref}as"} = $self->_merge_attr($attrs->{"${pref}as"}, $as);
652 my ($self, $left, $right) = @_;
655 (ref $_ eq 'ARRAY' and !@$_)
657 (ref $_ eq 'HASH' and ! keys %$_)
658 ) and $_ = undef for ($left, $right);
660 # either one of the two undef
661 if ( (defined $left) xor (defined $right) ) {
662 return defined $left ? $left : $right;
665 elsif ( ! defined $left ) {
669 return $self->result_source->schema->storage->_collapse_cond({ -and => [$left, $right] });
673 =head2 search_literal
675 B<CAVEAT>: C<search_literal> is provided for Class::DBI compatibility and
676 should only be used in that context. C<search_literal> is a convenience
677 method. It is equivalent to calling C<< $schema->search(\[]) >>, but if you
678 want to ensure columns are bound correctly, use L</search>.
680 See L<DBIx::Class::Manual::Cookbook/SEARCHING> and
681 L<DBIx::Class::Manual::FAQ/Searching> for searching techniques that do not
682 require C<search_literal>.
686 =item Arguments: $sql_fragment, @standalone_bind_values
688 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
692 my @cds = $cd_rs->search_literal('year = ? AND title = ?', qw/2001 Reload/);
693 my $newrs = $artist_rs->search_literal('name = ?', 'Metallica');
695 Pass a literal chunk of SQL to be added to the conditional part of the
698 Example of how to use C<search> instead of C<search_literal>
700 my @cds = $cd_rs->search_literal('cdid = ? AND (artist = ? OR artist = ?)', (2, 1, 2));
701 my @cds = $cd_rs->search(\[ 'cdid = ? AND (artist = ? OR artist = ?)', [ 'cdid', 2 ], [ 'artist', 1 ], [ 'artist', 2 ] ]);
706 my ($self, $sql, @bind) = @_;
708 if ( @bind && ref($bind[-1]) eq 'HASH' ) {
711 return $self->search(\[ $sql, map [ {} => $_ ], @bind ], ($attr || () ));
718 =item Arguments: \%columns_values | @pk_values, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
720 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
724 Finds and returns a single row based on supplied criteria. Takes either a
725 hashref with the same format as L</create> (including inference of foreign
726 keys from related objects), or a list of primary key values in the same
727 order as the L<primary columns|DBIx::Class::ResultSource/primary_columns>
728 declaration on the L</result_source>.
730 In either case an attempt is made to combine conditions already existing on
731 the resultset with the condition passed to this method.
733 To aid with preparing the correct query for the storage you may supply the
734 C<key> attribute, which is the name of a
735 L<unique constraint|DBIx::Class::ResultSource/add_unique_constraint> (the
736 unique constraint corresponding to the
737 L<primary columns|DBIx::Class::ResultSource/primary_columns> is always named
738 C<primary>). If the C<key> attribute has been supplied, and DBIC is unable
739 to construct a query that satisfies the named unique constraint fully (
740 non-NULL values for each column member of the constraint) an exception is
743 If no C<key> is specified, the search is carried over all unique constraints
744 which are fully defined by the available condition.
746 If no such constraint is found, C<find> currently defaults to a simple
747 C<< search->(\%column_values) >> which may or may not do what you expect.
748 Note that this fallback behavior may be deprecated in further versions. If
749 you need to search with arbitrary conditions - use L</search>. If the query
750 resulting from this fallback produces more than one row, a warning to the
751 effect is issued, though only the first row is constructed and returned as
754 In addition to C<key>, L</find> recognizes and applies standard
755 L<resultset attributes|/ATTRIBUTES> in the same way as L</search> does.
757 Note that if you have extra concerns about the correctness of the resulting
758 query you need to specify the C<key> attribute and supply the entire condition
759 as an argument to find (since it is not always possible to perform the
760 combination of the resultset condition with the supplied one, especially if
761 the resultset condition contains literal sql).
763 For example, to find a row by its primary key:
765 my $cd = $schema->resultset('CD')->find(5);
767 You can also find a row by a specific unique constraint:
769 my $cd = $schema->resultset('CD')->find(
771 artist => 'Massive Attack',
772 title => 'Mezzanine',
774 { key => 'cd_artist_title' }
777 See also L</find_or_create> and L</update_or_create>.
783 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
785 my $rsrc = $self->result_source;
788 if (exists $attrs->{key}) {
789 $constraint_name = defined $attrs->{key}
791 : $self->throw_exception("An undefined 'key' resultset attribute makes no sense")
795 # Parse out the condition from input
798 if (ref $_[0] eq 'HASH') {
799 $call_cond = { %{$_[0]} };
802 # if only values are supplied we need to default to 'primary'
803 $constraint_name = 'primary' unless defined $constraint_name;
805 my @c_cols = $rsrc->unique_constraint_columns($constraint_name);
807 $self->throw_exception(
808 "No constraint columns, maybe a malformed '$constraint_name' constraint?"
811 $self->throw_exception (
812 'find() expects either a column/value hashref, or a list of values '
813 . "corresponding to the columns of the specified unique constraint '$constraint_name'"
814 ) unless @c_cols == @_;
816 @{$call_cond}{@c_cols} = @_;
819 # process relationship data if any
820 for my $key (keys %$call_cond) {
822 length ref($call_cond->{$key})
824 my $relinfo = $rsrc->relationship_info($key)
826 # implicitly skip has_many's (likely MC)
827 (ref (my $val = delete $call_cond->{$key}) ne 'ARRAY' )
829 my ($rel_cond, $crosstable) = $rsrc->_resolve_condition(
830 $relinfo->{cond}, $val, $key, $key
833 $self->throw_exception("Complex condition via relationship '$key' is unsupported in find()")
834 if $crosstable or ref($rel_cond) ne 'HASH';
836 # supplement condition
837 # relationship conditions take precedence (?)
838 @{$call_cond}{keys %$rel_cond} = values %$rel_cond;
842 my $alias = exists $attrs->{alias} ? $attrs->{alias} : $self->{attrs}{alias};
844 if (defined $constraint_name) {
845 $final_cond = $self->_qualify_cond_columns (
847 $self->result_source->_minimal_valueset_satisfying_constraint(
848 constraint_name => $constraint_name,
849 values => ($self->_merge_with_rscond($call_cond))[0],
856 elsif ($self->{attrs}{accessor} and $self->{attrs}{accessor} eq 'single') {
857 # This means that we got here after a merger of relationship conditions
858 # in ::Relationship::Base::search_related (the row method), and furthermore
859 # the relationship is of the 'single' type. This means that the condition
860 # provided by the relationship (already attached to $self) is sufficient,
861 # as there can be only one row in the database that would satisfy the
865 my (@unique_queries, %seen_column_combinations, $ci, @fc_exceptions);
867 # no key was specified - fall down to heuristics mode:
868 # run through all unique queries registered on the resultset, and
869 # 'OR' all qualifying queries together
871 # always start from 'primary' if it exists at all
872 for my $c_name ( sort {
874 : $b eq 'primary' ? 1
876 } $rsrc->unique_constraint_names) {
878 next if $seen_column_combinations{
879 join "\x00", sort $rsrc->unique_constraint_columns($c_name)
883 push @unique_queries, $self->_qualify_cond_columns(
884 $self->result_source->_minimal_valueset_satisfying_constraint(
885 constraint_name => $c_name,
886 values => ($self->_merge_with_rscond($call_cond))[0],
887 columns_info => ($ci ||= $self->result_source->columns_info),
893 push @fc_exceptions, $_ if $_ =~ /\bFilterColumn\b/;
898 @unique_queries ? \@unique_queries
899 : @fc_exceptions ? $self->throw_exception(join "; ", map { $_ =~ /(.*) at .+ line \d+$/s } @fc_exceptions )
900 : $self->_non_unique_find_fallback ($call_cond, $attrs)
904 # Run the query, passing the result_class since it should propagate for find
905 my $rs = $self->search ($final_cond, {result_class => $self->result_class, %$attrs});
906 if ($rs->_resolved_attrs->{collapse}) {
908 carp "Query returned more than one row" if $rs->next;
916 # This is a stop-gap method as agreed during the discussion on find() cleanup:
917 # http://lists.scsys.co.uk/pipermail/dbix-class/2010-October/009535.html
919 # It is invoked when find() is called in legacy-mode with insufficiently-unique
920 # condition. It is provided for overrides until a saner way forward is devised
922 # *NOTE* This is not a public method, and it's *GUARANTEED* to disappear down
923 # the road. Please adjust your tests accordingly to catch this situation early
924 # DBIx::Class::ResultSet->can('_non_unique_find_fallback') is reasonable
926 # The method will not be removed without an adequately complete replacement
927 # for strict-mode enforcement
928 sub _non_unique_find_fallback {
929 my ($self, $cond, $attrs) = @_;
931 return $self->_qualify_cond_columns(
933 exists $attrs->{alias}
935 : $self->{attrs}{alias}
940 sub _qualify_cond_columns {
941 my ($self, $cond, $alias) = @_;
943 my %aliased = %$cond;
944 for (keys %aliased) {
945 $aliased{"$alias.$_"} = delete $aliased{$_}
952 sub _build_unique_cond {
954 '_build_unique_cond is a private method, and moreover is about to go '
955 . 'away. Please contact the development team at %s if you believe you '
956 . 'have a genuine use for this method, in order to discuss alternatives.',
957 DBIx::Class::_ENV_::HELP_URL,
960 my ($self, $constraint_name, $cond, $croak_on_null) = @_;
962 $self->result_source->_minimal_valueset_satisfying_constraint(
963 constraint_name => $constraint_name,
965 carp_on_nulls => !$croak_on_null
969 =head2 search_related
973 =item Arguments: $rel_name, $cond?, L<\%attrs?|/ATTRIBUTES>
975 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
979 $new_rs = $cd_rs->search_related('artist', {
983 Searches the specified relationship, optionally specifying a condition and
984 attributes for matching records. See L</ATTRIBUTES> for more information.
986 In list context, C<< ->all() >> is called implicitly on the resultset, thus
987 returning a list of result objects instead. To avoid that, use L</search_related_rs>.
989 See also L</search_related_rs>.
994 return shift->related_resultset(shift)->search(@_);
997 =head2 search_related_rs
999 This method works exactly the same as search_related, except that
1000 it guarantees a resultset, even in list context.
1004 sub search_related_rs {
1005 return shift->related_resultset(shift)->search_rs(@_);
1012 =item Arguments: none
1014 =item Return Value: L<$cursor|DBIx::Class::Cursor>
1018 Returns a storage-driven cursor to the given resultset. See
1019 L<DBIx::Class::Cursor> for more information.
1026 return $self->{cursor} ||= do {
1027 my $attrs = $self->_resolved_attrs;
1028 $self->result_source->storage->select(
1029 $attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs
1038 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1040 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1044 my $cd = $schema->resultset('CD')->single({ year => 2001 });
1046 Inflates the first result without creating a cursor if the resultset has
1047 any records in it; if not returns C<undef>. Used by L</find> as a lean version
1050 While this method can take an optional search condition (just like L</search>)
1051 being a fast-code-path it does not recognize search attributes. If you need to
1052 add extra joins or similar, call L</search> and then chain-call L</single> on the
1053 L<DBIx::Class::ResultSet> returned.
1059 As of 0.08100, this method enforces the assumption that the preceding
1060 query returns only one row. If more than one row is returned, you will receive
1063 Query returned more than one row
1065 In this case, you should be using L</next> or L</find> instead, or if you really
1066 know what you are doing, use the L</rows> attribute to explicitly limit the size
1069 This method will also throw an exception if it is called on a resultset prefetching
1070 has_many, as such a prefetch implies fetching multiple rows from the database in
1071 order to assemble the resulting object.
1078 my ($self, $where) = @_;
1080 $self->throw_exception('single() only takes search conditions, no attributes. You want ->search( $cond, $attrs )->single()');
1083 my $attrs = { %{$self->_resolved_attrs} };
1085 $self->throw_exception(
1086 'single() can not be used on resultsets collapsing a has_many. Use find( \%cond ) or next() instead'
1087 ) if $attrs->{collapse};
1090 if (defined $attrs->{where}) {
1093 [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ }
1094 $where, delete $attrs->{where} ]
1097 $attrs->{where} = $where;
1101 my $data = [ $self->result_source->storage->select_single(
1102 $attrs->{from}, $attrs->{select},
1103 $attrs->{where}, $attrs
1106 return undef unless @$data;
1107 $self->{_stashed_rows} = [ $data ];
1108 $self->_construct_results->[0];
1115 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1117 =item Return Value: L<$resultsetcolumn|DBIx::Class::ResultSetColumn>
1121 my $max_length = $rs->get_column('length')->max;
1123 Returns a L<DBIx::Class::ResultSetColumn> instance for a column of the ResultSet.
1128 my ($self, $column) = @_;
1129 my $new = DBIx::Class::ResultSetColumn->new($self, $column);
1137 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1139 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1143 # WHERE title LIKE '%blue%'
1144 $cd_rs = $rs->search_like({ title => '%blue%'});
1146 Performs a search, but uses C<LIKE> instead of C<=> as the condition. Note
1147 that this is simply a convenience method retained for ex Class::DBI users.
1148 You most likely want to use L</search> with specific operators.
1150 For more information, see L<DBIx::Class::Manual::Cookbook>.
1152 This method is deprecated and will be removed in 0.09. Use L<search()|/search>
1153 instead. An example conversion is:
1155 ->search_like({ foo => 'bar' });
1159 ->search({ foo => { like => 'bar' } });
1166 'search_like() is deprecated and will be removed in DBIC version 0.09.'
1167 .' Instead use ->search({ x => { -like => "y%" } })'
1168 .' (note the outer pair of {}s - they are important!)'
1170 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
1171 my $query = ref $_[0] eq 'HASH' ? { %{shift()} }: {@_};
1172 $query->{$_} = { 'like' => $query->{$_} } for keys %$query;
1173 return $class->search($query, { %$attrs });
1180 =item Arguments: $first, $last
1182 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1186 Returns a resultset or object list representing a subset of elements from the
1187 resultset slice is called on. Indexes are from 0, i.e., to get the first
1188 three records, call:
1190 my ($one, $two, $three) = $rs->slice(0, 2);
1195 my ($self, $min, $max) = @_;
1196 my $attrs = {}; # = { %{ $self->{attrs} || {} } };
1197 $attrs->{offset} = $self->{attrs}{offset} || 0;
1198 $attrs->{offset} += $min;
1199 $attrs->{rows} = ($max ? ($max - $min + 1) : 1);
1200 return $self->search(undef, $attrs);
1207 =item Arguments: none
1209 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1213 Returns the next element in the resultset (C<undef> is there is none).
