1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
|
# $Id$
package Data::ObjectDriver::Driver::Multiplexer;
use strict;
use warnings;
use Storable();
use base qw( Data::ObjectDriver Class::Accessor::Fast );
__PACKAGE__->mk_accessors(qw( on_search on_lookup drivers ));
use Carp qw( croak );
sub init {
my $driver = shift;
$driver->SUPER::init(@_);
my %param = @_;
for my $key (qw( on_search on_lookup drivers )) {
$driver->$key( $param{$key} );
}
return $driver;
}
sub lookup {
my $driver = shift;
my $subdriver = $driver->on_lookup;
croak "on_lookup is not defined in $driver"
unless $subdriver;
return $subdriver->lookup(@_);
}
sub fetch_data {
my $driver = shift;
my $subdriver = $driver->on_lookup;
croak "on_lookup is not defined in $driver"
unless $subdriver;
return $subdriver->fetch_data(@_);
}
sub lookup_multi {
my $driver = shift;
my $subdriver = $driver->on_lookup;
croak "on_lookup is not defined in $driver"
unless $subdriver;
return $subdriver->lookup_multi(@_);
}
sub exists {
my $driver = shift;
my($class, $terms, $args) = @_;
## just assume that the first driver declared is the more efficient one
my $sub_driver = $driver->drivers->[0];
return $sub_driver->exists(@_);
}
sub search {
my $driver = shift;
my($class, $terms, $args) = @_;
my $sub_driver = $driver->_find_sub_driver($terms)
or croak "No matching sub-driver found";
return $sub_driver->search(@_);
}
sub replace { shift->_exec_multiplexed('replace', @_) }
sub insert { shift->_exec_multiplexed('insert', @_) }
sub update { shift->_exec_multiplexed('update', @_) }
sub remove {
my $driver = shift;
my $removed = 0;
for my $sub_driver (@{ $driver->drivers }) {
$removed += $sub_driver->remove(@_);
}
if ($removed % 2) {
warn "remove count looks incorrect, we might miss one object";
}
return $removed || 0E0;
}
sub _find_sub_driver {
my $driver = shift;
my($terms) = @_;
for my $key (keys %$terms) {
if (my $sub_driver = $driver->on_search->{$key}) {
return $sub_driver;
}
}
}
sub _exec_multiplexed {
my $driver = shift;
my($meth, $obj, @args) = @_;
my $orig_obj = Storable::dclone($obj);
my $ret;
## We want to be sure to have the initial and final state of the object
## strictly identical as if we made only one call on $obj
## (Perhaps it's a bit overkill ? playing with 'changed_cols' may do the trick)
for my $sub_driver (@{ $driver->drivers }) {
$obj = Storable::dclone($orig_obj);
$ret = $sub_driver->$meth($obj, @args);
}
return $ret;
}
sub begin_work {
my $driver = shift;
$driver->SUPER::begin_work(@_);
for my $sub_driver (@{ $driver->drivers }) {
$sub_driver->begin_work;
}
}
sub commit {
my $driver = shift;
$driver->SUPER::commit(@_);
$driver->_end_txn('commit', @_);
}
sub rollback {
my $driver = shift;
$driver->SUPER::rollback(@_);
$driver->_end_txn('rollback', @_);
}
sub _end_txn {
my ($driver, $method) = @_;
for my $sub_driver (@{ $driver->drivers }) {
$sub_driver->$method;
}
}
1;
__END__
=head1 NAME
Data::ObjectDriver::Driver::Multiplexer - Multiplex multiple partitioned drivers
=head1 SYNOPSIS
package MappingTable;
use Foo;
use Bar;
my $foo_driver = Foo->driver;
my $bar_driver = Bar->driver;
__PACKAGE__->install_properties({
columns => [ qw( foo_id bar_id value ) ],
primary_key => 'foo_id',
driver => Data::ObjectDriver::Driver::Multiplexer->new(
on_search => {
foo_id => $foo_driver,
bar_id => $bar_driver,
},
on_lookup => $foo_driver,
drivers => [ $foo_driver, $bar_driver ],
),
});
=head1 DESCRIPTION
I<Data::ObjectDriver::Driver::Multiplexer> associates a set of drivers to
a particular class. In practice, this means that all INSERTs and DELETEs
are propagated to all associated drivers (for example, all associated
databases or tables in a database), and that SELECTs are sent to the
appropriate multiplexed driver, based on partitioning criteria.
Note that this driver has the following limitations currently:
=over 4
=item 1. It's very experimental.
=item 2. It's very experimental.
=item 3. IT'S VERY EXPERIMENTAL.
=item 4. This documentation you're reading is incomplete. the api is likely
to evolve
=back
=cut
|
