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 '%Lamb%' AND title LIKE '%Fear of Fours%'
42 Other queries might require slightly more complex logic:
44 my @albums = $schema->resultset('Album')->search({
47 artist => { 'like', '%Smashing Pumpkins%' },
48 title => 'Siamese Dream',
50 artist => 'Starchildren',
54 This results in the following C<WHERE> clause:
56 WHERE ( artist LIKE '%Smashing Pumpkins%' AND title = 'Siamese Dream' )
57 OR artist = 'Starchildren'
59 For more information on generating complex queries, see
60 L<SQL::Abstract/WHERE CLAUSES>.
62 =head2 Retrieve one and only one row from a resultset
64 Sometimes you need only the first "top" row of a resultset. While this
65 can be easily done with L<< $rs->first|DBIx::Class::ResultSet/first
66 >>, it is suboptimal, as a full blown cursor for the resultset will be
67 created and then immediately destroyed after fetching the first row
68 object. L<< $rs->single|DBIx::Class::ResultSet/single >> is designed
69 specifically for this case - it will grab the first returned result
70 without even instantiating a cursor.
72 Before replacing all your calls to C<first()> with C<single()> please observe the
79 While single() takes a search condition just like search() does, it does
80 _not_ accept search attributes. However one can always chain a single() to
83 my $top_cd = $cd_rs->search({}, { order_by => 'rating' })->single;
88 Since single() is the engine behind find(), it is designed to fetch a
89 single row per database query. Thus a warning will be issued when the
90 underlying SELECT returns more than one row. Sometimes however this usage
91 is valid: i.e. we have an arbitrary number of cd's but only one of them is
92 at the top of the charts at any given time. If you know what you are doing,
93 you can silence the warning by explicitly limiting the resultset size:
95 my $top_cd = $cd_rs->search ({}, { order_by => 'rating', rows => 1 })->single;
99 =head2 Arbitrary SQL through a custom ResultSource
101 Sometimes you have to run arbitrary SQL because your query is too complex
102 (e.g. it contains Unions, Sub-Selects, Stored Procedures, etc.) or has to
103 be optimized for your database in a special way, but you still want to
104 get the results as a L<DBIx::Class::ResultSet>.
106 The recommended way to accomplish this is by defining a separate
107 L<ResultSource::View|DBIx::Class::ResultSource::View> for your query.
109 package My::Schema::Result::UserFriendsComplex;
112 use base qw/DBIx::Class/;
114 __PACKAGE__->load_components('Core');
115 __PACKAGE__->table_class('DBIx::Class::ResultSource::View');
117 # ->table, ->add_columns, etc.
119 __PACKAGE__->result_source_instance->is_virtual(1);
120 __PACKAGE__->result_source_instance->view_definition(q[
121 SELECT u.* FROM user u
122 INNER JOIN user_friends f ON u.id = f.user_id
123 WHERE f.friend_user_id = ?
125 SELECT u.* FROM user u
126 INNER JOIN user_friends f ON u.id = f.friend_user_id
130 Next, you can execute your complex query using bind parameters like this:
132 my $friends = $schema->resultset( 'UserFriendsComplex' )->search( {},
134 bind => [ 12345, 12345 ]
138 ... and you'll get back a perfect L<DBIx::Class::ResultSet> (except, of course,
139 that you cannot modify the rows it contains, ie. cannot call L</update>,
140 L</delete>, ... on it).
142 Note that you cannot have bind parameters unless is_virtual is set to true.
144 =head2 Using specific columns
146 When you only want specific columns from a table, you can use
147 C<columns> to specify which ones you need. This is useful to avoid
148 loading columns with large amounts of data that you aren't about to
151 my $rs = $schema->resultset('Artist')->search(
154 columns => [qw/ name /]
159 # SELECT artist.name FROM artist
161 This is a shortcut for C<select> and C<as>, see below. C<columns>
162 cannot be used together with C<select> and C<as>.
164 =head2 Using database functions or stored procedures
166 The combination of C<select> and C<as> can be used to return the result of a
167 database function or stored procedure as a column value. You use C<select> to
168 specify the source for your column value (e.g. a column name, function, or
169 stored procedure name). You then use C<as> to set the column name you will use
170 to access the returned value:
172 my $rs = $schema->resultset('Artist')->search(
175 select => [ 'name', { LENGTH => 'name' } ],
176 as => [qw/ name name_length /],
181 # SELECT name name, LENGTH( name )
184 Note that the C< as > attribute has absolutely nothing to with the sql
185 syntax C< SELECT foo AS bar > (see the documentation in
186 L<DBIx::Class::ResultSet/ATTRIBUTES>). If your alias exists as a
187 column in your base class (i.e. it was added with C<add_columns>), you
188 just access it as normal. Our C<Artist> class has a C<name> column, so
189 we just use the C<name> accessor:
191 my $artist = $rs->first();
192 my $name = $artist->name();
194 If on the other hand the alias does not correspond to an existing column, you
195 have to fetch the value using the C<get_column> accessor:
197 my $name_length = $artist->get_column('name_length');
199 If you don't like using C<get_column>, you can always create an accessor for
200 any of your aliases using either of these:
202 # Define accessor manually:
203 sub name_length { shift->get_column('name_length'); }
205 # Or use DBIx::Class::AccessorGroup:
206 __PACKAGE__->mk_group_accessors('column' => 'name_length');
208 =head2 SELECT DISTINCT with multiple columns
210 my $rs = $schema->resultset('Artist')->search(
213 columns => [ qw/artist_id name rank/ ],
218 my $rs = $schema->resultset('Artist')->search(
221 columns => [ qw/artist_id name rank/ ],
222 group_by => [ qw/artist_id name rank/ ],
227 # SELECT me.artist_id, me.name, me.rank
229 # GROUP BY artist_id, name, rank
231 =head2 SELECT COUNT(DISTINCT colname)
233 my $rs = $schema->resultset('Artist')->search(
236 columns => [ qw/name/ ],
241 my $rs = $schema->resultset('Artist')->search(
244 columns => [ qw/name/ ],
245 group_by => [ qw/name/ ],
249 my $count = $rs->count;
252 # SELECT COUNT( * ) FROM (SELECT me.name FROM artist me GROUP BY me.name) count_subq:
254 =head2 Grouping results
256 L<DBIx::Class> supports C<GROUP BY> as follows:
258 my $rs = $schema->resultset('Artist')->search(
262 select => [ 'name', { count => 'cds.id' } ],
263 as => [qw/ name cd_count /],
264 group_by => [qw/ name /]
269 # SELECT name, COUNT( cd.id ) FROM artist
270 # LEFT JOIN cd ON artist.id = cd.artist
273 Please see L<DBIx::Class::ResultSet/ATTRIBUTES> documentation if you
274 are in any way unsure about the use of the attributes above (C< join
275 >, C< select >, C< as > and C< group_by >).
