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) = @_;
990 my $attrs = $self->_resolved_attrs;
991 my $keep_collapsing = $attrs->{collapse};
996 my $me_pref_col = $attrs->{_row_parser}->($row_ref);
999 if ($keep_collapsing) {
1001 # FIXME - we should be able to remove these 2 checks after the design validates
1002 $self->throw_exception ('Collapsing without a top-level collapse-set... can not happen')
1003 unless @{$me_ref_col->[2]};
1004 $self->throw_exception ('Top-level collapse-set contains a NULL-value... can not happen')
1005 if grep { ! defined $_ } @{$me_pref_col->[2]};
1007 my $main_ident = join "\x00", @{$me_pref_col->[2]};
1009 if (! $res_index->{$main_ident}) {
1010 # this is where we bail out IFF we are ordered, and the $main_ident changes
1012 $res_index->{$main_ident} = {
1014 index => scalar keys %$res_index,
1021 $container = $res_index->{$main_ident}{container};
1024 push @$container, [ @{$me_pref_col}[0,1] ];
1031 do { $row_ref = [$self->cursor->next]; $self->{stashed_row} = $row_ref if @$row_ref; scalar @$row_ref }
1034 # attempt collapse all rows with same collapse identity
1035 if (@to_collapse > 1) {
1037 while (@to_collapse) {
1038 $self->_merge_result(\@collapsed, shift @to_collapse);
1043 my $mepref_structs = $self->_collapse_result(\@row)
1046 my $rsrc = $self->result_source;
1047 my $res_class = $self->result_class;
1048 my $inflator = $res_class->can ('inflate_result');
1051 $res_class->$inflator ($rsrc, @$_)
1052 } (@$mepref_structs);
1054 if (my $f = $attrs->{record_filter}) {
1055 @objs = map { $f->($_) } @objs;
1063 # two arguments: $as_proto is an arrayref of column names,
1064 # $row_ref is an arrayref of the data. If none of the row data
1065 # is defined we return undef (that's copied from the old
1066 # _collapse_result). Next we decide whether we need to collapse
1067 # the resultset (i.e. we prefetch something) or not. $collapse
1068 # indicates that. The do-while loop will run once if we do not need
1069 # to collapse the result and will run as long as _merge_result returns
1070 # a true value. It will return undef if the current added row does not
1071 # match the previous row. A bit of stashing and cursor magic is
1072 # required so that the cursor is not mixed up.
1074 # "$rows" is a bit misleading. In the end, there should only be one
1075 # element in this arrayref.
1077 sub _collapse_result {
1078 my ( $self, $as_proto, $row_ref ) = @_;
1086 return undef unless $has_def;
1088 my $collapse = $self->_resolved_attrs->{collapse};
1090 my @row = @$row_ref;
1093 my $row = { map { $_ => $row[ $i++ ] } @$as_proto };
1094 $row = $self->result_source->_parse_row($row, $collapse);
1095 unless ( scalar @$rows ) {
1096 push( @$rows, $row );
1098 $collapse = undef unless ( $self->_merge_result( $rows, $row ) );
1101 && do { @row = $self->cursor->next; $self->{stashed_row} = \@row if @row; }
1108 # _merge_result accepts an arrayref of rows objects (again, an arrayref of two elements)
1109 # and a row object which should be merged into the first object.
1110 # First we try to find out whether $row is already in $rows. If this is the case
1111 # we try to merge them by iteration through their relationship data. We call
1112 # _merge_result again on them, so they get merged.
1114 # If we don't find the $row in $rows, we append it to $rows and return undef.
1115 # _merge_result returns 1 otherwise (i.e. $row has been found in $rows).
1118 my ( $self, $rows, $row ) = @_;
1119 my ( $columns, $rels ) = @$row;
1121 foreach my $seen (@$rows) {
1123 foreach my $column ( keys %$columns ) {
1124 if ( defined $seen->[0]->{$column} ^ defined $columns->{$column}
1125 or defined $columns->{$column}
1126 && $seen->[0]->{$column} ne $columns->{$column} )
1139 foreach my $rel ( keys %$rels ) {
1140 my $old_rows = $found->[1]->{$rel};
1141 $self->_merge_result(
1142 ref $found->[1]->{$rel}->[0] eq 'HASH' ? [ $found->[1]->{$rel} ]
1143 : $found->[1]->{$rel},
1144 ref $rels->{$rel}->[0] eq 'HASH' ? [ $rels->{$rel}->[0], $rels->{$rel}->[1] ]
1145 : $rels->{$rel}->[0]
1148 my $attrs = $self->_resolved_attrs;
1149 my ($keep_collapsing, $set_ident) = @{$attrs}{qw/collapse _collapse_ident/};
1151 # FIXME this is temporary, need to calculate in _resolved_attrs
1152 $set_ident ||= { me => [ $self->result_source->_pri_cols ], pref => {} };
1154 my @cur_row = @$row_ref;
1155 my (@to_collapse, $last_ident);
1158 my $row_hr = { map { $as_proto->[$_] => $cur_row[$_] } (0 .. $#$as_proto) };
1160 # see if we are switching to another object
1161 # this can be turned off and things will still work
1162 # since _merge_prefetch knows about _collapse_ident
1163 # my $cur_ident = [ @{$row_hr}{@$set_ident} ];
1165 $last_ident ||= $cur_ident;
1167 # if ($keep_collapsing = Test::Deep::eq_deeply ($cur_ident, $last_ident)) {
1168 # push @to_collapse, $self->result_source->_parse_row (
1175 do { @cur_row = $self->cursor->next; $self->{stashed_row} = \@cur_row if @cur_row; }
1178 die Dumper \@to_collapse;
1181 # attempt collapse all rows with same collapse identity
1182 if (@to_collapse > 1) {
1184 while (@to_collapse) {
1185 $self->_merge_result(\@collapsed, shift @to_collapse);
1187 @to_collapse = @collapsed;
1190 # still didn't fully collapse
1191 $self->throw_exception ('Resultset collapse failed (theoretically impossible). Maybe a wrong collapse_ident...?')
1192 if (@to_collapse > 1);
1194 return $to_collapse[0];
1198 # two arguments: $as_proto is an arrayref of 'as' column names,
1199 # $row_ref is an arrayref of the data. The do-while loop will run
1200 # once if we do not need to collapse the result and will run as long as
1201 # _merge_result returns a true value. It will return undef if the
1202 # current added row does not match the previous row, which in turn
1203 # means we need to stash the row for the subsequent ->next call
1204 sub _collapse_result {
1205 my ( $self, $as_proto, $row_ref ) = @_;
1207 my $attrs = $self->_resolved_attrs;
1208 my ($keep_collapsing, $set_ident) = @{$attrs}{qw/collapse _collapse_ident/};
1210 die Dumper [$as_proto, $row_ref, $keep_collapsing, $set_ident ];
1213 my @cur_row = @$row_ref;
1214 my (@to_collapse, $last_ident);
1217 my $row_hr = { map { $as_proto->[$_] => $cur_row[$_] } (0 .. $#$as_proto) };
1219 # see if we are switching to another object
1220 # this can be turned off and things will still work
1221 # since _merge_prefetch knows about _collapse_ident
1222 # my $cur_ident = [ @{$row_hr}{@$set_ident} ];
1224 $last_ident ||= $cur_ident;
1226 # if ($keep_collapsing = eq_deeply ($cur_ident, $last_ident)) {
1227 # push @to_collapse, $self->result_source->_parse_row (
1234 do { @cur_row = $self->cursor->next; $self->{stashed_row} = \@cur_row if @cur_row; }
1237 # attempt collapse all rows with same collapse identity
1241 # Takes an arrayref of me/pref pairs and a new me/pref pair that should
1242 # be merged on a preexisting matching me (or should be pushed into $merged
1243 # as a new me/pref pair for further invocations). It should be possible to
1244 # use this function to collapse complete ->all results, provided _collapse_result() is adjusted
1245 # to provide everything to this sub not to barf when $merged contains more than one
1247 sub _merge_prefetch {
1248 my ($self, $merged, $next_row) = @_;
1251 push @$merged, $next_row;
1257 =head2 result_source
1261 =item Arguments: $result_source?
1263 =item Return Value: $result_source
1267 An accessor for the primary ResultSource object from which this ResultSet
1274 =item Arguments: $result_class?
1276 =item Return Value: $result_class
1280 An accessor for the class to use when creating row objects. Defaults to
1281 C<< result_source->result_class >> - which in most cases is the name of the
1282 L<"table"|DBIx::Class::Manual::Glossary/"ResultSource"> class.
1284 Note that changing the result_class will also remove any components
1285 that were originally loaded in the source class via
1286 L<DBIx::Class::ResultSource/load_components>. Any overloaded methods
1287 in the original source class will not run.
1292 my ($self, $result_class) = @_;
1293 if ($result_class) {
1294 $self->ensure_class_loaded($result_class);
1295 $self->_result_class($result_class);
1296 $self->{attrs}{result_class} = $result_class if ref $self;
1298 $self->_result_class;
1305 =item Arguments: $cond, \%attrs??
