1 package DBIx::Class::ResultSet;
9 use Carp::Clan qw/^DBIx::Class/;
12 use DBIx::Class::ResultSetColumn;
13 use DBIx::Class::ResultSourceHandle;
16 use base qw/DBIx::Class/;
18 __PACKAGE__->mk_group_accessors('simple' => qw/_result_class _source_handle/);
22 DBIx::Class::ResultSet - Represents a query used for fetching a set of results.
26 my $users_rs = $schema->resultset('User');
27 my $registered_users_rs = $schema->resultset('User')->search({ registered => 1 });
28 my @cds_in_2005 = $schema->resultset('CD')->search({ year => 2005 })->all();
32 A ResultSet is an object which stores a set of conditions representing
33 a query. It is the backbone of DBIx::Class (i.e. the really
34 important/useful bit).
36 No SQL is executed on the database when a ResultSet is created, it
37 just stores all the conditions needed to create the query.
39 A basic ResultSet representing the data of an entire table is returned
40 by calling C<resultset> on a L<DBIx::Class::Schema> and passing in a
41 L<Source|DBIx::Class::Manual::Glossary/Source> name.
43 my $users_rs = $schema->resultset('User');
45 A new ResultSet is returned from calling L</search> on an existing
46 ResultSet. The new one will contain all the conditions of the
47 original, plus any new conditions added in the C<search> call.
49 A ResultSet is also an iterator. L</next> is used to return all the
50 L<DBIx::Class::Row>s the ResultSet represents.
52 The query that the ResultSet represents is B<only> executed against
53 the database when these methods are called:
73 =head2 Chaining resultsets
75 Let's say you've got a query that needs to be run to return some data
76 to the user. But, you have an authorization system in place that
77 prevents certain users from seeing certain information. So, you want
78 to construct the basic query in one method, but add constraints to it in
83 my $request = $self->get_request; # Get a request object somehow.
84 my $schema = $self->get_schema; # Get the DBIC schema object somehow.
86 my $cd_rs = $schema->resultset('CD')->search({
87 title => $request->param('title'),
88 year => $request->param('year'),
91 $self->apply_security_policy( $cd_rs );
96 sub apply_security_policy {
105 =head3 Resolving conditions and attributes
107 When a resultset is chained from another resultset, conditions and
108 attributes with the same keys need resolving.
110 L</join>, L</prefetch>, L</+select>, L</+as> attributes are merged
111 into the existing ones from the original resultset.
113 The L</where>, L</having> attribute, and any search conditions are
114 merged with an SQL C<AND> to the existing condition from the original
117 All other attributes are overridden by any new ones supplied in the
120 =head2 Multiple queries
122 Since a resultset just defines a query, you can do all sorts of
123 things with it with the same object.
125 # Don't hit the DB yet.
126 my $cd_rs = $schema->resultset('CD')->search({
127 title => 'something',
131 # Each of these hits the DB individually.
132 my $count = $cd_rs->count;
133 my $most_recent = $cd_rs->get_column('date_released')->max();
134 my @records = $cd_rs->all;
136 And it's not just limited to SELECT statements.
142 $cd_rs->create({ artist => 'Fred' });
144 Which is the same as:
146 $schema->resultset('CD')->create({
147 title => 'something',
152 See: L</search>, L</count>, L</get_column>, L</all>, L</create>.
156 If a resultset is used in a numeric context it returns the L</count>.
157 However, if it is used in a booleand context it is always true. So if
158 you want to check if a resultset has any results use C<if $rs != 0>.
159 C<if $rs> will always be true.
167 =item Arguments: $source, \%$attrs
169 =item Return Value: $rs
173 The resultset constructor. Takes a source object (usually a
174 L<DBIx::Class::ResultSourceProxy::Table>) and an attribute hash (see
175 L</ATTRIBUTES> below). Does not perform any queries -- these are
176 executed as needed by the other methods.
178 Generally you won't need to construct a resultset manually. You'll
179 automatically get one from e.g. a L</search> called in scalar context:
181 my $rs = $schema->resultset('CD')->search({ title => '100th Window' });
183 IMPORTANT: If called on an object, proxies to new_result instead so
185 my $cd = $schema->resultset('CD')->new({ title => 'Spoon' });
187 will return a CD object, not a ResultSet.
193 return $class->new_result(@_) if ref $class;
195 my ($source, $attrs) = @_;
196 $source = $source->handle
197 unless $source->isa('DBIx::Class::ResultSourceHandle');
198 $attrs = { %{$attrs||{}} };
200 if ($attrs->{page}) {
201 $attrs->{rows} ||= 10;
204 $attrs->{alias} ||= 'me';
206 # Creation of {} and bless separated to mitigate RH perl bug
207 # see https://bugzilla.redhat.com/show_bug.cgi?id=196836
209 _source_handle => $source,
210 cond => $attrs->{where},
219 $attrs->{result_class} || $source->resolve->result_class
229 =item Arguments: $cond, \%attrs?
231 =item Return Value: $resultset (scalar context), @row_objs (list context)
235 my @cds = $cd_rs->search({ year => 2001 }); # "... WHERE year = 2001"
236 my $new_rs = $cd_rs->search({ year => 2005 });
238 my $new_rs = $cd_rs->search([ { year => 2005 }, { year => 2004 } ]);
239 # year = 2005 OR year = 2004
241 If you need to pass in additional attributes but no additional condition,
242 call it as C<search(undef, \%attrs)>.
244 # "SELECT name, artistid FROM $artist_table"
245 my @all_artists = $schema->resultset('Artist')->search(undef, {
246 columns => [qw/name artistid/],
249 For a list of attributes that can be passed to C<search>, see
250 L</ATTRIBUTES>. For more examples of using this function, see
251 L<Searching|DBIx::Class::Manual::Cookbook/Searching>. For a complete
252 documentation for the first argument, see L<SQL::Abstract>.
254 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
260 my $rs = $self->search_rs( @_ );
261 return (wantarray ? $rs->all : $rs);
268 =item Arguments: $cond, \%attrs?
270 =item Return Value: $resultset
274 This method does the same exact thing as search() except it will
275 always return a resultset, even in list context.
282 # Special-case handling for (undef, undef).
283 if ( @_ == 2 && !defined $_[1] && !defined $_[0] ) {
288 $attrs = pop(@_) if @_ > 1 and ref $_[$#_] eq 'HASH';
289 my $our_attrs = { %{$self->{attrs}} };
290 my $having = delete $our_attrs->{having};
291 my $where = delete $our_attrs->{where};
295 my %safe = (alias => 1, cache => 1);
298 (@_ && defined($_[0])) # @_ == () or (undef)
300 (keys %$attrs # empty attrs or only 'safe' attrs
301 && List::Util::first { !$safe{$_} } keys %$attrs)
303 # no search, effectively just a clone
304 $rows = $self->get_cache;
307 my $new_attrs = { %{$our_attrs}, %{$attrs} };
309 # merge new attrs into inherited
310 foreach my $key (qw/join prefetch +select +as bind/) {
311 next unless exists $attrs->{$key};
312 $new_attrs->{$key} = $self->_merge_attr($our_attrs->{$key}, $attrs->{$key});
317 (@_ == 1 || ref $_[0] eq "HASH")
319 (ref $_[0] eq 'HASH')
321 (keys %{ $_[0] } > 0)
329 ? $self->throw_exception("Odd number of arguments to search")
336 if (defined $where) {
337 $new_attrs->{where} = (
338 defined $new_attrs->{where}
341 ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_
342 } $where, $new_attrs->{where}
349 $new_attrs->{where} = (
350 defined $new_attrs->{where}
353 ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_
354 } $cond, $new_attrs->{where}
360 if (defined $having) {
361 $new_attrs->{having} = (
362 defined $new_attrs->{having}
365 ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_
366 } $having, $new_attrs->{having}
372 my $rs = (ref $self)->new($self->result_source, $new_attrs);
374 $rs->set_cache($rows);
379 =head2 search_literal
383 =item Arguments: $sql_fragment, @bind_values
385 =item Return Value: $resultset (scalar context), @row_objs (list context)
389 my @cds = $cd_rs->search_literal('year = ? AND title = ?', qw/2001 Reload/);
390 my $newrs = $artist_rs->search_literal('name = ?', 'Metallica');
392 Pass a literal chunk of SQL to be added to the conditional part of the
395 CAVEAT: C<search_literal> is provided for Class::DBI compatibility and should
396 only be used in that context. C<search_literal> is a convenience method.
397 It is equivalent to calling $schema->search(\[]), but if you want to ensure
398 columns are bound correctly, use C<search>.
400 Example of how to use C<search> instead of C<search_literal>
402 my @cds = $cd_rs->search_literal('cdid = ? AND (artist = ? OR artist = ?)', (2, 1, 2));
403 my @cds = $cd_rs->search(\[ 'cdid = ? AND (artist = ? OR artist = ?)', [ 'cdid', 2 ], [ 'artist', 1 ], [ 'artist', 2 ] ]);
406 See L<DBIx::Class::Manual::Cookbook/Searching> and
407 L<DBIx::Class::Manual::FAQ/Searching> for searching techniques that do not
408 require C<search_literal>.
413 my ($self, $sql, @bind) = @_;
415 if ( @bind && ref($bind[-1]) eq 'HASH' ) {
418 return $self->search(\[ $sql, map [ __DUMMY__ => $_ ], @bind ], ($attr || () ));
425 =item Arguments: @values | \%cols, \%attrs?
