NAME

SYNOPSIS

 package MyApp::Schema::ResultSet::Foo;

 __PACKAGE__->load_components(qw{Helper::ResultSet::SetOperations});

 ...

 1;

And then elsewhere, like in a controller:

 my $rs1 = $rs->search({ foo => 'bar' });
 my $rs2 = $rs->search({ baz => 'biff' });
 for ($rs1->union($rs2)->all) { ... }

DESCRIPTION

Component throws exceptions if ResultSets have different ResultClasses or different "Columns Specs."

The basic idea here is that in SQL if you use a set operation they must be selecting the same columns names, so that the results will all match. The deal with the ResultClasses is that DBIC needs to inflate the results the same for the entire ResultSet, so if one were to try to apply something like a union in a table with the same column name but different classes DBIC wouldn't be doing what you would expect.

A nice way to use this is with DBIx::Class::ResultClass::HashRefInflator.

You might have something like the following sketch autocompletion code:

 my $rs1 = $schema->resultset('Album')->search({
    name => { -like => "$input%" }
 }, {
   columns => [qw( id name ), {
      tablename => \['?', [{} => 'album']],
   }],
 });

 my $rs2 = $schema->resultset('Artist')->search({
    name => { -like => "$input%" }
 }, {
   columns => [qw( id name ), {
      tablename => \['?', [{} => 'artist']],
   }],
 });

 my $rs3 = $schema->resultset('Song')->search({
    name => { -like => "$input%" }
 }, {
   columns => [qw( id name ), {
      tablename => \['?', [{} => 'song']],
   }],
 });

 $_->result_class('DBIx::Class::ResultClass::HashRefInflator')
   for ($rs1, $rs2, $rs3);

 my $data = [$rs1->union([$rs2, $rs3])->all];

METHODS

AUTHOR

COPYRIGHT AND LICENSE

This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself.