1 package DBIx::Class::ResultSet;
9 use Carp::Clan qw/^DBIx::Class/;
10 use DBIx::Class::Exception;
13 use DBIx::Class::ResultSetColumn;
14 use DBIx::Class::ResultSourceHandle;
18 use base qw/DBIx::Class/;
20 #use Test::Deep::NoTest (qw/eq_deeply/);
21 use Data::Dumper::Concise;
23 __PACKAGE__->mk_group_accessors('simple' => qw/_result_class _source_handle/);
27 DBIx::Class::ResultSet - Represents a query used for fetching a set of results.
31 my $users_rs = $schema->resultset('User');
32 my $registered_users_rs = $schema->resultset('User')->search({ registered => 1 });
33 my @cds_in_2005 = $schema->resultset('CD')->search({ year => 2005 })->all();
37 A ResultSet is an object which stores a set of conditions representing
38 a query. It is the backbone of DBIx::Class (i.e. the really
39 important/useful bit).
41 No SQL is executed on the database when a ResultSet is created, it
42 just stores all the conditions needed to create the query.
44 A basic ResultSet representing the data of an entire table is returned
45 by calling C<resultset> on a L<DBIx::Class::Schema> and passing in a
46 L<Source|DBIx::Class::Manual::Glossary/Source> name.
48 my $users_rs = $schema->resultset('User');
50 A new ResultSet is returned from calling L</search> on an existing
51 ResultSet. The new one will contain all the conditions of the
52 original, plus any new conditions added in the C<search> call.
54 A ResultSet also incorporates an implicit iterator. L</next> and L</reset>
55 can be used to walk through all the L<DBIx::Class::Row>s the ResultSet
58 The query that the ResultSet represents is B<only> executed against
59 the database when these methods are called:
60 L</find> L</next> L</all> L</first> L</single> L</count>
64 =head2 Chaining resultsets
66 Let's say you've got a query that needs to be run to return some data
67 to the user. But, you have an authorization system in place that
68 prevents certain users from seeing certain information. So, you want
69 to construct the basic query in one method, but add constraints to it in
74 my $request = $self->get_request; # Get a request object somehow.
75 my $schema = $self->get_schema; # Get the DBIC schema object somehow.
77 my $cd_rs = $schema->resultset('CD')->search({
78 title => $request->param('title'),
79 year => $request->param('year'),
82 $self->apply_security_policy( $cd_rs );
87 sub apply_security_policy {
96 =head3 Resolving conditions and attributes
98 When a resultset is chained from another resultset, conditions and
99 attributes with the same keys need resolving.
101 L</join>, L</prefetch>, L</+select>, L</+as> attributes are merged
102 into the existing ones from the original resultset.
104 The L</where>, L</having> attribute, and any search conditions are
105 merged with an SQL C<AND> to the existing condition from the original
108 All other attributes are overridden by any new ones supplied in the
111 =head2 Multiple queries
113 Since a resultset just defines a query, you can do all sorts of
114 things with it with the same object.
116 # Don't hit the DB yet.
117 my $cd_rs = $schema->resultset('CD')->search({
118 title => 'something',
122 # Each of these hits the DB individually.
123 my $count = $cd_rs->count;
124 my $most_recent = $cd_rs->get_column('date_released')->max();
125 my @records = $cd_rs->all;
127 And it's not just limited to SELECT statements.
133 $cd_rs->create({ artist => 'Fred' });
135 Which is the same as:
137 $schema->resultset('CD')->create({
138 title => 'something',
143 See: L</search>, L</count>, L</get_column>, L</all>, L</create>.
147 If a resultset is used in a numeric context it returns the L</count>.
148 However, if it is used in a boolean context it is always true. So if
149 you want to check if a resultset has any results use C<if $rs != 0>.
150 C<if $rs> will always be true.
158 =item Arguments: $source, \%$attrs
160 =item Return Value: $rs
164 The resultset constructor. Takes a source object (usually a
165 L<DBIx::Class::ResultSourceProxy::Table>) and an attribute hash (see
166 L</ATTRIBUTES> below). Does not perform any queries -- these are
167 executed as needed by the other methods.
169 Generally you won't need to construct a resultset manually. You'll
170 automatically get one from e.g. a L</search> called in scalar context:
172 my $rs = $schema->resultset('CD')->search({ title => '100th Window' });
174 IMPORTANT: If called on an object, proxies to new_result instead so
176 my $cd = $schema->resultset('CD')->new({ title => 'Spoon' });
178 will return a CD object, not a ResultSet.
184 return $class->new_result(@_) if ref $class;
186 my ($source, $attrs) = @_;
187 $source = $source->handle
188 unless $source->isa('DBIx::Class::ResultSourceHandle');
189 $attrs = { %{$attrs||{}} };
191 if ($attrs->{page}) {
192 $attrs->{rows} ||= 10;
195 $attrs->{alias} ||= 'me';
197 # Creation of {} and bless separated to mitigate RH perl bug
198 # see https://bugzilla.redhat.com/show_bug.cgi?id=196836
200 _source_handle => $source,
201 cond => $attrs->{where},
210 $attrs->{result_class} || $source->resolve->result_class
220 =item Arguments: $cond, \%attrs?
222 =item Return Value: $resultset (scalar context), @row_objs (list context)
226 my @cds = $cd_rs->search({ year => 2001 }); # "... WHERE year = 2001"
227 my $new_rs = $cd_rs->search({ year => 2005 });
229 my $new_rs = $cd_rs->search([ { year => 2005 }, { year => 2004 } ]);
230 # year = 2005 OR year = 2004
232 If you need to pass in additional attributes but no additional condition,
233 call it as C<search(undef, \%attrs)>.
235 # "SELECT name, artistid FROM $artist_table"
236 my @all_artists = $schema->resultset('Artist')->search(undef, {
237 columns => [qw/name artistid/],
240 For a list of attributes that can be passed to C<search>, see
241 L</ATTRIBUTES>. For more examples of using this function, see
242 L<Searching|DBIx::Class::Manual::Cookbook/Searching>. For a complete
243 documentation for the first argument, see L<SQL::Abstract>.
245 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
251 my $rs = $self->search_rs( @_ );
252 return (wantarray ? $rs->all : $rs);
259 =item Arguments: $cond, \%attrs?
261 =item Return Value: $resultset
265 This method does the same exact thing as search() except it will
266 always return a resultset, even in list context.
273 # Special-case handling for (undef, undef).
274 if ( @_ == 2 && !defined $_[1] && !defined $_[0] ) {
279 $attrs = pop(@_) if @_ > 1 and ref $_[$#_] eq 'HASH';
280 my $our_attrs = { %{$self->{attrs}} };
281 my $having = delete $our_attrs->{having};
282 my $where = delete $our_attrs->{where};
286 my %safe = (alias => 1, cache => 1);
289 (@_ && defined($_[0])) # @_ == () or (undef)
291 (keys %$attrs # empty attrs or only 'safe' attrs
292 && List::Util::first { !$safe{$_} } keys %$attrs)
294 # no search, effectively just a clone
295 $rows = $self->get_cache;
298 # reset the selector list
299 if (List::Util::first { exists $attrs->{$_} } qw{columns select as}) {
300 delete @{$our_attrs}{qw{select as columns +select +as +columns include_columns}};
303 my $new_attrs = { %{$our_attrs}, %{$attrs} };
305 # merge new attrs into inherited
306 foreach my $key (qw/join prefetch +select +as +columns include_columns bind/) {
307 next unless exists $attrs->{$key};
308 $new_attrs->{$key} = $self->_merge_attr($our_attrs->{$key}, $attrs->{$key});
313 (@_ == 1 || ref $_[0] eq "HASH")
315 (ref $_[0] eq 'HASH')
317 (keys %{ $_[0] } > 0)
325 ? $self->throw_exception("Odd number of arguments to search")
332 if (defined $where) {
333 $new_attrs->{where} = (
334 defined $new_attrs->{where}
337 ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_
338 } $where, $new_attrs->{where}
345 $new_attrs->{where} = (
346 defined $new_attrs->{where}
349 ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_
350 } $cond, $new_attrs->{where}
356 if (defined $having) {
357 $new_attrs->{having} = (
358 defined $new_attrs->{having}
361 ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_
362 } $having, $new_attrs->{having}
368 my $rs = (ref $self)->new($self->result_source, $new_attrs);
370 $rs->set_cache($rows) if ($rows);
375 =head2 search_literal
379 =item Arguments: $sql_fragment, @bind_values
381 =item Return Value: $resultset (scalar context), @row_objs (list context)
385 my @cds = $cd_rs->search_literal('year = ? AND title = ?', qw/2001 Reload/);
386 my $newrs = $artist_rs->search_literal('name = ?', 'Metallica');
388 Pass a literal chunk of SQL to be added to the conditional part of the
391 CAVEAT: C<search_literal> is provided for Class::DBI compatibility and should
392 only be used in that context. C<search_literal> is a convenience method.
393 It is equivalent to calling $schema->search(\[]), but if you want to ensure
394 columns are bound correctly, use C<search>.
396 Example of how to use C<search> instead of C<search_literal>
398 my @cds = $cd_rs->search_literal('cdid = ? AND (artist = ? OR artist = ?)', (2, 1, 2));
399 my @cds = $cd_rs->search(\[ 'cdid = ? AND (artist = ? OR artist = ?)', [ 'cdid', 2 ], [ 'artist', 1 ], [ 'artist', 2 ] ]);
402 See L<DBIx::Class::Manual::Cookbook/Searching> and
403 L<DBIx::Class::Manual::FAQ/Searching> for searching techniques that do not
404 require C<search_literal>.
409 my ($self, $sql, @bind) = @_;
411 if ( @bind && ref($bind[-1]) eq 'HASH' ) {
414 return $self->search(\[ $sql, map [ __DUMMY__ => $_ ], @bind ], ($attr || () ));
421 =item Arguments: @values | \%cols, \%attrs?
423 =item Return Value: $row_object | undef
427 Finds a row based on its primary key or unique constraint. For example, to find
428 a row by its primary key:
430 my $cd = $schema->resultset('CD')->find(5);
432 You can also find a row by a specific unique constraint using the C<key>
433 attribute. For example:
435 my $cd = $schema->resultset('CD')->find('Massive Attack', 'Mezzanine', {
436 key => 'cd_artist_title'
439 Additionally, you can specify the columns explicitly by name:
441 my $cd = $schema->resultset('CD')->find(
443 artist => 'Massive Attack',
444 title => 'Mezzanine',
446 { key => 'cd_artist_title' }
449 If the C<key> is specified as C<primary>, it searches only on the primary key.
451 If no C<key> is specified, it searches on all unique constraints defined on the
452 source for which column data is provided, including the primary key.
