3 DBIx::Class::Manual::Cookbook - Miscellaneous recipes
9 When you expect a large number of results, you can ask L<DBIx::Class> for a
10 paged resultset, which will fetch only a defined number of records at a time:
12 my $rs = $schema->resultset('Artist')->search(
15 page => 1, # page to return (defaults to 1)
16 rows => 10, # number of results per page
20 return $rs->all(); # all records for page 1
22 The C<page> attribute does not have to be specified in your search:
24 my $rs = $schema->resultset('Artist')->search(
31 return $rs->page(1); # DBIx::Class::ResultSet containing first 10 records
33 In either of the above cases, you can get a L<Data::Page> object for the
34 resultset (suitable for use in e.g. a template) using the C<pager> method:
38 =head2 Complex WHERE clauses
40 Sometimes you need to formulate a query using specific operators:
42 my @albums = $schema->resultset('Album')->search({
43 artist => { 'like', '%Lamb%' },
44 title => { 'like', '%Fear of Fours%' },
47 This results in something like the following C<WHERE> clause:
49 WHERE artist LIKE '%Lamb%' AND title LIKE '%Fear of Fours%'
51 Other queries might require slightly more complex logic:
53 my @albums = $schema->resultset('Album')->search({
56 artist => { 'like', '%Smashing Pumpkins%' },
57 title => 'Siamese Dream',
59 artist => 'Starchildren',
63 This results in the following C<WHERE> clause:
65 WHERE ( artist LIKE '%Smashing Pumpkins%' AND title = 'Siamese Dream' )
66 OR artist = 'Starchildren'
68 For more information on generating complex queries, see
69 L<SQL::Abstract/WHERE CLAUSES>.
71 =head2 Retrieve one and only one row from a resultset
73 Sometimes you need only the first "top" row of a resultset. While this can be
74 easily done with L<< $rs->first|DBIx::Class::ResultSet/first >>, it is suboptimal,
75 as a full blown cursor for the resultset will be created and then immediately
76 destroyed after fetching the first row object.
77 L<< $rs->single|DBIx::Class::ResultSet/single >> is
78 designed specifically for this case - it will grab the first returned result
79 without even instantiating a cursor.
81 Before replacing all your calls to C<first()> with C<single()> please observe the
87 While single() takes a search condition just like search() does, it does
88 _not_ accept search attributes. However one can always chain a single() to
91 my $top_cd = $cd_rs -> search({}, { order_by => 'rating' }) -> single;
95 Since single() is the engine behind find(), it is designed to fetch a
96 single row per database query. Thus a warning will be issued when the
97 underlying SELECT returns more than one row. Sometimes however this usage
98 is valid: i.e. we have an arbitrary number of cd's but only one of them is
99 at the top of the charts at any given time. If you know what you are doing,
100 you can silence the warning by explicitly limiting the resultset size:
102 my $top_cd = $cd_rs -> search ({}, { order_by => 'rating', rows => 1 }) -> single;
106 =head2 Arbitrary SQL through a custom ResultSource
108 Sometimes you have to run arbitrary SQL because your query is too complex
109 (e.g. it contains Unions, Sub-Selects, Stored Procedures, etc.) or has to
110 be optimized for your database in a special way, but you still want to
111 get the results as a L<DBIx::Class::ResultSet>.
112 The recommended way to accomplish this is by defining a separate ResultSource
113 for your query. You can then inject complete SQL statements using a scalar
114 reference (this is a feature of L<SQL::Abstract>).
116 Say you want to run a complex custom query on your user data, here's what
117 you have to add to your User class:
119 package My::Schema::User;
121 use base qw/DBIx::Class/;
123 # ->load_components, ->table, ->add_columns, etc.
125 # Make a new ResultSource based on the User class
126 my $source = __PACKAGE__->result_source_instance();
127 my $new_source = $source->new( $source );
128 $new_source->source_name( 'UserFriendsComplex' );
130 # Hand in your query as a scalar reference
131 # It will be added as a sub-select after FROM,
132 # so pay attention to the surrounding brackets!
133 $new_source->name( \<<SQL );
134 ( SELECT u.* FROM user u
135 INNER JOIN user_friends f ON u.id = f.user_id
136 WHERE f.friend_user_id = ?
138 SELECT u.* FROM user u
139 INNER JOIN user_friends f ON u.id = f.friend_user_id
140 WHERE f.user_id = ? )
143 # Finally, register your new ResultSource with your Schema
144 My::Schema->register_extra_source( 'UserFriendsComplex' => $new_source );
146 Next, you can execute your complex query using bind parameters like this:
148 my $friends = [ $schema->resultset( 'UserFriendsComplex' )->search( {},
150 bind => [ 12345, 12345 ]
154 ... and you'll get back a perfect L<DBIx::Class::ResultSet> (except, of course,
155 that you cannot modify the rows it contains, ie. cannot call L</update>,
156 L</delete>, ... on it).
158 If you prefer to have the definitions of these custom ResultSources in separate
159 files (instead of stuffing all of them into the same resultset class), you can
160 achieve the same with subclassing the resultset class and defining the
163 package My::Schema::UserFriendsComplex;
165 use My::Schema::User;
166 use base qw/My::Schema::User/;
168 __PACKAGE__->table('dummy'); # currently must be called before anything else
170 # Hand in your query as a scalar reference
171 # It will be added as a sub-select after FROM,
172 # so pay attention to the surrounding brackets!
173 __PACKAGE__->name( \<<SQL );
174 ( SELECT u.* FROM user u
175 INNER JOIN user_friends f ON u.id = f.user_id
176 WHERE f.friend_user_id = ?
178 SELECT u.* FROM user u
179 INNER JOIN user_friends f ON u.id = f.friend_user_id
180 WHERE f.user_id = ? )
185 =head2 Using specific columns
187 When you only want specific columns from a table, you can use
188 C<columns> to specify which ones you need. This is useful to avoid
189 loading columns with large amounts of data that you aren't about to
192 my $rs = $schema->resultset('Artist')->search(
195 columns => [qw/ name /]
200 # SELECT artist.name FROM artist
202 This is a shortcut for C<select> and C<as>, see below. C<columns>
203 cannot be used together with C<select> and C<as>.