277 =head2 Subqueries (EXPERIMENTAL)
279 You can write subqueries relatively easily in DBIC.
281 my $inside_rs = $schema->resultset('Artist')->search({
282 name => [ 'Billy Joel', 'Brittany Spears' ],
285 my $rs = $schema->resultset('CD')->search({
286 artist_id => { 'IN' => $inside_rs->get_column('id')->as_query },
289 The usual operators ( =, !=, IN, NOT IN, etc) are supported.
291 B<NOTE>: You have to explicitly use '=' when doing an equality comparison.
292 The following will B<not> work:
294 my $rs = $schema->resultset('CD')->search({
295 artist_id => $inside_rs->get_column('id')->as_query,
300 Subqueries are supported in the where clause (first hashref), and in the
301 from, select, and +select attributes.
303 =head3 Correlated subqueries
305 my $cdrs = $schema->resultset('CD');
306 my $rs = $cdrs->search({
308 '=' => $cdrs->search(
309 { artist_id => { '=' => \'me.artist_id' } },
311 )->get_column('year')->max_rs->as_query,
315 That creates the following SQL:
317 SELECT me.cdid, me.artist, me.title, me.year, me.genreid, me.single_track
320 SELECT MAX(inner.year)
322 WHERE artist_id = me.artist_id
327 Please note that subqueries are considered an experimental feature.
329 =head2 Predefined searches
331 You can write your own L<DBIx::Class::ResultSet> class by inheriting from it
332 and defining often used searches as methods:
334 package My::DBIC::ResultSet::CD;
337 use base 'DBIx::Class::ResultSet';
339 sub search_cds_ordered {
342 return $self->search(
344 { order_by => 'name DESC' },
350 To use your resultset, first tell DBIx::Class to create an instance of it
351 for you, in your My::DBIC::Schema::CD class:
353 # class definition as normal
354 __PACKAGE__->load_components(qw/ Core /);
355 __PACKAGE__->table('cd');
357 # tell DBIC to use the custom ResultSet class
358 __PACKAGE__->resultset_class('My::DBIC::ResultSet::CD');
360 Note that C<resultset_class> must be called after C<load_components> and C<table>, or you will get errors about missing methods.
362 Then call your new method in your code:
364 my $ordered_cds = $schema->resultset('CD')->search_cds_ordered();
366 =head2 Using SQL functions on the left hand side of a comparison
368 Using SQL functions on the left hand side of a comparison is generally
369 not a good idea since it requires a scan of the entire table. However,
370 it can be accomplished with C<DBIx::Class> when necessary.
372 If you do not have quoting on, simply include the function in your search
373 specification as you would any column:
375 $rs->search({ 'YEAR(date_of_birth)' => 1979 });
377 With quoting on, or for a more portable solution, use the C<where>
380 $rs->search({}, { where => \'YEAR(date_of_birth) = 1979' });
384 (When the bind args ordering bug is fixed, this technique will be better
385 and can replace the one above.)
387 With quoting on, or for a more portable solution, use the C<where> and
391 where => \'YEAR(date_of_birth) = ?',
397 =head1 JOINS AND PREFETCHING
399 =head2 Using joins and prefetch
401 You can use the C<join> attribute to allow searching on, or sorting your
402 results by, one or more columns in a related table.
404 This requires that you have defined the L<DBIx::Class::Relationship>. For example :
406 My::Schema::CD->has_many( artists => 'My::Schema::Artist', 'artist_id');
408 To return all CDs matching a particular artist name, you specify the name of the relationship ('artists'):
410 my $rs = $schema->resultset('CD')->search(
412 'artists.name' => 'Bob Marley'
415 join => 'artists', # join the artist table
420 # SELECT cd.* FROM cd
421 # JOIN artist ON cd.artist = artist.id
422 # WHERE artist.name = 'Bob Marley'
424 In that example both the join, and the condition use the relationship name rather than the table name
425 (see L<DBIx::Class::Manual::Joining> for more details on aliasing ).
427 If required, you can now sort on any column in the related tables by including
428 it in your C<order_by> attribute, (again using the aliased relation name rather than table name) :
430 my $rs = $schema->resultset('CD')->search(
432 'artists.name' => 'Bob Marley'
436 order_by => [qw/ artists.name /]
441 # SELECT cd.* FROM cd
442 # JOIN artist ON cd.artist = artist.id
443 # WHERE artist.name = 'Bob Marley'
444 # ORDER BY artist.name
446 Note that the C<join> attribute should only be used when you need to search or
447 sort using columns in a related table. Joining related tables when you only
448 need columns from the main table will make performance worse!
450 Now let's say you want to display a list of CDs, each with the name of the
451 artist. The following will work fine:
453 while (my $cd = $rs->next) {
454 print "CD: " . $cd->title . ", Artist: " . $cd->artist->name;
457 There is a problem however. We have searched both the C<cd> and C<artist> tables
458 in our main query, but we have only returned data from the C<cd> table. To get
459 the artist name for any of the CD objects returned, L<DBIx::Class> will go back
462 SELECT artist.* FROM artist WHERE artist.id = ?
464 A statement like the one above will run for each and every CD returned by our
465 main query. Five CDs, five extra queries. A hundred CDs, one hundred extra
468 Thankfully, L<DBIx::Class> has a C<prefetch> attribute to solve this problem.
469 This allows you to fetch results from related tables in advance:
471 my $rs = $schema->resultset('CD')->search(
473 'artists.name' => 'Bob Marley'
477 order_by => [qw/ artists.name /],
478 prefetch => 'artists' # return artist data too!
482 # Equivalent SQL (note SELECT from both "cd" and "artist"):
483 # SELECT cd.*, artist.* FROM cd
484 # JOIN artist ON cd.artist = artist.id
485 # WHERE artist.name = 'Bob Marley'
486 # ORDER BY artist.name
488 The code to print the CD list remains the same:
490 while (my $cd = $rs->next) {
491 print "CD: " . $cd->title . ", Artist: " . $cd->artist->name;
494 L<DBIx::Class> has now prefetched all matching data from the C<artist> table,
495 so no additional SQL statements are executed. You now have a much more
498 Also note that C<prefetch> should only be used when you know you will
499 definitely use data from a related table. Pre-fetching related tables when you
500 only need columns from the main table will make performance worse!