454 If your table does not have a primary key, you B<must> provide a value for the
455 C<key> attribute matching one of the unique constraints on the source.
457 In addition to C<key>, L</find> recognizes and applies standard
458 L<resultset attributes|/ATTRIBUTES> in the same way as L</search> does.
460 Note: If your query does not return only one row, a warning is generated:
462 Query returned more than one row
464 See also L</find_or_create> and L</update_or_create>. For information on how to
465 declare unique constraints, see
466 L<DBIx::Class::ResultSource/add_unique_constraint>.
472 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
474 # Default to the primary key, but allow a specific key
475 my @cols = exists $attrs->{key}
476 ? $self->result_source->unique_constraint_columns($attrs->{key})
477 : $self->result_source->primary_columns;
478 $self->throw_exception(
479 "Can't find unless a primary key is defined or unique constraint is specified"
482 # Parse out a hashref from input
484 if (ref $_[0] eq 'HASH') {
485 $input_query = { %{$_[0]} };
487 elsif (@_ == @cols) {
489 @{$input_query}{@cols} = @_;
492 # Compatibility: Allow e.g. find(id => $value)
493 carp "Find by key => value deprecated; please use a hashref instead";
497 my (%related, $info);
499 KEY: foreach my $key (keys %$input_query) {
500 if (ref($input_query->{$key})
501 && ($info = $self->result_source->relationship_info($key))) {
502 my $val = delete $input_query->{$key};
503 next KEY if (ref($val) eq 'ARRAY'); # has_many for multi_create
504 my $rel_q = $self->result_source->_resolve_condition(
505 $info->{cond}, $val, $key
507 die "Can't handle OR join condition in find" if ref($rel_q) eq 'ARRAY';
508 @related{keys %$rel_q} = values %$rel_q;
511 if (my @keys = keys %related) {
512 @{$input_query}{@keys} = values %related;
516 # Build the final query: Default to the disjunction of the unique queries,
517 # but allow the input query in case the ResultSet defines the query or the
518 # user is abusing find
519 my $alias = exists $attrs->{alias} ? $attrs->{alias} : $self->{attrs}{alias};
521 if (exists $attrs->{key}) {
522 my @unique_cols = $self->result_source->unique_constraint_columns($attrs->{key});
523 my $unique_query = $self->_build_unique_query($input_query, \@unique_cols);
524 $query = $self->_add_alias($unique_query, $alias);
526 elsif ($self->{attrs}{accessor} and $self->{attrs}{accessor} eq 'single') {
527 # This means that we got here after a merger of relationship conditions
528 # in ::Relationship::Base::search_related (the row method), and furthermore
529 # the relationship is of the 'single' type. This means that the condition
530 # provided by the relationship (already attached to $self) is sufficient,
531 # as there can be only one row in the database that would satisfy the
535 my @unique_queries = $self->_unique_queries($input_query, $attrs);
536 $query = @unique_queries
537 ? [ map { $self->_add_alias($_, $alias) } @unique_queries ]
538 : $self->_add_alias($input_query, $alias);
542 my $rs = $self->search ($query, {result_class => $self->result_class, %$attrs});
543 if ($rs->_resolved_attrs->{collapse}) {
545 carp "Query returned more than one row" if $rs->next;
555 # Add the specified alias to the specified query hash. A copy is made so the
556 # original query is not modified.
559 my ($self, $query, $alias) = @_;
561 my %aliased = %$query;
562 foreach my $col (grep { ! m/\./ } keys %aliased) {
563 $aliased{"$alias.$col"} = delete $aliased{$col};
571 # Build a list of queries which satisfy unique constraints.
573 sub _unique_queries {
574 my ($self, $query, $attrs) = @_;
576 my @constraint_names = exists $attrs->{key}
578 : $self->result_source->unique_constraint_names;
580 my $where = $self->_collapse_cond($self->{attrs}{where} || {});
581 my $num_where = scalar keys %$where;
583 my (@unique_queries, %seen_column_combinations);
584 foreach my $name (@constraint_names) {
585 my @constraint_cols = $self->result_source->unique_constraint_columns($name);
587 my $constraint_sig = join "\x00", sort @constraint_cols;
588 next if $seen_column_combinations{$constraint_sig}++;
590 my $unique_query = $self->_build_unique_query($query, \@constraint_cols);
592 my $num_cols = scalar @constraint_cols;
593 my $num_query = scalar keys %$unique_query;
595 my $total = $num_query + $num_where;
596 if ($num_query && ($num_query == $num_cols || $total == $num_cols)) {
597 # The query is either unique on its own or is unique in combination with
598 # the existing where clause
599 push @unique_queries, $unique_query;
603 return @unique_queries;
606 # _build_unique_query
608 # Constrain the specified query hash based on the specified column names.
610 sub _build_unique_query {
611 my ($self, $query, $unique_cols) = @_;
614 map { $_ => $query->{$_} }
615 grep { exists $query->{$_} }
620 =head2 search_related
624 =item Arguments: $rel, $cond, \%attrs?
626 =item Return Value: $new_resultset
630 $new_rs = $cd_rs->search_related('artist', {
634 Searches the specified relationship, optionally specifying a condition and
635 attributes for matching records. See L</ATTRIBUTES> for more information.
640 return shift->related_resultset(shift)->search(@_);
643 =head2 search_related_rs
645 This method works exactly the same as search_related, except that
646 it guarantees a resultset, even in list context.
650 sub search_related_rs {
651 return shift->related_resultset(shift)->search_rs(@_);
658 =item Arguments: none
660 =item Return Value: $cursor
664 Returns a storage-driven cursor to the given resultset. See
665 L<DBIx::Class::Cursor> for more information.
672 my $attrs = $self->_resolved_attrs_copy;
674 return $self->{cursor}
675 ||= $self->result_source->storage->select($attrs->{from}, $attrs->{select},
676 $attrs->{where},$attrs);
683 =item Arguments: $cond?
685 =item Return Value: $row_object?
689 my $cd = $schema->resultset('CD')->single({ year => 2001 });
691 Inflates the first result without creating a cursor if the resultset has
692 any records in it; if not returns nothing. Used by L</find> as a lean version of
695 While this method can take an optional search condition (just like L</search>)
696 being a fast-code-path it does not recognize search attributes. If you need to
697 add extra joins or similar, call L</search> and then chain-call L</single> on the
698 L<DBIx::Class::ResultSet> returned.
704 As of 0.08100, this method enforces the assumption that the preceding
705 query returns only one row. If more than one row is returned, you will receive
708 Query returned more than one row
710 In this case, you should be using L</next> or L</find> instead, or if you really
711 know what you are doing, use the L</rows> attribute to explicitly limit the size
714 This method will also throw an exception if it is called on a resultset prefetching
715 has_many, as such a prefetch implies fetching multiple rows from the database in
716 order to assemble the resulting object.
723 my ($self, $where) = @_;
725 $self->throw_exception('single() only takes search conditions, no attributes. You want ->search( $cond, $attrs )->single()');
728 my $attrs = $self->_resolved_attrs_copy;
730 if ($attrs->{collapse}) {
731 $self->throw_exception(
732 'single() can not be used on resultsets prefetching has_many. Use find( \%cond ) or next() instead'
737 if (defined $attrs->{where}) {
740 [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ }
741 $where, delete $attrs->{where} ]
744 $attrs->{where} = $where;
748 # XXX: Disabled since it doesn't infer uniqueness in all cases
749 # unless ($self->_is_unique_query($attrs->{where})) {
750 # carp "Query not guaranteed to return a single row"
751 # . "; please declare your unique constraints or use search instead";
754 my @data = $self->result_source->storage->select_single(
755 $attrs->{from}, $attrs->{select},
756 $attrs->{where}, $attrs
760 ? ($self->_construct_objects(@data))[0]
768 # Try to determine if the specified query is guaranteed to be unique, based on
769 # the declared unique constraints.
771 sub _is_unique_query {
772 my ($self, $query) = @_;
774 my $collapsed = $self->_collapse_query($query);
775 my $alias = $self->{attrs}{alias};
777 foreach my $name ($self->result_source->unique_constraint_names) {
778 my @unique_cols = map {
780 } $self->result_source->unique_constraint_columns($name);
782 # Count the values for each unique column
783 my %seen = map { $_ => 0 } @unique_cols;
785 foreach my $key (keys %$collapsed) {
786 my $aliased = $key =~ /\./ ? $key : "$alias.$key";
787 next unless exists $seen{$aliased}; # Additional constraints are okay
788 $seen{$aliased} = scalar keys %{ $collapsed->{$key} };
791 # If we get 0 or more than 1 value for a column, it's not necessarily unique
792 return 1 unless grep { $_ != 1 } values %seen;
800 # Recursively collapse the query, accumulating values for each column.
802 sub _collapse_query {
803 my ($self, $query, $collapsed) = @_;
807 if (ref $query eq 'ARRAY') {
808 foreach my $subquery (@$query) {
809 next unless ref $subquery; # -or
810 $collapsed = $self->_collapse_query($subquery, $collapsed);
813 elsif (ref $query eq 'HASH') {
814 if (keys %$query and (keys %$query)[0] eq '-and') {
815 foreach my $subquery (@{$query->{-and}}) {
816 $collapsed = $self->_collapse_query($subquery, $collapsed);
820 foreach my $col (keys %$query) {
821 my $value = $query->{$col};
822 $collapsed->{$col}{$value}++;
834 =item Arguments: $cond?
836 =item Return Value: $resultsetcolumn
840 my $max_length = $rs->get_column('length')->max;
842 Returns a L<DBIx::Class::ResultSetColumn> instance for a column of the ResultSet.
847 my ($self, $column) = @_;
848 my $new = DBIx::Class::ResultSetColumn->new($self, $column);
856 =item Arguments: $cond, \%attrs?
858 =item Return Value: $resultset (scalar context), @row_objs (list context)
862 # WHERE title LIKE '%blue%'
863 $cd_rs = $rs->search_like({ title => '%blue%'});
865 Performs a search, but uses C<LIKE> instead of C<=> as the condition. Note
866 that this is simply a convenience method retained for ex Class::DBI users.
867 You most likely want to use L</search> with specific operators.
869 For more information, see L<DBIx::Class::Manual::Cookbook>.
871 This method is deprecated and will be removed in 0.09. Use L</search()>
872 instead. An example conversion is:
874 ->search_like({ foo => 'bar' });
878 ->search({ foo => { like => 'bar' } });
885 'search_like() is deprecated and will be removed in DBIC version 0.09.'
886 .' Instead use ->search({ x => { -like => "y%" } })'
887 .' (note the outer pair of {}s - they are important!)'
889 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
890 my $query = ref $_[0] eq 'HASH' ? { %{shift()} }: {@_};
891 $query->{$_} = { 'like' => $query->{$_} } for keys %$query;
892 return $class->search($query, { %$attrs });
899 =item Arguments: $first, $last
901 =item Return Value: $resultset (scalar context), @row_objs (list context)
905 Returns a resultset or object list representing a subset of elements from the
906 resultset slice is called on. Indexes are from 0, i.e., to get the first
909 my ($one, $two, $three) = $rs->slice(0, 2);
914 my ($self, $min, $max) = @_;
915 my $attrs = {}; # = { %{ $self->{attrs} || {} } };
916 $attrs->{offset} = $self->{attrs}{offset} || 0;
917 $attrs->{offset} += $min;
918 $attrs->{rows} = ($max ? ($max - $min + 1) : 1);
919 return $self->search(undef(), $attrs);
920 #my $slice = (ref $self)->new($self->result_source, $attrs);
921 #return (wantarray ? $slice->all : $slice);
928 =item Arguments: none
930 =item Return Value: $result?
934 Returns the next element in the resultset (C<undef> is there is none).
936 Can be used to efficiently iterate over records in the resultset:
938 my $rs = $schema->resultset('CD')->search;
939 while (my $cd = $rs->next) {
943 Note that you need to store the resultset object, and call C<next> on it.
944 Calling C<< resultset('Table')->next >> repeatedly will always return the
945 first record from the resultset.
951 if (my $cache = $self->get_cache) {
952 $self->{all_cache_position} ||= 0;
953 return $cache->[$self->{all_cache_position}++];
955 if ($self->{attrs}{cache}) {
956 $self->{all_cache_position} = 1;
957 return ($self->all)[0];
959 if ($self->{stashed_objects}) {
960 my $obj = shift(@{$self->{stashed_objects}});
961 delete $self->{stashed_objects} unless @{$self->{stashed_objects}};
965 exists $self->{stashed_row}
966 ? @{delete $self->{stashed_row}}
967 : $self->cursor->next
969 return undef unless (@row);
970 my ($row, @more) = $self->_construct_objects(@row);
971 $self->{stashed_objects} = \@more if @more;
975 # takes a single DBI-row of data and coinstructs as many objects
976 # as the resultset attributes call for.
977 # This can be a bit of an action at a distance - it takes as an argument
978 # the *current* cursor-row (already taken off the $sth), but if
979 # collapsing is requested it will keep advancing the cursor either
980 # until the current row-object is assembled (the collapser was able to
981 # order the result sensibly) OR until the cursor is exhausted (an
982 # unordered collapsing resultset effectively triggers ->all)
984 # FIXME: why the *FUCK* do we pass around DBI data by copy?! Sadly needs
985 # assessment before changing...