1307 =item Return Value: $count
1311 Performs an SQL C<COUNT> with the same query as the resultset was built
1312 with to find the number of elements. Passing arguments is equivalent to
1313 C<< $rs->search ($cond, \%attrs)->count >>
1319 return $self->search(@_)->count if @_ and defined $_[0];
1320 return scalar @{ $self->get_cache } if $self->get_cache;
1322 my $attrs = $self->_resolved_attrs_copy;
1324 # this is a little optimization - it is faster to do the limit
1325 # adjustments in software, instead of a subquery
1326 my $rows = delete $attrs->{rows};
1327 my $offset = delete $attrs->{offset};
1330 if ($self->_has_resolved_attr (qw/collapse group_by/)) {
1331 $crs = $self->_count_subq_rs ($attrs);
1334 $crs = $self->_count_rs ($attrs);
1336 my $count = $crs->next;
1338 $count -= $offset if $offset;
1339 $count = $rows if $rows and $rows < $count;
1340 $count = 0 if ($count < 0);
1349 =item Arguments: $cond, \%attrs??
1351 =item Return Value: $count_rs
1355 Same as L</count> but returns a L<DBIx::Class::ResultSetColumn> object.
1356 This can be very handy for subqueries:
1358 ->search( { amount => $some_rs->count_rs->as_query } )
1360 As with regular resultsets the SQL query will be executed only after
1361 the resultset is accessed via L</next> or L</all>. That would return
1362 the same single value obtainable via L</count>.
1368 return $self->search(@_)->count_rs if @_;
1370 # this may look like a lack of abstraction (count() does about the same)
1371 # but in fact an _rs *must* use a subquery for the limits, as the
1372 # software based limiting can not be ported if this $rs is to be used
1373 # in a subquery itself (i.e. ->as_query)
1374 if ($self->_has_resolved_attr (qw/collapse group_by offset rows/)) {
1375 return $self->_count_subq_rs;
1378 return $self->_count_rs;
1383 # returns a ResultSetColumn object tied to the count query
1386 my ($self, $attrs) = @_;
1388 my $rsrc = $self->result_source;
1389 $attrs ||= $self->_resolved_attrs;
1391 my $tmp_attrs = { %$attrs };
1393 # take off any limits, record_filter is cdbi, and no point of ordering a count
1394 delete $tmp_attrs->{$_} for (qw/select as rows offset order_by record_filter/);
1396 # overwrite the selector (supplied by the storage)
1397 $tmp_attrs->{select} = $rsrc->storage->_count_select ($rsrc, $tmp_attrs);
1398 $tmp_attrs->{as} = 'count';
1400 my $tmp_rs = $rsrc->resultset_class->new($rsrc, $tmp_attrs)->get_column ('count');
1406 # same as above but uses a subquery
1408 sub _count_subq_rs {
1409 my ($self, $attrs) = @_;
1411 my $rsrc = $self->result_source;
1412 $attrs ||= $self->_resolved_attrs_copy;
1414 my $sub_attrs = { %$attrs };
1416 # extra selectors do not go in the subquery and there is no point of ordering it
1417 delete $sub_attrs->{$_} for qw/collapse select _prefetch_select as order_by/;
1419 # if we multi-prefetch we group_by primary keys only as this is what we would
1420 # get out of the rs via ->next/->all. We *DO WANT* to clobber old group_by regardless
1421 if ( $attrs->{collapse} ) {
1422 $sub_attrs->{group_by} = [ map { "$attrs->{alias}.$_" } ($rsrc->_pri_cols) ]
1425 $sub_attrs->{select} = $rsrc->storage->_subq_count_select ($rsrc, $attrs);
1427 # this is so that the query can be simplified e.g.
1428 # * ordering can be thrown away in things like Top limit
1429 $sub_attrs->{-for_count_only} = 1;
1431 my $sub_rs = $rsrc->resultset_class->new ($rsrc, $sub_attrs);
1434 -alias => 'count_subq',
1435 -source_handle => $rsrc->handle,
1436 count_subq => $sub_rs->as_query,
1439 # the subquery replaces this
1440 delete $attrs->{$_} for qw/where bind collapse group_by having having_bind rows offset/;
1442 return $self->_count_rs ($attrs);
1449 =head2 count_literal
1453 =item Arguments: $sql_fragment, @bind_values
1455 =item Return Value: $count
1459 Counts the results in a literal query. Equivalent to calling L</search_literal>
1460 with the passed arguments, then L</count>.
1464 sub count_literal { shift->search_literal(@_)->count; }
1470 =item Arguments: none
1472 =item Return Value: @objects
1476 Returns all elements in the resultset. Called implicitly if the resultset
1477 is returned in list context.
1484 $self->throw_exception("all() doesn't take any arguments, you probably wanted ->search(...)->all()");
1487 if (my $c = $self->get_cache) {
1493 if ($self->_resolved_attrs->{collapse}) {
1494 # Using $self->cursor->all is really just an optimisation.
1495 # If we're collapsing has_many prefetches it probably makes
1496 # very little difference, and this is cleaner than hacking
1497 # _construct_objects to survive the approach
1498 $self->cursor->reset;
1499 my @row = $self->cursor->next;
1501 push(@objects, $self->_construct_objects(@row));
1502 @row = (exists $self->{stashed_row}
1503 ? @{delete $self->{stashed_row}}
1504 : $self->cursor->next);
1507 @objects = map { $self->_construct_objects($_) } $self->cursor->all;
1510 $self->set_cache(\@objects) if $self->{attrs}{cache};
1519 =item Arguments: none
1521 =item Return Value: $self
1525 Resets the resultset's cursor, so you can iterate through the elements again.
1526 Implicitly resets the storage cursor, so a subsequent L</next> will trigger
1533 delete $self->{_attrs} if exists $self->{_attrs};
1534 $self->{all_cache_position} = 0;
1535 $self->cursor->reset;
1543 =item Arguments: none
1545 =item Return Value: $object?
1549 Resets the resultset and returns an object for the first result (if the
1550 resultset returns anything).
1555 return $_[0]->reset->next;
1561 # Determines whether and what type of subquery is required for the $rs operation.
1562 # If grouping is necessary either supplies its own, or verifies the current one
1563 # After all is done delegates to the proper storage method.
1565 sub _rs_update_delete {
1566 my ($self, $op, $values) = @_;
1568 my $rsrc = $self->result_source;
1570 # if a condition exists we need to strip all table qualifiers
1571 # if this is not possible we'll force a subquery below
1572 my $cond = $rsrc->schema->storage->_strip_cond_qualifiers ($self->{cond});
1574 my $needs_group_by_subq = $self->_has_resolved_attr (qw/collapse group_by -join/);
1575 my $needs_subq = $needs_group_by_subq || (not defined $cond) || $self->_has_resolved_attr(qw/row offset/);
1577 if ($needs_group_by_subq or $needs_subq) {
1579 # make a new $rs selecting only the PKs (that's all we really need)
1580 my $attrs = $self->_resolved_attrs_copy;
1582 delete $attrs->{$_} for qw/collapse select as/;
1583 $attrs->{columns} = [ map { "$attrs->{alias}.$_" } ($self->result_source->_pri_cols) ];
1585 if ($needs_group_by_subq) {
1586 # make sure no group_by was supplied, or if there is one - make sure it matches
1587 # the columns compiled above perfectly. Anything else can not be sanely executed
1588 # on most databases so croak right then and there
1590 if (my $g = $attrs->{group_by}) {
1591 my @current_group_by = map
1592 { $_ =~ /\./ ? $_ : "$attrs->{alias}.$_" }
1597 join ("\x00", sort @current_group_by)
1599 join ("\x00", sort @{$attrs->{columns}} )
1601 $self->throw_exception (
1602 "You have just attempted a $op operation on a resultset which does group_by"
1603 . ' on columns other than the primary keys, while DBIC internally needs to retrieve'
1604 . ' the primary keys in a subselect. All sane RDBMS engines do not support this'
1605 . ' kind of queries. Please retry the operation with a modified group_by or'
1606 . ' without using one at all.'
1611 $attrs->{group_by} = $attrs->{columns};
1615 my $subrs = (ref $self)->new($rsrc, $attrs);
1617 return $self->result_source->storage->_subq_update_delete($subrs, $op, $values);
1620 return $rsrc->storage->$op(
1622 $op eq 'update' ? $values : (),
1632 =item Arguments: \%values
1634 =item Return Value: $storage_rv
1638 Sets the specified columns in the resultset to the supplied values in a
1639 single query. Return value will be true if the update succeeded or false
1640 if no records were updated; exact type of success value is storage-dependent.
1645 my ($self, $values) = @_;
1646 $self->throw_exception('Values for update must be a hash')
1647 unless ref $values eq 'HASH';
1649 return $self->_rs_update_delete ('update', $values);
1656 =item Arguments: \%values
1658 =item Return Value: 1
1662 Fetches all objects and updates them one at a time. Note that C<update_all>
1663 will run DBIC cascade triggers, while L</update> will not.