205 =head2 Using database functions or stored procedures
207 The combination of C<select> and C<as> can be used to return the result of a
208 database function or stored procedure as a column value. You use C<select> to
209 specify the source for your column value (e.g. a column name, function, or
210 stored procedure name). You then use C<as> to set the column name you will use
211 to access the returned value:
213 my $rs = $schema->resultset('Artist')->search(
216 select => [ 'name', { LENGTH => 'name' } ],
217 as => [qw/ name name_length /],
222 # SELECT name name, LENGTH( name )
225 Note that the C< as > attribute has absolutely nothing to with the sql
226 syntax C< SELECT foo AS bar > (see the documentation in
227 L<DBIx::Class::ResultSet/ATTRIBUTES>). If your alias exists as a
228 column in your base class (i.e. it was added with C<add_columns>), you
229 just access it as normal. Our C<Artist> class has a C<name> column, so
230 we just use the C<name> accessor:
232 my $artist = $rs->first();
233 my $name = $artist->name();
235 If on the other hand the alias does not correspond to an existing column, you
236 have to fetch the value using the C<get_column> accessor:
238 my $name_length = $artist->get_column('name_length');
240 If you don't like using C<get_column>, you can always create an accessor for
241 any of your aliases using either of these:
243 # Define accessor manually:
244 sub name_length { shift->get_column('name_length'); }
246 # Or use DBIx::Class::AccessorGroup:
247 __PACKAGE__->mk_group_accessors('column' => 'name_length');
249 =head2 SELECT DISTINCT with multiple columns
251 my $rs = $schema->resultset('Foo')->search(
255 { distinct => [ $source->columns ] }
257 as => [ $source->columns ] # remember 'as' is not the same as SQL AS :-)
261 =head2 SELECT COUNT(DISTINCT colname)
263 my $rs = $schema->resultset('Foo')->search(
267 { count => { distinct => 'colname' } }
273 my $count = $rs->next->get_column('count');
275 =head2 Grouping results
277 L<DBIx::Class> supports C<GROUP BY> as follows:
279 my $rs = $schema->resultset('Artist')->search(
283 select => [ 'name', { count => 'cds.id' } ],
284 as => [qw/ name cd_count /],
285 group_by => [qw/ name /]
290 # SELECT name, COUNT( cd.id ) FROM artist
291 # LEFT JOIN cd ON artist.id = cd.artist
294 Please see L<DBIx::Class::ResultSet/ATTRIBUTES> documentation if you
295 are in any way unsure about the use of the attributes above (C< join
296 >, C< select >, C< as > and C< group_by >).
300 You can write subqueries relatively easily in DBIC.
302 my $inside_rs = $schema->resultset('Artist')->search({
303 name => [ 'Billy Joel', 'Brittany Spears' ],
306 my $rs = $schema->resultset('CD')->search({
307 artist_id => { 'IN' => $inside_rs->get_column('id')->as_query },
310 The usual operators ( =, !=, IN, NOT IN, etc) are supported.
312 B<NOTE>: You have to explicitly use '=' when doing an equality comparison.
313 The following will B<not> work:
315 my $rs = $schema->resultset('CD')->search({
316 artist_id => $inside_rs->get_column('id')->as_query,
321 Subqueries are supported in the where clause (first hashref), and in the
322 from, select, and +select attributes.
324 =head3 Correlated subqueries
326 my $cdrs = $schema->resultset('CD');
327 my $rs = $cdrs->search({
329 '=' => $cdrs->search(
330 { artistid => { '=' => \'me.artistid' } },
332 )->get_column('year')->max_rs->as_query,
336 That creates the following SQL:
338 SELECT me.cdid, me.artist, me.title, me.year, me.genreid, me.single_track
341 SELECT MAX(inner.year)
343 WHERE artistid = me.artistid
346 =head2 Predefined searches
348 You can write your own L<DBIx::Class::ResultSet> class by inheriting from it
349 and define often used searches as methods:
351 package My::DBIC::ResultSet::CD;
354 use base 'DBIx::Class::ResultSet';
356 sub search_cds_ordered {
359 return $self->search(
361 { order_by => 'name DESC' },
367 To use your resultset, first tell DBIx::Class to create an instance of it
368 for you, in your My::DBIC::Schema::CD class:
370 # class definition as normal
371 __PACKAGE__->load_components(qw/ Core /);
372 __PACKAGE__->table('cd');
374 # tell DBIC to use the custom ResultSet class
375 __PACKAGE__->resultset_class('My::DBIC::ResultSet::CD');
377 Note that C<resultset_class> must be called after C<load_components> and C<table>, or you will get errors about missing methods.
379 Then call your new method in your code:
381 my $ordered_cds = $schema->resultset('CD')->search_cds_ordered();
383 =head2 Using SQL functions on the left hand side of a comparison
385 Using SQL functions on the left hand side of a comparison is generally
386 not a good idea since it requires a scan of the entire table. However,
387 it can be accomplished with C<DBIx::Class> when necessary.
389 If you do not have quoting on, simply include the function in your search
390 specification as you would any column:
392 $rs->search({ 'YEAR(date_of_birth)' => 1979 });
394 With quoting on, or for a more portable solution, use the C<where>
397 $rs->search({}, { where => \'YEAR(date_of_birth) = 1979' });
401 (When the bind args ordering bug is fixed, this technique will be better
402 and can replace the one above.)
404 With quoting on, or for a more portable solution, use the C<where> and
408 where => \'YEAR(date_of_birth) = ?',
414 =head1 JOINS AND PREFETCHING
416 =head2 Using joins and prefetch
418 You can use the C<join> attribute to allow searching on, or sorting your
419 results by, one or more columns in a related table. To return all CDs matching
420 a particular artist name:
422 my $rs = $schema->resultset('CD')->search(
424 'artist.name' => 'Bob Marley'
427 join => 'artist', # join the artist table
432 # SELECT cd.* FROM cd
433 # JOIN artist ON cd.artist = artist.id
434 # WHERE artist.name = 'Bob Marley'
436 If required, you can now sort on any column in the related tables by including
437 it in your C<order_by> attribute:
439 my $rs = $schema->resultset('CD')->search(
441 'artist.name' => 'Bob Marley'
445 order_by => [qw/ artist.name /]
450 # SELECT cd.* FROM cd
451 # JOIN artist ON cd.artist = artist.id
452 # WHERE artist.name = 'Bob Marley'
453 # ORDER BY artist.name
455 Note that the C<join> attribute should only be used when you need to search or
456 sort using columns in a related table. Joining related tables when you only
457 need columns from the main table will make performance worse!
