DBIx::Class::Helper::ResultClass::Tee(3pm)

DBIx::Class::Helper::ResultClass::Tee3pm

Inflate to multiple result classes at the same time

Usage

   my ($hashref, $obj) = $rs->search(undef, {
      result_class => DBIx::Class::Helper::ResultClass::Tee->new(
         inner_classes => [ '::HRI', 'MyApp::Schema::Result::User'],
      ),
   })->first->@*;

(If you've never seen "->@*" before, check out "Postfix-Dereference-Syntax" in perlref, added in Perl v5.20!)

Overview

This result class has one obvious use case: when you have prefetched data and DBIx::Class::ResultClass::HashRefInflator is the simplest way to access all the data, but you still want to use some of the methods on your existing result class.

The other important raison d'être of this module is that it is an example of how to make a "parameterized" result class. It's almost a secret that DBIx::Class supports using objects to inflate results. This is an incredibly powerful feature that can be used to make consistent interfaces to do all kinds of things.

Once when I was at Micro Technology Services, Inc. I used it to efficiently do a "reverse synthetic, LIKE-ish join". The "relationship" was basically "foreign.name =~ self.name", which cannot actually be done if you want to go from within the database, but if you are able to load the entire foreign table into memory this can be done on-demand, and cached within the result class for (in our case) the duration of a request.

Releases

5 releases since Debian 9 (Stretch); the text changed 4 times.

  1. 9
  2. 10
  3. 11
  4. 12
  5. 13

Author

Arthur Axel "fREW" Schmidt <[email protected]>

From libdbix-class-helpers-perl 2.037000-1 in Debian 13 (Trixie). Parsed by carta from the page's man source.