1668 my ($self, $values) = @_;
1669 $self->throw_exception('Values for update_all must be a hash')
1670 unless ref $values eq 'HASH';
1671 foreach my $obj ($self->all) {
1672 $obj->set_columns($values)->update;
1681 =item Arguments: none
1683 =item Return Value: $storage_rv
1687 Deletes the contents of the resultset from its result source. Note that this
1688 will not run DBIC cascade triggers. See L</delete_all> if you need triggers
1689 to run. See also L<DBIx::Class::Row/delete>.
1691 Return value will be the amount of rows deleted; exact type of return value
1692 is storage-dependent.
1698 $self->throw_exception('delete does not accept any arguments')
1701 return $self->_rs_update_delete ('delete');
1708 =item Arguments: none
1710 =item Return Value: 1
1714 Fetches all objects and deletes them one at a time. Note that C<delete_all>
1715 will run DBIC cascade triggers, while L</delete> will not.
1721 $self->throw_exception('delete_all does not accept any arguments')
1724 $_->delete for $self->all;
1732 =item Arguments: \@data;
1736 Accepts either an arrayref of hashrefs or alternatively an arrayref of arrayrefs.
1737 For the arrayref of hashrefs style each hashref should be a structure suitable
1738 forsubmitting to a $resultset->create(...) method.
1740 In void context, C<insert_bulk> in L<DBIx::Class::Storage::DBI> is used
1741 to insert the data, as this is a faster method.
1743 Otherwise, each set of data is inserted into the database using
1744 L<DBIx::Class::ResultSet/create>, and the resulting objects are
1745 accumulated into an array. The array itself, or an array reference
1746 is returned depending on scalar or list context.
1748 Example: Assuming an Artist Class that has many CDs Classes relating:
1750 my $Artist_rs = $schema->resultset("Artist");
1752 ## Void Context Example
1753 $Artist_rs->populate([
1754 { artistid => 4, name => 'Manufactured Crap', cds => [
1755 { title => 'My First CD', year => 2006 },
1756 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
1759 { artistid => 5, name => 'Angsty-Whiny Girl', cds => [
1760 { title => 'My parents sold me to a record company', year => 2005 },
1761 { title => 'Why Am I So Ugly?', year => 2006 },
1762 { title => 'I Got Surgery and am now Popular', year => 2007 }
1767 ## Array Context Example
1768 my ($ArtistOne, $ArtistTwo, $ArtistThree) = $Artist_rs->populate([
1769 { name => "Artist One"},
1770 { name => "Artist Two"},
1771 { name => "Artist Three", cds=> [
1772 { title => "First CD", year => 2007},
1773 { title => "Second CD", year => 2008},
1777 print $ArtistOne->name; ## response is 'Artist One'
1778 print $ArtistThree->cds->count ## reponse is '2'
1780 For the arrayref of arrayrefs style, the first element should be a list of the
1781 fieldsnames to which the remaining elements are rows being inserted. For
1784 $Arstist_rs->populate([
1785 [qw/artistid name/],
1786 [100, 'A Formally Unknown Singer'],
1787 [101, 'A singer that jumped the shark two albums ago'],
1788 [102, 'An actually cool singer'],
1791 Please note an important effect on your data when choosing between void and
1792 wantarray context. Since void context goes straight to C<insert_bulk> in
1793 L<DBIx::Class::Storage::DBI> this will skip any component that is overriding
1794 C<insert>. So if you are using something like L<DBIx-Class-UUIDColumns> to
1795 create primary keys for you, you will find that your PKs are empty. In this
1796 case you will have to use the wantarray context in order to create those
1804 # cruft placed in standalone method
1805 my $data = $self->_normalize_populate_args(@_);
1807 if(defined wantarray) {
1809 foreach my $item (@$data) {
1810 push(@created, $self->create($item));
1812 return wantarray ? @created : \@created;
1814 my $first = $data->[0];
1816 # if a column is a registered relationship, and is a non-blessed hash/array, consider
1817 # it relationship data
1818 my (@rels, @columns);
1819 for (keys %$first) {
1820 my $ref = ref $first->{$_};
1821 $self->result_source->has_relationship($_) && ($ref eq 'ARRAY' or $ref eq 'HASH')
1827 my @pks = $self->result_source->primary_columns;
1829 ## do the belongs_to relationships
1830 foreach my $index (0..$#$data) {
1832 # delegate to create() for any dataset without primary keys with specified relationships
1833 if (grep { !defined $data->[$index]->{$_} } @pks ) {
1835 if (grep { ref $data->[$index]{$r} eq $_ } qw/HASH ARRAY/) { # a related set must be a HASH or AoH
1836 my @ret = $self->populate($data);
1842 foreach my $rel (@rels) {
1843 next unless ref $data->[$index]->{$rel} eq "HASH";
1844 my $result = $self->related_resultset($rel)->create($data->[$index]->{$rel});
1845 my ($reverse) = keys %{$self->result_source->reverse_relationship_info($rel)};
1846 my $related = $result->result_source->_resolve_condition(
1847 $result->result_source->relationship_info($reverse)->{cond},
1852 delete $data->[$index]->{$rel};
1853 $data->[$index] = {%{$data->[$index]}, %$related};
1855 push @columns, keys %$related if $index == 0;
1859 ## inherit the data locked in the conditions of the resultset
1860 my ($rs_data) = $self->_merge_cond_with_data({});
1861 delete @{$rs_data}{@columns};
1862 my @inherit_cols = keys %$rs_data;
1863 my @inherit_data = values %$rs_data;
1865 ## do bulk insert on current row
1866 $self->result_source->storage->insert_bulk(
1867 $self->result_source,
1868 [@columns, @inherit_cols],
1869 [ map { [ @$_{@columns}, @inherit_data ] } @$data ],
1872 ## do the has_many relationships
1873 foreach my $item (@$data) {
1875 foreach my $rel (@rels) {
1876 next unless $item->{$rel} && ref $item->{$rel} eq "ARRAY";
1878 my $parent = $self->find({map { $_ => $item->{$_} } @pks})
1879 || $self->throw_exception('Cannot find the relating object.');
1881 my $child = $parent->$rel;
1883 my $related = $child->result_source->_resolve_condition(
1884 $parent->result_source->relationship_info($rel)->{cond},
1889 my @rows_to_add = ref $item->{$rel} eq 'ARRAY' ? @{$item->{$rel}} : ($item->{$rel});
1890 my @populate = map { {%$_, %$related} } @rows_to_add;
1892 $child->populate( \@populate );
1899 # populate() argumnets went over several incarnations
1900 # What we ultimately support is AoH
1901 sub _normalize_populate_args {
1902 my ($self, $arg) = @_;
1904 if (ref $arg eq 'ARRAY') {
1905 if (ref $arg->[0] eq 'HASH') {
1908 elsif (ref $arg->[0] eq 'ARRAY') {
1910 my @colnames = @{$arg->[0]};
1911 foreach my $values (@{$arg}[1 .. $#$arg]) {
1912 push @ret, { map { $colnames[$_] => $values->[$_] } (0 .. $#colnames) };
1918 $self->throw_exception('Populate expects an arrayref of hashrefs or arrayref of arrayrefs');
1925 =item Arguments: none
1927 =item Return Value: $pager
1931 Return Value a L<Data::Page> object for the current resultset. Only makes
1932 sense for queries with a C<page> attribute.
1934 To get the full count of entries for a paged resultset, call
1935 C<total_entries> on the L<Data::Page> object.
1942 return $self->{pager} if $self->{pager};
1944 my $attrs = $self->{attrs};
1945 $self->throw_exception("Can't create pager for non-paged rs")
1946 unless $self->{attrs}{page};
1947 $attrs->{rows} ||= 10;
1949 # throw away the paging flags and re-run the count (possibly
1950 # with a subselect) to get the real total count
1951 my $count_attrs = { %$attrs };
1952 delete $count_attrs->{$_} for qw/rows offset page pager/;
1953 my $total_count = (ref $self)->new($self->result_source, $count_attrs)->count;
1955 return $self->{pager} = Data::Page->new(
1958 $self->{attrs}{page}
1966 =item Arguments: $page_number
1968 =item Return Value: $rs
1972 Returns a resultset for the $page_number page of the resultset on which page
1973 is called, where each page contains a number of rows equal to the 'rows'
1974 attribute set on the resultset (10 by default).
1979 my ($self, $page) = @_;
1980 return (ref $self)->new($self->result_source, { %{$self->{attrs}}, page => $page });
1987 =item Arguments: \%vals
1989 =item Return Value: $rowobject
1993 Creates a new row object in the resultset's result class and returns
1994 it. The row is not inserted into the database at this point, call
1995 L<DBIx::Class::Row/insert> to do that. Calling L<DBIx::Class::Row/in_storage>
1996 will tell you whether the row object has been inserted or not.
