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 return $rs->page(2); # records for page 2
24 You can get a L<Data::Page> object for the resultset (suitable for use
25 in e.g. a template) using the C<pager> method:
29 =head2 Complex WHERE clauses
31 Sometimes you need to formulate a query using specific operators:
33 my @albums = $schema->resultset('Album')->search({
34 artist => { 'like', '%Lamb%' },
35 title => { 'like', '%Fear of Fours%' },
38 This results in something like the following C<WHERE> clause:
40 WHERE artist LIKE ? AND title LIKE ?
42 And the following bind values for the placeholders: C<'%Lamb%'>, C<'%Fear of
45 Other queries might require slightly more complex logic:
47 my @albums = $schema->resultset('Album')->search({
50 artist => { 'like', '%Smashing Pumpkins%' },
51 title => 'Siamese Dream',
53 artist => 'Starchildren',
57 This results in the following C<WHERE> clause:
59 WHERE ( artist LIKE '%Smashing Pumpkins%' AND title = 'Siamese Dream' )
60 OR artist = 'Starchildren'
62 For more information on generating complex queries, see
63 L<SQL::Abstract/WHERE CLAUSES>.
65 =head2 Retrieve one and only one row from a resultset
67 Sometimes you need only the first "top" row of a resultset. While this
68 can be easily done with L<< $rs->first|DBIx::Class::ResultSet/first
69 >>, it is suboptimal, as a full blown cursor for the resultset will be
70 created and then immediately destroyed after fetching the first row
71 object. L<< $rs->single|DBIx::Class::ResultSet/single >> is designed
72 specifically for this case - it will grab the first returned result
73 without even instantiating a cursor.
75 Before replacing all your calls to C<first()> with C<single()> please observe the
82 While single() takes a search condition just like search() does, it does
83 _not_ accept search attributes. However one can always chain a single() to
86 my $top_cd = $cd_rs->search({}, { order_by => 'rating' })->single;
91 Since single() is the engine behind find(), it is designed to fetch a
92 single row per database query. Thus a warning will be issued when the
93 underlying SELECT returns more than one row. Sometimes however this usage
94 is valid: i.e. we have an arbitrary number of cd's but only one of them is
95 at the top of the charts at any given time. If you know what you are doing,
96 you can silence the warning by explicitly limiting the resultset size:
98 my $top_cd = $cd_rs->search ({}, { order_by => 'rating', rows => 1 })->single;
102 =head2 Arbitrary SQL through a custom ResultSource
104 Sometimes you have to run arbitrary SQL because your query is too complex
105 (e.g. it contains Unions, Sub-Selects, Stored Procedures, etc.) or has to
106 be optimized for your database in a special way, but you still want to
107 get the results as a L<DBIx::Class::ResultSet>.
109 This is accomplished by defining a
110 L<ResultSource::View|DBIx::Class::ResultSource::View> for your query,
111 almost like you would define a regular ResultSource.
113 package My::Schema::Result::UserFriendsComplex;
116 use base qw/DBIx::Class::Core/;
118 __PACKAGE__->table_class('DBIx::Class::ResultSource::View');
120 # For the time being this is necessary even for virtual views
121 __PACKAGE__->table($view_name);
124 # ->add_columns, etc.
127 # do not attempt to deploy() this view
128 __PACKAGE__->result_source_instance->is_virtual(1);
130 __PACKAGE__->result_source_instance->view_definition(q[
131 SELECT u.* FROM user u
132 INNER JOIN user_friends f ON u.id = f.user_id
133 WHERE f.friend_user_id = ?
135 SELECT u.* FROM user u
136 INNER JOIN user_friends f ON u.id = f.friend_user_id
140 Next, you can execute your complex query using bind parameters like this:
142 my $friends = $schema->resultset( 'UserFriendsComplex' )->search( {},
144 bind => [ 12345, 12345 ]
148 ... and you'll get back a perfect L<DBIx::Class::ResultSet> (except, of course,
149 that you cannot modify the rows it contains, e.g. cannot call L</update>,
150 L</delete>, ... on it).
152 Note that you cannot have bind parameters unless is_virtual is set to true.
158 If you're using the old deprecated C<< $rsrc_instance->name(\'( SELECT ...') >>
159 method for custom SQL execution, you are highly encouraged to update your code
160 to use a virtual view as above. If you do not want to change your code, and just
161 want to suppress the deprecation warning when you call
162 L<DBIx::Class::Schema/deploy>, add this line to your source definition, so that
163 C<deploy> will exclude this "table":
165 sub sqlt_deploy_hook { $_[1]->schema->drop_table ($_[1]) }
169 =head2 Using specific columns
171 When you only want specific columns from a table, you can use
172 C<columns> to specify which ones you need. This is useful to avoid
173 loading columns with large amounts of data that you aren't about to
176 my $rs = $schema->resultset('Artist')->search(
179 columns => [qw/ name /]
184 # SELECT artist.name FROM artist
186 This is a shortcut for C<select> and C<as>, see below. C<columns>
187 cannot be used together with C<select> and C<as>.
189 =head2 Using database functions or stored procedures
191 The combination of C<select> and C<as> can be used to return the result of a
192 database function or stored procedure as a column value. You use C<select> to
193 specify the source for your column value (e.g. a column name, function, or
194 stored procedure name). You then use C<as> to set the column name you will use
195 to access the returned value:
197 my $rs = $schema->resultset('Artist')->search(
200 select => [ 'name', { LENGTH => 'name' } ],
201 as => [qw/ name name_length /],
206 # SELECT name name, LENGTH( name )
209 Note that the C<as> attribute B<has absolutely nothing to do> with the SQL
210 syntax C< SELECT foo AS bar > (see the documentation in
211 L<DBIx::Class::ResultSet/ATTRIBUTES>). You can control the C<AS> part of the
212 generated SQL via the C<-as> field attribute as follows:
214 my $rs = $schema->resultset('Artist')->search(
219 '+select' => [ { count => 'cds.cdid', -as => 'amount_of_cds' } ],
220 '+as' => [qw/num_cds/],
221 order_by => { -desc => 'amount_of_cds' },
226 # SELECT me.artistid, me.name, me.rank, me.charfield, COUNT( cds.cdid ) AS amount_of_cds
227 # FROM artist me LEFT JOIN cd cds ON cds.artist = me.artistid
228 # GROUP BY me.artistid, me.name, me.rank, me.charfield
229 # ORDER BY amount_of_cds DESC
232 If your alias exists as a column in your base class (i.e. it was added with
233 L<add_columns|DBIx::Class::ResultSource/add_columns>), you just access it as
234 normal. Our C<Artist> class has a C<name> column, so we just use the C<name>
237 my $artist = $rs->first();
238 my $name = $artist->name();
240 If on the other hand the alias does not correspond to an existing column, you
241 have to fetch the value using the C<get_column> accessor:
243 my $name_length = $artist->get_column('name_length');
245 If you don't like using C<get_column>, you can always create an accessor for
246 any of your aliases using either of these:
248 # Define accessor manually:
249 sub name_length { shift->get_column('name_length'); }
251 # Or use DBIx::Class::AccessorGroup:
252 __PACKAGE__->mk_group_accessors('column' => 'name_length');
254 See also L</Using SQL functions on the left hand side of a comparison>.
256 =head2 SELECT DISTINCT with multiple columns
258 my $rs = $schema->resultset('Artist')->search(
261 columns => [ qw/artist_id name rank/ ],
266 my $rs = $schema->resultset('Artist')->search(
269 columns => [ qw/artist_id name rank/ ],
270 group_by => [ qw/artist_id name rank/ ],
275 # SELECT me.artist_id, me.name, me.rank
277 # GROUP BY artist_id, name, rank
279 =head2 SELECT COUNT(DISTINCT colname)
281 my $rs = $schema->resultset('Artist')->search(
284 columns => [ qw/name/ ],
289 my $rs = $schema->resultset('Artist')->search(
292 columns => [ qw/name/ ],
293 group_by => [ qw/name/ ],
297 my $count = $rs->count;
300 # SELECT COUNT( * ) FROM (SELECT me.name FROM artist me GROUP BY me.name) me:
302 =head2 Grouping results
304 L<DBIx::Class> supports C<GROUP BY> as follows:
306 my $rs = $schema->resultset('Artist')->search(
310 select => [ 'name', { count => 'cds.id' } ],
311 as => [qw/ name cd_count /],
312 group_by => [qw/ name /]
317 # SELECT name, COUNT( cd.id ) FROM artist
318 # LEFT JOIN cd ON artist.id = cd.artist
321 Please see L<DBIx::Class::ResultSet/ATTRIBUTES> documentation if you
322 are in any way unsure about the use of the attributes above (C< join
323 >, C< select >, C< as > and C< group_by >).