459 Now let's say you want to display a list of CDs, each with the name of the
460 artist. The following will work fine:
462 while (my $cd = $rs->next) {
463 print "CD: " . $cd->title . ", Artist: " . $cd->artist->name;
466 There is a problem however. We have searched both the C<cd> and C<artist> tables
467 in our main query, but we have only returned data from the C<cd> table. To get
468 the artist name for any of the CD objects returned, L<DBIx::Class> will go back
471 SELECT artist.* FROM artist WHERE artist.id = ?
473 A statement like the one above will run for each and every CD returned by our
474 main query. Five CDs, five extra queries. A hundred CDs, one hundred extra
477 Thankfully, L<DBIx::Class> has a C<prefetch> attribute to solve this problem.
478 This allows you to fetch results from related tables in advance:
480 my $rs = $schema->resultset('CD')->search(
482 'artist.name' => 'Bob Marley'
486 order_by => [qw/ artist.name /],
487 prefetch => 'artist' # return artist data too!
491 # Equivalent SQL (note SELECT from both "cd" and "artist"):
492 # SELECT cd.*, artist.* FROM cd
493 # JOIN artist ON cd.artist = artist.id
494 # WHERE artist.name = 'Bob Marley'
495 # ORDER BY artist.name
497 The code to print the CD list remains the same:
499 while (my $cd = $rs->next) {
500 print "CD: " . $cd->title . ", Artist: " . $cd->artist->name;
503 L<DBIx::Class> has now prefetched all matching data from the C<artist> table,
504 so no additional SQL statements are executed. You now have a much more
507 Note that as of L<DBIx::Class> 0.05999_01, C<prefetch> I<can> be used with
508 C<has_many> relationships.
510 Also note that C<prefetch> should only be used when you know you will
511 definitely use data from a related table. Pre-fetching related tables when you
512 only need columns from the main table will make performance worse!
514 =head2 Multiple joins
516 In the examples above, the C<join> attribute was a scalar. If you
517 pass an array reference instead, you can join to multiple tables. In
518 this example, we want to limit the search further, using
521 # Relationships defined elsewhere:
522 # CD->belongs_to('artist' => 'Artist');
523 # CD->has_one('liner_notes' => 'LinerNotes', 'cd');
524 my $rs = $schema->resultset('CD')->search(
526 'artist.name' => 'Bob Marley'
527 'liner_notes.notes' => { 'like', '%some text%' },
530 join => [qw/ artist liner_notes /],
531 order_by => [qw/ artist.name /],
536 # SELECT cd.*, artist.*, liner_notes.* FROM cd
537 # JOIN artist ON cd.artist = artist.id
538 # JOIN liner_notes ON cd.id = liner_notes.cd
539 # WHERE artist.name = 'Bob Marley'
540 # ORDER BY artist.name
542 =head2 Multi-step joins
544 Sometimes you want to join more than one relationship deep. In this example,
545 we want to find all C<Artist> objects who have C<CD>s whose C<LinerNotes>
546 contain a specific string:
548 # Relationships defined elsewhere:
549 # Artist->has_many('cds' => 'CD', 'artist');
550 # CD->has_one('liner_notes' => 'LinerNotes', 'cd');
552 my $rs = $schema->resultset('Artist')->search(
554 'liner_notes.notes' => { 'like', '%some text%' },
558 'cds' => 'liner_notes'
564 # SELECT artist.* FROM artist
565 # LEFT JOIN cd ON artist.id = cd.artist
566 # LEFT JOIN liner_notes ON cd.id = liner_notes.cd
567 # WHERE liner_notes.notes LIKE '%some text%'
569 Joins can be nested to an arbitrary level. So if we decide later that we
570 want to reduce the number of Artists returned based on who wrote the liner
573 # Relationship defined elsewhere:
574 # LinerNotes->belongs_to('author' => 'Person');
576 my $rs = $schema->resultset('Artist')->search(
578 'liner_notes.notes' => { 'like', '%some text%' },
579 'author.name' => 'A. Writer'
584 'liner_notes' => 'author'
591 # SELECT artist.* FROM artist
592 # LEFT JOIN cd ON artist.id = cd.artist
593 # LEFT JOIN liner_notes ON cd.id = liner_notes.cd
594 # LEFT JOIN author ON author.id = liner_notes.author
595 # WHERE liner_notes.notes LIKE '%some text%'
596 # AND author.name = 'A. Writer'
598 =head2 Multi-step and multiple joins
600 With various combinations of array and hash references, you can join
601 tables in any combination you desire. For example, to join Artist to
602 CD and Concert, and join CD to LinerNotes:
604 # Relationships defined elsewhere:
605 # Artist->has_many('concerts' => 'Concert', 'artist');
607 my $rs = $schema->resultset('Artist')->search(
620 # SELECT artist.* FROM artist
621 # LEFT JOIN cd ON artist.id = cd.artist
622 # LEFT JOIN liner_notes ON cd.id = liner_notes.cd
623 # LEFT JOIN concert ON artist.id = concert.artist
625 =head2 Multi-step prefetch
627 From 0.04999_05 onwards, C<prefetch> can be nested more than one relationship
628 deep using the same syntax as a multi-step join:
630 my $rs = $schema->resultset('Tag')->search(
640 # SELECT tag.*, cd.*, artist.* FROM tag
641 # JOIN cd ON tag.cd = cd.id
642 # JOIN artist ON cd.artist = artist.id
644 Now accessing our C<cd> and C<artist> relationships does not need additional
647 my $tag = $rs->first;
648 print $tag->cd->artist->name;
650 =head1 ROW-LEVEL OPERATIONS
652 =head2 Retrieving a row object's Schema
654 It is possible to get a Schema object from a row object like so:
656 my $schema = $cd->result_source->schema;
657 # use the schema as normal:
658 my $artist_rs = $schema->resultset('Artist');
660 This can be useful when you don't want to pass around a Schema object to every
663 =head2 Getting the value of the primary key for the last database insert
665 AKA getting last_insert_id
667 If you are using PK::Auto (which is a core component as of 0.07), this is
670 my $foo = $rs->create(\%blah);
672 my $id = $foo->id; # foo->my_primary_key_field will also work.