1215 Can be used to efficiently iterate over records in the resultset:
1217 my $rs = $schema->resultset('CD')->search;
1218 while (my $cd = $rs->next) {
1222 Note that you need to store the resultset object, and call C<next> on it.
1223 Calling C<< resultset('Table')->next >> repeatedly will always return the
1224 first record from the resultset.
1231 if (my $cache = $self->get_cache) {
1232 $self->{all_cache_position} ||= 0;
1233 return $cache->[$self->{all_cache_position}++];
1236 if ($self->{attrs}{cache}) {
1237 delete $self->{pager};
1238 $self->{all_cache_position} = 1;
1239 return ($self->all)[0];
1242 return shift(@{$self->{_stashed_results}}) if @{ $self->{_stashed_results}||[] };
1244 $self->{_stashed_results} = $self->_construct_results
1247 return shift @{$self->{_stashed_results}};
1250 # Constructs as many results as it can in one pass while respecting
1251 # cursor laziness. Several modes of operation:
1253 # * Always builds everything present in @{$self->{_stashed_rows}}
1254 # * If called with $fetch_all true - pulls everything off the cursor and
1255 # builds all result structures (or objects) in one pass
1256 # * If $self->_resolved_attrs->{collapse} is true, checks the order_by
1257 # and if the resultset is ordered properly by the left side:
1258 # * Fetches stuff off the cursor until the "master object" changes,
1259 # and saves the last extra row (if any) in @{$self->{_stashed_rows}}
1261 # * Just fetches, and collapses/constructs everything as if $fetch_all
1262 # was requested (there is no other way to collapse except for an
1264 # * If no collapse is requested - just get the next row, construct and
1266 sub _construct_results {
1267 my ($self, $fetch_all) = @_;
1269 my $rsrc = $self->result_source;
1270 my $attrs = $self->_resolved_attrs;
1275 ! $attrs->{order_by}
1279 my @pcols = $rsrc->primary_columns
1281 # default order for collapsing unless the user asked for something
1282 $attrs->{order_by} = [ map { join '.', $attrs->{alias}, $_} @pcols ];
1283 $attrs->{_ordered_for_collapse} = 1;
1284 $attrs->{_order_is_artificial} = 1;
1287 # this will be used as both initial raw-row collector AND as a RV of
1288 # _construct_results. Not regrowing the array twice matters a lot...
1289 # a surprising amount actually
1290 my $rows = delete $self->{_stashed_rows};
1292 my $cursor; # we may not need one at all
1294 my $did_fetch_all = $fetch_all;
1297 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1298 $rows = [ ($rows ? @$rows : ()), $self->cursor->all ];
1300 elsif( $attrs->{collapse} ) {
1302 # a cursor will need to be closed over in case of collapse
1303 $cursor = $self->cursor;
1305 $attrs->{_ordered_for_collapse} = (
1311 ->_extract_colinfo_of_stable_main_source_order_by_portion($attrs)
1313 ) unless defined $attrs->{_ordered_for_collapse};
1315 if (! $attrs->{_ordered_for_collapse}) {
1318 # instead of looping over ->next, use ->all in stealth mode
1319 # *without* calling a ->reset afterwards
1320 # FIXME ENCAPSULATION - encapsulation breach, cursor method additions pending
1321 if (! $cursor->{_done}) {
1322 $rows = [ ($rows ? @$rows : ()), $cursor->all ];
1323 $cursor->{_done} = 1;
1328 if (! $did_fetch_all and ! @{$rows||[]} ) {
1329 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1330 $cursor ||= $self->cursor;
1331 if (scalar (my @r = $cursor->next) ) {
1336 return undef unless @{$rows||[]};
1338 # sanity check - people are too clever for their own good
1339 if ($attrs->{collapse} and my $aliastypes = $attrs->{_last_sqlmaker_alias_map} ) {
1341 my $multiplied_selectors;
1342 for my $sel_alias ( grep { $_ ne $attrs->{alias} } keys %{ $aliastypes->{selecting} } ) {
1344 $aliastypes->{multiplying}{$sel_alias}
1346 $aliastypes->{premultiplied}{$sel_alias}
1348 $multiplied_selectors->{$_} = 1 for values %{$aliastypes->{selecting}{$sel_alias}{-seen_columns}}
1352 for my $i (0 .. $#{$attrs->{as}} ) {
1353 my $sel = $attrs->{select}[$i];
1355 if (ref $sel eq 'SCALAR') {
1358 elsif( ref $sel eq 'REF' and ref $$sel eq 'ARRAY' ) {
1362 $self->throw_exception(
1363 'Result collapse not possible - selection from a has_many source redirected to the main object'
1364 ) if ($multiplied_selectors->{$sel} and $attrs->{as}[$i] !~ /\./);
1368 # hotspot - skip the setter
1369 my $res_class = $self->_result_class;
1371 my $inflator_cref = $self->{_result_inflator}{cref} ||= do {
1372 $res_class->can ('inflate_result')
1373 or $self->throw_exception("Inflator $res_class does not provide an inflate_result() method");
1376 my $infmap = $attrs->{as};
1378 $self->{_result_inflator}{is_core_row} = ( (
1381 ( \&DBIx::Class::Row::inflate_result || die "No ::Row::inflate_result() - can't happen" )
1382 ) ? 1 : 0 ) unless defined $self->{_result_inflator}{is_core_row};
1384 $self->{_result_inflator}{is_hri} = ( (
1385 ! $self->{_result_inflator}{is_core_row}
1387 $inflator_cref == \&DBIx::Class::ResultClass::HashRefInflator::inflate_result
1388 ) ? 1 : 0 ) unless defined $self->{_result_inflator}{is_hri};
1391 if ($attrs->{_simple_passthrough_construction}) {
1392 # construct a much simpler array->hash folder for the one-table HRI cases right here
1393 if ($self->{_result_inflator}{is_hri}) {
1394 for my $r (@$rows) {
1395 $r = { map { $infmap->[$_] => $r->[$_] } 0..$#$infmap };
1398 # FIXME SUBOPTIMAL this is a very very very hot spot
1399 # while rather optimal we can *still* do much better, by
1400 # building a smarter Row::inflate_result(), and
1401 # switch to feeding it data via a much leaner interface
1403 # crude unscientific benchmarking indicated the shortcut eval is not worth it for
1404 # this particular resultset size
1405 elsif ( $self->{_result_inflator}{is_core_row} and @$rows < 60 ) {
1406 for my $r (@$rows) {
1407 $r = $inflator_cref->($res_class, $rsrc, { map { $infmap->[$_] => $r->[$_] } (0..$#$infmap) } );
1412 ( $self->{_result_inflator}{is_core_row}
1413 ? '$_ = $inflator_cref->($res_class, $rsrc, { %s }) for @$rows'
1414 # a custom inflator may be a multiplier/reductor - put it in direct list ctx
1415 : '@$rows = map { $inflator_cref->($res_class, $rsrc, { %s } ) } @$rows'
1417 ( join (', ', map { "\$infmap->[$_] => \$_->[$_]" } 0..$#$infmap ) )
1423 $self->{_result_inflator}{is_hri} ? 'hri'
1424 : $self->{_result_inflator}{is_core_row} ? 'classic_pruning'
1425 : 'classic_nonpruning'
1428 # $args and $attrs to _mk_row_parser are separated to delineate what is
1429 # core collapser stuff and what is dbic $rs specific
1430 @{$self->{_row_parser}{$parser_type}}{qw(cref nullcheck)} = $rsrc->_mk_row_parser({
1432 inflate_map => $infmap,
1433 collapse => $attrs->{collapse},
1434 premultiplied => $attrs->{_main_source_premultiplied},
1435 hri_style => $self->{_result_inflator}{is_hri},
1436 prune_null_branches => $self->{_result_inflator}{is_hri} || $self->{_result_inflator}{is_core_row},
1437 }, $attrs) unless $self->{_row_parser}{$parser_type}{cref};
1439 # column_info metadata historically hasn't been too reliable.
1440 # We need to start fixing this somehow (the collapse resolver
1441 # can't work without it). Add an explicit check for the *main*
1442 # result, hopefully this will gradually weed out such errors
1444 # FIXME - this is a temporary kludge that reduces performance
1445 # It is however necessary for the time being
1446 my ($unrolled_non_null_cols_to_check, $err);
1448 if (my $check_non_null_cols = $self->{_row_parser}{$parser_type}{nullcheck} ) {
1451 'Collapse aborted due to invalid ResultSource metadata - the following '
1452 . 'selections are declared non-nullable but NULLs were retrieved: '
1456 COL: for my $i (@$check_non_null_cols) {
1457 ! defined $_->[$i] and push @violating_idx, $i and next COL for @$rows;
1460 $self->throw_exception( $err . join (', ', map { "'$infmap->[$_]'" } @violating_idx ) )
1463 $unrolled_non_null_cols_to_check = join (',', @$check_non_null_cols);
1465 utf8::upgrade($unrolled_non_null_cols_to_check)
1466 if DBIx::Class::_ENV_::STRESSTEST_UTF8_UPGRADE_GENERATED_COLLAPSER_SOURCE;
1470 ($did_fetch_all or ! $attrs->{collapse}) ? undef
1471 : defined $unrolled_non_null_cols_to_check ? eval sprintf <<'EOS', $unrolled_non_null_cols_to_check
1473 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1474 my @r = $cursor->next or return;
1475 if (my @violating_idx = grep { ! defined $r[$_] } (%s) ) {
1476 $self->throw_exception( $err . join (', ', map { "'$infmap->[$_]'" } @violating_idx ) )
1482 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1483 my @r = $cursor->next or return;
1488 $self->{_row_parser}{$parser_type}{cref}->(
1490 $next_cref ? ( $next_cref, $self->{_stashed_rows} = [] ) : (),
1493 # simple in-place substitution, does not regrow $rows
1494 if ($self->{_result_inflator}{is_core_row}) {
1495 $_ = $inflator_cref->($res_class, $rsrc, @$_) for @$rows
1497 # Special-case multi-object HRI - there is no $inflator_cref pass at all
1498 elsif ( ! $self->{_result_inflator}{is_hri} ) {
1499 # the inflator may be a multiplier/reductor - put it in list ctx
1500 @$rows = map { $inflator_cref->($res_class, $rsrc, @$_) } @$rows;
1504 # The @$rows check seems odd at first - why wouldn't we want to warn
1505 # regardless? The issue is things like find() etc, where the user
1506 # *knows* only one result will come back. In these cases the ->all
1507 # is not a pessimization, but rather something we actually want
1509 'Unable to properly collapse has_many results in iterator mode due '
1510 . 'to order criteria - performed an eager cursor slurp underneath. '
1511 . 'Consider using ->all() instead'
1512 ) if ( ! $fetch_all and @$rows > 1 );
1517 =head2 result_source
1521 =item Arguments: L<$result_source?|DBIx::Class::ResultSource>
1523 =item Return Value: L<$result_source|DBIx::Class::ResultSource>
1527 An accessor for the primary ResultSource object from which this ResultSet
1534 =item Arguments: $result_class?
1536 =item Return Value: $result_class
1540 An accessor for the class to use when creating result objects. Defaults to
1541 C<< result_source->result_class >> - which in most cases is the name of the
1542 L<"table"|DBIx::Class::Manual::Glossary/"ResultSource"> class.
1544 Note that changing the result_class will also remove any components
1545 that were originally loaded in the source class via
1546 L<load_components|Class::C3::Componentised/load_components( @comps )>.
1547 Any overloaded methods in the original source class will not run.
1552 my ($self, $result_class) = @_;
1553 if ($result_class) {
1555 # don't fire this for an object
1556 $self->ensure_class_loaded($result_class)
1557 unless ref($result_class);
1559 if ($self->get_cache) {
1560 carp_unique('Changing the result_class of a ResultSet instance with cached results is a noop - the cache contents will not be altered');
1562 # FIXME ENCAPSULATION - encapsulation breach, cursor method additions pending
1563 elsif ($self->{cursor} && $self->{cursor}{_pos}) {
1564 $self->throw_exception('Changing the result_class of a ResultSet instance with an active cursor is not supported');
1567 $self->_result_class($result_class);
1569 delete $self->{_result_inflator};
1571 $self->_result_class;
1578 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1580 =item Return Value: $count
1584 Performs an SQL C<COUNT> with the same query as the resultset was built
1585 with to find the number of elements. Passing arguments is equivalent to
1586 C<< $rs->search ($cond, \%attrs)->count >>
1592 return $self->search(@_)->count if @_ and defined $_[0];
1593 return scalar @{ $self->get_cache } if $self->get_cache;
1595 my $attrs = { %{ $self->_resolved_attrs } };
1597 return $self->count_rs->first unless $attrs->{software_limit};
1599 # this is a little optimization - it is faster to do the limit
1600 # adjustments in software, instead of a subquery
1601 my ($rows, $offset) = delete @{$attrs}{qw/rows offset/};
1604 if ($self->_has_resolved_attr (qw/collapse group_by/)) {
1605 $crs = $self->_count_subq_rs ($attrs);
1608 $crs = $self->_count_rs ($attrs);
1610 my $count = $crs->next;
1612 $count -= $offset if $offset;
1613 $count = $rows if $rows and $rows < $count;
1614 $count = 0 if ($count < 0);
1623 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1625 =item Return Value: L<$count_rs|DBIx::Class::ResultSetColumn>
1629 Same as L</count> but returns a L<DBIx::Class::ResultSetColumn> object.
1630 This can be very handy for subqueries:
1632 ->search( { amount => $some_rs->count_rs->as_query } )
1634 As with regular resultsets the SQL query will be executed only after
1635 the resultset is accessed via L</next> or L</all>. That would return
1636 the same single value obtainable via L</count>.