987 sub _construct_objects {
988 my ($self, @row) = @_;
989 my $attrs = $self->_resolved_attrs;
990 my $keep_collapsing = $attrs->{collapse};
995 my $me_pref_col = $attrs->{_row_parser}->($row_ref);
998 if ($keep_collapsing) {
1000 # FIXME - we should be able to remove these 2 checks after the design validates
1001 $self->throw_exception ('Collapsing without a top-level collapse-set... can not happen')
1002 unless @{$me_ref_col->[2]};
1003 $self->throw_exception ('Top-level collapse-set contains a NULL-value... can not happen')
1004 if grep { ! defined $_ } @{$me_pref_col->[2]};
1006 my $main_ident = join "\x00", @{$me_pref_col->[2]};
1008 if (! $res_index->{$main_ident}) {
1009 # this is where we bail out IFF we are ordered, and the $main_ident changes
1011 $res_index->{$main_ident} = {
1013 index => scalar keys %$res_index,
1020 $container = $res_index->{$main_ident}{container};
1023 push @$container, [ @{$me_pref_col}[0,1] ];
1030 do { $row_ref = [$self->cursor->next]; $self->{stashed_row} = $row_ref if @$row_ref; scalar @$row_ref }
1033 # attempt collapse all rows with same collapse identity
1034 if (@to_collapse > 1) {
1036 while (@to_collapse) {
1037 $self->_merge_result(\@collapsed, shift @to_collapse);
1042 my $mepref_structs = $self->_collapse_result($attrs->{as}, \@row, $keep_collapsing)
1045 my $rsrc = $self->result_source;
1046 my $res_class = $self->result_class;
1047 my $inflator = $res_class->can ('inflate_result');
1050 $res_class->$inflator ($rsrc, @$mepref_structs);
1052 if (my $f = $attrs->{record_filter}) {
1053 @objs = map { $f->($_) } @objs;
1060 sub _collapse_result {
1061 my ( $self, $as_proto, $row_ref, $keep_collapsing ) = @_;
1062 my $collapse = $self->_resolved_attrs->{collapse};
1063 my $parser = $self->result_source->_mk_row_parser( $as_proto, $collapse );
1066 my $rel_register = {};
1068 my @row = @$row_ref;
1070 my $row = $parser->( \@row );
1073 $self->_check_register( $register, $row ) unless ( keys %$register );
1075 $self->_merge_result( $result, $row, $rel_register )
1077 || ( $collapse = $self->_check_register( $register, $row ) ) );
1081 && do { @row = $self->cursor->next; $self->{stashed_row} = \@row if @row; }
1083 # run this as long as there is a next row and we are not yet done collapsing
1091 sub _check_register {
1092 my ( $self, $register, $obj ) = @_;
1093 return undef unless ( ref $obj eq 'ARRAY' && ref $obj->[2] eq 'ARRAY' );
1094 my @ids = @{ $obj->[2] };
1095 while ( defined( my $id = shift @ids ) ) {
1096 return $register->{$id} if ( exists $register->{$id} && !@ids );
1097 $register->{$id} = @ids ? {} : $obj unless ( exists $register->{$id} );
1098 $register = $register->{$id};
1105 my ( $self, $result, $row, $register ) = @_;
1106 return @$result = @$row if ( @$result == 0 ); # initialize with $row
1108 my ( undef, $rels, $ids ) = @$result;
1109 my ( undef, $new_rels, $new_ids ) = @$row;
1111 use List::MoreUtils;
1112 my @rels = List::MoreUtils::uniq( keys %$rels, keys %$new_rels );
1113 foreach my $rel (@rels) {
1114 $register = $register->{$rel} ||= {};
1116 my $new_data = $new_rels->{$rel};
1117 my $data = $rels->{$rel};
1118 @$data = [@$data] unless ( ref $data->[0] eq 'ARRAY' );
1120 $self->_check_register( $register, $data->[0] )
1121 unless ( keys %$register );
1123 if ( my $found = $self->_check_register( $register, $new_data ) ) {
1124 $self->_merge_result( $found, $new_data, $register );
1127 push( @$data, $new_data );
1138 # two arguments: $as_proto is an arrayref of column names,
1139 # $row_ref is an arrayref of the data. If none of the row data
1140 # is defined we return undef (that's copied from the old
1141 # _collapse_result). Next we decide whether we need to collapse
1142 # the resultset (i.e. we prefetch something) or not. $collapse
1143 # indicates that. The do-while loop will run once if we do not need
1144 # to collapse the result and will run as long as _merge_result returns
1145 # a true value. It will return undef if the current added row does not
1146 # match the previous row. A bit of stashing and cursor magic is
1147 # required so that the cursor is not mixed up.
1149 # "$rows" is a bit misleading. In the end, there should only be one
1150 # element in this arrayref.
1152 sub _collapse_result {
1153 my ( $self, $as_proto, $row_ref ) = @_;
1161 return undef unless $has_def;
1163 my $collapse = $self->_resolved_attrs->{collapse};
1165 my @row = @$row_ref;
1168 my $row = { map { $_ => $row[ $i++ ] } @$as_proto };
1169 $row = $self->result_source->_parse_row($row, $collapse);
1170 unless ( scalar @$rows ) {
1171 push( @$rows, $row );
1173 $collapse = undef unless ( $self->_merge_result( $rows, $row ) );
1176 && do { @row = $self->cursor->next; $self->{stashed_row} = \@row if @row; }
1183 # _merge_result accepts an arrayref of rows objects (again, an arrayref of two elements)
1184 # and a row object which should be merged into the first object.
1185 # First we try to find out whether $row is already in $rows. If this is the case
1186 # we try to merge them by iteration through their relationship data. We call
1187 # _merge_result again on them, so they get merged.
1189 # If we don't find the $row in $rows, we append it to $rows and return undef.
1190 # _merge_result returns 1 otherwise (i.e. $row has been found in $rows).
1193 my ( $self, $rows, $row ) = @_;
1194 my ( $columns, $rels ) = @$row;
1196 foreach my $seen (@$rows) {
1198 foreach my $column ( keys %$columns ) {
1199 if ( defined $seen->[0]->{$column} ^ defined $columns->{$column}
1200 or defined $columns->{$column}
1201 && $seen->[0]->{$column} ne $columns->{$column} )
1214 foreach my $rel ( keys %$rels ) {
1215 my $old_rows = $found->[1]->{$rel};
1216 $self->_merge_result(
1217 ref $found->[1]->{$rel}->[0] eq 'HASH' ? [ $found->[1]->{$rel} ]
1218 : $found->[1]->{$rel},
1219 ref $rels->{$rel}->[0] eq 'HASH' ? [ $rels->{$rel}->[0], $rels->{$rel}->[1] ]
1220 : $rels->{$rel}->[0]
1223 my $attrs = $self->_resolved_attrs;
1224 my ($keep_collapsing, $set_ident) = @{$attrs}{qw/collapse _collapse_ident/};
1226 # FIXME this is temporary, need to calculate in _resolved_attrs
1227 $set_ident ||= { me => [ $self->result_source->_pri_cols ], pref => {} };
1229 my @cur_row = @$row_ref;
1230 my (@to_collapse, $last_ident);
1233 my $row_hr = { map { $as_proto->[$_] => $cur_row[$_] } (0 .. $#$as_proto) };
1235 # see if we are switching to another object
1236 # this can be turned off and things will still work
1237 # since _merge_prefetch knows about _collapse_ident
1238 # my $cur_ident = [ @{$row_hr}{@$set_ident} ];
1240 $last_ident ||= $cur_ident;
1242 # if ($keep_collapsing = Test::Deep::eq_deeply ($cur_ident, $last_ident)) {
1243 # push @to_collapse, $self->result_source->_parse_row (
1250 do { @cur_row = $self->cursor->next; $self->{stashed_row} = \@cur_row if @cur_row; }
1253 die Dumper \@to_collapse;
1256 # attempt collapse all rows with same collapse identity
1257 if (@to_collapse > 1) {
1259 while (@to_collapse) {
1260 $self->_merge_result(\@collapsed, shift @to_collapse);
1262 @to_collapse = @collapsed;
1265 # still didn't fully collapse
1266 $self->throw_exception ('Resultset collapse failed (theoretically impossible). Maybe a wrong collapse_ident...?')
1267 if (@to_collapse > 1);
1269 return $to_collapse[0];
1273 # two arguments: $as_proto is an arrayref of 'as' column names,
1274 # $row_ref is an arrayref of the data. The do-while loop will run
1275 # once if we do not need to collapse the result and will run as long as
1276 # _merge_result returns a true value. It will return undef if the
1277 # current added row does not match the previous row, which in turn
1278 # means we need to stash the row for the subsequent ->next call
1279 sub _collapse_result {
1280 my ( $self, $as_proto, $row_ref ) = @_;
1282 my $attrs = $self->_resolved_attrs;
1283 my ($keep_collapsing, $set_ident) = @{$attrs}{qw/collapse _collapse_ident/};
1285 die Dumper [$as_proto, $row_ref, $keep_collapsing, $set_ident ];
1288 my @cur_row = @$row_ref;
1289 my (@to_collapse, $last_ident);
1292 my $row_hr = { map { $as_proto->[$_] => $cur_row[$_] } (0 .. $#$as_proto) };
1294 # see if we are switching to another object
1295 # this can be turned off and things will still work
1296 # since _merge_prefetch knows about _collapse_ident
1297 # my $cur_ident = [ @{$row_hr}{@$set_ident} ];
1299 $last_ident ||= $cur_ident;
1301 # if ($keep_collapsing = eq_deeply ($cur_ident, $last_ident)) {
1302 # push @to_collapse, $self->result_source->_parse_row (
1309 do { @cur_row = $self->cursor->next; $self->{stashed_row} = \@cur_row if @cur_row; }
1312 # attempt collapse all rows with same collapse identity
1316 # Takes an arrayref of me/pref pairs and a new me/pref pair that should
1317 # be merged on a preexisting matching me (or should be pushed into $merged
1318 # as a new me/pref pair for further invocations). It should be possible to
1319 # use this function to collapse complete ->all results, provided _collapse_result() is adjusted
1320 # to provide everything to this sub not to barf when $merged contains more than one
1322 sub _merge_prefetch {
1323 my ($self, $merged, $next_row) = @_;
1326 push @$merged, $next_row;
1332 =head2 result_source
1336 =item Arguments: $result_source?
1338 =item Return Value: $result_source
1342 An accessor for the primary ResultSource object from which this ResultSet
1349 =item Arguments: $result_class?
1351 =item Return Value: $result_class
1355 An accessor for the class to use when creating row objects. Defaults to
1356 C<< result_source->result_class >> - which in most cases is the name of the
1357 L<"table"|DBIx::Class::Manual::Glossary/"ResultSource"> class.
1359 Note that changing the result_class will also remove any components
1360 that were originally loaded in the source class via
1361 L<DBIx::Class::ResultSource/load_components>. Any overloaded methods
1362 in the original source class will not run.
1367 my ($self, $result_class) = @_;
1368 if ($result_class) {
1369 $self->ensure_class_loaded($result_class);
1370 $self->_result_class($result_class);
1371 $self->{attrs}{result_class} = $result_class if ref $self;
1373 $self->_result_class;
1380 =item Arguments: $cond, \%attrs??
1382 =item Return Value: $count
1386 Performs an SQL C<COUNT> with the same query as the resultset was built
1387 with to find the number of elements. Passing arguments is equivalent to
1388 C<< $rs->search ($cond, \%attrs)->count >>
1394 return $self->search(@_)->count if @_ and defined $_[0];
1395 return scalar @{ $self->get_cache } if $self->get_cache;
1397 my $attrs = $self->_resolved_attrs_copy;
1399 # this is a little optimization - it is faster to do the limit
1400 # adjustments in software, instead of a subquery
1401 my $rows = delete $attrs->{rows};
1402 my $offset = delete $attrs->{offset};
1405 if ($self->_has_resolved_attr (qw/collapse group_by/)) {
1406 $crs = $self->_count_subq_rs ($attrs);
1409 $crs = $self->_count_rs ($attrs);
1411 my $count = $crs->next;
1413 $count -= $offset if $offset;
1414 $count = $rows if $rows and $rows < $count;
1415 $count = 0 if ($count < 0);
1424 =item Arguments: $cond, \%attrs??
1426 =item Return Value: $count_rs
1430 Same as L</count> but returns a L<DBIx::Class::ResultSetColumn> object.
1431 This can be very handy for subqueries:
1433 ->search( { amount => $some_rs->count_rs->as_query } )
1435 As with regular resultsets the SQL query will be executed only after
1436 the resultset is accessed via L</next> or L</all>. That would return
1437 the same single value obtainable via L</count>.
