1 package DBIx::Class::ResultSet;
5 use base qw/DBIx::Class/;
7 use DBIx::Class::ResultSetColumn;
8 use Scalar::Util qw/blessed weaken reftype/;
9 use DBIx::Class::_Util qw(
10 fail_on_internal_wantarray fail_on_internal_call UNRESOLVABLE_CONDITION
14 # not importing first() as it will clash with our own method
18 # De-duplication in _merge_attr() is disabled, but left in for reference
19 # (the merger is used for other things that ought not to be de-duped)
20 *__HM_DEDUP = sub () { 0 };
30 # this is real - CDBICompat overrides it with insanity
31 # yes, prototype won't matter, but that's for now ;)
34 __PACKAGE__->mk_group_accessors('simple' => qw/_result_class result_source/);
38 DBIx::Class::ResultSet - Represents a query used for fetching a set of results.
42 my $users_rs = $schema->resultset('User');
43 while( $user = $users_rs->next) {
44 print $user->username;
47 my $registered_users_rs = $schema->resultset('User')->search({ registered => 1 });
48 my @cds_in_2005 = $schema->resultset('CD')->search({ year => 2005 })->all();
52 A ResultSet is an object which stores a set of conditions representing
53 a query. It is the backbone of DBIx::Class (i.e. the really
54 important/useful bit).
56 No SQL is executed on the database when a ResultSet is created, it
57 just stores all the conditions needed to create the query.
59 A basic ResultSet representing the data of an entire table is returned
60 by calling C<resultset> on a L<DBIx::Class::Schema> and passing in a
61 L<Source|DBIx::Class::Manual::Glossary/Source> name.
63 my $users_rs = $schema->resultset('User');
65 A new ResultSet is returned from calling L</search> on an existing
66 ResultSet. The new one will contain all the conditions of the
67 original, plus any new conditions added in the C<search> call.
69 A ResultSet also incorporates an implicit iterator. L</next> and L</reset>
70 can be used to walk through all the L<DBIx::Class::Row>s the ResultSet
73 The query that the ResultSet represents is B<only> executed against
74 the database when these methods are called:
75 L</find>, L</next>, L</all>, L</first>, L</single>, L</count>.
77 If a resultset is used in a numeric context it returns the L</count>.
78 However, if it is used in a boolean context it is B<always> true. So if
79 you want to check if a resultset has any results, you must use C<if $rs
84 =head2 Chaining resultsets
86 Let's say you've got a query that needs to be run to return some data
87 to the user. But, you have an authorization system in place that
88 prevents certain users from seeing certain information. So, you want
89 to construct the basic query in one method, but add constraints to it in
94 my $request = $self->get_request; # Get a request object somehow.
95 my $schema = $self->result_source->schema;
97 my $cd_rs = $schema->resultset('CD')->search({
98 title => $request->param('title'),
99 year => $request->param('year'),
102 $cd_rs = $self->apply_security_policy( $cd_rs );
104 return $cd_rs->all();
107 sub apply_security_policy {
116 =head3 Resolving conditions and attributes
118 When a resultset is chained from another resultset (e.g.:
119 C<< my $new_rs = $old_rs->search(\%extra_cond, \%attrs) >>), conditions
120 and attributes with the same keys need resolving.
122 If any of L</columns>, L</select>, L</as> are present, they reset the
123 original selection, and start the selection "clean".
125 The L</join>, L</prefetch>, L</+columns>, L</+select>, L</+as> attributes
126 are merged into the existing ones from the original resultset.
128 The L</where> and L</having> attributes, and any search conditions, are
129 merged with an SQL C<AND> to the existing condition from the original
132 All other attributes are overridden by any new ones supplied in the
135 =head2 Multiple queries
137 Since a resultset just defines a query, you can do all sorts of
138 things with it with the same object.
140 # Don't hit the DB yet.
141 my $cd_rs = $schema->resultset('CD')->search({
142 title => 'something',
146 # Each of these hits the DB individually.
147 my $count = $cd_rs->count;
148 my $most_recent = $cd_rs->get_column('date_released')->max();
149 my @records = $cd_rs->all;
151 And it's not just limited to SELECT statements.
157 $cd_rs->create({ artist => 'Fred' });
159 Which is the same as:
161 $schema->resultset('CD')->create({
162 title => 'something',
167 See: L</search>, L</count>, L</get_column>, L</all>, L</create>.
169 =head2 Custom ResultSet classes
171 To add methods to your resultsets, you can subclass L<DBIx::Class::ResultSet>, similar to:
173 package MyApp::Schema::ResultSet::User;
178 use base 'DBIx::Class::ResultSet';
182 $self->search({ $self->current_source_alias . '.active' => 1 });
187 $self->search({ $self->current_source_alias . '.verified' => 0 });
190 sub created_n_days_ago {
191 my ($self, $days_ago) = @_;
193 $self->current_source_alias . '.create_date' => {
195 $self->result_source->schema->storage->datetime_parser->format_datetime(
196 DateTime->now( time_zone => 'UTC' )->subtract( days => $days_ago )
201 sub users_to_warn { shift->active->unverified->created_n_days_ago(7) }
205 See L<DBIx::Class::Schema/load_namespaces> on how DBIC can discover and
206 automatically attach L<Result|DBIx::Class::Manual::ResultClass>-specific
207 L<ResulSet|DBIx::Class::ResultSet> classes.
209 =head3 ResultSet subclassing with Moose and similar constructor-providers
211 Using L<Moose> or L<Moo> in your ResultSet classes is usually overkill, but
212 you may find it useful if your ResultSets contain a lot of business logic
213 (e.g. C<has xml_parser>, C<has json>, etc) or if you just prefer to organize
216 In order to write custom ResultSet classes with L<Moo> you need to use the
217 following template. The L<BUILDARGS|Moo/BUILDARGS> is necessary due to the
218 unusual signature of the L<constructor provided by DBIC
219 |DBIx::Class::ResultSet/new> C<< ->new($source, \%args) >>.
222 extends 'DBIx::Class::ResultSet';
223 sub BUILDARGS { $_[2] } # ::RS::new() expects my ($class, $rsrc, $args) = @_
229 If you want to build your custom ResultSet classes with L<Moose>, you need
230 a similar, though a little more elaborate template in order to interface the
231 inlining of the L<Moose>-provided
232 L<object constructor|Moose::Manual::Construction/WHERE'S THE CONSTRUCTOR?>,
235 package MyApp::Schema::ResultSet::User;
238 use MooseX::NonMoose;
239 extends 'DBIx::Class::ResultSet';
241 sub BUILDARGS { $_[2] } # ::RS::new() expects my ($class, $rsrc, $args) = @_
245 __PACKAGE__->meta->make_immutable;
249 The L<MooseX::NonMoose> is necessary so that the L<Moose> constructor does not
250 entirely overwrite the DBIC one (in contrast L<Moo> does this automatically).
251 Alternatively, you can skip L<MooseX::NonMoose> and get by with just L<Moose>
254 __PACKAGE__->meta->make_immutable(inline_constructor => 0);
262 =item Arguments: L<$source|DBIx::Class::ResultSource>, L<\%attrs?|/ATTRIBUTES>
264 =item Return Value: L<$resultset|/search>
268 The resultset constructor. Takes a source object (usually a
269 L<DBIx::Class::ResultSourceProxy::Table>) and an attribute hash (see
270 L</ATTRIBUTES> below). Does not perform any queries -- these are
271 executed as needed by the other methods.
273 Generally you never construct a resultset manually. Instead you get one
275 C<< $schema->L<resultset|DBIx::Class::Schema/resultset>('$source_name') >>
276 or C<< $another_resultset->L<search|/search>(...) >> (the later called in
279 my $rs = $schema->resultset('CD')->search({ title => '100th Window' });
285 If called on an object, proxies to L</new_result> instead, so
287 my $cd = $schema->resultset('CD')->new({ title => 'Spoon' });
289 will return a CD object, not a ResultSet, and is equivalent to:
291 my $cd = $schema->resultset('CD')->new_result({ title => 'Spoon' });
293 Please also keep in mind that many internals call L</new_result> directly,
294 so overloading this method with the idea of intercepting new result object
295 creation B<will not work>. See also warning pertaining to L</create>.
305 DBIx::Class::_ENV_::ASSERT_NO_INTERNAL_INDIRECT_CALLS and fail_on_internal_call;
306 return $class->new_result(@_);
309 my ($source, $attrs) = @_;
310 $source = $source->resolve
311 if $source->isa('DBIx::Class::ResultSourceHandle');
313 $attrs = { %{$attrs||{}} };
314 delete @{$attrs}{qw(_last_sqlmaker_alias_map _simple_passthrough_construction)};
316 if ($attrs->{page}) {
317 $attrs->{rows} ||= 10;
320 $attrs->{alias} ||= 'me';
323 result_source => $source,
324 cond => $attrs->{where},
329 # if there is a dark selector, this means we are already in a
330 # chain and the cleanup/sanification was taken care of by
332 $self->_normalize_selection($attrs)
333 unless $attrs->{_dark_selector};
336 $attrs->{result_class} || $source->result_class
346 =item Arguments: L<$cond|DBIx::Class::SQLMaker> | undef, L<\%attrs?|/ATTRIBUTES>
348 =item Return Value: $resultset (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
352 my @cds = $cd_rs->search({ year => 2001 }); # "... WHERE year = 2001"
353 my $new_rs = $cd_rs->search({ year => 2005 });
355 my $new_rs = $cd_rs->search([ { year => 2005 }, { year => 2004 } ]);
356 # year = 2005 OR year = 2004
358 In list context, C<< ->all() >> is called implicitly on the resultset, thus
359 returning a list of L<result|DBIx::Class::Manual::ResultClass> objects instead.
360 To avoid that, use L</search_rs>.
362 If you need to pass in additional attributes but no additional condition,
363 call it as C<search(undef, \%attrs)>.
365 # "SELECT name, artistid FROM $artist_table"
366 my @all_artists = $schema->resultset('Artist')->search(undef, {
367 columns => [qw/name artistid/],
370 For a list of attributes that can be passed to C<search>, see
371 L</ATTRIBUTES>. For more examples of using this function, see
372 L<Searching|DBIx::Class::Manual::Cookbook/SEARCHING>. For a complete
373 documentation for the first argument, see L<SQL::Abstract/"WHERE CLAUSES">
374 and its extension L<DBIx::Class::SQLMaker>.
376 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
380 Note that L</search> does not process/deflate any of the values passed in the
381 L<SQL::Abstract>-compatible search condition structure. This is unlike other
382 condition-bound methods L</new_result>, L</create> and L</find>. The user must ensure
383 manually that any value passed to this method will stringify to something the
384 RDBMS knows how to deal with. A notable example is the handling of L<DateTime>
385 objects, for more info see:
386 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
392 my $rs = $self->search_rs( @_ );
395 DBIx::Class::_ENV_::ASSERT_NO_INTERNAL_WANTARRAY and my $sog = fail_on_internal_wantarray;
398 elsif (defined wantarray) {
402 # we can be called by a relationship helper, which in
403 # turn may be called in void context due to some braindead
404 # overload or whatever else the user decided to be clever
405 # at this particular day. Thus limit the exception to
406 # external code calls only
407 $self->throw_exception ('->search is *not* a mutator, calling it in void context makes no sense')
408 if (caller)[0] !~ /^\QDBIx::Class::/;
418 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
420 =item Return Value: L<$resultset|/search>
424 This method does the same exact thing as search() except it will
425 always return a resultset, even in list context.
432 my $rsrc = $self->result_source;
433 my ($call_cond, $call_attrs);
435 # Special-case handling for (undef, undef) or (undef)
436 # Note that (foo => undef) is valid deprecated syntax
437 @_ = () if not scalar grep { defined $_ } @_;
443 # fish out attrs in the ($condref, $attr) case
444 elsif (@_ == 2 and ( ! defined $_[0] or (ref $_[0]) ne '') ) {
445 ($call_cond, $call_attrs) = @_;
448 $self->throw_exception('Odd number of arguments to search')
452 carp_unique 'search( %condition ) is deprecated, use search( \%condition ) instead'
453 unless $rsrc->result_class->isa('DBIx::Class::CDBICompat');
455 for my $i (0 .. $#_) {
457 $self->throw_exception ('All keys in condition key/value pairs must be plain scalars')
458 if (! defined $_[$i] or ref $_[$i] ne '');
464 # see if we can keep the cache (no $rs changes)
466 my %safe = (alias => 1, cache => 1);
467 if ( ! List::Util::first { !$safe{$_} } keys %$call_attrs and (
470 ref $call_cond eq 'HASH' && ! keys %$call_cond
472 ref $call_cond eq 'ARRAY' && ! @$call_cond
474 $cache = $self->get_cache;
477 my $old_attrs = { %{$self->{attrs}} };
478 my ($old_having, $old_where) = delete @{$old_attrs}{qw(having where)};
480 my $new_attrs = { %$old_attrs };
482 # take care of call attrs (only if anything is changing)
483 if ($call_attrs and keys %$call_attrs) {
485 # copy for _normalize_selection
486 $call_attrs = { %$call_attrs };
488 my @selector_attrs = qw/select as columns cols +select +as +columns include_columns/;
490 # reset the current selector list if new selectors are supplied
491 if (List::Util::first { exists $call_attrs->{$_} } qw/columns cols select as/) {
492 delete @{$old_attrs}{(@selector_attrs, '_dark_selector')};
495 # Normalize the new selector list (operates on the passed-in attr structure)
496 # Need to do it on every chain instead of only once on _resolved_attrs, in
497 # order to allow detection of empty vs partial 'as'
498 $call_attrs->{_dark_selector} = $old_attrs->{_dark_selector}
499 if $old_attrs->{_dark_selector};
500 $self->_normalize_selection ($call_attrs);
502 # start with blind overwriting merge, exclude selector attrs
503 $new_attrs = { %{$old_attrs}, %{$call_attrs} };
504 delete @{$new_attrs}{@selector_attrs};
506 for (@selector_attrs) {
507 $new_attrs->{$_} = $self->_merge_attr($old_attrs->{$_}, $call_attrs->{$_})
508 if ( exists $old_attrs->{$_} or exists $call_attrs->{$_} );
511 # older deprecated name, use only if {columns} is not there
512 if (my $c = delete $new_attrs->{cols}) {
513 carp_unique( "Resultset attribute 'cols' is deprecated, use 'columns' instead" );
514 if ($new_attrs->{columns}) {
515 carp "Resultset specifies both the 'columns' and the legacy 'cols' attributes - ignoring 'cols'";
518 $new_attrs->{columns} = $c;
523 # join/prefetch use their own crazy merging heuristics
524 foreach my $key (qw/join prefetch/) {
525 $new_attrs->{$key} = $self->_merge_joinpref_attr($old_attrs->{$key}, $call_attrs->{$key})
526 if exists $call_attrs->{$key};
529 # stack binds together
530 $new_attrs->{bind} = [ @{ $old_attrs->{bind} || [] }, @{ $call_attrs->{bind} || [] } ];
534 for ($old_where, $call_cond) {
536 $new_attrs->{where} = $self->_stack_cond (
537 $_, $new_attrs->{where}
542 if (defined $old_having) {
543 $new_attrs->{having} = $self->_stack_cond (
544 $old_having, $new_attrs->{having}
548 my $rs = (ref $self)->new($rsrc, $new_attrs);
550 $rs->set_cache($cache) if ($cache);
556 sub _normalize_selection {
557 my ($self, $attrs) = @_;
560 if ( exists $attrs->{include_columns} ) {
561 carp_unique( "Resultset attribute 'include_columns' is deprecated, use '+columns' instead" );
562 $attrs->{'+columns'} = $self->_merge_attr(
563 $attrs->{'+columns'}, delete $attrs->{include_columns}
567 # columns are always placed first, however
569 # Keep the X vs +X separation until _resolved_attrs time - this allows to
570 # delay the decision on whether to use a default select list ($rsrc->columns)
571 # allowing stuff like the remove_columns helper to work
573 # select/as +select/+as pairs need special handling - the amount of select/as
574 # elements in each pair does *not* have to be equal (think multicolumn
575 # selectors like distinct(foo, bar) ). If the selector is bare (no 'as'
576 # supplied at all) - try to infer the alias, either from the -as parameter
577 # of the selector spec, or use the parameter whole if it looks like a column
578 # name (ugly legacy heuristic). If all fails - leave the selector bare (which
579 # is ok as well), but make sure no more additions to the 'as' chain take place
580 for my $pref ('', '+') {
582 my ($sel, $as) = map {
583 my $key = "${pref}${_}";
585 my $val = [ ref $attrs->{$key} eq 'ARRAY'
587 : $attrs->{$key} || ()
589 delete $attrs->{$key};
593 if (! @$as and ! @$sel ) {
596 elsif (@$as and ! @$sel) {
597 $self->throw_exception(
598 "Unable to handle ${pref}as specification (@$as) without a corresponding ${pref}select"
602 # no as part supplied at all - try to deduce (unless explicit end of named selection is declared)
603 # if any @$as has been supplied we assume the user knows what (s)he is doing
604 # and blindly keep stacking up pieces
605 unless ($attrs->{_dark_selector}) {
608 if ( ref $_ eq 'HASH' and exists $_->{-as} ) {
609 push @$as, $_->{-as};
611 # assume any plain no-space, no-parenthesis string to be a column spec
612 # FIXME - this is retarded but is necessary to support shit like 'count(foo)'
613 elsif ( ! ref $_ and $_ =~ /^ [^\s\(\)]+ $/x) {
616 # if all else fails - raise a flag that no more aliasing will be allowed
618 $attrs->{_dark_selector} = {
620 string => ($dark_sel_dumper ||= do {
621 require Data::Dumper::Concise;
622 Data::Dumper::Concise::DumperObject()->Indent(0);
623 })->Values([$_])->Dump
631 elsif (@$as < @$sel) {
632 $self->throw_exception(
633 "Unable to handle an ${pref}as specification (@$as) with less elements than the corresponding ${pref}select"
636 elsif ($pref and $attrs->{_dark_selector}) {
637 $self->throw_exception(
638 "Unable to process named '+select', resultset contains an unnamed selector $attrs->{_dark_selector}{string}"
644 $attrs->{"${pref}select"} = $self->_merge_attr($attrs->{"${pref}select"}, $sel);
645 $attrs->{"${pref}as"} = $self->_merge_attr($attrs->{"${pref}as"}, $as);
650 my ($self, $left, $right) = @_;
653 (ref $_ eq 'ARRAY' and !@$_)
655 (ref $_ eq 'HASH' and ! keys %$_)
656 ) and $_ = undef for ($left, $right);
658 # either one of the two undef
659 if ( (defined $left) xor (defined $right) ) {
660 return defined $left ? $left : $right;
663 elsif ( ! defined $left ) {
667 return $self->result_source->schema->storage->_collapse_cond({ -and => [$left, $right] });
671 =head2 search_literal
673 B<CAVEAT>: C<search_literal> is provided for Class::DBI compatibility and
674 should only be used in that context. C<search_literal> is a convenience
675 method. It is equivalent to calling C<< $schema->search(\[]) >>, but if you
676 want to ensure columns are bound correctly, use L</search>.