427 =item Return Value: $row_object | undef
431 Finds a row based on its primary key or unique constraint. For example, to find
432 a row by its primary key:
434 my $cd = $schema->resultset('CD')->find(5);
436 You can also find a row by a specific unique constraint using the C<key>
437 attribute. For example:
439 my $cd = $schema->resultset('CD')->find('Massive Attack', 'Mezzanine', {
440 key => 'cd_artist_title'
443 Additionally, you can specify the columns explicitly by name:
445 my $cd = $schema->resultset('CD')->find(
447 artist => 'Massive Attack',
448 title => 'Mezzanine',
450 { key => 'cd_artist_title' }
453 If the C<key> is specified as C<primary>, it searches only on the primary key.
455 If no C<key> is specified, it searches on all unique constraints defined on the
456 source for which column data is provided, including the primary key.
458 If your table does not have a primary key, you B<must> provide a value for the
459 C<key> attribute matching one of the unique constraints on the source.
461 In addition to C<key>, L</find> recognizes and applies standard
462 L<resultset attributes|/ATTRIBUTES> in the same way as L</search> does.
464 Note: If your query does not return only one row, a warning is generated:
466 Query returned more than one row
468 See also L</find_or_create> and L</update_or_create>. For information on how to
469 declare unique constraints, see
470 L<DBIx::Class::ResultSource/add_unique_constraint>.
476 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
478 # Default to the primary key, but allow a specific key
479 my @cols = exists $attrs->{key}
480 ? $self->result_source->unique_constraint_columns($attrs->{key})
481 : $self->result_source->primary_columns;
482 $self->throw_exception(
483 "Can't find unless a primary key is defined or unique constraint is specified"
486 # Parse out a hashref from input
488 if (ref $_[0] eq 'HASH') {
489 $input_query = { %{$_[0]} };
491 elsif (@_ == @cols) {
493 @{$input_query}{@cols} = @_;
496 # Compatibility: Allow e.g. find(id => $value)
497 carp "Find by key => value deprecated; please use a hashref instead";
501 my (%related, $info);
503 KEY: foreach my $key (keys %$input_query) {
504 if (ref($input_query->{$key})
505 && ($info = $self->result_source->relationship_info($key))) {
506 my $val = delete $input_query->{$key};
507 next KEY if (ref($val) eq 'ARRAY'); # has_many for multi_create
508 my $rel_q = $self->result_source->resolve_condition(
509 $info->{cond}, $val, $key
511 die "Can't handle OR join condition in find" if ref($rel_q) eq 'ARRAY';
512 @related{keys %$rel_q} = values %$rel_q;
515 if (my @keys = keys %related) {
516 @{$input_query}{@keys} = values %related;
520 # Build the final query: Default to the disjunction of the unique queries,
521 # but allow the input query in case the ResultSet defines the query or the
522 # user is abusing find
523 my $alias = exists $attrs->{alias} ? $attrs->{alias} : $self->{attrs}{alias};
525 if (exists $attrs->{key}) {
526 my @unique_cols = $self->result_source->unique_constraint_columns($attrs->{key});
527 my $unique_query = $self->_build_unique_query($input_query, \@unique_cols);
528 $query = $self->_add_alias($unique_query, $alias);
531 my @unique_queries = $self->_unique_queries($input_query, $attrs);
532 $query = @unique_queries
533 ? [ map { $self->_add_alias($_, $alias) } @unique_queries ]
534 : $self->_add_alias($input_query, $alias);
539 my $rs = $self->search($query, $attrs);
540 if (keys %{$rs->_resolved_attrs->{collapse}}) {
542 carp "Query returned more than one row" if $rs->next;
550 if (keys %{$self->_resolved_attrs->{collapse}}) {
551 my $rs = $self->search($query);
553 carp "Query returned more than one row" if $rs->next;
557 return $self->single($query);
564 # Add the specified alias to the specified query hash. A copy is made so the
565 # original query is not modified.
568 my ($self, $query, $alias) = @_;
570 my %aliased = %$query;
571 foreach my $col (grep { ! m/\./ } keys %aliased) {
572 $aliased{"$alias.$col"} = delete $aliased{$col};
580 # Build a list of queries which satisfy unique constraints.
582 sub _unique_queries {
583 my ($self, $query, $attrs) = @_;
585 my @constraint_names = exists $attrs->{key}
587 : $self->result_source->unique_constraint_names;
589 my $where = $self->_collapse_cond($self->{attrs}{where} || {});
590 my $num_where = scalar keys %$where;
593 foreach my $name (@constraint_names) {
594 my @unique_cols = $self->result_source->unique_constraint_columns($name);
595 my $unique_query = $self->_build_unique_query($query, \@unique_cols);
597 my $num_cols = scalar @unique_cols;
598 my $num_query = scalar keys %$unique_query;
600 my $total = $num_query + $num_where;
601 if ($num_query && ($num_query == $num_cols || $total == $num_cols)) {
602 # The query is either unique on its own or is unique in combination with
603 # the existing where clause
604 push @unique_queries, $unique_query;
608 return @unique_queries;
611 # _build_unique_query
613 # Constrain the specified query hash based on the specified column names.
615 sub _build_unique_query {
616 my ($self, $query, $unique_cols) = @_;
619 map { $_ => $query->{$_} }
620 grep { exists $query->{$_} }
625 =head2 search_related
629 =item Arguments: $rel, $cond, \%attrs?
631 =item Return Value: $new_resultset
635 $new_rs = $cd_rs->search_related('artist', {
639 Searches the specified relationship, optionally specifying a condition and
640 attributes for matching records. See L</ATTRIBUTES> for more information.
645 return shift->related_resultset(shift)->search(@_);
648 =head2 search_related_rs
650 This method works exactly the same as search_related, except that
651 it guarantees a restultset, even in list context.
655 sub search_related_rs {
656 return shift->related_resultset(shift)->search_rs(@_);
663 =item Arguments: none
665 =item Return Value: $cursor
669 Returns a storage-driven cursor to the given resultset. See
670 L<DBIx::Class::Cursor> for more information.
677 my $attrs = { %{$self->_resolved_attrs} };
678 return $self->{cursor}
679 ||= $self->result_source->storage->select($attrs->{from}, $attrs->{select},
680 $attrs->{where},$attrs);
687 =item Arguments: $cond?
689 =item Return Value: $row_object?
693 my $cd = $schema->resultset('CD')->single({ year => 2001 });
695 Inflates the first result without creating a cursor if the resultset has
696 any records in it; if not returns nothing. Used by L</find> as a lean version of
699 While this method can take an optional search condition (just like L</search>)
700 being a fast-code-path it does not recognize search attributes. If you need to
701 add extra joins or similar, call L</search> and then chain-call L</single> on the
702 L<DBIx::Class::ResultSet> returned.
708 As of 0.08100, this method enforces the assumption that the preceeding
709 query returns only one row. If more than one row is returned, you will receive
712 Query returned more than one row
714 In this case, you should be using L</first> or L</find> instead, or if you really
715 know what you are doing, use the L</rows> attribute to explicitly limit the size
723 my ($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} };
730 if (defined $attrs->{where}) {
733 [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ }
734 $where, delete $attrs->{where} ]
737 $attrs->{where} = $where;
741 # XXX: Disabled since it doesn't infer uniqueness in all cases
742 # unless ($self->_is_unique_query($attrs->{where})) {
743 # carp "Query not guaranteed to return a single row"
744 # . "; please declare your unique constraints or use search instead";
747 my @data = $self->result_source->storage->select_single(
748 $attrs->{from}, $attrs->{select},
749 $attrs->{where}, $attrs
752 return (@data ? ($self->_construct_object(@data))[0] : undef);
757 # Try to determine if the specified query is guaranteed to be unique, based on
758 # the declared unique constraints.
760 sub _is_unique_query {
761 my ($self, $query) = @_;
763 my $collapsed = $self->_collapse_query($query);
764 my $alias = $self->{attrs}{alias};
766 foreach my $name ($self->result_source->unique_constraint_names) {
767 my @unique_cols = map {
769 } $self->result_source->unique_constraint_columns($name);
771 # Count the values for each unique column
772 my %seen = map { $_ => 0 } @unique_cols;
774 foreach my $key (keys %$collapsed) {
775 my $aliased = $key =~ /\./ ? $key : "$alias.$key";
776 next unless exists $seen{$aliased}; # Additional constraints are okay
777 $seen{$aliased} = scalar keys %{ $collapsed->{$key} };
780 # If we get 0 or more than 1 value for a column, it's not necessarily unique
781 return 1 unless grep { $_ != 1 } values %seen;
789 # Recursively collapse the query, accumulating values for each column.
791 sub _collapse_query {
792 my ($self, $query, $collapsed) = @_;
796 if (ref $query eq 'ARRAY') {
797 foreach my $subquery (@$query) {
798 next unless ref $subquery; # -or
799 # warn "ARRAY: " . Dumper $subquery;
800 $collapsed = $self->_collapse_query($subquery, $collapsed);
803 elsif (ref $query eq 'HASH') {
804 if (keys %$query and (keys %$query)[0] eq '-and') {
805 foreach my $subquery (@{$query->{-and}}) {
806 # warn "HASH: " . Dumper $subquery;
807 $collapsed = $self->_collapse_query($subquery, $collapsed);
811 # warn "LEAF: " . Dumper $query;
812 foreach my $col (keys %$query) {
813 my $value = $query->{$col};
814 $collapsed->{$col}{$value}++;
826 =item Arguments: $cond?
828 =item Return Value: $resultsetcolumn
832 my $max_length = $rs->get_column('length')->max;
834 Returns a L<DBIx::Class::ResultSetColumn> instance for a column of the ResultSet.
839 my ($self, $column) = @_;
840 my $new = DBIx::Class::ResultSetColumn->new($self, $column);
848 =item Arguments: $cond, \%attrs?
850 =item Return Value: $resultset (scalar context), @row_objs (list context)
854 # WHERE title LIKE '%blue%'
855 $cd_rs = $rs->search_like({ title => '%blue%'});
857 Performs a search, but uses C<LIKE> instead of C<=> as the condition. Note
858 that this is simply a convenience method retained for ex Class::DBI users.