502 =head2 Multiple joins
504 In the examples above, the C<join> attribute was a scalar. If you
505 pass an array reference instead, you can join to multiple tables. In
506 this example, we want to limit the search further, using
509 # Relationships defined elsewhere:
510 # CD->belongs_to('artist' => 'Artist');
511 # CD->has_one('liner_notes' => 'LinerNotes', 'cd');
512 my $rs = $schema->resultset('CD')->search(
514 'artist.name' => 'Bob Marley'
515 'liner_notes.notes' => { 'like', '%some text%' },
518 join => [qw/ artist liner_notes /],
519 order_by => [qw/ artist.name /],
524 # SELECT cd.*, artist.*, liner_notes.* FROM cd
525 # JOIN artist ON cd.artist = artist.id
526 # JOIN liner_notes ON cd.id = liner_notes.cd
527 # WHERE artist.name = 'Bob Marley'
528 # ORDER BY artist.name
530 =head2 Multi-step joins
532 Sometimes you want to join more than one relationship deep. In this example,
533 we want to find all C<Artist> objects who have C<CD>s whose C<LinerNotes>
534 contain a specific string:
536 # Relationships defined elsewhere:
537 # Artist->has_many('cds' => 'CD', 'artist');
538 # CD->has_one('liner_notes' => 'LinerNotes', 'cd');
540 my $rs = $schema->resultset('Artist')->search(
542 'liner_notes.notes' => { 'like', '%some text%' },
546 'cds' => 'liner_notes'
552 # SELECT artist.* FROM artist
553 # LEFT JOIN cd ON artist.id = cd.artist
554 # LEFT JOIN liner_notes ON cd.id = liner_notes.cd
555 # WHERE liner_notes.notes LIKE '%some text%'
557 Joins can be nested to an arbitrary level. So if we decide later that we
558 want to reduce the number of Artists returned based on who wrote the liner
561 # Relationship defined elsewhere:
562 # LinerNotes->belongs_to('author' => 'Person');
564 my $rs = $schema->resultset('Artist')->search(
566 'liner_notes.notes' => { 'like', '%some text%' },
567 'author.name' => 'A. Writer'
572 'liner_notes' => 'author'
579 # SELECT artist.* FROM artist
580 # LEFT JOIN cd ON artist.id = cd.artist
581 # LEFT JOIN liner_notes ON cd.id = liner_notes.cd
582 # LEFT JOIN author ON author.id = liner_notes.author
583 # WHERE liner_notes.notes LIKE '%some text%'
584 # AND author.name = 'A. Writer'
586 =head2 Multi-step and multiple joins
588 With various combinations of array and hash references, you can join
589 tables in any combination you desire. For example, to join Artist to
590 CD and Concert, and join CD to LinerNotes:
592 # Relationships defined elsewhere:
593 # Artist->has_many('concerts' => 'Concert', 'artist');
595 my $rs = $schema->resultset('Artist')->search(
608 # SELECT artist.* FROM artist
609 # LEFT JOIN cd ON artist.id = cd.artist
610 # LEFT JOIN liner_notes ON cd.id = liner_notes.cd
611 # LEFT JOIN concert ON artist.id = concert.artist
613 =head2 Multi-step prefetch
615 C<prefetch> can be nested more than one relationship
616 deep using the same syntax as a multi-step join:
618 my $rs = $schema->resultset('Tag')->search(
628 # SELECT tag.*, cd.*, artist.* FROM tag
629 # JOIN cd ON tag.cd = cd.id
630 # JOIN artist ON cd.artist = artist.id
632 Now accessing our C<cd> and C<artist> relationships does not need additional
635 my $tag = $rs->first;
636 print $tag->cd->artist->name;
638 =head1 ROW-LEVEL OPERATIONS
640 =head2 Retrieving a row object's Schema
642 It is possible to get a Schema object from a row object like so:
644 my $schema = $cd->result_source->schema;
645 # use the schema as normal:
646 my $artist_rs = $schema->resultset('Artist');
648 This can be useful when you don't want to pass around a Schema object to every
651 =head2 Getting the value of the primary key for the last database insert
653 AKA getting last_insert_id
655 Thanks to the core component PK::Auto, this is straightforward:
657 my $foo = $rs->create(\%blah);
659 my $id = $foo->id; # foo->my_primary_key_field will also work.
661 If you are not using autoincrementing primary keys, this will probably
662 not work, but then you already know the value of the last primary key anyway.
664 =head2 Stringification
666 Employ the standard stringification technique by using the L<overload>
669 To make an object stringify itself as a single column, use something
670 like this (replace C<name> with the column/method of your choice):
672 use overload '""' => sub { shift->name}, fallback => 1;
674 For more complex stringification, you can use an anonymous subroutine:
676 use overload '""' => sub { $_[0]->name . ", " .
677 $_[0]->address }, fallback => 1;
679 =head3 Stringification Example
681 Suppose we have two tables: C<Product> and C<Category>. The table
684 Product(id, Description, category)
685 Category(id, Description)
687 C<category> is a foreign key into the Category table.
689 If you have a Product object C<$obj> and write something like
693 things will not work as expected.
695 To obtain, for example, the category description, you should add this
696 method to the class defining the Category table:
698 use overload "" => sub {
701 return $self->Description;
704 =head2 Want to know if find_or_create found or created a row?
706 Just use C<find_or_new> instead, then check C<in_storage>:
708 my $obj = $rs->find_or_new({ blah => 'blarg' });
709 unless ($obj->in_storage) {
711 # do whatever else you wanted if it was a new row
714 =head2 Static sub-classing DBIx::Class result classes
716 AKA adding additional relationships/methods/etc. to a model for a
717 specific usage of the (shared) model.