1998 Passes the hashref of input on to L<DBIx::Class::Row/new>.
2003 my ($self, $values) = @_;
2004 $self->throw_exception( "new_result needs a hash" )
2005 unless (ref $values eq 'HASH');
2007 my ($merged_cond, $cols_from_relations) = $self->_merge_cond_with_data($values);
2011 @$cols_from_relations
2012 ? (-cols_from_relations => $cols_from_relations)
2014 -source_handle => $self->_source_handle,
2015 -result_source => $self->result_source, # DO NOT REMOVE THIS, REQUIRED
2018 return $self->result_class->new(\%new);
2021 # _merge_cond_with_data
2023 # Takes a simple hash of K/V data and returns its copy merged with the
2024 # condition already present on the resultset. Additionally returns an
2025 # arrayref of value/condition names, which were inferred from related
2026 # objects (this is needed for in-memory related objects)
2027 sub _merge_cond_with_data {
2028 my ($self, $data) = @_;
2030 my (%new_data, @cols_from_relations);
2032 my $alias = $self->{attrs}{alias};
2034 if (! defined $self->{cond}) {
2035 # just massage $data below
2037 elsif ($self->{cond} eq $DBIx::Class::ResultSource::UNRESOLVABLE_CONDITION) {
2038 %new_data = %{ $self->{attrs}{related_objects} || {} }; # nothing might have been inserted yet
2039 @cols_from_relations = keys %new_data;
2041 elsif (ref $self->{cond} ne 'HASH') {
2042 $self->throw_exception(
2043 "Can't abstract implicit construct, resultset condition not a hash"
2047 # precendence must be given to passed values over values inherited from
2048 # the cond, so the order here is important.
2049 my $collapsed_cond = $self->_collapse_cond($self->{cond});
2050 my %implied = %{$self->_remove_alias($collapsed_cond, $alias)};
2052 while ( my($col, $value) = each %implied ) {
2053 if (ref($value) eq 'HASH' && keys(%$value) && (keys %$value)[0] eq '=') {
2054 $new_data{$col} = $value->{'='};
2057 $new_data{$col} = $value if $self->_is_deterministic_value($value);
2063 %{ $self->_remove_alias($data, $alias) },
2066 return (\%new_data, \@cols_from_relations);
2069 # _is_deterministic_value
2071 # Make an effor to strip non-deterministic values from the condition,
2072 # to make sure new_result chokes less
2074 sub _is_deterministic_value {
2077 my $ref_type = ref $value;
2078 return 1 if $ref_type eq '' || $ref_type eq 'SCALAR';
2079 return 1 if Scalar::Util::blessed($value);
2083 # _has_resolved_attr
2085 # determines if the resultset defines at least one
2086 # of the attributes supplied
2088 # used to determine if a subquery is neccessary
2090 # supports some virtual attributes:
2092 # This will scan for any joins being present on the resultset.
2093 # It is not a mere key-search but a deep inspection of {from}
2096 sub _has_resolved_attr {
2097 my ($self, @attr_names) = @_;
2099 my $attrs = $self->_resolved_attrs;
2103 for my $n (@attr_names) {
2104 if (grep { $n eq $_ } (qw/-join/) ) {
2105 $extra_checks{$n}++;
2109 my $attr = $attrs->{$n};
2111 next if not defined $attr;
2113 if (ref $attr eq 'HASH') {
2114 return 1 if keys %$attr;
2116 elsif (ref $attr eq 'ARRAY') {
2124 # a resolved join is expressed as a multi-level from
2126 $extra_checks{-join}
2128 ref $attrs->{from} eq 'ARRAY'
2130 @{$attrs->{from}} > 1
2138 # Recursively collapse the condition.
2140 sub _collapse_cond {
2141 my ($self, $cond, $collapsed) = @_;
2145 if (ref $cond eq 'ARRAY') {
2146 foreach my $subcond (@$cond) {
2147 next unless ref $subcond; # -or
2148 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2151 elsif (ref $cond eq 'HASH') {
2152 if (keys %$cond and (keys %$cond)[0] eq '-and') {
2153 foreach my $subcond (@{$cond->{-and}}) {
2154 $collapsed = $self->_collapse_cond($subcond, $collapsed);
2158 foreach my $col (keys %$cond) {
2159 my $value = $cond->{$col};
2160 $collapsed->{$col} = $value;
2170 # Remove the specified alias from the specified query hash. A copy is made so
2171 # the original query is not modified.
2174 my ($self, $query, $alias) = @_;
2176 my %orig = %{ $query || {} };
2179 foreach my $key (keys %orig) {
2181 $unaliased{$key} = $orig{$key};
2184 $unaliased{$1} = $orig{$key}
2185 if $key =~ m/^(?:\Q$alias\E\.)?([^.]+)$/;
2195 =item Arguments: none
2197 =item Return Value: \[ $sql, @bind ]
2201 Returns the SQL query and bind vars associated with the invocant.
2203 This is generally used as the RHS for a subquery.
2210 my $attrs = $self->_resolved_attrs_copy;
2215 # my ($sql, \@bind, \%dbi_bind_attrs) = _select_args_to_query (...)
2216 # $sql also has no wrapping parenthesis in list ctx
2218 my $sqlbind = $self->result_source->storage
2219 ->_select_args_to_query ($attrs->{from}, $attrs->{select}, $attrs->{where}, $attrs);
2228 =item Arguments: \%vals, \%attrs?
2230 =item Return Value: $rowobject
2234 my $artist = $schema->resultset('Artist')->find_or_new(
2235 { artist => 'fred' }, { key => 'artists' });
2237 $cd->cd_to_producer->find_or_new({ producer => $producer },
2238 { key => 'primary });
2240 Find an existing record from this resultset, based on its primary
2241 key, or a unique constraint. If none exists, instantiate a new result
2242 object and return it. The object will not be saved into your storage
2243 until you call L<DBIx::Class::Row/insert> on it.
2245 You most likely want this method when looking for existing rows using
2246 a unique constraint that is not the primary key, or looking for
2249 If you want objects to be saved immediately, use L</find_or_create>
2252 B<Note>: Take care when using C<find_or_new> with a table having
2253 columns with default values that you intend to be automatically
2254 supplied by the database (e.g. an auto_increment primary key column).
2255 In normal usage, the value of such columns should NOT be included at
2256 all in the call to C<find_or_new>, even when set to C<undef>.
2262 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2263 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2264 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2267 return $self->new_result($hash);
2274 =item Arguments: \%vals
2276 =item Return Value: a L<DBIx::Class::Row> $object
2280 Attempt to create a single new row or a row with multiple related rows
2281 in the table represented by the resultset (and related tables). This
2282 will not check for duplicate rows before inserting, use
2283 L</find_or_create> to do that.
2285 To create one row for this resultset, pass a hashref of key/value
2286 pairs representing the columns of the table and the values you wish to
2287 store. If the appropriate relationships are set up, foreign key fields
2288 can also be passed an object representing the foreign row, and the
2289 value will be set to its primary key.
2291 To create related objects, pass a hashref of related-object column values
2292 B<keyed on the relationship name>. If the relationship is of type C<multi>
2293 (L<DBIx::Class::Relationship/has_many>) - pass an arrayref of hashrefs.
2294 The process will correctly identify columns holding foreign keys, and will
2295 transparently populate them from the keys of the corresponding relation.
2296 This can be applied recursively, and will work correctly for a structure
2297 with an arbitrary depth and width, as long as the relationships actually
2298 exists and the correct column data has been supplied.
2301 Instead of hashrefs of plain related data (key/value pairs), you may
2302 also pass new or inserted objects. New objects (not inserted yet, see
2303 L</new>), will be inserted into their appropriate tables.
2305 Effectively a shortcut for C<< ->new_result(\%vals)->insert >>.
2307 Example of creating a new row.
2309 $person_rs->create({
2310 name=>"Some Person",
2311 email=>"somebody@someplace.com"
2314 Example of creating a new row and also creating rows in a related C<has_many>
2315 or C<has_one> resultset. Note Arrayref.
2318 { artistid => 4, name => 'Manufactured Crap', cds => [
2319 { title => 'My First CD', year => 2006 },
2320 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
2325 Example of creating a new row and also creating a row in a related
2326 C<belongs_to>resultset. Note Hashref.
2329 title=>"Music for Silly Walks",
2332 name=>"Silly Musician",
2340 When subclassing ResultSet never attempt to override this method. Since
2341 it is a simple shortcut for C<< $self->new_result($attrs)->insert >>, a
2342 lot of the internals simply never call it, so your override will be
2343 bypassed more often than not. Override either L<new|DBIx::Class::Row/new>
2344 or L<insert|DBIx::Class::Row/insert> depending on how early in the
2345 L</create> process you need to intervene.
2352 my ($self, $attrs) = @_;
2353 $self->throw_exception( "create needs a hashref" )
2354 unless ref $attrs eq 'HASH';
2355 return $self->new_result($attrs)->insert;
2358 =head2 find_or_create
2362 =item Arguments: \%vals, \%attrs?
2364 =item Return Value: $rowobject
2368 $cd->cd_to_producer->find_or_create({ producer => $producer },
2369 { key => 'primary' });
2371 Tries to find a record based on its primary key or unique constraints; if none
2372 is found, creates one and returns that instead.