674 If you are not using autoincrementing primary keys, this will probably
675 not work, but then you already know the value of the last primary key anyway.
677 =head2 Stringification
679 Employ the standard stringification technique by using the C<overload>
682 To make an object stringify itself as a single column, use something
683 like this (replace C<foo> with the column/method of your choice):
685 use overload '""' => sub { shift->name}, fallback => 1;
687 For more complex stringification, you can use an anonymous subroutine:
689 use overload '""' => sub { $_[0]->name . ", " .
690 $_[0]->address }, fallback => 1;
692 =head3 Stringification Example
694 Suppose we have two tables: C<Product> and C<Category>. The table
697 Product(id, Description, category)
698 Category(id, Description)
700 C<category> is a foreign key into the Category table.
702 If you have a Product object C<$obj> and write something like
706 things will not work as expected.
708 To obtain, for example, the category description, you should add this
709 method to the class defining the Category table:
711 use overload "" => sub {
714 return $self->Description;
717 =head2 Want to know if find_or_create found or created a row?
719 Just use C<find_or_new> instead, then check C<in_storage>:
721 my $obj = $rs->find_or_new({ blah => 'blarg' });
722 unless ($obj->in_storage) {
724 # do whatever else you wanted if it was a new row
727 =head2 Dynamic Sub-classing DBIx::Class proxy classes
729 AKA multi-class object inflation from one table
731 L<DBIx::Class> classes are proxy classes, therefore some different
732 techniques need to be employed for more than basic subclassing. In
733 this example we have a single user table that carries a boolean bit
734 for admin. We would like like to give the admin users
735 objects(L<DBIx::Class::Row>) the same methods as a regular user but
736 also special admin only methods. It doesn't make sense to create two
737 seperate proxy-class files for this. We would be copying all the user
738 methods into the Admin class. There is a cleaner way to accomplish
741 Overriding the C<inflate_result> method within the User proxy-class
742 gives us the effect we want. This method is called by
743 L<DBIx::Class::ResultSet> when inflating a result from storage. So we
744 grab the object being returned, inspect the values we are looking for,
745 bless it if it's an admin object, and then return it. See the example
752 use base qw/DBIx::Class::Schema/;
754 __PACKAGE__->load_classes(qw/User/);
757 B<Proxy-Class definitions>
759 package DB::Schema::User;
763 use base qw/DBIx::Class/;
765 ### Defined what our admin class is for ensure_class_loaded
766 my $admin_class = __PACKAGE__ . '::Admin';
768 __PACKAGE__->load_components(qw/Core/);
770 __PACKAGE__->table('users');
772 __PACKAGE__->add_columns(qw/user_id email password
773 firstname lastname active
776 __PACKAGE__->set_primary_key('user_id');
780 my $ret = $self->next::method(@_);
781 if( $ret->admin ) {### If this is an admin rebless for extra functions
782 $self->ensure_class_loaded( $admin_class );
783 bless $ret, $admin_class;
789 print "I am a regular user.\n";
794 package DB::Schema::User::Admin;
798 use base qw/DB::Schema::User/;
802 print "I am an admin.\n";
808 print "I am doing admin stuff\n";
818 my $user_data = { email => 'someguy@place.com',
822 my $admin_data = { email => 'someadmin@adminplace.com',
826 my $schema = DB::Schema->connection('dbi:Pg:dbname=test');
828 $schema->resultset('User')->create( $user_data );
829 $schema->resultset('User')->create( $admin_data );
831 ### Now we search for them
832 my $user = $schema->resultset('User')->single( $user_data );
833 my $admin = $schema->resultset('User')->single( $admin_data );
835 print ref $user, "\n";
836 print ref $admin, "\n";
838 print $user->password , "\n"; # pass1
839 print $admin->password , "\n";# pass2; inherited from User
840 print $user->hello , "\n";# I am a regular user.
841 print $admin->hello, "\n";# I am an admin.
843 ### The statement below will NOT print
844 print "I can do admin stuff\n" if $user->can('do_admin_stuff');
845 ### The statement below will print
846 print "I can do admin stuff\n" if $admin->can('do_admin_stuff');
848 =head2 Skip row object creation for faster results
850 DBIx::Class is not built for speed, it's built for convenience and
851 ease of use, but sometimes you just need to get the data, and skip the
854 To do this simply use L<DBIx::Class::ResultClass::HashRefInflator>.
856 my $rs = $schema->resultset('CD');
858 $rs->result_class('DBIx::Class::ResultClass::HashRefInflator');
860 my $hash_ref = $rs->find(1);
864 =head2 Get raw data for blindingly fast results
866 If the L<HashRefInflator|DBIx::Class::ResultClass::HashRefInflator> solution
867 above is not fast enough for you, you can use a DBIx::Class to return values
868 exactly as they come out of the data base with none of the convenience methods
871 This is used like so:
873 my $cursor = $rs->cursor
874 while (my @vals = $cursor->next) {
875 # use $val[0..n] here
878 You will need to map the array offsets to particular columns (you can
879 use the I<select> attribute of C<search()> to force ordering).