721 package My::App::Schema;
723 use base DBIx::Class::Schema;
725 # load subclassed classes from My::App::Schema::Result/ResultSet
726 __PACKAGE__->load_namespaces;
728 # load classes from shared model
730 'My::Shared::Model::Result' => [qw/
737 B<Result-Subclass definition>
739 package My::App::Schema::Result::Baz;
743 use base My::Shared::Model::Result::Baz;
745 # WARNING: Make sure you call table() again in your subclass,
746 # otherwise DBIx::Class::ResultSourceProxy::Table will not be called
747 # and the class name is not correctly registered as a source
748 __PACKAGE__->table('baz');
750 sub additional_method {
751 return "I'm an additional method only needed by this app";
756 =head2 Dynamic Sub-classing DBIx::Class proxy classes
758 AKA multi-class object inflation from one table
760 L<DBIx::Class> classes are proxy classes, therefore some different
761 techniques need to be employed for more than basic subclassing. In
762 this example we have a single user table that carries a boolean bit
763 for admin. We would like like to give the admin users
764 objects (L<DBIx::Class::Row>) the same methods as a regular user but
765 also special admin only methods. It doesn't make sense to create two
766 seperate proxy-class files for this. We would be copying all the user
767 methods into the Admin class. There is a cleaner way to accomplish
770 Overriding the C<inflate_result> method within the User proxy-class
771 gives us the effect we want. This method is called by
772 L<DBIx::Class::ResultSet> when inflating a result from storage. So we
773 grab the object being returned, inspect the values we are looking for,
774 bless it if it's an admin object, and then return it. See the example
781 use base qw/DBIx::Class::Schema/;
783 __PACKAGE__->load_namespaces;
788 B<Proxy-Class definitions>
790 package My::Schema::Result::User;
794 use base qw/DBIx::Class/;
796 ### Define what our admin class is, for ensure_class_loaded()
797 my $admin_class = __PACKAGE__ . '::Admin';
799 __PACKAGE__->load_components(qw/Core/);
801 __PACKAGE__->table('users');
803 __PACKAGE__->add_columns(qw/user_id email password
804 firstname lastname active
807 __PACKAGE__->set_primary_key('user_id');
811 my $ret = $self->next::method(@_);
812 if( $ret->admin ) {### If this is an admin, rebless for extra functions
813 $self->ensure_class_loaded( $admin_class );
814 bless $ret, $admin_class;
820 print "I am a regular user.\n";
827 package My::Schema::Result::User::Admin;
831 use base qw/My::Schema::Result::User/;
833 # This line is important
834 __PACKAGE__->table('users');
838 print "I am an admin.\n";
844 print "I am doing admin stuff\n";
856 my $user_data = { email => 'someguy@place.com',
860 my $admin_data = { email => 'someadmin@adminplace.com',
864 my $schema = My::Schema->connection('dbi:Pg:dbname=test');
866 $schema->resultset('User')->create( $user_data );
867 $schema->resultset('User')->create( $admin_data );
869 ### Now we search for them
870 my $user = $schema->resultset('User')->single( $user_data );
871 my $admin = $schema->resultset('User')->single( $admin_data );
873 print ref $user, "\n";
874 print ref $admin, "\n";
876 print $user->password , "\n"; # pass1
877 print $admin->password , "\n";# pass2; inherited from User
878 print $user->hello , "\n";# I am a regular user.
879 print $admin->hello, "\n";# I am an admin.
881 ### The statement below will NOT print
882 print "I can do admin stuff\n" if $user->can('do_admin_stuff');
883 ### The statement below will print
884 print "I can do admin stuff\n" if $admin->can('do_admin_stuff');
886 =head2 Skip row object creation for faster results
888 DBIx::Class is not built for speed, it's built for convenience and
889 ease of use, but sometimes you just need to get the data, and skip the
892 To do this simply use L<DBIx::Class::ResultClass::HashRefInflator>.
894 my $rs = $schema->resultset('CD');
896 $rs->result_class('DBIx::Class::ResultClass::HashRefInflator');
898 my $hash_ref = $rs->find(1);
902 Beware, changing the Result class using
903 L<DBIx::Class::ResultSet/result_class> will replace any existing class
904 completely including any special components loaded using
905 load_components, eg L<DBIx::Class::InflateColumn::DateTime>.
907 =head2 Get raw data for blindingly fast results
909 If the L<HashRefInflator|DBIx::Class::ResultClass::HashRefInflator> solution
910 above is not fast enough for you, you can use a DBIx::Class to return values
911 exactly as they come out of the database with none of the convenience methods
914 This is used like so:
916 my $cursor = $rs->cursor
917 while (my @vals = $cursor->next) {
918 # use $val[0..n] here
921 You will need to map the array offsets to particular columns (you can
922 use the L<DBIx::Class::ResultSet/select> attribute of L<DBIx::Class::ResultSet/search> to force ordering).
924 =head1 RESULTSET OPERATIONS
926 =head2 Getting Schema from a ResultSet
928 To get the L<DBIx::Class::Schema> object from a ResultSet, do the following:
930 $rs->result_source->schema
932 =head2 Getting Columns Of Data
936 If you want to find the sum of a particular column there are several
937 ways, the obvious one is to use search:
939 my $rs = $schema->resultset('Items')->search(
942 select => [ { sum => 'Cost' } ],
943 as => [ 'total_cost' ], # remember this 'as' is for DBIx::Class::ResultSet not SQL
946 my $tc = $rs->first->get_column('total_cost');
948 Or, you can use the L<DBIx::Class::ResultSetColumn>, which gets
949 returned when you ask the C<ResultSet> for a column using
952 my $cost = $schema->resultset('Items')->get_column('Cost');
955 With this you can also do:
957 my $minvalue = $cost->min;
958 my $maxvalue = $cost->max;
960 Or just iterate through the values of this column only:
962 while ( my $c = $cost->next ) {
966 foreach my $c ($cost->all) {
970 C<ResultSetColumn> only has a limited number of built-in functions. If
971 you need one that it doesn't have, then you can use the C<func> method
974 my $avg = $cost->func('AVERAGE');
976 This will cause the following SQL statement to be run:
978 SELECT AVERAGE(Cost) FROM Items me
980 Which will of course only work if your database supports this function.
981 See L<DBIx::Class::ResultSetColumn> for more documentation.
983 =head2 Creating a result set from a set of rows
985 Sometimes you have a (set of) row objects that you want to put into a
986 resultset without the need to hit the DB again. You can do that by using the
987 L<set_cache|DBIx::Class::Resultset/set_cache> method:
989 my @uploadable_groups;
990 while (my $group = $groups->next) {
991 if ($group->can_upload($self)) {
992 push @uploadable_groups, $group;
995 my $new_rs = $self->result_source->resultset;
996 $new_rs->set_cache(\@uploadable_groups);
1000 =head1 USING RELATIONSHIPS
1002 =head2 Create a new row in a related table
1004 my $author = $book->create_related('author', { name => 'Fred'});
1006 =head2 Search in a related table
1008 Only searches for books named 'Titanic' by the author in $author.
1010 my $books_rs = $author->search_related('books', { name => 'Titanic' });
1012 =head2 Delete data in a related table
1014 Deletes only the book named Titanic by the author in $author.
1016 $author->delete_related('books', { name => 'Titanic' });
1018 =head2 Ordering a relationship result set
1020 If you always want a relation to be ordered, you can specify this when you
1021 create the relationship.