678 See L<DBIx::Class::Manual::Cookbook/SEARCHING> and
679 L<DBIx::Class::Manual::FAQ/Searching> for searching techniques that do not
680 require C<search_literal>.
684 =item Arguments: $sql_fragment, @standalone_bind_values
686 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
690 my @cds = $cd_rs->search_literal('year = ? AND title = ?', qw/2001 Reload/);
691 my $newrs = $artist_rs->search_literal('name = ?', 'Metallica');
693 Pass a literal chunk of SQL to be added to the conditional part of the
696 Example of how to use C<search> instead of C<search_literal>
698 my @cds = $cd_rs->search_literal('cdid = ? AND (artist = ? OR artist = ?)', (2, 1, 2));
699 my @cds = $cd_rs->search(\[ 'cdid = ? AND (artist = ? OR artist = ?)', [ 'cdid', 2 ], [ 'artist', 1 ], [ 'artist', 2 ] ]);
704 my ($self, $sql, @bind) = @_;
706 if ( @bind && ref($bind[-1]) eq 'HASH' ) {
709 return $self->search(\[ $sql, map [ {} => $_ ], @bind ], ($attr || () ));
716 =item Arguments: \%columns_values | @pk_values, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
718 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
722 Finds and returns a single row based on supplied criteria. Takes either a
723 hashref with the same format as L</create> (including inference of foreign
724 keys from related objects), or a list of primary key values in the same
725 order as the L<primary columns|DBIx::Class::ResultSource/primary_columns>
726 declaration on the L</result_source>.
728 In either case an attempt is made to combine conditions already existing on
729 the resultset with the condition passed to this method.
731 To aid with preparing the correct query for the storage you may supply the
732 C<key> attribute, which is the name of a
733 L<unique constraint|DBIx::Class::ResultSource/add_unique_constraint> (the
734 unique constraint corresponding to the
735 L<primary columns|DBIx::Class::ResultSource/primary_columns> is always named
736 C<primary>). If the C<key> attribute has been supplied, and DBIC is unable
737 to construct a query that satisfies the named unique constraint fully (
738 non-NULL values for each column member of the constraint) an exception is
741 If no C<key> is specified, the search is carried over all unique constraints
742 which are fully defined by the available condition.
744 If no such constraint is found, C<find> currently defaults to a simple
745 C<< search->(\%column_values) >> which may or may not do what you expect.
746 Note that this fallback behavior may be deprecated in further versions. If
747 you need to search with arbitrary conditions - use L</search>. If the query
748 resulting from this fallback produces more than one row, a warning to the
749 effect is issued, though only the first row is constructed and returned as
752 In addition to C<key>, L</find> recognizes and applies standard
753 L<resultset attributes|/ATTRIBUTES> in the same way as L</search> does.
755 Note that if you have extra concerns about the correctness of the resulting
756 query you need to specify the C<key> attribute and supply the entire condition
757 as an argument to find (since it is not always possible to perform the
758 combination of the resultset condition with the supplied one, especially if
759 the resultset condition contains literal sql).
761 For example, to find a row by its primary key:
763 my $cd = $schema->resultset('CD')->find(5);
765 You can also find a row by a specific unique constraint:
767 my $cd = $schema->resultset('CD')->find(
769 artist => 'Massive Attack',
770 title => 'Mezzanine',
772 { key => 'cd_artist_title' }
775 See also L</find_or_create> and L</update_or_create>.
781 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
783 my $rsrc = $self->result_source;
786 if (exists $attrs->{key}) {
787 $constraint_name = defined $attrs->{key}
789 : $self->throw_exception("An undefined 'key' resultset attribute makes no sense")
793 # Parse out the condition from input
796 if (ref $_[0] eq 'HASH') {
797 $call_cond = { %{$_[0]} };
800 # if only values are supplied we need to default to 'primary'
801 $constraint_name = 'primary' unless defined $constraint_name;
803 my @c_cols = $rsrc->unique_constraint_columns($constraint_name);
805 $self->throw_exception(
806 "No constraint columns, maybe a malformed '$constraint_name' constraint?"
809 $self->throw_exception (
810 'find() expects either a column/value hashref, or a list of values '
811 . "corresponding to the columns of the specified unique constraint '$constraint_name'"
812 ) unless @c_cols == @_;
814 @{$call_cond}{@c_cols} = @_;
817 # process relationship data if any
818 for my $key (keys %$call_cond) {
820 length ref($call_cond->{$key})
822 my $relinfo = $rsrc->relationship_info($key)
824 # implicitly skip has_many's (likely MC)
825 (ref (my $val = delete $call_cond->{$key}) ne 'ARRAY' )
827 my ($rel_cond, $crosstable) = $rsrc->_resolve_condition(
828 $relinfo->{cond}, $val, $key, $key
831 $self->throw_exception("Complex condition via relationship '$key' is unsupported in find()")
832 if $crosstable or ref($rel_cond) ne 'HASH';
834 # supplement condition
835 # relationship conditions take precedence (?)
836 @{$call_cond}{keys %$rel_cond} = values %$rel_cond;
840 my $alias = exists $attrs->{alias} ? $attrs->{alias} : $self->{attrs}{alias};
842 if (defined $constraint_name) {
843 $final_cond = $self->_qualify_cond_columns (
845 $self->result_source->_minimal_valueset_satisfying_constraint(
846 constraint_name => $constraint_name,
847 values => ($self->_merge_with_rscond($call_cond))[0],
854 elsif ($self->{attrs}{accessor} and $self->{attrs}{accessor} eq 'single') {
855 # This means that we got here after a merger of relationship conditions
856 # in ::Relationship::Base::search_related (the row method), and furthermore
857 # the relationship is of the 'single' type. This means that the condition
858 # provided by the relationship (already attached to $self) is sufficient,
859 # as there can be only one row in the database that would satisfy the
863 my (@unique_queries, %seen_column_combinations, $ci, @fc_exceptions);
865 # no key was specified - fall down to heuristics mode:
866 # run through all unique queries registered on the resultset, and
867 # 'OR' all qualifying queries together
869 # always start from 'primary' if it exists at all
870 for my $c_name ( sort {
872 : $b eq 'primary' ? 1
874 } $rsrc->unique_constraint_names) {
876 next if $seen_column_combinations{
877 join "\x00", sort $rsrc->unique_constraint_columns($c_name)
881 push @unique_queries, $self->_qualify_cond_columns(
882 $self->result_source->_minimal_valueset_satisfying_constraint(
883 constraint_name => $c_name,
884 values => ($self->_merge_with_rscond($call_cond))[0],
885 columns_info => ($ci ||= $self->result_source->columns_info),
891 push @fc_exceptions, $_ if $_ =~ /\bFilterColumn\b/;
896 @unique_queries ? \@unique_queries
897 : @fc_exceptions ? $self->throw_exception(join "; ", map { $_ =~ /(.*) at .+ line \d+$/s } @fc_exceptions )
898 : $self->_non_unique_find_fallback ($call_cond, $attrs)
902 # Run the query, passing the result_class since it should propagate for find
903 my $rs = $self->search ($final_cond, {result_class => $self->result_class, %$attrs});
904 if ($rs->_resolved_attrs->{collapse}) {
906 carp "Query returned more than one row" if $rs->next;
914 # This is a stop-gap method as agreed during the discussion on find() cleanup:
915 # http://lists.scsys.co.uk/pipermail/dbix-class/2010-October/009535.html
917 # It is invoked when find() is called in legacy-mode with insufficiently-unique
918 # condition. It is provided for overrides until a saner way forward is devised
920 # *NOTE* This is not a public method, and it's *GUARANTEED* to disappear down
921 # the road. Please adjust your tests accordingly to catch this situation early
922 # DBIx::Class::ResultSet->can('_non_unique_find_fallback') is reasonable
924 # The method will not be removed without an adequately complete replacement
925 # for strict-mode enforcement
926 sub _non_unique_find_fallback {
927 my ($self, $cond, $attrs) = @_;
929 return $self->_qualify_cond_columns(
931 exists $attrs->{alias}
933 : $self->{attrs}{alias}
938 sub _qualify_cond_columns {
939 my ($self, $cond, $alias) = @_;
941 my %aliased = %$cond;
942 for (keys %aliased) {
943 $aliased{"$alias.$_"} = delete $aliased{$_}
950 sub _build_unique_cond {
952 '_build_unique_cond is a private method, and moreover is about to go '
953 . 'away. Please contact the development team at %s if you believe you '
954 . 'have a genuine use for this method, in order to discuss alternatives.',
955 DBIx::Class::_ENV_::HELP_URL,
958 my ($self, $constraint_name, $cond, $croak_on_null) = @_;
960 $self->result_source->_minimal_valueset_satisfying_constraint(
961 constraint_name => $constraint_name,
963 carp_on_nulls => !$croak_on_null
967 =head2 search_related
971 =item Arguments: $rel_name, $cond?, L<\%attrs?|/ATTRIBUTES>
973 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
977 $new_rs = $cd_rs->search_related('artist', {
981 Searches the specified relationship, optionally specifying a condition and
982 attributes for matching records. See L</ATTRIBUTES> for more information.
984 In list context, C<< ->all() >> is called implicitly on the resultset, thus
985 returning a list of result objects instead. To avoid that, use L</search_related_rs>.
987 See also L</search_related_rs>.
992 return shift->related_resultset(shift)->search(@_);
995 =head2 search_related_rs
997 This method works exactly the same as search_related, except that
998 it guarantees a resultset, even in list context.
1002 sub search_related_rs {
1003 return shift->related_resultset(shift)->search_rs(@_);
1010 =item Arguments: none
1012 =item Return Value: L<$cursor|DBIx::Class::Cursor>
1016 Returns a storage-driven cursor to the given resultset. See
1017 L<DBIx::Class::Cursor> for more information.
1024 return $self->{cursor} ||= do {
1025 my $attrs = $self->_resolved_attrs;
1026 $self->result_source->storage->select(
1027 $attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs
1036 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1038 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1042 my $cd = $schema->resultset('CD')->single({ year => 2001 });
1044 Inflates the first result without creating a cursor if the resultset has
1045 any records in it; if not returns C<undef>. Used by L</find> as a lean version
1048 While this method can take an optional search condition (just like L</search>)
1049 being a fast-code-path it does not recognize search attributes. If you need to
1050 add extra joins or similar, call L</search> and then chain-call L</single> on the
1051 L<DBIx::Class::ResultSet> returned.
1057 As of 0.08100, this method enforces the assumption that the preceding
1058 query returns only one row. If more than one row is returned, you will receive
1061 Query returned more than one row
1063 In this case, you should be using L</next> or L</find> instead, or if you really
1064 know what you are doing, use the L</rows> attribute to explicitly limit the size
1067 This method will also throw an exception if it is called on a resultset prefetching
1068 has_many, as such a prefetch implies fetching multiple rows from the database in
1069 order to assemble the resulting object.
1076 my ($self, $where) = @_;
1078 $self->throw_exception('single() only takes search conditions, no attributes. You want ->search( $cond, $attrs )->single()');
1081 my $attrs = { %{$self->_resolved_attrs} };
1083 $self->throw_exception(
1084 'single() can not be used on resultsets collapsing a has_many. Use find( \%cond ) or next() instead'
1085 ) if $attrs->{collapse};
1088 if (defined $attrs->{where}) {
1091 [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ }
1092 $where, delete $attrs->{where} ]
1095 $attrs->{where} = $where;
1099 my $data = [ $self->result_source->storage->select_single(
1100 $attrs->{from}, $attrs->{select},
1101 $attrs->{where}, $attrs
1104 return undef unless @$data;
1105 $self->{_stashed_rows} = [ $data ];
1106 $self->_construct_results->[0];
1113 =item Arguments: L<$cond?|DBIx::Class::SQLMaker>
1115 =item Return Value: L<$resultsetcolumn|DBIx::Class::ResultSetColumn>
1119 my $max_length = $rs->get_column('length')->max;
1121 Returns a L<DBIx::Class::ResultSetColumn> instance for a column of the ResultSet.
1126 my ($self, $column) = @_;
1127 my $new = DBIx::Class::ResultSetColumn->new($self, $column);
1135 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1137 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1141 # WHERE title LIKE '%blue%'
1142 $cd_rs = $rs->search_like({ title => '%blue%'});
1144 Performs a search, but uses C<LIKE> instead of C<=> as the condition. Note
1145 that this is simply a convenience method retained for ex Class::DBI users.
1146 You most likely want to use L</search> with specific operators.
1148 For more information, see L<DBIx::Class::Manual::Cookbook>.
1150 This method is deprecated and will be removed in 0.09. Use L</search()>
1151 instead. An example conversion is:
1153 ->search_like({ foo => 'bar' });
1157 ->search({ foo => { like => 'bar' } });
1164 'search_like() is deprecated and will be removed in DBIC version 0.09.'
1165 .' Instead use ->search({ x => { -like => "y%" } })'
1166 .' (note the outer pair of {}s - they are important!)'
1168 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
1169 my $query = ref $_[0] eq 'HASH' ? { %{shift()} }: {@_};
1170 $query->{$_} = { 'like' => $query->{$_} } for keys %$query;
1171 return $class->search($query, { %$attrs });
1178 =item Arguments: $first, $last
1180 =item Return Value: L<$resultset|/search> (scalar context) | L<@result_objs|DBIx::Class::Manual::ResultClass> (list context)
1184 Returns a resultset or object list representing a subset of elements from the
1185 resultset slice is called on. Indexes are from 0, i.e., to get the first
1186 three records, call:
1188 my ($one, $two, $three) = $rs->slice(0, 2);
1193 my ($self, $min, $max) = @_;
1194 my $attrs = {}; # = { %{ $self->{attrs} || {} } };
1195 $attrs->{offset} = $self->{attrs}{offset} || 0;
1196 $attrs->{offset} += $min;
1197 $attrs->{rows} = ($max ? ($max - $min + 1) : 1);
1198 return $self->search(undef, $attrs);
1205 =item Arguments: none
1207 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1211 Returns the next element in the resultset (C<undef> is there is none).
1213 Can be used to efficiently iterate over records in the resultset:
1215 my $rs = $schema->resultset('CD')->search;
1216 while (my $cd = $rs->next) {
1220 Note that you need to store the resultset object, and call C<next> on it.
1221 Calling C<< resultset('Table')->next >> repeatedly will always return the
1222 first record from the resultset.