1443 return $self->search(@_)->count_rs if @_;
1445 # this may look like a lack of abstraction (count() does about the same)
1446 # but in fact an _rs *must* use a subquery for the limits, as the
1447 # software based limiting can not be ported if this $rs is to be used
1448 # in a subquery itself (i.e. ->as_query)
1449 if ($self->_has_resolved_attr (qw/collapse group_by offset rows/)) {
1450 return $self->_count_subq_rs;
1453 return $self->_count_rs;
1458 # returns a ResultSetColumn object tied to the count query
1461 my ($self, $attrs) = @_;
1463 my $rsrc = $self->result_source;
1464 $attrs ||= $self->_resolved_attrs;
1466 my $tmp_attrs = { %$attrs };
1468 # take off any limits, record_filter is cdbi, and no point of ordering a count
1469 delete $tmp_attrs->{$_} for (qw/select as rows offset order_by record_filter/);
1471 # overwrite the selector (supplied by the storage)
1472 $tmp_attrs->{select} = $rsrc->storage->_count_select ($rsrc, $tmp_attrs);
1473 $tmp_attrs->{as} = 'count';
1475 my $tmp_rs = $rsrc->resultset_class->new($rsrc, $tmp_attrs)->get_column ('count');
1481 # same as above but uses a subquery
1483 sub _count_subq_rs {
1484 my ($self, $attrs) = @_;
1486 my $rsrc = $self->result_source;
1487 $attrs ||= $self->_resolved_attrs_copy;
1489 my $sub_attrs = { %$attrs };
1491 # extra selectors do not go in the subquery and there is no point of ordering it
1492 delete $sub_attrs->{$_} for qw/collapse select _prefetch_select as order_by/;
1494 # if we multi-prefetch we group_by primary keys only as this is what we would
1495 # get out of the rs via ->next/->all. We *DO WANT* to clobber old group_by regardless
1496 if ( $attrs->{collapse} ) {
1497 $sub_attrs->{group_by} = [ map { "$attrs->{alias}.$_" } ($rsrc->_pri_cols) ]
1500 $sub_attrs->{select} = $rsrc->storage->_subq_count_select ($rsrc, $attrs);
1502 # this is so that the query can be simplified e.g.
1503 # * ordering can be thrown away in things like Top limit
1504 $sub_attrs->{-for_count_only} = 1;
1506 my $sub_rs = $rsrc->resultset_class->new ($rsrc, $sub_attrs);
1509 -alias => 'count_subq',
1510 -source_handle => $rsrc->handle,
1511 count_subq => $sub_rs->as_query,
1514 # the subquery replaces this
1515 delete $attrs->{$_} for qw/where bind collapse group_by having having_bind rows offset/;
1517 return $self->_count_rs ($attrs);
1524 =head2 count_literal
1528 =item Arguments: $sql_fragment, @bind_values
1530 =item Return Value: $count
1534 Counts the results in a literal query. Equivalent to calling L</search_literal>
1535 with the passed arguments, then L</count>.
1539 sub count_literal { shift->search_literal(@_)->count; }
1545 =item Arguments: none
1547 =item Return Value: @objects
1551 Returns all elements in the resultset. Called implicitly if the resultset
1552 is returned in list context.
1559 $self->throw_exception("all() doesn't take any arguments, you probably wanted ->search(...)->all()");
1562 if (my $c = $self->get_cache) {
1568 if ($self->_resolved_attrs->{collapse}) {
1569 # Using $self->cursor->all is really just an optimisation.
1570 # If we're collapsing has_many prefetches it probably makes
1571 # very little difference, and this is cleaner than hacking
1572 # _construct_objects to survive the approach
1573 $self->cursor->reset;
1574 my @row = $self->cursor->next;
1576 push(@objects, $self->_construct_objects(@row));
1577 @row = (exists $self->{stashed_row}
1578 ? @{delete $self->{stashed_row}}
1579 : $self->cursor->next);
1582 @objects = map { $self->_construct_objects(@$_) } $self->cursor->all;
1585 $self->set_cache(\@objects) if $self->{attrs}{cache};
1594 =item Arguments: none
1596 =item Return Value: $self
1600 Resets the resultset's cursor, so you can iterate through the elements again.
1601 Implicitly resets the storage cursor, so a subsequent L</next> will trigger
1608 delete $self->{_attrs} if exists $self->{_attrs};
1609 $self->{all_cache_position} = 0;
1610 $self->cursor->reset;
1618 =item Arguments: none
1620 =item Return Value: $object?
1624 Resets the resultset and returns an object for the first result (if the
1625 resultset returns anything).
1630 return $_[0]->reset->next;
1636 # Determines whether and what type of subquery is required for the $rs operation.
1637 # If grouping is necessary either supplies its own, or verifies the current one
1638 # After all is done delegates to the proper storage method.
1640 sub _rs_update_delete {
1641 my ($self, $op, $values) = @_;
1643 my $rsrc = $self->result_source;
1645 # if a condition exists we need to strip all table qualifiers
1646 # if this is not possible we'll force a subquery below
1647 my $cond = $rsrc->schema->storage->_strip_cond_qualifiers ($self->{cond});
1649 my $needs_group_by_subq = $self->_has_resolved_attr (qw/collapse group_by -join/);
1650 my $needs_subq = $needs_group_by_subq || (not defined $cond) || $self->_has_resolved_attr(qw/row offset/);
1652 if ($needs_group_by_subq or $needs_subq) {
1654 # make a new $rs selecting only the PKs (that's all we really need)
1655 my $attrs = $self->_resolved_attrs_copy;
1657 delete $attrs->{$_} for qw/collapse select as/;
1658 $attrs->{columns} = [ map { "$attrs->{alias}.$_" } ($self->result_source->_pri_cols) ];
1660 if ($needs_group_by_subq) {
1661 # make sure no group_by was supplied, or if there is one - make sure it matches
1662 # the columns compiled above perfectly. Anything else can not be sanely executed
1663 # on most databases so croak right then and there
1665 if (my $g = $attrs->{group_by}) {
1666 my @current_group_by = map
1667 { $_ =~ /\./ ? $_ : "$attrs->{alias}.$_" }
1672 join ("\x00", sort @current_group_by)
1674 join ("\x00", sort @{$attrs->{columns}} )
1676 $self->throw_exception (
1677 "You have just attempted a $op operation on a resultset which does group_by"
1678 . ' on columns other than the primary keys, while DBIC internally needs to retrieve'
1679 . ' the primary keys in a subselect. All sane RDBMS engines do not support this'
1680 . ' kind of queries. Please retry the operation with a modified group_by or'
1681 . ' without using one at all.'
1686 $attrs->{group_by} = $attrs->{columns};
1690 my $subrs = (ref $self)->new($rsrc, $attrs);
1692 return $self->result_source->storage->_subq_update_delete($subrs, $op, $values);
1695 return $rsrc->storage->$op(
1697 $op eq 'update' ? $values : (),
1707 =item Arguments: \%values
1709 =item Return Value: $storage_rv
1713 Sets the specified columns in the resultset to the supplied values in a
1714 single query. Return value will be true if the update succeeded or false
1715 if no records were updated; exact type of success value is storage-dependent.
1720 my ($self, $values) = @_;
1721 $self->throw_exception('Values for update must be a hash')
1722 unless ref $values eq 'HASH';
1724 return $self->_rs_update_delete ('update', $values);
1731 =item Arguments: \%values
1733 =item Return Value: 1
1737 Fetches all objects and updates them one at a time. Note that C<update_all>
1738 will run DBIC cascade triggers, while L</update> will not.
1743 my ($self, $values) = @_;
1744 $self->throw_exception('Values for update_all must be a hash')
1745 unless ref $values eq 'HASH';
1746 foreach my $obj ($self->all) {
1747 $obj->set_columns($values)->update;
1756 =item Arguments: none
1758 =item Return Value: $storage_rv
1762 Deletes the contents of the resultset from its result source. Note that this
1763 will not run DBIC cascade triggers. See L</delete_all> if you need triggers
1764 to run. See also L<DBIx::Class::Row/delete>.
1766 Return value will be the amount of rows deleted; exact type of return value
1767 is storage-dependent.
1773 $self->throw_exception('delete does not accept any arguments')
1776 return $self->_rs_update_delete ('delete');
1783 =item Arguments: none
1785 =item Return Value: 1
1789 Fetches all objects and deletes them one at a time. Note that C<delete_all>
1790 will run DBIC cascade triggers, while L</delete> will not.
1796 $self->throw_exception('delete_all does not accept any arguments')
1799 $_->delete for $self->all;
1807 =item Arguments: \@data;
1811 Accepts either an arrayref of hashrefs or alternatively an arrayref of arrayrefs.
1812 For the arrayref of hashrefs style each hashref should be a structure suitable
1813 forsubmitting to a $resultset->create(...) method.
1815 In void context, C<insert_bulk> in L<DBIx::Class::Storage::DBI> is used
1816 to insert the data, as this is a faster method.
1818 Otherwise, each set of data is inserted into the database using
1819 L<DBIx::Class::ResultSet/create>, and the resulting objects are
1820 accumulated into an array. The array itself, or an array reference
1821 is returned depending on scalar or list context.
1823 Example: Assuming an Artist Class that has many CDs Classes relating:
1825 my $Artist_rs = $schema->resultset("Artist");
1827 ## Void Context Example
1828 $Artist_rs->populate([
1829 { artistid => 4, name => 'Manufactured Crap', cds => [
1830 { title => 'My First CD', year => 2006 },
1831 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
1834 { artistid => 5, name => 'Angsty-Whiny Girl', cds => [
1835 { title => 'My parents sold me to a record company', year => 2005 },
1836 { title => 'Why Am I So Ugly?', year => 2006 },
1837 { title => 'I Got Surgery and am now Popular', year => 2007 }
1842 ## Array Context Example
1843 my ($ArtistOne, $ArtistTwo, $ArtistThree) = $Artist_rs->populate([
1844 { name => "Artist One"},
1845 { name => "Artist Two"},
1846 { name => "Artist Three", cds=> [
1847 { title => "First CD", year => 2007},
1848 { title => "Second CD", year => 2008},
1852 print $ArtistOne->name; ## response is 'Artist One'
1853 print $ArtistThree->cds->count ## reponse is '2'
1855 For the arrayref of arrayrefs style, the first element should be a list of the
1856 fieldsnames to which the remaining elements are rows being inserted. For
1859 $Arstist_rs->populate([
1860 [qw/artistid name/],
1861 [100, 'A Formally Unknown Singer'],
1862 [101, 'A singer that jumped the shark two albums ago'],
1863 [102, 'An actually cool singer'],
1866 Please note an important effect on your data when choosing between void and
1867 wantarray context. Since void context goes straight to C<insert_bulk> in
1868 L<DBIx::Class::Storage::DBI> this will skip any component that is overriding
1869 C<insert>. So if you are using something like L<DBIx-Class-UUIDColumns> to
1870 create primary keys for you, you will find that your PKs are empty. In this
1871 case you will have to use the wantarray context in order to create those
1879 # cruft placed in standalone method
1880 my $data = $self->_normalize_populate_args(@_);
1882 if(defined wantarray) {
1884 foreach my $item (@$data) {
1885 push(@created, $self->create($item));
1887 return wantarray ? @created : \@created;
1889 my $first = $data->[0];
1891 # if a column is a registered relationship, and is a non-blessed hash/array, consider
1892 # it relationship data
1893 my (@rels, @columns);
1894 for (keys %$first) {
1895 my $ref = ref $first->{$_};
1896 $self->result_source->has_relationship($_) && ($ref eq 'ARRAY' or $ref eq 'HASH')
1902 my @pks = $self->result_source->primary_columns;
1904 ## do the belongs_to relationships
1905 foreach my $index (0..$#$data) {
1907 # delegate to create() for any dataset without primary keys with specified relationships
1908 if (grep { !defined $data->[$index]->{$_} } @pks ) {
1910 if (grep { ref $data->[$index]{$r} eq $_ } qw/HASH ARRAY/) { # a related set must be a HASH or AoH
1911 my @ret = $self->populate($data);
1917 foreach my $rel (@rels) {
1918 next unless ref $data->[$index]->{$rel} eq "HASH";
1919 my $result = $self->related_resultset($rel)->create($data->[$index]->{$rel});
1920 my ($reverse) = keys %{$self->result_source->reverse_relationship_info($rel)};
1921 my $related = $result->result_source->_resolve_condition(
1922 $result->result_source->relationship_info($reverse)->{cond},
1927 delete $data->[$index]->{$rel};
1928 $data->[$index] = {%{$data->[$index]}, %$related};
1930 push @columns, keys %$related if $index == 0;
1934 ## inherit the data locked in the conditions of the resultset
1935 my ($rs_data) = $self->_merge_cond_with_data({});
1936 delete @{$rs_data}{@columns};
1937 my @inherit_cols = keys %$rs_data;
1938 my @inherit_data = values %$rs_data;
1940 ## do bulk insert on current row
1941 $self->result_source->storage->insert_bulk(
1942 $self->result_source,
1943 [@columns, @inherit_cols],
1944 [ map { [ @$_{@columns}, @inherit_data ] } @$data ],
1947 ## do the has_many relationships
1948 foreach my $item (@$data) {
1950 foreach my $rel (@rels) {
1951 next unless $item->{$rel} && ref $item->{$rel} eq "ARRAY";
1953 my $parent = $self->find({map { $_ => $item->{$_} } @pks})
1954 || $self->throw_exception('Cannot find the relating object.');
1956 my $child = $parent->$rel;
1958 my $related = $child->result_source->_resolve_condition(
1959 $parent->result_source->relationship_info($rel)->{cond},
1964 my @rows_to_add = ref $item->{$rel} eq 'ARRAY' ? @{$item->{$rel}} : ($item->{$rel});
1965 my @populate = map { {%$_, %$related} } @rows_to_add;
1967 $child->populate( \@populate );
1974 # populate() argumnets went over several incarnations
1975 # What we ultimately support is AoH
1976 sub _normalize_populate_args {
1977 my ($self, $arg) = @_;
1979 if (ref $arg eq 'ARRAY') {
1980 if (ref $arg->[0] eq 'HASH') {
1983 elsif (ref $arg->[0] eq 'ARRAY') {
1985 my @colnames = @{$arg->[0]};
1986 foreach my $values (@{$arg}[1 .. $#$arg]) {
1987 push @ret, { map { $colnames[$_] => $values->[$_] } (0 .. $#colnames) };
1993 $self->throw_exception('Populate expects an arrayref of hashrefs or arrayref of arrayrefs');
2000 =item Arguments: none
2002 =item Return Value: $pager
2006 Return Value a L<Data::Page> object for the current resultset. Only makes
2007 sense for queries with a C<page> attribute.
