Commit | Line | Data |
89c0a5a2 |
1 | package DBIx::Class::ResultSet; |
2 | |
3 | use strict; |
4 | use warnings; |
5 | use overload |
ebaefbc2 |
6 | '0+' => \&count, |
a910dc57 |
7 | 'bool' => sub { 1; }, |
89c0a5a2 |
8 | fallback => 1; |
3c5b25c5 |
9 | use Data::Page; |
ea20d0fd |
10 | use Storable; |
89c0a5a2 |
11 | |
701da8c4 |
12 | use base qw/DBIx::Class/; |
13 | __PACKAGE__->load_components(qw/AccessorGroup/); |
a50bcd52 |
14 | __PACKAGE__->mk_group_accessors('simple' => qw/result_source result_class/); |
701da8c4 |
15 | |
ee38fa40 |
16 | =head1 NAME |
17 | |
bfab575a |
18 | DBIx::Class::ResultSet - Responsible for fetching and creating resultset. |
ee38fa40 |
19 | |
bfab575a |
20 | =head1 SYNOPSIS |
ee38fa40 |
21 | |
a33df5d4 |
22 | my $rs = $schema->resultset('User')->search(registered => 1); |
23 | my @rows = $schema->resultset('Foo')->search(bar => 'baz'); |
ee38fa40 |
24 | |
25 | =head1 DESCRIPTION |
26 | |
bfab575a |
27 | The resultset is also known as an iterator. It is responsible for handling |
a33df5d4 |
28 | queries that may return an arbitrary number of rows, e.g. via L</search> |
bfab575a |
29 | or a C<has_many> relationship. |
ee38fa40 |
30 | |
a33df5d4 |
31 | In the examples below, the following table classes are used: |
32 | |
33 | package MyApp::Schema::Artist; |
34 | use base qw/DBIx::Class/; |
f4409169 |
35 | __PACKAGE__->load_components(qw/Core/); |
a33df5d4 |
36 | __PACKAGE__->table('artist'); |
37 | __PACKAGE__->add_columns(qw/artistid name/); |
38 | __PACKAGE__->set_primary_key('artistid'); |
39 | __PACKAGE__->has_many(cds => 'MyApp::Schema::CD'); |
40 | 1; |
41 | |
42 | package MyApp::Schema::CD; |
43 | use base qw/DBIx::Class/; |
f4409169 |
44 | __PACKAGE__->load_components(qw/Core/); |
45 | __PACKAGE__->table('cd'); |
a33df5d4 |
46 | __PACKAGE__->add_columns(qw/cdid artist title year/); |
47 | __PACKAGE__->set_primary_key('cdid'); |
48 | __PACKAGE__->belongs_to(artist => 'MyApp::Schema::Artist'); |
49 | 1; |
50 | |
ee38fa40 |
51 | =head1 METHODS |
52 | |
87c4e602 |
53 | =head2 new |
54 | |
55 | =head3 Arguments: ($source, \%$attrs) |
ee38fa40 |
56 | |
a33df5d4 |
57 | The resultset constructor. Takes a source object (usually a |
181a28f4 |
58 | L<DBIx::Class::ResultSourceProxy::Table>) and an attribute hash (see L</ATTRIBUTES> |
a33df5d4 |
59 | below). Does not perform any queries -- these are executed as needed by the |
60 | other methods. |
61 | |
62 | Generally you won't need to construct a resultset manually. You'll |
63 | automatically get one from e.g. a L</search> called in scalar context: |
64 | |
65 | my $rs = $schema->resultset('CD')->search({ title => '100th Window' }); |
ee38fa40 |
66 | |
67 | =cut |
68 | |
89c0a5a2 |
69 | sub new { |
fea3d045 |
70 | my $class = shift; |
f9db5527 |
71 | return $class->new_result(@_) if ref $class; |
fea3d045 |
72 | my ($source, $attrs) = @_; |
b98e75f6 |
73 | #use Data::Dumper; warn Dumper($attrs); |
ea20d0fd |
74 | $attrs = Storable::dclone($attrs || {}); # { %{ $attrs || {} } }; |
c7ce65e6 |
75 | my %seen; |
6aeb9185 |
76 | my $alias = ($attrs->{alias} ||= 'me'); |
a9433341 |
77 | if ($attrs->{cols} || !$attrs->{select}) { |
78 | delete $attrs->{as} if $attrs->{cols}; |
976f3686 |
79 | my @cols = ($attrs->{cols} |
80 | ? @{delete $attrs->{cols}} |
a9433341 |
81 | : $source->columns); |
6aeb9185 |
82 | $attrs->{select} = [ map { m/\./ ? $_ : "${alias}.$_" } @cols ]; |
976f3686 |
83 | } |
6aeb9185 |
84 | $attrs->{as} ||= [ map { m/^$alias\.(.*)$/ ? $1 : $_ } @{$attrs->{select}} ]; |
5ac6a044 |
85 | if (my $include = delete $attrs->{include_columns}) { |
86 | push(@{$attrs->{select}}, @$include); |
87 | push(@{$attrs->{as}}, map { m/([^\.]+)$/; $1; } @$include); |
88 | } |
976f3686 |
89 | #use Data::Dumper; warn Dumper(@{$attrs}{qw/select as/}); |
fea3d045 |
90 | $attrs->{from} ||= [ { $alias => $source->from } ]; |
8fab5eef |
91 | $attrs->{seen_join} ||= {}; |
b52e9bf8 |
92 | if (my $join = delete $attrs->{join}) { |
93 | foreach my $j (ref $join eq 'ARRAY' |
94 | ? (@{$join}) : ($join)) { |
c7ce65e6 |
95 | if (ref $j eq 'HASH') { |
96 | $seen{$_} = 1 foreach keys %$j; |
97 | } else { |
98 | $seen{$j} = 1; |
99 | } |
100 | } |
8fab5eef |
101 | push(@{$attrs->{from}}, $source->resolve_join($join, $attrs->{alias}, $attrs->{seen_join})); |
c7ce65e6 |
102 | } |
54540863 |
103 | $attrs->{group_by} ||= $attrs->{select} if delete $attrs->{distinct}; |
b3e8ac9b |
104 | |
a86b1efe |
105 | $attrs->{order_by} = [ $attrs->{order_by} ] |
106 | if $attrs->{order_by} && !ref($attrs->{order_by}); |
107 | $attrs->{order_by} ||= []; |
108 | |
0f66a01b |
109 | my $collapse = {}; |
110 | |
b3e8ac9b |
111 | if (my $prefetch = delete $attrs->{prefetch}) { |
0f66a01b |
112 | my @pre_order; |
b3e8ac9b |
113 | foreach my $p (ref $prefetch eq 'ARRAY' |
114 | ? (@{$prefetch}) : ($prefetch)) { |
115 | if( ref $p eq 'HASH' ) { |
116 | foreach my $key (keys %$p) { |
117 | push(@{$attrs->{from}}, $source->resolve_join($p, $attrs->{alias})) |
118 | unless $seen{$key}; |
119 | } |
120 | } |
121 | else { |
122 | push(@{$attrs->{from}}, $source->resolve_join($p, $attrs->{alias})) |
123 | unless $seen{$p}; |
124 | } |
a86b1efe |
125 | my @prefetch = $source->resolve_prefetch( |
0f66a01b |
126 | $p, $attrs->{alias}, {}, \@pre_order, $collapse); |
b3e8ac9b |
127 | #die Dumper \@cols; |
489709af |
128 | push(@{$attrs->{select}}, map { $_->[0] } @prefetch); |
129 | push(@{$attrs->{as}}, map { $_->[1] } @prefetch); |
b3e8ac9b |
130 | } |
0f66a01b |
131 | push(@{$attrs->{order_by}}, @pre_order); |
fef5d100 |
132 | } |
b3e8ac9b |