1229 if (my $cache = $self->get_cache) {
1230 $self->{all_cache_position} ||= 0;
1231 return $cache->[$self->{all_cache_position}++];
1234 if ($self->{attrs}{cache}) {
1235 delete $self->{pager};
1236 $self->{all_cache_position} = 1;
1237 return ($self->all)[0];
1240 return shift(@{$self->{_stashed_results}}) if @{ $self->{_stashed_results}||[] };
1242 $self->{_stashed_results} = $self->_construct_results
1245 return shift @{$self->{_stashed_results}};
1248 # Constructs as many results as it can in one pass while respecting
1249 # cursor laziness. Several modes of operation:
1251 # * Always builds everything present in @{$self->{_stashed_rows}}
1252 # * If called with $fetch_all true - pulls everything off the cursor and
1253 # builds all result structures (or objects) in one pass
1254 # * If $self->_resolved_attrs->{collapse} is true, checks the order_by
1255 # and if the resultset is ordered properly by the left side:
1256 # * Fetches stuff off the cursor until the "master object" changes,
1257 # and saves the last extra row (if any) in @{$self->{_stashed_rows}}
1259 # * Just fetches, and collapses/constructs everything as if $fetch_all
1260 # was requested (there is no other way to collapse except for an
1262 # * If no collapse is requested - just get the next row, construct and
1264 sub _construct_results {
1265 my ($self, $fetch_all) = @_;
1267 my $rsrc = $self->result_source;
1268 my $attrs = $self->_resolved_attrs;
1273 ! $attrs->{order_by}
1277 my @pcols = $rsrc->primary_columns
1279 # default order for collapsing unless the user asked for something
1280 $attrs->{order_by} = [ map { join '.', $attrs->{alias}, $_} @pcols ];
1281 $attrs->{_ordered_for_collapse} = 1;
1282 $attrs->{_order_is_artificial} = 1;
1285 # this will be used as both initial raw-row collector AND as a RV of
1286 # _construct_results. Not regrowing the array twice matters a lot...
1287 # a surprising amount actually
1288 my $rows = delete $self->{_stashed_rows};
1290 my $cursor; # we may not need one at all
1292 my $did_fetch_all = $fetch_all;
1295 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1296 $rows = [ ($rows ? @$rows : ()), $self->cursor->all ];
1298 elsif( $attrs->{collapse} ) {
1300 # a cursor will need to be closed over in case of collapse
1301 $cursor = $self->cursor;
1303 $attrs->{_ordered_for_collapse} = (
1309 ->_extract_colinfo_of_stable_main_source_order_by_portion($attrs)
1311 ) unless defined $attrs->{_ordered_for_collapse};
1313 if (! $attrs->{_ordered_for_collapse}) {
1316 # instead of looping over ->next, use ->all in stealth mode
1317 # *without* calling a ->reset afterwards
1318 # FIXME ENCAPSULATION - encapsulation breach, cursor method additions pending
1319 if (! $cursor->{_done}) {
1320 $rows = [ ($rows ? @$rows : ()), $cursor->all ];
1321 $cursor->{_done} = 1;
1326 if (! $did_fetch_all and ! @{$rows||[]} ) {
1327 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1328 $cursor ||= $self->cursor;
1329 if (scalar (my @r = $cursor->next) ) {
1334 return undef unless @{$rows||[]};
1336 # sanity check - people are too clever for their own good
1337 if ($attrs->{collapse} and my $aliastypes = $attrs->{_last_sqlmaker_alias_map} ) {
1339 my $multiplied_selectors;
1340 for my $sel_alias ( grep { $_ ne $attrs->{alias} } keys %{ $aliastypes->{selecting} } ) {
1342 $aliastypes->{multiplying}{$sel_alias}
1344 $aliastypes->{premultiplied}{$sel_alias}
1346 $multiplied_selectors->{$_} = 1 for values %{$aliastypes->{selecting}{$sel_alias}{-seen_columns}}
1350 for my $i (0 .. $#{$attrs->{as}} ) {
1351 my $sel = $attrs->{select}[$i];
1353 if (ref $sel eq 'SCALAR') {
1356 elsif( ref $sel eq 'REF' and ref $$sel eq 'ARRAY' ) {
1360 $self->throw_exception(
1361 'Result collapse not possible - selection from a has_many source redirected to the main object'
1362 ) if ($multiplied_selectors->{$sel} and $attrs->{as}[$i] !~ /\./);
1366 # hotspot - skip the setter
1367 my $res_class = $self->_result_class;
1369 my $inflator_cref = $self->{_result_inflator}{cref} ||= do {
1370 $res_class->can ('inflate_result')
1371 or $self->throw_exception("Inflator $res_class does not provide an inflate_result() method");
1374 my $infmap = $attrs->{as};
1376 $self->{_result_inflator}{is_core_row} = ( (
1379 ( \&DBIx::Class::Row::inflate_result || die "No ::Row::inflate_result() - can't happen" )
1380 ) ? 1 : 0 ) unless defined $self->{_result_inflator}{is_core_row};
1382 $self->{_result_inflator}{is_hri} = ( (
1383 ! $self->{_result_inflator}{is_core_row}
1386 require DBIx::Class::ResultClass::HashRefInflator
1388 DBIx::Class::ResultClass::HashRefInflator->can('inflate_result')
1390 ) ? 1 : 0 ) unless defined $self->{_result_inflator}{is_hri};
1393 if ($attrs->{_simple_passthrough_construction}) {
1394 # construct a much simpler array->hash folder for the one-table HRI cases right here
1395 if ($self->{_result_inflator}{is_hri}) {
1396 for my $r (@$rows) {
1397 $r = { map { $infmap->[$_] => $r->[$_] } 0..$#$infmap };
1400 # FIXME SUBOPTIMAL this is a very very very hot spot
1401 # while rather optimal we can *still* do much better, by
1402 # building a smarter Row::inflate_result(), and
1403 # switch to feeding it data via a much leaner interface
1405 # crude unscientific benchmarking indicated the shortcut eval is not worth it for
1406 # this particular resultset size
1407 elsif ( $self->{_result_inflator}{is_core_row} and @$rows < 60 ) {
1408 for my $r (@$rows) {
1409 $r = $inflator_cref->($res_class, $rsrc, { map { $infmap->[$_] => $r->[$_] } (0..$#$infmap) } );
1414 ( $self->{_result_inflator}{is_core_row}
1415 ? '$_ = $inflator_cref->($res_class, $rsrc, { %s }) for @$rows'
1416 # a custom inflator may be a multiplier/reductor - put it in direct list ctx
1417 : '@$rows = map { $inflator_cref->($res_class, $rsrc, { %s } ) } @$rows'
1419 ( join (', ', map { "\$infmap->[$_] => \$_->[$_]" } 0..$#$infmap ) )
1425 $self->{_result_inflator}{is_hri} ? 'hri'
1426 : $self->{_result_inflator}{is_core_row} ? 'classic_pruning'
1427 : 'classic_nonpruning'
1430 # $args and $attrs to _mk_row_parser are separated to delineate what is
1431 # core collapser stuff and what is dbic $rs specific
1432 @{$self->{_row_parser}{$parser_type}}{qw(cref nullcheck)} = $rsrc->_mk_row_parser({
1434 inflate_map => $infmap,
1435 collapse => $attrs->{collapse},
1436 premultiplied => $attrs->{_main_source_premultiplied},
1437 hri_style => $self->{_result_inflator}{is_hri},
1438 prune_null_branches => $self->{_result_inflator}{is_hri} || $self->{_result_inflator}{is_core_row},
1439 }, $attrs) unless $self->{_row_parser}{$parser_type}{cref};
1441 # column_info metadata historically hasn't been too reliable.
1442 # We need to start fixing this somehow (the collapse resolver
1443 # can't work without it). Add an explicit check for the *main*
1444 # result, hopefully this will gradually weed out such errors
1446 # FIXME - this is a temporary kludge that reduces performance
1447 # It is however necessary for the time being
1448 my ($unrolled_non_null_cols_to_check, $err);
1450 if (my $check_non_null_cols = $self->{_row_parser}{$parser_type}{nullcheck} ) {
1453 'Collapse aborted due to invalid ResultSource metadata - the following '
1454 . 'selections are declared non-nullable but NULLs were retrieved: '
1458 COL: for my $i (@$check_non_null_cols) {
1459 ! defined $_->[$i] and push @violating_idx, $i and next COL for @$rows;
1462 $self->throw_exception( $err . join (', ', map { "'$infmap->[$_]'" } @violating_idx ) )
1465 $unrolled_non_null_cols_to_check = join (',', @$check_non_null_cols);
1469 ($did_fetch_all or ! $attrs->{collapse}) ? undef
1470 : defined $unrolled_non_null_cols_to_check ? eval sprintf <<'EOS', $unrolled_non_null_cols_to_check
1472 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1473 my @r = $cursor->next or return;
1474 if (my @violating_idx = grep { ! defined $r[$_] } (%s) ) {
1475 $self->throw_exception( $err . join (', ', map { "'$infmap->[$_]'" } @violating_idx ) )
1481 # FIXME SUBOPTIMAL - we can do better, cursor->next/all (well diff. methods) should return a ref
1482 my @r = $cursor->next or return;
1487 $self->{_row_parser}{$parser_type}{cref}->(
1489 $next_cref ? ( $next_cref, $self->{_stashed_rows} = [] ) : (),
1492 # simple in-place substitution, does not regrow $rows
1493 if ($self->{_result_inflator}{is_core_row}) {
1494 $_ = $inflator_cref->($res_class, $rsrc, @$_) for @$rows
1496 # Special-case multi-object HRI - there is no $inflator_cref pass at all
1497 elsif ( ! $self->{_result_inflator}{is_hri} ) {
1498 # the inflator may be a multiplier/reductor - put it in list ctx
1499 @$rows = map { $inflator_cref->($res_class, $rsrc, @$_) } @$rows;
1503 # The @$rows check seems odd at first - why wouldn't we want to warn
1504 # regardless? The issue is things like find() etc, where the user
1505 # *knows* only one result will come back. In these cases the ->all
1506 # is not a pessimization, but rather something we actually want
1508 'Unable to properly collapse has_many results in iterator mode due '
1509 . 'to order criteria - performed an eager cursor slurp underneath. '
1510 . 'Consider using ->all() instead'
1511 ) if ( ! $fetch_all and @$rows > 1 );
1516 =head2 result_source
1520 =item Arguments: L<$result_source?|DBIx::Class::ResultSource>
1522 =item Return Value: L<$result_source|DBIx::Class::ResultSource>
1526 An accessor for the primary ResultSource object from which this ResultSet
1533 =item Arguments: $result_class?
1535 =item Return Value: $result_class
1539 An accessor for the class to use when creating result objects. Defaults to
1540 C<< result_source->result_class >> - which in most cases is the name of the
1541 L<"table"|DBIx::Class::Manual::Glossary/"ResultSource"> class.
1543 Note that changing the result_class will also remove any components
1544 that were originally loaded in the source class via
1545 L<DBIx::Class::ResultSource/load_components>. Any overloaded methods
1546 in the original source class will not run.
1551 my ($self, $result_class) = @_;
1552 if ($result_class) {
1554 # don't fire this for an object
1555 $self->ensure_class_loaded($result_class)
1556 unless ref($result_class);
1558 if ($self->get_cache) {
1559 carp_unique('Changing the result_class of a ResultSet instance with cached results is a noop - the cache contents will not be altered');
1561 # FIXME ENCAPSULATION - encapsulation breach, cursor method additions pending
1562 elsif ($self->{cursor} && $self->{cursor}{_pos}) {
1563 $self->throw_exception('Changing the result_class of a ResultSet instance with an active cursor is not supported');
1566 $self->_result_class($result_class);
1568 delete $self->{_result_inflator};
1570 $self->_result_class;
1577 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1579 =item Return Value: $count
1583 Performs an SQL C<COUNT> with the same query as the resultset was built
1584 with to find the number of elements. Passing arguments is equivalent to
1585 C<< $rs->search ($cond, \%attrs)->count >>
1591 return $self->search(@_)->count if @_ and defined $_[0];
1592 return scalar @{ $self->get_cache } if $self->get_cache;
1594 my $attrs = { %{ $self->_resolved_attrs } };
1596 # this is a little optimization - it is faster to do the limit
1597 # adjustments in software, instead of a subquery
1598 my ($rows, $offset) = delete @{$attrs}{qw/rows offset/};
1601 if ($self->_has_resolved_attr (qw/collapse group_by/)) {
1602 $crs = $self->_count_subq_rs ($attrs);
1605 $crs = $self->_count_rs ($attrs);
1607 my $count = $crs->next;
1609 $count -= $offset if $offset;
1610 $count = $rows if $rows and $rows < $count;
1611 $count = 0 if ($count < 0);
1620 =item Arguments: L<$cond|DBIx::Class::SQLMaker>, L<\%attrs?|/ATTRIBUTES>
1622 =item Return Value: L<$count_rs|DBIx::Class::ResultSetColumn>
1626 Same as L</count> but returns a L<DBIx::Class::ResultSetColumn> object.
1627 This can be very handy for subqueries:
1629 ->search( { amount => $some_rs->count_rs->as_query } )
1631 As with regular resultsets the SQL query will be executed only after
1632 the resultset is accessed via L</next> or L</all>. That would return
1633 the same single value obtainable via L</count>.
1639 return $self->search(@_)->count_rs if @_;
1641 # this may look like a lack of abstraction (count() does about the same)
1642 # but in fact an _rs *must* use a subquery for the limits, as the
1643 # software based limiting can not be ported if this $rs is to be used
1644 # in a subquery itself (i.e. ->as_query)
1645 if ($self->_has_resolved_attr (qw/collapse group_by offset rows/)) {
1646 return $self->_count_subq_rs($self->{_attrs});
1649 return $self->_count_rs($self->{_attrs});
1654 # returns a ResultSetColumn object tied to the count query
1657 my ($self, $attrs) = @_;
1659 my $rsrc = $self->result_source;
1661 my $tmp_attrs = { %$attrs };
1662 # take off any limits, record_filter is cdbi, and no point of ordering nor locking a count
1663 delete @{$tmp_attrs}{qw/rows offset order_by record_filter for/};
1665 # overwrite the selector (supplied by the storage)
1666 $rsrc->resultset_class->new($rsrc, {
1668 select => $rsrc->storage->_count_select ($rsrc, $attrs),
1670 })->get_column ('count');
1674 # same as above but uses a subquery
1676 sub _count_subq_rs {
1677 my ($self, $attrs) = @_;
1679 my $rsrc = $self->result_source;
1681 my $sub_attrs = { %$attrs };
1682 # extra selectors do not go in the subquery and there is no point of ordering it, nor locking it
1683 delete @{$sub_attrs}{qw/collapse columns as select order_by for/};
1685 # if we multi-prefetch we group_by something unique, as this is what we would
1686 # get out of the rs via ->next/->all. We *DO WANT* to clobber old group_by regardless
1687 if ( $attrs->{collapse} ) {
1688 $sub_attrs->{group_by} = [ map { "$attrs->{alias}.$_" } @{
1689 $rsrc->_identifying_column_set || $self->throw_exception(
1690 'Unable to construct a unique group_by criteria properly collapsing the '
1691 . 'has_many prefetch before count()'
1696 # Calculate subquery selector
1697 if (my $g = $sub_attrs->{group_by}) {
1699 my $sql_maker = $rsrc->storage->sql_maker;
1701 # necessary as the group_by may refer to aliased functions
1703 for my $sel (@{$attrs->{select}}) {
1704 $sel_index->{$sel->{-as}} = $sel
1705 if (ref $sel eq 'HASH' and $sel->{-as});
1708 # anything from the original select mentioned on the group-by needs to make it to the inner selector
1709 # also look for named aggregates referred in the having clause
1710 # having often contains scalarrefs - thus parse it out entirely
1712 if ($attrs->{having}) {
1713 local $sql_maker->{having_bind};
1714 local $sql_maker->{quote_char} = $sql_maker->{quote_char};
1715 local $sql_maker->{name_sep} = $sql_maker->{name_sep};
1716 unless (defined $sql_maker->{quote_char} and length $sql_maker->{quote_char}) {
1717 $sql_maker->{quote_char} = [ "\x00", "\xFF" ];
1718 # if we don't unset it we screw up retarded but unfortunately working
1719 # 'MAX(foo.bar)' => { '>', 3 }
1720 $sql_maker->{name_sep} = '';
1723 my ($lquote, $rquote, $sep) = map { quotemeta $_ } ($sql_maker->_quote_chars, $sql_maker->name_sep);
1725 my $having_sql = $sql_maker->_parse_rs_attrs ({ having => $attrs->{having} });
1728 # search for both a proper quoted qualified string, for a naive unquoted scalarref
1729 # and if all fails for an utterly naive quoted scalar-with-function
1730 while ($having_sql =~ /
1731 $rquote $sep $lquote (.+?) $rquote
1733 [\s,] \w+ \. (\w+) [\s,]
1735 [\s,] $lquote (.+?) $rquote [\s,]
1737 my $part = $1 || $2 || $3; # one of them matched if we got here
1738 unless ($seen_having{$part}++) {
1745 my $colpiece = $sel_index->{$_} || $_;
1747 # unqualify join-based group_by's. Arcane but possible query
1748 # also horrible horrible hack to alias a column (not a func.)