327 You can write subqueries relatively easily in DBIC.
329 my $inside_rs = $schema->resultset('Artist')->search({
330 name => [ 'Billy Joel', 'Brittany Spears' ],
333 my $rs = $schema->resultset('CD')->search({
334 artist_id => { -in => $inside_rs->get_column('id')->as_query },
337 The usual operators ( '=', '!=', -in, -not_in, etc.) are supported.
339 B<NOTE>: You have to explicitly use '=' when doing an equality comparison.
340 The following will B<not> work:
342 my $rs = $schema->resultset('CD')->search({
343 artist_id => $inside_rs->get_column('id')->as_query, # does NOT work
348 Subqueries are supported in the where clause (first hashref), and in the
349 from, select, and +select attributes.
351 =head3 Correlated subqueries
353 my $cdrs = $schema->resultset('CD');
354 my $rs = $cdrs->search({
356 '=' => $cdrs->search(
357 { artist_id => { -ident => 'me.artist_id' } },
358 { alias => 'sub_query' }
359 )->get_column('year')->max_rs->as_query,
363 That creates the following SQL:
365 SELECT me.cdid, me.artist, me.title, me.year, me.genreid, me.single_track
368 SELECT MAX(sub_query.year)
370 WHERE artist_id = me.artist_id
373 =head2 Predefined searches
375 You can define frequently used searches as methods by subclassing
376 L<DBIx::Class::ResultSet>:
378 package My::DBIC::ResultSet::CD;
381 use base 'DBIx::Class::ResultSet';
383 sub search_cds_ordered {
386 return $self->search(
388 { order_by => 'name DESC' },
394 If you're using L<DBIx::Class::Schema/load_namespaces>, simply place the file
395 into the C<ResultSet> directory next to your C<Result> directory, and it will
396 be automatically loaded.
398 If however you are still using L<DBIx::Class::Schema/load_classes>, first tell
399 DBIx::Class to create an instance of the ResultSet class for you, in your
400 My::DBIC::Schema::CD class:
402 # class definition as normal
403 use base 'DBIx::Class::Core';
404 __PACKAGE__->table('cd');
406 # tell DBIC to use the custom ResultSet class
407 __PACKAGE__->resultset_class('My::DBIC::ResultSet::CD');
409 Note that C<resultset_class> must be called after C<load_components> and C<table>, or you will get errors about missing methods.
411 Then call your new method in your code:
413 my $ordered_cds = $schema->resultset('CD')->search_cds_ordered();
415 =head2 Using SQL functions on the left hand side of a comparison
417 Using SQL functions on the left hand side of a comparison is generally not a
418 good idea since it requires a scan of the entire table. (Unless your RDBMS
419 supports indexes on expressions - including return values of functions - and
420 you create an index on the return value of the function in question.) However,
421 it can be accomplished with C<DBIx::Class> when necessary by resorting to
425 \[ 'YEAR(date_of_birth) = ?', 1979 ]
429 # SELECT * FROM employee WHERE YEAR(date_of_birth) = ?
431 $rs->search({ -and => [
433 \[ 'YEAR(date_of_birth) = ?', 1979 ]
437 # SELECT * FROM employee WHERE name = ? AND YEAR(date_of_birth) = ?
439 Note: the syntax for specifying the bind value's datatype and value is
440 explained in L<DBIx::Class::ResultSet/DBIC BIND VALUES>.
442 See also L<SQL::Abstract/Literal SQL with placeholders and bind values
445 =head2 Software Limits
447 When your RDBMS does not have a working SQL limit mechanism (e.g. Sybase ASE)
448 and L<GenericSubQ|SQL::Abstract::Limit/GenericSubQ> is either too slow or does
449 not work at all, you can try the
450 L<software_limit|DBIx::Class::ResultSet/software_limit>
451 L<DBIx::Class::ResultSet> attribute, which skips over records to simulate limits
456 my $paged_rs = $rs->search({}, {
459 order_by => [ 'me.last_name' ],
463 You can set it as a default for your schema by placing the following in your
466 __PACKAGE__->default_resultset_attributes({ software_limit => 1 });
468 B<WARNING:> If you are dealing with large resultsets and your L<DBI> or
469 ODBC/ADO driver does not have proper cursor support (i.e. it loads the whole
470 resultset into memory) then this feature will be extremely slow and use huge
471 amounts of memory at best, and may cause your process to run out of memory and
472 cause instability on your server at worst, beware!
474 =head1 JOINS AND PREFETCHING
476 =head2 Using joins and prefetch
478 You can use the C<join> attribute to allow searching on, or sorting your
479 results by, one or more columns in a related table.
481 This requires that you have defined the L<DBIx::Class::Relationship>. For example :
483 My::Schema::CD->has_many( artists => 'My::Schema::Artist', 'artist_id');
485 To return all CDs matching a particular artist name, you specify the name of the relationship ('artists'):
487 my $rs = $schema->resultset('CD')->search(
489 'artists.name' => 'Bob Marley'
492 join => 'artists', # join the artist table
497 # SELECT cd.* FROM cd
498 # JOIN artist ON cd.artist = artist.id
499 # WHERE artist.name = 'Bob Marley'
501 In that example both the join, and the condition use the relationship name rather than the table name
502 (see L<DBIx::Class::Manual::Joining> for more details on aliasing ).
504 If required, you can now sort on any column in the related tables by including
505 it in your C<order_by> attribute, (again using the aliased relation name rather than table name) :
507 my $rs = $schema->resultset('CD')->search(
509 'artists.name' => 'Bob Marley'
513 order_by => [qw/ artists.name /]
518 # SELECT cd.* FROM cd
519 # JOIN artist ON cd.artist = artist.id
520 # WHERE artist.name = 'Bob Marley'
521 # ORDER BY artist.name
523 Note that the C<join> attribute should only be used when you need to search or
524 sort using columns in a related table. Joining related tables when you only
525 need columns from the main table will make performance worse!
527 Now let's say you want to display a list of CDs, each with the name of the
528 artist. The following will work fine:
530 while (my $cd = $rs->next) {
531 print "CD: " . $cd->title . ", Artist: " . $cd->artist->name;
534 There is a problem however. We have searched both the C<cd> and C<artist> tables
535 in our main query, but we have only returned data from the C<cd> table. To get
536 the artist name for any of the CD objects returned, L<DBIx::Class> will go back
539 SELECT artist.* FROM artist WHERE artist.id = ?
541 A statement like the one above will run for each and every CD returned by our
542 main query. Five CDs, five extra queries. A hundred CDs, one hundred extra
545 Thankfully, L<DBIx::Class> has a C<prefetch> attribute to solve this problem.
546 This allows you to fetch results from related tables in advance:
548 my $rs = $schema->resultset('CD')->search(
550 'artists.name' => 'Bob Marley'
554 order_by => [qw/ artists.name /],
555 prefetch => 'artists' # return artist data too!
559 # Equivalent SQL (note SELECT from both "cd" and "artist"):
560 # SELECT cd.*, artist.* FROM cd
561 # JOIN artist ON cd.artist = artist.id
562 # WHERE artist.name = 'Bob Marley'
563 # ORDER BY artist.name
565 The code to print the CD list remains the same:
567 while (my $cd = $rs->next) {
568 print "CD: " . $cd->title . ", Artist: " . $cd->artist->name;
571 L<DBIx::Class> has now prefetched all matching data from the C<artist> table,
572 so no additional SQL statements are executed. You now have a much more
575 Also note that C<prefetch> should only be used when you know you will
576 definitely use data from a related table. Pre-fetching related tables when you
577 only need columns from the main table will make performance worse!