133 | |
6aeb9185 |
134 | if ($attrs->{page}) { |
135 | $attrs->{rows} ||= 10; |
136 | $attrs->{offset} ||= 0; |
137 | $attrs->{offset} += ($attrs->{rows} * ($attrs->{page} - 1)); |
138 | } |
0f66a01b |
139 | |
140 | #if (keys %{$collapse}) { |
141 | # use Data::Dumper; warn Dumper($collapse); |
142 | #} |
143 | |
89c0a5a2 |
144 | my $new = { |
701da8c4 |
145 | result_source => $source, |
a50bcd52 |
146 | result_class => $attrs->{result_class} || $source->result_class, |
89c0a5a2 |
147 | cond => $attrs->{where}, |
0a3c5b43 |
148 | from => $attrs->{from}, |
0f66a01b |
149 | collapse => $collapse, |
3c5b25c5 |
150 | count => undef, |
93b004d3 |
151 | page => delete $attrs->{page}, |
3c5b25c5 |
152 | pager => undef, |
89c0a5a2 |
153 | attrs => $attrs }; |
2f5911b2 |
154 | bless ($new, $class); |
9229f20a |
155 | return $new; |
89c0a5a2 |
156 | } |
157 | |
bfab575a |
158 | =head2 search |
0a3c5b43 |
159 | |
87f0da6a |
160 | my @obj = $rs->search({ foo => 3 }); # "... WHERE foo = 3" |
161 | my $new_rs = $rs->search({ foo => 3 }); |
162 | |
6009260a |
163 | If you need to pass in additional attributes but no additional condition, |
a33df5d4 |
164 | call it as C<search({}, \%attrs);>. |
87f0da6a |
165 | |
a33df5d4 |
166 | # "SELECT foo, bar FROM $class_table" |
167 | my @all = $class->search({}, { cols => [qw/foo bar/] }); |
0a3c5b43 |
168 | |
169 | =cut |
170 | |
171 | sub search { |
172 | my $self = shift; |
173 | |
ff7bb7a1 |
174 | my $rs; |
175 | if( @_ ) { |
176 | |
177 | my $attrs = { %{$self->{attrs}} }; |
8839560b |
178 | my $having = delete $attrs->{having}; |
ff7bb7a1 |
179 | if (@_ > 1 && ref $_[$#_] eq 'HASH') { |
180 | $attrs = { %$attrs, %{ pop(@_) } }; |
181 | } |
6009260a |
182 | |
3e0e9e27 |
183 | my $where = (@_ |
184 | ? ((@_ == 1 || ref $_[0] eq "HASH") |
185 | ? shift |
186 | : ((@_ % 2) |
187 | ? $self->throw_exception( |
188 | "Odd number of arguments to search") |
189 | : {@_})) |
190 | : undef()); |
ff7bb7a1 |
191 | if (defined $where) { |
192 | $where = (defined $attrs->{where} |
ad3d2d7c |
193 | ? { '-and' => |
194 | [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ } |
195 | $where, $attrs->{where} ] } |
0a3c5b43 |
196 | : $where); |
ff7bb7a1 |
197 | $attrs->{where} = $where; |
198 | } |
0a3c5b43 |
199 | |
8839560b |
200 | if (defined $having) { |
201 | $having = (defined $attrs->{having} |
202 | ? { '-and' => |
203 | [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ } |
204 | $having, $attrs->{having} ] } |
205 | : $having); |
206 | $attrs->{having} = $having; |
207 | } |
208 | |
ff7bb7a1 |
209 | $rs = (ref $self)->new($self->result_source, $attrs); |
210 | } |
211 | else { |
212 | $rs = $self; |
213 | $rs->reset(); |
214 | } |
0a3c5b43 |
215 | return (wantarray ? $rs->all : $rs); |
216 | } |
217 | |
87f0da6a |
218 | =head2 search_literal |
219 | |
6009260a |
220 | my @obj = $rs->search_literal($literal_where_cond, @bind); |
221 | my $new_rs = $rs->search_literal($literal_where_cond, @bind); |
222 | |
223 | Pass a literal chunk of SQL to be added to the conditional part of the |
87f0da6a |
224 | resultset. |
6009260a |
225 | |
bfab575a |
226 | =cut |
fd9f5466 |
227 | |
6009260a |
228 | sub search_literal { |
229 | my ($self, $cond, @vals) = @_; |
230 | my $attrs = (ref $vals[$#vals] eq 'HASH' ? { %{ pop(@vals) } } : {}); |
231 | $attrs->{bind} = [ @{$self->{attrs}{bind}||[]}, @vals ]; |
232 | return $self->search(\$cond, $attrs); |
233 | } |
0a3c5b43 |
234 | |
87c4e602 |
235 | =head2 find |
236 | |
237 | =head3 Arguments: (@colvalues) | (\%cols, \%attrs?) |
87f0da6a |
238 | |
239 | Finds a row based on its primary key or unique constraint. For example: |
240 | |
87f0da6a |
241 | my $cd = $schema->resultset('CD')->find(5); |
242 | |
243 | Also takes an optional C<key> attribute, to search by a specific key or unique |
244 | constraint. For example: |
245 | |
fd9f5466 |
246 | my $cd = $schema->resultset('CD')->find( |
87f0da6a |
247 | { |
248 | artist => 'Massive Attack', |
249 | title => 'Mezzanine', |
250 | }, |
251 | { key => 'artist_title' } |
252 | ); |
253 | |
a33df5d4 |
254 | See also L</find_or_create> and L</update_or_create>. |
255 | |
87f0da6a |
256 | =cut |
716b3d29 |
257 | |
258 | sub find { |
259 | my ($self, @vals) = @_; |
260 | my $attrs = (@vals > 1 && ref $vals[$#vals] eq 'HASH' ? pop(@vals) : {}); |
87f0da6a |
261 | |
701da8c4 |
262 | my @cols = $self->result_source->primary_columns; |
87f0da6a |
263 | if (exists $attrs->{key}) { |
701da8c4 |
264 | my %uniq = $self->result_source->unique_constraints; |
87f0da6a |
265 | $self->( "Unknown key " . $attrs->{key} . " on " . $self->name ) |
266 | unless exists $uniq{$attrs->{key}}; |
267 | @cols = @{ $uniq{$attrs->{key}} }; |
268 | } |
269 | #use Data::Dumper; warn Dumper($attrs, @vals, @cols); |
701da8c4 |
270 | $self->throw_exception( "Can't find unless a primary key or unique constraint is defined" ) |
87f0da6a |
271 | unless @cols; |
272 | |
716b3d29 |
273 | my $query; |
274 | if (ref $vals[0] eq 'HASH') { |
01bc091e |
275 | $query = { %{$vals[0]} }; |
87f0da6a |
276 | } elsif (@cols == @vals) { |
716b3d29 |
277 | $query = {}; |
87f0da6a |
278 | @{$query}{@cols} = @vals; |
716b3d29 |
279 | } else { |
280 | $query = {@vals}; |
281 | } |
01bc091e |
282 | foreach (keys %$query) { |
283 | next if m/\./; |
284 | $query->{$self->{attrs}{alias}.'.'.$_} = delete $query->{$_}; |
285 | } |
716b3d29 |
286 | #warn Dumper($query); |
a04ab285 |
287 | return (keys %$attrs |
288 | ? $self->search($query,$attrs)->single |
289 | : $self->single($query)); |
716b3d29 |
290 | } |