1749 # (probably need to introduce SQLA syntax)
1750 if ($colpiece =~ /\./ && $colpiece !~ /^$attrs->{alias}\./) {
1753 $colpiece = \ sprintf ('%s AS %s', map { $sql_maker->_quote ($_) } ($colpiece, $as) );
1755 push @{$sub_attrs->{select}}, $colpiece;
1759 my @pcols = map { "$attrs->{alias}.$_" } ($rsrc->primary_columns);
1760 $sub_attrs->{select} = @pcols ? \@pcols : [ 1 ];
1763 return $rsrc->resultset_class
1764 ->new ($rsrc, $sub_attrs)
1766 ->search ({}, { columns => { count => $rsrc->storage->_count_select ($rsrc, $attrs) } })
1767 ->get_column ('count');
1771 =head2 count_literal
1773 B<CAVEAT>: C<count_literal> is provided for Class::DBI compatibility and
1774 should only be used in that context. See L</search_literal> for further info.
1778 =item Arguments: $sql_fragment, @standalone_bind_values
1780 =item Return Value: $count
1784 Counts the results in a literal query. Equivalent to calling L</search_literal>
1785 with the passed arguments, then L</count>.
1789 sub count_literal { shift->search_literal(@_)->count; }
1795 =item Arguments: none
1797 =item Return Value: L<@result_objs|DBIx::Class::Manual::ResultClass>
1801 Returns all elements in the resultset.
1808 $self->throw_exception("all() doesn't take any arguments, you probably wanted ->search(...)->all()");
1811 delete @{$self}{qw/_stashed_rows _stashed_results/};
1813 if (my $c = $self->get_cache) {
1817 $self->cursor->reset;
1819 my $objs = $self->_construct_results('fetch_all') || [];
1821 $self->set_cache($objs) if $self->{attrs}{cache};
1830 =item Arguments: none
1832 =item Return Value: $self
1836 Resets the resultset's cursor, so you can iterate through the elements again.
1837 Implicitly resets the storage cursor, so a subsequent L</next> will trigger
1845 delete @{$self}{qw/_stashed_rows _stashed_results/};
1846 $self->{all_cache_position} = 0;
1847 $self->cursor->reset;
1855 =item Arguments: none
1857 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass> | undef
1861 L<Resets|/reset> the resultset (causing a fresh query to storage) and returns
1862 an object for the first result (or C<undef> if the resultset is empty).
1867 return $_[0]->reset->next;
1873 # Determines whether and what type of subquery is required for the $rs operation.
1874 # If grouping is necessary either supplies its own, or verifies the current one
1875 # After all is done delegates to the proper storage method.
1877 sub _rs_update_delete {
1878 my ($self, $op, $values) = @_;
1880 my $rsrc = $self->result_source;
1881 my $storage = $rsrc->schema->storage;
1883 my $attrs = { %{$self->_resolved_attrs} };
1885 my $join_classifications;
1886 my ($existing_group_by) = delete @{$attrs}{qw(group_by _grouped_by_distinct)};
1888 # do we need a subquery for any reason?
1890 defined $existing_group_by
1892 # if {from} is unparseable wrap a subq
1893 ref($attrs->{from}) ne 'ARRAY'
1895 # limits call for a subq
1896 $self->_has_resolved_attr(qw/rows offset/)
1899 # simplify the joinmap, so we can further decide if a subq is necessary
1900 if (!$needs_subq and @{$attrs->{from}} > 1) {
1902 ($attrs->{from}, $join_classifications) =
1903 $storage->_prune_unused_joins ($attrs);
1905 # any non-pruneable non-local restricting joins imply subq
1906 $needs_subq = defined List::Util::first { $_ ne $attrs->{alias} } keys %{ $join_classifications->{restricting} || {} };
1909 # check if the head is composite (by now all joins are thrown out unless $needs_subq)
1911 (ref $attrs->{from}[0]) ne 'HASH'
1913 ref $attrs->{from}[0]{ $attrs->{from}[0]{-alias} }
1917 # do we need anything like a subquery?
1918 if (! $needs_subq) {
1919 # Most databases do not allow aliasing of tables in UPDATE/DELETE. Thus
1920 # a condition containing 'me' or other table prefixes will not work
1921 # at all. Tell SQLMaker to dequalify idents via a gross hack.
1923 my $sqla = $rsrc->storage->sql_maker;
1924 local $sqla->{_dequalify_idents} = 1;
1925 \[ $sqla->_recurse_where($self->{cond}) ];
1929 # we got this far - means it is time to wrap a subquery
1930 my $idcols = $rsrc->_identifying_column_set || $self->throw_exception(
1932 "Unable to perform complex resultset %s() without an identifying set of columns on source '%s'",
1938 # make a new $rs selecting only the PKs (that's all we really need for the subq)
1939 delete $attrs->{$_} for qw/select as collapse/;
1940 $attrs->{columns} = [ map { "$attrs->{alias}.$_" } @$idcols ];
1942 # this will be consumed by the pruner waaaaay down the stack
1943 $attrs->{_force_prune_multiplying_joins} = 1;
1945 my $subrs = (ref $self)->new($rsrc, $attrs);
1947 if (@$idcols == 1) {
1948 $cond = { $idcols->[0] => { -in => $subrs->as_query } };
1950 elsif ($storage->_use_multicolumn_in) {
1951 # no syntax for calling this properly yet
1952 # !!! EXPERIMENTAL API !!! WILL CHANGE !!!
1953 $cond = $storage->sql_maker->_where_op_multicolumn_in (
1954 $idcols, # how do I convey a list of idents...? can binds reside on lhs?
1959 # if all else fails - get all primary keys and operate over a ORed set
1960 # wrap in a transaction for consistency
1961 # this is where the group_by/multiplication starts to matter
1965 # we do not need to check pre-multipliers, since if the premulti is there, its
1966 # parent (who is multi) will be there too
1967 keys %{ $join_classifications->{multiplying} || {} }
1969 # make sure if there is a supplied group_by it matches the columns compiled above
1970 # perfectly. Anything else can not be sanely executed on most databases so croak
1971 # right then and there
1972 if ($existing_group_by) {
1973 my @current_group_by = map
1974 { $_ =~ /\./ ? $_ : "$attrs->{alias}.$_" }
1979 join ("\x00", sort @current_group_by)
1981 join ("\x00", sort @{$attrs->{columns}} )
1983 $self->throw_exception (
1984 "You have just attempted a $op operation on a resultset which does group_by"
1985 . ' on columns other than the primary keys, while DBIC internally needs to retrieve'
1986 . ' the primary keys in a subselect. All sane RDBMS engines do not support this'
1987 . ' kind of queries. Please retry the operation with a modified group_by or'
1988 . ' without using one at all.'
1993 $subrs = $subrs->search({}, { group_by => $attrs->{columns} });
1996 $guard = $storage->txn_scope_guard;
1998 for my $row ($subrs->cursor->all) {
2000 { $idcols->[$_] => $row->[$_] }
2007 my $res = $cond ? $storage->$op (
2009 $op eq 'update' ? $values : (),
2013 $guard->commit if $guard;
2022 =item Arguments: \%values
2024 =item Return Value: $underlying_storage_rv
2028 Sets the specified columns in the resultset to the supplied values in a
2029 single query. Note that this will not run any accessor/set_column/update
2030 triggers, nor will it update any result object instances derived from this
2031 resultset (this includes the contents of the L<resultset cache|/set_cache>
2032 if any). See L</update_all> if you need to execute any on-update
2033 triggers or cascades defined either by you or a
2034 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
2036 The return value is a pass through of what the underlying
2037 storage backend returned, and may vary. See L<DBI/execute> for the most
2042 Note that L</update> does not process/deflate any of the values passed in.
2043 This is unlike the corresponding L<DBIx::Class::Row/update>. The user must
2044 ensure manually that any value passed to this method will stringify to
2045 something the RDBMS knows how to deal with. A notable example is the
2046 handling of L<DateTime> objects, for more info see:
2047 L<DBIx::Class::Manual::Cookbook/Formatting DateTime objects in queries>.
2052 my ($self, $values) = @_;
2053 $self->throw_exception('Values for update must be a hash')
2054 unless ref $values eq 'HASH';
2056 return $self->_rs_update_delete ('update', $values);
2063 =item Arguments: \%values
2065 =item Return Value: 1
2069 Fetches all objects and updates them one at a time via
2070 L<DBIx::Class::Row/update>. Note that C<update_all> will run DBIC defined
2071 triggers, while L</update> will not.
2076 my ($self, $values) = @_;
2077 $self->throw_exception('Values for update_all must be a hash')
2078 unless ref $values eq 'HASH';
2080 my $guard = $self->result_source->schema->txn_scope_guard;
2081 $_->update({%$values}) for $self->all; # shallow copy - update will mangle it
2090 =item Arguments: none
2092 =item Return Value: $underlying_storage_rv
2096 Deletes the rows matching this resultset in a single query. Note that this
2097 will not run any delete triggers, nor will it alter the
2098 L<in_storage|DBIx::Class::Row/in_storage> status of any result object instances
2099 derived from this resultset (this includes the contents of the
2100 L<resultset cache|/set_cache> if any). See L</delete_all> if you need to
2101 execute any on-delete triggers or cascades defined either by you or a
2102 L<result component|DBIx::Class::Manual::Component/WHAT IS A COMPONENT>.
2104 The return value is a pass through of what the underlying storage backend
2105 returned, and may vary. See L<DBI/execute> for the most common case.
2111 $self->throw_exception('delete does not accept any arguments')
2114 return $self->_rs_update_delete ('delete');
2121 =item Arguments: none
2123 =item Return Value: 1
2127 Fetches all objects and deletes them one at a time via
2128 L<DBIx::Class::Row/delete>. Note that C<delete_all> will run DBIC defined
2129 triggers, while L</delete> will not.
2135 $self->throw_exception('delete_all does not accept any arguments')
2138 my $guard = $self->result_source->schema->txn_scope_guard;
2139 $_->delete for $self->all;
2148 =item Arguments: [ \@column_list, \@row_values+ ] | [ \%col_data+ ]
2150 =item Return Value: L<\@result_objects|DBIx::Class::Manual::ResultClass> (scalar context) | L<@result_objects|DBIx::Class::Manual::ResultClass> (list context)
2154 Accepts either an arrayref of hashrefs or alternatively an arrayref of
2161 The context of this method call has an important effect on what is
2162 submitted to storage. In void context data is fed directly to fastpath
2163 insertion routines provided by the underlying storage (most often
2164 L<DBI/execute_for_fetch>), bypassing the L<new|DBIx::Class::Row/new> and
2165 L<insert|DBIx::Class::Row/insert> calls on the
2166 L<Result|DBIx::Class::Manual::ResultClass> class, including any
2167 augmentation of these methods provided by components. For example if you
2168 are using something like L<DBIx::Class::UUIDColumns> to create primary
2169 keys for you, you will find that your PKs are empty. In this case you
2170 will have to explicitly force scalar or list context in order to create
2175 In non-void (scalar or list) context, this method is simply a wrapper
2176 for L</create>. Depending on list or scalar context either a list of
2177 L<Result|DBIx::Class::Manual::ResultClass> objects or an arrayref
2178 containing these objects is returned.
2180 When supplying data in "arrayref of arrayrefs" invocation style, the
2181 first element should be a list of column names and each subsequent
2182 element should be a data value in the earlier specified column order.
2185 $schema->resultset("Artist")->populate([
2186 [ qw( artistid name ) ],
2187 [ 100, 'A Formally Unknown Singer' ],
2188 [ 101, 'A singer that jumped the shark two albums ago' ],
2189 [ 102, 'An actually cool singer' ],
2192 For the arrayref of hashrefs style each hashref should be a structure
2193 suitable for passing to L</create>. Multi-create is also permitted with
2196 $schema->resultset("Artist")->populate([
2197 { artistid => 4, name => 'Manufactured Crap', cds => [
2198 { title => 'My First CD', year => 2006 },
2199 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2202 { artistid => 5, name => 'Angsty-Whiny Girl', cds => [
2203 { title => 'My parents sold me to a record company', year => 2005 },
2204 { title => 'Why Am I So Ugly?', year => 2006 },
2205 { title => 'I Got Surgery and am now Popular', year => 2007 }
2210 If you attempt a void-context multi-create as in the example above (each
2211 Artist also has the related list of CDs), and B<do not> supply the
2212 necessary autoinc foreign key information, this method will proxy to the
2213 less efficient L</create>, and then throw the Result objects away. In this
2214 case there are obviously no benefits to using this method over L</create>.
