package DBIx::Class::Storage::DBI::Replicated;
BEGIN {
- use Carp::Clan qw/^DBIx::Class/;
use DBIx::Class;
- croak('The following modules are required for Replication ' . DBIx::Class::Optional::Dependencies->req_missing_for ('replicated') )
+ die('The following modules are required for Replication ' . DBIx::Class::Optional::Dependencies->req_missing_for ('replicated') . "\n" )
unless DBIx::Class::Optional::Dependencies->req_ok_for ('replicated');
}
use DBIx::Class::Storage::DBI::Replicated::Types qw/BalancerClassNamePart DBICSchema DBICStorageDBI/;
use MooseX::Types::Moose qw/ClassName HashRef Object/;
use Scalar::Util 'reftype';
-use Hash::Merge 'merge';
-use List::Util qw/min max/;
+use Hash::Merge;
+use List::Util qw/min max reduce/;
+use Context::Preserve 'preserve_context';
+use Try::Tiny;
use namespace::clean -except => 'meta';
+=encoding utf8
+
=head1 NAME
DBIx::Class::Storage::DBI::Replicated - BETA Replicated database support
=head1 SYNOPSIS
The Following example shows how to change an existing $schema to a replicated
-storage type, add some replicated (readonly) databases, and perform reporting
+storage type, add some replicated (read-only) databases, and perform reporting
tasks.
You should set the 'storage_type attribute to a replicated type. You should
that the Pool object should get.
my $schema = Schema::Class->clone;
- $schema->storage_type( ['::DBI::Replicated', {balancer=>'::Random'}] );
+ $schema->storage_type(['::DBI::Replicated', { balancer_type => '::Random' }]);
$schema->connection(...);
-Next, you need to add in the Replicants. Basically this is an array of
+Next, you need to add in the Replicants. Basically this is an array of
arrayrefs, where each arrayref is database connect information. Think of these
arguments as what you'd pass to the 'normal' $schema->connect method.
You can force a given query to use a particular storage using the search
attribute 'force_pool'. For example:
- my $RS = $schema->resultset('Source')->search(undef, {force_pool=>'master'});
+ my $rs = $schema->resultset('Source')->search(undef, {force_pool=>'master'});
-Now $RS will force everything (both reads and writes) to use whatever was setup
-as the master storage. 'master' is hardcoded to always point to the Master,
+Now $rs will force everything (both reads and writes) to use whatever was setup
+as the master storage. 'master' is hardcoded to always point to the Master,
but you can also use any Replicant name. Please see:
L<DBIx::Class::Storage::DBI::Replicated::Pool> and the replicants attribute for more.
in a transaction when you are reading and writing to the same tables at the
same time, since your replicants will often lag a bit behind the master.
-See L<DBIx::Class::Storage::DBI::Replicated::Instructions> for more help and
-walkthroughs.
+If you have a multi-statement read only transaction you can force it to select
+a random server in the pool by:
+
+ my $rs = $schema->resultset('Source')->search( undef,
+ { force_pool => $db->storage->read_handler->next_storage }
+ );
=head1 DESCRIPTION
Warning: This class is marked BETA. This has been running a production
website using MySQL native replication as its backend and we have some decent
test coverage but the code hasn't yet been stressed by a variety of databases.
-Individual DB's may have quirks we are not aware of. Please use this in first
+Individual DBs may have quirks we are not aware of. Please use this in first
development and pass along your experiences/bug fixes.
This class implements replicated data store for DBI. Currently you can define
to all existing storages. This way our storage class is a drop in replacement
for L<DBIx::Class::Storage::DBI>.
-Read traffic is spread across the replicants (slaves) occuring to a user
+Read traffic is spread across the replicants (slaves) occurring to a user
selected algorithm. The default algorithm is random weighted.