859 You most likely want to use L</search> with specific operators.
861 For more information, see L<DBIx::Class::Manual::Cookbook>.
863 This method is deprecated and will be removed in 0.09. Use L</search()>
864 instead. An example conversion is:
866 ->search_like({ foo => 'bar' });
870 ->search({ foo => { like => 'bar' } });
877 'search_like() is deprecated and will be removed in 0.09.',
878 'Instead use ->search({ x => { -like => "y%" } })',
879 '(note the outer pair of {}s - they are important!)'
881 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
882 my $query = ref $_[0] eq 'HASH' ? { %{shift()} }: {@_};
883 $query->{$_} = { 'like' => $query->{$_} } for keys %$query;
884 return $class->search($query, { %$attrs });
891 =item Arguments: $first, $last
893 =item Return Value: $resultset (scalar context), @row_objs (list context)
897 Returns a resultset or object list representing a subset of elements from the
898 resultset slice is called on. Indexes are from 0, i.e., to get the first
901 my ($one, $two, $three) = $rs->slice(0, 2);
906 my ($self, $min, $max) = @_;
907 my $attrs = {}; # = { %{ $self->{attrs} || {} } };
908 $attrs->{offset} = $self->{attrs}{offset} || 0;
909 $attrs->{offset} += $min;
910 $attrs->{rows} = ($max ? ($max - $min + 1) : 1);
911 return $self->search(undef(), $attrs);
912 #my $slice = (ref $self)->new($self->result_source, $attrs);
913 #return (wantarray ? $slice->all : $slice);
920 =item Arguments: none
922 =item Return Value: $result?
926 Returns the next element in the resultset (C<undef> is there is none).
928 Can be used to efficiently iterate over records in the resultset:
930 my $rs = $schema->resultset('CD')->search;
931 while (my $cd = $rs->next) {
935 Note that you need to store the resultset object, and call C<next> on it.
936 Calling C<< resultset('Table')->next >> repeatedly will always return the
937 first record from the resultset.
943 if (my $cache = $self->get_cache) {
944 $self->{all_cache_position} ||= 0;
945 return $cache->[$self->{all_cache_position}++];
947 if ($self->{attrs}{cache}) {
948 $self->{all_cache_position} = 1;
949 return ($self->all)[0];
951 if ($self->{stashed_objects}) {
952 my $obj = shift(@{$self->{stashed_objects}});
953 delete $self->{stashed_objects} unless @{$self->{stashed_objects}};
957 exists $self->{stashed_row}
958 ? @{delete $self->{stashed_row}}
959 : $self->cursor->next
961 return undef unless (@row);
962 my ($row, @more) = $self->_construct_object(@row);
963 $self->{stashed_objects} = \@more if @more;
967 sub _construct_object {
968 my ($self, @row) = @_;
969 my $info = $self->_collapse_result($self->{_attrs}{as}, \@row);
970 my @new = $self->result_class->inflate_result($self->result_source, @$info);
971 @new = $self->{_attrs}{record_filter}->(@new)
972 if exists $self->{_attrs}{record_filter};
976 sub _collapse_result {
977 my ($self, $as_proto, $row) = @_;
981 # 'foo' => [ undef, 'foo' ]
982 # 'foo.bar' => [ 'foo', 'bar' ]
983 # 'foo.bar.baz' => [ 'foo.bar', 'baz' ]
985 my @construct_as = map { [ (/^(?:(.*)\.)?([^.]+)$/) ] } @$as_proto;
987 my %collapse = %{$self->{_attrs}{collapse}||{}};
991 # if we're doing collapsing (has_many prefetch) we need to grab records
992 # until the PK changes, so fill @pri_index. if not, we leave it empty so
993 # we know we don't have to bother.
995 # the reason for not using the collapse stuff directly is because if you
996 # had for e.g. two artists in a row with no cds, the collapse info for
997 # both would be NULL (undef) so you'd lose the second artist
999 # store just the index so we can check the array positions from the row
1000 # without having to contruct the full hash
1002 if (keys %collapse) {
1003 my %pri = map { ($_ => 1) } $self->result_source->primary_columns;
1004 foreach my $i (0 .. $#construct_as) {
1005 next if defined($construct_as[$i][0]); # only self table
1006 if (delete $pri{$construct_as[$i][1]}) {
1007 push(@pri_index, $i);
1009 last unless keys %pri; # short circuit (Johnny Five Is Alive!)
1013 # no need to do an if, it'll be empty if @pri_index is empty anyway
1015 my %pri_vals = map { ($_ => $copy[$_]) } @pri_index;
1019 do { # no need to check anything at the front, we always want the first row
1023 foreach my $this_as (@construct_as) {
1024 $const{$this_as->[0]||''}{$this_as->[1]} = shift(@copy);
1027 push(@const_rows, \%const);
1029 } until ( # no pri_index => no collapse => drop straight out
1032 do { # get another row, stash it, drop out if different PK
1034 @copy = $self->cursor->next;
1035 $self->{stashed_row} = \@copy;
1037 # last thing in do block, counts as true if anything doesn't match
1039 # check xor defined first for NULL vs. NOT NULL then if one is
1040 # defined the other must be so check string equality
1043 (defined $pri_vals{$_} ^ defined $copy[$_])
1044 || (defined $pri_vals{$_} && ($pri_vals{$_} ne $copy[$_]))
1049 my $alias = $self->{attrs}{alias};
1056 foreach my $const (@const_rows) {
1057 scalar @const_keys or do {
1058 @const_keys = sort { length($a) <=> length($b) } keys %$const;
1060 foreach my $key (@const_keys) {
1063 my @parts = split(/\./, $key);
1065 my $data = $const->{$key};
1066 foreach my $p (@parts) {
1067 $target = $target->[1]->{$p} ||= [];
1069 if ($cur eq ".${key}" && (my @ckey = @{$collapse{$cur}||[]})) {
1070 # collapsing at this point and on final part
1071 my $pos = $collapse_pos{$cur};
1072 CK: foreach my $ck (@ckey) {
1073 if (!defined $pos->{$ck} || $pos->{$ck} ne $data->{$ck}) {
1074 $collapse_pos{$cur} = $data;
1075 delete @collapse_pos{ # clear all positioning for sub-entries
1076 grep { m/^\Q${cur}.\E/ } keys %collapse_pos
1083 if (exists $collapse{$cur}) {
1084 $target = $target->[-1];
1087 $target->[0] = $data;
1089 $info->[0] = $const->{$key};
1097 =head2 result_source
1101 =item Arguments: $result_source?
1103 =item Return Value: $result_source
1107 An accessor for the primary ResultSource object from which this ResultSet
1114 =item Arguments: $result_class?
1116 =item Return Value: $result_class
1120 An accessor for the class to use when creating row objects. Defaults to
1121 C<< result_source->result_class >> - which in most cases is the name of the
1122 L<"table"|DBIx::Class::Manual::Glossary/"ResultSource"> class.
1124 Note that changing the result_class will also remove any components
1125 that were originally loaded in the source class via
1126 L<DBIx::Class::ResultSource/load_components>. Any overloaded methods
1127 in the original source class will not run.
1132 my ($self, $result_class) = @_;
1133 if ($result_class) {
1134 $self->ensure_class_loaded($result_class);
1135 $self->_result_class($result_class);
1137 $self->_result_class;
1144 =item Arguments: $cond, \%attrs??
1146 =item Return Value: $count
1150 Performs an SQL C<COUNT> with the same query as the resultset was built
1151 with to find the number of elements. If passed arguments, does a search
1152 on the resultset and counts the results of that.
1158 return $self->search(@_)->count if @_ and defined $_[0];
1159 return scalar @{ $self->get_cache } if $self->get_cache;
1160 my $count = $self->_count;
1161 return 0 unless $count;
1163 # need to take offset from resolved attrs
1165 $count -= $self->{_attrs}{offset} if $self->{_attrs}{offset};
1166 $count = $self->{attrs}{rows} if
1167 $self->{attrs}{rows} and $self->{attrs}{rows} < $count;
1168 $count = 0 if ($count < 0);
1172 sub _count { # Separated out so pager can get the full count
1174 my $attrs = { %{$self->_resolved_attrs} };
1176 if (my $group_by = $attrs->{group_by}) {
1177 delete $attrs->{order_by};
1179 $attrs->{select} = $group_by;
1180 $attrs->{from} = [ { 'mesub' => (ref $self)->new($self->result_source, $attrs)->cursor->as_query } ];
1181 delete $attrs->{where};
1184 $attrs->{select} = { count => '*' };
1185 $attrs->{as} = [qw/count/];
1187 # offset, order by, group by, where and page are not needed to count. record_filter is cdbi
1188 delete $attrs->{$_} for qw/rows offset order_by group_by page pager record_filter/;
1190 my $tmp_rs = (ref $self)->new($self->result_source, $attrs);
1191 my ($count) = $tmp_rs->cursor->next;
1199 =head2 count_literal
1203 =item Arguments: $sql_fragment, @bind_values
1205 =item Return Value: $count
1209 Counts the results in a literal query. Equivalent to calling L</search_literal>
1210 with the passed arguments, then L</count>.
1214 sub count_literal { shift->search_literal(@_)->count; }
1220 =item Arguments: none
1222 =item Return Value: @objects
1226 Returns all elements in the resultset. Called implicitly if the resultset
1227 is returned in list context.
1234 $self->throw_exception("all() doesn't take any arguments, you probably wanted ->search(...)->all()");
1237 return @{ $self->get_cache } if $self->get_cache;
1241 # TODO: don't call resolve here
1242 if (keys %{$self->_resolved_attrs->{collapse}}) {
1243 # if ($self->{attrs}{prefetch}) {
1244 # Using $self->cursor->all is really just an optimisation.