291 | |
b52e9bf8 |
292 | =head2 search_related |
293 | |
294 | $rs->search_related('relname', $cond?, $attrs?); |
295 | |
a33df5d4 |
296 | Search the specified relationship. Optionally specify a condition for matching |
297 | records. |
298 | |
b52e9bf8 |
299 | =cut |
300 | |
6aeb9185 |
301 | sub search_related { |
64acc2bc |
302 | return shift->related_resultset(shift)->search(@_); |
6aeb9185 |
303 | } |
b52e9bf8 |
304 | |
bfab575a |
305 | =head2 cursor |
ee38fa40 |
306 | |
bfab575a |
307 | Returns a storage-driven cursor to the given resultset. |
ee38fa40 |
308 | |
309 | =cut |
310 | |
73f58123 |
311 | sub cursor { |
312 | my ($self) = @_; |
701da8c4 |
313 | my ($attrs) = $self->{attrs}; |
6aeb9185 |
314 | $attrs = { %$attrs }; |
73f58123 |
315 | return $self->{cursor} |
701da8c4 |
316 | ||= $self->result_source->storage->select($self->{from}, $attrs->{select}, |
73f58123 |
317 | $attrs->{where},$attrs); |
318 | } |
319 | |
a04ab285 |
320 | =head2 single |
321 | |
322 | Inflates the first result without creating a cursor |
323 | |
324 | =cut |
325 | |
326 | sub single { |
327 | my ($self, $extra) = @_; |
328 | my ($attrs) = $self->{attrs}; |
329 | $attrs = { %$attrs }; |
330 | if ($extra) { |
331 | if (defined $attrs->{where}) { |
332 | $attrs->{where} = { |
333 | '-and' |
334 | => [ map { ref $_ eq 'ARRAY' ? [ -or => $_ ] : $_ } |
335 | delete $attrs->{where}, $extra ] |
336 | }; |
337 | } else { |
338 | $attrs->{where} = $extra; |
339 | } |
340 | } |
341 | my @data = $self->result_source->storage->select_single( |
342 | $self->{from}, $attrs->{select}, |
343 | $attrs->{where},$attrs); |
344 | return (@data ? $self->_construct_object(@data) : ()); |
345 | } |
346 | |
347 | |
87f0da6a |
348 | =head2 search_like |
349 | |
a33df5d4 |
350 | Perform a search, but use C<LIKE> instead of equality as the condition. Note |
351 | that this is simply a convenience method; you most likely want to use |
352 | L</search> with specific operators. |
353 | |
354 | For more information, see L<DBIx::Class::Manual::Cookbook>. |
87f0da6a |
355 | |
356 | =cut |
58a4bd18 |
357 | |
358 | sub search_like { |
359 | my $class = shift; |
360 | my $attrs = { }; |
361 | if (@_ > 1 && ref $_[$#_] eq 'HASH') { |
362 | $attrs = pop(@_); |
363 | } |
364 | my $query = ref $_[0] eq "HASH" ? { %{shift()} }: {@_}; |
365 | $query->{$_} = { 'like' => $query->{$_} } for keys %$query; |
366 | return $class->search($query, { %$attrs }); |
367 | } |
368 | |
87c4e602 |
369 | =head2 slice |
370 | |
371 | =head3 Arguments: ($first, $last) |
ee38fa40 |
372 | |
bfab575a |
373 | Returns a subset of elements from the resultset. |
ee38fa40 |
374 | |
375 | =cut |
376 | |
89c0a5a2 |
377 | sub slice { |
378 | my ($self, $min, $max) = @_; |
379 | my $attrs = { %{ $self->{attrs} || {} } }; |
6aeb9185 |
380 | $attrs->{offset} ||= 0; |
381 | $attrs->{offset} += $min; |
89c0a5a2 |
382 | $attrs->{rows} = ($max ? ($max - $min + 1) : 1); |
701da8c4 |
383 | my $slice = (ref $self)->new($self->result_source, $attrs); |
89c0a5a2 |
384 | return (wantarray ? $slice->all : $slice); |
385 | } |
386 | |
87f0da6a |
387 | =head2 next |
ee38fa40 |
388 | |
a33df5d4 |
389 | Returns the next element in the resultset (C<undef> is there is none). |
390 | |
391 | Can be used to efficiently iterate over records in the resultset: |
392 | |
393 | my $rs = $schema->resultset('CD')->search({}); |
394 | while (my $cd = $rs->next) { |
395 | print $cd->title; |
396 | } |
ee38fa40 |
397 | |
398 | =cut |
399 | |
89c0a5a2 |
400 | sub next { |
401 | my ($self) = @_; |
3e0e9e27 |
402 | my $cache; |
403 | if( @{$cache = $self->{all_cache} || []}) { |
64acc2bc |
404 | $self->{all_cache_position} ||= 0; |
405 | my $obj = $cache->[$self->{all_cache_position}]; |
406 | $self->{all_cache_position}++; |
407 | return $obj; |
408 | } |
3e0e9e27 |
409 | if ($self->{attrs}{cache}) { |
0f66a01b |
410 | $self->{all_cache_position} = 1; |
3e0e9e27 |
411 | return ($self->all)[0]; |
412 | } |
0f66a01b |
413 | my @row = (exists $self->{stashed_row} |
414 | ? @{delete $self->{stashed_row}} |
415 | : $self->cursor->next); |
a953d8d9 |
416 | # warn Dumper(\@row); use Data::Dumper; |
89c0a5a2 |
417 | return unless (@row); |
c7ce65e6 |
418 | return $self->_construct_object(@row); |
419 | } |
420 | |
421 | sub _construct_object { |
422 | my ($self, @row) = @_; |
b3e8ac9b |
423 | my @as = @{ $self->{attrs}{as} }; |
0f66a01b |
424 | |
425 | my $info = $self->_collapse_result(\@as, \@row); |
426 | |
b3e8ac9b |
427 | #use Data::Dumper; warn Dumper(\@as, $info); |
a50bcd52 |
428 | my $new = $self->result_class->inflate_result($self->result_source, @$info); |
0f66a01b |
429 | |
33ce49d6 |
430 | $new = $self->{attrs}{record_filter}->($new) |
431 | if exists $self->{attrs}{record_filter}; |
f9cc31dd |
432 | |
33ce49d6 |
433 | return $new; |
89c0a5a2 |
434 | } |
435 | |
0f66a01b |
436 | sub _collapse_result { |
437 | my ($self, $as, $row, $prefix) = @_; |
438 | |
439 | my %const; |
440 | |
441 | my @copy = @$row; |
442 | foreach my $as (@$as) { |
443 | if (defined $prefix && !($as =~ s/\Q${prefix}\E\.//)) { |
444 | shift @copy; |
445 | next; |
446 | } |
447 | $as =~ /^(?:(.*)\.)?([^\.]+)$/; |
448 | $const{$1||''}{$2} = shift @copy; |
449 | } |
450 | |
451 | #warn "@cols -> @row"; |
452 | my $info = [ {}, {} ]; |
453 | foreach my $key (keys %const) { |
454 | if (length $key) { |
455 | my $target = $info; |
456 | my @parts = split(/\./, $key); |
457 | foreach my $p (@parts) { |
458 | $target = $target->[1]->{$p} ||= []; |
459 | } |
460 | $target->[0] = $const{$key}; |
461 | } else { |