1642 return $self->search(@_)->count_rs if @_;
1644 # this may look like a lack of abstraction (count() does about the same)
1645 # but in fact an _rs *must* use a subquery for the limits, as the
1646 # software based limiting can not be ported if this $rs is to be used
1647 # in a subquery itself (i.e. ->as_query)
1648 if ($self->_has_resolved_attr (qw/collapse group_by offset rows/)) {
1649 return $self->_count_subq_rs($self->{_attrs});
1652 return $self->_count_rs($self->{_attrs});
1657 # returns a ResultSetColumn object tied to the count query
1660 my ($self, $attrs) = @_;
1662 my $rsrc = $self->result_source;
1664 my $tmp_attrs = { %$attrs };
1665 # take off any limits, record_filter is cdbi, and no point of ordering nor locking a count
1666 delete @{$tmp_attrs}{qw/rows offset order_by record_filter for/};
1668 # overwrite the selector (supplied by the storage)
1669 $rsrc->resultset_class->new($rsrc, {
1671 select => $rsrc->storage->_count_select ($rsrc, $attrs),
1673 })->get_column ('count');
1677 # same as above but uses a subquery
1679 sub _count_subq_rs {
1680 my ($self, $attrs) = @_;
1682 my $rsrc = $self->result_source;
1684 my $sub_attrs = { %$attrs };
1685 # extra selectors do not go in the subquery and there is no point of ordering it, nor locking it
1686 delete @{$sub_attrs}{qw/collapse columns as select order_by for/};
1688 # if we multi-prefetch we group_by something unique, as this is what we would
1689 # get out of the rs via ->next/->all. We *DO WANT* to clobber old group_by regardless
1690 if ( $attrs->{collapse} ) {
1691 $sub_attrs->{group_by} = [ map { "$attrs->{alias}.$_" } @{
1692 $rsrc->_identifying_column_set || $self->throw_exception(
1693 'Unable to construct a unique group_by criteria properly collapsing the '
1694 . 'has_many prefetch before count()'
1699 # Calculate subquery selector
1700 if (my $g = $sub_attrs->{group_by}) {
1702 my $sql_maker = $rsrc->storage->sql_maker;
1704 # necessary as the group_by may refer to aliased functions
1706 for my $sel (@{$attrs->{select}}) {
1707 $sel_index->{$sel->{-as}} = $sel
1708 if (ref $sel eq 'HASH' and $sel->{-as});
1711 # anything from the original select mentioned on the group-by needs to make it to the inner selector
1712 # also look for named aggregates referred in the having clause
1713 # having often contains scalarrefs - thus parse it out entirely
1715 if ($attrs->{having}) {
1716 local $sql_maker->{having_bind};
1717 local $sql_maker->{quote_char} = $sql_maker->{quote_char};
1718 local $sql_maker->{name_sep} = $sql_maker->{name_sep};
1719 unless (defined $sql_maker->{quote_char} and length $sql_maker->{quote_char}) {
1720 $sql_maker->{quote_char} = [ "\x00", "\xFF" ];
1721 # if we don't unset it we screw up retarded but unfortunately working
1722 # 'MAX(foo.bar)' => { '>', 3 }
1723 $sql_maker->{name_sep} = '';
1726 my ($lquote, $rquote, $sep) = map { quotemeta $_ } ($sql_maker->_quote_chars, $sql_maker->name_sep);
1728 my $having_sql = $sql_maker->_parse_rs_attrs ({ having => $attrs->{having} });
1731 # search for both a proper quoted qualified string, for a naive unquoted scalarref
1732 # and if all fails for an utterly naive quoted scalar-with-function
1733 while ($having_sql =~ /
1734 $rquote $sep $lquote (.+?) $rquote
1736 [\s,] \w+ \. (\w+) [\s,]
1738 [\s,] $lquote (.+?) $rquote [\s,]
1740 my $part = $1 || $2 || $3; # one of them matched if we got here
1741 unless ($seen_having{$part}++) {
1748 my $colpiece = $sel_index->{$_} || $_;
1750 # unqualify join-based group_by's. Arcane but possible query
1751 # also horrible horrible hack to alias a column (not a func.)
1752 # (probably need to introduce SQLA syntax)
1753 if ($colpiece =~ /\./ && $colpiece !~ /^$attrs->{alias}\./) {
1756 $colpiece = \ sprintf ('%s AS %s', map { $sql_maker->_quote ($_) } ($colpiece, $as) );
1758 push @{$sub_attrs->{select}}, $colpiece;
1762 my @pcols = map { "$attrs->{alias}.$_" } ($rsrc->primary_columns);
1763 $sub_attrs->{select} = @pcols ? \@pcols : [ 1 ];
1766 return $rsrc->resultset_class
1767 ->new ($rsrc, $sub_attrs)
1769 ->search ({}, { columns => { count => $rsrc->storage->_count_select ($rsrc, $attrs) } })
1770 ->get_column ('count');
1774 =head2 count_literal
1776 B<CAVEAT>: C<count_literal> is provided for Class::DBI compatibility and
1777 should only be used in that context. See L</search_literal> for further info.
1781 =item Arguments: $sql_fragment, @standalone_bind_values
1783 =item Return Value: $count
1787 Counts the results in a literal query. Equivalent to calling L</search_literal>
1788 with the passed arguments, then L</count>.
1792 sub count_literal { shift->search_literal(@_)->count; }
1798 =item Arguments: none
1800 =item Return Value: L<@result_objs|DBIx::Class::Manual::ResultClass>
1804 Returns all elements in the resultset.
1811 $self->throw_exception("all() doesn't take any arguments, you probably wanted ->search(...)->all()");
1814 delete @{$self}{qw/_stashed_rows _stashed_results/};
1816 if (my $c = $self->get_cache) {
1820 $self->cursor->reset;
1822 my $objs = $self->_construct_results('fetch_all') || [];
1824 $self->set_cache($objs) if $self->{attrs}{cache};
1833 =item Arguments: none
1835 =item Return Value: $self
1839 Resets the resultset's cursor, so you can iterate through the elements again.
1840 Implicitly resets the storage cursor, so a subsequent L</next> will trigger
1848 delete @{$self}{qw/_stashed_rows _stashed_results/};
1849 $self->{all_cache_position} = 0;
1850 $self->cursor->reset;
1858 =item Arguments: none
1860 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1864 L<Resets|/reset> the resultset (causing a fresh query to storage) and returns
1865 an object for the first result (or C<undef> if the resultset is empty).
1870 return $_[0]->reset->next;
1876 # Determines whether and what type of subquery is required for the $rs operation.
1877 # If grouping is necessary either supplies its own, or verifies the current one
1878 # After all is done delegates to the proper storage method.
1880 sub _rs_update_delete {
1881 my ($self, $op, $values) = @_;
1883 my $rsrc = $self->result_source;
1884 my $storage = $rsrc->schema->storage;
1886 my $attrs = { %{$self->_resolved_attrs} };
1888 my $join_classifications;
1889 my ($existing_group_by) = delete @{$attrs}{qw(group_by _grouped_by_distinct)};
1891 # do we need a subquery for any reason?
1893 defined $existing_group_by
1895 # if {from} is unparseable wrap a subq
1896 ref($attrs->{from}) ne 'ARRAY'
1898 # limits call for a subq
1899 $self->_has_resolved_attr(qw/rows offset/)
1902 # simplify the joinmap, so we can further decide if a subq is necessary
1903 if (!$needs_subq and @{$attrs->{from}} > 1) {
1905 ($attrs->{from}, $join_classifications) =
1906 $storage->_prune_unused_joins ($attrs);
1908 # any non-pruneable non-local restricting joins imply subq
1909 $needs_subq = defined List::Util::first { $_ ne $attrs->{alias} } keys %{ $join_classifications->{restricting} || {} };
1912 # check if the head is composite (by now all joins are thrown out unless $needs_subq)
1914 (ref $attrs->{from}[0]) ne 'HASH'
1916 ref $attrs->{from}[0]{ $attrs->{from}[0]{-alias} }
1920 # do we need anything like a subquery?
1921 if (! $needs_subq) {
1922 # Most databases do not allow aliasing of tables in UPDATE/DELETE. Thus
1923 # a condition containing 'me' or other table prefixes will not work
1924 # at all. Tell SQLMaker to dequalify idents via a gross hack.
1926 my $sqla = $rsrc->storage->sql_maker;
1927 local $sqla->{_dequalify_idents} = 1;
1928 \[ $sqla->_recurse_where($self->{cond}) ];
1932 # we got this far - means it is time to wrap a subquery
1933 my $idcols = $rsrc->_identifying_column_set || $self->throw_exception(
1935 "Unable to perform complex resultset %s() without an identifying set of columns on source '%s'",
1941 # make a new $rs selecting only the PKs (that's all we really need for the subq)
1942 delete $attrs->{$_} for qw/select as collapse/;
1943 $attrs->{columns} = [ map { "$attrs->{alias}.$_" } @$idcols ];
1945 # this will be consumed by the pruner waaaaay down the stack
1946 $attrs->{_force_prune_multiplying_joins} = 1;
1948 my $subrs = (ref $self)->new($rsrc, $attrs);
1950 if (@$idcols == 1) {
1951 $cond = { $idcols->[0] => { -in => $subrs->as_query } };
1953 elsif ($storage->_use_multicolumn_in) {
1954 # no syntax for calling this properly yet
1955 # !!! EXPERIMENTAL API !!! WILL CHANGE !!!
1956 $cond = $storage->sql_maker->_where_op_multicolumn_in (
1957 $idcols, # how do I convey a list of idents...? can binds reside on lhs?
1962 # if all else fails - get all primary keys and operate over a ORed set
1963 # wrap in a transaction for consistency
1964 # this is where the group_by/multiplication starts to matter
1968 # we do not need to check pre-multipliers, since if the premulti is there, its
1969 # parent (who is multi) will be there too
1970 keys %{ $join_classifications->{multiplying} || {} }
1972 # make sure if there is a supplied group_by it matches the columns compiled above
1973 # perfectly. Anything else can not be sanely executed on most databases so croak
1974 # right then and there
1975 if ($existing_group_by) {
1976 my @current_group_by = map
1977 { $_ =~ /\./ ? $_ : "$attrs->{alias}.$_" }
1982 join ("\x00", sort @current_group_by)
1984 join ("\x00", sort @{$attrs->{columns}} )
1986 $self->throw_exception (
1987 "You have just attempted a $op operation on a resultset which does group_by"
1988 . ' on columns other than the primary keys, while DBIC internally needs to retrieve'
1989 . ' the primary keys in a subselect. All sane RDBMS engines do not support this'
1990 . ' kind of queries. Please retry the operation with a modified group_by or'
1991 . ' without using one at all.'
1996 $subrs = $subrs->search({}, { group_by => $attrs->{columns} });
1999 $guard = $storage->txn_scope_guard;
2001 for my $row ($subrs->cursor->all) {
2003 { $idcols->[$_] => $row->[$_] }
2010 my $res = $cond ? $storage->$op (
2012 $op eq 'update' ? $values : (),
2016 $guard->commit if $guard;
2025 =item Arguments: \%values
2027 =item Return Value: $underlying_storage_rv
2031 Sets the specified columns in the resultset to the supplied values in a
2032 single query. Note that this will not run any accessor/set_column/update
2033 triggers, nor will it update any result object instances derived from this
2034 resultset (this includes the contents of the L<resultset cache|/set_cache>
2035 if any). See L</update_all> if you need to execute any on-update
2036 triggers or cascades defined either by you or a
2037 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
2039 The return value is a pass through of what the underlying
2040 storage backend returned, and may vary. See L<DBI/execute> for the most
2045 Note that L</update> does not process/deflate any of the values passed in.
2046 This is unlike the corresponding L<DBIx::Class::Row/update>. The user must
2047 ensure manually that any value passed to this method will stringify to
2048 something the RDBMS knows how to deal with. A notable example is the
2049 handling of L<DateTime> objects, for more info see:
2050 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
2055 my ($self, $values) = @_;
2056 $self->throw_exception('Values for update must be a hash')
2057 unless ref $values eq 'HASH';
2059 return $self->_rs_update_delete ('update', $values);
2066 =item Arguments: \%values
2068 =item Return Value: 1
2072 Fetches all objects and updates them one at a time via
2073 L<DBIx::Class::Row/update>. Note that C<update_all> will run DBIC defined
2074 triggers, while L</update> will not.
2079 my ($self, $values) = @_;
2080 $self->throw_exception('Values for update_all must be a hash')
2081 unless ref $values eq 'HASH';
2083 my $guard = $self->result_source->schema->txn_scope_guard;
2084 $_->update({%$values}) for $self->all; # shallow copy - update will mangle it
2093 =item Arguments: none
2095 =item Return Value: $underlying_storage_rv
2099 Deletes the rows matching this resultset in a single query. Note that this
2100 will not run any delete triggers, nor will it alter the
2101 L<in_storage|DBIx::Class::Row/in_storage> status of any result object instances
2102 derived from this resultset (this includes the contents of the
2103 L<resultset cache|/set_cache> if any). See L</delete_all> if you need to
2104 execute any on-delete triggers or cascades defined either by you or a
2105 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
2107 The return value is a pass through of what the underlying storage backend
2108 returned, and may vary. See L<DBI/execute> for the most common case.
2114 $self->throw_exception('delete does not accept any arguments')
2117 return $self->_rs_update_delete ('delete');
2124 =item Arguments: none
2126 =item Return Value: 1
2130 Fetches all objects and deletes them one at a time via
2131 L<DBIx::Class::Row/delete>. Note that C<delete_all> will run DBIC defined
2132 triggers, while L</delete> will not.
2138 $self->throw_exception('delete_all does not accept any arguments')
2141 my $guard = $self->result_source->schema->txn_scope_guard;
2142 $_->delete for $self->all;
2151 =item Arguments: [ \@column_list, \@row_values+ ] | [ \%col_data+ ]
2153 =item Return Value: L<\@result_objects|DBIx::Class::Manual::ResultClass> (scalar context) | L<@result_objects|DBIx::Class::Manual::ResultClass> (list context)
2157 Accepts either an arrayref of hashrefs or alternatively an arrayref of
2164 The context of this method call has an important effect on what is
2165 submitted to storage. In void context data is fed directly to fastpath
2166 insertion routines provided by the underlying storage (most often
2167 L<DBI/execute_for_fetch>), bypassing the L<new|DBIx::Class::Row/new> and
2168 L<insert|DBIx::Class::Row/insert> calls on the
2169 L<Result|DBIx::Class::Manual::ResultClass> class, including any
2170 augmentation of these methods provided by components. For example if you
2171 are using something like L<DBIx::Class::UUIDColumns> to create primary
2172 keys for you, you will find that your PKs are empty. In this case you
2173 will have to explicitly force scalar or list context in order to create
2178 In non-void (scalar or list) context, this method is simply a wrapper
2179 for L</create>. Depending on list or scalar context either a list of
2180 L<Result|DBIx::Class::Manual::ResultClass> objects or an arrayref
2181 containing these objects is returned.
2183 When supplying data in "arrayref of arrayrefs" invocation style, the
2184 first element should be a list of column names and each subsequent
2185 element should be a data value in the earlier specified column order.
2188 $schema->resultset("Artist")->populate([
2189 [ qw( artistid name ) ],
2190 [ 100, 'A Formally Unknown Singer' ],
2191 [ 101, 'A singer that jumped the shark two albums ago' ],
2192 [ 102, 'An actually cool singer' ],
2195 For the arrayref of hashrefs style each hashref should be a structure
2196 suitable for passing to L</create>. Multi-create is also permitted with
2199 $schema->resultset("Artist")->populate([
2200 { artistid => 4, name => 'Manufactured Crap', cds => [
2201 { title => 'My First CD', year => 2006 },
2202 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2205 { artistid => 5, name => 'Angsty-Whiny Girl', cds => [
2206 { title => 'My parents sold me to a record company', year => 2005 },
2207 { title => 'Why Am I So Ugly?', year => 2006 },
2208 { title => 'I Got Surgery and am now Popular', year => 2007 }
2213 If you attempt a void-context multi-create as in the example above (each
2214 Artist also has the related list of CDs), and B<do not> supply the
2215 necessary autoinc foreign key information, this method will proxy to the
2216 less efficient L</create>, and then throw the Result objects away. In this
2217 case there are obviously no benefits to using this method over L</create>.