1245 # If we're collapsing has_many prefetches it probably makes
1246 # very little difference, and this is cleaner than hacking
1247 # _construct_object to survive the approach
1248 my @row = $self->cursor->next;
1250 push(@obj, $self->_construct_object(@row));
1251 @row = (exists $self->{stashed_row}
1252 ? @{delete $self->{stashed_row}}
1253 : $self->cursor->next);
1256 @obj = map { $self->_construct_object(@$_) } $self->cursor->all;
1259 $self->set_cache(\@obj) if $self->{attrs}{cache};
1267 =item Arguments: none
1269 =item Return Value: $self
1273 Resets the resultset's cursor, so you can iterate through the elements again.
1279 delete $self->{_attrs} if exists $self->{_attrs};
1280 $self->{all_cache_position} = 0;
1281 $self->cursor->reset;
1289 =item Arguments: none
1291 =item Return Value: $object?
1295 Resets the resultset and returns an object for the first result (if the
1296 resultset returns anything).
1301 return $_[0]->reset->next;
1304 # _cond_for_update_delete
1306 # update/delete require the condition to be modified to handle
1307 # the differing SQL syntax available. This transforms the $self->{cond}
1308 # appropriately, returning the new condition.
1310 sub _cond_for_update_delete {
1311 my ($self, $full_cond) = @_;
1314 $full_cond ||= $self->{cond};
1315 # No-op. No condition, we're updating/deleting everything
1316 return $cond unless ref $full_cond;
1318 if (ref $full_cond eq 'ARRAY') {
1322 foreach my $key (keys %{$_}) {
1324 $hash{$1} = $_->{$key};
1330 elsif (ref $full_cond eq 'HASH') {
1331 if ((keys %{$full_cond})[0] eq '-and') {
1334 my @cond = @{$full_cond->{-and}};
1335 for (my $i = 0; $i < @cond; $i++) {
1336 my $entry = $cond[$i];
1339 if (ref $entry eq 'HASH') {
1340 $hash = $self->_cond_for_update_delete($entry);
1343 $entry =~ /([^.]+)$/;
1344 $hash->{$1} = $cond[++$i];
1347 push @{$cond->{-and}}, $hash;
1351 foreach my $key (keys %{$full_cond}) {
1353 $cond->{$1} = $full_cond->{$key};
1358 $self->throw_exception(
1359 "Can't update/delete on resultset with condition unless hash or array"
1371 =item Arguments: \%values
1373 =item Return Value: $storage_rv
1377 Sets the specified columns in the resultset to the supplied values in a
1378 single query. Return value will be true if the update succeeded or false
1379 if no records were updated; exact type of success value is storage-dependent.
1384 my ($self, $values) = @_;
1385 $self->throw_exception("Values for update must be a hash")
1386 unless ref $values eq 'HASH';
1388 carp( 'WARNING! Currently $rs->update() does not generate proper SQL'
1389 . ' on joined resultsets, and may affect rows well outside of the'
1390 . ' contents of $rs. Use at your own risk' )
1391 if ( $self->{attrs}{seen_join} );
1393 my $cond = $self->_cond_for_update_delete;
1395 return $self->result_source->storage->update(
1396 $self->result_source, $values, $cond
1404 =item Arguments: \%values
1406 =item Return Value: 1
1410 Fetches all objects and updates them one at a time. Note that C<update_all>
1411 will run DBIC cascade triggers, while L</update> will not.
1416 my ($self, $values) = @_;
1417 $self->throw_exception("Values for update must be a hash")
1418 unless ref $values eq 'HASH';
1419 foreach my $obj ($self->all) {
1420 $obj->set_columns($values)->update;
1429 =item Arguments: none
1431 =item Return Value: 1
1435 Deletes the contents of the resultset from its result source. Note that this
1436 will not run DBIC cascade triggers. See L</delete_all> if you need triggers
1437 to run. See also L<DBIx::Class::Row/delete>.
1439 delete may not generate correct SQL for a query with joins or a resultset
1440 chained from a related resultset. In this case it will generate a warning:-
1442 WARNING! Currently $rs->delete() does not generate proper SQL on
1443 joined resultsets, and may delete rows well outside of the contents
1444 of $rs. Use at your own risk
1446 In these cases you may find that delete_all is more appropriate, or you
1447 need to respecify your query in a way that can be expressed without a join.
1453 $self->throw_exception("Delete should not be passed any arguments")
1455 carp( 'WARNING! Currently $rs->delete() does not generate proper SQL'
1456 . ' on joined resultsets, and may delete rows well outside of the'
1457 . ' contents of $rs. Use at your own risk' )
1458 if ( $self->{attrs}{seen_join} );
1459 my $cond = $self->_cond_for_update_delete;
1461 $self->result_source->storage->delete($self->result_source, $cond);
1469 =item Arguments: none
1471 =item Return Value: 1
1475 Fetches all objects and deletes them one at a time. Note that C<delete_all>
1476 will run DBIC cascade triggers, while L</delete> will not.
1482 $_->delete for $self->all;
1490 =item Arguments: \@data;
1494 Accepts either an arrayref of hashrefs or alternatively an arrayref of arrayrefs.
1495 For the arrayref of hashrefs style each hashref should be a structure suitable
1496 forsubmitting to a $resultset->create(...) method.
1498 In void context, C<insert_bulk> in L<DBIx::Class::Storage::DBI> is used
1499 to insert the data, as this is a faster method.
1501 Otherwise, each set of data is inserted into the database using
1502 L<DBIx::Class::ResultSet/create>, and a arrayref of the resulting row
1503 objects is returned.
1505 Example: Assuming an Artist Class that has many CDs Classes relating:
1507 my $Artist_rs = $schema->resultset("Artist");
1509 ## Void Context Example
1510 $Artist_rs->populate([
1511 { artistid => 4, name => 'Manufactured Crap', cds => [
1512 { title => 'My First CD', year => 2006 },
1513 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
1516 { artistid => 5, name => 'Angsty-Whiny Girl', cds => [
1517 { title => 'My parents sold me to a record company' ,year => 2005 },
1518 { title => 'Why Am I So Ugly?', year => 2006 },
1519 { title => 'I Got Surgery and am now Popular', year => 2007 }
1524 ## Array Context Example
1525 my ($ArtistOne, $ArtistTwo, $ArtistThree) = $Artist_rs->populate([
1526 { name => "Artist One"},
1527 { name => "Artist Two"},
1528 { name => "Artist Three", cds=> [
1529 { title => "First CD", year => 2007},
1530 { title => "Second CD", year => 2008},
1534 print $ArtistOne->name; ## response is 'Artist One'
1535 print $ArtistThree->cds->count ## reponse is '2'
1537 For the arrayref of arrayrefs style, the first element should be a list of the
1538 fieldsnames to which the remaining elements are rows being inserted. For
1541 $Arstist_rs->populate([
1542 [qw/artistid name/],
1543 [100, 'A Formally Unknown Singer'],
1544 [101, 'A singer that jumped the shark two albums ago'],
1545 [102, 'An actually cool singer.'],
1548 Please note an important effect on your data when choosing between void and
1549 wantarray context. Since void context goes straight to C<insert_bulk> in
1550 L<DBIx::Class::Storage::DBI> this will skip any component that is overriding
1551 c<insert>. So if you are using something like L<DBIx-Class-UUIDColumns> to
1552 create primary keys for you, you will find that your PKs are empty. In this
1553 case you will have to use the wantarray context in order to create those
1559 my $self = shift @_;
1560 my $data = ref $_[0][0] eq 'HASH'
1561 ? $_[0] : ref $_[0][0] eq 'ARRAY' ? $self->_normalize_populate_args($_[0]) :
1562 $self->throw_exception('Populate expects an arrayref of hashes or arrayref of arrayrefs');
1564 if(defined wantarray) {
1566 foreach my $item (@$data) {
1567 push(@created, $self->create($item));
1571 my ($first, @rest) = @$data;
1573 my @names = grep {!ref $first->{$_}} keys %$first;
1574 my @rels = grep { $self->result_source->has_relationship($_) } keys %$first;
1575 my @pks = $self->result_source->primary_columns;
1577 ## do the belongs_to relationships
1578 foreach my $index (0..$#$data) {
1579 if( grep { !defined $data->[$index]->{$_} } @pks ) {
1580 my @ret = $self->populate($data);
1584 foreach my $rel (@rels) {
1585 next unless $data->[$index]->{$rel} && ref $data->[$index]->{$rel} eq "HASH";
1586 my $result = $self->related_resultset($rel)->create($data->[$index]->{$rel});
1587 my ($reverse) = keys %{$self->result_source->reverse_relationship_info($rel)};
1588 my $related = $result->result_source->resolve_condition(
1589 $result->result_source->relationship_info($reverse)->{cond},
1594 delete $data->[$index]->{$rel};
1595 $data->[$index] = {%{$data->[$index]}, %$related};
1597 push @names, keys %$related if $index == 0;
1601 ## do bulk insert on current row
1602 my @values = map { [ @$_{@names} ] } @$data;
1604 $self->result_source->storage->insert_bulk(
1605 $self->result_source,
1610 ## do the has_many relationships
1611 foreach my $item (@$data) {
1613 foreach my $rel (@rels) {
1614 next unless $item->{$rel} && ref $item->{$rel} eq "ARRAY";
1616 my $parent = $self->find(map {{$_=>$item->{$_}} } @pks)
1617 || $self->throw_exception('Cannot find the relating object.');
1619 my $child = $parent->$rel;
1621 my $related = $child->result_source->resolve_condition(
1622 $parent->result_source->relationship_info($rel)->{cond},
1627 my @rows_to_add = ref $item->{$rel} eq 'ARRAY' ? @{$item->{$rel}} : ($item->{$rel});
1628 my @populate = map { {%$_, %$related} } @rows_to_add;
1630 $child->populate( \@populate );
1636 =head2 _normalize_populate_args ($args)
1638 Private method used by L</populate> to normalize its incoming arguments. Factored
1639 out in case you want to subclass and accept new argument structures to the
1640 L</populate> method.
1644 sub _normalize_populate_args {
1645 my ($self, $data) = @_;
1646 my @names = @{shift(@$data)};
1647 my @results_to_create;
1648 foreach my $datum (@$data) {
1649 my %result_to_create;
1650 foreach my $index (0..$#names) {
1651 $result_to_create{$names[$index]} = $$datum[$index];
1653 push @results_to_create, \%result_to_create;
1655 return \@results_to_create;
1662 =item Arguments: none
1664 =item Return Value: $pager
1668 Return Value a L<Data::Page> object for the current resultset. Only makes
1669 sense for queries with a C<page> attribute.