2221 # this is naive and just a quick check
2222 # the types will need to be checked more thoroughly when the
2223 # multi-source populate gets added
2225 ref $_[0] eq 'ARRAY'
2227 ( @{$_[0]} or return )
2229 ( ref $_[0][0] eq 'HASH' or ref $_[0][0] eq 'ARRAY' )
2232 ) or $self->throw_exception('Populate expects an arrayref of hashrefs or arrayref of arrayrefs');
2234 # FIXME - no cref handling
2235 # At this point assume either hashes or arrays
2237 if(defined wantarray) {
2238 my (@results, $guard);
2240 if (ref $data->[0] eq 'ARRAY') {
2241 # column names only, nothing to do
2242 return if @$data == 1;
2244 $guard = $self->result_source->schema->storage->txn_scope_guard
2248 { my $vals = $_; $self->new_result({ map { $data->[0][$_] => $vals->[$_] } 0..$#{$data->[0]} })->insert }
2249 @{$data}[1 .. $#$data]
2254 $guard = $self->result_source->schema->storage->txn_scope_guard
2257 @results = map { $self->new_result($_)->insert } @$data;
2260 $guard->commit if $guard;
2261 return wantarray ? @results : \@results;
2264 # we have to deal with *possibly incomplete* related data
2265 # this means we have to walk the data structure twice
2266 # whether we want this or not
2267 # jnap, I hate you ;)
2268 my $rsrc = $self->result_source;
2269 my $rel_info = { map { $_ => $rsrc->relationship_info($_) } $rsrc->relationships };
2271 my ($colinfo, $colnames, $slices_with_rels);
2275 for my $i (0 .. $#$data) {
2277 my $current_slice_seen_rel_infos;
2279 ### Determine/Supplement collists
2280 ### BEWARE - This is a hot piece of code, a lot of weird idioms were used
2281 if( ref $data->[$i] eq 'ARRAY' ) {
2283 # positional(!) explicit column list
2285 # column names only, nothing to do
2286 return if @$data == 1;
2288 $colinfo->{$data->[0][$_]} = { pos => $_, name => $data->[0][$_] } and push @$colnames, $data->[0][$_]
2289 for 0 .. $#{$data->[0]};
2296 for (values %$colinfo) {
2297 if ($_->{is_rel} ||= (
2298 $rel_info->{$_->{name}}
2301 ref $data->[$i][$_->{pos}] eq 'ARRAY'
2303 ref $data->[$i][$_->{pos}] eq 'HASH'
2305 ( defined blessed $data->[$i][$_->{pos}] and $data->[$i][$_->{pos}]->isa('DBIx::Class::Row') )
2311 # moar sanity check... sigh
2312 for ( ref $data->[$i][$_->{pos}] eq 'ARRAY' ? @{$data->[$i][$_->{pos}]} : $data->[$i][$_->{pos}] ) {
2313 if ( defined blessed $_ and $_->isa('DBIx::Class::Row' ) ) {
2314 carp_unique("Fast-path populate() with supplied related objects is not possible - falling back to regular create()");
2315 return my $throwaway = $self->populate(@_);
2319 push @$current_slice_seen_rel_infos, $rel_info->{$_->{name}};
2324 if ($current_slice_seen_rel_infos) {
2325 push @$slices_with_rels, { map { $colnames->[$_] => $data->[$i][$_] } 0 .. $#$colnames };
2327 # this is needed further down to decide whether or not to fallback to create()
2328 $colinfo->{$colnames->[$_]}{seen_null} ||= ! defined $data->[$i][$_]
2329 for 0 .. $#$colnames;
2332 elsif( ref $data->[$i] eq 'HASH' ) {
2334 for ( sort keys %{$data->[$i]} ) {
2336 $colinfo->{$_} ||= do {
2338 $self->throw_exception("Column '$_' must be present in supplied explicit column list")
2339 if $data_start; # it will be 0 on AoH, 1 on AoA
2341 push @$colnames, $_;
2344 { pos => $#$colnames, name => $_ }
2347 if ($colinfo->{$_}{is_rel} ||= (
2351 ref $data->[$i]{$_} eq 'ARRAY'
2353 ref $data->[$i]{$_} eq 'HASH'
2355 ( defined blessed $data->[$i]{$_} and $data->[$i]{$_}->isa('DBIx::Class::Row') )
2361 # moar sanity check... sigh
2362 for ( ref $data->[$i]{$_} eq 'ARRAY' ? @{$data->[$i]{$_}} : $data->[$i]{$_} ) {
2363 if ( defined blessed $_ and $_->isa('DBIx::Class::Row' ) ) {
2364 carp_unique("Fast-path populate() with supplied related objects is not possible - falling back to regular create()");
2365 return my $throwaway = $self->populate(@_);
2369 push @$current_slice_seen_rel_infos, $rel_info->{$_};
2373 if ($current_slice_seen_rel_infos) {
2374 push @$slices_with_rels, $data->[$i];
2376 # this is needed further down to decide whether or not to fallback to create()
2377 $colinfo->{$_}{seen_null} ||= ! defined $data->[$i]{$_}
2378 for keys %{$data->[$i]};
2382 $self->throw_exception('Unexpected populate() data structure member type: ' . ref $data->[$i] );
2386 { $_->{attrs}{is_depends_on} }
2387 @{ $current_slice_seen_rel_infos || [] }
2389 carp_unique("Fast-path populate() of belongs_to relationship data is not possible - falling back to regular create()");
2390 return my $throwaway = $self->populate(@_);
2394 if( $slices_with_rels ) {
2396 # need to exclude the rel "columns"
2397 $colnames = [ grep { ! $colinfo->{$_}{is_rel} } @$colnames ];
2399 # extra sanity check - ensure the main source is in fact identifiable
2400 # the localizing of nullability is insane, but oh well... the use-case is legit
2401 my $ci = $rsrc->columns_info($colnames);
2403 $ci->{$_} = { %{$ci->{$_}}, is_nullable => 0 }
2404 for grep { ! $colinfo->{$_}{seen_null} } keys %$ci;
2406 unless( $rsrc->_identifying_column_set($ci) ) {
2407 carp_unique("Fast-path populate() of non-uniquely identifiable rows with related data is not possible - falling back to regular create()");
2408 return my $throwaway = $self->populate(@_);
2412 ### inherit the data locked in the conditions of the resultset
2413 my ($rs_data) = $self->_merge_with_rscond({});
2414 delete @{$rs_data}{@$colnames}; # passed-in stuff takes precedence
2416 # if anything left - decompose rs_data
2418 if (keys %$rs_data) {
2419 push @$rs_data_vals, $rs_data->{$_}
2420 for sort keys %$rs_data;
2425 $guard = $rsrc->schema->storage->txn_scope_guard
2426 if $slices_with_rels;
2428 ### main source data
2429 # FIXME - need to switch entirely to a coderef-based thing,
2430 # so that large sets aren't copied several times... I think
2431 $rsrc->storage->_insert_bulk(
2433 [ @$colnames, sort keys %$rs_data ],
2435 ref $data->[$_] eq 'ARRAY'
2437 $slices_with_rels ? [ @{$data->[$_]}[0..$#$colnames], @{$rs_data_vals||[]} ] # the collist changed
2438 : $rs_data_vals ? [ @{$data->[$_]}, @$rs_data_vals ]
2441 : [ @{$data->[$_]}{@$colnames}, @{$rs_data_vals||[]} ]
2442 } $data_start .. $#$data ],
2445 ### do the children relationships
2446 if ( $slices_with_rels ) {
2447 my @rels = grep { $colinfo->{$_}{is_rel} } keys %$colinfo
2448 or die 'wtf... please report a bug with DBIC_TRACE=1 output (stacktrace)';
2450 for my $sl (@$slices_with_rels) {
2452 my ($main_proto, $main_proto_rs);
2453 for my $rel (@rels) {
2454 next unless defined $sl->{$rel};
2458 (map { $_ => $sl->{$_} } @$colnames),
2461 unless (defined $colinfo->{$rel}{rs}) {
2463 $colinfo->{$rel}{rs} = $rsrc->related_source($rel)->resultset;
2465 $colinfo->{$rel}{fk_map} = { reverse %{ $rsrc->_resolve_relationship_condition(
2467 self_alias => "\xFE", # irrelevant
2468 foreign_alias => "\xFF", # irrelevant
2469 )->{identity_map} || {} } };
2473 $colinfo->{$rel}{rs}->search({ map # only so that we inherit them values properly, no actual search
2476 ( $main_proto_rs ||= $rsrc->resultset->search($main_proto) )
2477 ->get_column( $colinfo->{$rel}{fk_map}{$_} )
2481 keys %{$colinfo->{$rel}{fk_map}}
2482 })->populate( ref $sl->{$rel} eq 'ARRAY' ? $sl->{$rel} : [ $sl->{$rel} ] );
2489 $guard->commit if $guard;
2496 =item Arguments: none
2498 =item Return Value: L<$pager|Data::Page>
2502 Returns a L<Data::Page> object for the current resultset. Only makes
2503 sense for queries with a C<page> attribute.
2505 To get the full count of entries for a paged resultset, call
2506 C<total_entries> on the L<Data::Page> object.
2513 return $self->{pager} if $self->{pager};
2515 my $attrs = $self->{attrs};
2516 if (!defined $attrs->{page}) {
2517 $self->throw_exception("Can't create pager for non-paged rs");
2519 elsif ($attrs->{page} <= 0) {
2520 $self->throw_exception('Invalid page number (page-numbers are 1-based)');
2522 $attrs->{rows} ||= 10;
2524 # throw away the paging flags and re-run the count (possibly
2525 # with a subselect) to get the real total count
2526 my $count_attrs = { %$attrs };
2527 delete @{$count_attrs}{qw/rows offset page pager/};
2529 my $total_rs = (ref $self)->new($self->result_source, $count_attrs);
2531 require DBIx::Class::ResultSet::Pager;
2532 return $self->{pager} = DBIx::Class::ResultSet::Pager->new(
2533 sub { $total_rs->count }, #lazy-get the total
2535 $self->{attrs}{page},
2543 =item Arguments: $page_number
2545 =item Return Value: L<$resultset|/search>
2549 Returns a resultset for the $page_number page of the resultset on which page
2550 is called, where each page contains a number of rows equal to the 'rows'
2551 attribute set on the resultset (10 by default).
2556 my ($self, $page) = @_;
2557 return (ref $self)->new($self->result_source, { %{$self->{attrs}}, page => $page });
2564 =item Arguments: \%col_data
2566 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2570 Creates a new result object in the resultset's result class and returns
2571 it. The row is not inserted into the database at this point, call
2572 L<DBIx::Class::Row/insert> to do that. Calling L<DBIx::Class::Row/in_storage>
2573 will tell you whether the result object has been inserted or not.
2575 Passes the hashref of input on to L<DBIx::Class::Row/new>.
2580 my ($self, $values) = @_;
2582 $self->throw_exception( "new_result takes only one argument - a hashref of values" )
2585 $self->throw_exception( "Result object instantiation requires a hashref as argument" )
2586 unless (ref $values eq 'HASH');
2588 my ($merged_cond, $cols_from_relations) = $self->_merge_with_rscond($values);
2590 my $new = $self->result_class->new({
2592 ( @$cols_from_relations
2593 ? (-cols_from_relations => $cols_from_relations)
2596 -result_source => $self->result_source, # DO NOT REMOVE THIS, REQUIRED
2600 reftype($new) eq 'HASH'
2606 carp_unique (sprintf (
2607 "%s->new returned a blessed empty hashref - a strong indicator something is wrong with its inheritance chain",
2608 $self->result_class,
2615 # _merge_with_rscond
2617 # Takes a simple hash of K/V data and returns its copy merged with the
2618 # condition already present on the resultset. Additionally returns an
2619 # arrayref of value/condition names, which were inferred from related
2620 # objects (this is needed for in-memory related objects)
2621 sub _merge_with_rscond {
2622 my ($self, $data) = @_;
2624 my ($implied_data, @cols_from_relations);
2626 my $alias = $self->{attrs}{alias};
2628 if (! defined $self->{cond}) {
2629 # just massage $data below
2631 elsif ($self->{cond} eq UNRESOLVABLE_CONDITION) {
2632 $implied_data = $self->{attrs}{related_objects}; # nothing might have been inserted yet
2633 @cols_from_relations = keys %{ $implied_data || {} };
2636 my $eqs = $self->result_source->schema->storage->_extract_fixed_condition_columns($self->{cond}, 'consider_nulls');
2637 $implied_data = { map {
2638 ( ($eqs->{$_}||'') eq UNRESOLVABLE_CONDITION ) ? () : ( $_ => $eqs->{$_} )
2644 { %{ $self->_remove_alias($_, $alias) } }
2645 # precedence must be given to passed values over values inherited from
2646 # the cond, so the order here is important.
2647 ( $implied_data||(), $data)
2649 \@cols_from_relations
2653 # _has_resolved_attr
2655 # determines if the resultset defines at least one
2656 # of the attributes supplied
2658 # used to determine if a subquery is necessary
2660 # supports some virtual attributes:
2662 # This will scan for any joins being present on the resultset.
2663 # It is not a mere key-search but a deep inspection of {from}
2666 sub _has_resolved_attr {
2667 my ($self, @attr_names) = @_;
2669 my $attrs = $self->_resolved_attrs;
2673 for my $n (@attr_names) {
2674 if (grep { $n eq $_ } (qw/-join/) ) {
2675 $extra_checks{$n}++;
2679 my $attr = $attrs->{$n};
2681 next if not defined $attr;
2683 if (ref $attr eq 'HASH') {
2684 return 1 if keys %$attr;
2686 elsif (ref $attr eq 'ARRAY') {
2694 # a resolved join is expressed as a multi-level from
2696 $extra_checks{-join}
2698 ref $attrs->{from} eq 'ARRAY'
2700 @{$attrs->{from}} > 1
2708 # Remove the specified alias from the specified query hash. A copy is made so
2709 # the original query is not modified.
2712 my ($self, $query, $alias) = @_;
2714 my %orig = %{ $query || {} };
2717 foreach my $key (keys %orig) {
2719 $unaliased{$key} = $orig{$key};
2722 $unaliased{$1} = $orig{$key}
2723 if $key =~ m/^(?:\Q$alias\E\.)?([^.]+)$/;
2733 =item Arguments: none
2735 =item Return Value: \[ $sql, L<@bind_values|/DBIC BIND VALUES> ]
2739 Returns the SQL query and bind vars associated with the invocant.
2741 This is generally used as the RHS for a subquery.
2748 my $attrs = { %{ $self->_resolved_attrs } };
2750 my $aq = $self->result_source->storage->_select_args_to_query (
2751 $attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs
2761 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2763 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2767 my $artist = $schema->resultset('Artist')->find_or_new(
2768 { artist => 'fred' }, { key => 'artists' });
2770 $cd->cd_to_producer->find_or_new({ producer => $producer },
2771 { key => 'primary' });
2773 Find an existing record from this resultset using L</find>. if none exists,
2774 instantiate a new result object and return it. The object will not be saved
2775 into your storage until you call L<DBIx::Class::Row/insert> on it.
2777 You most likely want this method when looking for existing rows using a unique
2778 constraint that is not the primary key, or looking for related rows.
2780 If you want objects to be saved immediately, use L</find_or_create> instead.
2782 B<Note>: Make sure to read the documentation of L</find> and understand the
2783 significance of the C<key> attribute, as its lack may skew your search, and
2784 subsequently result in spurious new objects.
2786 B<Note>: Take care when using C<find_or_new> with a table having
2787 columns with default values that you intend to be automatically
2788 supplied by the database (e.g. an auto_increment primary key column).
2789 In normal usage, the value of such columns should NOT be included at
2790 all in the call to C<find_or_new>, even when set to C<undef>.
2796 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2797 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2798 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2801 return $self->new_result($hash);
2808 =item Arguments: \%col_data
2810 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2814 Attempt to create a single new row or a row with multiple related rows
2815 in the table represented by the resultset (and related tables). This
2816 will not check for duplicate rows before inserting, use
2817 L</find_or_create> to do that.
2819 To create one row for this resultset, pass a hashref of key/value
2820 pairs representing the columns of the table and the values you wish to
2821 store. If the appropriate relationships are set up, foreign key fields
2822 can also be passed an object representing the foreign row, and the
2823 value will be set to its primary key.
2825 To create related objects, pass a hashref of related-object column values
2826 B<keyed on the relationship name>. If the relationship is of type C<multi>
2827 (L<DBIx::Class::Relationship/has_many>) - pass an arrayref of hashrefs.
2828 The process will correctly identify columns holding foreign keys, and will
2829 transparently populate them from the keys of the corresponding relation.
2830 This can be applied recursively, and will work correctly for a structure
2831 with an arbitrary depth and width, as long as the relationships actually
2832 exists and the correct column data has been supplied.
2834 Instead of hashrefs of plain related data (key/value pairs), you may
2835 also pass new or inserted objects. New objects (not inserted yet, see
2836 L</new_result>), will be inserted into their appropriate tables.
2838 Effectively a shortcut for C<< ->new_result(\%col_data)->insert >>.
2840 Example of creating a new row.
2842 $person_rs->create({
2843 name=>"Some Person",
2844 email=>"somebody@someplace.com"
2847 Example of creating a new row and also creating rows in a related C<has_many>
2848 or C<has_one> resultset. Note Arrayref.
2851 { artistid => 4, name => 'Manufactured Crap', cds => [
2852 { title => 'My First CD', year => 2006 },
2853 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2858 Example of creating a new row and also creating a row in a related
2859 C<belongs_to> resultset. Note Hashref.
2862 title=>"Music for Silly Walks",
2865 name=>"Silly Musician",
2873 When subclassing ResultSet never attempt to override this method. Since
2874 it is a simple shortcut for C<< $self->new_result($attrs)->insert >>, a
2875 lot of the internals simply never call it, so your override will be
2876 bypassed more often than not. Override either L<DBIx::Class::Row/new>
2877 or L<DBIx::Class::Row/insert> depending on how early in the
2878 L</create> process you need to intervene. See also warning pertaining to
2886 #my ($self, $col_data) = @_;
2887 DBIx::Class::_ENV_::ASSERT_NO_INTERNAL_INDIRECT_CALLS and fail_on_internal_call;
2888 return shift->new_result(shift)->insert;
2891 =head2 find_or_create
2895 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2897 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2901 $cd->cd_to_producer->find_or_create({ producer => $producer },
2902 { key => 'primary' });
2904 Tries to find a record based on its primary key or unique constraints; if none
2905 is found, creates one and returns that instead.
2907 my $cd = $schema->resultset('CD')->find_or_create({
2909 artist => 'Massive Attack',
2910 title => 'Mezzanine',
2914 Also takes an optional C<key> attribute, to search by a specific key or unique
2915 constraint. For example:
2917 my $cd = $schema->resultset('CD')->find_or_create(
2919 artist => 'Massive Attack',
2920 title => 'Mezzanine',
2922 { key => 'cd_artist_title' }
2925 B<Note>: Make sure to read the documentation of L</find> and understand the
2926 significance of the C<key> attribute, as its lack may skew your search, and
2927 subsequently result in spurious row creation.
2929 B<Note>: Because find_or_create() reads from the database and then
2930 possibly inserts based on the result, this method is subject to a race
2931 condition. Another process could create a record in the table after
2932 the find has completed and before the create has started. To avoid
2933 this problem, use find_or_create() inside a transaction.
2935 B<Note>: Take care when using C<find_or_create> with a table having
2936 columns with default values that you intend to be automatically
2937 supplied by the database (e.g. an auto_increment primary key column).
2938 In normal usage, the value of such columns should NOT be included at
2939 all in the call to C<find_or_create>, even when set to C<undef>.
2941 See also L</find> and L</update_or_create>. For information on how to declare
2942 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2944 If you need to know if an existing row was found or a new one created use
2945 L</find_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
2946 to call L<DBIx::Class::Row/insert> to save the newly created row to the
2949 my $cd = $schema->resultset('CD')->find_or_new({
2951 artist => 'Massive Attack',
2952 title => 'Mezzanine',
2956 if( !$cd->in_storage ) {
2963 sub find_or_create {
2965 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2966 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2967 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2970 return $self->new_result($hash)->insert;
2973 =head2 update_or_create
2977 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
2979 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
2983 $resultset->update_or_create({ col => $val, ... });
2985 Like L</find_or_create>, but if a row is found it is immediately updated via
2986 C<< $found_row->update (\%col_data) >>.
2989 Takes an optional C<key> attribute to search on a specific unique constraint.
2992 # In your application
2993 my $cd = $schema->resultset('CD')->update_or_create(
2995 artist => 'Massive Attack',
2996 title => 'Mezzanine',
2999 { key => 'cd_artist_title' }
3002 $cd->cd_to_producer->update_or_create({
3003 producer => $producer,
3009 B<Note>: Make sure to read the documentation of L</find> and understand the
3010 significance of the C<key> attribute, as its lack may skew your search, and
3011 subsequently result in spurious row creation.