2224 # this is naive and just a quick check
2225 # the types will need to be checked more thoroughly when the
2226 # multi-source populate gets added
2228 ref $_[0] eq 'ARRAY'
2230 ( @{$_[0]} or return )
2232 ( ref $_[0][0] eq 'HASH' or ref $_[0][0] eq 'ARRAY' )
2235 ) or $self->throw_exception('Populate expects an arrayref of hashrefs or arrayref of arrayrefs');
2237 # FIXME - no cref handling
2238 # At this point assume either hashes or arrays
2240 if(defined wantarray) {
2241 my (@results, $guard);
2243 if (ref $data->[0] eq 'ARRAY') {
2244 # column names only, nothing to do
2245 return if @$data == 1;
2247 $guard = $self->result_source->schema->storage->txn_scope_guard
2251 { my $vals = $_; $self->new_result({ map { $data->[0][$_] => $vals->[$_] } 0..$#{$data->[0]} })->insert }
2252 @{$data}[1 .. $#$data]
2257 $guard = $self->result_source->schema->storage->txn_scope_guard
2260 @results = map { $self->new_result($_)->insert } @$data;
2263 $guard->commit if $guard;
2264 return wantarray ? @results : \@results;
2267 # we have to deal with *possibly incomplete* related data
2268 # this means we have to walk the data structure twice
2269 # whether we want this or not
2270 # jnap, I hate you ;)
2271 my $rsrc = $self->result_source;
2272 my $rel_info = { map { $_ => $rsrc->relationship_info($_) } $rsrc->relationships };
2274 my ($colinfo, $colnames, $slices_with_rels);
2278 for my $i (0 .. $#$data) {
2280 my $current_slice_seen_rel_infos;
2282 ### Determine/Supplement collists
2283 ### BEWARE - This is a hot piece of code, a lot of weird idioms were used
2284 if( ref $data->[$i] eq 'ARRAY' ) {
2286 # positional(!) explicit column list
2288 # column names only, nothing to do
2289 return if @$data == 1;
2291 $colinfo->{$data->[0][$_]} = { pos => $_, name => $data->[0][$_] } and push @$colnames, $data->[0][$_]
2292 for 0 .. $#{$data->[0]};
2299 for (values %$colinfo) {
2300 if ($_->{is_rel} ||= (
2301 $rel_info->{$_->{name}}
2304 ref $data->[$i][$_->{pos}] eq 'ARRAY'
2306 ref $data->[$i][$_->{pos}] eq 'HASH'
2308 ( defined blessed $data->[$i][$_->{pos}] and $data->[$i][$_->{pos}]->isa('DBIx::Class::Row') )
2314 # moar sanity check... sigh
2315 for ( ref $data->[$i][$_->{pos}] eq 'ARRAY' ? @{$data->[$i][$_->{pos}]} : $data->[$i][$_->{pos}] ) {
2316 if ( defined blessed $_ and $_->isa('DBIx::Class::Row' ) ) {
2317 carp_unique("Fast-path populate() with supplied related objects is not possible - falling back to regular create()");
2318 return my $throwaway = $self->populate(@_);
2322 push @$current_slice_seen_rel_infos, $rel_info->{$_->{name}};
2327 if ($current_slice_seen_rel_infos) {
2328 push @$slices_with_rels, { map { $colnames->[$_] => $data->[$i][$_] } 0 .. $#$colnames };
2330 # this is needed further down to decide whether or not to fallback to create()
2331 $colinfo->{$colnames->[$_]}{seen_null} ||= ! defined $data->[$i][$_]
2332 for 0 .. $#$colnames;
2335 elsif( ref $data->[$i] eq 'HASH' ) {
2337 for ( sort keys %{$data->[$i]} ) {
2339 $colinfo->{$_} ||= do {
2341 $self->throw_exception("Column '$_' must be present in supplied explicit column list")
2342 if $data_start; # it will be 0 on AoH, 1 on AoA
2344 push @$colnames, $_;
2347 { pos => $#$colnames, name => $_ }
2350 if ($colinfo->{$_}{is_rel} ||= (
2354 ref $data->[$i]{$_} eq 'ARRAY'
2356 ref $data->[$i]{$_} eq 'HASH'
2358 ( defined blessed $data->[$i]{$_} and $data->[$i]{$_}->isa('DBIx::Class::Row') )
2364 # moar sanity check... sigh
2365 for ( ref $data->[$i]{$_} eq 'ARRAY' ? @{$data->[$i]{$_}} : $data->[$i]{$_} ) {
2366 if ( defined blessed $_ and $_->isa('DBIx::Class::Row' ) ) {
2367 carp_unique("Fast-path populate() with supplied related objects is not possible - falling back to regular create()");
2368 return my $throwaway = $self->populate(@_);
2372 push @$current_slice_seen_rel_infos, $rel_info->{$_};
2376 if ($current_slice_seen_rel_infos) {
2377 push @$slices_with_rels, $data->[$i];
2379 # this is needed further down to decide whether or not to fallback to create()
2380 $colinfo->{$_}{seen_null} ||= ! defined $data->[$i]{$_}
2381 for keys %{$data->[$i]};
2385 $self->throw_exception('Unexpected populate() data structure member type: ' . ref $data->[$i] );
2389 { $_->{attrs}{is_depends_on} }
2390 @{ $current_slice_seen_rel_infos || [] }
2392 carp_unique("Fast-path populate() of belongs_to relationship data is not possible - falling back to regular create()");
2393 return my $throwaway = $self->populate(@_);
2397 if( $slices_with_rels ) {
2399 # need to exclude the rel "columns"
2400 $colnames = [ grep { ! $colinfo->{$_}{is_rel} } @$colnames ];
2402 # extra sanity check - ensure the main source is in fact identifiable
2403 # the localizing of nullability is insane, but oh well... the use-case is legit
2404 my $ci = $rsrc->columns_info($colnames);
2406 $ci->{$_} = { %{$ci->{$_}}, is_nullable => 0 }
2407 for grep { ! $colinfo->{$_}{seen_null} } keys %$ci;
2409 unless( $rsrc->_identifying_column_set($ci) ) {
2410 carp_unique("Fast-path populate() of non-uniquely identifiable rows with related data is not possible - falling back to regular create()");
2411 return my $throwaway = $self->populate(@_);
2415 ### inherit the data locked in the conditions of the resultset
2416 my ($rs_data) = $self->_merge_with_rscond({});
2417 delete @{$rs_data}{@$colnames}; # passed-in stuff takes precedence
2419 # if anything left - decompose rs_data
2421 if (keys %$rs_data) {
2422 push @$rs_data_vals, $rs_data->{$_}
2423 for sort keys %$rs_data;
2428 $guard = $rsrc->schema->storage->txn_scope_guard
2429 if $slices_with_rels;
2431 ### main source data
2432 # FIXME - need to switch entirely to a coderef-based thing,
2433 # so that large sets aren't copied several times... I think
2434 $rsrc->storage->_insert_bulk(
2436 [ @$colnames, sort keys %$rs_data ],
2438 ref $data->[$_] eq 'ARRAY'
2440 $slices_with_rels ? [ @{$data->[$_]}[0..$#$colnames], @{$rs_data_vals||[]} ] # the collist changed
2441 : $rs_data_vals ? [ @{$data->[$_]}, @$rs_data_vals ]
2444 : [ @{$data->[$_]}{@$colnames}, @{$rs_data_vals||[]} ]
2445 } $data_start .. $#$data ],
2448 ### do the children relationships
2449 if ( $slices_with_rels ) {
2450 my @rels = grep { $colinfo->{$_}{is_rel} } keys %$colinfo
2451 or die 'wtf... please report a bug with DBIC_TRACE=1 output (stacktrace)';
2453 for my $sl (@$slices_with_rels) {
2455 my ($main_proto, $main_proto_rs);
2456 for my $rel (@rels) {
2457 next unless defined $sl->{$rel};
2461 (map { $_ => $sl->{$_} } @$colnames),
2464 unless (defined $colinfo->{$rel}{rs}) {
2466 $colinfo->{$rel}{rs} = $rsrc->related_source($rel)->resultset;
2468 $colinfo->{$rel}{fk_map} = { reverse %{ $rsrc->_resolve_relationship_condition(
2470 self_alias => "\xFE", # irrelevant
2471 foreign_alias => "\xFF", # irrelevant
2472 )->{identity_map} || {} } };
2476 $colinfo->{$rel}{rs}->search({ map # only so that we inherit them values properly, no actual search
2479 ( $main_proto_rs ||= $rsrc->resultset->search($main_proto) )
2480 ->get_column( $colinfo->{$rel}{fk_map}{$_} )
2484 keys %{$colinfo->{$rel}{fk_map}}
2485 })->populate( ref $sl->{$rel} eq 'ARRAY' ? $sl->{$rel} : [ $sl->{$rel} ] );
2492 $guard->commit if $guard;
2499 =item Arguments: none
2501 =item Return Value: L<$pager|Data::Page>
2505 Returns a L<Data::Page> object for the current resultset. Only makes
2506 sense for queries with a C<page> attribute.
2508 To get the full count of entries for a paged resultset, call
2509 C<total_entries> on the L<Data::Page> object.
2516 return $self->{pager} if $self->{pager};
2518 my $attrs = $self->{attrs};
2519 if (!defined $attrs->{page}) {
2520 $self->throw_exception("Can't create pager for non-paged rs");
2522 elsif ($attrs->{page} <= 0) {
2523 $self->throw_exception('Invalid page number (page-numbers are 1-based)');
2525 $attrs->{rows} ||= 10;
2527 # throw away the paging flags and re-run the count (possibly
2528 # with a subselect) to get the real total count
2529 my $count_attrs = { %$attrs };
2530 delete @{$count_attrs}{qw/rows offset page pager/};
2532 my $total_rs = (ref $self)->new($self->result_source, $count_attrs);
2534 require DBIx::Class::ResultSet::Pager;
2535 return $self->{pager} = DBIx::Class::ResultSet::Pager->new(
2536 sub { $total_rs->count }, #lazy-get the total
2538 $self->{attrs}{page},
2546 =item Arguments: $page_number
2548 =item Return Value: L<$resultset|/search>
2552 Returns a resultset for the $page_number page of the resultset on which page
2553 is called, where each page contains a number of rows equal to the 'rows'
2554 attribute set on the resultset (10 by default).
2559 my ($self, $page) = @_;
2560 return (ref $self)->new($self->result_source, { %{$self->{attrs}}, page => $page });
2567 =item Arguments: \%col_data
2569 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2573 Creates a new result object in the resultset's result class and returns
2574 it. The row is not inserted into the database at this point, call
2575 L<DBIx::Class::Row/insert> to do that. Calling L<DBIx::Class::Row/in_storage>
2576 will tell you whether the result object has been inserted or not.
2578 Passes the hashref of input on to L<DBIx::Class::Row/new>.
2583 my ($self, $values) = @_;
2585 $self->throw_exception( "new_result takes only one argument - a hashref of values" )
2588 $self->throw_exception( "Result object instantiation requires a hashref as argument" )
2589 unless (ref $values eq 'HASH');
2591 my ($merged_cond, $cols_from_relations) = $self->_merge_with_rscond($values);
2593 my $new = $self->result_class->new({
2595 ( @$cols_from_relations
2596 ? (-cols_from_relations => $cols_from_relations)
2599 -result_source => $self->result_source, # DO NOT REMOVE THIS, REQUIRED
2603 reftype($new) eq 'HASH'
2609 carp_unique (sprintf (
2610 "%s->new returned a blessed empty hashref - a strong indicator something is wrong with its inheritance chain",
2611 $self->result_class,
2618 # _merge_with_rscond
2620 # Takes a simple hash of K/V data and returns its copy merged with the
2621 # condition already present on the resultset. Additionally returns an
2622 # arrayref of value/condition names, which were inferred from related
2623 # objects (this is needed for in-memory related objects)
2624 sub _merge_with_rscond {
2625 my ($self, $data) = @_;
2627 my ($implied_data, @cols_from_relations);
2629 my $alias = $self->{attrs}{alias};
2631 if (! defined $self->{cond}) {
2632 # just massage $data below
2634 elsif ($self->{cond} eq UNRESOLVABLE_CONDITION) {
2635 $implied_data = $self->{attrs}{related_objects}; # nothing might have been inserted yet
2636 @cols_from_relations = keys %{ $implied_data || {} };
2639 my $eqs = $self->result_source->schema->storage->_extract_fixed_condition_columns($self->{cond}, 'consider_nulls');
2640 $implied_data = { map {
2641 ( ($eqs->{$_}||'') eq UNRESOLVABLE_CONDITION ) ? () : ( $_ => $eqs->{$_} )
2647 { %{ $self->_remove_alias($_, $alias) } }
2648 # precedence must be given to passed values over values inherited from
2649 # the cond, so the order here is important.
2650 ( $implied_data||(), $data)
2652 \@cols_from_relations
2656 # _has_resolved_attr
2658 # determines if the resultset defines at least one
2659 # of the attributes supplied
2661 # used to determine if a subquery is necessary
2663 # supports some virtual attributes:
2665 # This will scan for any joins being present on the resultset.
2666 # It is not a mere key-search but a deep inspection of {from}
2669 sub _has_resolved_attr {
2670 my ($self, @attr_names) = @_;
2672 my $attrs = $self->_resolved_attrs;
2676 for my $n (@attr_names) {
2677 if (grep { $n eq $_ } (qw/-join/) ) {
2678 $extra_checks{$n}++;
2682 my $attr = $attrs->{$n};
2684 next if not defined $attr;
2686 if (ref $attr eq 'HASH') {
2687 return 1 if keys %$attr;
2689 elsif (ref $attr eq 'ARRAY') {
2697 # a resolved join is expressed as a multi-level from
2699 $extra_checks{-join}
2701 ref $attrs->{from} eq 'ARRAY'
2703 @{$attrs->{from}} > 1
2711 # Remove the specified alias from the specified query hash. A copy is made so
2712 # the original query is not modified.