1023 To order C<< $book->pages >> by descending page_number, create the relation
1026 __PACKAGE__->has_many('pages' => 'Page', 'book', { order_by => \'page_number DESC'} );
1028 =head2 Filtering a relationship result set
1030 If you want to get a filtered result set, you can just add add to $attr as follows:
1032 __PACKAGE__->has_many('pages' => 'Page', 'book', { where => { scrap => 0 } } );
1034 =head2 Many-to-many relationships
1036 This is straightforward using L<ManyToMany|DBIx::Class::Relationship/many_to_many>:
1039 use base 'DBIx::Class';
1040 __PACKAGE__->load_components('Core');
1041 __PACKAGE__->table('user');
1042 __PACKAGE__->add_columns(qw/id name/);
1043 __PACKAGE__->set_primary_key('id');
1044 __PACKAGE__->has_many('user_address' => 'My::UserAddress', 'user');
1045 __PACKAGE__->many_to_many('addresses' => 'user_address', 'address');
1047 package My::UserAddress;
1048 use base 'DBIx::Class';
1049 __PACKAGE__->load_components('Core');
1050 __PACKAGE__->table('user_address');
1051 __PACKAGE__->add_columns(qw/user address/);
1052 __PACKAGE__->set_primary_key(qw/user address/);
1053 __PACKAGE__->belongs_to('user' => 'My::User');
1054 __PACKAGE__->belongs_to('address' => 'My::Address');
1056 package My::Address;
1057 use base 'DBIx::Class';
1058 __PACKAGE__->load_components('Core');
1059 __PACKAGE__->table('address');
1060 __PACKAGE__->add_columns(qw/id street town area_code country/);
1061 __PACKAGE__->set_primary_key('id');
1062 __PACKAGE__->has_many('user_address' => 'My::UserAddress', 'address');
1063 __PACKAGE__->many_to_many('users' => 'user_address', 'user');
1065 $rs = $user->addresses(); # get all addresses for a user
1066 $rs = $address->users(); # get all users for an address
1068 =head2 Relationships across DB schemas
1070 Mapping relationships across L<DB schemas|DBIx::Class::Manual::Glossary/DB schema>
1071 is easy as long as the schemas themselves are all accessible via the same DBI
1072 connection. In most cases, this means that they are on the same database host
1073 as each other and your connecting database user has the proper permissions to them.
1075 To accomplish this one only needs to specify the DB schema name in the table
1076 declaration, like so...
1078 package MyDatabase::Main::Artist;
1079 use base qw/DBIx::Class/;
1080 __PACKAGE__->load_components(qw/PK::Auto Core/);
1082 __PACKAGE__->table('database1.artist'); # will use "database1.artist" in FROM clause
1084 __PACKAGE__->add_columns(qw/ artist_id name /);
1085 __PACKAGE__->set_primary_key('artist_id');
1086 __PACKAGE__->has_many('cds' => 'MyDatabase::Main::Cd');
1090 Whatever string you specify there will be used to build the "FROM" clause in SQL
1093 The big drawback to this is you now have DB schema names hardcoded in your
1094 class files. This becomes especially troublesome if you have multiple instances
1095 of your application to support a change lifecycle (e.g. DEV, TEST, PROD) and
1096 the DB schemas are named based on the environment (e.g. database1_dev).
1098 However, one can dynamically "map" to the proper DB schema by overriding the
1099 L<connection|DBIx::Class::Schama/connection> method in your Schema class and
1100 building a renaming facility, like so:
1102 package MyDatabase::Schema;
1105 extends 'DBIx::Class::Schema';
1107 around connection => sub {
1108 my ( $inner, $self, $dsn, $username, $pass, $attr ) = ( shift, @_ );
1110 my $postfix = delete $attr->{schema_name_postfix};
1115 $self->append_db_name($postfix);
1119 sub append_db_name {
1120 my ( $self, $postfix ) = @_;
1124 { $_->name =~ /^\w+\./mx }
1126 { $self->source($_) }
1129 foreach my $source (@sources_with_db) {
1130 my $name = $source->name;
1131 $name =~ s{^(\w+)\.}{${1}${postfix}\.}mx;
1133 $source->name($name);
1139 By overridding the L<connection|DBIx::Class::Schama/connection>
1140 method and extracting a custom option from the provided \%attr hashref one can
1141 then simply iterate over all the Schema's ResultSources, renaming them as
1144 To use this facility, simply add or modify the \%attr hashref that is passed to
1145 L<connection|DBIx::Class::Schama/connect>, as follows:
1148 = MyDatabase::Schema->connect(
1153 schema_name_postfix => '_dev'
1154 # ... Other options as desired ...
1157 Obviously, one could accomplish even more advanced mapping via a hash map or a
1162 As of version 0.04001, there is improved transaction support in
1163 L<DBIx::Class::Storage> and L<DBIx::Class::Schema>. Here is an
1164 example of the recommended way to use it:
1166 my $genus = $schema->resultset('Genus')->find(12);
1168 my $coderef2 = sub {
1173 my $coderef1 = sub {
1174 $genus->add_to_species({ name => 'troglodyte' });
1177 $schema->txn_do($coderef2); # Can have a nested transaction. Only the outer will actualy commit
1178 return $genus->species;
1183 $rs = $schema->txn_do($coderef1);
1186 if ($@) { # Transaction failed
1187 die "the sky is falling!" #
1188 if ($@ =~ /Rollback failed/); # Rollback failed
1190 deal_with_failed_transaction();
1193 Nested transactions will work as expected. That is, only the outermost
1194 transaction will actually issue a commit to the $dbh, and a rollback
1195 at any level of any transaction will cause the entire nested
1196 transaction to fail. Support for savepoints and for true nested
1197 transactions (for databases that support them) will hopefully be added
1202 =head2 Creating Schemas From An Existing Database
1204 L<DBIx::Class::Schema::Loader> will connect to a database and create a
1205 L<DBIx::Class::Schema> and associated sources by examining the database.
1207 The recommend way of achieving this is to use the
1208 L<make_schema_at|DBIx::Class::Schema::Loader/make_schema_at> method:
1210 perl -MDBIx::Class::Schema::Loader=make_schema_at,dump_to_dir:./lib \
1211 -e 'make_schema_at("My::Schema", { debug => 1 }, [ "dbi:Pg:dbname=foo","postgres" ])'
1213 This will create a tree of files rooted at C<./lib/My/Schema/> containing
1214 source definitions for all the tables found in the C<foo> database.
1216 =head2 Creating DDL SQL
1218 The following functionality requires you to have L<SQL::Translator>
1219 (also known as "SQL Fairy") installed.
1221 To create a set of database-specific .sql files for the above schema:
1223 my $schema = My::Schema->connect($dsn);
1224 $schema->create_ddl_dir(['MySQL', 'SQLite', 'PostgreSQL'],
1229 By default this will create schema files in the current directory, for
1230 MySQL, SQLite and PostgreSQL, using the $VERSION from your Schema.pm.
1232 To create a new database using the schema:
1234 my $schema = My::Schema->connect($dsn);
1235 $schema->deploy({ add_drop_tables => 1});
1237 To import created .sql files using the mysql client:
1239 mysql -h "host" -D "database" -u "user" -p < My_Schema_1.0_MySQL.sql
1241 To create C<ALTER TABLE> conversion scripts to update a database to a
1242 newer version of your schema at a later point, first set a new
1243 C<$VERSION> in your Schema file, then:
1245 my $schema = My::Schema->connect($dsn);
1246 $schema->create_ddl_dir(['MySQL', 'SQLite', 'PostgreSQL'],
1252 This will produce new database-specific .sql files for the new version
1253 of the schema, plus scripts to convert from version 0.1 to 0.2. This
1254 requires that the files for 0.1 as created above are available in the
1255 given directory to diff against.