881 =head1 RESULTSET OPERATIONS
883 =head2 Getting Schema from a ResultSet
885 To get the schema object from a result set, do the following:
887 $rs->result_source->schema
889 =head2 Getting Columns Of Data
893 If you want to find the sum of a particular column there are several
894 ways, the obvious one is to use search:
896 my $rs = $schema->resultset('Items')->search(
899 select => [ { sum => 'Cost' } ],
900 as => [ 'total_cost' ], # remember this 'as' is for DBIx::Class::ResultSet not SQL
903 my $tc = $rs->first->get_column('total_cost');
905 Or, you can use the L<DBIx::Class::ResultSetColumn>, which gets
906 returned when you ask the C<ResultSet> for a column using
909 my $cost = $schema->resultset('Items')->get_column('Cost');
912 With this you can also do:
914 my $minvalue = $cost->min;
915 my $maxvalue = $cost->max;
917 Or just iterate through the values of this column only:
919 while ( my $c = $cost->next ) {
923 foreach my $c ($cost->all) {
927 C<ResultSetColumn> only has a limited number of built-in functions, if
928 you need one that it doesn't have, then you can use the C<func> method
931 my $avg = $cost->func('AVERAGE');
933 This will cause the following SQL statement to be run:
935 SELECT AVERAGE(Cost) FROM Items me
937 Which will of course only work if your database supports this function.
938 See L<DBIx::Class::ResultSetColumn> for more documentation.
940 =head2 Creating a result set from a set of rows
942 Sometimes you have a (set of) row objects that you want to put into a
943 resultset without the need to hit the DB again. You can do that by using the
944 L<set_cache|DBIx::Class::Resultset/set_cache> method:
946 my @uploadable_groups;
947 while (my $group = $groups->next) {
948 if ($group->can_upload($self)) {
949 push @uploadable_groups, $group;
952 my $new_rs = $self->result_source->resultset;
953 $new_rs->set_cache(\@uploadable_groups);
957 =head1 USING RELATIONSHIPS
959 =head2 Create a new row in a related table
961 my $author = $book->create_related('author', { name => 'Fred'});
963 =head2 Search in a related table
965 Only searches for books named 'Titanic' by the author in $author.
967 my $books_rs = $author->search_related('books', { name => 'Titanic' });
969 =head2 Delete data in a related table
971 Deletes only the book named Titanic by the author in $author.
973 $author->delete_related('books', { name => 'Titanic' });
975 =head2 Ordering a relationship result set
977 If you always want a relation to be ordered, you can specify this when you
978 create the relationship.
980 To order C<< $book->pages >> by descending page_number, create the relation
983 __PACKAGE__->has_many('pages' => 'Page', 'book', { order_by => \'page_number DESC'} );
985 =head2 Filtering a relationship result set
987 If you want to get a filtered result set, you can just add add to $attr as follows:
989 __PACKAGE__->has_many('pages' => 'Page', 'book', { where => { scrap => 0 } } );
991 =head2 Many-to-many relationships
993 This is straightforward using L<ManyToMany|DBIx::Class::Relationship/many_to_many>:
996 use base 'DBIx::Class';
997 __PACKAGE__->load_components('Core');
998 __PACKAGE__->table('user');
999 __PACKAGE__->add_columns(qw/id name/);
1000 __PACKAGE__->set_primary_key('id');
1001 __PACKAGE__->has_many('user_address' => 'My::UserAddress', 'user');
1002 __PACKAGE__->many_to_many('addresses' => 'user_address', 'address');
1004 package My::UserAddress;
1005 use base 'DBIx::Class';
1006 __PACKAGE__->load_components('Core');
1007 __PACKAGE__->table('user_address');
1008 __PACKAGE__->add_columns(qw/user address/);
1009 __PACKAGE__->set_primary_key(qw/user address/);
1010 __PACKAGE__->belongs_to('user' => 'My::User');
1011 __PACKAGE__->belongs_to('address' => 'My::Address');
1013 package My::Address;
1014 use base 'DBIx::Class';
1015 __PACKAGE__->load_components('Core');
1016 __PACKAGE__->table('address');
1017 __PACKAGE__->add_columns(qw/id street town area_code country/);
1018 __PACKAGE__->set_primary_key('id');
1019 __PACKAGE__->has_many('user_address' => 'My::UserAddress', 'address');
1020 __PACKAGE__->many_to_many('users' => 'user_address', 'user');
1022 $rs = $user->addresses(); # get all addresses for a user
1023 $rs = $address->users(); # get all users for an address
1027 As of version 0.04001, there is improved transaction support in
1028 L<DBIx::Class::Storage> and L<DBIx::Class::Schema>. Here is an
1029 example of the recommended way to use it:
1031 my $genus = $schema->resultset('Genus')->find(12);
1033 my $coderef2 = sub {
1038 my $coderef1 = sub {
1039 $genus->add_to_species({ name => 'troglodyte' });
1042 $schema->txn_do($coderef2); # Can have a nested transaction. Only the outer will actualy commit
1043 return $genus->species;
1048 $rs = $schema->txn_do($coderef1);
1051 if ($@) { # Transaction failed
1052 die "the sky is falling!" #
1053 if ($@ =~ /Rollback failed/); # Rollback failed
1055 deal_with_failed_transaction();
1058 Nested transactions will work as expected. That is, only the outermost
1059 transaction will actually issue a commit to the $dbh, and a rollback
1060 at any level of any transaction will cause the entire nested
1061 transaction to fail. Support for savepoints and for true nested
1062 transactions (for databases that support them) will hopefully be added
1067 =head2 Creating Schemas From An Existing Database
1069 L<DBIx::Class::Schema::Loader> will connect to a database and create a
1070 L<DBIx::Class::Schema> and associated sources by examining the database.
1072 The recommend way of achieving this is to use the
1073 L<make_schema_at|DBIx::Class::Schema::Loader/make_schema_at> method:
1075 perl -MDBIx::Class::Schema::Loader=make_schema_at,dump_to_dir:./lib \
1076 -e 'make_schema_at("My::Schema", { debug => 1 }, [ "dbi:Pg:dbname=foo","postgres" ])'
1078 This will create a tree of files rooted at C<./lib/My/Schema/> containing
1079 source definitions for all the tables found in the C<foo> database.
1081 =head2 Creating DDL SQL
1083 The following functionality requires you to have L<SQL::Translator>
1084 (also known as "SQL Fairy") installed.