2715 my ($self, $query, $alias) = @_;
2717 my %orig = %{ $query || {} };
2720 foreach my $key (keys %orig) {
2722 $unaliased{$key} = $orig{$key};
2725 $unaliased{$1} = $orig{$key}
2726 if $key =~ m/^(?:\Q$alias\E\.)?([^.]+)$/;
2736 =item Arguments: none
2738 =item Return Value: \[ $sql, L<@bind_values|/DBIC BIND VALUES> ]
2742 Returns the SQL query and bind vars associated with the invocant.
2744 This is generally used as the RHS for a subquery.
2751 my $attrs = { %{ $self->_resolved_attrs } };
2753 my $aq = $self->result_source->storage->_select_args_to_query (
2754 $attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs
2764 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2766 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2770 my $artist = $schema->resultset('Artist')->find_or_new(
2771 { artist => 'fred' }, { key => 'artists' });
2773 $cd->cd_to_producer->find_or_new({ producer => $producer },
2774 { key => 'primary' });
2776 Find an existing record from this resultset using L</find>. if none exists,
2777 instantiate a new result object and return it. The object will not be saved
2778 into your storage until you call L<DBIx::Class::Row/insert> on it.
2780 You most likely want this method when looking for existing rows using a unique
2781 constraint that is not the primary key, or looking for related rows.
2783 If you want objects to be saved immediately, use L</find_or_create> instead.
2785 B<Note>: Make sure to read the documentation of L</find> and understand the
2786 significance of the C<key> attribute, as its lack may skew your search, and
2787 subsequently result in spurious new objects.
2789 B<Note>: Take care when using C<find_or_new> with a table having
2790 columns with default values that you intend to be automatically
2791 supplied by the database (e.g. an auto_increment primary key column).
2792 In normal usage, the value of such columns should NOT be included at
2793 all in the call to C<find_or_new>, even when set to C<undef>.
2799 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2800 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2801 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2804 return $self->new_result($hash);
2811 =item Arguments: \%col_data
2813 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2817 Attempt to create a single new row or a row with multiple related rows
2818 in the table represented by the resultset (and related tables). This
2819 will not check for duplicate rows before inserting, use
2820 L</find_or_create> to do that.
2822 To create one row for this resultset, pass a hashref of key/value
2823 pairs representing the columns of the table and the values you wish to
2824 store. If the appropriate relationships are set up, foreign key fields
2825 can also be passed an object representing the foreign row, and the
2826 value will be set to its primary key.
2828 To create related objects, pass a hashref of related-object column values
2829 B<keyed on the relationship name>. If the relationship is of type C<multi>
2830 (L<DBIx::Class::Relationship/has_many>) - pass an arrayref of hashrefs.
2831 The process will correctly identify columns holding foreign keys, and will
2832 transparently populate them from the keys of the corresponding relation.
2833 This can be applied recursively, and will work correctly for a structure
2834 with an arbitrary depth and width, as long as the relationships actually
2835 exists and the correct column data has been supplied.
2837 Instead of hashrefs of plain related data (key/value pairs), you may
2838 also pass new or inserted objects. New objects (not inserted yet, see
2839 L</new_result>), will be inserted into their appropriate tables.
2841 Effectively a shortcut for C<< ->new_result(\%col_data)->insert >>.
2843 Example of creating a new row.
2845 $person_rs->create({
2846 name=>"Some Person",
2847 email=>"somebody@someplace.com"
2850 Example of creating a new row and also creating rows in a related C<has_many>
2851 or C<has_one> resultset. Note Arrayref.
2854 { artistid => 4, name => 'Manufactured Crap', cds => [
2855 { title => 'My First CD', year => 2006 },
2856 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2861 Example of creating a new row and also creating a row in a related
2862 C<belongs_to> resultset. Note Hashref.
2865 title=>"Music for Silly Walks",
2868 name=>"Silly Musician",
2876 When subclassing ResultSet never attempt to override this method. Since
2877 it is a simple shortcut for C<< $self->new_result($attrs)->insert >>, a
2878 lot of the internals simply never call it, so your override will be
2879 bypassed more often than not. Override either L<DBIx::Class::Row/new>
2880 or L<DBIx::Class::Row/insert> depending on how early in the
2881 L</create> process you need to intervene. See also warning pertaining to
2889 #my ($self, $col_data) = @_;
2890 DBIx::Class::_ENV_::ASSERT_NO_INTERNAL_INDIRECT_CALLS and fail_on_internal_call;
2891 return shift->new_result(shift)->insert;
2894 =head2 find_or_create
2898 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2900 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2904 $cd->cd_to_producer->find_or_create({ producer => $producer },
2905 { key => 'primary' });
2907 Tries to find a record based on its primary key or unique constraints; if none
2908 is found, creates one and returns that instead.
2910 my $cd = $schema->resultset('CD')->find_or_create({
2912 artist => 'Massive Attack',
2913 title => 'Mezzanine',
2917 Also takes an optional C<key> attribute, to search by a specific key or unique
2918 constraint. For example:
2920 my $cd = $schema->resultset('CD')->find_or_create(
2922 artist => 'Massive Attack',
2923 title => 'Mezzanine',
2925 { key => 'cd_artist_title' }
2928 B<Note>: Make sure to read the documentation of L</find> and understand the
2929 significance of the C<key> attribute, as its lack may skew your search, and
2930 subsequently result in spurious row creation.
2932 B<Note>: Because find_or_create() reads from the database and then
2933 possibly inserts based on the result, this method is subject to a race
2934 condition. Another process could create a record in the table after
2935 the find has completed and before the create has started. To avoid
2936 this problem, use find_or_create() inside a transaction.
2938 B<Note>: Take care when using C<find_or_create> with a table having
2939 columns with default values that you intend to be automatically
2940 supplied by the database (e.g. an auto_increment primary key column).
2941 In normal usage, the value of such columns should NOT be included at
2942 all in the call to C<find_or_create>, even when set to C<undef>.
2944 See also L</find> and L</update_or_create>. For information on how to declare
2945 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2947 If you need to know if an existing row was found or a new one created use
2948 L</find_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
2949 to call L<DBIx::Class::Row/insert> to save the newly created row to the
2952 my $cd = $schema->resultset('CD')->find_or_new({
2954 artist => 'Massive Attack',
2955 title => 'Mezzanine',
2959 if( !$cd->in_storage ) {
2966 sub find_or_create {
2968 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2969 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2970 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2973 return $self->new_result($hash)->insert;
2976 =head2 update_or_create
2980 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2982 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2986 $resultset->update_or_create({ col => $val, ... });
2988 Like L</find_or_create>, but if a row is found it is immediately updated via
2989 C<< $found_row->update (\%col_data) >>.
2992 Takes an optional C<key> attribute to search on a specific unique constraint.
2995 # In your application
2996 my $cd = $schema->resultset('CD')->update_or_create(
2998 artist => 'Massive Attack',
2999 title => 'Mezzanine',
3002 { key => 'cd_artist_title' }
3005 $cd->cd_to_producer->update_or_create({
3006 producer => $producer,
3012 B<Note>: Make sure to read the documentation of L</find> and understand the
3013 significance of the C<key> attribute, as its lack may skew your search, and
3014 subsequently result in spurious row creation.
3016 B<Note>: Take care when using C<update_or_create> with a table having
3017 columns with default values that you intend to be automatically
3018 supplied by the database (e.g. an auto_increment primary key column).
3019 In normal usage, the value of such columns should NOT be included at
3020 all in the call to C<update_or_create>, even when set to C<undef>.
3022 See also L</find> and L</find_or_create>. For information on how to declare
3023 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
3025 If you need to know if an existing row was updated or a new one created use
3026 L</update_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
3027 to call L<DBIx::Class::Row/insert> to save the newly created row to the
3032 sub update_or_create {
3034 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
3035 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
3037 my $row = $self->find($cond, $attrs);
3039 $row->update($cond);
3043 return $self->new_result($cond)->insert;
3046 =head2 update_or_new
3050 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
3052 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
3056 $resultset->update_or_new({ col => $val, ... });
3058 Like L</find_or_new> but if a row is found it is immediately updated via
3059 C<< $found_row->update (\%col_data) >>.
3063 # In your application
3064 my $cd = $schema->resultset('CD')->update_or_new(
3066 artist => 'Massive Attack',
3067 title => 'Mezzanine',
3070 { key => 'cd_artist_title' }
3073 if ($cd->in_storage) {
3074 # the cd was updated
3077 # the cd is not yet in the database, let's insert it
3081 B<Note>: Make sure to read the documentation of L</find> and understand the
3082 significance of the C<key> attribute, as its lack may skew your search, and
3083 subsequently result in spurious new objects.
3085 B<Note>: Take care when using C<update_or_new> with a table having
3086 columns with default values that you intend to be automatically
3087 supplied by the database (e.g. an auto_increment primary key column).
3088 In normal usage, the value of such columns should NOT be included at
3089 all in the call to C<update_or_new>, even when set to C<undef>.
3091 See also L</find>, L</find_or_create> and L</find_or_new>.
3097 my $attrs = ( @_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {} );
3098 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
3100 my $row = $self->find( $cond, $attrs );
3101 if ( defined $row ) {
3102 $row->update($cond);
3106 return $self->new_result($cond);
3113 =item Arguments: none
3115 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass> | undef
3119 Gets the contents of the cache for the resultset, if the cache is set.
3121 The cache is populated either by using the L</prefetch> attribute to
3122 L</search> or by calling L</set_cache>.
3134 =item Arguments: L<\@result_objs|DBIx::Class::Manual::ResultClass>
3136 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass>
3140 Sets the contents of the cache for the resultset. Expects an arrayref
3141 of objects of the same class as those produced by the resultset. Note that
3142 if the cache is set, the resultset will return the cached objects rather
3143 than re-querying the database even if the cache attr is not set.
3145 The contents of the cache can also be populated by using the
3146 L</prefetch> attribute to L</search>.
3151 my ( $self, $data ) = @_;
3152 $self->throw_exception("set_cache requires an arrayref")
3153 if defined($data) && (ref $data ne 'ARRAY');
3154 $self->{all_cache} = $data;
3161 =item Arguments: none
3163 =item Return Value: undef
3167 Clears the cache for the resultset.
3172 shift->set_cache(undef);
3179 =item Arguments: none
3181 =item Return Value: true, if the resultset has been paginated
3189 return !!$self->{attrs}{page};
3196 =item Arguments: none
3198 =item Return Value: true, if the resultset has been ordered with C<order_by>.
3206 return scalar $self->result_source->storage->_extract_order_criteria($self->{attrs}{order_by});
3209 =head2 related_resultset
3213 =item Arguments: $rel_name
3215 =item Return Value: L<$resultset|/search>
3219 Returns a related resultset for the supplied relationship name.
3221 $artist_rs = $schema->resultset('CD')->related_resultset('Artist');
3225 sub related_resultset {
3226 $_[0]->throw_exception(
3227 'Extra arguments to $rs->related_resultset() were always quietly '
3228 . 'discarded without consideration, you need to switch to '
3229 . '...->related_resultset( $relname )->search_rs( $search, $args ) instead.'
3232 return $_[0]->{related_resultsets}{$_[1]}
3233 if defined $_[0]->{related_resultsets}{$_[1]};
3235 my ($self, $rel) = @_;
3237 return $self->{related_resultsets}{$rel} = do {
3238 my $rsrc = $self->result_source;
3239 my $rel_info = $rsrc->relationship_info($rel);
3241 $self->throw_exception(
3242 "search_related: result source '" . $rsrc->source_name .
3243 "' has no such relationship $rel")
3246 my $attrs = $self->_chain_relationship($rel);
3248 my $storage = $rsrc->schema->storage;
3250 # Previously this atribute was deleted (instead of being set as it is now)
3251 # Doing so seems to be harmless in all available test permutations
3252 # See also 01d59a6a6 and mst's comment below
3254 $attrs->{alias} = $storage->relname_to_table_alias(
3256 $attrs->{seen_join}{$rel}
3259 # since this is search_related, and we already slid the select window inwards
3260 # (the select/as attrs were deleted in the beginning), we need to flip all
3261 # left joins to inner, so we get the expected results
3262 # read the comment on top of the actual function to see what this does
3263 $attrs->{from} = $storage->_inner_join_to_node( $attrs->{from}, $attrs->{alias} );
3265 #XXX - temp fix for result_class bug. There likely is a more elegant fix -groditi
3266 delete $attrs->{result_class};
3270 # The reason we do this now instead of passing the alias to the
3271 # search_rs below is that if you wrap/overload resultset on the
3272 # source you need to know what alias it's -going- to have for things
3273 # to work sanely (e.g. RestrictWithObject wants to be able to add
3274 # extra query restrictions, and these may need to be $alias.)
3275 # -- mst ~ 2007 (01d59a6a6)
3277 # FIXME - this seems to be no longer neccessary (perhaps due to the
3278 # advances in relcond resolution. Testing DBIC::S::RWO and its only
3279 # dependent (as of Jun 2015 ) does not yield any difference with or
3280 # without this line. Nevertheless keep it as is for now, to minimize
3281 # churn, there is enough potential for breakage in 0.0829xx as it is
3282 # -- ribasushi Jun 2015
3284 my $rel_source = $rsrc->related_source($rel);
3285 local $rel_source->resultset_attributes->{alias} = $attrs->{alias};
3287 $rel_source->resultset->search_rs( undef, $attrs );
3290 if (my $cache = $self->get_cache) {
3291 my @related_cache = map
3292 { $_->related_resultset($rel)->get_cache || () }
3296 $new->set_cache([ map @$_, @related_cache ]) if @related_cache == @$cache;
3303 =head2 current_source_alias
3307 =item Arguments: none
3309 =item Return Value: $source_alias
3313 Returns the current table alias for the result source this resultset is built
3314 on, that will be used in the SQL query. Usually it is C<me>.
3316 Currently the source alias that refers to the result set returned by a
3317 L</search>/L</find> family method depends on how you got to the resultset: it's
3318 C<me> by default, but eg. L</search_related> aliases it to the related result
3319 source name (and keeps C<me> referring to the original result set). The long
3320 term goal is to make L<DBIx::Class> always alias the current resultset as C<me>
3321 (and make this method unnecessary).