462 | $info->[0] = $const{$key}; |
463 | } |
464 | } |
465 | |
466 | if (!defined($prefix) && keys %{$self->{collapse}}) { |
467 | my ($c) = sort { length $a <=> length $b } keys %{$self->{collapse}}; |
468 | #warn "Collapsing ${c}"; |
469 | my $target = $info; |
470 | #warn Data::Dumper::Dumper($target); |
471 | foreach my $p (split(/\./, $c)) { |
472 | $target = $target->[1]->{$p}; |
473 | } |
474 | my @co_key = @{$self->{collapse}{$c}}; |
475 | my %co_check = map { ($_, $target->[0]->{$_}); } @co_key; |
476 | my $tree = $self->_collapse_result($as, $row, $c); |
477 | #warn Data::Dumper::Dumper($target); |
478 | my (@final, @raw); |
479 | while ( !(grep { $co_check{$_} ne $tree->[0]->{$_} } @co_key) ) { |
480 | push(@final, $tree); |
481 | last unless (@raw = $self->cursor->next); |
482 | $row = $self->{stashed_row} = \@raw; |
483 | $tree = $self->_collapse_result($as, $row, $c); |
484 | } |
485 | @{$target} = @final; |
486 | #warn Data::Dumper::Dumper($target); |
487 | } |
488 | |
489 | #warn Dumper($info); |
490 | |
491 | return $info; |
492 | } |
493 | |
87c4e602 |
494 | =head2 result_source |
701da8c4 |
495 | |
496 | Returns a reference to the result source for this recordset. |
497 | |
498 | =cut |
499 | |
500 | |
bfab575a |
501 | =head2 count |
ee38fa40 |
502 | |
bfab575a |
503 | Performs an SQL C<COUNT> with the same query as the resultset was built |
6009260a |
504 | with to find the number of elements. If passed arguments, does a search |
505 | on the resultset and counts the results of that. |
ee38fa40 |
506 | |
bda4c2b8 |
507 | Note: When using C<count> with C<group_by>, L<DBIX::Class> emulates C<GROUP BY> |
508 | using C<COUNT( DISTINCT( columns ) )>. Some databases (notably SQLite) do |
509 | not support C<DISTINCT> with multiple columns. If you are using such a |
510 | database, you should only use columns from the main table in your C<group_by> |
511 | clause. |
512 | |
ee38fa40 |
513 | =cut |
514 | |
89c0a5a2 |
515 | sub count { |
6009260a |
516 | my $self = shift; |
517 | return $self->search(@_)->count if @_ && defined $_[0]; |
6aeb9185 |
518 | unless (defined $self->{count}) { |
64acc2bc |
519 | return scalar @{ $self->get_cache } |
520 | if @{ $self->get_cache }; |
15c382be |
521 | my $group_by; |
522 | my $select = { 'count' => '*' }; |
8839560b |
523 | my $attrs = { %{ $self->{attrs} } }; |
524 | if( $group_by = delete $attrs->{group_by} ) { |
525 | delete $attrs->{having}; |
dec2517f |
526 | my @distinct = (ref $group_by ? @$group_by : ($group_by)); |
15c382be |
527 | # todo: try CONCAT for multi-column pk |
528 | my @pk = $self->result_source->primary_columns; |
529 | if( scalar(@pk) == 1 ) { |
530 | my $pk = shift(@pk); |
8839560b |
531 | my $alias = $attrs->{alias}; |
15c382be |
532 | my $re = qr/^($alias\.)?$pk$/; |
d0f1e63f |
533 | foreach my $column ( @distinct) { |
15c382be |
534 | if( $column =~ $re ) { |
535 | @distinct = ( $column ); |
536 | last; |
537 | } |
538 | } |
539 | } |
540 | |
541 | $select = { count => { 'distinct' => \@distinct } }; |
542 | #use Data::Dumper; die Dumper $select; |
543 | } |
544 | |
8839560b |
545 | $attrs->{select} = $select; |
546 | $attrs->{as} = [ 'count' ]; |
ea20d0fd |
547 | # offset, order by and page are not needed to count. record_filter is cdbi |
548 | delete $attrs->{$_} for qw/rows offset order_by page pager record_filter/; |
3c5b25c5 |
549 | |
701da8c4 |
550 | ($self->{count}) = (ref $self)->new($self->result_source, $attrs)->cursor->next; |
3c5b25c5 |
551 | } |
552 | return 0 unless $self->{count}; |
6aeb9185 |
553 | my $count = $self->{count}; |
554 | $count -= $self->{attrs}{offset} if $self->{attrs}{offset}; |
555 | $count = $self->{attrs}{rows} if |
556 | ($self->{attrs}{rows} && $self->{attrs}{rows} < $count); |
557 | return $count; |
89c0a5a2 |
558 | } |
559 | |
bfab575a |
560 | =head2 count_literal |
6009260a |
561 | |
a33df5d4 |
562 | Calls L</search_literal> with the passed arguments, then L</count>. |
6009260a |
563 | |
564 | =cut |
565 | |
566 | sub count_literal { shift->search_literal(@_)->count; } |
567 | |
bfab575a |
568 | =head2 all |
ee38fa40 |
569 | |
bfab575a |
570 | Returns all elements in the resultset. Called implictly if the resultset |
571 | is returned in list context. |
ee38fa40 |
572 | |
573 | =cut |
574 | |
89c0a5a2 |
575 | sub all { |
576 | my ($self) = @_; |
64acc2bc |
577 | return @{ $self->get_cache } |
578 | if @{ $self->get_cache }; |
579 | if( $self->{attrs}->{cache} ) { |
580 | my @obj = map { $self->_construct_object(@$_); } |
581 | $self->cursor->all; |
582 | $self->set_cache( \@obj ); |
3e0e9e27 |
583 | return @obj; |
64acc2bc |
584 | } |
c7ce65e6 |
585 | return map { $self->_construct_object(@$_); } |
73f58123 |
586 | $self->cursor->all; |
89c0a5a2 |
587 | } |
588 | |
bfab575a |
589 | =head2 reset |
ee38fa40 |
590 | |
bfab575a |
591 | Resets the resultset's cursor, so you can iterate through the elements again. |
ee38fa40 |
592 | |
593 | =cut |
594 | |
89c0a5a2 |
595 | sub reset { |
596 | my ($self) = @_; |
64acc2bc |
597 | $self->{all_cache_position} = 0; |
73f58123 |
598 | $self->cursor->reset; |
89c0a5a2 |
599 | return $self; |
600 | } |
601 | |
bfab575a |
602 | =head2 first |
ee38fa40 |
603 | |
bfab575a |
604 | Resets the resultset and returns the first element. |
ee38fa40 |
605 | |
606 | =cut |
607 | |
89c0a5a2 |
608 | sub first { |
609 | return $_[0]->reset->next; |
610 | } |
611 | |
87c4e602 |
612 | =head2 update |
613 | |
614 | =head3 Arguments: (\%values) |
c01ab172 |
615 | |
a33df5d4 |
616 | Sets the specified columns in the resultset to the supplied values. |