=head1 NOTES
-The consistency betweeen master and replicants is database specific. The Pool
+The consistency between master and replicants is database specific. The Pool
gives you a method to validate its replicants, removing and replacing them
when they fail/pass predefined criteria. Please make careful use of the ways
to force a query to run against Master when needed.
=head2 pool_type
-Contains the classname which will instantiate the L</pool> object. Defaults
+Contains the classname which will instantiate the L</pool> object. Defaults
to: L<DBIx::Class::Storage::DBI::Replicated::Pool>.
=cut
=head2 pool
-Is a <DBIx::Class::Storage::DBI::Replicated::Pool> or derived class. This is a
+Is a L<DBIx::Class::Storage::DBI::Replicated::Pool> or derived class. This is a
container class for one or more replicated databases.
=cut
=head2 balancer
-Is a <DBIx::Class::Storage::DBI::Replicated::Balancer> or derived class. This
-is a class that takes a pool (<DBIx::Class::Storage::DBI::Replicated::Pool>)
+Is a L<DBIx::Class::Storage::DBI::Replicated::Balancer> or derived class. This
+is a class that takes a pool (L<DBIx::Class::Storage::DBI::Replicated::Pool>)
=cut
=head1 ATTRIBUTES IMPLEMENTING THE DBIx::Storage::DBI INTERFACE
-The following methods are delegated all the methods required for the
+The following methods are delegated all the methods required for the
L<DBIx::Class::Storage::DBI> interface.
-=head2 read_handler
-
-Defines an object that implements the read side of L<BIx::Class::Storage::DBI>.
-
-=cut
-
-has 'read_handler' => (
- is=>'rw',
- isa=>Object,
- lazy_build=>1,
- handles=>[qw/
- select
- select_single
- columns_info_for
- _dbh_columns_info_for
- _select
- /],
-);
-
-=head2 write_handler
-
-Defines an object that implements the write side of L<BIx::Class::Storage::DBI>,
-as well as methods that don't write or read that can be called on only one
-storage, methods that return a C<$dbh>, and any methods that don't make sense to
-run on a replicant.
-
=cut
-has 'write_handler' => (
- is=>'ro',
- isa=>Object,
- lazy_build=>1,
- handles=>[qw/
+my $method_dispatch = {
+ writer => [qw/
on_connect_do
on_disconnect_do
on_connect_call
build_datetime_parser
last_insert_id
insert
- insert_bulk
update
delete
dbh
txn_commit
txn_rollback
txn_scope_guard
- sth
+ _exec_txn_rollback
+ _exec_txn_begin
+ _exec_txn_commit
deploy
with_deferred_fk_checks
dbh_do
- reload_row
- with_deferred_fk_checks
_prep_for_execute
-
- backup
is_datatype_numeric
_count_select
- _subq_count_select
- _subq_update_delete
svp_rollback
svp_begin
svp_release
relname_to_table_alias
- _straight_join_to_node
_dbh_last_insert_id
- _fix_bind_params
_default_dbi_connect_attributes
_dbi_connect_info
+ _dbic_connect_attributes
auto_savepoint
- _sqlt_version_ok
+ _query_start
_query_end
+ _format_for_trace
+ _dbi_attrs_for_bind
bind_attribute_by_data_type
transaction_depth
_dbh
_select_args
- _dbh_execute_array
- _sql_maker_args
+ _dbh_execute_for_fetch
_sql_maker
- _query_start
- _sqlt_version_error
- _per_row_update_delete
- _dbh_begin_work
_dbh_execute_inserts_with_no_binds
_select_args_to_query
+ _gen_sql_bind
_svp_generate_name
- _multipk_update_delete
- source_bind_attributes
_normalize_connect_info
_parse_connect_do
- _dbh_commit
- _execute_array