3323 Thus it's currently necessary to use this method in predefined queries (see
3324 L<DBIx::Class::Manual::Cookbook/Predefined searches>) when referring to the
3325 source alias of the current result set:
3327 # in a result set class
3329 my ($self, $user) = @_;
3331 my $me = $self->current_source_alias;
3333 return $self->search({
3334 "$me.modified" => $user->id,
3338 The alias of L<newly created resultsets|/search> can be altered by the
3339 L<alias attribute|/alias>.
3343 sub current_source_alias {
3344 return (shift->{attrs} || {})->{alias} || 'me';
3347 =head2 as_subselect_rs
3351 =item Arguments: none
3353 =item Return Value: L<$resultset|/search>
3357 Act as a barrier to SQL symbols. The resultset provided will be made into a
3358 "virtual view" by including it as a subquery within the from clause. From this
3359 point on, any joined tables are inaccessible to ->search on the resultset (as if
3360 it were simply where-filtered without joins). For example:
3362 my $rs = $schema->resultset('Bar')->search({'x.name' => 'abc'},{ join => 'x' });
3364 # 'x' now pollutes the query namespace
3366 # So the following works as expected
3367 my $ok_rs = $rs->search({'x.other' => 1});
3369 # But this doesn't: instead of finding a 'Bar' related to two x rows (abc and
3370 # def) we look for one row with contradictory terms and join in another table
3371 # (aliased 'x_2') which we never use
3372 my $broken_rs = $rs->search({'x.name' => 'def'});
3374 my $rs2 = $rs->as_subselect_rs;
3376 # doesn't work - 'x' is no longer accessible in $rs2, having been sealed away
3377 my $not_joined_rs = $rs2->search({'x.other' => 1});
3379 # works as expected: finds a 'table' row related to two x rows (abc and def)
3380 my $correctly_joined_rs = $rs2->search({'x.name' => 'def'});
3382 Another example of when one might use this would be to select a subset of
3383 columns in a group by clause:
3385 my $rs = $schema->resultset('Bar')->search(undef, {
3386 group_by => [qw{ id foo_id baz_id }],
3387 })->as_subselect_rs->search(undef, {
3388 columns => [qw{ id foo_id }]
3391 In the above example normally columns would have to be equal to the group by,
3392 but because we isolated the group by into a subselect the above works.
3396 sub as_subselect_rs {
3399 my $attrs = $self->_resolved_attrs;
3401 my $fresh_rs = (ref $self)->new (
3402 $self->result_source
3405 # these pieces will be locked in the subquery
3406 delete $fresh_rs->{cond};
3407 delete @{$fresh_rs->{attrs}}{qw/where bind/};
3409 return $fresh_rs->search( {}, {
3411 $attrs->{alias} => $self->as_query,
3412 -alias => $attrs->{alias},
3413 -rsrc => $self->result_source,
3415 alias => $attrs->{alias},
3419 # This code is called by search_related, and makes sure there
3420 # is clear separation between the joins before, during, and
3421 # after the relationship. This information is needed later
3422 # in order to properly resolve prefetch aliases (any alias
3423 # with a relation_chain_depth less than the depth of the
3424 # current prefetch is not considered)
3426 # The increments happen twice per join. An even number means a
3427 # relationship specified via a search_related, whereas an odd
3428 # number indicates a join/prefetch added via attributes
3430 # Also this code will wrap the current resultset (the one we
3431 # chain to) in a subselect IFF it contains limiting attributes
3432 sub _chain_relationship {
3433 my ($self, $rel) = @_;
3434 my $source = $self->result_source;
3435 my $attrs = { %{$self->{attrs}||{}} };
3437 # we need to take the prefetch the attrs into account before we
3438 # ->_resolve_join as otherwise they get lost - captainL
3439 my $join = $self->_merge_joinpref_attr( $attrs->{join}, $attrs->{prefetch} );
3441 delete @{$attrs}{qw/join prefetch collapse group_by distinct _grouped_by_distinct select as columns +select +as +columns/};
3443 my $seen = { %{ (delete $attrs->{seen_join}) || {} } };
3446 my @force_subq_attrs = qw/offset rows group_by having/;
3449 ($attrs->{from} && ref $attrs->{from} ne 'ARRAY')
3451 $self->_has_resolved_attr (@force_subq_attrs)
3453 # Nuke the prefetch (if any) before the new $rs attrs
3454 # are resolved (prefetch is useless - we are wrapping
3455 # a subquery anyway).
3456 my $rs_copy = $self->search;
3457 $rs_copy->{attrs}{join} = $self->_merge_joinpref_attr (
3458 $rs_copy->{attrs}{join},
3459 delete $rs_copy->{attrs}{prefetch},
3464 -alias => $attrs->{alias},
3465 $attrs->{alias} => $rs_copy->as_query,
3467 delete @{$attrs}{@force_subq_attrs, qw/where bind/};
3468 $seen->{-relation_chain_depth} = 0;
3470 elsif ($attrs->{from}) { #shallow copy suffices
3471 $from = [ @{$attrs->{from}} ];
3476 -alias => $attrs->{alias},
3477 $attrs->{alias} => $source->from,
3481 my $jpath = ($seen->{-relation_chain_depth})
3482 ? $from->[-1][0]{-join_path}
3485 my @requested_joins = $source->_resolve_join(
3492 push @$from, @requested_joins;
3494 $seen->{-relation_chain_depth}++;
3496 # if $self already had a join/prefetch specified on it, the requested
3497 # $rel might very well be already included. What we do in this case
3498 # is effectively a no-op (except that we bump up the chain_depth on
3499 # the join in question so we could tell it *is* the search_related)
3502 # we consider the last one thus reverse
3503 for my $j (reverse @requested_joins) {
3504 my ($last_j) = keys %{$j->[0]{-join_path}[-1]};
3505 if ($rel eq $last_j) {
3506 $j->[0]{-relation_chain_depth}++;
3512 unless ($already_joined) {
3513 push @$from, $source->_resolve_join(
3521 $seen->{-relation_chain_depth}++;
3523 return {%$attrs, from => $from, seen_join => $seen};
3526 sub _resolved_attrs {
3528 return $self->{_attrs} if $self->{_attrs};
3530 my $attrs = { %{ $self->{attrs} || {} } };
3531 my $source = $attrs->{result_source} = $self->result_source;
3532 my $alias = $attrs->{alias};
3534 $self->throw_exception("Specifying distinct => 1 in conjunction with collapse => 1 is unsupported")
3535 if $attrs->{collapse} and $attrs->{distinct};
3538 # Sanity check the paging attributes
3539 # SQLMaker does it too, but in case of a software_limit we'll never get there
3540 if (defined $attrs->{offset}) {
3541 $self->throw_exception('A supplied offset attribute must be a non-negative integer')
3542 if ( $attrs->{offset} =~ /[^0-9]/ or $attrs->{offset} < 0 );
3544 if (defined $attrs->{rows}) {
3545 $self->throw_exception("The rows attribute must be a positive integer if present")
3546 if ( $attrs->{rows} =~ /[^0-9]/ or $attrs->{rows} <= 0 );
3550 # default selection list
3551 $attrs->{columns} = [ $source->columns ]
3552 unless List::Util::first { exists $attrs->{$_} } qw/columns cols select as/;
3554 # merge selectors together
3555 for (qw/columns select as/) {
3556 $attrs->{$_} = $self->_merge_attr($attrs->{$_}, delete $attrs->{"+$_"})
3557 if $attrs->{$_} or $attrs->{"+$_"};
3560 # disassemble columns
3562 if (my $cols = delete $attrs->{columns}) {
3563 for my $c (ref $cols eq 'ARRAY' ? @$cols : $cols) {
3564 if (ref $c eq 'HASH') {
3565 for my $as (sort keys %$c) {
3566 push @sel, $c->{$as};
3577 # when trying to weed off duplicates later do not go past this point -
3578 # everything added from here on is unbalanced "anyone's guess" stuff
3579 my $dedup_stop_idx = $#as;
3581 push @as, @{ ref $attrs->{as} eq 'ARRAY' ? $attrs->{as} : [ $attrs->{as} ] }
3583 push @sel, @{ ref $attrs->{select} eq 'ARRAY' ? $attrs->{select} : [ $attrs->{select} ] }
3584 if $attrs->{select};
3586 # assume all unqualified selectors to apply to the current alias (legacy stuff)
3587 $_ = (ref $_ or $_ =~ /\./) ? $_ : "$alias.$_" for @sel;
3589 # disqualify all $alias.col as-bits (inflate-map mandated)
3590 $_ = ($_ =~ /^\Q$alias.\E(.+)$/) ? $1 : $_ for @as;
3592 # de-duplicate the result (remove *identical* select/as pairs)
3593 # and also die on duplicate {as} pointing to different {select}s
3594 # not using a c-style for as the condition is prone to shrinkage
3597 while ($i <= $dedup_stop_idx) {
3598 if ($seen->{"$sel[$i] \x00\x00 $as[$i]"}++) {
3603 elsif ($seen->{$as[$i]}++) {
3604 $self->throw_exception(
3605 "inflate_result() alias '$as[$i]' specified twice with different SQL-side {select}-ors"
3613 $attrs->{select} = \@sel;
3614 $attrs->{as} = \@as;
3616 $attrs->{from} ||= [{
3618 -alias => $self->{attrs}{alias},
3619 $self->{attrs}{alias} => $source->from,
3622 if ( $attrs->{join} || $attrs->{prefetch} ) {
3624 $self->throw_exception ('join/prefetch can not be used with a custom {from}')
3625 if ref $attrs->{from} ne 'ARRAY';
3627 my $join = (delete $attrs->{join}) || {};
3629 if ( defined $attrs->{prefetch} ) {
3630 $join = $self->_merge_joinpref_attr( $join, $attrs->{prefetch} );
3633 $attrs->{from} = # have to copy here to avoid corrupting the original
3635 @{ $attrs->{from} },
3636 $source->_resolve_join(
3639 { %{ $attrs->{seen_join} || {} } },
3640 ( $attrs->{seen_join} && keys %{$attrs->{seen_join}})
3641 ? $attrs->{from}[-1][0]{-join_path}
3649 for my $attr (qw(order_by group_by)) {
3651 if ( defined $attrs->{$attr} ) {
3653 ref( $attrs->{$attr} ) eq 'ARRAY'
3654 ? [ @{ $attrs->{$attr} } ]
3655 : [ $attrs->{$attr} || () ]
3658 delete $attrs->{$attr} unless @{$attrs->{$attr}};
3663 # set collapse default based on presence of prefetch
3666 defined $attrs->{prefetch}
3668 $prefetch = $self->_merge_joinpref_attr( {}, delete $attrs->{prefetch} )
3670 $self->throw_exception("Specifying prefetch in conjunction with an explicit collapse => 0 is unsupported")
3671 if defined $attrs->{collapse} and ! $attrs->{collapse};
3673 $attrs->{collapse} = 1;
3677 # run through the resulting joinstructure (starting from our current slot)
3678 # and unset collapse if proven unnecessary
3680 # also while we are at it find out if the current root source has
3681 # been premultiplied by previous related_source chaining
3683 # this allows to predict whether a root object with all other relation
3684 # data set to NULL is in fact unique
3685 if ($attrs->{collapse}) {
3687 if (ref $attrs->{from} eq 'ARRAY') {
3689 if (@{$attrs->{from}} == 1) {
3690 # no joins - no collapse
3691 $attrs->{collapse} = 0;
3694 # find where our table-spec starts
3695 my @fromlist = @{$attrs->{from}};
3697 my $t = shift @fromlist;
3700 # me vs join from-spec distinction - a ref means non-root
3701 if (ref $t eq 'ARRAY') {
3703 $is_multi ||= ! $t->{-is_single};
3705 last if ($t->{-alias} && $t->{-alias} eq $alias);
3706 $attrs->{_main_source_premultiplied} ||= $is_multi;
3709 # no non-singles remaining, nor any premultiplication - nothing to collapse
3711 ! $attrs->{_main_source_premultiplied}
3713 ! List::Util::first { ! $_->[0]{-is_single} } @fromlist
3715 $attrs->{collapse} = 0;
3721 # if we can not analyze the from - err on the side of safety
3722 $attrs->{_main_source_premultiplied} = 1;
3727 # generate the distinct induced group_by before injecting the prefetched select/as parts
3728 if (delete $attrs->{distinct}) {
3729 if ($attrs->{group_by}) {
3730 carp_unique ("Useless use of distinct on a grouped resultset ('distinct' is ignored when a 'group_by' is present)");
3733 $attrs->{_grouped_by_distinct} = 1;
3734 # distinct affects only the main selection part, not what prefetch may add below
3735 ($attrs->{group_by}, my $new_order) = $source->storage->_group_over_selection($attrs);
3737 # FIXME possibly ignore a rewritten order_by (may turn out to be an issue)
3738 # The thinking is: if we are collapsing the subquerying prefetch engine will
3739 # rip stuff apart for us anyway, and we do not want to have a potentially
3740 # function-converted external order_by
3741 # ( there is an explicit if ( collapse && _grouped_by_distinct ) check in DBIHacks )
3742 $attrs->{order_by} = $new_order unless $attrs->{collapse};
3747 # generate selections based on the prefetch helper
3750 $self->throw_exception("Unable to prefetch, resultset contains an unnamed selector $attrs->{_dark_selector}{string}")
3751 if $attrs->{_dark_selector};
3753 # this is a separate structure (we don't look in {from} directly)
3754 # as the resolver needs to shift things off the lists to work
3755 # properly (identical-prefetches on different branches)
3756 my $joined_node_aliases_map = {};
3757 if (ref $attrs->{from} eq 'ARRAY') {
3759 my $start_depth = $attrs->{seen_join}{-relation_chain_depth} || 0;
3761 for my $j ( @{$attrs->{from}}[1 .. $#{$attrs->{from}} ] ) {
3762 next unless $j->[0]{-alias};
3763 next unless $j->[0]{-join_path};
3764 next if ($j->[0]{-relation_chain_depth} || 0) < $start_depth;
3766 my @jpath = map { keys %$_ } @{$j->[0]{-join_path}};
3768 my $p = $joined_node_aliases_map;
3769 $p = $p->{$_} ||= {} for @jpath[ ($start_depth/2) .. $#jpath]; #only even depths are actual jpath boundaries
3770 push @{$p->{-join_aliases} }, $j->[0]{-alias};
3774 ( push @{$attrs->{select}}, $_->[0] ) and ( push @{$attrs->{as}}, $_->[1] )
3775 for $source->_resolve_selection_from_prefetch( $prefetch, $joined_node_aliases_map );
3779 $attrs->{_simple_passthrough_construction} = !(
3782 grep { $_ =~ /\./ } @{$attrs->{as}}
3786 # if both page and offset are specified, produce a combined offset
3787 # even though it doesn't make much sense, this is what pre 081xx has
3789 if (my $page = delete $attrs->{page}) {
3791 ($attrs->{rows} * ($page - 1))
3793 ($attrs->{offset} || 0)
3797 return $self->{_attrs} = $attrs;
3801 my ($self, $attr) = @_;
3803 if (ref $attr eq 'HASH') {
3804 return $self->_rollout_hash($attr);
3805 } elsif (ref $attr eq 'ARRAY') {
3806 return $self->_rollout_array($attr);
3812 sub _rollout_array {
3813 my ($self, $attr) = @_;
3816 foreach my $element (@{$attr}) {
3817 if (ref $element eq 'HASH') {
3818 push( @rolled_array, @{ $self->_rollout_hash( $element ) } );
3819 } elsif (ref $element eq 'ARRAY') {
3820 # XXX - should probably recurse here