1086 To create a set of database-specific .sql files for the above schema:
1088 my $schema = My::Schema->connect($dsn);
1089 $schema->create_ddl_dir(['MySQL', 'SQLite', 'PostgreSQL'],
1094 By default this will create schema files in the current directory, for
1095 MySQL, SQLite and PostgreSQL, using the $VERSION from your Schema.pm.
1097 To create a new database using the schema:
1099 my $schema = My::Schema->connect($dsn);
1100 $schema->deploy({ add_drop_tables => 1});
1102 To import created .sql files using the mysql client:
1104 mysql -h "host" -D "database" -u "user" -p < My_Schema_1.0_MySQL.sql
1106 To create C<ALTER TABLE> conversion scripts to update a database to a
1107 newer version of your schema at a later point, first set a new
1108 C<$VERSION> in your Schema file, then:
1110 my $schema = My::Schema->connect($dsn);
1111 $schema->create_ddl_dir(['MySQL', 'SQLite', 'PostgreSQL'],
1117 This will produce new database-specific .sql files for the new version
1118 of the schema, plus scripts to convert from version 0.1 to 0.2. This
1119 requires that the files for 0.1 as created above are available in the
1120 given directory to diff against.
1122 =head2 Select from dual
1124 Dummy tables are needed by some databases to allow calling functions
1125 or expressions that aren't based on table content, for examples of how
1126 this applies to various database types, see:
1127 L<http://troels.arvin.dk/db/rdbms/#other-dummy_table>.
1129 Note: If you're using Oracles dual table don't B<ever> do anything
1130 other than a select, if you CRUD on your dual table you *will* break
1133 Make a table class as you would for any other table
1135 package MyAppDB::Dual;
1138 use base 'DBIx::Class';
1139 __PACKAGE__->load_components("Core");
1140 __PACKAGE__->table("Dual");
1141 __PACKAGE__->add_columns(
1143 { data_type => "VARCHAR2", is_nullable => 0, size => 1 },
1146 Once you've loaded your table class select from it using C<select>
1147 and C<as> instead of C<columns>
1149 my $rs = $schema->resultset('Dual')->search(undef,
1150 { select => [ 'sydate' ],
1155 All you have to do now is be careful how you access your resultset, the below
1156 will not work because there is no column called 'now' in the Dual table class
1158 while (my $dual = $rs->next) {
1159 print $dual->now."\n";
1161 # Can't locate object method "now" via package "MyAppDB::Dual" at headshot.pl line 23.
1163 You could of course use 'dummy' in C<as> instead of 'now', or C<add_columns> to
1164 your Dual class for whatever you wanted to select from dual, but that's just
1165 silly, instead use C<get_column>
1167 while (my $dual = $rs->next) {
1168 print $dual->get_column('now')."\n";
1173 my $cursor = $rs->cursor;
1174 while (my @vals = $cursor->next) {
1175 print $vals[0]."\n";
1178 Or use L<DBIx::Class::ResultClass::HashRefInflator>
1180 $rs->result_class('DBIx::Class::ResultClass::HashRefInflator');
1181 while ( my $dual = $rs->next ) {
1182 print $dual->{now}."\n";
1185 Here are some example C<select> conditions to illustrate the different syntax
1186 you could use for doing stuff like
1187 C<oracles.heavily(nested(functions_can('take', 'lots'), OF), 'args')>
1189 # get a sequence value
1190 select => [ 'A_SEQ.nextval' ],
1192 # get create table sql
1193 select => [ { 'dbms_metadata.get_ddl' => [ "'TABLE'", "'ARTIST'" ]} ],
1195 # get a random num between 0 and 100
1196 select => [ { "trunc" => [ { "dbms_random.value" => [0,100] } ]} ],
1199 select => [ { 'extract' => [ \'year from sysdate' ] } ],
1202 select => [ {'round' => [{'cos' => [ \'180 * 3.14159265359/180' ]}]}],
1204 # which day of the week were you born on?
1205 select => [{'to_char' => [{'to_date' => [ "'25-DEC-1980'", "'dd-mon-yyyy'" ]}, "'day'"]}],
1207 # select 16 rows from dual
1208 select => [ "'hello'" ],
1210 group_by => [ 'cube( 1, 2, 3, 4 )' ],
1214 =head2 Adding Indexes And Functions To Your SQL
1216 Often you will want indexes on columns on your table to speed up searching. To
1217 do this, create a method called C<sqlt_deploy_hook> in the relevant source
1218 class (refer to the advanced
1219 L<callback system|DBIx::Class::ResultSource/sqlt_deploy_callback> if you wish
1220 to share a hook between multiple sources):
1222 package My::Schema::Artist;
1224 __PACKAGE__->table('artist');
1225 __PACKAGE__->add_columns(id => { ... }, name => { ... })
1227 sub sqlt_deploy_hook {
1228 my ($self, $sqlt_table) = @_;
1230 $sqlt_table->add_index(name => 'idx_name', fields => ['name']);
1235 Sometimes you might want to change the index depending on the type of the
1236 database for which SQL is being generated:
1238 my ($db_type = $sqlt_table->schema->translator->producer_type)
1239 =~ s/^SQL::Translator::Producer:://;
1241 You can also add hooks to the schema level to stop certain tables being
1248 sub sqlt_deploy_hook {
1249 my ($self, $sqlt_schema) = @_;
1251 $sqlt_schema->drop_table('table_name');
1254 You could also add views, procedures or triggers to the output using
1255 L<SQL::Translator::Schema/add_view>,
1256 L<SQL::Translator::Schema/add_procedure> or
1257 L<SQL::Translator::Schema/add_trigger>.
1260 =head2 Schema versioning
1262 The following example shows simplistically how you might use DBIx::Class to
1263 deploy versioned schemas to your customers. The basic process is as follows:
1269 Create a DBIx::Class schema
1281 Modify schema to change functionality
1285 Deploy update to customers
1289 B<Create a DBIx::Class schema>
1291 This can either be done manually, or generated from an existing database as
1292 described under L</Creating Schemas From An Existing Database>
1296 Call L<DBIx::Class::Schema/create_ddl_dir> as above under L</Creating DDL SQL>.
1298 B<Deploy to customers>
1300 There are several ways you could deploy your schema. These are probably
1301 beyond the scope of this recipe, but might include:
1307 Require customer to apply manually using their RDBMS.
1311 Package along with your app, making database dump/schema update/tests
1312 all part of your install.
1316 B<Modify the schema to change functionality>
1318 As your application evolves, it may be necessary to modify your schema
1319 to change functionality. Once the changes are made to your schema in
1320 DBIx::Class, export the modified schema and the conversion scripts as
1321 in L</Creating DDL SQL>.