1671 To get the full count of entries for a paged resultset, call
1672 C<total_entries> on the L<Data::Page> object.
1678 my $attrs = $self->{attrs};
1679 $self->throw_exception("Can't create pager for non-paged rs")
1680 unless $self->{attrs}{page};
1681 $attrs->{rows} ||= 10;
1682 return $self->{pager} ||= Data::Page->new(
1683 $self->_count, $attrs->{rows}, $self->{attrs}{page});
1690 =item Arguments: $page_number
1692 =item Return Value: $rs
1696 Returns a resultset for the $page_number page of the resultset on which page
1697 is called, where each page contains a number of rows equal to the 'rows'
1698 attribute set on the resultset (10 by default).
1703 my ($self, $page) = @_;
1704 return (ref $self)->new($self->result_source, { %{$self->{attrs}}, page => $page });
1711 =item Arguments: \%vals
1713 =item Return Value: $rowobject
1717 Creates a new row object in the resultset's result class and returns
1718 it. The row is not inserted into the database at this point, call
1719 L<DBIx::Class::Row/insert> to do that. Calling L<DBIx::Class::Row/in_storage>
1720 will tell you whether the row object has been inserted or not.
1722 Passes the hashref of input on to L<DBIx::Class::Row/new>.
1727 my ($self, $values) = @_;
1728 $self->throw_exception( "new_result needs a hash" )
1729 unless (ref $values eq 'HASH');
1732 my $alias = $self->{attrs}{alias};
1735 defined $self->{cond}
1736 && $self->{cond} eq $DBIx::Class::ResultSource::UNRESOLVABLE_CONDITION
1738 %new = %{ $self->{attrs}{related_objects} || {} }; # nothing might have been inserted yet
1739 $new{-from_resultset} = [ keys %new ] if keys %new;
1741 $self->throw_exception(
1742 "Can't abstract implicit construct, condition not a hash"
1743 ) if ($self->{cond} && !(ref $self->{cond} eq 'HASH'));
1745 my $collapsed_cond = (
1747 ? $self->_collapse_cond($self->{cond})
1751 # precendence must be given to passed values over values inherited from
1752 # the cond, so the order here is important.
1753 my %implied = %{$self->_remove_alias($collapsed_cond, $alias)};
1754 while( my($col,$value) = each %implied ){
1755 if(ref($value) eq 'HASH' && keys(%$value) && (keys %$value)[0] eq '='){
1756 $new{$col} = $value->{'='};
1759 $new{$col} = $value if $self->_is_deterministic_value($value);
1765 %{ $self->_remove_alias($values, $alias) },
1766 -source_handle => $self->_source_handle,
1767 -result_source => $self->result_source, # DO NOT REMOVE THIS, REQUIRED
1770 return $self->result_class->new(\%new);
1773 # _is_deterministic_value
1775 # Make an effor to strip non-deterministic values from the condition,
1776 # to make sure new_result chokes less
1778 sub _is_deterministic_value {
1781 my $ref_type = ref $value;
1782 return 1 if $ref_type eq '' || $ref_type eq 'SCALAR';
1783 return 1 if Scalar::Util::blessed($value);
1789 # Recursively collapse the condition.
1791 sub _collapse_cond {
1792 my ($self, $cond, $collapsed) = @_;
1796 if (ref $cond eq 'ARRAY') {
1797 foreach my $subcond (@$cond) {
1798 next unless ref $subcond; # -or
1799 # warn "ARRAY: " . Dumper $subcond;
1800 $collapsed = $self->_collapse_cond($subcond, $collapsed);
1803 elsif (ref $cond eq 'HASH') {
1804 if (keys %$cond and (keys %$cond)[0] eq '-and') {
1805 foreach my $subcond (@{$cond->{-and}}) {
1806 # warn "HASH: " . Dumper $subcond;
1807 $collapsed = $self->_collapse_cond($subcond, $collapsed);
1811 # warn "LEAF: " . Dumper $cond;
1812 foreach my $col (keys %$cond) {
1813 my $value = $cond->{$col};
1814 $collapsed->{$col} = $value;
1824 # Remove the specified alias from the specified query hash. A copy is made so
1825 # the original query is not modified.
1828 my ($self, $query, $alias) = @_;
1830 my %orig = %{ $query || {} };
1833 foreach my $key (keys %orig) {
1835 $unaliased{$key} = $orig{$key};
1838 $unaliased{$1} = $orig{$key}
1839 if $key =~ m/^(?:\Q$alias\E\.)?([^.]+)$/;
1845 =head2 as_query (EXPERIMENTAL)
1849 =item Arguments: none
1851 =item Return Value: \[ $sql, @bind ]
1855 Returns the SQL query and bind vars associated with the invocant.
1857 This is generally used as the RHS for a subquery.
1859 B<NOTE>: This feature is still experimental.
1863 sub as_query { return shift->cursor->as_query(@_) }
1869 =item Arguments: \%vals, \%attrs?
1871 =item Return Value: $rowobject
1875 my $artist = $schema->resultset('Artist')->find_or_new(
1876 { artist => 'fred' }, { key => 'artists' });
1878 $cd->cd_to_producer->find_or_new({ producer => $producer },
1879 { key => 'primary });
1881 Find an existing record from this resultset, based on its primary
1882 key, or a unique constraint. If none exists, instantiate a new result
1883 object and return it. The object will not be saved into your storage
1884 until you call L<DBIx::Class::Row/insert> on it.
1886 You most likely want this method when looking for existing rows using
1887 a unique constraint that is not the primary key, or looking for
1890 If you want objects to be saved immediately, use L</find_or_create> instead.
1892 B<Note>: C<find_or_new> is probably not what you want when creating a
1893 new row in a table that uses primary keys supplied by the
1894 database. Passing in a primary key column with a value of I<undef>
1895 will cause L</find> to attempt to search for a row with a value of
1902 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
1903 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
1904 my $exists = $self->find($hash, $attrs);
1905 return defined $exists ? $exists : $self->new_result($hash);
1912 =item Arguments: \%vals
1914 =item Return Value: a L<DBIx::Class::Row> $object
1918 Attempt to create a single new row or a row with multiple related rows
1919 in the table represented by the resultset (and related tables). This
1920 will not check for duplicate rows before inserting, use
1921 L</find_or_create> to do that.
1923 To create one row for this resultset, pass a hashref of key/value
1924 pairs representing the columns of the table and the values you wish to
1925 store. If the appropriate relationships are set up, foreign key fields
1926 can also be passed an object representing the foreign row, and the
1927 value will be set to its primary key.
1929 To create related objects, pass a hashref for the value if the related
1930 item is a foreign key relationship (L<DBIx::Class::Relationship/belongs_to>),
1931 and use the name of the relationship as the key. (NOT the name of the field,
1932 necessarily). For C<has_many> and C<has_one> relationships, pass an arrayref
1933 of hashrefs containing the data for each of the rows to create in the foreign
1934 tables, again using the relationship name as the key.
1936 Instead of hashrefs of plain related data (key/value pairs), you may
1937 also pass new or inserted objects. New objects (not inserted yet, see
1938 L</new>), will be inserted into their appropriate tables.
1940 Effectively a shortcut for C<< ->new_result(\%vals)->insert >>.
1942 Example of creating a new row.
1944 $person_rs->create({
1945 name=>"Some Person",
1946 email=>"somebody@someplace.com"
1949 Example of creating a new row and also creating rows in a related C<has_many>
1950 or C<has_one> resultset. Note Arrayref.
1953 { artistid => 4, name => 'Manufactured Crap', cds => [
1954 { title => 'My First CD', year => 2006 },
1955 { title => 'Yet More Tweeny-Pop crap', year => 2007 },
1960 Example of creating a new row and also creating a row in a related
1961 C<belongs_to>resultset. Note Hashref.
1964 title=>"Music for Silly Walks",
1967 name=>"Silly Musician",
1974 my ($self, $attrs) = @_;
1975 $self->throw_exception( "create needs a hashref" )
1976 unless ref $attrs eq 'HASH';
1977 return $self->new_result($attrs)->insert;
1980 =head2 find_or_create
1984 =item Arguments: \%vals, \%attrs?
1986 =item Return Value: $rowobject
1990 $cd->cd_to_producer->find_or_create({ producer => $producer },
1991 { key => 'primary });
1993 Tries to find a record based on its primary key or unique constraints; if none
1994 is found, creates one and returns that instead.
1996 my $cd = $schema->resultset('CD')->find_or_create({
1998 artist => 'Massive Attack',
1999 title => 'Mezzanine',
2003 Also takes an optional C<key> attribute, to search by a specific key or unique
2004 constraint. For example:
2006 my $cd = $schema->resultset('CD')->find_or_create(
2008 artist => 'Massive Attack',
2009 title => 'Mezzanine',
2011 { key => 'cd_artist_title' }
2014 B<Note>: Because find_or_create() reads from the database and then
2015 possibly inserts based on the result, this method is subject to a race
2016 condition. Another process could create a record in the table after
2017 the find has completed and before the create has started. To avoid
2018 this problem, use find_or_create() inside a transaction.