2009 To get the full count of entries for a paged resultset, call
2010 C<total_entries> on the L<Data::Page> object.
2017 return $self->{pager} if $self->{pager};
2019 my $attrs = $self->{attrs};
2020 $self->throw_exception("Can't create pager for non-paged rs")
2021 unless $self->{attrs}{page};
2022 $attrs->{rows} ||= 10;
2024 # throw away the paging flags and re-run the count (possibly
2025 # with a subselect) to get the real total count
2026 my $count_attrs = { %$attrs };
2027 delete $count_attrs->{$_} for qw/rows offset page pager/;
2028 my $total_count = (ref $self)->new($self->result_source, $count_attrs)->count;
2030 return $self->{pager} = Data::Page->new(
2033 $self->{attrs}{page}
2041 =item Arguments: $page_number
2043 =item Return Value: $rs
2047 Returns a resultset for the $page_number page of the resultset on which page
2048 is called, where each page contains a number of rows equal to the 'rows'
2049 attribute set on the resultset (10 by default).
2054 my ($self, $page) = @_;
2055 return (ref $self)->new($self->result_source, { %{$self->{attrs}}, page => $page });
2062 =item Arguments: \%vals
2064 =item Return Value: $rowobject
2068 Creates a new row object in the resultset's result class and returns
2069 it. The row is not inserted into the database at this point, call
2070 L<DBIx::Class::Row/insert> to do that. Calling L<DBIx::Class::Row/in_storage>
2071 will tell you whether the row object has been inserted or not.
2073 Passes the hashref of input on to L<DBIx::Class::Row/new>.
2078 my ($self, $values) = @_;
2079 $self->throw_exception( "new_result needs a hash" )
2080 unless (ref $values eq 'HASH');
2082 my ($merged_cond, $cols_from_relations) = $self->_merge_cond_with_data($values);
2086 @$cols_from_relations
2087 ? (-cols_from_relations => $cols_from_relations)
2089 -source_handle => $self->_source_handle,
2090 -result_source => $self->result_source, # DO NOT REMOVE THIS, REQUIRED
2093 return $self->result_class->new(\%new);
2096 # _merge_cond_with_data
2098 # Takes a simple hash of K/V data and returns its copy merged with the
2099 # condition already present on the resultset. Additionally returns an
2100 # arrayref of value/condition names, which were inferred from related
2101 # objects (this is needed for in-memory related objects)
2102 sub _merge_cond_with_data {
2103 my ($self, $data) = @_;
2105 my (%new_data, @cols_from_relations);
2107 my $alias = $self->{attrs}{alias};
2109 if (! defined $self->{cond}) {
2110 # just massage $data below
2112 elsif ($self->{cond} eq $DBIx::Class::ResultSource::UNRESOLVABLE_CONDITION) {
2113 %new_data = %{ $self->{attrs}{related_objects} || {} }; # nothing might have been inserted yet
2114 @cols_from_relations = keys %new_data;
2116 elsif (ref $self->{cond} ne 'HASH') {
2117 $self->throw_exception(
2118 "Can't abstract implicit construct, resultset condition not a hash"
2122 # precendence must be given to passed values over values inherited from
2123 # the cond, so the order here is important.
2124 my $collapsed_cond = $self->_collapse_cond($self->{cond});
2125 my %implied = %{$self->_remove_alias($collapsed_cond, $alias)};
2127 while ( my($col, $value) = each %implied ) {
2128 if (ref($value) eq 'HASH' && keys(%$value) && (keys %$value)[0] eq '=') {
2129 $new_data{$col} = $value->{'='};
2132 $new_data{$col} = $value if $self->_is_deterministic_value($value);
2138 %{ $self->_remove_alias($data, $alias) },
2141 return (\%new_data, \@cols_from_relations);
2144 # _is_deterministic_value
2146 # Make an effor to strip non-deterministic values from the condition,
2147 # to make sure new_result chokes less
2149 sub _is_deterministic_value {
2152 my $ref_type = ref $value;
2153 return 1 if $ref_type eq '' || $ref_type eq 'SCALAR';
2154 return 1 if Scalar::Util::blessed($value);
2158 # _has_resolved_attr
2160 # determines if the resultset defines at least one
2161 # of the attributes supplied
2163 # used to determine if a subquery is neccessary
2165 # supports some virtual attributes:
2167 # This will scan for any joins being present on the resultset.
2168 # It is not a mere key-search but a deep inspection of {from}
2171 sub _has_resolved_attr {
2172 my ($self, @attr_names) = @_;
2174 my $attrs = $self->_resolved_attrs;
2178 for my $n (@attr_names) {
2179 if (grep { $n eq $_ } (qw/-join/) ) {
2180 $extra_checks{$n}++;
2184 my $attr = $attrs->{$n};
2186 next if not defined $attr;
2188 if (ref $attr eq 'HASH') {
2189 return 1 if keys %$attr;
2191 elsif (ref $attr eq 'ARRAY') {
2199 # a resolved join is expressed as a multi-level from
2201 $extra_checks{-join}
2203 ref $attrs->{from} eq 'ARRAY'
2205 @{$attrs->{from}} > 1
2213 # Recursively collapse the condition.
2215 sub _collapse_cond {
2216 my ($self, $cond, $collapsed) = @_;
2220 if (ref $cond eq 'ARRAY') {
2221 foreach my $subcond (@$cond) {
2222 next unless ref $subcond; # -or
2223 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2226 elsif (ref $cond eq 'HASH') {
2227 if (keys %$cond and (keys %$cond)[0] eq '-and') {
2228 foreach my $subcond (@{$cond->{-and}}) {
2229 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2233 foreach my $col (keys %$cond) {
2234 my $value = $cond->{$col};
2235 $collapsed->{$col} = $value;
2245 # Remove the specified alias from the specified query hash. A copy is made so
2246 # the original query is not modified.
2249 my ($self, $query, $alias) = @_;
2251 my %orig = %{ $query || {} };
2254 foreach my $key (keys %orig) {
2256 $unaliased{$key} = $orig{$key};
2259 $unaliased{$1} = $orig{$key}
2260 if $key =~ m/^(?:\Q$alias\E\.)?([^.]+)$/;
2270 =item Arguments: none
2272 =item Return Value: \[ $sql, @bind ]
2276 Returns the SQL query and bind vars associated with the invocant.
2278 This is generally used as the RHS for a subquery.
2285 my $attrs = $self->_resolved_attrs_copy;
2290 # my ($sql, \@bind, \%dbi_bind_attrs) = _select_args_to_query (...)
2291 # $sql also has no wrapping parenthesis in list ctx
2293 my $sqlbind = $self->result_source->storage
2294 ->_select_args_to_query ($attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs);
2303 =item Arguments: \%vals, \%attrs?
2305 =item Return Value: $rowobject
2309 my $artist = $schema->resultset('Artist')->find_or_new(
2310 { artist => 'fred' }, { key => 'artists' });
2312 $cd->cd_to_producer->find_or_new({ producer => $producer },
2313 { key => 'primary });
2315 Find an existing record from this resultset, based on its primary
2316 key, or a unique constraint. If none exists, instantiate a new result
2317 object and return it. The object will not be saved into your storage
2318 until you call L<DBIx::Class::Row/insert> on it.
2320 You most likely want this method when looking for existing rows using
2321 a unique constraint that is not the primary key, or looking for
2324 If you want objects to be saved immediately, use L</find_or_create>
2327 B<Note>: Take care when using C<find_or_new> with a table having
2328 columns with default values that you intend to be automatically
2329 supplied by the database (e.g. an auto_increment primary key column).
2330 In normal usage, the value of such columns should NOT be included at
2331 all in the call to C<find_or_new>, even when set to C<undef>.
2337 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2338 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2339 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2342 return $self->new_result($hash);
2349 =item Arguments: \%vals
2351 =item Return Value: a L<DBIx::Class::Row> $object
2355 Attempt to create a single new row or a row with multiple related rows
2356 in the table represented by the resultset (and related tables). This
2357 will not check for duplicate rows before inserting, use
2358 L</find_or_create> to do that.
2360 To create one row for this resultset, pass a hashref of key/value
2361 pairs representing the columns of the table and the values you wish to
2362 store. If the appropriate relationships are set up, foreign key fields
2363 can also be passed an object representing the foreign row, and the
2364 value will be set to its primary key.
2366 To create related objects, pass a hashref of related-object column values
2367 B<keyed on the relationship name>. If the relationship is of type C<multi>
2368 (L<DBIx::Class::Relationship/has_many>) - pass an arrayref of hashrefs.
2369 The process will correctly identify columns holding foreign keys, and will
2370 transparently populate them from the keys of the corresponding relation.
2371 This can be applied recursively, and will work correctly for a structure
2372 with an arbitrary depth and width, as long as the relationships actually
2373 exists and the correct column data has been supplied.
2376 Instead of hashrefs of plain related data (key/value pairs), you may
2377 also pass new or inserted objects. New objects (not inserted yet, see
2378 L</new>), will be inserted into their appropriate tables.
2380 Effectively a shortcut for C<< ->new_result(\%vals)->insert >>.
2382 Example of creating a new row.
2384 $person_rs->create({
2385 name=>"Some Person",
2386 email=>"somebody@someplace.com"
2389 Example of creating a new row and also creating rows in a related C<has_many>
2390 or C<has_one> resultset. Note Arrayref.
2393 { artistid => 4, name => 'Manufactured Crap', cds => [
2394 { title => 'My First CD', year => 2006 },
2395 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2400 Example of creating a new row and also creating a row in a related
2401 C<belongs_to>resultset. Note Hashref.
2404 title=>"Music for Silly Walks",
2407 name=>"Silly Musician",
2415 When subclassing ResultSet never attempt to override this method. Since
2416 it is a simple shortcut for C<< $self->new_result($attrs)->insert >>, a
2417 lot of the internals simply never call it, so your override will be
2418 bypassed more often than not. Override either L<new|DBIx::Class::Row/new>
2419 or L<insert|DBIx::Class::Row/insert> depending on how early in the
2420 L</create> process you need to intervene.
2427 my ($self, $attrs) = @_;
2428 $self->throw_exception( "create needs a hashref" )
2429 unless ref $attrs eq 'HASH';
2430 return $self->new_result($attrs)->insert;
2433 =head2 find_or_create
2437 =item Arguments: \%vals, \%attrs?