3013 B<Note>: Take care when using C<update_or_create> with a table having
3014 columns with default values that you intend to be automatically
3015 supplied by the database (e.g. an auto_increment primary key column).
3016 In normal usage, the value of such columns should NOT be included at
3017 all in the call to C<update_or_create>, even when set to C<undef>.
3019 See also L</find> and L</find_or_create>. For information on how to declare
3020 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
3022 If you need to know if an existing row was updated or a new one created use
3023 L</update_or_new> and L<DBIx::Class::Row/in_storage> instead. Don't forget
3024 to call L<DBIx::Class::Row/insert> to save the newly created row to the
3029 sub update_or_create {
3031 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
3032 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
3034 my $row = $self->find($cond, $attrs);
3036 $row->update($cond);
3040 return $self->new_result($cond)->insert;
3043 =head2 update_or_new
3047 =item Arguments: \%col_data, { key => $unique_constraint, L<%attrs|/ATTRIBUTES> }?
3049 =item Return Value: L<$result|DBIx::Class::Manual::ResultClass>
3053 $resultset->update_or_new({ col => $val, ... });
3055 Like L</find_or_new> but if a row is found it is immediately updated via
3056 C<< $found_row->update (\%col_data) >>.
3060 # In your application
3061 my $cd = $schema->resultset('CD')->update_or_new(
3063 artist => 'Massive Attack',
3064 title => 'Mezzanine',
3067 { key => 'cd_artist_title' }
3070 if ($cd->in_storage) {
3071 # the cd was updated
3074 # the cd is not yet in the database, let's insert it
3078 B<Note>: Make sure to read the documentation of L</find> and understand the
3079 significance of the C<key> attribute, as its lack may skew your search, and
3080 subsequently result in spurious new objects.
3082 B<Note>: Take care when using C<update_or_new> with a table having
3083 columns with default values that you intend to be automatically
3084 supplied by the database (e.g. an auto_increment primary key column).
3085 In normal usage, the value of such columns should NOT be included at
3086 all in the call to C<update_or_new>, even when set to C<undef>.
3088 See also L</find>, L</find_or_create> and L</find_or_new>.
3094 my $attrs = ( @_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {} );
3095 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
3097 my $row = $self->find( $cond, $attrs );
3098 if ( defined $row ) {
3099 $row->update($cond);
3103 return $self->new_result($cond);
3110 =item Arguments: none
3112 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass> | undef
3116 Gets the contents of the cache for the resultset, if the cache is set.
3118 The cache is populated either by using the L</prefetch> attribute to
3119 L</search> or by calling L</set_cache>.
3131 =item Arguments: L<\@result_objs|DBIx::Class::Manual::ResultClass>
3133 =item Return Value: L<\@result_objs|DBIx::Class::Manual::ResultClass>
3137 Sets the contents of the cache for the resultset. Expects an arrayref
3138 of objects of the same class as those produced by the resultset. Note that
3139 if the cache is set, the resultset will return the cached objects rather
3140 than re-querying the database even if the cache attr is not set.
3142 The contents of the cache can also be populated by using the
3143 L</prefetch> attribute to L</search>.
3148 my ( $self, $data ) = @_;
3149 $self->throw_exception("set_cache requires an arrayref")
3150 if defined($data) && (ref $data ne 'ARRAY');
3151 $self->{all_cache} = $data;
3158 =item Arguments: none
3160 =item Return Value: undef
3164 Clears the cache for the resultset.
3169 shift->set_cache(undef);
3176 =item Arguments: none
3178 =item Return Value: true, if the resultset has been paginated
3186 return !!$self->{attrs}{page};
3193 =item Arguments: none
3195 =item Return Value: true, if the resultset has been ordered with C<order_by>.
3203 return scalar $self->result_source->storage->_extract_order_criteria($self->{attrs}{order_by});
3206 =head2 related_resultset
3210 =item Arguments: $rel_name
3212 =item Return Value: L<$resultset|/search>
3216 Returns a related resultset for the supplied relationship name.
3218 $artist_rs = $schema->resultset('CD')->related_resultset('Artist');
3222 sub related_resultset {
3223 my ($self, $rel) = @_;
3225 return $self->{related_resultsets}{$rel}
3226 if defined $self->{related_resultsets}{$rel};
3228 return $self->{related_resultsets}{$rel} = do {
3229 my $rsrc = $self->result_source;
3230 my $rel_info = $rsrc->relationship_info($rel);
3232 $self->throw_exception(
3233 "search_related: result source '" . $rsrc->source_name .
3234 "' has no such relationship $rel")
3237 my $attrs = $self->_chain_relationship($rel);
3239 my $join_count = $attrs->{seen_join}{$rel};
3241 my $alias = $self->result_source->storage
3242 ->relname_to_table_alias($rel, $join_count);
3244 # since this is search_related, and we already slid the select window inwards
3245 # (the select/as attrs were deleted in the beginning), we need to flip all
3246 # left joins to inner, so we get the expected results
3247 # read the comment on top of the actual function to see what this does
3248 $attrs->{from} = $rsrc->schema->storage->_inner_join_to_node ($attrs->{from}, $alias);
3251 #XXX - temp fix for result_class bug. There likely is a more elegant fix -groditi
3252 delete @{$attrs}{qw(result_class alias)};
3254 my $rel_source = $rsrc->related_source($rel);
3258 # The reason we do this now instead of passing the alias to the
3259 # search_rs below is that if you wrap/overload resultset on the
3260 # source you need to know what alias it's -going- to have for things
3261 # to work sanely (e.g. RestrictWithObject wants to be able to add
3262 # extra query restrictions, and these may need to be $alias.)
3264 my $rel_attrs = $rel_source->resultset_attributes;
3265 local $rel_attrs->{alias} = $alias;
3267 $rel_source->resultset
3271 where => $attrs->{where},
3275 if (my $cache = $self->get_cache) {
3276 my @related_cache = map
3277 { $_->related_resultset($rel)->get_cache || () }
3281 $new->set_cache([ map @$_, @related_cache ]) if @related_cache == @$cache;
3288 =head2 current_source_alias
3292 =item Arguments: none
3294 =item Return Value: $source_alias
3298 Returns the current table alias for the result source this resultset is built
3299 on, that will be used in the SQL query. Usually it is C<me>.
3301 Currently the source alias that refers to the result set returned by a
3302 L</search>/L</find> family method depends on how you got to the resultset: it's
3303 C<me> by default, but eg. L</search_related> aliases it to the related result
3304 source name (and keeps C<me> referring to the original result set). The long
3305 term goal is to make L<DBIx::Class> always alias the current resultset as C<me>
3306 (and make this method unnecessary).
3308 Thus it's currently necessary to use this method in predefined queries (see
3309 L<DBIx::Class::Manual::Cookbook/Predefined searches>) when referring to the
3310 source alias of the current result set:
3312 # in a result set class
3314 my ($self, $user) = @_;
3316 my $me = $self->current_source_alias;
3318 return $self->search({
3319 "$me.modified" => $user->id,
3325 sub current_source_alias {
3326 return (shift->{attrs} || {})->{alias} || 'me';
3329 =head2 as_subselect_rs
3333 =item Arguments: none
3335 =item Return Value: L<$resultset|/search>
3339 Act as a barrier to SQL symbols. The resultset provided will be made into a
3340 "virtual view" by including it as a subquery within the from clause. From this
3341 point on, any joined tables are inaccessible to ->search on the resultset (as if
3342 it were simply where-filtered without joins). For example:
3344 my $rs = $schema->resultset('Bar')->search({'x.name' => 'abc'},{ join => 'x' });
3346 # 'x' now pollutes the query namespace
3348 # So the following works as expected
3349 my $ok_rs = $rs->search({'x.other' => 1});
3351 # But this doesn't: instead of finding a 'Bar' related to two x rows (abc and
3352 # def) we look for one row with contradictory terms and join in another table
3353 # (aliased 'x_2') which we never use
3354 my $broken_rs = $rs->search({'x.name' => 'def'});
3356 my $rs2 = $rs->as_subselect_rs;
3358 # doesn't work - 'x' is no longer accessible in $rs2, having been sealed away
3359 my $not_joined_rs = $rs2->search({'x.other' => 1});
3361 # works as expected: finds a 'table' row related to two x rows (abc and def)
3362 my $correctly_joined_rs = $rs2->search({'x.name' => 'def'});
3364 Another example of when one might use this would be to select a subset of
3365 columns in a group by clause:
3367 my $rs = $schema->resultset('Bar')->search(undef, {
3368 group_by => [qw{ id foo_id baz_id }],
3369 })->as_subselect_rs->search(undef, {
3370 columns => [qw{ id foo_id }]
3373 In the above example normally columns would have to be equal to the group by,
3374 but because we isolated the group by into a subselect the above works.
3378 sub as_subselect_rs {
3381 my $attrs = $self->_resolved_attrs;
3383 my $fresh_rs = (ref $self)->new (
3384 $self->result_source
3387 # these pieces will be locked in the subquery
3388 delete $fresh_rs->{cond};
3389 delete @{$fresh_rs->{attrs}}{qw/where bind/};
3391 return $fresh_rs->search( {}, {
3393 $attrs->{alias} => $self->as_query,
3394 -alias => $attrs->{alias},
3395 -rsrc => $self->result_source,
3397 alias => $attrs->{alias},
3401 # This code is called by search_related, and makes sure there
3402 # is clear separation between the joins before, during, and
3403 # after the relationship. This information is needed later
3404 # in order to properly resolve prefetch aliases (any alias
3405 # with a relation_chain_depth less than the depth of the
3406 # current prefetch is not considered)
3408 # The increments happen twice per join. An even number means a
3409 # relationship specified via a search_related, whereas an odd
3410 # number indicates a join/prefetch added via attributes
3412 # Also this code will wrap the current resultset (the one we
3413 # chain to) in a subselect IFF it contains limiting attributes
3414 sub _chain_relationship {
3415 my ($self, $rel) = @_;
3416 my $source = $self->result_source;
3417 my $attrs = { %{$self->{attrs}||{}} };
3419 # we need to take the prefetch the attrs into account before we
3420 # ->_resolve_join as otherwise they get lost - captainL
3421 my $join = $self->_merge_joinpref_attr( $attrs->{join}, $attrs->{prefetch} );
3423 delete @{$attrs}{qw/join prefetch collapse group_by distinct _grouped_by_distinct select as columns +select +as +columns/};
3425 my $seen = { %{ (delete $attrs->{seen_join}) || {} } };
3428 my @force_subq_attrs = qw/offset rows group_by having/;
3431 ($attrs->{from} && ref $attrs->{from} ne 'ARRAY')
3433 $self->_has_resolved_attr (@force_subq_attrs)
3435 # Nuke the prefetch (if any) before the new $rs attrs
3436 # are resolved (prefetch is useless - we are wrapping
3437 # a subquery anyway).
3438 my $rs_copy = $self->search;
3439 $rs_copy->{attrs}{join} = $self->_merge_joinpref_attr (
3440 $rs_copy->{attrs}{join},
3441 delete $rs_copy->{attrs}{prefetch},
3446 -alias => $attrs->{alias},
3447 $attrs->{alias} => $rs_copy->as_query,
3449 delete @{$attrs}{@force_subq_attrs, qw/where bind/};
3450 $seen->{-relation_chain_depth} = 0;
3452 elsif ($attrs->{from}) { #shallow copy suffices
3453 $from = [ @{$attrs->{from}} ];
3458 -alias => $attrs->{alias},
3459 $attrs->{alias} => $source->from,
3463 my $jpath = ($seen->{-relation_chain_depth})
3464 ? $from->[-1][0]{-join_path}
3467 my @requested_joins = $source->_resolve_join(
3474 push @$from, @requested_joins;
3476 $seen->{-relation_chain_depth}++;
3478 # if $self already had a join/prefetch specified on it, the requested
3479 # $rel might very well be already included. What we do in this case
3480 # is effectively a no-op (except that we bump up the chain_depth on
3481 # the join in question so we could tell it *is* the search_related)
3484 # we consider the last one thus reverse
3485 for my $j (reverse @requested_joins) {
3486 my ($last_j) = keys %{$j->[0]{-join_path}[-1]};
3487 if ($rel eq $last_j) {
3488 $j->[0]{-relation_chain_depth}++;
3494 unless ($already_joined) {
3495 push @$from, $source->_resolve_join(
3503 $seen->{-relation_chain_depth}++;
3505 return {%$attrs, from => $from, seen_join => $seen};
3508 sub _resolved_attrs {
3510 return $self->{_attrs} if $self->{_attrs};
3512 my $attrs = { %{ $self->{attrs} || {} } };
3513 my $source = $attrs->{result_source} = $self->result_source;
3514 my $alias = $attrs->{alias};
3516 $self->throw_exception("Specifying distinct => 1 in conjunction with collapse => 1 is unsupported")
3517 if $attrs->{collapse} and $attrs->{distinct};
3519 # default selection list
3520 $attrs->{columns} = [ $source->columns ]
3521 unless List::Util::first { exists $attrs->{$_} } qw/columns cols select as/;
3523 # merge selectors together
3524 for (qw/columns select as/) {
3525 $attrs->{$_} = $self->_merge_attr($attrs->{$_}, delete $attrs->{"+$_"})
3526 if $attrs->{$_} or $attrs->{"+$_"};
3529 # disassemble columns
3531 if (my $cols = delete $attrs->{columns}) {
3532 for my $c (ref $cols eq 'ARRAY' ? @$cols : $cols) {
3533 if (ref $c eq 'HASH') {
3534 for my $as (sort keys %$c) {
3535 push @sel, $c->{$as};
3546 # when trying to weed off duplicates later do not go past this point -
3547 # everything added from here on is unbalanced "anyone's guess" stuff
3548 my $dedup_stop_idx = $#as;
3550 push @as, @{ ref $attrs->{as} eq 'ARRAY' ? $attrs->{as} : [ $attrs->{as} ] }
3552 push @sel, @{ ref $attrs->{select} eq 'ARRAY' ? $attrs->{select} : [ $attrs->{select} ] }
3553 if $attrs->{select};
3555 # assume all unqualified selectors to apply to the current alias (legacy stuff)
3556 $_ = (ref $_ or $_ =~ /\./) ? $_ : "$alias.$_" for @sel;
3558 # disqualify all $alias.col as-bits (inflate-map mandated)
3559 $_ = ($_ =~ /^\Q$alias.\E(.+)$/) ? $1 : $_ for @as;
3561 # de-duplicate the result (remove *identical* select/as pairs)
3562 # and also die on duplicate {as} pointing to different {select}s
3563 # not using a c-style for as the condition is prone to shrinkage
3566 while ($i <= $dedup_stop_idx) {
3567 if ($seen->{"$sel[$i] \x00\x00 $as[$i]"}++) {
3572 elsif ($seen->{$as[$i]}++) {
3573 $self->throw_exception(
3574 "inflate_result() alias '$as[$i]' specified twice with different SQL-side {select}-ors"
3582 $attrs->{select} = \@sel;
3583 $attrs->{as} = \@as;
3585 $attrs->{from} ||= [{
3587 -alias => $self->{attrs}{alias},
3588 $self->{attrs}{alias} => $source->from,
3591 if ( $attrs->{join} || $attrs->{prefetch} ) {
3593 $self->throw_exception ('join/prefetch can not be used with a custom {from}')
3594 if ref $attrs->{from} ne 'ARRAY';
3596 my $join = (delete $attrs->{join}) || {};
3598 if ( defined $attrs->{prefetch} ) {
3599 $join = $self->_merge_joinpref_attr( $join, $attrs->{prefetch} );
3602 $attrs->{from} = # have to copy here to avoid corrupting the original
3604 @{ $attrs->{from} },
3605 $source->_resolve_join(
3608 { %{ $attrs->{seen_join} || {} } },
3609 ( $attrs->{seen_join} && keys %{$attrs->{seen_join}})
3610 ? $attrs->{from}[-1][0]{-join_path}
3617 if ( defined $attrs->{order_by} ) {
3618 $attrs->{order_by} = (
3619 ref( $attrs->{order_by} ) eq 'ARRAY'
3620 ? [ @{ $attrs->{order_by} } ]
3621 : [ $attrs->{order_by} || () ]
3625 if ($attrs->{group_by} and ref $attrs->{group_by} ne 'ARRAY') {
3626 $attrs->{group_by} = [ $attrs->{group_by} ];
3630 # generate selections based on the prefetch helper
3631 my ($prefetch, @prefetch_select, @prefetch_as);