579 =head2 Multiple joins
581 In the examples above, the C<join> attribute was a scalar. If you
582 pass an array reference instead, you can join to multiple tables. In
583 this example, we want to limit the search further, using
586 # Relationships defined elsewhere:
587 # CD->belongs_to('artist' => 'Artist');
588 # CD->has_one('liner_notes' => 'LinerNotes', 'cd');
589 my $rs = $schema->resultset('CD')->search(
591 'artist.name' => 'Bob Marley'
592 'liner_notes.notes' => { 'like', '%some text%' },
595 join => [qw/ artist liner_notes /],
596 order_by => [qw/ artist.name /],
601 # SELECT cd.*, artist.*, liner_notes.* FROM cd
602 # JOIN artist ON cd.artist = artist.id
603 # JOIN liner_notes ON cd.id = liner_notes.cd
604 # WHERE artist.name = 'Bob Marley' AND liner_notes.notes LIKE '%some text%'
605 # ORDER BY artist.name
607 =head2 Multi-step joins
609 Sometimes you want to join more than one relationship deep. In this example,
610 we want to find all C<Artist> objects who have C<CD>s whose C<LinerNotes>
611 contain a specific string:
613 # Relationships defined elsewhere:
614 # Artist->has_many('cds' => 'CD', 'artist');
615 # CD->has_one('liner_notes' => 'LinerNotes', 'cd');
617 my $rs = $schema->resultset('Artist')->search(
619 'liner_notes.notes' => { 'like', '%some text%' },
623 'cds' => 'liner_notes'
629 # SELECT artist.* FROM artist
630 # LEFT JOIN cd ON artist.id = cd.artist
631 # LEFT JOIN liner_notes ON cd.id = liner_notes.cd
632 # WHERE liner_notes.notes LIKE '%some text%'
634 Joins can be nested to an arbitrary level. So if we decide later that we
635 want to reduce the number of Artists returned based on who wrote the liner
638 # Relationship defined elsewhere:
639 # LinerNotes->belongs_to('author' => 'Person');
641 my $rs = $schema->resultset('Artist')->search(
643 'liner_notes.notes' => { 'like', '%some text%' },
644 'author.name' => 'A. Writer'
649 'liner_notes' => 'author'
656 # SELECT artist.* FROM artist
657 # LEFT JOIN cd ON artist.id = cd.artist
658 # LEFT JOIN liner_notes ON cd.id = liner_notes.cd
659 # LEFT JOIN author ON author.id = liner_notes.author
660 # WHERE liner_notes.notes LIKE '%some text%'
661 # AND author.name = 'A. Writer'
663 =head2 Multi-step and multiple joins
665 With various combinations of array and hash references, you can join
666 tables in any combination you desire. For example, to join Artist to
667 CD and Concert, and join CD to LinerNotes:
669 # Relationships defined elsewhere:
670 # Artist->has_many('concerts' => 'Concert', 'artist');
672 my $rs = $schema->resultset('Artist')->search(
685 # SELECT artist.* FROM artist
686 # LEFT JOIN cd ON artist.id = cd.artist
687 # LEFT JOIN liner_notes ON cd.id = liner_notes.cd
688 # LEFT JOIN concert ON artist.id = concert.artist
690 =head2 Multi-step prefetch
692 C<prefetch> can be nested more than one relationship
693 deep using the same syntax as a multi-step join:
695 my $rs = $schema->resultset('Tag')->search(
705 # SELECT tag.*, cd.*, artist.* FROM tag
706 # JOIN cd ON tag.cd = cd.id
707 # JOIN artist ON cd.artist = artist.id
709 Now accessing our C<cd> and C<artist> relationships does not need additional
712 my $tag = $rs->first;
713 print $tag->cd->artist->name;
715 =head1 ROW-LEVEL OPERATIONS
717 =head2 Retrieving a result object's Schema
719 It is possible to get a Schema object from a result object like so:
721 my $schema = $cd->result_source->schema;
722 # use the schema as normal:
723 my $artist_rs = $schema->resultset('Artist');
725 This can be useful when you don't want to pass around a Schema object to every
728 =head2 Getting the value of the primary key for the last database insert
730 AKA getting last_insert_id
732 Thanks to the core component PK::Auto, this is straightforward:
734 my $foo = $rs->create(\%blah);
736 my $id = $foo->id; # foo->my_primary_key_field will also work.
738 If you are not using autoincrementing primary keys, this will probably
739 not work, but then you already know the value of the last primary key anyway.
741 =head2 Stringification
743 Employ the standard stringification technique by using the L<overload>
746 To make an object stringify itself as a single column, use something
747 like this (replace C<name> with the column/method of your choice):
749 use overload '""' => sub { shift->name}, fallback => 1;
751 For more complex stringification, you can use an anonymous subroutine:
753 use overload '""' => sub { $_[0]->name . ", " .
754 $_[0]->address }, fallback => 1;
756 =head3 Stringification Example
758 Suppose we have two tables: C<Product> and C<Category>. The table
761 Product(id, Description, category)
762 Category(id, Description)
764 C<category> is a foreign key into the Category table.
766 If you have a Product object C<$obj> and write something like
770 things will not work as expected.
772 To obtain, for example, the category description, you should add this
773 method to the class defining the Category table:
775 use overload "" => sub {
778 return $self->Description;
781 =head2 Want to know if find_or_create found or created a row?
783 Just use C<find_or_new> instead, then check C<in_storage>:
785 my $obj = $rs->find_or_new({ blah => 'blarg' });
786 unless ($obj->in_storage) {
788 # do whatever else you wanted if it was a new row
791 =head2 Static sub-classing DBIx::Class result classes
793 AKA adding additional relationships/methods/etc. to a model for a
794 specific usage of the (shared) model.
798 package My::App::Schema;
800 use base 'DBIx::Class::Schema';
802 # load subclassed classes from My::App::Schema::Result/ResultSet
803 __PACKAGE__->load_namespaces;
805 # load classes from shared model
807 'My::Shared::Model::Result' => [qw/
814 B<Result-Subclass definition>
816 package My::App::Schema::Result::Baz;
820 use base 'My::Shared::Model::Result::Baz';
822 # WARNING: Make sure you call table() again in your subclass,
823 # otherwise DBIx::Class::ResultSourceProxy::Table will not be called
824 # and the class name is not correctly registered as a source
825 __PACKAGE__->table('baz');
827 sub additional_method {
828 return "I'm an additional method only needed by this app";
833 =head2 Dynamic Sub-classing DBIx::Class proxy classes
835 AKA multi-class object inflation from one table
837 L<DBIx::Class> classes are proxy classes, therefore some different
838 techniques need to be employed for more than basic subclassing. In
839 this example we have a single user table that carries a boolean bit
840 for admin. We would like like to give the admin users
841 objects (L<DBIx::Class::Row>) the same methods as a regular user but
842 also special admin only methods. It doesn't make sense to create two
843 separate proxy-class files for this. We would be copying all the user
844 methods into the Admin class. There is a cleaner way to accomplish
847 Overriding the C<inflate_result> method within the User proxy-class
848 gives us the effect we want. This method is called by
849 L<DBIx::Class::ResultSet> when inflating a result from storage. So we
850 grab the object being returned, inspect the values we are looking for,
851 bless it if it's an admin object, and then return it. See the example
858 use base qw/DBIx::Class::Schema/;
860 __PACKAGE__->load_namespaces;
865 B<Proxy-Class definitions>
867 package My::Schema::Result::User;
871 use base qw/DBIx::Class::Core/;
873 ### Define what our admin class is, for ensure_class_loaded()
874 my $admin_class = __PACKAGE__ . '::Admin';
876 __PACKAGE__->table('users');
878 __PACKAGE__->add_columns(qw/user_id email password
879 firstname lastname active
882 __PACKAGE__->set_primary_key('user_id');
886 my $ret = $self->next::method(@_);
887 if( $ret->admin ) {### If this is an admin, rebless for extra functions
888 $self->ensure_class_loaded( $admin_class );
889 bless $ret, $admin_class;
895 print "I am a regular user.\n";
902 package My::Schema::Result::User::Admin;
906 use base qw/My::Schema::Result::User/;
908 # This line is important
909 __PACKAGE__->table('users');
913 print "I am an admin.\n";
919 print "I am doing admin stuff\n";
931 my $user_data = { email => 'someguy@place.com',
935 my $admin_data = { email => 'someadmin@adminplace.com',
939 my $schema = My::Schema->connection('dbi:Pg:dbname=test');
941 $schema->resultset('User')->create( $user_data );
942 $schema->resultset('User')->create( $admin_data );
944 ### Now we search for them
945 my $user = $schema->resultset('User')->single( $user_data );
946 my $admin = $schema->resultset('User')->single( $admin_data );
948 print ref $user, "\n";
949 print ref $admin, "\n";
951 print $user->password , "\n"; # pass1
952 print $admin->password , "\n";# pass2; inherited from User
953 print $user->hello , "\n";# I am a regular user.