c01ab172 |
617 | |
618 | =cut |
619 | |
620 | sub update { |
621 | my ($self, $values) = @_; |
701da8c4 |
622 | $self->throw_exception("Values for update must be a hash") unless ref $values eq 'HASH'; |
623 | return $self->result_source->storage->update( |
624 | $self->result_source->from, $values, $self->{cond}); |
c01ab172 |
625 | } |
626 | |
87c4e602 |
627 | =head2 update_all |
628 | |
629 | =head3 Arguments: (\%values) |
c01ab172 |
630 | |
a33df5d4 |
631 | Fetches all objects and updates them one at a time. Note that C<update_all> |
632 | will run cascade triggers while L</update> will not. |
c01ab172 |
633 | |
634 | =cut |
635 | |
636 | sub update_all { |
637 | my ($self, $values) = @_; |
701da8c4 |
638 | $self->throw_exception("Values for update must be a hash") unless ref $values eq 'HASH'; |
c01ab172 |
639 | foreach my $obj ($self->all) { |
640 | $obj->set_columns($values)->update; |
641 | } |
642 | return 1; |
643 | } |
644 | |
bfab575a |
645 | =head2 delete |
ee38fa40 |
646 | |
c01ab172 |
647 | Deletes the contents of the resultset from its result source. |
ee38fa40 |
648 | |
649 | =cut |
650 | |
28927b50 |
651 | sub delete { |
89c0a5a2 |
652 | my ($self) = @_; |
ca4b5ab7 |
653 | my $del = {}; |
654 | $self->throw_exception("Can't delete on resultset with condition unless hash or array") |
655 | unless (ref($self->{cond}) eq 'HASH' || ref($self->{cond}) eq 'ARRAY'); |
656 | if (ref $self->{cond} eq 'ARRAY') { |
657 | $del = [ map { my %hash; |
658 | foreach my $key (keys %{$_}) { |
659 | $key =~ /([^\.]+)$/; |
660 | $hash{$1} = $_->{$key}; |
661 | }; \%hash; } @{$self->{cond}} ]; |
662 | } elsif ((keys %{$self->{cond}})[0] eq '-and') { |
663 | $del->{-and} = [ map { my %hash; |
664 | foreach my $key (keys %{$_}) { |
665 | $key =~ /([^\.]+)$/; |
666 | $hash{$1} = $_->{$key}; |
667 | }; \%hash; } @{$self->{cond}{-and}} ]; |
668 | } else { |
669 | foreach my $key (keys %{$self->{cond}}) { |
670 | $key =~ /([^\.]+)$/; |
671 | $del->{$1} = $self->{cond}{$key}; |
672 | } |
673 | } |
674 | $self->result_source->storage->delete($self->result_source->from, $del); |
89c0a5a2 |
675 | return 1; |
676 | } |
677 | |
c01ab172 |
678 | =head2 delete_all |
679 | |
a33df5d4 |
680 | Fetches all objects and deletes them one at a time. Note that C<delete_all> |
681 | will run cascade triggers while L</delete> will not. |
c01ab172 |
682 | |
683 | =cut |
684 | |
685 | sub delete_all { |
686 | my ($self) = @_; |
687 | $_->delete for $self->all; |
688 | return 1; |
689 | } |
28927b50 |
690 | |
bfab575a |
691 | =head2 pager |
ee38fa40 |
692 | |
693 | Returns a L<Data::Page> object for the current resultset. Only makes |
a33df5d4 |
694 | sense for queries with a C<page> attribute. |
ee38fa40 |
695 | |
696 | =cut |
697 | |
3c5b25c5 |
698 | sub pager { |
699 | my ($self) = @_; |
700 | my $attrs = $self->{attrs}; |
701da8c4 |
701 | $self->throw_exception("Can't create pager for non-paged rs") unless $self->{page}; |
6aeb9185 |
702 | $attrs->{rows} ||= 10; |
703 | $self->count; |
704 | return $self->{pager} ||= Data::Page->new( |
93b004d3 |
705 | $self->{count}, $attrs->{rows}, $self->{page}); |
3c5b25c5 |
706 | } |
707 | |
87c4e602 |
708 | =head2 page |
709 | |
710 | =head3 Arguments: ($page_num) |
ee38fa40 |
711 | |
bfab575a |
712 | Returns a new resultset for the specified page. |
ee38fa40 |
713 | |
714 | =cut |
715 | |
3c5b25c5 |
716 | sub page { |
717 | my ($self, $page) = @_; |
6aeb9185 |
718 | my $attrs = { %{$self->{attrs}} }; |
3c5b25c5 |
719 | $attrs->{page} = $page; |
701da8c4 |
720 | return (ref $self)->new($self->result_source, $attrs); |
fea3d045 |
721 | } |
722 | |
87c4e602 |
723 | =head2 new_result |
724 | |
725 | =head3 Arguments: (\%vals) |
fea3d045 |
726 | |
87f0da6a |
727 | Creates a result in the resultset's result class. |
fea3d045 |
728 | |
729 | =cut |
730 | |
731 | sub new_result { |
732 | my ($self, $values) = @_; |
701da8c4 |
733 | $self->throw_exception( "new_result needs a hash" ) |
fea3d045 |
734 | unless (ref $values eq 'HASH'); |
701da8c4 |
735 | $self->throw_exception( "Can't abstract implicit construct, condition not a hash" ) |
fea3d045 |
736 | if ($self->{cond} && !(ref $self->{cond} eq 'HASH')); |
737 | my %new = %$values; |
738 | my $alias = $self->{attrs}{alias}; |
739 | foreach my $key (keys %{$self->{cond}||{}}) { |
740 | $new{$1} = $self->{cond}{$key} if ($key =~ m/^(?:$alias\.)?([^\.]+)$/); |
741 | } |
a50bcd52 |
742 | my $obj = $self->result_class->new(\%new); |
701da8c4 |
743 | $obj->result_source($self->result_source) if $obj->can('result_source'); |
097d3227 |
744 | $obj; |
fea3d045 |
745 | } |
746 | |
87c4e602 |
747 | =head2 create |
748 | |
749 | =head3 Arguments: (\%vals) |
fea3d045 |
750 | |
87f0da6a |
751 | Inserts a record into the resultset and returns the object. |
fea3d045 |
752 | |
a33df5d4 |
753 | Effectively a shortcut for C<< ->new_result(\%vals)->insert >>. |
fea3d045 |
754 | |
755 | =cut |
756 | |
757 | sub create { |
758 | my ($self, $attrs) = @_; |
701da8c4 |
759 | $self->throw_exception( "create needs a hashref" ) unless ref $attrs eq 'HASH'; |
fea3d045 |
760 | return $self->new_result($attrs)->insert; |
3c5b25c5 |
761 | } |
762 | |
87c4e602 |
763 | =head2 find_or_create |
764 | |
765 | =head3 Arguments: (\%vals, \%attrs?) |
87f0da6a |
766 | |
767 | $class->find_or_create({ key => $val, ... }); |
c2b15ecc |
768 | |
fd9f5466 |
769 | Searches for a record matching the search condition; if it doesn't find one, |
770 | creates one and returns that instead. |
87f0da6a |
771 | |
87f0da6a |