1257 =head2 Select from dual
1259 Dummy tables are needed by some databases to allow calling functions
1260 or expressions that aren't based on table content, for examples of how
1261 this applies to various database types, see:
1262 L<http://troels.arvin.dk/db/rdbms/#other-dummy_table>.
1264 Note: If you're using Oracles dual table don't B<ever> do anything
1265 other than a select, if you CRUD on your dual table you *will* break
1268 Make a table class as you would for any other table
1270 package MyAppDB::Dual;
1273 use base 'DBIx::Class';
1274 __PACKAGE__->load_components("Core");
1275 __PACKAGE__->table("Dual");
1276 __PACKAGE__->add_columns(
1278 { data_type => "VARCHAR2", is_nullable => 0, size => 1 },
1281 Once you've loaded your table class select from it using C<select>
1282 and C<as> instead of C<columns>
1284 my $rs = $schema->resultset('Dual')->search(undef,
1285 { select => [ 'sydate' ],
1290 All you have to do now is be careful how you access your resultset, the below
1291 will not work because there is no column called 'now' in the Dual table class
1293 while (my $dual = $rs->next) {
1294 print $dual->now."\n";
1296 # Can't locate object method "now" via package "MyAppDB::Dual" at headshot.pl line 23.
1298 You could of course use 'dummy' in C<as> instead of 'now', or C<add_columns> to
1299 your Dual class for whatever you wanted to select from dual, but that's just
1300 silly, instead use C<get_column>
1302 while (my $dual = $rs->next) {
1303 print $dual->get_column('now')."\n";
1308 my $cursor = $rs->cursor;
1309 while (my @vals = $cursor->next) {
1310 print $vals[0]."\n";
1313 In case you're going to use this "trick" together with L<DBIx::Class::Schema/deploy> or
1314 L<DBIx::Class::Schema/create_ddl_dir> a table called "dual" will be created in your
1315 current schema. This would overlap "sys.dual" and you could not fetch "sysdate" or
1316 "sequence.nextval" anymore from dual. To avoid this problem, just tell
1317 L<SQL::Translator> to not create table dual:
1320 add_drop_table => 1,
1321 parser_args => { sources => [ grep $_ ne 'Dual', schema->sources ] },
1323 $schema->create_ddl_dir( [qw/Oracle/], undef, './sql', undef, $sqlt_args );
1325 Or use L<DBIx::Class::ResultClass::HashRefInflator>
1327 $rs->result_class('DBIx::Class::ResultClass::HashRefInflator');
1328 while ( my $dual = $rs->next ) {
1329 print $dual->{now}."\n";
1332 Here are some example C<select> conditions to illustrate the different syntax
1333 you could use for doing stuff like
1334 C<oracles.heavily(nested(functions_can('take', 'lots'), OF), 'args')>
1336 # get a sequence value
1337 select => [ 'A_SEQ.nextval' ],
1339 # get create table sql
1340 select => [ { 'dbms_metadata.get_ddl' => [ "'TABLE'", "'ARTIST'" ]} ],
1342 # get a random num between 0 and 100
1343 select => [ { "trunc" => [ { "dbms_random.value" => [0,100] } ]} ],
1346 select => [ { 'extract' => [ \'year from sysdate' ] } ],
1349 select => [ {'round' => [{'cos' => [ \'180 * 3.14159265359/180' ]}]}],
1351 # which day of the week were you born on?
1352 select => [{'to_char' => [{'to_date' => [ "'25-DEC-1980'", "'dd-mon-yyyy'" ]}, "'day'"]}],
1354 # select 16 rows from dual
1355 select => [ "'hello'" ],
1357 group_by => [ 'cube( 1, 2, 3, 4 )' ],
1361 =head2 Adding Indexes And Functions To Your SQL
1363 Often you will want indexes on columns on your table to speed up searching. To
1364 do this, create a method called C<sqlt_deploy_hook> in the relevant source
1365 class (refer to the advanced
1366 L<callback system|DBIx::Class::ResultSource/sqlt_deploy_callback> if you wish
1367 to share a hook between multiple sources):
1369 package My::Schema::Result::Artist;
1371 __PACKAGE__->table('artist');
1372 __PACKAGE__->add_columns(id => { ... }, name => { ... })
1374 sub sqlt_deploy_hook {
1375 my ($self, $sqlt_table) = @_;
1377 $sqlt_table->add_index(name => 'idx_name', fields => ['name']);
1382 Sometimes you might want to change the index depending on the type of the
1383 database for which SQL is being generated:
1385 my ($db_type = $sqlt_table->schema->translator->producer_type)
1386 =~ s/^SQL::Translator::Producer:://;
1388 You can also add hooks to the schema level to stop certain tables being
1395 sub sqlt_deploy_hook {
1396 my ($self, $sqlt_schema) = @_;
1398 $sqlt_schema->drop_table('table_name');
1401 You could also add views, procedures or triggers to the output using
1402 L<SQL::Translator::Schema/add_view>,
1403 L<SQL::Translator::Schema/add_procedure> or
1404 L<SQL::Translator::Schema/add_trigger>.
1407 =head2 Schema versioning
1409 The following example shows simplistically how you might use DBIx::Class to
1410 deploy versioned schemas to your customers. The basic process is as follows:
1416 Create a DBIx::Class schema
1428 Modify schema to change functionality
1432 Deploy update to customers
1436 B<Create a DBIx::Class schema>
1438 This can either be done manually, or generated from an existing database as
1439 described under L</Creating Schemas From An Existing Database>
1443 Call L<DBIx::Class::Schema/create_ddl_dir> as above under L</Creating DDL SQL>.
1445 B<Deploy to customers>
1447 There are several ways you could deploy your schema. These are probably
1448 beyond the scope of this recipe, but might include:
1454 Require customer to apply manually using their RDBMS.
1458 Package along with your app, making database dump/schema update/tests
1459 all part of your install.
1463 B<Modify the schema to change functionality>
1465 As your application evolves, it may be necessary to modify your schema
1466 to change functionality. Once the changes are made to your schema in
1467 DBIx::Class, export the modified schema and the conversion scripts as
1468 in L</Creating DDL SQL>.