954 print $admin->hello, "\n";# I am an admin.
956 ### The statement below will NOT print
957 print "I can do admin stuff\n" if $user->can('do_admin_stuff');
958 ### The statement below will print
959 print "I can do admin stuff\n" if $admin->can('do_admin_stuff');
961 Alternatively you can use L<DBIx::Class::DynamicSubclass> that implements
962 exactly the above functionality.
964 =head2 Skip result object creation for faster results
966 DBIx::Class is not built for speed, it's built for convenience and
967 ease of use, but sometimes you just need to get the data, and skip the
970 To do this simply use L<DBIx::Class::ResultClass::HashRefInflator>.
972 my $rs = $schema->resultset('CD');
974 $rs->result_class('DBIx::Class::ResultClass::HashRefInflator');
976 my $hash_ref = $rs->find(1);
980 Beware, changing the Result class using
981 L<DBIx::Class::ResultSet/result_class> will replace any existing class
982 completely including any special components loaded using
983 load_components, eg L<DBIx::Class::InflateColumn::DateTime>.
985 =head2 Get raw data for blindingly fast results
987 If the L<HashRefInflator|DBIx::Class::ResultClass::HashRefInflator> solution
988 above is not fast enough for you, you can use a DBIx::Class to return values
989 exactly as they come out of the database with none of the convenience methods
992 This is used like so:
994 my $cursor = $rs->cursor
995 while (my @vals = $cursor->next) {
996 # use $val[0..n] here
999 You will need to map the array offsets to particular columns (you can
1000 use the L<DBIx::Class::ResultSet/select> attribute of L<DBIx::Class::ResultSet/search> to force ordering).
1002 =head1 RESULTSET OPERATIONS
1004 =head2 Getting Schema from a ResultSet
1006 To get the L<DBIx::Class::Schema> object from a ResultSet, do the following:
1008 $rs->result_source->schema
1010 =head2 Getting Columns Of Data
1012 AKA Aggregating Data
1014 If you want to find the sum of a particular column there are several
1015 ways, the obvious one is to use search:
1017 my $rs = $schema->resultset('Items')->search(
1020 select => [ { sum => 'Cost' } ],
1021 as => [ 'total_cost' ], # remember this 'as' is for DBIx::Class::ResultSet not SQL
1024 my $tc = $rs->first->get_column('total_cost');
1026 Or, you can use the L<DBIx::Class::ResultSetColumn>, which gets
1027 returned when you ask the C<ResultSet> for a column using
1030 my $cost = $schema->resultset('Items')->get_column('Cost');
1031 my $tc = $cost->sum;
1033 With this you can also do:
1035 my $minvalue = $cost->min;
1036 my $maxvalue = $cost->max;
1038 Or just iterate through the values of this column only:
1040 while ( my $c = $cost->next ) {
1044 foreach my $c ($cost->all) {
1048 C<ResultSetColumn> only has a limited number of built-in functions. If
1049 you need one that it doesn't have, then you can use the C<func> method
1052 my $avg = $cost->func('AVERAGE');
1054 This will cause the following SQL statement to be run:
1056 SELECT AVERAGE(Cost) FROM Items me
1058 Which will of course only work if your database supports this function.
1059 See L<DBIx::Class::ResultSetColumn> for more documentation.
1061 =head2 Creating a result set from a set of rows
1063 Sometimes you have a (set of) result objects that you want to put into a
1064 resultset without the need to hit the DB again. You can do that by using the
1065 L<set_cache|DBIx::Class::Resultset/set_cache> method:
1067 my @uploadable_groups;
1068 while (my $group = $groups->next) {
1069 if ($group->can_upload($self)) {
1070 push @uploadable_groups, $group;
1073 my $new_rs = $self->result_source->resultset;
1074 $new_rs->set_cache(\@uploadable_groups);
1078 =head1 USING RELATIONSHIPS
1080 =head2 Create a new row in a related table
1082 my $author = $book->create_related('author', { name => 'Fred'});
1084 =head2 Search in a related table
1086 Only searches for books named 'Titanic' by the author in $author.
1088 my $books_rs = $author->search_related('books', { name => 'Titanic' });
1090 =head2 Delete data in a related table
1092 Deletes only the book named Titanic by the author in $author.
1094 $author->delete_related('books', { name => 'Titanic' });
1096 =head2 Ordering a relationship result set
1098 If you always want a relation to be ordered, you can specify this when you
1099 create the relationship.
1101 To order C<< $book->pages >> by descending page_number, create the relation
1104 __PACKAGE__->has_many('pages' => 'Page', 'book', { order_by => { -desc => 'page_number'} } );
1106 =head2 Filtering a relationship result set
1108 If you want to get a filtered result set, you can just add add to $attr as follows:
1110 __PACKAGE__->has_many('pages' => 'Page', 'book', { where => { scrap => 0 } } );
1112 =head2 Many-to-many relationship bridges
1114 This is straightforward using L<ManyToMany|DBIx::Class::Relationship/many_to_many>:
1117 use base 'DBIx::Class::Core';
1118 __PACKAGE__->table('user');
1119 __PACKAGE__->add_columns(qw/id name/);
1120 __PACKAGE__->set_primary_key('id');
1121 __PACKAGE__->has_many('user_address' => 'My::UserAddress', 'user');
1122 __PACKAGE__->many_to_many('addresses' => 'user_address', 'address');
1124 package My::UserAddress;
1125 use base 'DBIx::Class::Core';
1126 __PACKAGE__->table('user_address');
1127 __PACKAGE__->add_columns(qw/user address/);
1128 __PACKAGE__->set_primary_key(qw/user address/);
1129 __PACKAGE__->belongs_to('user' => 'My::User');
1130 __PACKAGE__->belongs_to('address' => 'My::Address');
1132 package My::Address;
1133 use base 'DBIx::Class::Core';
1134 __PACKAGE__->table('address');
1135 __PACKAGE__->add_columns(qw/id street town area_code country/);
1136 __PACKAGE__->set_primary_key('id');
1137 __PACKAGE__->has_many('user_address' => 'My::UserAddress', 'address');
1138 __PACKAGE__->many_to_many('users' => 'user_address', 'user');
1140 $rs = $user->addresses(); # get all addresses for a user
1141 $rs = $address->users(); # get all users for an address
1143 my $address = $user->add_to_addresses( # returns a My::Address instance,
1144 # NOT a My::UserAddress instance!
1146 country => 'United Kingdom',
1153 =head2 Relationships across DB schemas
1155 Mapping relationships across L<DB schemas|DBIx::Class::Manual::Glossary/DB schema>
1156 is easy as long as the schemas themselves are all accessible via the same DBI
1157 connection. In most cases, this means that they are on the same database host
1158 as each other and your connecting database user has the proper permissions to them.
1160 To accomplish this one only needs to specify the DB schema name in the table
1161 declaration, like so...
1163 package MyDatabase::Main::Artist;
1164 use base qw/DBIx::Class::Core/;
1166 __PACKAGE__->table('database1.artist'); # will use "database1.artist" in FROM clause
1168 __PACKAGE__->add_columns(qw/ artist_id name /);
1169 __PACKAGE__->set_primary_key('artist_id');
1170 __PACKAGE__->has_many('cds' => 'MyDatabase::Main::Cd');
1174 Whatever string you specify there will be used to build the "FROM" clause in SQL
1177 The big drawback to this is you now have DB schema names hardcoded in your
1178 class files. This becomes especially troublesome if you have multiple instances
1179 of your application to support a change lifecycle (e.g. DEV, TEST, PROD) and
1180 the DB schemas are named based on the environment (e.g. database1_dev).