3632 $prefetch = $self->_merge_joinpref_attr( {}, delete $attrs->{prefetch} )
3633 if defined $attrs->{prefetch};
3637 $self->throw_exception("Unable to prefetch, resultset contains an unnamed selector $attrs->{_dark_selector}{string}")
3638 if $attrs->{_dark_selector};
3640 $self->throw_exception("Specifying prefetch in conjunction with an explicit collapse => 0 is unsupported")
3641 if defined $attrs->{collapse} and ! $attrs->{collapse};
3643 $attrs->{collapse} = 1;
3645 # this is a separate structure (we don't look in {from} directly)
3646 # as the resolver needs to shift things off the lists to work
3647 # properly (identical-prefetches on different branches)
3649 if (ref $attrs->{from} eq 'ARRAY') {
3651 my $start_depth = $attrs->{seen_join}{-relation_chain_depth} || 0;
3653 for my $j ( @{$attrs->{from}}[1 .. $#{$attrs->{from}} ] ) {
3654 next unless $j->[0]{-alias};
3655 next unless $j->[0]{-join_path};
3656 next if ($j->[0]{-relation_chain_depth} || 0) < $start_depth;
3658 my @jpath = map { keys %$_ } @{$j->[0]{-join_path}};
3661 $p = $p->{$_} ||= {} for @jpath[ ($start_depth/2) .. $#jpath]; #only even depths are actual jpath boundaries
3662 push @{$p->{-join_aliases} }, $j->[0]{-alias};
3666 my @prefetch = $source->_resolve_prefetch( $prefetch, $alias, $join_map );
3668 # save these for after distinct resolution
3669 @prefetch_select = map { $_->[0] } @prefetch;
3670 @prefetch_as = map { $_->[1] } @prefetch;
3673 # run through the resulting joinstructure (starting from our current slot)
3674 # and unset collapse if proven unnecessary
3676 # also while we are at it find out if the current root source has
3677 # been premultiplied by previous related_source chaining
3679 # this allows to predict whether a root object with all other relation
3680 # data set to NULL is in fact unique
3681 if ($attrs->{collapse}) {
3683 if (ref $attrs->{from} eq 'ARRAY') {
3685 if (@{$attrs->{from}} == 1) {
3686 # no joins - no collapse
3687 $attrs->{collapse} = 0;
3690 # find where our table-spec starts
3691 my @fromlist = @{$attrs->{from}};
3693 my $t = shift @fromlist;
3696 # me vs join from-spec distinction - a ref means non-root
3697 if (ref $t eq 'ARRAY') {
3699 $is_multi ||= ! $t->{-is_single};
3701 last if ($t->{-alias} && $t->{-alias} eq $alias);
3702 $attrs->{_main_source_premultiplied} ||= $is_multi;
3705 # no non-singles remaining, nor any premultiplication - nothing to collapse
3707 ! $attrs->{_main_source_premultiplied}
3709 ! List::Util::first { ! $_->[0]{-is_single} } @fromlist
3711 $attrs->{collapse} = 0;
3717 # if we can not analyze the from - err on the side of safety
3718 $attrs->{_main_source_premultiplied} = 1;
3722 # generate the distinct induced group_by before injecting the prefetched select/as parts
3723 if (delete $attrs->{distinct}) {
3724 if ($attrs->{group_by}) {
3725 carp_unique ("Useless use of distinct on a grouped resultset ('distinct' is ignored when a 'group_by' is present)");
3728 $attrs->{_grouped_by_distinct} = 1;
3729 # distinct affects only the main selection part, not what prefetch may add below
3730 ($attrs->{group_by}, my $new_order) = $source->storage->_group_over_selection($attrs);
3732 # FIXME possibly ignore a rewritten order_by (may turn out to be an issue)
3733 # The thinking is: if we are collapsing the subquerying prefetch engine will
3734 # rip stuff apart for us anyway, and we do not want to have a potentially
3735 # function-converted external order_by
3736 # ( there is an explicit if ( collapse && _grouped_by_distinct ) check in DBIHacks )
3737 $attrs->{order_by} = $new_order unless $attrs->{collapse};
3741 # inject prefetch-bound selection (if any)
3742 push @{$attrs->{select}}, @prefetch_select;
3743 push @{$attrs->{as}}, @prefetch_as;
3745 $attrs->{_simple_passthrough_construction} = !(
3748 grep { $_ =~ /\./ } @{$attrs->{as}}
3751 # if both page and offset are specified, produce a combined offset
3752 # even though it doesn't make much sense, this is what pre 081xx has
3754 if (my $page = delete $attrs->{page}) {
3756 ($attrs->{rows} * ($page - 1))
3758 ($attrs->{offset} || 0)
3762 return $self->{_attrs} = $attrs;
3766 my ($self, $attr) = @_;
3768 if (ref $attr eq 'HASH') {
3769 return $self->_rollout_hash($attr);
3770 } elsif (ref $attr eq 'ARRAY') {
3771 return $self->_rollout_array($attr);
3777 sub _rollout_array {
3778 my ($self, $attr) = @_;
3781 foreach my $element (@{$attr}) {
3782 if (ref $element eq 'HASH') {
3783 push( @rolled_array, @{ $self->_rollout_hash( $element ) } );
3784 } elsif (ref $element eq 'ARRAY') {
3785 # XXX - should probably recurse here
3786 push( @rolled_array, @{$self->_rollout_array($element)} );
3788 push( @rolled_array, $element );
3791 return \@rolled_array;
3795 my ($self, $attr) = @_;
3798 foreach my $key (keys %{$attr}) {
3799 push( @rolled_array, { $key => $attr->{$key} } );
3801 return \@rolled_array;
3804 sub _calculate_score {
3805 my ($self, $a, $b) = @_;
3807 if (defined $a xor defined $b) {
3810 elsif (not defined $a) {
3814 if (ref $b eq 'HASH') {
3815 my ($b_key) = keys %{$b};
3816 if (ref $a eq 'HASH') {
3817 my ($a_key) = keys %{$a};
3818 if ($a_key eq $b_key) {
3819 return (1 + $self->_calculate_score( $a->{$a_key}, $b->{$b_key} ));
3824 return ($a eq $b_key) ? 1 : 0;
3827 if (ref $a eq 'HASH') {
3828 my ($a_key) = keys %{$a};
3829 return ($b eq $a_key) ? 1 : 0;
3831 return ($b eq $a) ? 1 : 0;
3836 sub _merge_joinpref_attr {
3837 my ($self, $orig, $import) = @_;
3839 return $import unless defined($orig);
3840 return $orig unless defined($import);
3842 $orig = $self->_rollout_attr($orig);
3843 $import = $self->_rollout_attr($import);
3846 foreach my $import_element ( @{$import} ) {
3847 # find best candidate from $orig to merge $b_element into
3848 my $best_candidate = { position => undef, score => 0 }; my $position = 0;
3849 foreach my $orig_element ( @{$orig} ) {
3850 my $score = $self->_calculate_score( $orig_element, $import_element );
3851 if ($score > $best_candidate->{score}) {
3852 $best_candidate->{position} = $position;
3853 $best_candidate->{score} = $score;
3857 my ($import_key) = ( ref $import_element eq 'HASH' ) ? keys %{$import_element} : ($import_element);
3858 $import_key = '' if not defined $import_key;
3860 if ($best_candidate->{score} == 0 || exists $seen_keys->{$import_key}) {
3861 push( @{$orig}, $import_element );
3863 my $orig_best = $orig->[$best_candidate->{position}];
3864 # merge orig_best and b_element together and replace original with merged
3865 if (ref $orig_best ne 'HASH') {
3866 $orig->[$best_candidate->{position}] = $import_element;
3867 } elsif (ref $import_element eq 'HASH') {
3868 my ($key) = keys %{$orig_best};
3869 $orig->[$best_candidate->{position}] = { $key => $self->_merge_joinpref_attr($orig_best->{$key}, $import_element->{$key}) };
3872 $seen_keys->{$import_key} = 1; # don't merge the same key twice
3875 return @$orig ? $orig : ();
3883 require Hash::Merge;
3884 my $hm = Hash::Merge->new;
3886 $hm->specify_behavior({
3889 my ($defl, $defr) = map { defined $_ } (@_[0,1]);
3891 if ($defl xor $defr) {
3892 return [ $defl ? $_[0] : $_[1] ];
3897 elsif (__HM_DEDUP and $_[0] eq $_[1]) {
3901 return [$_[0], $_[1]];
3905 return $_[1] if !defined $_[0];
3906 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3907 return [$_[0], @{$_[1]}]
3910 return [] if !defined $_[0] and !keys %{$_[1]};
3911 return [ $_[1] ] if !defined $_[0];
3912 return [ $_[0] ] if !keys %{$_[1]};
3913 return [$_[0], $_[1]]
3918 return $_[0] if !defined $_[1];
3919 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3920 return [@{$_[0]}, $_[1]]
3923 my @ret = @{$_[0]} or return $_[1];
3924 return [ @ret, @{$_[1]} ] unless __HM_DEDUP;
3925 my %idx = map { $_ => 1 } @ret;
3926 push @ret, grep { ! defined $idx{$_} } (@{$_[1]});
3930 return [ $_[1] ] if ! @{$_[0]};
3931 return $_[0] if !keys %{$_[1]};
3932 return $_[0] if __HM_DEDUP and List::Util::first { $_ eq $_[1] } @{$_[0]};
3933 return [ @{$_[0]}, $_[1] ];
3938 return [] if !keys %{$_[0]} and !defined $_[1];
3939 return [ $_[0] ] if !defined $_[1];
3940 return [ $_[1] ] if !keys %{$_[0]};
3941 return [$_[0], $_[1]]
3944 return [] if !keys %{$_[0]} and !@{$_[1]};
3945 return [ $_[0] ] if !@{$_[1]};
3946 return $_[1] if !keys %{$_[0]};
3947 return $_[1] if __HM_DEDUP and List::Util::first { $_ eq $_[0] } @{$_[1]};
3948 return [ $_[0], @{$_[1]} ];
3951 return [] if !keys %{$_[0]} and !keys %{$_[1]};
3952 return [ $_[0] ] if !keys %{$_[1]};
3953 return [ $_[1] ] if !keys %{$_[0]};
3954 return [ $_[0] ] if $_[0] eq $_[1];
3955 return [ $_[0], $_[1] ];
3958 } => 'DBIC_RS_ATTR_MERGER');
3962 return $hm->merge ($_[1], $_[2]);
3966 sub STORABLE_freeze {
3967 my ($self, $cloning) = @_;
3968 my $to_serialize = { %$self };
3970 # A cursor in progress can't be serialized (and would make little sense anyway)
3971 # the parser can be regenerated (and can't be serialized)
3972 delete @{$to_serialize}{qw/cursor _row_parser _result_inflator/};
3974 # nor is it sensical to store a not-yet-fired-count pager
3975 if ($to_serialize->{pager} and ref $to_serialize->{pager}{total_entries} eq 'CODE') {
3976 delete $to_serialize->{pager};
3979 Storable::nfreeze($to_serialize);
3982 # need this hook for symmetry
3984 my ($self, $cloning, $serialized) = @_;
3986 %$self = %{ Storable::thaw($serialized) };
3992 =head2 throw_exception
3994 See L<DBIx::Class::Schema/throw_exception> for details.
3998 sub throw_exception {
4001 if (ref $self and my $rsrc = $self->result_source) {
4002 $rsrc->throw_exception(@_)
4005 DBIx::Class::Exception->throw(@_);
4013 # XXX: FIXME: Attributes docs need clearing up
4017 Attributes are used to refine a ResultSet in various ways when
4018 searching for data. They can be passed to any method which takes an
4019 C<\%attrs> argument. See L</search>, L</search_rs>, L</find>,
4022 Default attributes can be set on the result class using
4023 L<DBIx::Class::ResultSource/resultset_attributes>. (Please read
4024 the CAVEATS on that feature before using it!)
4026 These are in no particular order:
4032 =item Value: ( $order_by | \@order_by | \%order_by )
4036 Which column(s) to order the results by.
4038 [The full list of suitable values is documented in
4039 L<SQL::Abstract/"ORDER BY CLAUSES">; the following is a summary of
4042 If a single column name, or an arrayref of names is supplied, the
4043 argument is passed through directly to SQL. The hashref syntax allows
4044 for connection-agnostic specification of ordering direction:
4046 For descending order:
4048 order_by => { -desc => [qw/col1 col2 col3/] }
4050 For explicit ascending order:
4052 order_by => { -asc => 'col' }
4054 The old scalarref syntax (i.e. order_by => \'year DESC') is still
4055 supported, although you are strongly encouraged to use the hashref
4056 syntax as outlined above.
4062 =item Value: \@columns | \%columns | $column
4066 Shortcut to request a particular set of columns to be retrieved. Each
4067 column spec may be a string (a table column name), or a hash (in which
4068 case the key is the C<as> value, and the value is used as the C<select>
4069 expression). Adds the L</current_source_alias> onto the start of any column without a C<.> in
4070 it and sets C<select> from that, then auto-populates C<as> from
4071 C<select> as normal. (You may also use the C<cols> attribute, as in
4072 earlier versions of DBIC, but this is deprecated)
4074 Essentially C<columns> does the same as L</select> and L</as>.
4076 columns => [ 'some_column', { dbic_slot => 'another_column' } ]
4080 select => [qw(some_column another_column)],
4081 as => [qw(some_column dbic_slot)]
4083 If you want to individually retrieve related columns (in essence perform
4084 manual prefetch) you have to make sure to specify the correct inflation slot
4085 chain such that it matches existing relationships:
4087 my $rs = $schema->resultset('Artist')->search({}, {
4088 # required to tell DBIC to collapse has_many relationships
4090 join => { cds => 'tracks'},
4092 'cds.cdid' => 'cds.cdid',
4093 'cds.tracks.title' => 'tracks.title',
4099 B<NOTE:> You B<MUST> explicitly quote C<'+columns'> when using this attribute.
4100 Not doing so causes Perl to incorrectly interpret C<+columns> as a bareword
4101 with a unary plus operator before it, which is the same as simply C<columns>.
4105 =item Value: \@extra_columns
4109 Indicates additional columns to be selected from storage. Works the same as
4110 L</columns> but adds columns to the current selection. (You may also use the
4111 C<include_columns> attribute, as in earlier versions of DBIC, but this is
4114 $schema->resultset('CD')->search(undef, {
4115 '+columns' => ['artist.name'],
4119 would return all CDs and include a 'name' column to the information
4120 passed to object inflation. Note that the 'artist' is the name of the
4121 column (or relationship) accessor, and 'name' is the name of the column
4122 accessor in the related table.
4128 =item Value: \@select_columns
4132 Indicates which columns should be selected from the storage. You can use
4133 column names, or in the case of RDBMS back ends, function or stored procedure
4136 $rs = $schema->resultset('Employee')->search(undef, {
4139 { count => 'employeeid' },
4140 { max => { length => 'name' }, -as => 'longest_name' }
4145 SELECT name, COUNT( employeeid ), MAX( LENGTH( name ) ) AS longest_name FROM employee
4147 B<NOTE:> You will almost always need a corresponding L</as> attribute when you
4148 use L</select>, to instruct DBIx::Class how to store the result of the column.
4150 Also note that the L</as> attribute has B<nothing to do> with the SQL-side
4151 C<AS> identifier aliasing. You B<can> alias a function (so you can use it e.g.
4152 in an C<ORDER BY> clause), however this is done via the C<-as> B<select
4153 function attribute> supplied as shown in the example above.
4157 B<NOTE:> You B<MUST> explicitly quote C<'+select'> when using this attribute.
4158 Not doing so causes Perl to incorrectly interpret C<+select> as a bareword
4159 with a unary plus operator before it, which is the same as simply C<select>.
4163 =item Value: \@extra_select_columns
4167 Indicates additional columns to be selected from storage. Works the same as
4168 L</select> but adds columns to the current selection, instead of specifying
4169 a new explicit list.
4175 =item Value: \@inflation_names
4179 Indicates DBIC-side names for object inflation. That is L</as> indicates the
4180 slot name in which the column value will be stored within the
4181 L<Row|DBIx::Class::Row> object. The value will then be accessible via this
4182 identifier by the C<get_column> method (or via the object accessor B<if one
4183 with the same name already exists>) as shown below.
4185 The L</as> attribute has B<nothing to do> with the SQL-side identifier
4186 aliasing C<AS>. See L</select> for details.
4188 $rs = $schema->resultset('Employee')->search(undef, {
4191 { count => 'employeeid' },
4192 { max => { length => 'name' }, -as => 'longest_name' }
4201 If the object against which the search is performed already has an accessor
4202 matching a column name specified in C<as>, the value can be retrieved using
4203 the accessor as normal:
4205 my $name = $employee->name();
4207 If on the other hand an accessor does not exist in the object, you need to
4208 use C<get_column> instead:
4210 my $employee_count = $employee->get_column('employee_count');
4212 You can create your own accessors if required - see
4213 L<DBIx::Class::Manual::Cookbook> for details.
4217 B<NOTE:> You B<MUST> explicitly quote C<'+as'> when using this attribute.
4218 Not doing so causes Perl to incorrectly interpret C<+as> as a bareword
4219 with a unary plus operator before it, which is the same as simply C<as>.