2439 =item Return Value: $rowobject
2443 $cd->cd_to_producer->find_or_create({ producer => $producer },
2444 { key => 'primary' });
2446 Tries to find a record based on its primary key or unique constraints; if none
2447 is found, creates one and returns that instead.
2449 my $cd = $schema->resultset('CD')->find_or_create({
2451 artist => 'Massive Attack',
2452 title => 'Mezzanine',
2456 Also takes an optional C<key> attribute, to search by a specific key or unique
2457 constraint. For example:
2459 my $cd = $schema->resultset('CD')->find_or_create(
2461 artist => 'Massive Attack',
2462 title => 'Mezzanine',
2464 { key => 'cd_artist_title' }
2467 B<Note>: Because find_or_create() reads from the database and then
2468 possibly inserts based on the result, this method is subject to a race
2469 condition. Another process could create a record in the table after
2470 the find has completed and before the create has started. To avoid
2471 this problem, use find_or_create() inside a transaction.
2473 B<Note>: Take care when using C<find_or_create> with a table having
2474 columns with default values that you intend to be automatically
2475 supplied by the database (e.g. an auto_increment primary key column).
2476 In normal usage, the value of such columns should NOT be included at
2477 all in the call to C<find_or_create>, even when set to C<undef>.
2479 See also L</find> and L</update_or_create>. For information on how to declare
2480 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2484 sub find_or_create {
2486 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2487 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2488 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2491 return $self->create($hash);
2494 =head2 update_or_create
2498 =item Arguments: \%col_values, { key => $unique_constraint }?
2500 =item Return Value: $rowobject
2504 $resultset->update_or_create({ col => $val, ... });
2506 First, searches for an existing row matching one of the unique constraints
2507 (including the primary key) on the source of this resultset. If a row is
2508 found, updates it with the other given column values. Otherwise, creates a new
2511 Takes an optional C<key> attribute to search on a specific unique constraint.
2514 # In your application
2515 my $cd = $schema->resultset('CD')->update_or_create(
2517 artist => 'Massive Attack',
2518 title => 'Mezzanine',
2521 { key => 'cd_artist_title' }
2524 $cd->cd_to_producer->update_or_create({
2525 producer => $producer,
2532 If no C<key> is specified, it searches on all unique constraints defined on the
2533 source, including the primary key.
2535 If the C<key> is specified as C<primary>, it searches only on the primary key.
2537 See also L</find> and L</find_or_create>. For information on how to declare
2538 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2540 B<Note>: Take care when using C<update_or_create> with a table having
2541 columns with default values that you intend to be automatically
2542 supplied by the database (e.g. an auto_increment primary key column).
2543 In normal usage, the value of such columns should NOT be included at
2544 all in the call to C<update_or_create>, even when set to C<undef>.
2548 sub update_or_create {
2550 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2551 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2553 my $row = $self->find($cond, $attrs);
2555 $row->update($cond);
2559 return $self->create($cond);
2562 =head2 update_or_new
2566 =item Arguments: \%col_values, { key => $unique_constraint }?
2568 =item Return Value: $rowobject
2572 $resultset->update_or_new({ col => $val, ... });
2574 First, searches for an existing row matching one of the unique constraints
2575 (including the primary key) on the source of this resultset. If a row is
2576 found, updates it with the other given column values. Otherwise, instantiate
2577 a new result object and return it. The object will not be saved into your storage
2578 until you call L<DBIx::Class::Row/insert> on it.
2580 Takes an optional C<key> attribute to search on a specific unique constraint.
2583 # In your application
2584 my $cd = $schema->resultset('CD')->update_or_new(
2586 artist => 'Massive Attack',
2587 title => 'Mezzanine',
2590 { key => 'cd_artist_title' }
2593 if ($cd->in_storage) {
2594 # the cd was updated
2597 # the cd is not yet in the database, let's insert it
2601 B<Note>: Take care when using C<update_or_new> with a table having
2602 columns with default values that you intend to be automatically
2603 supplied by the database (e.g. an auto_increment primary key column).
2604 In normal usage, the value of such columns should NOT be included at
2605 all in the call to C<update_or_new>, even when set to C<undef>.
2607 See also L</find>, L</find_or_create> and L</find_or_new>.
2613 my $attrs = ( @_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {} );
2614 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2616 my $row = $self->find( $cond, $attrs );
2617 if ( defined $row ) {
2618 $row->update($cond);
2622 return $self->new_result($cond);
2629 =item Arguments: none
2631 =item Return Value: \@cache_objects?
2635 Gets the contents of the cache for the resultset, if the cache is set.
2637 The cache is populated either by using the L</prefetch> attribute to
2638 L</search> or by calling L</set_cache>.
2650 =item Arguments: \@cache_objects
2652 =item Return Value: \@cache_objects
2656 Sets the contents of the cache for the resultset. Expects an arrayref
2657 of objects of the same class as those produced by the resultset. Note that
2658 if the cache is set the resultset will return the cached objects rather
2659 than re-querying the database even if the cache attr is not set.
2661 The contents of the cache can also be populated by using the
2662 L</prefetch> attribute to L</search>.
2667 my ( $self, $data ) = @_;
2668 $self->throw_exception("set_cache requires an arrayref")
2669 if defined($data) && (ref $data ne 'ARRAY');
2670 $self->{all_cache} = $data;
2677 =item Arguments: none
2679 =item Return Value: []
2683 Clears the cache for the resultset.
2688 shift->set_cache(undef);
2695 =item Arguments: none
2697 =item Return Value: true, if the resultset has been paginated
2705 return !!$self->{attrs}{page};
2712 =item Arguments: none
2714 =item Return Value: true, if the resultset has been ordered with C<order_by>.
2722 return scalar $self->result_source->storage->_parse_order_by($self->{attrs}{order_by});
2725 =head2 related_resultset
2729 =item Arguments: $relationship_name
2731 =item Return Value: $resultset
2735 Returns a related resultset for the supplied relationship name.
2737 $artist_rs = $schema->resultset('CD')->related_resultset('Artist');
2741 sub related_resultset {
2742 my ($self, $rel) = @_;
2744 $self->{related_resultsets} ||= {};
2745 return $self->{related_resultsets}{$rel} ||= do {
2746 my $rsrc = $self->result_source;
2747 my $rel_info = $rsrc->relationship_info($rel);
2749 $self->throw_exception(
2750 "search_related: result source '" . $rsrc->source_name .
2751 "' has no such relationship $rel")
2754 my $attrs = $self->_chain_relationship($rel);
2756 my $join_count = $attrs->{seen_join}{$rel};
2758 my $alias = $self->result_source->storage
2759 ->relname_to_table_alias($rel, $join_count);
2761 # since this is search_related, and we already slid the select window inwards
2762 # (the select/as attrs were deleted in the beginning), we need to flip all
2763 # left joins to inner, so we get the expected results
2764 # read the comment on top of the actual function to see what this does
2765 $attrs->{from} = $rsrc->schema->storage->_straight_join_to_node ($attrs->{from}, $alias);
2768 #XXX - temp fix for result_class bug. There likely is a more elegant fix -groditi
2769 delete @{$attrs}{qw(result_class alias)};
2773 if (my $cache = $self->get_cache) {
2774 if ($cache->[0] && $cache->[0]->related_resultset($rel)->get_cache) {
2775 $new_cache = [ map { @{$_->related_resultset($rel)->get_cache} }
2780 my $rel_source = $rsrc->related_source($rel);
2784 # The reason we do this now instead of passing the alias to the
2785 # search_rs below is that if you wrap/overload resultset on the
2786 # source you need to know what alias it's -going- to have for things
2787 # to work sanely (e.g. RestrictWithObject wants to be able to add
2788 # extra query restrictions, and these may need to be $alias.)
2790 my $rel_attrs = $rel_source->resultset_attributes;
2791 local $rel_attrs->{alias} = $alias;
2793 $rel_source->resultset
2797 where => $attrs->{where},
2800 $new->set_cache($new_cache) if $new_cache;
2805 =head2 current_source_alias
2809 =item Arguments: none
2811 =item Return Value: $source_alias
2815 Returns the current table alias for the result source this resultset is built
2816 on, that will be used in the SQL query. Usually it is C<me>.
2818 Currently the source alias that refers to the result set returned by a
2819 L</search>/L</find> family method depends on how you got to the resultset: it's
2820 C<me> by default, but eg. L</search_related> aliases it to the related result
2821 source name (and keeps C<me> referring to the original result set). The long
2822 term goal is to make L<DBIx::Class> always alias the current resultset as C<me>
2823 (and make this method unnecessary).
2825 Thus it's currently necessary to use this method in predefined queries (see
2826 L<DBIx::Class::Manual::Cookbook/Predefined searches>) when referring to the
2827 source alias of the current result set:
2829 # in a result set class
2831 my ($self, $user) = @_;
2833 my $me = $self->current_source_alias;
2835 return $self->search(
2836 "$me.modified" => $user->id,
2842 sub current_source_alias {
2845 return ($self->{attrs} || {})->{alias} || 'me';
2848 =head2 as_subselect_rs
2852 =item Arguments: none
2854 =item Return Value: $resultset
2858 Act as a barrier to SQL symbols. The resultset provided will be made into a
2859 "virtual view" by including it as a subquery within the from clause. From this
2860 point on, any joined tables are inaccessible to ->search on the resultset (as if
2861 it were simply where-filtered without joins). For example:
2863 my $rs = $schema->resultset('Bar')->search({'x.name' => 'abc'},{ join => 'x' });
2865 # 'x' now pollutes the query namespace
2867 # So the following works as expected
2868 my $ok_rs = $rs->search({'x.other' => 1});
2870 # But this doesn't: instead of finding a 'Bar' related to two x rows (abc and
2871 # def) we look for one row with contradictory terms and join in another table
2872 # (aliased 'x_2') which we never use
2873 my $broken_rs = $rs->search({'x.name' => 'def'});
2875 my $rs2 = $rs->as_subselect_rs;
2877 # doesn't work - 'x' is no longer accessible in $rs2, having been sealed away
2878 my $not_joined_rs = $rs2->search({'x.other' => 1});
2880 # works as expected: finds a 'table' row related to two x rows (abc and def)
2881 my $correctly_joined_rs = $rs2->search({'x.name' => 'def'});
2883 Another example of when one might use this would be to select a subset of
2884 columns in a group by clause:
2886 my $rs = $schema->resultset('Bar')->search(undef, {
2887 group_by => [qw{ id foo_id baz_id }],
2888 })->as_subselect_rs->search(undef, {
2889 columns => [qw{ id foo_id }]
2892 In the above example normally columns would have to be equal to the group by,
2893 but because we isolated the group by into a subselect the above works.
2897 sub as_subselect_rs {
2900 return $self->result_source->resultset->search( undef, {
2901 alias => $self->current_source_alias,
2903 $self->current_source_alias => $self->as_query,
2904 -alias => $self->current_source_alias,
2905 -source_handle => $self->result_source->handle,
2910 # This code is called by search_related, and makes sure there
2911 # is clear separation between the joins before, during, and
2912 # after the relationship. This information is needed later
2913 # in order to properly resolve prefetch aliases (any alias
2914 # with a relation_chain_depth less than the depth of the
2915 # current prefetch is not considered)
2917 # The increments happen twice per join. An even number means a
2918 # relationship specified via a search_related, whereas an odd
2919 # number indicates a join/prefetch added via attributes
2921 # Also this code will wrap the current resultset (the one we
2922 # chain to) in a subselect IFF it contains limiting attributes
2923 sub _chain_relationship {
2924 my ($self, $rel) = @_;
2925 my $source = $self->result_source;
2926 my $attrs = { %{$self->{attrs}||{}} };
2928 # we need to take the prefetch the attrs into account before we
2929 # ->_resolve_join as otherwise they get lost - captainL
2930 my $join = $self->_merge_attr( $attrs->{join}, $attrs->{prefetch} );
2932 delete @{$attrs}{qw/join prefetch collapse distinct select as columns +select +as +columns/};
2934 my $seen = { %{ (delete $attrs->{seen_join}) || {} } };
2937 my @force_subq_attrs = qw/offset rows group_by having/;
2940 ($attrs->{from} && ref $attrs->{from} ne 'ARRAY')
2942 $self->_has_resolved_attr (@force_subq_attrs)
2944 # Nuke the prefetch (if any) before the new $rs attrs
2945 # are resolved (prefetch is useless - we are wrapping
2946 # a subquery anyway).