- _placeholders_supported
- _verify_pid
savepoints
- _sqlt_minimum_version
_sql_maker_opts
_conn_pid
- _typeless_placeholders_supported
- _conn_tid
_dbh_autocommit
_native_data_type
_get_dbh
sql_maker_class
- _dbh_rollback
- _adjust_select_args_for_complex_prefetch
- _resolve_ident_sources
- _resolve_column_info
- _prune_unused_joins
- _strip_cond_qualifiers
- _parse_order_by
- _resolve_aliastypes_from_select_args
+ insert_bulk
+ _insert_bulk
_execute
_do_query
- _dbh_sth
_dbh_execute
+ /, Class::MOP::Class->initialize('DBIx::Class::Storage::DBIHacks')->get_method_list ],
+ reader => [qw/
+ select
+ select_single
+ columns_info_for
+ _dbh_columns_info_for
+ _select
/],
+ unimplemented => [qw/
+ _arm_global_destructor
+ _verify_pid
+
+ get_use_dbms_capability
+ set_use_dbms_capability
+ get_dbms_capability
+ set_dbms_capability
+ _dbh_details
+ _dbh_get_info
+
+ _determine_connector_driver
+ _extract_driver_from_connect_info
+ _describe_connection
+ _warn_undetermined_driver
+
+ sql_limit_dialect
+ sql_quote_char
+ sql_name_sep
+
+ _prefetch_autovalues
+ _perform_autoinc_retrieval
+ _autoinc_supplied_for_op
+
+ _resolve_bindattrs
+
+ _max_column_bytesize
+ _is_lob_type
+ _is_binary_lob_type
+ _is_binary_type
+ _is_text_lob_type
+
+ _prepare_sth
+ _bind_sth_params
+ /,(
+ # the capability framework
+ # not sure if CMOP->initialize does evil things to DBIC::S::DBI, fix if a problem
+ grep
+ { $_ =~ /^ _ (?: use | supports | determine_supports ) _ /x }
+ ( Class::MOP::Class->initialize('DBIx::Class::Storage::DBI')->get_all_method_names )
+ )],
+};
+
+if (DBIx::Class::_ENV_::DBICTEST) {
+
+ my $seen;
+ for my $type (keys %$method_dispatch) {
+ for (@{$method_dispatch->{$type}}) {
+ push @{$seen->{$_}}, $type;
+ }
+ }
+
+ if (my @dupes = grep { @{$seen->{$_}} > 1 } keys %$seen) {
+ die(join "\n", '',
+ 'The following methods show up multiple times in ::Storage::DBI::Replicated handlers:',
+ (map { "$_: " . (join ', ', @{$seen->{$_}}) } sort @dupes),
+ '',
+ );
+ }
+
+ if (my @cant = grep { ! DBIx::Class::Storage::DBI->can($_) } keys %$seen) {
+ die(join "\n", '',
+ '::Storage::DBI::Replicated specifies handling of the following *NON EXISTING* ::Storage::DBI methods:',
+ @cant,
+ '',
+ );
+ }
+}
+
+for my $method (@{$method_dispatch->{unimplemented}}) {
+ __PACKAGE__->meta->add_method($method, sub {
+ my $self = shift;
+ $self->throw_exception("$method() must not be called on ".(blessed $self).' objects');
+ });
+}
+
+=head2 read_handler
+
+Defines an object that implements the read side of L<BIx::Class::Storage::DBI>.
+
+=cut
+
+has 'read_handler' => (
+ is=>'rw',
+ isa=>Object,
+ lazy_build=>1,
+ handles=>$method_dispatch->{reader},
+);
+
+=head2 write_handler
+
+Defines an object that implements the write side of L<BIx::Class::Storage::DBI>,
+as well as methods that don't write or read that can be called on only one
+storage, methods that return a C<$dbh>, and any methods that don't make sense to
+run on a replicant.
+
+=cut
+
+has 'write_handler' => (
+ is=>'ro',
+ isa=>Object,
+ lazy_build=>1,
+ handles=>$method_dispatch->{writer},
);
+
+
has _master_connect_info_opts =>
(is => 'rw', isa => HashRef, default => sub { {} });
=head2 around: connect_info
-Preserve master's C<connect_info> options (for merging with replicants.)