1470 B<Deploy update to customers>
1472 Add the L<DBIx::Class::Schema::Versioned> schema component to your
1473 Schema class. This will add a new table to your database called
1474 C<dbix_class_schema_vesion> which will keep track of which version is installed
1475 and warn if the user trys to run a newer schema version than the
1476 database thinks it has.
1478 Alternatively, you can send the conversion sql scripts to your
1481 =head2 Setting quoting for the generated SQL.
1483 If the database contains column names with spaces and/or reserved words, they
1484 need to be quoted in the SQL queries. This is done using:
1486 __PACKAGE__->storage->sql_maker->quote_char([ qw/[ ]/] );
1487 __PACKAGE__->storage->sql_maker->name_sep('.');
1489 The first sets the quote characters. Either a pair of matching
1490 brackets, or a C<"> or C<'>:
1492 __PACKAGE__->storage->sql_maker->quote_char('"');
1494 Check the documentation of your database for the correct quote
1495 characters to use. C<name_sep> needs to be set to allow the SQL
1496 generator to put the quotes the correct place.
1498 In most cases you should set these as part of the arguments passed to
1499 L<DBIx::Class::Schema/connect>:
1501 my $schema = My::Schema->connect(
1511 =head2 Setting limit dialect for SQL::Abstract::Limit
1513 In some cases, SQL::Abstract::Limit cannot determine the dialect of
1514 the remote SQL server by looking at the database handle. This is a
1515 common problem when using the DBD::JDBC, since the DBD-driver only
1516 know that in has a Java-driver available, not which JDBC driver the
1517 Java component has loaded. This specifically sets the limit_dialect
1518 to Microsoft SQL-server (See more names in SQL::Abstract::Limit
1521 __PACKAGE__->storage->sql_maker->limit_dialect('mssql');
1523 The JDBC bridge is one way of getting access to a MSSQL server from a platform
1524 that Microsoft doesn't deliver native client libraries for. (e.g. Linux)
1526 The limit dialect can also be set at connect time by specifying a
1527 C<limit_dialect> key in the final hash as shown above.
1529 =head2 Working with PostgreSQL array types
1531 You can also assign values to PostgreSQL array columns by passing array
1532 references in the C<\%columns> (C<\%vals>) hashref of the
1533 L<DBIx::Class::ResultSet/create> and L<DBIx::Class::Row/update> family of
1536 $resultset->create({
1537 numbers => [1, 2, 3]
1542 numbers => [1, 2, 3]
1549 In conditions (eg. C<\%cond> in the L<DBIx::Class::ResultSet/search> family of
1550 methods) you cannot directly use array references (since this is interpreted as
1551 a list of values to be C<OR>ed), but you can use the following syntax to force
1552 passing them as bind values:
1556 numbers => \[ '= ?', [numbers => [1, 2, 3]] ]
1560 See L<SQL::Abstract/array_datatypes> and L<SQL::Abstract/Literal SQL with
1561 placeholders and bind values (subqueries)> for more explanation. Note that
1562 L<DBIx::Class> sets L<SQL::Abstract/bindtype> to C<columns>, so you must pass
1563 the bind values (the C<[1, 2, 3]> arrayref in the above example) wrapped in
1564 arrayrefs together with the column name, like this: C<< [column_name => value]
1567 =head1 BOOTSTRAPPING/MIGRATING
1569 =head2 Easy migration from class-based to schema-based setup
1571 You want to start using the schema-based approach to L<DBIx::Class>
1572 (see L<SchemaIntro.pod>), but have an established class-based setup with lots
1573 of existing classes that you don't want to move by hand. Try this nifty script
1577 use SQL::Translator;
1579 my $schema = MyDB->schema_instance;
1581 my $translator = SQL::Translator->new(
1582 debug => $debug || 0,
1583 trace => $trace || 0,
1584 no_comments => $no_comments || 0,
1585 show_warnings => $show_warnings || 0,
1586 add_drop_table => $add_drop_table || 0,
1587 validate => $validate || 0,
1589 'DBIx::Schema' => $schema,
1592 'prefix' => 'My::Schema',
1596 $translator->parser('SQL::Translator::Parser::DBIx::Class');
1597 $translator->producer('SQL::Translator::Producer::DBIx::Class::File');
1599 my $output = $translator->translate(@args) or die
1600 "Error: " . $translator->error;
1604 You could use L<Module::Find> to search for all subclasses in the MyDB::*
1605 namespace, which is currently left as an exercise for the reader.
1607 =head1 OVERLOADING METHODS
1609 L<DBIx::Class> uses the L<Class::C3> package, which provides for redispatch of
1610 method calls, useful for things like default values and triggers. You have to
1611 use calls to C<next::method> to overload methods. More information on using
1612 L<Class::C3> with L<DBIx::Class> can be found in
1613 L<DBIx::Class::Manual::Component>.
1615 =head2 Setting default values for a row
1617 It's as simple as overriding the C<new> method. Note the use of
1621 my ( $class, $attrs ) = @_;
1623 $attrs->{foo} = 'bar' unless defined $attrs->{foo};
1625 my $new = $class->next::method($attrs);
1630 For more information about C<next::method>, look in the L<Class::C3>
1631 documentation. See also L<DBIx::Class::Manual::Component> for more
1632 ways to write your own base classes to do this.
1634 People looking for ways to do "triggers" with DBIx::Class are probably
1635 just looking for this.
1637 =head2 Changing one field whenever another changes
1639 For example, say that you have three columns, C<id>, C<number>, and
1640 C<squared>. You would like to make changes to C<number> and have
1641 C<squared> be automagically set to the value of C<number> squared.
1642 You can accomplish this by overriding C<store_column>:
1645 my ( $self, $name, $value ) = @_;
1646 if ($name eq 'number') {
1647 $self->squared($value * $value);
1649 $self->next::method($name, $value);
1652 Note that the hard work is done by the call to C<next::method>, which
1653 redispatches your call to store_column in the superclass(es).
1655 =head2 Automatically creating related objects
1657 You might have a class C<Artist> which has many C<CD>s. Further, you
1658 want to create a C<CD> object every time you insert an C<Artist> object.
1659 You can accomplish this by overriding C<insert> on your objects:
1662 my ( $self, @args ) = @_;
1663 $self->next::method(@args);
1664 $self->cds->new({})->fill_from_artist($self)->insert;
1668 where C<fill_from_artist> is a method you specify in C<CD> which sets
1669 values in C<CD> based on the data in the C<Artist> object you pass in.
1671 =head2 Wrapping/overloading a column accessor
1675 Say you have a table "Camera" and want to associate a description
1676 with each camera. For most cameras, you'll be able to generate the description from
1677 the other columns. However, in a few special cases you may want to associate a
1678 custom description with a camera.