2947 my $rs_copy = $self->search;
2948 $rs_copy->{attrs}{join} = $self->_merge_attr (
2949 $rs_copy->{attrs}{join},
2950 delete $rs_copy->{attrs}{prefetch},
2954 -source_handle => $source->handle,
2955 -alias => $attrs->{alias},
2956 $attrs->{alias} => $rs_copy->as_query,
2958 delete @{$attrs}{@force_subq_attrs, 'where'};
2959 $seen->{-relation_chain_depth} = 0;
2961 elsif ($attrs->{from}) { #shallow copy suffices
2962 $from = [ @{$attrs->{from}} ];
2966 -source_handle => $source->handle,
2967 -alias => $attrs->{alias},
2968 $attrs->{alias} => $source->from,
2972 my $jpath = ($seen->{-relation_chain_depth})
2973 ? $from->[-1][0]{-join_path}
2976 my @requested_joins = $source->_resolve_join(
2983 push @$from, @requested_joins;
2985 $seen->{-relation_chain_depth}++;
2987 # if $self already had a join/prefetch specified on it, the requested
2988 # $rel might very well be already included. What we do in this case
2989 # is effectively a no-op (except that we bump up the chain_depth on
2990 # the join in question so we could tell it *is* the search_related)
2993 # we consider the last one thus reverse
2994 for my $j (reverse @requested_joins) {
2995 my ($last_j) = keys %{$j->[0]{-join_path}[-1]};
2996 if ($rel eq $last_j) {
2997 $j->[0]{-relation_chain_depth}++;
3003 unless ($already_joined) {
3004 push @$from, $source->_resolve_join(
3012 $seen->{-relation_chain_depth}++;
3014 return {%$attrs, from => $from, seen_join => $seen};
3017 # too many times we have to do $attrs = { %{$self->_resolved_attrs} }
3018 sub _resolved_attrs_copy {
3020 return { %{$self->_resolved_attrs (@_)} };
3023 sub _resolved_attrs {
3025 return $self->{_attrs} if $self->{_attrs};
3027 my $attrs = { %{ $self->{attrs} || {} } };
3028 my $source = $self->result_source;
3029 my $alias = $attrs->{alias};
3031 $attrs->{columns} ||= delete $attrs->{cols} if exists $attrs->{cols};
3034 # build columns (as long as select isn't set) into a set of as/select hashes
3035 unless ( $attrs->{select} ) {
3038 if ( ref $attrs->{columns} eq 'ARRAY' ) {
3039 @cols = @{ delete $attrs->{columns}}
3040 } elsif ( defined $attrs->{columns} ) {
3041 @cols = delete $attrs->{columns}
3043 @cols = $source->columns
3047 if ( ref $_ eq 'HASH' ) {
3050 my $key = /^\Q${alias}.\E(.+)$/
3056 push @colbits, { $key => $value };
3061 # add the additional columns on
3062 foreach (qw{include_columns +columns}) {
3063 if ( $attrs->{$_} ) {
3064 my @list = ( ref($attrs->{$_}) eq 'ARRAY' )
3065 ? @{ delete $attrs->{$_} }
3066 : delete $attrs->{$_};
3068 if ( ref($_) eq 'HASH' ) {
3071 my $key = ( split /\./, $_ )[-1];
3072 my $value = ( /\./ ? $_ : "$alias.$_" );
3073 push @colbits, { $key => $value };
3079 # start with initial select items
3080 if ( $attrs->{select} ) {
3082 ( ref $attrs->{select} eq 'ARRAY' )
3083 ? [ @{ $attrs->{select} } ]
3084 : [ $attrs->{select} ];
3086 if ( $attrs->{as} ) {
3089 ref $attrs->{as} eq 'ARRAY'
3090 ? [ @{ $attrs->{as} } ]
3094 $attrs->{as} = [ map {
3095 m/^\Q${alias}.\E(.+)$/
3098 } @{ $attrs->{select} }
3103 # otherwise we intialise select & as to empty
3104 $attrs->{select} = [];
3108 # now add colbits to select/as
3109 push @{ $attrs->{select} }, map values %{$_}, @colbits;
3110 push @{ $attrs->{as} }, map keys %{$_}, @colbits;
3112 if ( my $adds = delete $attrs->{'+select'} ) {
3113 $adds = [$adds] unless ref $adds eq 'ARRAY';
3114 push @{ $attrs->{select} },
3115 map { /\./ || ref $_ ? $_ : "$alias.$_" } @$adds;
3117 if ( my $adds = delete $attrs->{'+as'} ) {
3118 $adds = [$adds] unless ref $adds eq 'ARRAY';
3119 push @{ $attrs->{as} }, @$adds;
3122 $attrs->{from} ||= [{
3123 -source_handle => $source->handle,
3124 -alias => $self->{attrs}{alias},
3125 $self->{attrs}{alias} => $source->from,
3128 if ( $attrs->{join} || $attrs->{prefetch} ) {
3130 $self->throw_exception ('join/prefetch can not be used with a custom {from}')
3131 if ref $attrs->{from} ne 'ARRAY';
3133 my $join = delete $attrs->{join} || {};
3135 if ( defined $attrs->{prefetch} ) {
3136 $join = $self->_merge_attr( $join, $attrs->{prefetch} );
3139 $attrs->{from} = # have to copy here to avoid corrupting the original
3141 @{ $attrs->{from} },
3142 $source->_resolve_join(
3145 { %{ $attrs->{seen_join} || {} } },
3146 ( $attrs->{seen_join} && keys %{$attrs->{seen_join}})
3147 ? $attrs->{from}[-1][0]{-join_path}
3154 if ( defined $attrs->{order_by} ) {
3155 $attrs->{order_by} = (
3156 ref( $attrs->{order_by} ) eq 'ARRAY'
3157 ? [ @{ $attrs->{order_by} } ]
3158 : [ $attrs->{order_by} || () ]
3162 if ($attrs->{group_by} and ref $attrs->{group_by} ne 'ARRAY') {
3163 $attrs->{group_by} = [ $attrs->{group_by} ];
3166 # generate the distinct induced group_by early, as prefetch will be carried via a
3167 # subquery (since a group_by is present)
3168 if (delete $attrs->{distinct}) {
3169 if ($attrs->{group_by}) {
3170 carp ("Useless use of distinct on a grouped resultset ('distinct' is ignored when a 'group_by' is present)");
3173 $attrs->{group_by} = [ grep { !ref($_) || (ref($_) ne 'HASH') } @{$attrs->{select}} ];
3175 # add any order_by parts that are not already present in the group_by
3176 # we need to be careful not to add any named functions/aggregates
3177 # i.e. select => [ ... { count => 'foo', -as 'foocount' } ... ]
3178 my %already_grouped = map { $_ => 1 } (@{$attrs->{group_by}});
3180 my $storage = $self->result_source->schema->storage;
3182 my $rs_column_list = $storage->_resolve_column_info ($attrs->{from});
3184 for my $chunk ($storage->_parse_order_by($attrs->{order_by})) {
3185 if ($rs_column_list->{$chunk} && not $already_grouped{$chunk}++) {
3186 push @{$attrs->{group_by}}, $chunk;
3192 # generate selections based on the prefetch helper
3193 if ( my $prefetch = delete $attrs->{prefetch} ) {
3194 $attrs->{collapse} = 1;
3196 # this is a separate structure (we don't look in {from} directly)
3197 # as the resolver needs to shift things off the lists to work
3198 # properly (identical-prefetches on different branches)
3200 if (ref $attrs->{from} eq 'ARRAY') {
3202 my $start_depth = $attrs->{seen_join}{-relation_chain_depth} || 0;
3204 for my $j ( @{$attrs->{from}}[1 .. $#{$attrs->{from}} ] ) {
3205 next unless $j->[0]{-alias};
3206 next unless $j->[0]{-join_path};
3207 next if ($j->[0]{-relation_chain_depth} || 0) < $start_depth;
3209 my @jpath = map { keys %$_ } @{$j->[0]{-join_path}};
3212 $p = $p->{$_} ||= {} for @jpath[ ($start_depth/2) .. $#jpath]; #only even depths are actual jpath boundaries
3213 push @{$p->{-join_aliases} }, $j->[0]{-alias};
3217 my @prefetch = $source->_resolve_prefetch( $prefetch, $alias, $join_map );
3219 # we need to somehow mark which columns came from prefetch
3220 $attrs->{_prefetch_select} = [ map { $_->[0] } @prefetch ];
3222 push @{ $attrs->{select} }, @{$attrs->{_prefetch_select}};
3223 push @{ $attrs->{as} }, (map { $_->[1] } @prefetch);
3226 # run through the resulting joinstructure (starting from our current slot)
3227 # and unset collapse if proven unnesessary
3228 if ($attrs->{collapse} && ref $attrs->{from} eq 'ARRAY') {
3230 if (@{$attrs->{from}} > 1) {
3232 # find where our table-spec starts and consider only things after us
3233 my @fromlist = @{$attrs->{from}};
3235 my $t = shift @fromlist;
3236 $t = $t->[0] if ref $t eq 'ARRAY'; #me vs join from-spec mismatch
3237 last if ($t->{-alias} && $t->{-alias} eq $alias);
3241 $attrs->{collapse} = ! $_->[0]{-is_single}
3246 # no joins - no collapse
3247 $attrs->{collapse} = 0;
3251 # if both page and offset are specified, produce a combined offset
3252 # even though it doesn't make much sense, this is what pre 081xx has
3254 if (my $page = delete $attrs->{page}) {
3256 ($attrs->{rows} * ($page - 1))
3258 ($attrs->{offset} || 0)
3262 return $self->{_attrs} = $attrs;
3266 my ($self, $attr) = @_;
3268 if (ref $attr eq 'HASH') {
3269 return $self->_rollout_hash($attr);
3270 } elsif (ref $attr eq 'ARRAY') {
3271 return $self->_rollout_array($attr);
3277 sub _rollout_array {
3278 my ($self, $attr) = @_;
3281 foreach my $element (@{$attr}) {
3282 if (ref $element eq 'HASH') {
3283 push( @rolled_array, @{ $self->_rollout_hash( $element ) } );
3284 } elsif (ref $element eq 'ARRAY') {
3285 # XXX - should probably recurse here
3286 push( @rolled_array, @{$self->_rollout_array($element)} );
3288 push( @rolled_array, $element );
3291 return \@rolled_array;
3295 my ($self, $attr) = @_;
3298 foreach my $key (keys %{$attr}) {
3299 push( @rolled_array, { $key => $attr->{$key} } );
3301 return \@rolled_array;
3304 sub _calculate_score {
3305 my ($self, $a, $b) = @_;
3307 if (defined $a xor defined $b) {
3310 elsif (not defined $a) {
3314 if (ref $b eq 'HASH') {
3315 my ($b_key) = keys %{$b};
3316 if (ref $a eq 'HASH') {
3317 my ($a_key) = keys %{$a};
3318 if ($a_key eq $b_key) {
3319 return (1 + $self->_calculate_score( $a->{$a_key}, $b->{$b_key} ));
3324 return ($a eq $b_key) ? 1 : 0;
3327 if (ref $a eq 'HASH') {
3328 my ($a_key) = keys %{$a};
3329 return ($b eq $a_key) ? 1 : 0;
3331 return ($b eq $a) ? 1 : 0;
3337 my ($self, $orig, $import) = @_;
3339 return $import unless defined($orig);
3340 return $orig unless defined($import);
3342 $orig = $self->_rollout_attr($orig);
3343 $import = $self->_rollout_attr($import);
3346 foreach my $import_element ( @{$import} ) {
3347 # find best candidate from $orig to merge $b_element into
3348 my $best_candidate = { position => undef, score => 0 }; my $position = 0;
3349 foreach my $orig_element ( @{$orig} ) {
3350 my $score = $self->_calculate_score( $orig_element, $import_element );
3351 if ($score > $best_candidate->{score}) {
3352 $best_candidate->{position} = $position;
3353 $best_candidate->{score} = $score;
3357 my ($import_key) = ( ref $import_element eq 'HASH' ) ? keys %{$import_element} : ($import_element);
3359 if ($best_candidate->{score} == 0 || exists $seen_keys->{$import_key}) {
3360 push( @{$orig}, $import_element );
3362 my $orig_best = $orig->[$best_candidate->{position}];
3363 # merge orig_best and b_element together and replace original with merged
3364 if (ref $orig_best ne 'HASH') {
3365 $orig->[$best_candidate->{position}] = $import_element;
3366 } elsif (ref $import_element eq 'HASH') {
3367 my ($key) = keys %{$orig_best};
3368 $orig->[$best_candidate->{position}] = { $key => $self->_merge_attr($orig_best->{$key}, $import_element->{$key}) };
3371 $seen_keys->{$import_key} = 1; # don't merge the same key twice
3381 $self->_source_handle($_[0]->handle);
3383 $self->_source_handle->resolve;