1182 However, one can dynamically "map" to the proper DB schema by overriding the
1183 L<connection|DBIx::Class::Schama/connection> method in your Schema class and
1184 building a renaming facility, like so:
1186 package MyDatabase::Schema;
1189 extends 'DBIx::Class::Schema';
1191 around connection => sub {
1192 my ( $inner, $self, $dsn, $username, $pass, $attr ) = ( shift, @_ );
1194 my $postfix = delete $attr->{schema_name_postfix};
1199 $self->append_db_name($postfix);
1203 sub append_db_name {
1204 my ( $self, $postfix ) = @_;
1208 { $_->name =~ /^\w+\./mx }
1210 { $self->source($_) }
1213 foreach my $source (@sources_with_db) {
1214 my $name = $source->name;
1215 $name =~ s{^(\w+)\.}{${1}${postfix}\.}mx;
1217 $source->name($name);
1223 By overridding the L<connection|DBIx::Class::Schama/connection>
1224 method and extracting a custom option from the provided \%attr hashref one can
1225 then simply iterate over all the Schema's ResultSources, renaming them as
1228 To use this facility, simply add or modify the \%attr hashref that is passed to
1229 L<connection|DBIx::Class::Schama/connect>, as follows:
1232 = MyDatabase::Schema->connect(
1237 schema_name_postfix => '_dev'
1238 # ... Other options as desired ...
1241 Obviously, one could accomplish even more advanced mapping via a hash map or a
1246 =head2 Transactions with txn_do
1248 As of version 0.04001, there is improved transaction support in
1249 L<DBIx::Class::Storage> and L<DBIx::Class::Schema>. Here is an
1250 example of the recommended way to use it:
1252 my $genus = $schema->resultset('Genus')->find(12);
1254 my $coderef2 = sub {
1259 my $coderef1 = sub {
1260 $genus->add_to_species({ name => 'troglodyte' });
1263 $schema->txn_do($coderef2); # Can have a nested transaction. Only the outer will actualy commit
1264 return $genus->species;
1270 $rs = $schema->txn_do($coderef1);
1272 # Transaction failed
1273 die "the sky is falling!" #
1274 if ($_ =~ /Rollback failed/); # Rollback failed
1276 deal_with_failed_transaction();
1279 Note: by default C<txn_do> will re-run the coderef one more time if an
1280 error occurs due to client disconnection (e.g. the server is bounced).
1281 You need to make sure that your coderef can be invoked multiple times
1282 without terrible side effects.
1284 Nested transactions will work as expected. That is, only the outermost
1285 transaction will actually issue a commit to the $dbh, and a rollback
1286 at any level of any transaction will cause the entire nested
1287 transaction to fail.
1289 =head2 Nested transactions and auto-savepoints
1291 If savepoints are supported by your RDBMS, it is possible to achieve true
1292 nested transactions with minimal effort. To enable auto-savepoints via nested
1293 transactions, supply the C<< auto_savepoint = 1 >> connection attribute.
1295 Here is an example of true nested transactions. In the example, we start a big
1296 task which will create several rows. Generation of data for each row is a
1297 fragile operation and might fail. If we fail creating something, depending on
1298 the type of failure, we want to abort the whole task, or only skip the failed
1301 my $schema = MySchema->connect("dbi:Pg:dbname=my_db");
1303 # Start a transaction. Every database change from here on will only be
1304 # committed into the database if the try block succeeds.
1308 $schema->txn_do(sub {
1311 my $job = $schema->resultset('Job')->create({ name=> 'big job' });
1312 # SQL: INSERT INTO job ( name) VALUES ( 'big job' );
1316 # Start a nested transaction, which in fact sets a savepoint.
1318 $schema->txn_do(sub {
1319 # SQL: SAVEPOINT savepoint_0;
1321 my $thing = $schema->resultset('Thing')->create({ job=>$job->id });
1322 # SQL: INSERT INTO thing ( job) VALUES ( 1 );
1325 # This will generate an error, thus setting $@
1327 $thing->update({force_fail=>'foo'});
1328 # SQL: UPDATE thing SET force_fail = 'foo'
1329 # WHERE ( id = 42 );
1333 # SQL: ROLLBACK TO SAVEPOINT savepoint_0;
1335 # There was an error while creating a $thing. Depending on the error
1336 # we want to abort the whole transaction, or only rollback the
1337 # changes related to the creation of this $thing
1339 # Abort the whole job
1340 if ($_ =~ /horrible_problem/) {
1341 print "something horrible happend, aborting job!";
1342 die $_; # rethrow error
1345 # Ignore this $thing, report the error, and continue with the
1347 print "Cannot create thing: $_";
1349 # There was no error, so save all changes since the last
1352 # SQL: RELEASE SAVEPOINT savepoint_0;
1360 # There was an error while handling the $job. Rollback all changes
1361 # since the transaction started, including the already committed
1362 # ('released') savepoints. There will be neither a new $job nor any
1363 # $thing entry in the database.
1367 print "ERROR: $exception\n";
1370 # There was no error while handling the $job. Commit all changes.
1371 # Only now other connections can see the newly created $job and
1379 In this example it might be hard to see where the rollbacks, releases and
1380 commits are happening, but it works just the same as for plain L<<txn_do>>: If
1381 the C<try>-block around C<txn_do> fails, a rollback is issued. If the C<try>
1382 succeeds, the transaction is committed (or the savepoint released).
1384 While you can get more fine-grained control using C<svp_begin>, C<svp_release>
1385 and C<svp_rollback>, it is strongly recommended to use C<txn_do> with coderefs.
1387 =head2 Simple Transactions with DBIx::Class::Storage::TxnScopeGuard
1389 An easy way to use transactions is with
1390 L<DBIx::Class::Storage::TxnScopeGuard>. See L</Automatically creating
1391 related objects> for an example.
1393 Note that unlike txn_do, TxnScopeGuard will only make sure the connection is
1394 alive when issuing the C<BEGIN> statement. It will not (and really can not)
1395 retry if the server goes away mid-operations, unlike C<txn_do>.
1399 =head2 Creating Schemas From An Existing Database
1401 L<DBIx::Class::Schema::Loader> will connect to a database and create a
1402 L<DBIx::Class::Schema> and associated sources by examining the database.
1404 The recommend way of achieving this is to use the L<dbicdump> utility or the
1405 L<Catalyst> helper, as described in
1406 L<Manual::Intro|DBIx::Class::Manual::Intro/Using DBIx::Class::Schema::Loader>.
1408 Alternatively, use the
1409 L<make_schema_at|DBIx::Class::Schema::Loader/make_schema_at> method:
1411 perl -MDBIx::Class::Schema::Loader=make_schema_at,dump_to_dir:./lib \
1412 -e 'make_schema_at("My::Schema", \
1413 { db_schema => 'myschema', components => ["InflateColumn::DateTime"] }, \
1414 [ "dbi:Pg:dbname=foo", "username", "password" ])'
1416 This will create a tree of files rooted at C<./lib/My/Schema/> containing source
1417 definitions for all the tables found in the C<myschema> schema in the C<foo>
1420 =head2 Creating DDL SQL
1422 The following functionality requires you to have L<SQL::Translator>
1423 (also known as "SQL Fairy") installed.
1425 To create a set of database-specific .sql files for the above schema:
1427 my $schema = My::Schema->connect($dsn);
1428 $schema->create_ddl_dir(['MySQL', 'SQLite', 'PostgreSQL'],
1433 By default this will create schema files in the current directory, for
1434 MySQL, SQLite and PostgreSQL, using the $VERSION from your Schema.pm.
1436 To create a new database using the schema:
1438 my $schema = My::Schema->connect($dsn);
1439 $schema->deploy({ add_drop_table => 1});
1441 To import created .sql files using the mysql client:
1443 mysql -h "host" -D "database" -u "user" -p < My_Schema_1.0_MySQL.sql
1445 To create C<ALTER TABLE> conversion scripts to update a database to a
1446 newer version of your schema at a later point, first set a new
1447 C<$VERSION> in your Schema file, then:
1449 my $schema = My::Schema->connect($dsn);
1450 $schema->create_ddl_dir(['MySQL', 'SQLite', 'PostgreSQL'],
1456 This will produce new database-specific .sql files for the new version
1457 of the schema, plus scripts to convert from version 0.1 to 0.2. This
1458 requires that the files for 0.1 as created above are available in the
1459 given directory to diff against.