2020 B<Note>: C<find_or_create> is probably not what you want when creating
2021 a new row in a table that uses primary keys supplied by the
2022 database. Passing in a primary key column with a value of I<undef>
2023 will cause L</find> to attempt to search for a row with a value of
2026 See also L</find> and L</update_or_create>. For information on how to declare
2027 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2031 sub find_or_create {
2033 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2034 my $hash = ref $_[0] eq 'HASH' ? shift : {@_};
2035 my $exists = $self->find($hash, $attrs);
2036 return defined $exists ? $exists : $self->create($hash);
2039 =head2 update_or_create
2043 =item Arguments: \%col_values, { key => $unique_constraint }?
2045 =item Return Value: $rowobject
2049 $resultset->update_or_create({ col => $val, ... });
2051 First, searches for an existing row matching one of the unique constraints
2052 (including the primary key) on the source of this resultset. If a row is
2053 found, updates it with the other given column values. Otherwise, creates a new
2056 Takes an optional C<key> attribute to search on a specific unique constraint.
2059 # In your application
2060 my $cd = $schema->resultset('CD')->update_or_create(
2062 artist => 'Massive Attack',
2063 title => 'Mezzanine',
2066 { key => 'cd_artist_title' }
2069 $cd->cd_to_producer->update_or_create({
2070 producer => $producer,
2077 If no C<key> is specified, it searches on all unique constraints defined on the
2078 source, including the primary key.
2080 If the C<key> is specified as C<primary>, it searches only on the primary key.
2082 See also L</find> and L</find_or_create>. For information on how to declare
2083 unique constraints, see L<DBIx::Class::ResultSource/add_unique_constraint>.
2085 B<Note>: C<update_or_create> is probably not what you want when
2086 looking for a row in a table that uses primary keys supplied by the
2087 database, unless you actually have a key value. Passing in a primary
2088 key column with a value of I<undef> will cause L</find> to attempt to
2089 search for a row with a value of I<NULL>.
2093 sub update_or_create {
2095 my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {});
2096 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2098 my $row = $self->find($cond, $attrs);
2100 $row->update($cond);
2104 return $self->create($cond);
2107 =head2 update_or_new
2111 =item Arguments: \%col_values, { key => $unique_constraint }?
2113 =item Return Value: $rowobject
2117 $resultset->update_or_new({ col => $val, ... });
2119 First, searches for an existing row matching one of the unique constraints
2120 (including the primary key) on the source of this resultset. If a row is
2121 found, updates it with the other given column values. Otherwise, instantiate
2122 a new result object and return it. The object will not be saved into your storage
2123 until you call L<DBIx::Class::Row/insert> on it.
2125 Takes an optional C<key> attribute to search on a specific unique constraint.
2128 # In your application
2129 my $cd = $schema->resultset('CD')->update_or_new(
2131 artist => 'Massive Attack',
2132 title => 'Mezzanine',
2135 { key => 'cd_artist_title' }
2138 if ($cd->in_storage) {
2139 # the cd was updated
2142 # the cd is not yet in the database, let's insert it
2146 See also L</find>, L</find_or_create> and L<find_or_new>.
2152 my $attrs = ( @_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {} );
2153 my $cond = ref $_[0] eq 'HASH' ? shift : {@_};
2155 my $row = $self->find( $cond, $attrs );
2156 if ( defined $row ) {
2157 $row->update($cond);
2161 return $self->new_result($cond);
2168 =item Arguments: none
2170 =item Return Value: \@cache_objects?
2174 Gets the contents of the cache for the resultset, if the cache is set.
2176 The cache is populated either by using the L</prefetch> attribute to
2177 L</search> or by calling L</set_cache>.
2189 =item Arguments: \@cache_objects
2191 =item Return Value: \@cache_objects
2195 Sets the contents of the cache for the resultset. Expects an arrayref
2196 of objects of the same class as those produced by the resultset. Note that
2197 if the cache is set the resultset will return the cached objects rather
2198 than re-querying the database even if the cache attr is not set.
2200 The contents of the cache can also be populated by using the
2201 L</prefetch> attribute to L</search>.
2206 my ( $self, $data ) = @_;
2207 $self->throw_exception("set_cache requires an arrayref")
2208 if defined($data) && (ref $data ne 'ARRAY');
2209 $self->{all_cache} = $data;
2216 =item Arguments: none
2218 =item Return Value: []
2222 Clears the cache for the resultset.
2227 shift->set_cache(undef);
2230 =head2 related_resultset
2234 =item Arguments: $relationship_name
2236 =item Return Value: $resultset
2240 Returns a related resultset for the supplied relationship name.
2242 $artist_rs = $schema->resultset('CD')->related_resultset('Artist');
2246 sub related_resultset {
2247 my ($self, $rel) = @_;
2249 $self->{related_resultsets} ||= {};
2250 return $self->{related_resultsets}{$rel} ||= do {
2251 my $rel_obj = $self->result_source->relationship_info($rel);
2253 $self->throw_exception(
2254 "search_related: result source '" . $self->result_source->source_name .
2255 "' has no such relationship $rel")
2258 my ($from,$seen) = $self->_resolve_from($rel);
2260 my $join_count = $seen->{$rel};
2261 my $alias = ($join_count > 1 ? join('_', $rel, $join_count) : $rel);
2263 #XXX - temp fix for result_class bug. There likely is a more elegant fix -groditi
2264 my %attrs = %{$self->{attrs}||{}};
2265 delete @attrs{qw(result_class alias)};
2269 if (my $cache = $self->get_cache) {
2270 if ($cache->[0] && $cache->[0]->related_resultset($rel)->get_cache) {
2271 $new_cache = [ map { @{$_->related_resultset($rel)->get_cache} }
2276 my $rel_source = $self->result_source->related_source($rel);
2280 # The reason we do this now instead of passing the alias to the
2281 # search_rs below is that if you wrap/overload resultset on the
2282 # source you need to know what alias it's -going- to have for things
2283 # to work sanely (e.g. RestrictWithObject wants to be able to add
2284 # extra query restrictions, and these may need to be $alias.)
2286 my $attrs = $rel_source->resultset_attributes;
2287 local $attrs->{alias} = $alias;
2289 $rel_source->resultset
2297 where => $self->{cond},
2302 $new->set_cache($new_cache) if $new_cache;
2307 =head2 current_source_alias
2311 =item Arguments: none
2313 =item Return Value: $source_alias
2317 Returns the current table alias for the result source this resultset is built
2318 on, that will be used in the SQL query. Usually it is C<me>.
2320 Currently the source alias that refers to the result set returned by a
2321 L</search>/L</find> family method depends on how you got to the resultset: it's
2322 C<me> by default, but eg. L</search_related> aliases it to the related result
2323 source name (and keeps C<me> referring to the original result set). The long
2324 term goal is to make L<DBIx::Class> always alias the current resultset as C<me>
2325 (and make this method unnecessary).