2374 my $cd = $schema->resultset('CD')->find_or_create({
2376 artist => 'Massive Attack',
2377 title => 'Mezzanine',
2381 Also takes an optional C<key> attribute, to search by a specific key or unique
2382 constraint. For example:
2384 my $cd = $schema->resultset('CD')->find_or_create(
2386 artist => 'Massive Attack',
2387 title => 'Mezzanine',
2389 { key => 'cd_artist_title' }
2392 B<Note>: Because find_or_create() reads from the database and then
2393 possibly inserts based on the result, this method is subject to a race
2394 condition. Another process could create a record in the table after
2395 the find has completed and before the create has started. To avoid
2396 this problem, use find_or_create() inside a transaction.
2398 B<Note>: Take care when using C<find_or_create> with a table having
2399 columns with default values that you intend to be automatically
2400 supplied by the database (e.g. an auto_increment primary key column).
2401 In normal usage, the value of such columns should NOT be included at
2402 all in the call to C<find_or_create>, even when set to C<undef>.
2404 See also L</find> and L</update_or_create>. For information on how to declare
2405 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2409 sub find_or_create {
2411 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2412 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2413 if (keys %$hash and my $row = $self->find($hash, $attrs) ) {
2416 return $self->create($hash);
2419 =head2 update_or_create
2423 =item Arguments: \%col_values, { key => $unique_constraint }?
2425 =item Return Value: $rowobject
2429 $resultset->update_or_create({ col => $val, ... });
2431 First, searches for an existing row matching one of the unique constraints
2432 (including the primary key) on the source of this resultset. If a row is
2433 found, updates it with the other given column values. Otherwise, creates a new
2436 Takes an optional C<key> attribute to search on a specific unique constraint.
2439 # In your application
2440 my $cd = $schema->resultset('CD')->update_or_create(
2442 artist => 'Massive Attack',
2443 title => 'Mezzanine',
2446 { key => 'cd_artist_title' }
2449 $cd->cd_to_producer->update_or_create({
2450 producer => $producer,
2457 If no C<key> is specified, it searches on all unique constraints defined on the
2458 source, including the primary key.
2460 If the C<key> is specified as C<primary>, it searches only on the primary key.
2462 See also L</find> and L</find_or_create>. For information on how to declare
2463 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2465 B<Note>: Take care when using C<update_or_create> with a table having
2466 columns with default values that you intend to be automatically
2467 supplied by the database (e.g. an auto_increment primary key column).
2468 In normal usage, the value of such columns should NOT be included at
2469 all in the call to C<update_or_create>, even when set to C<undef>.
2473 sub update_or_create {
2475 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2476 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2478 my $row = $self->find($cond, $attrs);
2480 $row->update($cond);
2484 return $self->create($cond);
2487 =head2 update_or_new
2491 =item Arguments: \%col_values, { key => $unique_constraint }?
2493 =item Return Value: $rowobject
2497 $resultset->update_or_new({ col => $val, ... });
2499 First, searches for an existing row matching one of the unique constraints
2500 (including the primary key) on the source of this resultset. If a row is
2501 found, updates it with the other given column values. Otherwise, instantiate
2502 a new result object and return it. The object will not be saved into your storage
2503 until you call L<DBIx::Class::Row/insert> on it.
2505 Takes an optional C<key> attribute to search on a specific unique constraint.
2508 # In your application
2509 my $cd = $schema->resultset('CD')->update_or_new(
2511 artist => 'Massive Attack',
2512 title => 'Mezzanine',
2515 { key => 'cd_artist_title' }
2518 if ($cd->in_storage) {
2519 # the cd was updated
2522 # the cd is not yet in the database, let's insert it
2526 B<Note>: Take care when using C<update_or_new> with a table having
2527 columns with default values that you intend to be automatically
2528 supplied by the database (e.g. an auto_increment primary key column).
2529 In normal usage, the value of such columns should NOT be included at
2530 all in the call to C<update_or_new>, even when set to C<undef>.
2532 See also L</find>, L</find_or_create> and L</find_or_new>.
2538 my $attrs = ( @_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {} );
2539 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2541 my $row = $self->find( $cond, $attrs );
2542 if ( defined $row ) {
2543 $row->update($cond);
2547 return $self->new_result($cond);
2554 =item Arguments: none
2556 =item Return Value: \@cache_objects?
2560 Gets the contents of the cache for the resultset, if the cache is set.
2562 The cache is populated either by using the L</prefetch> attribute to
2563 L</search> or by calling L</set_cache>.
2575 =item Arguments: \@cache_objects
2577 =item Return Value: \@cache_objects
2581 Sets the contents of the cache for the resultset. Expects an arrayref
2582 of objects of the same class as those produced by the resultset. Note that
2583 if the cache is set the resultset will return the cached objects rather
2584 than re-querying the database even if the cache attr is not set.
2586 The contents of the cache can also be populated by using the
2587 L</prefetch> attribute to L</search>.
2592 my ( $self, $data ) = @_;
2593 $self->throw_exception("set_cache requires an arrayref")
2594 if defined($data) && (ref $data ne 'ARRAY');
2595 $self->{all_cache} = $data;
2602 =item Arguments: none
2604 =item Return Value: []
2608 Clears the cache for the resultset.
2613 shift->set_cache(undef);
2620 =item Arguments: none
2622 =item Return Value: true, if the resultset has been paginated
2630 return !!$self->{attrs}{page};
2637 =item Arguments: none
2639 =item Return Value: true, if the resultset has been ordered with C<order_by>.
2647 return scalar $self->result_source->storage->_parse_order_by($self->{attrs}{order_by});
2650 =head2 related_resultset
2654 =item Arguments: $relationship_name
2656 =item Return Value: $resultset
2660 Returns a related resultset for the supplied relationship name.
2662 $artist_rs = $schema->resultset('CD')->related_resultset('Artist');
2666 sub related_resultset {
2667 my ($self, $rel) = @_;
2669 $self->{related_resultsets} ||= {};
2670 return $self->{related_resultsets}{$rel} ||= do {
2671 my $rsrc = $self->result_source;
2672 my $rel_info = $rsrc->relationship_info($rel);
2674 $self->throw_exception(
2675 "search_related: result source '" . $rsrc->source_name .
2676 "' has no such relationship $rel")
2679 my $attrs = $self->_chain_relationship($rel);
2681 my $join_count = $attrs->{seen_join}{$rel};
2683 my $alias = $self->result_source->storage
2684 ->relname_to_table_alias($rel, $join_count);
2686 # since this is search_related, and we already slid the select window inwards
2687 # (the select/as attrs were deleted in the beginning), we need to flip all
2688 # left joins to inner, so we get the expected results
2689 # read the comment on top of the actual function to see what this does
2690 $attrs->{from} = $rsrc->schema->storage->_straight_join_to_node ($attrs->{from}, $alias);
2693 #XXX - temp fix for result_class bug. There likely is a more elegant fix -groditi
2694 delete @{$attrs}{qw(result_class alias)};
2698 if (my $cache = $self->get_cache) {
2699 if ($cache->[0] && $cache->[0]->related_resultset($rel)->get_cache) {
2700 $new_cache = [ map { @{$_->related_resultset($rel)->get_cache} }
2705 my $rel_source = $rsrc->related_source($rel);
2709 # The reason we do this now instead of passing the alias to the
2710 # search_rs below is that if you wrap/overload resultset on the
2711 # source you need to know what alias it's -going- to have for things
2712 # to work sanely (e.g. RestrictWithObject wants to be able to add
2713 # extra query restrictions, and these may need to be $alias.)
2715 my $rel_attrs = $rel_source->resultset_attributes;
2716 local $rel_attrs->{alias} = $alias;
2718 $rel_source->resultset
2722 where => $attrs->{where},
2725 $new->set_cache($new_cache) if $new_cache;
2730 =head2 current_source_alias
2734 =item Arguments: none
2736 =item Return Value: $source_alias
2740 Returns the current table alias for the result source this resultset is built
2741 on, that will be used in the SQL query. Usually it is C<me>.
2743 Currently the source alias that refers to the result set returned by a
2744 L</search>/L</find> family method depends on how you got to the resultset: it's
2745 C<me> by default, but eg. L</search_related> aliases it to the related result
2746 source name (and keeps C<me> referring to the original result set). The long
2747 term goal is to make L<DBIx::Class> always alias the current resultset as C<me>
2748 (and make this method unnecessary).