772 | my $cd = $schema->resultset('CD')->find_or_create({ |
773 | cdid => 5, |
774 | artist => 'Massive Attack', |
775 | title => 'Mezzanine', |
776 | year => 2005, |
777 | }); |
778 | |
779 | Also takes an optional C<key> attribute, to search by a specific key or unique |
780 | constraint. For example: |
781 | |
782 | my $cd = $schema->resultset('CD')->find_or_create( |
783 | { |
784 | artist => 'Massive Attack', |
785 | title => 'Mezzanine', |
786 | }, |
787 | { key => 'artist_title' } |
788 | ); |
789 | |
790 | See also L</find> and L</update_or_create>. |
791 | |
c2b15ecc |
792 | =cut |
793 | |
794 | sub find_or_create { |
795 | my $self = shift; |
87f0da6a |
796 | my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {}); |
797 | my $hash = ref $_[0] eq "HASH" ? shift : {@_}; |
798 | my $exists = $self->find($hash, $attrs); |
c2b15ecc |
799 | return defined($exists) ? $exists : $self->create($hash); |
800 | } |
801 | |
87f0da6a |
802 | =head2 update_or_create |
803 | |
804 | $class->update_or_create({ key => $val, ... }); |
805 | |
806 | First, search for an existing row matching one of the unique constraints |
807 | (including the primary key) on the source of this resultset. If a row is |
808 | found, update it with the other given column values. Otherwise, create a new |
809 | row. |
810 | |
811 | Takes an optional C<key> attribute to search on a specific unique constraint. |
812 | For example: |
813 | |
814 | # In your application |
815 | my $cd = $schema->resultset('CD')->update_or_create( |
816 | { |
817 | artist => 'Massive Attack', |
818 | title => 'Mezzanine', |
819 | year => 1998, |
820 | }, |
821 | { key => 'artist_title' } |
822 | ); |
823 | |
824 | If no C<key> is specified, it searches on all unique constraints defined on the |
825 | source, including the primary key. |
826 | |
827 | If the C<key> is specified as C<primary>, search only on the primary key. |
828 | |
a33df5d4 |
829 | See also L</find> and L</find_or_create>. |
830 | |
87f0da6a |
831 | =cut |
832 | |
833 | sub update_or_create { |
834 | my $self = shift; |
835 | |
836 | my $attrs = (@_ > 1 && ref $_[$#_] eq 'HASH' ? pop(@_) : {}); |
837 | my $hash = ref $_[0] eq "HASH" ? shift : {@_}; |
838 | |
701da8c4 |
839 | my %unique_constraints = $self->result_source->unique_constraints; |
87f0da6a |
840 | my @constraint_names = (exists $attrs->{key} |
841 | ? ($attrs->{key}) |
842 | : keys %unique_constraints); |
843 | |
844 | my @unique_hashes; |
845 | foreach my $name (@constraint_names) { |
846 | my @unique_cols = @{ $unique_constraints{$name} }; |
847 | my %unique_hash = |
848 | map { $_ => $hash->{$_} } |
849 | grep { exists $hash->{$_} } |
850 | @unique_cols; |
851 | |
852 | push @unique_hashes, \%unique_hash |
853 | if (scalar keys %unique_hash == scalar @unique_cols); |
854 | } |
855 | |
856 | my $row; |
857 | if (@unique_hashes) { |
858 | $row = $self->search(\@unique_hashes, { rows => 1 })->first; |
859 | if ($row) { |
860 | $row->set_columns($hash); |
861 | $row->update; |
862 | } |
863 | } |
864 | |
865 | unless ($row) { |
866 | $row = $self->create($hash); |
867 | } |
868 | |
869 | return $row; |
870 | } |
871 | |
64acc2bc |
872 | =head2 get_cache |
873 | |
874 | Gets the contents of the cache for the resultset. |
875 | |
876 | =cut |
877 | |
878 | sub get_cache { |
879 | my $self = shift; |
880 | return $self->{all_cache} || []; |
881 | } |
882 | |
883 | =head2 set_cache |
884 | |
885 | Sets the contents of the cache for the resultset. Expects an arrayref of objects of the same class as those produced by the resultset. |
886 | |
887 | =cut |
888 | |
889 | sub set_cache { |
890 | my ( $self, $data ) = @_; |
891 | $self->throw_exception("set_cache requires an arrayref") |
892 | if ref $data ne 'ARRAY'; |
a50bcd52 |
893 | my $result_class = $self->result_class; |
64acc2bc |
894 | foreach( @$data ) { |
895 | $self->throw_exception("cannot cache object of type '$_', expected '$result_class'") |
896 | if ref $_ ne $result_class; |
897 | } |
898 | $self->{all_cache} = $data; |
899 | } |
900 | |
901 | =head2 clear_cache |
902 | |
903 | Clears the cache for the resultset. |
904 | |
905 | =cut |
906 | |
907 | sub clear_cache { |
908 | my $self = shift; |
909 | $self->set_cache([]); |
910 | } |
911 | |
912 | =head2 related_resultset |
913 | |
914 | Returns a related resultset for the supplied relationship name. |
915 | |
916 | $rs = $rs->related_resultset('foo'); |
917 | |
918 | =cut |
919 | |
920 | sub related_resultset { |
921 | my ( $self, $rel, @rest ) = @_; |
922 | $self->{related_resultsets} ||= {}; |
923 | my $resultsets = $self->{related_resultsets}; |
924 | if( !exists $resultsets->{$rel} ) { |
925 | #warn "fetching related resultset for rel '$rel'"; |
926 | my $rel_obj = $self->result_source->relationship_info($rel); |
927 | $self->throw_exception( |
928 | "search_related: result source '" . $self->result_source->name . |
929 | "' has no such relationship ${rel}") |
930 | unless $rel_obj; #die Dumper $self->{attrs}; |
a86b1efe |
931 | my $rs = $self->search(undef, { join => $rel }); |
932 | #if( $self->{attrs}->{cache} ) { |
933 | # $rs = $self->search(undef); |
934 | #} |
935 | #else { |
936 | #} |
64acc2bc |
937 | #use Data::Dumper; die Dumper $rs->{attrs};#$rs = $self->search( undef ); |
938 | #use Data::Dumper; warn Dumper $self->{attrs}, Dumper $rs->{attrs}; |
939 | my $alias = (defined $rs->{attrs}{seen_join}{$rel} |
940 | && $rs->{attrs}{seen_join}{$rel} > 1 |
941 | ? join('_', $rel, $rs->{attrs}{seen_join}{$rel}) |
942 | : $rel); |
943 | $resultsets->{$rel} = |