1323 B<Deploy update to customers>
1325 Add the L<DBIx::Class::Schema::Versioned> schema component to your
1326 Schema class. This will add a new table to your database called
1327 C<dbix_class_schema_vesion> which will keep track of which version is installed
1328 and warn if the user trys to run a newer schema version than the
1329 database thinks it has.
1331 Alternatively, you can send the conversion sql scripts to your
1334 =head2 Setting quoting for the generated SQL.
1336 If the database contains column names with spaces and/or reserved words, they
1337 need to be quoted in the SQL queries. This is done using:
1339 __PACKAGE__->storage->sql_maker->quote_char([ qw/[ ]/] );
1340 __PACKAGE__->storage->sql_maker->name_sep('.');
1342 The first sets the quote characters. Either a pair of matching
1343 brackets, or a C<"> or C<'>:
1345 __PACKAGE__->storage->sql_maker->quote_char('"');
1347 Check the documentation of your database for the correct quote
1348 characters to use. C<name_sep> needs to be set to allow the SQL
1349 generator to put the quotes the correct place.
1351 In most cases you should set these as part of the arguments passed to
1352 L<DBIx::Class::Schema/connect>:
1354 my $schema = My::Schema->connect(
1364 =head2 Setting limit dialect for SQL::Abstract::Limit
1366 In some cases, SQL::Abstract::Limit cannot determine the dialect of
1367 the remote SQL server by looking at the database handle. This is a
1368 common problem when using the DBD::JDBC, since the DBD-driver only
1369 know that in has a Java-driver available, not which JDBC driver the
1370 Java component has loaded. This specifically sets the limit_dialect
1371 to Microsoft SQL-server (See more names in SQL::Abstract::Limit
1374 __PACKAGE__->storage->sql_maker->limit_dialect('mssql');
1376 The JDBC bridge is one way of getting access to a MSSQL server from a platform
1377 that Microsoft doesn't deliver native client libraries for. (e.g. Linux)
1379 The limit dialect can also be set at connect time by specifying a
1380 C<limit_dialect> key in the final hash as shown above.
1382 =head2 Working with PostgreSQL array types
1384 You can also assign values to PostgreSQL array columns by passing array
1385 references in the C<\%columns> (C<\%vals>) hashref of the
1386 L<DBIx::Class::ResultSet/create> and L<DBIx::Class::Row/update> family of
1389 $resultset->create({
1390 numbers => [1, 2, 3]
1395 numbers => [1, 2, 3]
1402 In conditions (eg. C<\%cond> in the L<DBIx::Class::ResultSet/search> family of
1403 methods) you cannot directly use array references (since this is interpreted as
1404 a list of values to be C<OR>ed), but you can use the following syntax to force
1405 passing them as bind values:
1409 numbers => \[ '= ?', [numbers => [1, 2, 3]] ]
1413 See L<SQL::Abstract/array_datatypes> and L<SQL::Abstract/Literal SQL with
1414 placeholders and bind values (subqueries)> for more explanation. Note that
1415 L<DBIx::Class> sets L<SQL::Abstract/bindtype> to C<columns>, so you must pass
1416 the bind values (the C<[1, 2, 3]> arrayref in the above example) wrapped in
1417 arrayrefs together with the column name, like this: C<< [column_name => value]
1420 =head1 BOOTSTRAPPING/MIGRATING
1422 =head2 Easy migration from class-based to schema-based setup
1424 You want to start using the schema-based approach to L<DBIx::Class>
1425 (see L<SchemaIntro.pod>), but have an established class-based setup with lots
1426 of existing classes that you don't want to move by hand. Try this nifty script
1430 use SQL::Translator;
1432 my $schema = MyDB->schema_instance;
1434 my $translator = SQL::Translator->new(
1435 debug => $debug || 0,
1436 trace => $trace || 0,
1437 no_comments => $no_comments || 0,
1438 show_warnings => $show_warnings || 0,
1439 add_drop_table => $add_drop_table || 0,
1440 validate => $validate || 0,
1442 'DBIx::Schema' => $schema,
1445 'prefix' => 'My::Schema',
1449 $translator->parser('SQL::Translator::Parser::DBIx::Class');
1450 $translator->producer('SQL::Translator::Producer::DBIx::Class::File');
1452 my $output = $translator->translate(@args) or die
1453 "Error: " . $translator->error;
1457 You could use L<Module::Find> to search for all subclasses in the MyDB::*
1458 namespace, which is currently left as an exercise for the reader.
1460 =head1 OVERLOADING METHODS
1462 L<DBIx::Class> uses the L<Class::C3> package, which provides for redispatch of
1463 method calls, useful for things like default values and triggers. You have to
1464 use calls to C<next::method> to overload methods. More information on using
1465 L<Class::C3> with L<DBIx::Class> can be found in
1466 L<DBIx::Class::Manual::Component>.
1468 =head2 Setting default values for a row
1470 It's as simple as overriding the C<new> method. Note the use of
1474 my ( $class, $attrs ) = @_;
1476 $attrs->{foo} = 'bar' unless defined $attrs->{foo};
1478 my $new = $class->next::method($attrs);
1483 For more information about C<next::method>, look in the L<Class::C3>
1484 documentation. See also L<DBIx::Class::Manual::Component> for more
1485 ways to write your own base classes to do this.
1487 People looking for ways to do "triggers" with DBIx::Class are probably
1488 just looking for this.
1490 =head2 Changing one field whenever another changes
1492 For example, say that you have three columns, C<id>, C<number>, and
1493 C<squared>. You would like to make changes to C<number> and have
1494 C<squared> be automagically set to the value of C<number> squared.
1495 You can accomplish this by overriding C<store_column>:
1498 my ( $self, $name, $value ) = @_;
1499 if ($name eq 'number') {
1500 $self->squared($value * $value);
1502 $self->next::method($name, $value);
1505 Note that the hard work is done by the call to C<next::method>, which
1506 redispatches your call to store_column in the superclass(es).
1508 =head2 Automatically creating related objects
1510 You might have a class C<Artist> which has many C<CD>s. Further, if you
1511 want to create a C<CD> object every time you insert an C<Artist> object.