2750 Thus it's currently necessary to use this method in predefined queries (see
2751 L<DBIx::Class::Manual::Cookbook/Predefined searches>) when referring to the
2752 source alias of the current result set:
2754 # in a result set class
2756 my ($self, $user) = @_;
2758 my $me = $self->current_source_alias;
2760 return $self->search(
2761 "$me.modified" => $user->id,
2767 sub current_source_alias {
2770 return ($self->{attrs} || {})->{alias} || 'me';
2773 =head2 as_subselect_rs
2777 =item Arguments: none
2779 =item Return Value: $resultset
2783 Act as a barrier to SQL symbols. The resultset provided will be made into a
2784 "virtual view" by including it as a subquery within the from clause. From this
2785 point on, any joined tables are inaccessible to ->search on the resultset (as if
2786 it were simply where-filtered without joins). For example:
2788 my $rs = $schema->resultset('Bar')->search({'x.name' => 'abc'},{ join => 'x' });
2790 # 'x' now pollutes the query namespace
2792 # So the following works as expected
2793 my $ok_rs = $rs->search({'x.other' => 1});
2795 # But this doesn't: instead of finding a 'Bar' related to two x rows (abc and
2796 # def) we look for one row with contradictory terms and join in another table
2797 # (aliased 'x_2') which we never use
2798 my $broken_rs = $rs->search({'x.name' => 'def'});
2800 my $rs2 = $rs->as_subselect_rs;
2802 # doesn't work - 'x' is no longer accessible in $rs2, having been sealed away
2803 my $not_joined_rs = $rs2->search({'x.other' => 1});
2805 # works as expected: finds a 'table' row related to two x rows (abc and def)
2806 my $correctly_joined_rs = $rs2->search({'x.name' => 'def'});
2808 Another example of when one might use this would be to select a subset of
2809 columns in a group by clause:
2811 my $rs = $schema->resultset('Bar')->search(undef, {
2812 group_by => [qw{ id foo_id baz_id }],
2813 })->as_subselect_rs->search(undef, {
2814 columns => [qw{ id foo_id }]
2817 In the above example normally columns would have to be equal to the group by,
2818 but because we isolated the group by into a subselect the above works.
2822 sub as_subselect_rs {
2825 return $self->result_source->resultset->search( undef, {
2826 alias => $self->current_source_alias,
2828 $self->current_source_alias => $self->as_query,
2829 -alias => $self->current_source_alias,
2830 -source_handle => $self->result_source->handle,
2835 # This code is called by search_related, and makes sure there
2836 # is clear separation between the joins before, during, and
2837 # after the relationship. This information is needed later
2838 # in order to properly resolve prefetch aliases (any alias
2839 # with a relation_chain_depth less than the depth of the
2840 # current prefetch is not considered)
2842 # The increments happen twice per join. An even number means a
2843 # relationship specified via a search_related, whereas an odd
2844 # number indicates a join/prefetch added via attributes
2846 # Also this code will wrap the current resultset (the one we
2847 # chain to) in a subselect IFF it contains limiting attributes
2848 sub _chain_relationship {
2849 my ($self, $rel) = @_;
2850 my $source = $self->result_source;
2851 my $attrs = { %{$self->{attrs}||{}} };
2853 # we need to take the prefetch the attrs into account before we
2854 # ->_resolve_join as otherwise they get lost - captainL
2855 my $join = $self->_merge_attr( $attrs->{join}, $attrs->{prefetch} );
2857 delete @{$attrs}{qw/join prefetch collapse distinct select as columns +select +as +columns/};
2859 my $seen = { %{ (delete $attrs->{seen_join}) || {} } };
2862 my @force_subq_attrs = qw/offset rows group_by having/;
2865 ($attrs->{from} && ref $attrs->{from} ne 'ARRAY')
2867 $self->_has_resolved_attr (@force_subq_attrs)
2869 # Nuke the prefetch (if any) before the new $rs attrs
2870 # are resolved (prefetch is useless - we are wrapping
2871 # a subquery anyway).
2872 my $rs_copy = $self->search;
2873 $rs_copy->{attrs}{join} = $self->_merge_attr (
2874 $rs_copy->{attrs}{join},
2875 delete $rs_copy->{attrs}{prefetch},
2879 -source_handle => $source->handle,
2880 -alias => $attrs->{alias},
2881 $attrs->{alias} => $rs_copy->as_query,
2883 delete @{$attrs}{@force_subq_attrs, 'where'};
2884 $seen->{-relation_chain_depth} = 0;
2886 elsif ($attrs->{from}) { #shallow copy suffices
2887 $from = [ @{$attrs->{from}} ];
2891 -source_handle => $source->handle,
2892 -alias => $attrs->{alias},
2893 $attrs->{alias} => $source->from,
2897 my $jpath = ($seen->{-relation_chain_depth})
2898 ? $from->[-1][0]{-join_path}
2901 my @requested_joins = $source->_resolve_join(
2908 push @$from, @requested_joins;
2910 $seen->{-relation_chain_depth}++;
2912 # if $self already had a join/prefetch specified on it, the requested
2913 # $rel might very well be already included. What we do in this case
2914 # is effectively a no-op (except that we bump up the chain_depth on
2915 # the join in question so we could tell it *is* the search_related)
2918 # we consider the last one thus reverse
2919 for my $j (reverse @requested_joins) {
2920 my ($last_j) = keys %{$j->[0]{-join_path}[-1]};
2921 if ($rel eq $last_j) {
2922 $j->[0]{-relation_chain_depth}++;
2928 unless ($already_joined) {
2929 push @$from, $source->_resolve_join(
2937 $seen->{-relation_chain_depth}++;
2939 return {%$attrs, from => $from, seen_join => $seen};
2942 # too many times we have to do $attrs = { %{$self->_resolved_attrs} }
2943 sub _resolved_attrs_copy {
2945 return { %{$self->_resolved_attrs (@_)} };
2948 sub _resolved_attrs {
2950 return $self->{_attrs} if $self->{_attrs};
2952 my $attrs = { %{ $self->{attrs} || {} } };
2953 my $source = $self->result_source;
2954 my $alias = $attrs->{alias};
2956 $attrs->{columns} ||= delete $attrs->{cols} if exists $attrs->{cols};
2959 # build columns (as long as select isn't set) into a set of as/select hashes
2960 unless ( $attrs->{select} ) {
2963 if ( ref $attrs->{columns} eq 'ARRAY' ) {
2964 @cols = @{ delete $attrs->{columns}}
2965 } elsif ( defined $attrs->{columns} ) {
2966 @cols = delete $attrs->{columns}
2968 @cols = $source->columns
2972 if ( ref $_ eq 'HASH' ) {
2975 my $key = /^\Q${alias}.\E(.+)$/
2981 push @colbits, { $key => $value };
2986 # add the additional columns on
2987 foreach (qw{include_columns +columns}) {
2988 if ( $attrs->{$_} ) {
2989 my @list = ( ref($attrs->{$_}) eq 'ARRAY' )
2990 ? @{ delete $attrs->{$_} }
2991 : delete $attrs->{$_};
2993 if ( ref($_) eq 'HASH' ) {
2996 my $key = ( split /\./, $_ )[-1];
2997 my $value = ( /\./ ? $_ : "$alias.$_" );
2998 push @colbits, { $key => $value };
3004 # start with initial select items
3005 if ( $attrs->{select} ) {
3007 ( ref $attrs->{select} eq 'ARRAY' )
3008 ? [ @{ $attrs->{select} } ]
3009 : [ $attrs->{select} ];
3011 if ( $attrs->{as} ) {
3014 ref $attrs->{as} eq 'ARRAY'
3015 ? [ @{ $attrs->{as} } ]
3019 $attrs->{as} = [ map {
3020 m/^\Q${alias}.\E(.+)$/
3023 } @{ $attrs->{select} }
3028 # otherwise we intialise select & as to empty
3029 $attrs->{select} = [];
3033 # now add colbits to select/as
3034 push @{ $attrs->{select} }, map values %{$_}, @colbits;
3035 push @{ $attrs->{as} }, map keys %{$_}, @colbits;
3037 if ( my $adds = delete $attrs->{'+select'} ) {
3038 $adds = [$adds] unless ref $adds eq 'ARRAY';
3039 push @{ $attrs->{select} },
3040 map { /\./ || ref $_ ? $_ : "$alias.$_" } @$adds;
3042 if ( my $adds = delete $attrs->{'+as'} ) {
3043 $adds = [$adds] unless ref $adds eq 'ARRAY';
3044 push @{ $attrs->{as} }, @$adds;
3047 $attrs->{from} ||= [{
3048 -source_handle => $source->handle,
3049 -alias => $self->{attrs}{alias},
3050 $self->{attrs}{alias} => $source->from,
3053 if ( $attrs->{join} || $attrs->{prefetch} ) {
3055 $self->throw_exception ('join/prefetch can not be used with a custom {from}')
3056 if ref $attrs->{from} ne 'ARRAY';
3058 my $join = delete $attrs->{join} || {};
3060 if ( defined $attrs->{prefetch} ) {
3061 $join = $self->_merge_attr( $join, $attrs->{prefetch} );
3064 $attrs->{from} = # have to copy here to avoid corrupting the original
3066 @{ $attrs->{from} },
3067 $source->_resolve_join(
3070 { %{ $attrs->{seen_join} || {} } },
3071 ( $attrs->{seen_join} && keys %{$attrs->{seen_join}})
3072 ? $attrs->{from}[-1][0]{-join_path}
