From Code to Community: Sponsoring The Perl and Raku Conference 2025 Learn more

NAME

DBIx::Class::InflateColumn::JSON2Object - convert JSON columns to Perl objects

VERSION

version 0.907

SYNOPSIS

# In a DBIx::Class Result
... lots of DBIx::Class code...
DBIx::Class::InflateColumn::JSON2Object->fixed_class({
column=>'data', # a column storing JSON
class=>'MyApp::SomeClass',
});
# later, in some code far, far away...
my $row = $schema->resultset('SomeTable')->find(42);
my $obj = $row->data;
$obj->foo; # $obj ISA MyApp::SomeClass
# store a hash as JSON after "validating" it through MyApp::SomeClass
$schema->resultset('SomeTable')->create({
data => {
foo=>'bar',
};
})
# you can also use it to just deflate/inflate JSON to a Perl hash
DBIx::Class::InflateColumn::JSON2Object->no_class({
column=>'args',
});
# or have a complex set of objects
DBIx::Class::InflateColumn::JSON2Object->class_in_column({
data_column => 'object', # some JSON
class_column => 'type', # here we store the name of the object
namespace => 'MyApp::Object', # the namespace of the objects
});
# or store a list of objects (of one class!) into an array
DBIx::Class::InflateColumn::JSON2Object->array_of_class({
column=>'data', # a column storing a JSON list
class=>'MyApp::SomeElementClass',
});

DESCRIPTION

TODO: short overview

Booleans, JSON, oh my...

TODO: describe problem and the (hacky/crappy?) solution

METHODS

Please note that you can pass more than one HASHREF per method to install several inflator/deflators at once.

no_class

Install a JSON inflator/deflator for each column.

DBIx::Class::InflateColumn::JSON2Object->no_class({
column=>'args'
});

You can pass a Perl datastructure to the row and it will be stored as JSON:

$resultset->create( {
id => 123,
args => {
some => 'data',
more => [1,1,2,3,5,8]
}
});
# will be stored as '{"some":"data","more":[1,1,2,3,5,8]}'

You can also access the data directly as a Perl hash:

my $row = $resultset->find(123);
$row->args->{some}; # 'data'
$row->args->{more}[5]; # 8

fixed_class

If plain JSON is to wobbly for you, you can define Moose objects and have them serialized to JSON. Not only can you now add some custom methods to the objects, but you can (ab)use the object initalisation and all features Moose provides to define your objects.

# Define a class
use Moose;
has 'name' => ( is=>'ro', isa=>'Str');
has 'amount => ( is=>'ro', isa=>'Int');
__PACKAGE__->meta->make_immutable;
# In your Result class, map the class to a column
__PACKAGE__->add_columns(
"sweets",
{ data_type => "jsonb", is_nullable => 1 }
);
DBIx::Class::InflateColumn::JSON2Object->fixed_class({
column=>'sweets',
class=>'YourApp::Sweets',
});

Now you can pass a plain hash, a raw JSON string, or an initiated object to the row, and always get back the object:

my $thing = $schema->resultset('Thing')->create({
sweets => {
name => 'Manner Schnitten',
amount => 10,
}
});
ref($thing->sweets); # 'YourApp::Sweets'

class_in_column

Sometimes you have a set of similar objects you want to store. TODO: explain oe1.article.paragraph

TODO ->class_in_column(..)

You can pass an object to the row and the infered type will be stored together with the JSON-payload:

TODO

To get back the object:

TODO

array_in_class

Sometimes you want to store a list of objects without packing them into a "Set"-Object. This only works if all objects are of the same class, as we do not want to include the class name inside the object (c.f. MooseX::Storage, KiokuDB).

TODO ->array_in_class(..)

Just pass an arrayref of object to the row and it will be stored as JSON:

TODO

And you get back the initated list of objects:

TODO

THANKS

AUTHOR

Thomas Klausner <domm@plix.at>

COPYRIGHT AND LICENSE

This software is copyright (c) 2017 - 2021 by Thomas Klausner.

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