1512 You can accomplish this by overriding C<insert> on your objects:
1515 my ( $self, @args ) = @_;
1516 $self->next::method(@args);
1517 $self->cds->new({})->fill_from_artist($self)->insert;
1521 where C<fill_from_artist> is a method you specify in C<CD> which sets
1522 values in C<CD> based on the data in the C<Artist> object you pass in.
1524 =head2 Wrapping/overloading a column accessor
1528 Say you have a table "Camera" and want to associate a description
1529 with each camera. For most cameras, you'll be able to generate the description from
1530 the other columns. However, in a few special cases you may want to associate a
1531 custom description with a camera.
1535 In your database schema, define a description field in the "Camera" table that
1536 can contain text and null values.
1538 In DBIC, we'll overload the column accessor to provide a sane default if no
1539 custom description is defined. The accessor will either return or generate the
1540 description, depending on whether the field is null or not.
1542 First, in your "Camera" schema class, define the description field as follows:
1544 __PACKAGE__->add_columns(description => { accessor => '_description' });
1546 Next, we'll define the accessor-wrapper subroutine:
1551 # If there is an update to the column, we'll let the original accessor
1553 return $self->_description(@_) if @_;
1555 # Fetch the column value.
1556 my $description = $self->_description;
1558 # If there's something in the description field, then just return that.
1559 return $description if defined $description && length $descripton;
1561 # Otherwise, generate a description.
1562 return $self->generate_description;
1565 =head1 DEBUGGING AND PROFILING
1567 =head2 DBIx::Class objects with Data::Dumper
1569 L<Data::Dumper> can be a very useful tool for debugging, but sometimes it can
1570 be hard to find the pertinent data in all the data it can generate.
1571 Specifically, if one naively tries to use it like so,
1575 my $cd = $schema->resultset('CD')->find(1);
1578 several pages worth of data from the CD object's schema and result source will
1579 be dumped to the screen. Since usually one is only interested in a few column
1580 values of the object, this is not very helpful.
1582 Luckily, it is possible to modify the data before L<Data::Dumper> outputs
1583 it. Simply define a hook that L<Data::Dumper> will call on the object before
1584 dumping it. For example,
1591 result_source => undef,
1599 local $Data::Dumper::Freezer = '_dumper_hook';
1601 my $cd = $schema->resultset('CD')->find(1);
1603 # dumps $cd without its ResultSource
1605 If the structure of your schema is such that there is a common base class for
1606 all your table classes, simply put a method similar to C<_dumper_hook> in the
1607 base class and set C<$Data::Dumper::Freezer> to its name and L<Data::Dumper>
1608 will automagically clean up your data before printing it. See
1609 L<Data::Dumper/EXAMPLES> for more information.
1613 When you enable L<DBIx::Class::Storage>'s debugging it prints the SQL
1614 executed as well as notifications of query completion and transaction
1615 begin/commit. If you'd like to profile the SQL you can subclass the
1616 L<DBIx::Class::Storage::Statistics> class and write your own profiling
1619 package My::Profiler;
1622 use base 'DBIx::Class::Storage::Statistics';
1624 use Time::HiRes qw(time);
1633 $self->print("Executing $sql: ".join(', ', @params)."\n");
1642 my $elapsed = sprintf("%0.4f", time() - $start);
1643 $self->print("Execution took $elapsed seconds.\n");
1649 You can then install that class as the debugging object:
1651 __PACKAGE__->storage->debugobj(new My::Profiler());
1652 __PACKAGE__->storage->debug(1);
1654 A more complicated example might involve storing each execution of SQL in an
1662 my $elapsed = time() - $start;
1663 push(@{ $calls{$sql} }, {
1669 You could then create average, high and low execution times for an SQL
1670 statement and dig down to see if certain parameters cause aberrant behavior.
1671 You might want to check out L<DBIx::Class::QueryLog> as well.
1673 =head1 STARTUP SPEED
1675 L<DBIx::Class|DBIx::Class> programs can have a significant startup delay
1676 as the ORM loads all the relevant classes. This section examines
1677 techniques for reducing the startup delay.
1679 These tips are are listed in order of decreasing effectiveness - so the
1680 first tip, if applicable, should have the greatest effect on your
1683 =head2 Statically Define Your Schema
1686 L<DBIx::Class::Schema::Loader|DBIx::Class::Schema::Loader> to build the
1687 classes dynamically based on the database schema then there will be a
1688 significant startup delay.
1690 For production use a statically defined schema (which can be generated
1691 using L<DBIx::Class::Schema::Loader|DBIx::Class::Schema::Loader> to dump
1692 the database schema once - see
1693 L<make_schema_at|DBIx::Class::Schema::Loader/make_schema_at> and
1694 L<dump_directory|DBIx::Class::Schema::Loader/dump_directory> for more
1695 details on creating static schemas from a database).
1697 =head2 Move Common Startup into a Base Class
1699 Typically L<DBIx::Class> result classes start off with
1701 use base qw/DBIx::Class/;
1702 __PACKAGE__->load_components(qw/InflateColumn::DateTime Core/);
1704 If this preamble is moved into a common base class:-
1708 use base qw/DBIx::Class/;
1709 __PACKAGE__->load_components(qw/InflateColumn::DateTime Core/);
1712 and each result class then uses this as a base:-
1714 use base qw/MyDBICbase/;
1716 then the load_components is only performed once, which can result in a
1717 considerable startup speedup for schemas with many classes.
1719 =head2 Explicitly List Schema Result Classes
1721 The schema class will normally contain
1723 __PACKAGE__->load_classes();
1725 to load the result classes. This will use L<Module::Find|Module::Find>
1726 to find and load the appropriate modules. Explicitly defining the
1727 classes you wish to load will remove the overhead of
1728 L<Module::Find|Module::Find> and the related directory operations:-
1730 __PACKAGE__->load_classes(qw/ CD Artist Track /);
1732 If you are instead using the L<load_namespaces|DBIx::Class::Schema/load_namespaces>
1733 syntax to load the appropriate classes there is not a direct alternative
1734 avoiding L<Module::Find|Module::Find>.