2327 Thus it's currently necessary to use this method in predefined queries (see
2328 L<DBIx::Class::Manual::Cookbook/Predefined searches>) when referring to the
2329 source alias of the current result set:
2331 # in a result set class
2333 my ($self, $user) = @_;
2335 my $me = $self->current_source_alias;
2337 return $self->search(
2338 "$me.modified" => $user->id,
2344 sub current_source_alias {
2347 return ($self->{attrs} || {})->{alias} || 'me';
2351 my ($self, $extra_join) = @_;
2352 my $source = $self->result_source;
2353 my $attrs = $self->{attrs};
2355 my $from = $attrs->{from}
2356 || [ { $attrs->{alias} => $source->from } ];
2358 my $seen = { %{$attrs->{seen_join}||{}} };
2360 my $join = ($attrs->{join}
2361 ? [ $attrs->{join}, $extra_join ]
2364 # we need to take the prefetch the attrs into account before we
2365 # ->resolve_join as otherwise they get lost - captainL
2366 my $merged = $self->_merge_attr( $join, $attrs->{prefetch} );
2370 ($join ? $source->resolve_join($merged, $attrs->{alias}, $seen) : ()),
2373 return ($from,$seen);
2376 sub _resolved_attrs {
2378 return $self->{_attrs} if $self->{_attrs};
2380 my $attrs = { %{ $self->{attrs} || {} } };
2381 my $source = $self->result_source;
2382 my $alias = $attrs->{alias};
2384 $attrs->{columns} ||= delete $attrs->{cols} if exists $attrs->{cols};
2387 # build columns (as long as select isn't set) into a set of as/select hashes
2388 unless ( $attrs->{select} ) {
2390 ( ref($_) eq 'HASH' )
2394 /^\Q${alias}.\E(.+)$/
2405 } ( ref($attrs->{columns}) eq 'ARRAY' ) ? @{ delete $attrs->{columns}} : (delete $attrs->{columns} || $source->columns );
2407 # add the additional columns on
2408 foreach ( 'include_columns', '+columns' ) {
2409 push @colbits, map {
2410 ( ref($_) eq 'HASH' )
2412 : { ( split( /\./, $_ ) )[-1] => ( /\./ ? $_ : "${alias}.$_" ) }
2413 } ( ref($attrs->{$_}) eq 'ARRAY' ) ? @{ delete $attrs->{$_} } : delete $attrs->{$_} if ( $attrs->{$_} );
2416 # start with initial select items
2417 if ( $attrs->{select} ) {
2419 ( ref $attrs->{select} eq 'ARRAY' )
2420 ? [ @{ $attrs->{select} } ]
2421 : [ $attrs->{select} ];
2425 ref $attrs->{as} eq 'ARRAY'
2426 ? [ @{ $attrs->{as} } ]
2429 : [ map { m/^\Q${alias}.\E(.+)$/ ? $1 : $_ } @{ $attrs->{select} } ]
2434 # otherwise we intialise select & as to empty
2435 $attrs->{select} = [];
2439 # now add colbits to select/as
2440 push( @{ $attrs->{select} }, map { values( %{$_} ) } @colbits );
2441 push( @{ $attrs->{as} }, map { keys( %{$_} ) } @colbits );
2444 if ( $adds = delete $attrs->{'+select'} ) {
2445 $adds = [$adds] unless ref $adds eq 'ARRAY';
2447 @{ $attrs->{select} },
2448 map { /\./ || ref $_ ? $_ : "${alias}.$_" } @$adds
2451 if ( $adds = delete $attrs->{'+as'} ) {
2452 $adds = [$adds] unless ref $adds eq 'ARRAY';
2453 push( @{ $attrs->{as} }, @$adds );
2456 $attrs->{from} ||= [ { $self->{attrs}{alias} => $source->from } ];
2458 if ( exists $attrs->{join} || exists $attrs->{prefetch} ) {
2459 my $join = delete $attrs->{join} || {};
2461 if ( defined $attrs->{prefetch} ) {
2462 $join = $self->_merge_attr( $join, $attrs->{prefetch} );
2466 $attrs->{from} = # have to copy here to avoid corrupting the original
2468 @{ $attrs->{from} },
2469 $source->resolve_join(
2470 $join, $alias, { %{ $attrs->{seen_join} || {} } }
2476 $attrs->{group_by} ||= $attrs->{select}
2477 if delete $attrs->{distinct};
2478 if ( $attrs->{order_by} ) {
2479 $attrs->{order_by} = (
2480 ref( $attrs->{order_by} ) eq 'ARRAY'
2481 ? [ @{ $attrs->{order_by} } ]
2482 : [ $attrs->{order_by} ]
2486 $attrs->{order_by} = [];
2489 my $collapse = $attrs->{collapse} || {};
2490 if ( my $prefetch = delete $attrs->{prefetch} ) {
2491 $prefetch = $self->_merge_attr( {}, $prefetch );
2493 my $seen = { %{ $attrs->{seen_join} || {} } };
2494 foreach my $p ( ref $prefetch eq 'ARRAY' ? @$prefetch : ($prefetch) ) {
2496 # bring joins back to level of current class
2498 $source->resolve_prefetch( $p, $alias, $seen, \@pre_order, $collapse );
2499 push( @{ $attrs->{select} }, map { $_->[0] } @prefetch );
2500 push( @{ $attrs->{as} }, map { $_->[1] } @prefetch );
2502 push( @{ $attrs->{order_by} }, @pre_order );
2504 $attrs->{collapse} = $collapse;
2506 if ( $attrs->{page} ) {
2507 $attrs->{offset} ||= 0;
2508 $attrs->{offset} += ( $attrs->{rows} * ( $attrs->{page} - 1 ) );
2511 return $self->{_attrs} = $attrs;
2515 my ($self, $attr) = @_;
2517 if (ref $attr eq 'HASH') {
2518 return $self->_rollout_hash($attr);
2519 } elsif (ref $attr eq 'ARRAY') {
2520 return $self->_rollout_array($attr);
2526 sub _rollout_array {
2527 my ($self, $attr) = @_;
2530 foreach my $element (@{$attr}) {
2531 if (ref $element eq 'HASH') {
2532 push( @rolled_array, @{ $self->_rollout_hash( $element ) } );
2533 } elsif (ref $element eq 'ARRAY') {
2534 # XXX - should probably recurse here
2535 push( @rolled_array, @{$self->_rollout_array($element)} );
2537 push( @rolled_array, $element );
2540 return \@rolled_array;
2544 my ($self, $attr) = @_;
2547 foreach my $key (keys %{$attr}) {
2548 push( @rolled_array, { $key => $attr->{$key} } );
2550 return \@rolled_array;
2553 sub _calculate_score {
2554 my ($self, $a, $b) = @_;
2556 if (ref $b eq 'HASH') {
2557 my ($b_key) = keys %{$b};
2558 if (ref $a eq 'HASH') {
2559 my ($a_key) = keys %{$a};
2560 if ($a_key eq $b_key) {
2561 return (1 + $self->_calculate_score( $a->{$a_key}, $b->{$b_key} ));
2566 return ($a eq $b_key) ? 1 : 0;
2569 if (ref $a eq 'HASH') {
2570 my ($a_key) = keys %{$a};
2571 return ($b eq $a_key) ? 1 : 0;
2573 return ($b eq $a) ? 1 : 0;
2579 my ($self, $orig, $import) = @_;
2581 return $import unless defined($orig);
2582 return $orig unless defined($import);
2584 $orig = $self->_rollout_attr($orig);
2585 $import = $self->_rollout_attr($import);
2588 foreach my $import_element ( @{$import} ) {
2589 # find best candidate from $orig to merge $b_element into
2590 my $best_candidate = { position => undef, score => 0 }; my $position = 0;
2591 foreach my $orig_element ( @{$orig} ) {
2592 my $score = $self->_calculate_score( $orig_element, $import_element );
2593 if ($score > $best_candidate->{score}) {
2594 $best_candidate->{position} = $position;
2595 $best_candidate->{score} = $score;
2599 my ($import_key) = ( ref $import_element eq 'HASH' ) ? keys %{$import_element} : ($import_element);
2601 if ($best_candidate->{score} == 0 || exists $seen_keys->{$import_key}) {
2602 push( @{$orig}, $import_element );
2604 my $orig_best = $orig->[$best_candidate->{position}];
2605 # merge orig_best and b_element together and replace original with merged
2606 if (ref $orig_best ne 'HASH') {
2607 $orig->[$best_candidate->{position}] = $import_element;
2608 } elsif (ref $import_element eq 'HASH') {
2609 my ($key) = keys %{$orig_best};
2610 $orig->[$best_candidate->{position}] = { $key => $self->_merge_attr($orig_best->{$key}, $import_element->{$key}) };
2613 $seen_keys->{$import_key} = 1; # don't merge the same key twice
2623 $self->_source_handle($_[0]->handle);
2625 $self->_source_handle->resolve;
2629 =head2 throw_exception
2631 See L<DBIx::Class::Schema/throw_exception> for details.
2635 sub throw_exception {
2637 if (ref $self && $self->_source_handle->schema) {
2638 $self->_source_handle->schema->throw_exception(@_)
2645 # XXX: FIXME: Attributes docs need clearing up
2649 Attributes are used to refine a ResultSet in various ways when
2650 searching for data. They can be passed to any method which takes an
2651 C<\%attrs> argument. See L</search>, L</search_rs>, L</find>,
2654 These are in no particular order:
2660 =item Value: ( $order_by | \@order_by | \%order_by )
2664 Which column(s) to order the results by. If a single column name, or
2665 an arrayref of names is supplied, the argument is passed through
2666 directly to SQL. The hashref syntax allows for connection-agnostic
2667 specification of ordering direction:
2669 For descending order:
2671 order_by => { -desc => [qw/col1 col2 col3/] }
2673 For explicit ascending order:
2675 order_by => { -asc => 'col' }
2677 The old scalarref syntax (i.e. order_by => \'year DESC') is still
2678 supported, although you are strongly encouraged to use the hashref
2679 syntax as outlined above.
2685 =item Value: \@columns
2689 Shortcut to request a particular set of columns to be retrieved. Each
2690 column spec may be a string (a table column name), or a hash (in which
2691 case the key is the C<as> value, and the value is used as the C<select>
2692 expression). Adds C<me.> onto the start of any column without a C<.> in
2693 it and sets C<select> from that, then auto-populates C<as> from
2694 C<select> as normal. (You may also use the C<cols> attribute, as in
2695 earlier versions of DBIC.)
2701 =item Value: \@columns
2705 Indicates additional columns to be selected from storage. Works the same
2706 as L</columns> but adds columns to the selection. (You may also use the
2707 C<include_columns> attribute, as in earlier versions of DBIC). For
2710 $schema->resultset('CD')->search(undef, {
2711 '+columns' => ['artist.name'],
2715 would return all CDs and include a 'name' column to the information
2716 passed to object inflation. Note that the 'artist' is the name of the
2717 column (or relationship) accessor, and 'name' is the name of the column
2718 accessor in the related table.
2720 =head2 include_columns
2724 =item Value: \@columns
2728 Deprecated. Acts as a synonym for L</+columns> for backward compatibility.
2734 =item Value: \@select_columns
2738 Indicates which columns should be selected from the storage. You can use
2739 column names, or in the case of RDBMS back ends, function or stored procedure
2742 $rs = $schema->resultset('Employee')->search(undef, {
2745 { count => 'employeeid' },
2750 When you use function/stored procedure names and do not supply an C<as>
2751 attribute, the column names returned are storage-dependent. E.g. MySQL would
2752 return a column named C<count(employeeid)> in the above example.
2758 Indicates additional columns to be selected from storage. Works the same as
2759 L</select> but adds columns to the selection.
2767 Indicates additional column names for those added via L</+select>. See L</as>.
2775 =item Value: \@inflation_names
2779 Indicates column names for object inflation. That is, C<as>
2780 indicates the name that the column can be accessed as via the
2781 C<get_column> method (or via the object accessor, B<if one already
2782 exists>). It has nothing to do with the SQL code C<SELECT foo AS bar>.
2784 The C<as> attribute is used in conjunction with C<select>,
2785 usually when C<select> contains one or more function or stored
2788 $rs = $schema->resultset('Employee')->search(undef, {
2791 { count => 'employeeid' }
2793 as => ['name', 'employee_count'],
2796 my $employee = $rs->first(); # get the first Employee
2798 If the object against which the search is performed already has an accessor
2799 matching a column name specified in C<as>, the value can be retrieved using
2800 the accessor as normal:
2802 my $name = $employee->name();
2804 If on the other hand an accessor does not exist in the object, you need to
2805 use C<get_column> instead:
2807 my $employee_count = $employee->get_column('employee_count');
2809 You can create your own accessors if required - see
2810 L<DBIx::Class::Manual::Cookbook> for details.
2812 Please note: This will NOT insert an C<AS employee_count> into the SQL
2813 statement produced, it is used for internal access only. Thus
2814 attempting to use the accessor in an C<order_by> clause or similar
2815 will fail miserably.