3387 =head2 throw_exception
3389 See L<DBIx::Class::Schema/throw_exception> for details.
3393 sub throw_exception {
3396 if (ref $self && $self->_source_handle->schema) {
3397 $self->_source_handle->schema->throw_exception(@_)
3400 DBIx::Class::Exception->throw(@_);
3404 # XXX: FIXME: Attributes docs need clearing up
3408 Attributes are used to refine a ResultSet in various ways when
3409 searching for data. They can be passed to any method which takes an
3410 C<\%attrs> argument. See L</search>, L</search_rs>, L</find>,
3413 These are in no particular order:
3419 =item Value: ( $order_by | \@order_by | \%order_by )
3423 Which column(s) to order the results by.
3425 [The full list of suitable values is documented in
3426 L<SQL::Abstract/"ORDER BY CLAUSES">; the following is a summary of
3429 If a single column name, or an arrayref of names is supplied, the
3430 argument is passed through directly to SQL. The hashref syntax allows
3431 for connection-agnostic specification of ordering direction:
3433 For descending order:
3435 order_by => { -desc => [qw/col1 col2 col3/] }
3437 For explicit ascending order:
3439 order_by => { -asc => 'col' }
3441 The old scalarref syntax (i.e. order_by => \'year DESC') is still
3442 supported, although you are strongly encouraged to use the hashref
3443 syntax as outlined above.
3449 =item Value: \@columns
3453 Shortcut to request a particular set of columns to be retrieved. Each
3454 column spec may be a string (a table column name), or a hash (in which
3455 case the key is the C<as> value, and the value is used as the C<select>
3456 expression). Adds C<me.> onto the start of any column without a C<.> in
3457 it and sets C<select> from that, then auto-populates C<as> from
3458 C<select> as normal. (You may also use the C<cols> attribute, as in
3459 earlier versions of DBIC.)
3465 =item Value: \@columns
3469 Indicates additional columns to be selected from storage. Works the same
3470 as L</columns> but adds columns to the selection. (You may also use the
3471 C<include_columns> attribute, as in earlier versions of DBIC). For
3474 $schema->resultset('CD')->search(undef, {
3475 '+columns' => ['artist.name'],
3479 would return all CDs and include a 'name' column to the information
3480 passed to object inflation. Note that the 'artist' is the name of the
3481 column (or relationship) accessor, and 'name' is the name of the column
3482 accessor in the related table.
3484 =head2 include_columns
3488 =item Value: \@columns
3492 Deprecated. Acts as a synonym for L</+columns> for backward compatibility.
3498 =item Value: \@select_columns
3502 Indicates which columns should be selected from the storage. You can use
3503 column names, or in the case of RDBMS back ends, function or stored procedure
3506 $rs = $schema->resultset('Employee')->search(undef, {
3509 { count => 'employeeid' },
3514 When you use function/stored procedure names and do not supply an C<as>
3515 attribute, the column names returned are storage-dependent. E.g. MySQL would
3516 return a column named C<count(employeeid)> in the above example.
3518 B<NOTE:> You will almost always need a corresponding 'as' entry when you use
3525 Indicates additional columns to be selected from storage. Works the same as
3526 L</select> but adds columns to the selection.
3534 Indicates additional column names for those added via L</+select>. See L</as>.
3542 =item Value: \@inflation_names
3546 Indicates column names for object inflation. That is, C<as>
3547 indicates the name that the column can be accessed as via the
3548 C<get_column> method (or via the object accessor, B<if one already
3549 exists>). It has nothing to do with the SQL code C<SELECT foo AS bar>.
3551 The C<as> attribute is used in conjunction with C<select>,
3552 usually when C<select> contains one or more function or stored
3555 $rs = $schema->resultset('Employee')->search(undef, {
3558 { count => 'employeeid' }
3560 as => ['name', 'employee_count'],
3563 my $employee = $rs->first(); # get the first Employee
3565 If the object against which the search is performed already has an accessor
3566 matching a column name specified in C<as>, the value can be retrieved using
3567 the accessor as normal:
3569 my $name = $employee->name();
3571 If on the other hand an accessor does not exist in the object, you need to
3572 use C<get_column> instead:
3574 my $employee_count = $employee->get_column('employee_count');
3576 You can create your own accessors if required - see
3577 L<DBIx::Class::Manual::Cookbook> for details.
3579 Please note: This will NOT insert an C<AS employee_count> into the SQL
3580 statement produced, it is used for internal access only. Thus
3581 attempting to use the accessor in an C<order_by> clause or similar
3582 will fail miserably.
3584 To get around this limitation, you can supply literal SQL to your
3585 C<select> attribute that contains the C<AS alias> text, e.g.
3587 select => [\'myfield AS alias']
3593 =item Value: ($rel_name | \@rel_names | \%rel_names)
3597 Contains a list of relationships that should be joined for this query. For
3600 # Get CDs by Nine Inch Nails
3601 my $rs = $schema->resultset('CD')->search(
3602 { 'artist.name' => 'Nine Inch Nails' },
3603 { join => 'artist' }
3606 Can also contain a hash reference to refer to the other relation's relations.
3609 package MyApp::Schema::Track;
3610 use base qw/DBIx::Class/;
3611 __PACKAGE__->table('track');
3612 __PACKAGE__->add_columns(qw/trackid cd position title/);
3613 __PACKAGE__->set_primary_key('trackid');
3614 __PACKAGE__->belongs_to(cd => 'MyApp::Schema::CD');
3617 # In your application
3618 my $rs = $schema->resultset('Artist')->search(
3619 { 'track.title' => 'Teardrop' },
3621 join => { cd => 'track' },
3622 order_by => 'artist.name',
3626 You need to use the relationship (not the table) name in conditions,
3627 because they are aliased as such. The current table is aliased as "me", so
3628 you need to use me.column_name in order to avoid ambiguity. For example:
3630 # Get CDs from 1984 with a 'Foo' track
3631 my $rs = $schema->resultset('CD')->search(
3634 'tracks.name' => 'Foo'
3636 { join => 'tracks' }
3639 If the same join is supplied twice, it will be aliased to <rel>_2 (and
3640 similarly for a third time). For e.g.
3642 my $rs = $schema->resultset('Artist')->search({
3643 'cds.title' => 'Down to Earth',
3644 'cds_2.title' => 'Popular',
3646 join => [ qw/cds cds/ ],
3649 will return a set of all artists that have both a cd with title 'Down
3650 to Earth' and a cd with title 'Popular'.
3652 If you want to fetch related objects from other tables as well, see C<prefetch>
3655 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
3661 =item Value: ($rel_name | \@rel_names | \%rel_names)
3665 Contains one or more relationships that should be fetched along with
3666 the main query (when they are accessed afterwards the data will
3667 already be available, without extra queries to the database). This is
3668 useful for when you know you will need the related objects, because it
3669 saves at least one query:
3671 my $rs = $schema->resultset('Tag')->search(
3680 The initial search results in SQL like the following:
3682 SELECT tag.*, cd.*, artist.* FROM tag
3683 JOIN cd ON tag.cd = cd.cdid
3684 JOIN artist ON cd.artist = artist.artistid
3686 L<DBIx::Class> has no need to go back to the database when we access the
3687 C<cd> or C<artist> relationships, which saves us two SQL statements in this
3690 Simple prefetches will be joined automatically, so there is no need
3691 for a C<join> attribute in the above search.
3693 C<prefetch> can be used with the following relationship types: C<belongs_to>,
3694 C<has_one> (or if you're using C<add_relationship>, any relationship declared
3695 with an accessor type of 'single' or 'filter'). A more complex example that
3696 prefetches an artists cds, the tracks on those cds, and the tags associated
3697 with that artist is given below (assuming many-to-many from artists to tags):
3699 my $rs = $schema->resultset('Artist')->search(
3703 { cds => 'tracks' },
3704 { artist_tags => 'tags' }
3710 B<NOTE:> If you specify a C<prefetch> attribute, the C<join> and C<select>
3711 attributes will be ignored.
3713 B<CAVEATs>: Prefetch does a lot of deep magic. As such, it may not behave
3714 exactly as you might expect.
3720 Prefetch uses the L</cache> to populate the prefetched relationships. This
3721 may or may not be what you want.
3725 If you specify a condition on a prefetched relationship, ONLY those
3726 rows that match the prefetched condition will be fetched into that relationship.
3727 This means that adding prefetch to a search() B<may alter> what is returned by
3728 traversing a relationship. So, if you have C<< Artist->has_many(CDs) >> and you do
3730 my $artist_rs = $schema->resultset('Artist')->search({
3736 my $count = $artist_rs->first->cds->count;
3738 my $artist_rs_prefetch = $artist_rs->search( {}, { prefetch => 'cds' } );
3740 my $prefetch_count = $artist_rs_prefetch->first->cds->count;
3742 cmp_ok( $count, '==', $prefetch_count, "Counts should be the same" );
3744 that cmp_ok() may or may not pass depending on the datasets involved. This
3745 behavior may or may not survive the 0.09 transition.
3757 Makes the resultset paged and specifies the page to retrieve. Effectively
3758 identical to creating a non-pages resultset and then calling ->page($page)
3761 If L<rows> attribute is not specified it defaults to 10 rows per page.
3763 When you have a paged resultset, L</count> will only return the number
3764 of rows in the page. To get the total, use the L</pager> and call
3765 C<total_entries> on it.
3775 Specifies the maximum number of rows for direct retrieval or the number of
3776 rows per page if the page attribute or method is used.
3782 =item Value: $offset
3786 Specifies the (zero-based) row number for the first row to be returned, or the
3787 of the first row of the first page if paging is used.
3793 =item Value: \@columns
3797 A arrayref of columns to group by. Can include columns of joined tables.
3799 group_by => [qw/ column1 column2 ... /]
3805 =item Value: $condition
3809 HAVING is a select statement attribute that is applied between GROUP BY and
3810 ORDER BY. It is applied to the after the grouping calculations have been
3813 having => { 'count(employee)' => { '>=', 100 } }
3819 =item Value: (0 | 1)
3823 Set to 1 to group by all columns. If the resultset already has a group_by
3824 attribute, this setting is ignored and an appropriate warning is issued.
3830 Adds to the WHERE clause.
3832 # only return rows WHERE deleted IS NULL for all searches
3833 __PACKAGE__->resultset_attributes({ where => { deleted => undef } }); )
3835 Can be overridden by passing C<< { where => undef } >> as an attribute
3842 Set to 1 to cache search results. This prevents extra SQL queries if you
3843 revisit rows in your ResultSet:
3845 my $resultset = $schema->resultset('Artist')->search( undef, { cache => 1 } );
3847 while( my $artist = $resultset->next ) {
3851 $rs->first; # without cache, this would issue a query
3853 By default, searches are not cached.
3855 For more examples of using these attributes, see
3856 L<DBIx::Class::Manual::Cookbook>.
3862 =item Value: ( 'update' | 'shared' )
3866 Set to 'update' for a SELECT ... FOR UPDATE or 'shared' for a SELECT