944 | $self->result_source->schema->resultset($rel_obj->{class} |
945 | )->search( undef, |
946 | { %{$rs->{attrs}}, |
947 | alias => $alias, |
948 | select => undef(), |
949 | as => undef() } |
950 | )->search(@rest); |
951 | } |
952 | return $resultsets->{$rel}; |
953 | } |
954 | |
701da8c4 |
955 | =head2 throw_exception |
956 | |
957 | See Schema's throw_exception |
958 | |
959 | =cut |
960 | |
961 | sub throw_exception { |
962 | my $self=shift; |
963 | $self->result_source->schema->throw_exception(@_); |
964 | } |
965 | |
40dbc108 |
966 | =head1 ATTRIBUTES |
076652e8 |
967 | |
a33df5d4 |
968 | The resultset takes various attributes that modify its behavior. Here's an |
969 | overview of them: |
bfab575a |
970 | |
971 | =head2 order_by |
076652e8 |
972 | |
a33df5d4 |
973 | Which column(s) to order the results by. This is currently passed through |
974 | directly to SQL, so you can give e.g. C<foo DESC> for a descending order. |
076652e8 |
975 | |
87c4e602 |
976 | =head2 cols |
977 | |
978 | =head3 Arguments: (arrayref) |
976f3686 |
979 | |
a33df5d4 |
980 | Shortcut to request a particular set of columns to be retrieved. Adds |
981 | C<me.> onto the start of any column without a C<.> in it and sets C<select> |
982 | from that, then auto-populates C<as> from C<select> as normal. |
976f3686 |
983 | |
87c4e602 |
984 | =head2 include_columns |
985 | |
986 | =head3 Arguments: (arrayref) |
5ac6a044 |
987 | |
988 | Shortcut to include additional columns in the returned results - for example |
989 | |
990 | { include_columns => ['foo.name'], join => ['foo'] } |
991 | |
992 | would add a 'name' column to the information passed to object inflation |
993 | |
87c4e602 |
994 | =head2 select |
995 | |
996 | =head3 Arguments: (arrayref) |
976f3686 |
997 | |
4a28c340 |
998 | Indicates which columns should be selected from the storage. You can use |
999 | column names, or in the case of RDBMS back ends, function or stored procedure |
1000 | names: |
1001 | |
1002 | $rs = $schema->resultset('Foo')->search( |
1003 | {}, |
1004 | { |
cf7b40ed |
1005 | select => [ |
4a28c340 |
1006 | 'column_name', |
1007 | { count => 'column_to_count' }, |
1008 | { sum => 'column_to_sum' } |
cf7b40ed |
1009 | ] |
4a28c340 |
1010 | } |
1011 | ); |
1012 | |
1013 | When you use function/stored procedure names and do not supply an C<as> |
1014 | attribute, the column names returned are storage-dependent. E.g. MySQL would |
1015 | return a column named C<count(column_to_count)> in the above example. |
976f3686 |
1016 | |
87c4e602 |
1017 | =head2 as |
1018 | |
1019 | =head3 Arguments: (arrayref) |
076652e8 |
1020 | |
4a28c340 |
1021 | Indicates column names for object inflation. This is used in conjunction with |
1022 | C<select>, usually when C<select> contains one or more function or stored |
1023 | procedure names: |
1024 | |
1025 | $rs = $schema->resultset('Foo')->search( |
1026 | {}, |
1027 | { |
cf7b40ed |
1028 | select => [ |
4a28c340 |
1029 | 'column1', |
1030 | { count => 'column2' } |
cf7b40ed |
1031 | ], |
4a28c340 |
1032 | as => [qw/ column1 column2_count /] |
1033 | } |
1034 | ); |
1035 | |
1036 | my $foo = $rs->first(); # get the first Foo |
1037 | |
1038 | If the object against which the search is performed already has an accessor |
1039 | matching a column name specified in C<as>, the value can be retrieved using |
1040 | the accessor as normal: |
1041 | |
1042 | my $column1 = $foo->column1(); |
1043 | |
1044 | If on the other hand an accessor does not exist in the object, you need to |
1045 | use C<get_column> instead: |
1046 | |
1047 | my $column2_count = $foo->get_column('column2_count'); |
1048 | |
1049 | You can create your own accessors if required - see |
1050 | L<DBIx::Class::Manual::Cookbook> for details. |
ee38fa40 |
1051 | |
bfab575a |
1052 | =head2 join |
ee38fa40 |
1053 | |
a33df5d4 |
1054 | Contains a list of relationships that should be joined for this query. For |
1055 | example: |
1056 | |
1057 | # Get CDs by Nine Inch Nails |
1058 | my $rs = $schema->resultset('CD')->search( |
1059 | { 'artist.name' => 'Nine Inch Nails' }, |
1060 | { join => 'artist' } |
1061 | ); |
1062 | |
1063 | Can also contain a hash reference to refer to the other relation's relations. |
1064 | For example: |
1065 | |
1066 | package MyApp::Schema::Track; |
1067 | use base qw/DBIx::Class/; |
1068 | __PACKAGE__->table('track'); |
1069 | __PACKAGE__->add_columns(qw/trackid cd position title/); |
1070 | __PACKAGE__->set_primary_key('trackid'); |
1071 | __PACKAGE__->belongs_to(cd => 'MyApp::Schema::CD'); |
1072 | 1; |
1073 | |
1074 | # In your application |
1075 | my $rs = $schema->resultset('Artist')->search( |
1076 | { 'track.title' => 'Teardrop' }, |
1077 | { |
1078 | join => { cd => 'track' }, |
1079 | order_by => 'artist.name', |
1080 | } |
1081 | ); |
1082 | |
2cb360cc |
1083 | If the same join is supplied twice, it will be aliased to <rel>_2 (and |
1084 | similarly for a third time). For e.g. |
1085 | |
1086 | my $rs = $schema->resultset('Artist')->search( |
1087 | { 'cds.title' => 'Foo', |
1088 | 'cds_2.title' => 'Bar' }, |
1089 | { join => [ qw/cds cds/ ] }); |
1090 | |
1091 | will return a set of all artists that have both a cd with title Foo and a cd |
1092 | with title Bar. |
1093 | |
1094 | If you want to fetch related objects from other tables as well, see C<prefetch> |
ae1c90a1 |
1095 | below. |
ee38fa40 |
1096 | |
87c4e602 |
1097 | =head2 prefetch |
1098 | |
1099 | =head3 Arguments: arrayref/hashref |
ee38fa40 |
1100 | |
ae1c90a1 |
1101 | Contains one or more relationships that should be fetched along with the main |