2817 To get around this limitation, you can supply literal SQL to your
2818 C<select> attibute that contains the C<AS alias> text, eg:
2820 select => [\'myfield AS alias']
2826 =item Value: ($rel_name | \@rel_names | \%rel_names)
2830 Contains a list of relationships that should be joined for this query. For
2833 # Get CDs by Nine Inch Nails
2834 my $rs = $schema->resultset('CD')->search(
2835 { 'artist.name' => 'Nine Inch Nails' },
2836 { join => 'artist' }
2839 Can also contain a hash reference to refer to the other relation's relations.
2842 package MyApp::Schema::Track;
2843 use base qw/DBIx::Class/;
2844 __PACKAGE__->table('track');
2845 __PACKAGE__->add_columns(qw/trackid cd position title/);
2846 __PACKAGE__->set_primary_key('trackid');
2847 __PACKAGE__->belongs_to(cd => 'MyApp::Schema::CD');
2850 # In your application
2851 my $rs = $schema->resultset('Artist')->search(
2852 { 'track.title' => 'Teardrop' },
2854 join => { cd => 'track' },
2855 order_by => 'artist.name',
2859 You need to use the relationship (not the table) name in conditions,
2860 because they are aliased as such. The current table is aliased as "me", so
2861 you need to use me.column_name in order to avoid ambiguity. For example:
2863 # Get CDs from 1984 with a 'Foo' track
2864 my $rs = $schema->resultset('CD')->search(
2867 'tracks.name' => 'Foo'
2869 { join => 'tracks' }
2872 If the same join is supplied twice, it will be aliased to <rel>_2 (and
2873 similarly for a third time). For e.g.
2875 my $rs = $schema->resultset('Artist')->search({
2876 'cds.title' => 'Down to Earth',
2877 'cds_2.title' => 'Popular',
2879 join => [ qw/cds cds/ ],
2882 will return a set of all artists that have both a cd with title 'Down
2883 to Earth' and a cd with title 'Popular'.
2885 If you want to fetch related objects from other tables as well, see C<prefetch>
2888 For more help on using joins with search, see L<DBIx::Class::Manual::Joining>.
2894 =item Value: ($rel_name | \@rel_names | \%rel_names)
2898 Contains one or more relationships that should be fetched along with
2899 the main query (when they are accessed afterwards the data will
2900 already be available, without extra queries to the database). This is
2901 useful for when you know you will need the related objects, because it
2902 saves at least one query:
2904 my $rs = $schema->resultset('Tag')->search(
2913 The initial search results in SQL like the following:
2915 SELECT tag.*, cd.*, artist.* FROM tag
2916 JOIN cd ON tag.cd = cd.cdid
2917 JOIN artist ON cd.artist = artist.artistid
2919 L<DBIx::Class> has no need to go back to the database when we access the
2920 C<cd> or C<artist> relationships, which saves us two SQL statements in this
2923 Simple prefetches will be joined automatically, so there is no need
2924 for a C<join> attribute in the above search.
2926 C<prefetch> can be used with the following relationship types: C<belongs_to>,
2927 C<has_one> (or if you're using C<add_relationship>, any relationship declared
2928 with an accessor type of 'single' or 'filter'). A more complex example that
2929 prefetches an artists cds, the tracks on those cds, and the tags associted
2930 with that artist is given below (assuming many-to-many from artists to tags):
2932 my $rs = $schema->resultset('Artist')->search(
2936 { cds => 'tracks' },
2937 { artist_tags => 'tags' }
2943 B<NOTE:> If you specify a C<prefetch> attribute, the C<join> and C<select>
2944 attributes will be ignored.
2954 Makes the resultset paged and specifies the page to retrieve. Effectively
2955 identical to creating a non-pages resultset and then calling ->page($page)
2958 If L<rows> attribute is not specified it defualts to 10 rows per page.
2960 When you have a paged resultset, L</count> will only return the number
2961 of rows in the page. To get the total, use the L</pager> and call
2962 C<total_entries> on it.
2972 Specifes the maximum number of rows for direct retrieval or the number of
2973 rows per page if the page attribute or method is used.
2979 =item Value: $offset
2983 Specifies the (zero-based) row number for the first row to be returned, or the
2984 of the first row of the first page if paging is used.
2990 =item Value: \@columns
2994 A arrayref of columns to group by. Can include columns of joined tables.
2996 group_by => [qw/ column1 column2 ... /]
3002 =item Value: $condition
3006 HAVING is a select statement attribute that is applied between GROUP BY and
3007 ORDER BY. It is applied to the after the grouping calculations have been
3010 having => { 'count(employee)' => { '>=', 100 } }
3016 =item Value: (0 | 1)
3020 Set to 1 to group by all columns.
3026 Adds to the WHERE clause.
3028 # only return rows WHERE deleted IS NULL for all searches
3029 __PACKAGE__->resultset_attributes({ where => { deleted => undef } }); )
3031 Can be overridden by passing C<{ where => undef }> as an attribute
3038 Set to 1 to cache search results. This prevents extra SQL queries if you
3039 revisit rows in your ResultSet:
3041 my $resultset = $schema->resultset('Artist')->search( undef, { cache => 1 } );
3043 while( my $artist = $resultset->next ) {
3047 $rs->first; # without cache, this would issue a query
3049 By default, searches are not cached.
3051 For more examples of using these attributes, see
3052 L<DBIx::Class::Manual::Cookbook>.
3058 =item Value: \@from_clause
3062 The C<from> attribute gives you manual control over the C<FROM> clause of SQL
3063 statements generated by L<DBIx::Class>, allowing you to express custom C<JOIN>
3066 NOTE: Use this on your own risk. This allows you to shoot off your foot!
3068 C<join> will usually do what you need and it is strongly recommended that you
3069 avoid using C<from> unless you cannot achieve the desired result using C<join>.
3070 And we really do mean "cannot", not just tried and failed. Attempting to use
3071 this because you're having problems with C<join> is like trying to use x86
3072 ASM because you've got a syntax error in your C. Trust us on this.
3074 Now, if you're still really, really sure you need to use this (and if you're
3075 not 100% sure, ask the mailing list first), here's an explanation of how this
3078 The syntax is as follows -
3081 { <alias1> => <table1> },
3083 { <alias2> => <table2>, -join_type => 'inner|left|right' },
3084 [], # nested JOIN (optional)
3085 { <table1.column1> => <table2.column2>, ... (more conditions) },
3087 # More of the above [ ] may follow for additional joins
3094 ON <table1.column1> = <table2.column2>
3095 <more joins may follow>
3097 An easy way to follow the examples below is to remember the following:
3099 Anything inside "[]" is a JOIN
3100 Anything inside "{}" is a condition for the enclosing JOIN
3102 The following examples utilize a "person" table in a family tree application.
3103 In order to express parent->child relationships, this table is self-joined:
3105 # Person->belongs_to('father' => 'Person');
3106 # Person->belongs_to('mother' => 'Person');
3108 C<from> can be used to nest joins. Here we return all children with a father,
3109 then search against all mothers of those children:
3111 $rs = $schema->resultset('Person')->search(
3114 alias => 'mother', # alias columns in accordance with "from"
3116 { mother => 'person' },
3119 { child => 'person' },
3121 { father => 'person' },
3122 { 'father.person_id' => 'child.father_id' }
3125 { 'mother.person_id' => 'child.mother_id' }
3132 # SELECT mother.* FROM person mother
3135 # JOIN person father
3136 # ON ( father.person_id = child.father_id )
3138 # ON ( mother.person_id = child.mother_id )
3140 The type of any join can be controlled manually. To search against only people
3141 with a father in the person table, we could explicitly use C<INNER JOIN>:
3143 $rs = $schema->resultset('Person')->search(
3146 alias => 'child', # alias columns in accordance with "from"
3148 { child => 'person' },
3150 { father => 'person', -join_type => 'inner' },
3151 { 'father.id' => 'child.father_id' }
3158 # SELECT child.* FROM person child
3159 # INNER JOIN person father ON child.father_id = father.id
3161 If you need to express really complex joins or you need a subselect, you
3162 can supply literal SQL to C<from> via a scalar reference. In this case
3163 the contents of the scalar will replace the table name asscoiated with the
3166 WARNING: This technique might very well not work as expected on chained
3167 searches - you have been warned.
3169 # Assuming the Event resultsource is defined as:
3171 MySchema::Event->add_columns (
3174 is_auto_increment => 1,
3183 MySchema::Event->set_primary_key ('sequence');
3185 # This will get back the latest event for every location. The column
3186 # selector is still provided by DBIC, all we do is add a JOIN/WHERE
3187 # combo to limit the resultset
3189 $rs = $schema->resultset('Event');
3190 $table = $rs->result_source->name;
3191 $latest = $rs->search (
3194 (SELECT e1.* FROM $table e1
3196 ON e1.location = e2.location
3197 AND e1.sequence < e2.sequence
3198 WHERE e2.sequence is NULL
3203 # Equivalent SQL (with the DBIC chunks added):
3205 SELECT me.sequence, me.location, me.type FROM
3206 (SELECT e1.* FROM events e1
3208 ON e1.location = e2.location
3209 AND e1.sequence < e2.sequence
3210 WHERE e2.sequence is NULL
3217 =item Value: ( 'update' | 'shared' )
3221 Set to 'update' for a SELECT ... FOR UPDATE or 'shared' for a SELECT