1682 In your database schema, define a description field in the "Camera" table that
1683 can contain text and null values.
1685 In DBIC, we'll overload the column accessor to provide a sane default if no
1686 custom description is defined. The accessor will either return or generate the
1687 description, depending on whether the field is null or not.
1689 First, in your "Camera" schema class, define the description field as follows:
1691 __PACKAGE__->add_columns(description => { accessor => '_description' });
1693 Next, we'll define the accessor-wrapper subroutine:
1698 # If there is an update to the column, we'll let the original accessor
1700 return $self->_description(@_) if @_;
1702 # Fetch the column value.
1703 my $description = $self->_description;
1705 # If there's something in the description field, then just return that.
1706 return $description if defined $description && length $descripton;
1708 # Otherwise, generate a description.
1709 return $self->generate_description;
1712 =head1 DEBUGGING AND PROFILING
1714 =head2 DBIx::Class objects with Data::Dumper
1716 L<Data::Dumper> can be a very useful tool for debugging, but sometimes it can
1717 be hard to find the pertinent data in all the data it can generate.
1718 Specifically, if one naively tries to use it like so,
1722 my $cd = $schema->resultset('CD')->find(1);
1725 several pages worth of data from the CD object's schema and result source will
1726 be dumped to the screen. Since usually one is only interested in a few column
1727 values of the object, this is not very helpful.
1729 Luckily, it is possible to modify the data before L<Data::Dumper> outputs
1730 it. Simply define a hook that L<Data::Dumper> will call on the object before
1731 dumping it. For example,
1738 result_source => undef,
1746 local $Data::Dumper::Freezer = '_dumper_hook';
1748 my $cd = $schema->resultset('CD')->find(1);
1750 # dumps $cd without its ResultSource
1752 If the structure of your schema is such that there is a common base class for
1753 all your table classes, simply put a method similar to C<_dumper_hook> in the
1754 base class and set C<$Data::Dumper::Freezer> to its name and L<Data::Dumper>
1755 will automagically clean up your data before printing it. See
1756 L<Data::Dumper/EXAMPLES> for more information.
1760 When you enable L<DBIx::Class::Storage>'s debugging it prints the SQL
1761 executed as well as notifications of query completion and transaction
1762 begin/commit. If you'd like to profile the SQL you can subclass the
1763 L<DBIx::Class::Storage::Statistics> class and write your own profiling
1766 package My::Profiler;
1769 use base 'DBIx::Class::Storage::Statistics';
1771 use Time::HiRes qw(time);
1780 $self->print("Executing $sql: ".join(', ', @params)."\n");
1789 my $elapsed = sprintf("%0.4f", time() - $start);
1790 $self->print("Execution took $elapsed seconds.\n");
1796 You can then install that class as the debugging object:
1798 __PACKAGE__->storage->debugobj(new My::Profiler());
1799 __PACKAGE__->storage->debug(1);
1801 A more complicated example might involve storing each execution of SQL in an
1809 my $elapsed = time() - $start;
1810 push(@{ $calls{$sql} }, {
1816 You could then create average, high and low execution times for an SQL
1817 statement and dig down to see if certain parameters cause aberrant behavior.
1818 You might want to check out L<DBIx::Class::QueryLog> as well.
1820 =head1 STARTUP SPEED
1822 L<DBIx::Class|DBIx::Class> programs can have a significant startup delay
1823 as the ORM loads all the relevant classes. This section examines
1824 techniques for reducing the startup delay.
1826 These tips are are listed in order of decreasing effectiveness - so the
1827 first tip, if applicable, should have the greatest effect on your
1830 =head2 Statically Define Your Schema
1833 L<DBIx::Class::Schema::Loader|DBIx::Class::Schema::Loader> to build the
1834 classes dynamically based on the database schema then there will be a
1835 significant startup delay.
1837 For production use a statically defined schema (which can be generated
1838 using L<DBIx::Class::Schema::Loader|DBIx::Class::Schema::Loader> to dump
1839 the database schema once - see
1840 L<make_schema_at|DBIx::Class::Schema::Loader/make_schema_at> and
1841 L<dump_directory|DBIx::Class::Schema::Loader/dump_directory> for more
1842 details on creating static schemas from a database).
1844 =head2 Move Common Startup into a Base Class
1846 Typically L<DBIx::Class> result classes start off with
1848 use base qw/DBIx::Class/;
1849 __PACKAGE__->load_components(qw/InflateColumn::DateTime Core/);
1851 If this preamble is moved into a common base class:-
1855 use base qw/DBIx::Class/;
1856 __PACKAGE__->load_components(qw/InflateColumn::DateTime Core/);
1859 and each result class then uses this as a base:-
1861 use base qw/MyDBICbase/;
1863 then the load_components is only performed once, which can result in a
1864 considerable startup speedup for schemas with many classes.
1866 =head2 Explicitly List Schema Result Classes
1868 The schema class will normally contain
1870 __PACKAGE__->load_classes();
1872 to load the result classes. This will use L<Module::Find|Module::Find>
1873 to find and load the appropriate modules. Explicitly defining the
1874 classes you wish to load will remove the overhead of
1875 L<Module::Find|Module::Find> and the related directory operations:
1877 __PACKAGE__->load_classes(qw/ CD Artist Track /);
1879 If you are instead using the L<load_namespaces|DBIx::Class::Schema/load_namespaces>
1880 syntax to load the appropriate classes there is not a direct alternative
1881 avoiding L<Module::Find|Module::Find>.
1885 =head2 Cached statements
1887 L<DBIx::Class> normally caches all statements with L<< prepare_cached()|DBI/prepare_cached >>.
1888 This is normally a good idea, but if too many statements are cached, the database may use too much
1889 memory and may eventually run out and fail entirely. If you suspect this may be the case, you may want
1890 to examine DBI's L<< CachedKids|DBI/CachedKidsCachedKids_(hash_ref) >> hash:
1892 # print all currently cached prepared statements
1893 print for keys %{$schema->storage->dbh->{CachedKids}};
1894 # get a count of currently cached prepared statements
1895 my $count = scalar keys %{$schema->storage->dbh->{CachedKids}};
1897 If it's appropriate, you can simply clear these statements, automatically deallocating them in the
1900 my $kids = $schema->storage->dbh->{CachedKids};
1901 delete @{$kids}{keys %$kids} if scalar keys %$kids > 100;
1903 But what you probably want is to expire unused statements and not those that are used frequently.
1904 You can accomplish this with L<Tie::Cache> or L<Tie::Cache::LRU>:
1908 my $schema = DB::Main->connect($dbi_dsn, $user, $pass, {
1909 on_connect_do => sub { tie %{shift->_dbh->{CachedKids}}, 'Tie::Cache', 100 },