3821 push( @rolled_array, @{$self->_rollout_array($element)} );
3823 push( @rolled_array, $element );
3826 return \@rolled_array;
3830 my ($self, $attr) = @_;
3833 foreach my $key (keys %{$attr}) {
3834 push( @rolled_array, { $key => $attr->{$key} } );
3836 return \@rolled_array;
3839 sub _calculate_score {
3840 my ($self, $a, $b) = @_;
3842 if (defined $a xor defined $b) {
3845 elsif (not defined $a) {
3849 if (ref $b eq 'HASH') {
3850 my ($b_key) = keys %{$b};
3851 $b_key = '' if ! defined $b_key;
3852 if (ref $a eq 'HASH') {
3853 my ($a_key) = keys %{$a};
3854 $a_key = '' if ! defined $a_key;
3855 if ($a_key eq $b_key) {
3856 return (1 + $self->_calculate_score( $a->{$a_key}, $b->{$b_key} ));
3861 return ($a eq $b_key) ? 1 : 0;
3864 if (ref $a eq 'HASH') {
3865 my ($a_key) = keys %{$a};
3866 return ($b eq $a_key) ? 1 : 0;
3868 return ($b eq $a) ? 1 : 0;
3873 sub _merge_joinpref_attr {
3874 my ($self, $orig, $import) = @_;
3876 return $import unless defined($orig);
3877 return $orig unless defined($import);
3879 $orig = $self->_rollout_attr($orig);
3880 $import = $self->_rollout_attr($import);
3883 foreach my $import_element ( @{$import} ) {
3884 # find best candidate from $orig to merge $b_element into
3885 my $best_candidate = { position => undef, score => 0 }; my $position = 0;
3886 foreach my $orig_element ( @{$orig} ) {
3887 my $score = $self->_calculate_score( $orig_element, $import_element );
3888 if ($score > $best_candidate->{score}) {
3889 $best_candidate->{position} = $position;
3890 $best_candidate->{score} = $score;
3894 my ($import_key) = ( ref $import_element eq 'HASH' ) ? keys %{$import_element} : ($import_element);
3895 $import_key = '' if not defined $import_key;
3897 if ($best_candidate->{score} == 0 || exists $seen_keys->{$import_key}) {
3898 push( @{$orig}, $import_element );
3900 my $orig_best = $orig->[$best_candidate->{position}];
3901 # merge orig_best and b_element together and replace original with merged
3902 if (ref $orig_best ne 'HASH') {
3903 $orig->[$best_candidate->{position}] = $import_element;
3904 } elsif (ref $import_element eq 'HASH') {
3905 my ($key) = keys %{$orig_best};
3906 $orig->[$best_candidate->{position}] = { $key => $self->_merge_joinpref_attr($orig_best->{$key}, $import_element->{$key}) };
3909 $seen_keys->{$import_key} = 1; # don't merge the same key twice
3912 return @$orig ? $orig : ();
3920 require Hash::Merge;
3921 my $hm = Hash::Merge->new;
3923 $hm->specify_behavior({
3926 my ($defl, $defr) = map { defined $_ } (@_[0,1]);
3928 if ($defl xor $defr) {
3929 return [ $defl ? $_[0] : $_[1] ];
3934 elsif (__HM_DEDUP and $_[0] eq $_[1]) {
3938 return [$_[0], $_[1]];
3942 return $_[1] if !defined $_[0];
3943 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3944 return [$_[0], @{$_[1]}]
3947 return [] if !defined $_[0] and !keys %{$_[1]};
3948 return [ $_[1] ] if !defined $_[0];
3949 return [ $_[0] ] if !keys %{$_[1]};
3950 return [$_[0], $_[1]]
3955 return $_[0] if !defined $_[1];
3956 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3957 return [@{$_[0]}, $_[1]]
3960 my @ret = @{$_[0]} or return $_[1];
3961 return [ @ret, @{$_[1]} ] unless __HM_DEDUP;
3962 my %idx = map { $_ => 1 } @ret;
3963 push @ret, grep { ! defined $idx{$_} } (@{$_[1]});
3967 return [ $_[1] ] if ! @{$_[0]};
3968 return $_[0] if !keys %{$_[1]};
3969 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3970 return [ @{$_[0]}, $_[1] ];
3975 return [] if !keys %{$_[0]} and !defined $_[1];
3976 return [ $_[0] ] if !defined $_[1];
3977 return [ $_[1] ] if !keys %{$_[0]};
3978 return [$_[0], $_[1]]
3981 return [] if !keys %{$_[0]} and !@{$_[1]};
3982 return [ $_[0] ] if !@{$_[1]};
3983 return $_[1] if !keys %{$_[0]};
3984 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3985 return [ $_[0], @{$_[1]} ];
3988 return [] if !keys %{$_[0]} and !keys %{$_[1]};
3989 return [ $_[0] ] if !keys %{$_[1]};
3990 return [ $_[1] ] if !keys %{$_[0]};
3991 return [ $_[0] ] if $_[0] eq $_[1];
3992 return [ $_[0], $_[1] ];
3995 } => 'DBIC_RS_ATTR_MERGER');
3999 return $hm->merge ($_[1], $_[2]);
4003 sub STORABLE_freeze {
4004 my ($self, $cloning) = @_;
4005 my $to_serialize = { %$self };
4007 # A cursor in progress can't be serialized (and would make little sense anyway)
4008 # the parser can be regenerated (and can't be serialized)
4009 delete @{$to_serialize}{qw/cursor _row_parser _result_inflator/};
4011 # nor is it sensical to store a not-yet-fired-count pager
4012 if ($to_serialize->{pager} and ref $to_serialize->{pager}{total_entries} eq 'CODE') {
4013 delete $to_serialize->{pager};
4016 Storable::nfreeze($to_serialize);
4019 # need this hook for symmetry
4021 my ($self, $cloning, $serialized) = @_;
4023 %$self = %{ Storable::thaw($serialized) };
4029 =head2 throw_exception
4031 See L<DBIx::Class::Schema/throw_exception> for details.
4035 sub throw_exception {
4038 if (ref $self and my $rsrc = $self->result_source) {
4039 $rsrc->throw_exception(@_)
4042 DBIx::Class::Exception->throw(@_);
4050 # XXX: FIXME: Attributes docs need clearing up
4054 Attributes are used to refine a ResultSet in various ways when
4055 searching for data. They can be passed to any method which takes an
4056 C<\%attrs> argument. See L</search>, L</search_rs>, L</find>,
4059 Default attributes can be set on the result class using
4060 L<DBIx::Class::ResultSource/resultset_attributes>. (Please read
4061 the CAVEATS on that feature before using it!)
4063 These are in no particular order:
4069 =item Value: ( $order_by | \@order_by | \%order_by )
4073 Which column(s) to order the results by.
4075 [The full list of suitable values is documented in
4076 L<SQL::Abstract/"ORDER BY CLAUSES">; the following is a summary of
4079 If a single column name, or an arrayref of names is supplied, the
4080 argument is passed through directly to SQL. The hashref syntax allows
4081 for connection-agnostic specification of ordering direction:
4083 For descending order:
4085 order_by => { -desc => [qw/col1 col2 col3/] }
4087 For explicit ascending order:
4089 order_by => { -asc => 'col' }
4091 The old scalarref syntax (i.e. order_by => \'year DESC') is still
4092 supported, although you are strongly encouraged to use the hashref
4093 syntax as outlined above.
4099 =item Value: \@columns | \%columns | $column
4103 Shortcut to request a particular set of columns to be retrieved. Each
4104 column spec may be a string (a table column name), or a hash (in which
4105 case the key is the C<as> value, and the value is used as the C<select>
4106 expression). Adds the L</current_source_alias> onto the start of any column without a C<.> in
4107 it and sets C<select> from that, then auto-populates C<as> from
4108 C<select> as normal. (You may also use the C<cols> attribute, as in
4109 earlier versions of DBIC, but this is deprecated)
4111 Essentially C<columns> does the same as L</select> and L</as>.
4113 columns => [ 'some_column', { dbic_slot => 'another_column' } ]
4117 select => [qw(some_column another_column)],
4118 as => [qw(some_column dbic_slot)]
4120 If you want to individually retrieve related columns (in essence perform
4121 manual L</prefetch>) you have to make sure to specify the correct inflation slot
4122 chain such that it matches existing relationships:
4124 my $rs = $schema->resultset('Artist')->search({}, {
4125 # required to tell DBIC to collapse has_many relationships
4127 join => { cds => 'tracks' },
4129 'cds.cdid' => 'cds.cdid',
4130 'cds.tracks.title' => 'tracks.title',
4134 Like elsewhere, literal SQL or literal values can be included by using a
4135 scalar reference or a literal bind value, and these values will be available
4136 in the result with C<get_column> (see also
4137 L<SQL::Abstract/Literal SQL and value type operators>):
4139 # equivalent SQL: SELECT 1, 'a string', IF(my_column,?,?) ...
4140 # bind values: $true_value, $false_value
4144 bar => \q{'a string'},
4145 baz => \[ 'IF(my_column,?,?)', $true_value, $false_value ],
4151 B<NOTE:> You B<MUST> explicitly quote C<'+columns'> when using this attribute.
4152 Not doing so causes Perl to incorrectly interpret C<+columns> as a bareword
4153 with a unary plus operator before it, which is the same as simply C<columns>.
4157 =item Value: \@extra_columns
4161 Indicates additional columns to be selected from storage. Works the same as
4162 L</columns> but adds columns to the current selection. (You may also use the
4163 C<include_columns> attribute, as in earlier versions of DBIC, but this is
4166 $schema->resultset('CD')->search(undef, {
4167 '+columns' => ['artist.name'],
4171 would return all CDs and include a 'name' column to the information
4172 passed to object inflation. Note that the 'artist' is the name of the
4173 column (or relationship) accessor, and 'name' is the name of the column
4174 accessor in the related table.
4180 =item Value: \@select_columns
4184 Indicates which columns should be selected from the storage. You can use
4185 column names, or in the case of RDBMS back ends, function or stored procedure
4188 $rs = $schema->resultset('Employee')->search(undef, {
4191 { count => 'employeeid' },
4192 { max => { length => 'name' }, -as => 'longest_name' }
4197 SELECT name, COUNT( employeeid ), MAX( LENGTH( name ) ) AS longest_name FROM employee
4199 B<NOTE:> You will almost always need a corresponding L</as> attribute when you
4200 use L</select>, to instruct DBIx::Class how to store the result of the column.
4202 Also note that the L</as> attribute has B<nothing to do> with the SQL-side
4203 C<AS> identifier aliasing. You B<can> alias a function (so you can use it e.g.
4204 in an C<ORDER BY> clause), however this is done via the C<-as> B<select
4205 function attribute> supplied as shown in the example above.
4209 B<NOTE:> You B<MUST> explicitly quote C<'+select'> when using this attribute.
4210 Not doing so causes Perl to incorrectly interpret C<+select> as a bareword
4211 with a unary plus operator before it, which is the same as simply C<select>.
4215 =item Value: \@extra_select_columns
4219 Indicates additional columns to be selected from storage. Works the same as
4220 L</select> but adds columns to the current selection, instead of specifying
4221 a new explicit list.
4227 =item Value: \@inflation_names
4231 Indicates DBIC-side names for object inflation. That is L</as> indicates the
4232 slot name in which the column value will be stored within the
4233 L<Row|DBIx::Class::Row> object. The value will then be accessible via this
4234 identifier by the C<get_column> method (or via the object accessor B<if one
4235 with the same name already exists>) as shown below.
4237 The L</as> attribute has B<nothing to do> with the SQL-side identifier
4238 aliasing C<AS>. See L</select> for details.
4240 $rs = $schema->resultset('Employee')->search(undef, {
4243 { count => 'employeeid' },
4244 { max => { length => 'name' }, -as => 'longest_name' }
4253 If the object against which the search is performed already has an accessor
4254 matching a column name specified in C<as>, the value can be retrieved using
4255 the accessor as normal:
4257 my $name = $employee->name();
4259 If on the other hand an accessor does not exist in the object, you need to
4260 use C<get_column> instead:
4262 my $employee_count = $employee->get_column('employee_count');
4264 You can create your own accessors if required - see
4265 L<DBIx::Class::Manual::Cookbook> for details.
4269 B<NOTE:> You B<MUST> explicitly quote C<'+as'> when using this attribute.
4270 Not doing so causes Perl to incorrectly interpret C<+as> as a bareword
4271 with a unary plus operator before it, which is the same as simply C<as>.
4275 =item Value: \@extra_inflation_names
4279 Indicates additional inflation names for selectors added via L</+select>. See L</as>.
4285 =item Value: ($rel_name | \@rel_names | \%rel_names)
4289 Contains a list of relationships that should be joined for this query. For
4292 # Get CDs by Nine Inch Nails
4293 my $rs = $schema->resultset('CD')->search(
4294 { 'artist.name' => 'Nine Inch Nails' },
4295 { join => 'artist' }
4298 Can also contain a hash reference to refer to the other relation's relations.
4301 package MyApp::Schema::Track;
4302 use base qw/DBIx::Class/;
4303 __PACKAGE__->table('track');
4304 __PACKAGE__->add_columns(qw/trackid cd position title/);
4305 __PACKAGE__->set_primary_key('trackid');
4306 __PACKAGE__->belongs_to(cd => 'MyApp::Schema::CD');
4309 # In your application
4310 my $rs = $schema->resultset('Artist')->search(
4311 { 'track.title' => 'Teardrop' },
4313 join => { cd => 'track' },
4314 order_by => 'artist.name',
4318 You need to use the relationship (not the table) name in conditions,
4319 because they are aliased as such. The current table is aliased as "me", so
4320 you need to use me.column_name in order to avoid ambiguity. For example:
4322 # Get CDs from 1984 with a 'Foo' track
4323 my $rs = $schema->resultset('CD')->search(
4326 'tracks.name' => 'Foo'
4328 { join => 'tracks' }
4331 If the same join is supplied twice, it will be aliased to <rel>_2 (and
4332 similarly for a third time). For e.g.
4334 my $rs = $schema->resultset('Artist')->search({
4335 'cds.title' => 'Down to Earth',
4336 'cds_2.title' => 'Popular',
4338 join => [ qw/cds cds/ ],
4341 will return a set of all artists that have both a cd with title 'Down
4342 to Earth' and a cd with title 'Popular'.
4344 If you want to fetch related objects from other tables as well, see L</prefetch>
4347 NOTE: An internal join-chain pruner will discard certain joins while
4348 constructing the actual SQL query, as long as the joins in question do not
4349 affect the retrieved result. This for example includes 1:1 left joins
4350 that are not part of the restriction specification (WHERE/HAVING) nor are
4351 a part of the query selection.