1461 =head2 Select from dual
1463 Dummy tables are needed by some databases to allow calling functions
1464 or expressions that aren't based on table content, for examples of how
1465 this applies to various database types, see:
1466 L<http://troels.arvin.dk/db/rdbms/#other-dummy_table>.
1468 Note: If you're using Oracles dual table don't B<ever> do anything
1469 other than a select, if you CRUD on your dual table you *will* break
1472 Make a table class as you would for any other table
1474 package MyAppDB::Dual;
1477 use base 'DBIx::Class::Core';
1478 __PACKAGE__->table("Dual");
1479 __PACKAGE__->add_columns(
1481 { data_type => "VARCHAR2", is_nullable => 0, size => 1 },
1484 Once you've loaded your table class select from it using C<select>
1485 and C<as> instead of C<columns>
1487 my $rs = $schema->resultset('Dual')->search(undef,
1488 { select => [ 'sydate' ],
1493 All you have to do now is be careful how you access your resultset, the below
1494 will not work because there is no column called 'now' in the Dual table class
1496 while (my $dual = $rs->next) {
1497 print $dual->now."\n";
1499 # Can't locate object method "now" via package "MyAppDB::Dual" at headshot.pl line 23.
1501 You could of course use 'dummy' in C<as> instead of 'now', or C<add_columns> to
1502 your Dual class for whatever you wanted to select from dual, but that's just
1503 silly, instead use C<get_column>
1505 while (my $dual = $rs->next) {
1506 print $dual->get_column('now')."\n";
1511 my $cursor = $rs->cursor;
1512 while (my @vals = $cursor->next) {
1513 print $vals[0]."\n";
1516 In case you're going to use this "trick" together with L<DBIx::Class::Schema/deploy> or
1517 L<DBIx::Class::Schema/create_ddl_dir> a table called "dual" will be created in your
1518 current schema. This would overlap "sys.dual" and you could not fetch "sysdate" or
1519 "sequence.nextval" anymore from dual. To avoid this problem, just tell
1520 L<SQL::Translator> to not create table dual:
1523 add_drop_table => 1,
1524 parser_args => { sources => [ grep $_ ne 'Dual', schema->sources ] },
1526 $schema->create_ddl_dir( [qw/Oracle/], undef, './sql', undef, $sqlt_args );
1528 Or use L<DBIx::Class::ResultClass::HashRefInflator>
1530 $rs->result_class('DBIx::Class::ResultClass::HashRefInflator');
1531 while ( my $dual = $rs->next ) {
1532 print $dual->{now}."\n";
1535 Here are some example C<select> conditions to illustrate the different syntax
1536 you could use for doing stuff like
1537 C<oracles.heavily(nested(functions_can('take', 'lots'), OF), 'args')>
1539 # get a sequence value
1540 select => [ 'A_SEQ.nextval' ],
1542 # get create table sql
1543 select => [ { 'dbms_metadata.get_ddl' => [ "'TABLE'", "'ARTIST'" ]} ],
1545 # get a random num between 0 and 100
1546 select => [ { "trunc" => [ { "dbms_random.value" => [0,100] } ]} ],
1549 select => [ { 'extract' => [ \'year from sysdate' ] } ],
1552 select => [ {'round' => [{'cos' => [ \'180 * 3.14159265359/180' ]}]}],
1554 # which day of the week were you born on?
1555 select => [{'to_char' => [{'to_date' => [ "'25-DEC-1980'", "'dd-mon-yyyy'" ]}, "'day'"]}],
1557 # select 16 rows from dual
1558 select => [ "'hello'" ],
1560 group_by => [ 'cube( 1, 2, 3, 4 )' ],
1564 =head2 Adding Indexes And Functions To Your SQL
1566 Often you will want indexes on columns on your table to speed up searching. To
1567 do this, create a method called C<sqlt_deploy_hook> in the relevant source
1568 class (refer to the advanced
1569 L<callback system|DBIx::Class::ResultSource/sqlt_deploy_callback> if you wish
1570 to share a hook between multiple sources):
1572 package My::Schema::Result::Artist;
1574 __PACKAGE__->table('artist');
1575 __PACKAGE__->add_columns(id => { ... }, name => { ... })
1577 sub sqlt_deploy_hook {
1578 my ($self, $sqlt_table) = @_;
1580 $sqlt_table->add_index(name => 'idx_name', fields => ['name']);
1585 Sometimes you might want to change the index depending on the type of the
1586 database for which SQL is being generated:
1588 my ($db_type = $sqlt_table->schema->translator->producer_type)
1589 =~ s/^SQL::Translator::Producer:://;
1591 You can also add hooks to the schema level to stop certain tables being
1598 sub sqlt_deploy_hook {
1599 my ($self, $sqlt_schema) = @_;
1601 $sqlt_schema->drop_table('table_name');
1604 You could also add views, procedures or triggers to the output using
1605 L<SQL::Translator::Schema/add_view>,
1606 L<SQL::Translator::Schema/add_procedure> or
1607 L<SQL::Translator::Schema/add_trigger>.
1610 =head2 Schema versioning
1612 The following example shows simplistically how you might use DBIx::Class to
1613 deploy versioned schemas to your customers. The basic process is as follows:
1619 Create a DBIx::Class schema
1631 Modify schema to change functionality
1635 Deploy update to customers
1639 B<Create a DBIx::Class schema>
1641 This can either be done manually, or generated from an existing database as
1642 described under L</Creating Schemas From An Existing Database>
1646 Call L<DBIx::Class::Schema/create_ddl_dir> as above under L</Creating DDL SQL>.
1648 B<Deploy to customers>
1650 There are several ways you could deploy your schema. These are probably
1651 beyond the scope of this recipe, but might include:
1657 Require customer to apply manually using their RDBMS.
1661 Package along with your app, making database dump/schema update/tests
1662 all part of your install.
1666 B<Modify the schema to change functionality>
1668 As your application evolves, it may be necessary to modify your schema
1669 to change functionality. Once the changes are made to your schema in
1670 DBIx::Class, export the modified schema and the conversion scripts as
1671 in L</Creating DDL SQL>.
1673 B<Deploy update to customers>
1675 Add the L<DBIx::Class::Schema::Versioned> schema component to your
1676 Schema class. This will add a new table to your database called
1677 C<dbix_class_schema_vesion> which will keep track of which version is installed
1678 and warn if the user tries to run a newer schema version than the
1679 database thinks it has.
1681 Alternatively, you can send the conversion SQL scripts to your
1684 =head2 Setting quoting for the generated SQL
1686 If the database contains column names with spaces and/or reserved words, they
1687 need to be quoted in the SQL queries. This is done using:
1689 $schema->storage->sql_maker->quote_char([ qw/[ ]/] );
1690 $schema->storage->sql_maker->name_sep('.');
1692 The first sets the quote characters. Either a pair of matching
1693 brackets, or a C<"> or C<'>:
1695 $schema->storage->sql_maker->quote_char('"');
1697 Check the documentation of your database for the correct quote
1698 characters to use. C<name_sep> needs to be set to allow the SQL
1699 generator to put the quotes the correct place, and defaults to
1700 C<.> if not supplied.
1702 In most cases you should set these as part of the arguments passed to
1703 L<DBIx::Class::Schema/connect>:
1705 my $schema = My::Schema->connect(
1715 In some cases, quoting will be required for all users of a schema. To enforce
1716 this, you can also overload the C<connection> method for your schema class:
1720 my $rv = $self->next::method( @_ );
1721 $rv->storage->sql_maker->quote_char([ qw/[ ]/ ]);
1722 $rv->storage->sql_maker->name_sep('.');
1726 =head2 Working with PostgreSQL array types
1728 You can also assign values to PostgreSQL array columns by passing array
1729 references in the C<\%columns> (C<\%vals>) hashref of the
1730 L<DBIx::Class::ResultSet/create> and L<DBIx::Class::Row/update> family of
1733 $resultset->create({
1734 numbers => [1, 2, 3]
1739 numbers => [1, 2, 3]
1746 In conditions (e.g. C<\%cond> in the L<DBIx::Class::ResultSet/search> family of
1747 methods) you cannot directly use array references (since this is interpreted as
1748 a list of values to be C<OR>ed), but you can use the following syntax to force
1749 passing them as bind values:
1753 numbers => \[ '= ?', [numbers => [1, 2, 3]] ]
1757 See L<SQL::Abstract/array_datatypes> and L<SQL::Abstract/Literal SQL with
1758 placeholders and bind values (subqueries)> for more explanation. Note that
1759 L<DBIx::Class> sets L<SQL::Abstract/bindtype> to C<columns>, so you must pass
1760 the bind values (the C<[1, 2, 3]> arrayref in the above example) wrapped in
1761 arrayrefs together with the column name, like this:
1762 C<< [column_name => value] >>.
1764 =head2 Formatting DateTime objects in queries
1766 To ensure C<WHERE> conditions containing L<DateTime> arguments are properly
1767 formatted to be understood by your RDBMS, you must use the C<DateTime>
1768 formatter returned by L<DBIx::Class::Storage::DBI/datetime_parser> to format
1769 any L<DateTime> objects you pass to L<search|DBIx::Class::ResultSet/search>
1770 conditions. Any L<Storage|DBIx::Class::Storage> object attached to your
1771 L<Schema|DBIx::Class::Schema> provides a correct C<DateTime> formatter, so
1772 all you have to do is:
1774 my $dtf = $schema->storage->datetime_parser;
1775 my $rs = $schema->resultset('users')->search(
1779 $dtf->format_datetime($dt_start),
1780 $dtf->format_datetime($dt_end),
1786 Without doing this the query will contain the simple stringification of the
1787 C<DateTime> object, which almost never matches the RDBMS expectations.
1789 This kludge is necessary only for conditions passed to
1790 L<DBIx::Class::ResultSet/search>, whereas
1791 L<create|DBIx::Class::ResultSet/create>,
1792 L<find|DBIx::Class::ResultSet/find>,
1793 L<DBIx::Class::Row/update> (but not L<DBIx::Class::ResultSet/update>) are all
1794 L<DBIx::Class::InflateColumn>-aware and will do the right thing when supplied
1795 an inflated C<DateTime> object.
1797 =head2 Using Unicode
1799 When using unicode character data there are two alternatives -
1800 either your database supports unicode characters (including setting
1801 the utf8 flag on the returned string), or you need to encode/decode
1802 data appropriately each time a string field is inserted into or
1803 retrieved from the database. It is better to avoid
1804 encoding/decoding data and to use your database's own unicode
1805 capabilities if at all possible.
1807 The L<DBIx::Class::UTF8Columns> component handles storing selected
1808 unicode columns in a database that does not directly support
1809 unicode. If used with a database that does correctly handle unicode
1810 then strange and unexpected data corrupt B<will> occur.
1812 The Catalyst Wiki Unicode page at
1813 L<http://wiki.catalystframework.org/wiki/tutorialsandhowtos/using_unicode>
1814 has additional information on the use of Unicode with Catalyst and
1817 The following databases do correctly handle unicode data:-
1821 MySQL supports unicode, and will correctly flag utf8 data from the
1822 database if the C<mysql_enable_utf8> is set in the connect options.
1824 my $schema = My::Schema->connection('dbi:mysql:dbname=test',
1826 { mysql_enable_utf8 => 1} );
1829 When set, a data retrieved from a textual column type (char,
1830 varchar, etc) will have the UTF-8 flag turned on if necessary. This
1831 enables character semantics on that string. You will also need to
1832 ensure that your database / table / column is configured to use
1833 UTF8. See Chapter 10 of the mysql manual for details.
1835 See L<DBD::mysql> for further details.
1839 Information about Oracle support for unicode can be found in
1840 L<DBD::Oracle/Unicode>.
1844 PostgreSQL supports unicode if the character set is correctly set
1845 at database creation time. Additionally the C<pg_enable_utf8>
1846 should be set to ensure unicode data is correctly marked.
1848 my $schema = My::Schema->connection('dbi:Pg:dbname=test',
1850 { pg_enable_utf8 => 1} );
1852 Further information can be found in L<DBD::Pg>.
1856 SQLite version 3 and above natively use unicode internally. To
1857 correctly mark unicode strings taken from the database, the
1858 C<sqlite_unicode> flag should be set at connect time (in versions
1859 of L<DBD::SQLite> prior to 1.27 this attribute was named
1862 my $schema = My::Schema->connection('dbi:SQLite:/tmp/test.db',
1864 { sqlite_unicode => 1} );
1866 =head1 BOOTSTRAPPING/MIGRATING
1868 =head2 Easy migration from class-based to schema-based setup
1870 You want to start using the schema-based approach to L<DBIx::Class>
1871 (see L<DBIx::Class::Manual::Intro/Setting it up manually>), but have an
1872 established class-based setup with lots of existing classes that you don't
1873 want to move by hand. Try this nifty script instead:
1876 use SQL::Translator;
1878 my $schema = MyDB->schema_instance;
1880 my $translator = SQL::Translator->new(
1881 debug => $debug || 0,
1882 trace => $trace || 0,
1883 no_comments => $no_comments || 0,
1884 show_warnings => $show_warnings || 0,
1885 add_drop_table => $add_drop_table || 0,
1886 validate => $validate || 0,
1888 'DBIx::Schema' => $schema,
1891 'prefix' => 'My::Schema',
1895 $translator->parser('SQL::Translator::Parser::DBIx::Class');
1896 $translator->producer('SQL::Translator::Producer::DBIx::Class::File');
1898 my $output = $translator->translate(@args) or die
1899 "Error: " . $translator->error;
1903 You could use L<Module::Find> to search for all subclasses in the MyDB::*
1904 namespace, which is currently left as an exercise for the reader.
1906 =head1 OVERLOADING METHODS
1908 L<DBIx::Class> uses the L<Class::C3> package, which provides for redispatch of
1909 method calls, useful for things like default values and triggers. You have to
1910 use calls to C<next::method> to overload methods. More information on using
1911 L<Class::C3> with L<DBIx::Class> can be found in
1912 L<DBIx::Class::Manual::Component>.
1914 =head2 Setting default values for a row
1916 It's as simple as overriding the C<new> method. Note the use of
1920 my ( $class, $attrs ) = @_;
1922 $attrs->{foo} = 'bar' unless defined $attrs->{foo};
1924 my $new = $class->next::method($attrs);
1929 For more information about C<next::method>, look in the L<Class::C3>
1930 documentation. See also L<DBIx::Class::Manual::Component> for more
1931 ways to write your own base classes to do this.
1933 People looking for ways to do "triggers" with DBIx::Class are probably
1934 just looking for this.
1936 =head2 Changing one field whenever another changes
1938 For example, say that you have three columns, C<id>, C<number>, and
1939 C<squared>. You would like to make changes to C<number> and have
1940 C<squared> be automagically set to the value of C<number> squared.
1941 You can accomplish this by wrapping the C<number> accessor with the C<around>
1942 method modifier, available through either L<Class::Method::Modifiers>,
1943 L<Moose|Moose::Manual::MethodModifiers> or L<Moose-like|Moo> modules):
1945 around number => sub {
1946 my ($orig, $self) = (shift, shift);
1950 $self->squared( $value * $value );
1956 Note that the hard work is done by the call to C<< $self->$orig >>, which
1957 redispatches your call to store_column in the superclass(es).
1959 Generally, if this is a calculation your database can easily do, try
1960 and avoid storing the calculated value, it is safer to calculate when
1961 needed, than rely on the data being in sync.
1963 =head2 Automatically creating related objects
1965 You might have a class C<Artist> which has many C<CD>s. Further, you
1966 want to create a C<CD> object every time you insert an C<Artist> object.
1967 You can accomplish this by overriding C<insert> on your objects:
1970 my ( $self, @args ) = @_;
1971 $self->next::method(@args);
1972 $self->create_related ('cds', \%initial_cd_data );
1976 If you want to wrap the two inserts in a transaction (for consistency,
1977 an excellent idea), you can use the awesome
1978 L<DBIx::Class::Storage::TxnScopeGuard>:
1981 my ( $self, @args ) = @_;
1983 my $guard = $self->result_source->schema->txn_scope_guard;
1985 $self->next::method(@args);
1986 $self->create_related ('cds', \%initial_cd_data );
1994 =head2 Wrapping/overloading a column accessor
1998 Say you have a table "Camera" and want to associate a description
1999 with each camera. For most cameras, you'll be able to generate the description from
2000 the other columns. However, in a few special cases you may want to associate a
2001 custom description with a camera.