4223 =item Value: \@extra_inflation_names
4227 Indicates additional inflation names for selectors added via L</+select>. See L</as>.
4233 =item Value: ($rel_name | \@rel_names | \%rel_names)
4237 Contains a list of relationships that should be joined for this query. For
4240 # Get CDs by Nine Inch Nails
4241 my $rs = $schema->resultset('CD')->search(
4242 { 'artist.name' => 'Nine Inch Nails' },
4243 { join => 'artist' }
4246 Can also contain a hash reference to refer to the other relation's relations.
4249 package MyApp::Schema::Track;
4250 use base qw/DBIx::Class/;
4251 __PACKAGE__->table('track');
4252 __PACKAGE__->add_columns(qw/trackid cd position title/);
4253 __PACKAGE__->set_primary_key('trackid');
4254 __PACKAGE__->belongs_to(cd => 'MyApp::Schema::CD');
4257 # In your application
4258 my $rs = $schema->resultset('Artist')->search(
4259 { 'track.title' => 'Teardrop' },
4261 join => { cd => 'track' },
4262 order_by => 'artist.name',
4266 You need to use the relationship (not the table) name in conditions,
4267 because they are aliased as such. The current table is aliased as "me", so
4268 you need to use me.column_name in order to avoid ambiguity. For example:
4270 # Get CDs from 1984 with a 'Foo' track
4271 my $rs = $schema->resultset('CD')->search(
4274 'tracks.name' => 'Foo'
4276 { join => 'tracks' }
4279 If the same join is supplied twice, it will be aliased to <rel>_2 (and
4280 similarly for a third time). For e.g.
4282 my $rs = $schema->resultset('Artist')->search({
4283 'cds.title' => 'Down to Earth',
4284 'cds_2.title' => 'Popular',
4286 join => [ qw/cds cds/ ],
4289 will return a set of all artists that have both a cd with title 'Down
4290 to Earth' and a cd with title 'Popular'.
4292 If you want to fetch related objects from other tables as well, see L</prefetch>
4295 NOTE: An internal join-chain pruner will discard certain joins while
4296 constructing the actual SQL query, as long as the joins in question do not
4297 affect the retrieved result. This for example includes 1:1 left joins
4298 that are not part of the restriction specification (WHERE/HAVING) nor are
4299 a part of the query selection.
4301 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
4307 =item Value: (0 | 1)
4311 When set to a true value, indicates that any rows fetched from joined has_many
4312 relationships are to be aggregated into the corresponding "parent" object. For
4313 example, the resultset:
4315 my $rs = $schema->resultset('CD')->search({}, {
4316 '+columns' => [ qw/ tracks.title tracks.position / ],
4321 While executing the following query:
4323 SELECT me.*, tracks.title, tracks.position
4325 LEFT JOIN track tracks
4326 ON tracks.cdid = me.cdid
4328 Will return only as many objects as there are rows in the CD source, even
4329 though the result of the query may span many rows. Each of these CD objects
4330 will in turn have multiple "Track" objects hidden behind the has_many
4331 generated accessor C<tracks>. Without C<< collapse => 1 >>, the return values
4332 of this resultset would be as many CD objects as there are tracks (a "Cartesian
4333 product"), with each CD object containing exactly one of all fetched Track data.
4335 When a collapse is requested on a non-ordered resultset, an order by some
4336 unique part of the main source (the left-most table) is inserted automatically.
4337 This is done so that the resultset is allowed to be "lazy" - calling
4338 L<< $rs->next|/next >> will fetch only as many rows as it needs to build the next
4339 object with all of its related data.
4341 If an L</order_by> is already declared, and orders the resultset in a way that
4342 makes collapsing as described above impossible (e.g. C<< ORDER BY
4343 has_many_rel.column >> or C<ORDER BY RANDOM()>), DBIC will automatically
4344 switch to "eager" mode and slurp the entire resultset before constructing the
4345 first object returned by L</next>.
4347 Setting this attribute on a resultset that does not join any has_many
4348 relations is a no-op.
4350 For a more in-depth discussion, see L</PREFETCHING>.
4356 =item Value: ($rel_name | \@rel_names | \%rel_names)
4360 This attribute is a shorthand for specifying a L</join> spec, adding all
4361 columns from the joined related sources as L</+columns> and setting
4362 L</collapse> to a true value. It can be thought of as a rough B<superset>
4363 of the L</join> attribute.
4365 For example, the following two queries are equivalent:
4367 my $rs = $schema->resultset('Artist')->search({}, {
4368 prefetch => { cds => ['genre', 'tracks' ] },
4373 my $rs = $schema->resultset('Artist')->search({}, {
4374 join => { cds => ['genre', 'tracks' ] },
4378 { +{ "cds.$_" => "cds.$_" } }
4379 $schema->source('Artist')->related_source('cds')->columns
4382 { +{ "cds.genre.$_" => "genre.$_" } }
4383 $schema->source('Artist')->related_source('cds')->related_source('genre')->columns
4386 { +{ "cds.tracks.$_" => "tracks.$_" } }
4387 $schema->source('Artist')->related_source('cds')->related_source('tracks')->columns
4392 Both producing the following SQL:
4394 SELECT me.artistid, me.name, me.rank, me.charfield,
4395 cds.cdid, cds.artist, cds.title, cds.year, cds.genreid, cds.single_track,
4396 genre.genreid, genre.name,
4397 tracks.trackid, tracks.cd, tracks.position, tracks.title, tracks.last_updated_on, tracks.last_updated_at
4400 ON cds.artist = me.artistid
4401 LEFT JOIN genre genre
4402 ON genre.genreid = cds.genreid
4403 LEFT JOIN track tracks
4404 ON tracks.cd = cds.cdid
4405 ORDER BY me.artistid
4407 While L</prefetch> implies a L</join>, it is ok to mix the two together, as
4408 the arguments are properly merged and generally do the right thing. For
4409 example, you may want to do the following:
4411 my $artists_and_cds_without_genre = $schema->resultset('Artist')->search(
4412 { 'genre.genreid' => undef },
4414 join => { cds => 'genre' },
4419 Which generates the following SQL:
4421 SELECT me.artistid, me.name, me.rank, me.charfield,
4422 cds.cdid, cds.artist, cds.title, cds.year, cds.genreid, cds.single_track
4425 ON cds.artist = me.artistid
4426 LEFT JOIN genre genre
4427 ON genre.genreid = cds.genreid
4428 WHERE genre.genreid IS NULL
4429 ORDER BY me.artistid
4431 For a more in-depth discussion, see L</PREFETCHING>.
4437 =item Value: $source_alias
4441 Sets the source alias for the query. Normally, this defaults to C<me>, but
4442 nested search queries (sub-SELECTs) might need specific aliases set to
4443 reference inner queries. For example:
4446 ->related_resultset('CDs')
4447 ->related_resultset('Tracks')
4449 'track.id' => { -ident => 'none_search.id' },
4453 my $ids = $self->search({
4456 alias => 'none_search',
4457 group_by => 'none_search.id',
4458 })->get_column('id')->as_query;
4460 $self->search({ id => { -in => $ids } })
4462 This attribute is directly tied to L</current_source_alias>.
4472 Makes the resultset paged and specifies the page to retrieve. Effectively
4473 identical to creating a non-pages resultset and then calling ->page($page)
4476 If L</rows> attribute is not specified it defaults to 10 rows per page.
4478 When you have a paged resultset, L</count> will only return the number
4479 of rows in the page. To get the total, use the L</pager> and call
4480 C<total_entries> on it.
4490 Specifies the maximum number of rows for direct retrieval or the number of
4491 rows per page if the page attribute or method is used.
4497 =item Value: $offset
4501 Specifies the (zero-based) row number for the first row to be returned, or the
4502 of the first row of the first page if paging is used.
4504 =head2 software_limit
4508 =item Value: (0 | 1)
4512 When combined with L</rows> and/or L</offset> the generated SQL will not
4513 include any limit dialect stanzas. Instead the entire result will be selected
4514 as if no limits were specified, and DBIC will perform the limit locally, by
4515 artificially advancing and finishing the resulting L</cursor>.
4517 This is the recommended way of performing resultset limiting when no sane RDBMS
4518 implementation is available (e.g.
4519 L<Sybase ASE|DBIx::Class::Storage::DBI::Sybase::ASE> using the
4520 L<Generic Sub Query|DBIx::Class::SQLMaker::LimitDialects/GenericSubQ> hack)
4526 =item Value: \@columns
4530 A arrayref of columns to group by. Can include columns of joined tables.
4532 group_by => [qw/ column1 column2 ... /]
4538 =item Value: $condition
4542 The HAVING operator specifies a B<secondary> condition applied to the set
4543 after the grouping calculations have been done. In other words it is a
4544 constraint just like L</where> (and accepting the same
4545 L<SQL::Abstract syntax|SQL::Abstract/WHERE CLAUSES>) applied to the data
4546 as it exists after GROUP BY has taken place. Specifying L</having> without
4547 L</group_by> is a logical mistake, and a fatal error on most RDBMS engines.
4551 having => { 'count_employee' => { '>=', 100 } }
4553 or with an in-place function in which case literal SQL is required:
4555 having => \[ 'count(employee) >= ?', 100 ]
4561 =item Value: (0 | 1)
4565 Set to 1 to automatically generate a L</group_by> clause based on the selection
4566 (including intelligent handling of L</order_by> contents). Note that the group
4567 criteria calculation takes place over the B<final> selection. This includes
4568 any L</+columns>, L</+select> or L</order_by> additions in subsequent
4569 L</search> calls, and standalone columns selected via
4570 L<DBIx::Class::ResultSetColumn> (L</get_column>). A notable exception are the
4571 extra selections specified via L</prefetch> - such selections are explicitly
4572 excluded from group criteria calculations.
4574 If the final ResultSet also explicitly defines a L</group_by> attribute, this
4575 setting is ignored and an appropriate warning is issued.
4581 Adds to the WHERE clause.
4583 # only return rows WHERE deleted IS NULL for all searches
4584 __PACKAGE__->resultset_attributes({ where => { deleted => undef } });
4586 Can be overridden by passing C<< { where => undef } >> as an attribute
4589 For more complicated where clauses see L<SQL::Abstract/WHERE CLAUSES>.
4595 Set to 1 to cache search results. This prevents extra SQL queries if you
4596 revisit rows in your ResultSet:
4598 my $resultset = $schema->resultset('Artist')->search( undef, { cache => 1 } );
4600 while( my $artist = $resultset->next ) {
4604 $rs->first; # without cache, this would issue a query
4606 By default, searches are not cached.
4608 For more examples of using these attributes, see
4609 L<DBIx::Class::Manual::Cookbook>.
4615 =item Value: ( 'update' | 'shared' | \$scalar )
4619 Set to 'update' for a SELECT ... FOR UPDATE or 'shared' for a SELECT
4620 ... FOR SHARED. If \$scalar is passed, this is taken directly and embedded in the
4625 DBIx::Class supports arbitrary related data prefetching from multiple related
4626 sources. Any combination of relationship types and column sets are supported.
4627 If L<collapsing|/collapse> is requested, there is an additional requirement of
4628 selecting enough data to make every individual object uniquely identifiable.
4630 Here are some more involved examples, based on the following relationship map:
4633 My::Schema::CD->belongs_to( artist => 'My::Schema::Artist' );
4634 My::Schema::CD->might_have( liner_note => 'My::Schema::LinerNotes' );
4635 My::Schema::CD->has_many( tracks => 'My::Schema::Track' );
4637 My::Schema::Artist->belongs_to( record_label => 'My::Schema::RecordLabel' );
4639 My::Schema::Track->has_many( guests => 'My::Schema::Guest' );
4643 my $rs = $schema->resultset('Tag')->search(
4652 The initial search results in SQL like the following:
4654 SELECT tag.*, cd.*, artist.* FROM tag
4655 JOIN cd ON tag.cd = cd.cdid
4656 JOIN artist ON cd.artist = artist.artistid
4658 L<DBIx::Class> has no need to go back to the database when we access the
4659 C<cd> or C<artist> relationships, which saves us two SQL statements in this
4662 Simple prefetches will be joined automatically, so there is no need
4663 for a C<join> attribute in the above search.
4665 The L</prefetch> attribute can be used with any of the relationship types
4666 and multiple prefetches can be specified together. Below is a more complex
4667 example that prefetches a CD's artist, its liner notes (if present),
4668 the cover image, the tracks on that CD, and the guests on those
4671 my $rs = $schema->resultset('CD')->search(
4675 { artist => 'record_label'}, # belongs_to => belongs_to
4676 'liner_note', # might_have
4677 'cover_image', # has_one
4678 { tracks => 'guests' }, # has_many => has_many
4683 This will produce SQL like the following:
4685 SELECT cd.*, artist.*, record_label.*, liner_note.*, cover_image.*,
4689 ON artist.artistid = me.artistid
4690 JOIN record_label record_label
4691 ON record_label.labelid = artist.labelid
4692 LEFT JOIN track tracks
4693 ON tracks.cdid = me.cdid
4694 LEFT JOIN guest guests
4695 ON guests.trackid = track.trackid
4696 LEFT JOIN liner_notes liner_note
4697 ON liner_note.cdid = me.cdid
4698 JOIN cd_artwork cover_image
4699 ON cover_image.cdid = me.cdid
4702 Now the C<artist>, C<record_label>, C<liner_note>, C<cover_image>,
4703 C<tracks>, and C<guests> of the CD will all be available through the
4704 relationship accessors without the need for additional queries to the
4709 Prefetch does a lot of deep magic. As such, it may not behave exactly
4710 as you might expect.
4716 Prefetch uses the L</cache> to populate the prefetched relationships. This
4717 may or may not be what you want.
4721 If you specify a condition on a prefetched relationship, ONLY those
4722 rows that match the prefetched condition will be fetched into that relationship.
4723 This means that adding prefetch to a search() B<may alter> what is returned by
4724 traversing a relationship. So, if you have C<< Artist->has_many(CDs) >> and you do
4726 my $artist_rs = $schema->resultset('Artist')->search({
4732 my $count = $artist_rs->first->cds->count;
4734 my $artist_rs_prefetch = $artist_rs->search( {}, { prefetch => 'cds' } );
4736 my $prefetch_count = $artist_rs_prefetch->first->cds->count;
4738 cmp_ok( $count, '==', $prefetch_count, "Counts should be the same" );
4740 That cmp_ok() may or may not pass depending on the datasets involved. In other
4741 words the C<WHERE> condition would apply to the entire dataset, just like
4742 it would in regular SQL. If you want to add a condition only to the "right side"
4743 of a C<LEFT JOIN> - consider declaring and using a L<relationship with a custom
4744 condition|DBIx::Class::Relationship::Base/condition>
4748 =head1 DBIC BIND VALUES
4750 Because DBIC may need more information to bind values than just the column name
4751 and value itself, it uses a special format for both passing and receiving bind
4752 values. Each bind value should be composed of an arrayref of
4753 C<< [ \%args => $val ] >>. The format of C<< \%args >> is currently:
4759 If present (in any form), this is what is being passed directly to bind_param.
4760 Note that different DBD's expect different bind args. (e.g. DBD::SQLite takes
4761 a single numerical type, while DBD::Pg takes a hashref if bind options.)
4763 If this is specified, all other bind options described below are ignored.
4767 If present, this is used to infer the actual bind attribute by passing to
4768 C<< $resolved_storage->bind_attribute_by_data_type() >>. Defaults to the
4769 "data_type" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>.
4771 Note that the data type is somewhat freeform (hence the sqlt_ prefix);
4772 currently drivers are expected to "Do the Right Thing" when given a common
4773 datatype name. (Not ideal, but that's what we got at this point.)
4777 Currently used to correctly allocate buffers for bind_param_inout().
4778 Defaults to "size" from the L<add_columns column info|DBIx::Class::ResultSource/add_columns>,
4779 or to a sensible value based on the "data_type".
4783 Used to fill in missing sqlt_datatype and sqlt_size attributes (if they are
4784 explicitly specified they are never overridden). Also used by some weird DBDs,
4785 where the column name should be available at bind_param time (e.g. Oracle).
4789 For backwards compatibility and convenience, the following shortcuts are
4792 [ $name => $val ] === [ { dbic_colname => $name }, $val ]
4793 [ \$dt => $val ] === [ { sqlt_datatype => $dt }, $val ]
4794 [ undef, $val ] === [ {}, $val ]
4795 $val === [ {}, $val ]
4797 =head1 AUTHOR AND CONTRIBUTORS
4799 See L<AUTHOR|DBIx::Class/AUTHOR> and L<CONTRIBUTORS|DBIx::Class/CONTRIBUTORS> in DBIx::Class
4803 You may distribute this code under the same terms as Perl itself.