bfab575a |
1102 | query (when they are accessed afterwards they will have already been |
a33df5d4 |
1103 | "prefetched"). This is useful for when you know you will need the related |
ae1c90a1 |
1104 | objects, because it saves at least one query: |
1105 | |
1106 | my $rs = $schema->resultset('Tag')->search( |
1107 | {}, |
1108 | { |
1109 | prefetch => { |
1110 | cd => 'artist' |
1111 | } |
1112 | } |
1113 | ); |
1114 | |
1115 | The initial search results in SQL like the following: |
1116 | |
1117 | SELECT tag.*, cd.*, artist.* FROM tag |
1118 | JOIN cd ON tag.cd = cd.cdid |
1119 | JOIN artist ON cd.artist = artist.artistid |
1120 | |
1121 | L<DBIx::Class> has no need to go back to the database when we access the |
1122 | C<cd> or C<artist> relationships, which saves us two SQL statements in this |
1123 | case. |
1124 | |
2cb360cc |
1125 | Simple prefetches will be joined automatically, so there is no need |
1126 | for a C<join> attribute in the above search. If you're prefetching to |
1127 | depth (e.g. { cd => { artist => 'label' } or similar), you'll need to |
1128 | specify the join as well. |
ae1c90a1 |
1129 | |
1130 | C<prefetch> can be used with the following relationship types: C<belongs_to>, |
2cb360cc |
1131 | C<has_one> (or if you're using C<add_relationship>, any relationship declared |
1132 | with an accessor type of 'single' or 'filter'). |
ee38fa40 |
1133 | |
87c4e602 |
1134 | =head2 from |
1135 | |
1136 | =head3 Arguments: (arrayref) |
ee38fa40 |
1137 | |
4a28c340 |
1138 | The C<from> attribute gives you manual control over the C<FROM> clause of SQL |
1139 | statements generated by L<DBIx::Class>, allowing you to express custom C<JOIN> |
1140 | clauses. |
ee38fa40 |
1141 | |
a33df5d4 |
1142 | NOTE: Use this on your own risk. This allows you to shoot off your foot! |
4a28c340 |
1143 | C<join> will usually do what you need and it is strongly recommended that you |
1144 | avoid using C<from> unless you cannot achieve the desired result using C<join>. |
1145 | |
1146 | In simple terms, C<from> works as follows: |
1147 | |
1148 | [ |
1149 | { <alias> => <table>, -join-type => 'inner|left|right' } |
1150 | [] # nested JOIN (optional) |
1151 | { <table.column> = <foreign_table.foreign_key> } |
1152 | ] |
1153 | |
1154 | JOIN |
1155 | <alias> <table> |
1156 | [JOIN ...] |
1157 | ON <table.column> = <foreign_table.foreign_key> |
1158 | |
1159 | An easy way to follow the examples below is to remember the following: |
1160 | |
1161 | Anything inside "[]" is a JOIN |
1162 | Anything inside "{}" is a condition for the enclosing JOIN |
1163 | |
1164 | The following examples utilize a "person" table in a family tree application. |
1165 | In order to express parent->child relationships, this table is self-joined: |
1166 | |
1167 | # Person->belongs_to('father' => 'Person'); |
1168 | # Person->belongs_to('mother' => 'Person'); |
1169 | |
1170 | C<from> can be used to nest joins. Here we return all children with a father, |
1171 | then search against all mothers of those children: |
1172 | |
1173 | $rs = $schema->resultset('Person')->search( |
1174 | {}, |
1175 | { |
1176 | alias => 'mother', # alias columns in accordance with "from" |
1177 | from => [ |
1178 | { mother => 'person' }, |
1179 | [ |
1180 | [ |
1181 | { child => 'person' }, |
1182 | [ |
1183 | { father => 'person' }, |
1184 | { 'father.person_id' => 'child.father_id' } |
1185 | ] |
1186 | ], |
1187 | { 'mother.person_id' => 'child.mother_id' } |
fd9f5466 |
1188 | ], |
4a28c340 |
1189 | ] |
1190 | }, |
1191 | ); |
1192 | |
1193 | # Equivalent SQL: |
1194 | # SELECT mother.* FROM person mother |
1195 | # JOIN ( |
1196 | # person child |
1197 | # JOIN person father |
1198 | # ON ( father.person_id = child.father_id ) |
1199 | # ) |
1200 | # ON ( mother.person_id = child.mother_id ) |
1201 | |
1202 | The type of any join can be controlled manually. To search against only people |
1203 | with a father in the person table, we could explicitly use C<INNER JOIN>: |
1204 | |
1205 | $rs = $schema->resultset('Person')->search( |
1206 | {}, |
1207 | { |
1208 | alias => 'child', # alias columns in accordance with "from" |
1209 | from => [ |
1210 | { child => 'person' }, |
1211 | [ |
1212 | { father => 'person', -join-type => 'inner' }, |
1213 | { 'father.id' => 'child.father_id' } |
1214 | ], |
1215 | ] |
1216 | }, |
1217 | ); |
1218 | |
1219 | # Equivalent SQL: |
1220 | # SELECT child.* FROM person child |
1221 | # INNER JOIN person father ON child.father_id = father.id |
ee38fa40 |
1222 | |
bfab575a |
1223 | =head2 page |
076652e8 |
1224 | |
a33df5d4 |
1225 | For a paged resultset, specifies which page to retrieve. Leave unset |
bfab575a |
1226 | for an unpaged resultset. |
076652e8 |
1227 | |
bfab575a |
1228 | =head2 rows |
076652e8 |
1229 | |
4a28c340 |
1230 | For a paged resultset, how many rows per page: |
1231 | |
1232 | rows => 10 |
1233 | |
1234 | Can also be used to simulate an SQL C<LIMIT>. |
076652e8 |
1235 | |
87c4e602 |
1236 | =head2 group_by |
1237 | |
1238 | =head3 Arguments: (arrayref) |
54540863 |
1239 | |
bda4c2b8 |
1240 | A arrayref of columns to group by. Can include columns of joined tables. |
54540863 |
1241 | |
675ce4a6 |
1242 | group_by => [qw/ column1 column2 ... /] |
1243 | |
54540863 |
1244 | =head2 distinct |
1245 | |
a33df5d4 |
1246 | Set to 1 to group by all columns. |
1247 | |
1248 | For more examples of using these attributes, see |
1249 | L<DBIx::Class::Manual::Cookbook>. |
54540863 |
1250 | |
bfab575a |
1251 | =cut |
076652e8 |
1252 | |
89c0a5a2 |
1253 | 1; |