2005 In your database schema, define a description field in the "Camera" table that
2006 can contain text and null values.
2008 In DBIC, we'll overload the column accessor to provide a sane default if no
2009 custom description is defined. The accessor will either return or generate the
2010 description, depending on whether the field is null or not.
2012 First, in your "Camera" schema class, define the description field as follows:
2014 __PACKAGE__->add_columns(description => { accessor => '_description' });
2016 Next, we'll define the accessor-wrapper subroutine:
2021 # If there is an update to the column, we'll let the original accessor
2023 return $self->_description(@_) if @_;
2025 # Fetch the column value.
2026 my $description = $self->_description;
2028 # If there's something in the description field, then just return that.
2029 return $description if defined $description && length $descripton;
2031 # Otherwise, generate a description.
2032 return $self->generate_description;
2035 =head1 DEBUGGING AND PROFILING
2037 =head2 DBIx::Class objects with Data::Dumper
2039 L<Data::Dumper> can be a very useful tool for debugging, but sometimes it can
2040 be hard to find the pertinent data in all the data it can generate.
2041 Specifically, if one naively tries to use it like so,
2045 my $cd = $schema->resultset('CD')->find(1);
2048 several pages worth of data from the CD object's schema and result source will
2049 be dumped to the screen. Since usually one is only interested in a few column
2050 values of the object, this is not very helpful.
2052 Luckily, it is possible to modify the data before L<Data::Dumper> outputs
2053 it. Simply define a hook that L<Data::Dumper> will call on the object before
2054 dumping it. For example,
2061 result_source => undef,
2069 local $Data::Dumper::Freezer = '_dumper_hook';
2071 my $cd = $schema->resultset('CD')->find(1);
2073 # dumps $cd without its ResultSource
2075 If the structure of your schema is such that there is a common base class for
2076 all your table classes, simply put a method similar to C<_dumper_hook> in the
2077 base class and set C<$Data::Dumper::Freezer> to its name and L<Data::Dumper>
2078 will automagically clean up your data before printing it. See
2079 L<Data::Dumper/EXAMPLES> for more information.
2083 When you enable L<DBIx::Class::Storage>'s debugging it prints the SQL
2084 executed as well as notifications of query completion and transaction
2085 begin/commit. If you'd like to profile the SQL you can subclass the
2086 L<DBIx::Class::Storage::Statistics> class and write your own profiling
2089 package My::Profiler;
2092 use base 'DBIx::Class::Storage::Statistics';
2094 use Time::HiRes qw(time);
2103 $self->print("Executing $sql: ".join(', ', @params)."\n");
2112 my $elapsed = sprintf("%0.4f", time() - $start);
2113 $self->print("Execution took $elapsed seconds.\n");
2119 You can then install that class as the debugging object:
2121 __PACKAGE__->storage->debugobj(new My::Profiler());
2122 __PACKAGE__->storage->debug(1);
2124 A more complicated example might involve storing each execution of SQL in an
2132 my $elapsed = time() - $start;
2133 push(@{ $calls{$sql} }, {
2139 You could then create average, high and low execution times for an SQL
2140 statement and dig down to see if certain parameters cause aberrant behavior.
2141 You might want to check out L<DBIx::Class::QueryLog> as well.
2143 =head1 IMPROVING PERFORMANCE
2149 Install L<Class::XSAccessor> to speed up L<Class::Accessor::Grouped>.
2153 On Perl 5.8 install L<Class::C3::XS>.
2157 L<prefetch|DBIx::Class::ResultSet/prefetch> relationships, where possible. See
2158 L</Using joins and prefetch>.
2162 Use L<populate|DBIx::Class::ResultSet/populate> in void context to insert data
2163 when you don't need the resulting L<result|DBIx::Class::Manual::ResultClass> objects,
2164 if possible, but see the caveats.
2166 When inserting many rows, for best results, populate a large number of rows at a
2167 time, but not so large that the table is locked for an unacceptably long time.
2169 If using L<create|DBIx::Class::ResultSet/create> instead, use a transaction and
2170 commit every C<X> rows; where C<X> gives you the best performance without
2171 locking the table for too long.
2175 When selecting many rows, if you don't need full-blown L<DBIx::Class::Row>
2176 objects, consider using L<DBIx::Class::ResultClass::HashRefInflator>.
2180 See also L</STARTUP SPEED> and L</MEMORY USAGE> in this document.
2184 =head1 STARTUP SPEED
2186 L<DBIx::Class|DBIx::Class> programs can have a significant startup delay
2187 as the ORM loads all the relevant classes. This section examines
2188 techniques for reducing the startup delay.
2190 These tips are are listed in order of decreasing effectiveness - so the
2191 first tip, if applicable, should have the greatest effect on your
2194 =head2 Statically Define Your Schema
2197 L<DBIx::Class::Schema::Loader|DBIx::Class::Schema::Loader> to build the
2198 classes dynamically based on the database schema then there will be a
2199 significant startup delay.
2201 For production use a statically defined schema (which can be generated
2202 using L<DBIx::Class::Schema::Loader|DBIx::Class::Schema::Loader> to dump
2203 the database schema once - see
2204 L<make_schema_at|DBIx::Class::Schema::Loader/make_schema_at> and
2205 L<dump_directory|DBIx::Class::Schema::Loader/dump_directory> for more
2206 details on creating static schemas from a database).
2208 =head2 Move Common Startup into a Base Class
2210 Typically L<DBIx::Class> result classes start off with
2212 use base qw/DBIx::Class::Core/;
2213 __PACKAGE__->load_components(qw/InflateColumn::DateTime/);
2215 If this preamble is moved into a common base class:-
2219 use base qw/DBIx::Class::Core/;
2220 __PACKAGE__->load_components(qw/InflateColumn::DateTime/);
2223 and each result class then uses this as a base:-
2225 use base qw/MyDBICbase/;
2227 then the load_components is only performed once, which can result in a
2228 considerable startup speedup for schemas with many classes.
2230 =head2 Explicitly List Schema Result Classes
2232 The schema class will normally contain
2234 __PACKAGE__->load_classes();
2236 to load the result classes. This will use L<Module::Find|Module::Find>
2237 to find and load the appropriate modules. Explicitly defining the
2238 classes you wish to load will remove the overhead of
2239 L<Module::Find|Module::Find> and the related directory operations:
2241 __PACKAGE__->load_classes(qw/ CD Artist Track /);
2243 If you are instead using the L<load_namespaces|DBIx::Class::Schema/load_namespaces>
2244 syntax to load the appropriate classes there is not a direct alternative
2245 avoiding L<Module::Find|Module::Find>.
2249 =head2 Cached statements
2251 L<DBIx::Class> normally caches all statements with L<< prepare_cached()|DBI/prepare_cached >>.
2252 This is normally a good idea, but if too many statements are cached, the database may use too much
2253 memory and may eventually run out and fail entirely. If you suspect this may be the case, you may want
2254 to examine DBI's L<< CachedKids|DBI/CachedKidsCachedKids_(hash_ref) >> hash:
2256 # print all currently cached prepared statements
2257 print for keys %{$schema->storage->dbh->{CachedKids}};
2258 # get a count of currently cached prepared statements
2259 my $count = scalar keys %{$schema->storage->dbh->{CachedKids}};
2261 If it's appropriate, you can simply clear these statements, automatically deallocating them in the
2264 my $kids = $schema->storage->dbh->{CachedKids};
2265 delete @{$kids}{keys %$kids} if scalar keys %$kids > 100;
2267 But what you probably want is to expire unused statements and not those that are used frequently.
2268 You can accomplish this with L<Tie::Cache> or L<Tie::Cache::LRU>:
2272 my $schema = DB::Main->connect($dbi_dsn, $user, $pass, {
2273 on_connect_do => sub { tie %{shift->_dbh->{CachedKids}}, 'Tie::Cache', 100 },