3079 if ( defined $attrs->{order_by} ) {
3080 $attrs->{order_by} = (
3081 ref( $attrs->{order_by} ) eq 'ARRAY'
3082 ? [ @{ $attrs->{order_by} } ]
3083 : [ $attrs->{order_by} || () ]
3087 if ($attrs->{group_by} and ref $attrs->{group_by} ne 'ARRAY') {
3088 $attrs->{group_by} = [ $attrs->{group_by} ];
3091 # generate the distinct induced group_by early, as prefetch will be carried via a
3092 # subquery (since a group_by is present)
3093 if (delete $attrs->{distinct}) {
3094 if ($attrs->{group_by}) {
3095 carp ("Useless use of distinct on a grouped resultset ('distinct' is ignored when a 'group_by' is present)");
3098 $attrs->{group_by} = [ grep { !ref($_) || (ref($_) ne 'HASH') } @{$attrs->{select}} ];
3100 # add any order_by parts that are not already present in the group_by
3101 # we need to be careful not to add any named functions/aggregates
3102 # i.e. select => [ ... { count => 'foo', -as 'foocount' } ... ]
3103 my %already_grouped = map { $_ => 1 } (@{$attrs->{group_by}});
3105 my $storage = $self->result_source->schema->storage;
3107 my $rs_column_list = $storage->_resolve_column_info ($attrs->{from});
3109 for my $chunk ($storage->_parse_order_by($attrs->{order_by})) {
3110 if ($rs_column_list->{$chunk} && not $already_grouped{$chunk}++) {
3111 push @{$attrs->{group_by}}, $chunk;
3117 # generate selections based on the prefetch helper
3118 if ( my $prefetch = delete $attrs->{prefetch} ) {
3119 $attrs->{collapse} = 1;
3121 # this is a separate structure (we don't look in {from} directly)
3122 # as the resolver needs to shift things off the lists to work
3123 # properly (identical-prefetches on different branches)
3125 if (ref $attrs->{from} eq 'ARRAY') {
3127 my $start_depth = $attrs->{seen_join}{-relation_chain_depth} || 0;
3129 for my $j ( @{$attrs->{from}}[1 .. $#{$attrs->{from}} ] ) {
3130 next unless $j->[0]{-alias};
3131 next unless $j->[0]{-join_path};
3132 next if ($j->[0]{-relation_chain_depth} || 0) < $start_depth;
3134 my @jpath = map { keys %$_ } @{$j->[0]{-join_path}};
3137 $p = $p->{$_} ||= {} for @jpath[ ($start_depth/2) .. $#jpath]; #only even depths are actual jpath boundaries
3138 push @{$p->{-join_aliases} }, $j->[0]{-alias};
3142 my @prefetch = $source->_resolve_prefetch( $prefetch, $alias, $join_map );
3144 # we need to somehow mark which columns came from prefetch
3145 $attrs->{_prefetch_select} = [ map { $_->[0] } @prefetch ];
3147 push @{ $attrs->{select} }, @{$attrs->{_prefetch_select}};
3148 push @{ $attrs->{as} }, (map { $_->[1] } @prefetch);
3151 # run through the resulting joinstructure (starting from our current slot)
3152 # and unset collapse if proven unnesessary
3153 if ($attrs->{collapse} && ref $attrs->{from} eq 'ARRAY') {
3155 if (@{$attrs->{from}} > 1) {
3157 # find where our table-spec starts and consider only things after us
3158 my @fromlist = @{$attrs->{from}};
3160 my $t = shift @fromlist;
3161 $t = $t->[0] if ref $t eq 'ARRAY'; #me vs join from-spec mismatch
3162 last if ($t->{-alias} && $t->{-alias} eq $alias);
3166 $attrs->{collapse} = ! $_->[0]{-is_single}
3171 # no joins - no collapse
3172 $attrs->{collapse} = 0;
3176 # the row parser generates differently depending on whether collapsing is requested
3177 # the need to look at {select} is temporary
3178 $attrs->{_row_parser} = $source->_mk_row_parser (
3179 @{$attrs}{qw/as collapse select/}
3182 # if both page and offset are specified, produce a combined offset
3183 # even though it doesn't make much sense, this is what pre 081xx has
3185 if (my $page = delete $attrs->{page}) {
3187 ($attrs->{rows} * ($page - 1))
3189 ($attrs->{offset} || 0)
3193 return $self->{_attrs} = $attrs;
3197 my ($self, $attr) = @_;
3199 if (ref $attr eq 'HASH') {
3200 return $self->_rollout_hash($attr);
3201 } elsif (ref $attr eq 'ARRAY') {
3202 return $self->_rollout_array($attr);
3208 sub _rollout_array {
3209 my ($self, $attr) = @_;
3212 foreach my $element (@{$attr}) {
3213 if (ref $element eq 'HASH') {
3214 push( @rolled_array, @{ $self->_rollout_hash( $element ) } );
3215 } elsif (ref $element eq 'ARRAY') {
3216 # XXX - should probably recurse here
3217 push( @rolled_array, @{$self->_rollout_array($element)} );
3219 push( @rolled_array, $element );
3222 return \@rolled_array;
3226 my ($self, $attr) = @_;
3229 foreach my $key (keys %{$attr}) {
3230 push( @rolled_array, { $key => $attr->{$key} } );
3232 return \@rolled_array;
3235 sub _calculate_score {
3236 my ($self, $a, $b) = @_;
3238 if (defined $a xor defined $b) {
3241 elsif (not defined $a) {
3245 if (ref $b eq 'HASH') {
3246 my ($b_key) = keys %{$b};
3247 if (ref $a eq 'HASH') {
3248 my ($a_key) = keys %{$a};
3249 if ($a_key eq $b_key) {
3250 return (1 + $self->_calculate_score( $a->{$a_key}, $b->{$b_key} ));
3255 return ($a eq $b_key) ? 1 : 0;
3258 if (ref $a eq 'HASH') {
3259 my ($a_key) = keys %{$a};
3260 return ($b eq $a_key) ? 1 : 0;
3262 return ($b eq $a) ? 1 : 0;
3268 my ($self, $orig, $import) = @_;
3270 return $import unless defined($orig);
3271 return $orig unless defined($import);
3273 $orig = $self->_rollout_attr($orig);
3274 $import = $self->_rollout_attr($import);
3277 foreach my $import_element ( @{$import} ) {
3278 # find best candidate from $orig to merge $b_element into
3279 my $best_candidate = { position => undef, score => 0 }; my $position = 0;
3280 foreach my $orig_element ( @{$orig} ) {
3281 my $score = $self->_calculate_score( $orig_element, $import_element );
3282 if ($score > $best_candidate->{score}) {
3283 $best_candidate->{position} = $position;
3284 $best_candidate->{score} = $score;
3288 my ($import_key) = ( ref $import_element eq 'HASH' ) ? keys %{$import_element} : ($import_element);
3290 if ($best_candidate->{score} == 0 || exists $seen_keys->{$import_key}) {
3291 push( @{$orig}, $import_element );
3293 my $orig_best = $orig->[$best_candidate->{position}];
3294 # merge orig_best and b_element together and replace original with merged
3295 if (ref $orig_best ne 'HASH') {
3296 $orig->[$best_candidate->{position}] = $import_element;
3297 } elsif (ref $import_element eq 'HASH') {
3298 my ($key) = keys %{$orig_best};
3299 $orig->[$best_candidate->{position}] = { $key => $self->_merge_attr($orig_best->{$key}, $import_element->{$key}) };
3302 $seen_keys->{$import_key} = 1; # don't merge the same key twice
3312 $self->_source_handle($_[0]->handle);
3314 $self->_source_handle->resolve;
3318 =head2 throw_exception
3320 See L<DBIx::Class::Schema/throw_exception> for details.
3324 sub throw_exception {
3327 if (ref $self && $self->_source_handle->schema) {
3328 $self->_source_handle->schema->throw_exception(@_)
3331 DBIx::Class::Exception->throw(@_);
3335 # XXX: FIXME: Attributes docs need clearing up
3339 Attributes are used to refine a ResultSet in various ways when
3340 searching for data. They can be passed to any method which takes an
3341 C<\%attrs> argument. See L</search>, L</search_rs>, L</find>,
3344 These are in no particular order:
3350 =item Value: ( $order_by | \@order_by | \%order_by )
3354 Which column(s) to order the results by.
3356 [The full list of suitable values is documented in
3357 L<SQL::Abstract/"ORDER BY CLAUSES">; the following is a summary of
3360 If a single column name, or an arrayref of names is supplied, the
3361 argument is passed through directly to SQL. The hashref syntax allows
3362 for connection-agnostic specification of ordering direction:
3364 For descending order:
3366 order_by => { -desc => [qw/col1 col2 col3/] }
3368 For explicit ascending order:
3370 order_by => { -asc => 'col' }
3372 The old scalarref syntax (i.e. order_by => \'year DESC') is still
3373 supported, although you are strongly encouraged to use the hashref
3374 syntax as outlined above.
3380 =item Value: \@columns
3384 Shortcut to request a particular set of columns to be retrieved. Each
3385 column spec may be a string (a table column name), or a hash (in which
3386 case the key is the C<as> value, and the value is used as the C<select>
3387 expression). Adds C<me.> onto the start of any column without a C<.> in
3388 it and sets C<select> from that, then auto-populates C<as> from
3389 C<select> as normal. (You may also use the C<cols> attribute, as in
3390 earlier versions of DBIC.)