4353 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
4359 =item Value: (0 | 1)
4363 When set to a true value, indicates that any rows fetched from joined has_many
4364 relationships are to be aggregated into the corresponding "parent" object. For
4365 example, the resultset:
4367 my $rs = $schema->resultset('CD')->search({}, {
4368 '+columns' => [ qw/ tracks.title tracks.position / ],
4373 While executing the following query:
4375 SELECT me.*, tracks.title, tracks.position
4377 LEFT JOIN track tracks
4378 ON tracks.cdid = me.cdid
4380 Will return only as many objects as there are rows in the CD source, even
4381 though the result of the query may span many rows. Each of these CD objects
4382 will in turn have multiple "Track" objects hidden behind the has_many
4383 generated accessor C<tracks>. Without C<< collapse => 1 >>, the return values
4384 of this resultset would be as many CD objects as there are tracks (a "Cartesian
4385 product"), with each CD object containing exactly one of all fetched Track data.
4387 When a collapse is requested on a non-ordered resultset, an order by some
4388 unique part of the main source (the left-most table) is inserted automatically.
4389 This is done so that the resultset is allowed to be "lazy" - calling
4390 L<< $rs->next|/next >> will fetch only as many rows as it needs to build the next
4391 object with all of its related data.
4393 If an L</order_by> is already declared, and orders the resultset in a way that
4394 makes collapsing as described above impossible (e.g. C<< ORDER BY
4395 has_many_rel.column >> or C<ORDER BY RANDOM()>), DBIC will automatically
4396 switch to "eager" mode and slurp the entire resultset before constructing the
4397 first object returned by L</next>.
4399 Setting this attribute on a resultset that does not join any has_many
4400 relations is a no-op.
4402 For a more in-depth discussion, see L</PREFETCHING>.
4408 =item Value: ($rel_name | \@rel_names | \%rel_names)
4412 This attribute is a shorthand for specifying a L</join> spec, adding all
4413 columns from the joined related sources as L</+columns> and setting
4414 L</collapse> to a true value. It can be thought of as a rough B<superset>
4415 of the L</join> attribute.
4417 For example, the following two queries are equivalent:
4419 my $rs = $schema->resultset('Artist')->search({}, {
4420 prefetch => { cds => ['genre', 'tracks' ] },
4425 my $rs = $schema->resultset('Artist')->search({}, {
4426 join => { cds => ['genre', 'tracks' ] },
4430 { +{ "cds.$_" => "cds.$_" } }
4431 $schema->source('Artist')->related_source('cds')->columns
4434 { +{ "cds.genre.$_" => "genre.$_" } }
4435 $schema->source('Artist')->related_source('cds')->related_source('genre')->columns
4438 { +{ "cds.tracks.$_" => "tracks.$_" } }
4439 $schema->source('Artist')->related_source('cds')->related_source('tracks')->columns
4444 Both producing the following SQL:
4446 SELECT me.artistid, me.name, me.rank, me.charfield,
4447 cds.cdid, cds.artist, cds.title, cds.year, cds.genreid, cds.single_track,
4448 genre.genreid, genre.name,
4449 tracks.trackid, tracks.cd, tracks.position, tracks.title, tracks.last_updated_on, tracks.last_updated_at
4452 ON cds.artist = me.artistid
4453 LEFT JOIN genre genre
4454 ON genre.genreid = cds.genreid
4455 LEFT JOIN track tracks
4456 ON tracks.cd = cds.cdid
4457 ORDER BY me.artistid
4459 While L</prefetch> implies a L</join>, it is ok to mix the two together, as
4460 the arguments are properly merged and generally do the right thing. For
4461 example, you may want to do the following:
4463 my $artists_and_cds_without_genre = $schema->resultset('Artist')->search(
4464 { 'genre.genreid' => undef },
4466 join => { cds => 'genre' },
4471 Which generates the following SQL:
4473 SELECT me.artistid, me.name, me.rank, me.charfield,
4474 cds.cdid, cds.artist, cds.title, cds.year, cds.genreid, cds.single_track
4477 ON cds.artist = me.artistid
4478 LEFT JOIN genre genre
4479 ON genre.genreid = cds.genreid
4480 WHERE genre.genreid IS NULL
4481 ORDER BY me.artistid
4483 For a more in-depth discussion, see L</PREFETCHING>.
4489 =item Value: $source_alias
4493 Sets the source alias for the query. Normally, this defaults to C<me>, but
4494 nested search queries (sub-SELECTs) might need specific aliases set to
4495 reference inner queries. For example:
4498 ->related_resultset('CDs')
4499 ->related_resultset('Tracks')
4501 'track.id' => { -ident => 'none_search.id' },
4505 my $ids = $self->search({
4508 alias => 'none_search',
4509 group_by => 'none_search.id',
4510 })->get_column('id')->as_query;
4512 $self->search({ id => { -in => $ids } })
4514 This attribute is directly tied to L</current_source_alias>.
4524 Makes the resultset paged and specifies the page to retrieve. Effectively
4525 identical to creating a non-pages resultset and then calling ->page($page)
4528 If L</rows> attribute is not specified it defaults to 10 rows per page.
4530 When you have a paged resultset, L</count> will only return the number
4531 of rows in the page. To get the total, use the L</pager> and call
4532 C<total_entries> on it.
4542 Specifies the maximum number of rows for direct retrieval or the number of
4543 rows per page if the page attribute or method is used.
4549 =item Value: $offset
4553 Specifies the (zero-based) row number for the first row to be returned, or the
4554 of the first row of the first page if paging is used.
4556 =head2 software_limit
4560 =item Value: (0 | 1)
4564 When combined with L</rows> and/or L</offset> the generated SQL will not
4565 include any limit dialect stanzas. Instead the entire result will be selected
4566 as if no limits were specified, and DBIC will perform the limit locally, by
4567 artificially advancing and finishing the resulting L</cursor>.
4569 This is the recommended way of performing resultset limiting when no sane RDBMS
4570 implementation is available (e.g.
4571 L<Sybase ASE|DBIx::Class::Storage::DBI::Sybase::ASE> using the
4572 L<Generic Sub Query|DBIx::Class::SQLMaker::LimitDialects/GenericSubQ> hack)
4578 =item Value: \@columns
4582 A arrayref of columns to group by. Can include columns of joined tables.
4584 group_by => [qw/ column1 column2 ... /]
4590 =item Value: $condition
4594 The HAVING operator specifies a B<secondary> condition applied to the set
4595 after the grouping calculations have been done. In other words it is a
4596 constraint just like L</where> (and accepting the same
4597 L<SQL::Abstract syntax|SQL::Abstract/WHERE CLAUSES>) applied to the data
4598 as it exists after GROUP BY has taken place. Specifying L</having> without
4599 L</group_by> is a logical mistake, and a fatal error on most RDBMS engines.
4603 having => { 'count_employee' => { '>=', 100 } }
4605 or with an in-place function in which case literal SQL is required:
4607 having => \[ 'count(employee) >= ?', 100 ]
4613 =item Value: (0 | 1)
4617 Set to 1 to automatically generate a L</group_by> clause based on the selection
4618 (including intelligent handling of L</order_by> contents). Note that the group
4619 criteria calculation takes place over the B<final> selection. This includes
4620 any L</+columns>, L</+select> or L</order_by> additions in subsequent
4621 L</search> calls, and standalone columns selected via
4622 L<DBIx::Class::ResultSetColumn> (L</get_column>). A notable exception are the
4623 extra selections specified via L</prefetch> - such selections are explicitly
4624 excluded from group criteria calculations.
4626 If the final ResultSet also explicitly defines a L</group_by> attribute, this
4627 setting is ignored and an appropriate warning is issued.
4631 Adds extra conditions to the resultset, combined with the preexisting C<WHERE>
4632 conditions, same as the B<first> argument to the L<search operator|/search>
4634 # only return rows WHERE deleted IS NULL for all searches
4635 __PACKAGE__->resultset_attributes({ where => { deleted => undef } });
4637 Note that the above example is
4638 L<strongly discouraged|DBIx::Class::ResultSource/resultset_attributes>.
4642 Set to 1 to cache search results. This prevents extra SQL queries if you
4643 revisit rows in your ResultSet:
4645 my $resultset = $schema->resultset('Artist')->search( undef, { cache => 1 } );
4647 while( my $artist = $resultset->next ) {
4651 $resultset->first; # without cache, this would issue a query
4653 By default, searches are not cached.
4655 For more examples of using these attributes, see
4656 L<DBIx::Class::Manual::Cookbook>.
4662 =item Value: ( 'update' | 'shared' | \$scalar )
4666 Set to 'update' for a SELECT ... FOR UPDATE or 'shared' for a SELECT
4667 ... FOR SHARED. If \$scalar is passed, this is taken directly and embedded in the
4672 DBIx::Class supports arbitrary related data prefetching from multiple related
4673 sources. Any combination of relationship types and column sets are supported.
4674 If L<collapsing|/collapse> is requested, there is an additional requirement of
4675 selecting enough data to make every individual object uniquely identifiable.
4677 Here are some more involved examples, based on the following relationship map:
4680 My::Schema::CD->belongs_to( artist => 'My::Schema::Artist' );
4681 My::Schema::CD->might_have( liner_note => 'My::Schema::LinerNotes' );
4682 My::Schema::CD->has_many( tracks => 'My::Schema::Track' );
4684 My::Schema::Artist->belongs_to( record_label => 'My::Schema::RecordLabel' );
4686 My::Schema::Track->has_many( guests => 'My::Schema::Guest' );
4690 my $rs = $schema->resultset('Tag')->search(
4699 The initial search results in SQL like the following:
4701 SELECT tag.*, cd.*, artist.* FROM tag
4702 JOIN cd ON tag.cd = cd.cdid
4703 JOIN artist ON cd.artist = artist.artistid
4705 L<DBIx::Class> has no need to go back to the database when we access the
4706 C<cd> or C<artist> relationships, which saves us two SQL statements in this
4709 Simple prefetches will be joined automatically, so there is no need
4710 for a C<join> attribute in the above search.
4712 The L</prefetch> attribute can be used with any of the relationship types
4713 and multiple prefetches can be specified together. Below is a more complex
4714 example that prefetches a CD's artist, its liner notes (if present),
4715 the cover image, the tracks on that CD, and the guests on those
4718 my $rs = $schema->resultset('CD')->search(
4722 { artist => 'record_label'}, # belongs_to => belongs_to
4723 'liner_note', # might_have
4724 'cover_image', # has_one
4725 { tracks => 'guests' }, # has_many => has_many
4730 This will produce SQL like the following:
4732 SELECT cd.*, artist.*, record_label.*, liner_note.*, cover_image.*,
4736 ON artist.artistid = me.artistid
4737 JOIN record_label record_label
4738 ON record_label.labelid = artist.labelid
4739 LEFT JOIN track tracks
4740 ON tracks.cdid = me.cdid
4741 LEFT JOIN guest guests
4742 ON guests.trackid = track.trackid
4743 LEFT JOIN liner_notes liner_note
4744 ON liner_note.cdid = me.cdid
4745 JOIN cd_artwork cover_image
4746 ON cover_image.cdid = me.cdid
4749 Now the C<artist>, C<record_label>, C<liner_note>, C<cover_image>,
4750 C<tracks>, and C<guests> of the CD will all be available through the
4751 relationship accessors without the need for additional queries to the
4756 Prefetch does a lot of deep magic. As such, it may not behave exactly
4757 as you might expect.
4763 Prefetch uses the L</cache> to populate the prefetched relationships. This
4764 may or may not be what you want.
4768 If you specify a condition on a prefetched relationship, ONLY those
4769 rows that match the prefetched condition will be fetched into that relationship.
4770 This means that adding prefetch to a search() B<may alter> what is returned by
4771 traversing a relationship. So, if you have C<< Artist->has_many(CDs) >> and you do
4773 my $artist_rs = $schema->resultset('Artist')->search({
4779 my $count = $artist_rs->first->cds->count;
4781 my $artist_rs_prefetch = $artist_rs->search( {}, { prefetch => 'cds' } );
4783 my $prefetch_count = $artist_rs_prefetch->first->cds->count;
4785 cmp_ok( $count, '==', $prefetch_count, "Counts should be the same" );
4787 That cmp_ok() may or may not pass depending on the datasets involved. In other
4788 words the C<WHERE> condition would apply to the entire dataset, just like
4789 it would in regular SQL. If you want to add a condition only to the "right side"
4790 of a C<LEFT JOIN> - consider declaring and using a L<relationship with a custom
4791 condition|DBIx::Class::Relationship::Base/condition>
4795 =head1 DBIC BIND VALUES
4797 Because DBIC may need more information to bind values than just the column name
4798 and value itself, it uses a special format for both passing and receiving bind
4799 values. Each bind value should be composed of an arrayref of
4800 C<< [ \%args => $val ] >>. The format of C<< \%args >> is currently:
4806 If present (in any form), this is what is being passed directly to bind_param.
4807 Note that different DBD's expect different bind args. (e.g. DBD::SQLite takes
4808 a single numerical type, while DBD::Pg takes a hashref if bind options.)
4810 If this is specified, all other bind options described below are ignored.
4814 If present, this is used to infer the actual bind attribute by passing to
4815 C<< $resolved_storage->bind_attribute_by_data_type() >>. Defaults to the
4816 "data_type" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>.
4818 Note that the data type is somewhat freeform (hence the sqlt_ prefix);
4819 currently drivers are expected to "Do the Right Thing" when given a common
4820 datatype name. (Not ideal, but that's what we got at this point.)
4824 Currently used to correctly allocate buffers for bind_param_inout().
4825 Defaults to "size" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>,
4826 or to a sensible value based on the "data_type".
4830 Used to fill in missing sqlt_datatype and sqlt_size attributes (if they are
4831 explicitly specified they are never overridden). Also used by some weird DBDs,
4832 where the column name should be available at bind_param time (e.g. Oracle).
4836 For backwards compatibility and convenience, the following shortcuts are
4839 [ $name => $val ] === [ { dbic_colname => $name }, $val ]
4840 [ \$dt => $val ] === [ { sqlt_datatype => $dt }, $val ]
4841 [ undef, $val ] === [ {}, $val ]
4842 $val === [ {}, $val ]
4844 =head1 FURTHER QUESTIONS?
4846 Check the list of L<additional DBIC resources|DBIx::Class/GETTING HELP/SUPPORT>.
4848 =head1 COPYRIGHT AND LICENSE
4850 This module is free software L<copyright|DBIx::Class/COPYRIGHT AND LICENSE>
4851 by the L<DBIx::Class (DBIC) authors|DBIx::Class/AUTHORS>. You can
4852 redistribute it and/or modify it under the same terms as the
4853 L<DBIx::Class library|DBIx::Class/COPYRIGHT AND LICENSE>.