3396 =item Value: \@columns
3400 Indicates additional columns to be selected from storage. Works the same
3401 as L</columns> but adds columns to the selection. (You may also use the
3402 C<include_columns> attribute, as in earlier versions of DBIC). For
3405 $schema->resultset('CD')->search(undef, {
3406 '+columns' => ['artist.name'],
3410 would return all CDs and include a 'name' column to the information
3411 passed to object inflation. Note that the 'artist' is the name of the
3412 column (or relationship) accessor, and 'name' is the name of the column
3413 accessor in the related table.
3415 =head2 include_columns
3419 =item Value: \@columns
3423 Deprecated. Acts as a synonym for L</+columns> for backward compatibility.
3429 =item Value: \@select_columns
3433 Indicates which columns should be selected from the storage. You can use
3434 column names, or in the case of RDBMS back ends, function or stored procedure
3437 $rs = $schema->resultset('Employee')->search(undef, {
3440 { count => 'employeeid' },
3445 When you use function/stored procedure names and do not supply an C<as>
3446 attribute, the column names returned are storage-dependent. E.g. MySQL would
3447 return a column named C<count(employeeid)> in the above example.
3449 B<NOTE:> You will almost always need a corresponding 'as' entry when you use
3456 Indicates additional columns to be selected from storage. Works the same as
3457 L</select> but adds columns to the selection.
3465 Indicates additional column names for those added via L</+select>. See L</as>.
3473 =item Value: \@inflation_names
3477 Indicates column names for object inflation. That is, C<as>
3478 indicates the name that the column can be accessed as via the
3479 C<get_column> method (or via the object accessor, B<if one already
3480 exists>). It has nothing to do with the SQL code C<SELECT foo AS bar>.
3482 The C<as> attribute is used in conjunction with C<select>,
3483 usually when C<select> contains one or more function or stored
3486 $rs = $schema->resultset('Employee')->search(undef, {
3489 { count => 'employeeid' }
3491 as => ['name', 'employee_count'],
3494 my $employee = $rs->first(); # get the first Employee
3496 If the object against which the search is performed already has an accessor
3497 matching a column name specified in C<as>, the value can be retrieved using
3498 the accessor as normal:
3500 my $name = $employee->name();
3502 If on the other hand an accessor does not exist in the object, you need to
3503 use C<get_column> instead:
3505 my $employee_count = $employee->get_column('employee_count');
3507 You can create your own accessors if required - see
3508 L<DBIx::Class::Manual::Cookbook> for details.
3510 Please note: This will NOT insert an C<AS employee_count> into the SQL
3511 statement produced, it is used for internal access only. Thus
3512 attempting to use the accessor in an C<order_by> clause or similar
3513 will fail miserably.
3515 To get around this limitation, you can supply literal SQL to your
3516 C<select> attribute that contains the C<AS alias> text, e.g.
3518 select => [\'myfield AS alias']
3524 =item Value: ($rel_name | \@rel_names | \%rel_names)
3528 Contains a list of relationships that should be joined for this query. For
3531 # Get CDs by Nine Inch Nails
3532 my $rs = $schema->resultset('CD')->search(
3533 { 'artist.name' => 'Nine Inch Nails' },
3534 { join => 'artist' }
3537 Can also contain a hash reference to refer to the other relation's relations.
3540 package MyApp::Schema::Track;
3541 use base qw/DBIx::Class/;
3542 __PACKAGE__->table('track');
3543 __PACKAGE__->add_columns(qw/trackid cd position title/);
3544 __PACKAGE__->set_primary_key('trackid');
3545 __PACKAGE__->belongs_to(cd => 'MyApp::Schema::CD');
3548 # In your application
3549 my $rs = $schema->resultset('Artist')->search(
3550 { 'track.title' => 'Teardrop' },
3552 join => { cd => 'track' },
3553 order_by => 'artist.name',
3557 You need to use the relationship (not the table) name in conditions,
3558 because they are aliased as such. The current table is aliased as "me", so
3559 you need to use me.column_name in order to avoid ambiguity. For example:
3561 # Get CDs from 1984 with a 'Foo' track
3562 my $rs = $schema->resultset('CD')->search(
3565 'tracks.name' => 'Foo'
3567 { join => 'tracks' }
3570 If the same join is supplied twice, it will be aliased to <rel>_2 (and
3571 similarly for a third time). For e.g.
3573 my $rs = $schema->resultset('Artist')->search({
3574 'cds.title' => 'Down to Earth',
3575 'cds_2.title' => 'Popular',
3577 join => [ qw/cds cds/ ],
3580 will return a set of all artists that have both a cd with title 'Down
3581 to Earth' and a cd with title 'Popular'.
3583 If you want to fetch related objects from other tables as well, see C<prefetch>
3586 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
3592 =item Value: ($rel_name | \@rel_names | \%rel_names)
3596 Contains one or more relationships that should be fetched along with
3597 the main query (when they are accessed afterwards the data will
3598 already be available, without extra queries to the database). This is
3599 useful for when you know you will need the related objects, because it
3600 saves at least one query:
3602 my $rs = $schema->resultset('Tag')->search(
3611 The initial search results in SQL like the following:
3613 SELECT tag.*, cd.*, artist.* FROM tag
3614 JOIN cd ON tag.cd = cd.cdid
3615 JOIN artist ON cd.artist = artist.artistid
3617 L<DBIx::Class> has no need to go back to the database when we access the
3618 C<cd> or C<artist> relationships, which saves us two SQL statements in this
3621 Simple prefetches will be joined automatically, so there is no need
3622 for a C<join> attribute in the above search.
3624 C<prefetch> can be used with the following relationship types: C<belongs_to>,
3625 C<has_one> (or if you're using C<add_relationship>, any relationship declared
3626 with an accessor type of 'single' or 'filter'). A more complex example that
3627 prefetches an artists cds, the tracks on those cds, and the tags associated
3628 with that artist is given below (assuming many-to-many from artists to tags):
3630 my $rs = $schema->resultset('Artist')->search(
3634 { cds => 'tracks' },
3635 { artist_tags => 'tags' }
3641 B<NOTE:> If you specify a C<prefetch> attribute, the C<join> and C<select>
3642 attributes will be ignored.
3644 B<CAVEATs>: Prefetch does a lot of deep magic. As such, it may not behave
3645 exactly as you might expect.
3651 Prefetch uses the L</cache> to populate the prefetched relationships. This
3652 may or may not be what you want.
3656 If you specify a condition on a prefetched relationship, ONLY those
3657 rows that match the prefetched condition will be fetched into that relationship.
3658 This means that adding prefetch to a search() B<may alter> what is returned by
3659 traversing a relationship. So, if you have C<< Artist->has_many(CDs) >> and you do
3661 my $artist_rs = $schema->resultset('Artist')->search({
3667 my $count = $artist_rs->first->cds->count;
3669 my $artist_rs_prefetch = $artist_rs->search( {}, { prefetch => 'cds' } );
3671 my $prefetch_count = $artist_rs_prefetch->first->cds->count;
3673 cmp_ok( $count, '==', $prefetch_count, "Counts should be the same" );
3675 that cmp_ok() may or may not pass depending on the datasets involved. This
3676 behavior may or may not survive the 0.09 transition.
3688 Makes the resultset paged and specifies the page to retrieve. Effectively
3689 identical to creating a non-pages resultset and then calling ->page($page)
3692 If L<rows> attribute is not specified it defaults to 10 rows per page.
3694 When you have a paged resultset, L</count> will only return the number
3695 of rows in the page. To get the total, use the L</pager> and call
3696 C<total_entries> on it.
3706 Specifies the maximum number of rows for direct retrieval or the number of
3707 rows per page if the page attribute or method is used.
3713 =item Value: $offset
3717 Specifies the (zero-based) row number for the first row to be returned, or the
3718 of the first row of the first page if paging is used.
3724 =item Value: \@columns
3728 A arrayref of columns to group by. Can include columns of joined tables.
3730 group_by => [qw/ column1 column2 ... /]
3736 =item Value: $condition
3740 HAVING is a select statement attribute that is applied between GROUP BY and
3741 ORDER BY. It is applied to the after the grouping calculations have been
3744 having => { 'count(employee)' => { '>=', 100 } }
3750 =item Value: (0 | 1)
3754 Set to 1 to group by all columns. If the resultset already has a group_by
3755 attribute, this setting is ignored and an appropriate warning is issued.
3761 Adds to the WHERE clause.
3763 # only return rows WHERE deleted IS NULL for all searches
3764 __PACKAGE__->resultset_attributes({ where => { deleted => undef } }); )
3766 Can be overridden by passing C<< { where => undef } >> as an attribute
3773 Set to 1 to cache search results. This prevents extra SQL queries if you
3774 revisit rows in your ResultSet:
3776 my $resultset = $schema->resultset('Artist')->search( undef, { cache => 1 } );
3778 while( my $artist = $resultset->next ) {
3782 $rs->first; # without cache, this would issue a query
3784 By default, searches are not cached.
3786 For more examples of using these attributes, see
3787 L<DBIx::Class::Manual::Cookbook>.
3793 =item Value: ( 'update' | 'shared' )
3797 Set to 'update' for a SELECT ... FOR UPDATE or 'shared' for a SELECT