-Also set any Replicated related options from connect_info, such as
+Preserves master's C<connect_info> options (for merging with replicants.)
+Also sets any Replicated-related options from connect_info, such as
C<pool_type>, C<pool_args>, C<balancer_type> and C<balancer_args>.
=cut
around connect_info => sub {
my ($next, $self, $info, @extra) = @_;
- my $wantarray = wantarray;
+ $self->throw_exception(
+ 'connect_info can not be retrieved from a replicated storage - '
+ . 'accessor must be called on a specific pool instance'
+ ) unless defined $info;
+
+ my $merge = Hash::Merge->new('LEFT_PRECEDENT');
my %opts;
for my $arg (@$info) {
next unless (reftype($arg)||'') eq 'HASH';
- %opts = %{ merge($arg, \%opts) };
+ %opts = %{ $merge->merge($arg, \%opts) };
}
delete $opts{dsn};
if $opts{pool_type};
$self->pool_args(
- merge((delete $opts{pool_args} || {}), $self->pool_args)
+ $merge->merge((delete $opts{pool_args} || {}), $self->pool_args)
);
- $self->pool($self->_build_pool)
- if $self->pool;
+ ## Since we possibly changed the pool_args, we need to clear the current
+ ## pool object so that next time it is used it will be rebuilt.
+ $self->clear_pool;
}
if (@opts{qw/balancer_type balancer_args/}) {
if $opts{balancer_type};
$self->balancer_args(
- merge((delete $opts{balancer_args} || {}), $self->balancer_args)
+ $merge->merge((delete $opts{balancer_args} || {}), $self->balancer_args)
);
$self->balancer($self->_build_balancer)
$self->_master_connect_info_opts(\%opts);
- my (@res, $res);
- if ($wantarray) {
- @res = $self->$next($info, @extra);
- } else {
- $res = $self->$next($info, @extra);
- }
-
- # Make sure master is blessed into the correct class and apply role to it.
- my $master = $self->master;
- $master->_determine_driver;
- Moose::Meta::Class->initialize(ref $master);
-
- DBIx::Class::Storage::DBI::Replicated::WithDSN->meta->apply($master);
+ return preserve_context {
+ $self->$next($info, @extra);
+ } after => sub {
+ # Make sure master is blessed into the correct class and apply role to it.
+ my $master = $self->master;
+ $master->_determine_driver;
+ Moose::Meta::Class->initialize(ref $master);
- # link pool back to master
- $self->pool->master($master);
+ DBIx::Class::Storage::DBI::Replicated::WithDSN->meta->apply($master);
- $wantarray ? @res : $res;
+ # link pool back to master
+ $self->pool->master($master);
+ };
};
=head1 METHODS
=cut
sub BUILDARGS {
- my ($class, $schema, $storage_type_args, @args) = @_;
+ my ($class, $schema, $storage_type_args, @args) = @_;
return {
schema=>$schema,
$self->throw_exception('too many hashrefs in connect_info')
if @hashes > 2;
- my %opts = %{ merge(reverse @hashes) };
+ my $merge = Hash::Merge->new('LEFT_PRECEDENT');
+ my %opts = %{ $merge->merge(reverse @hashes) };
# delete them
splice @$r, $i+1, ($#{$r} - $i), ();
delete $master_opts{dbh_maker};
# merge with master
- %opts = %{ merge(\%opts, \%master_opts) };
+ %opts = %{ $merge->merge(\%opts, \%master_opts) };
# update
$r->[$i] = \%opts;
=head2 all_storages
-Returns an array of of all the connected storage backends. The first element
-in the returned array is the master, and the remainings are each of the
+Returns an array of all the connected storage backends. The first element
+in the returned array is the master, and the rest are each of the
replicants.
=cut
=head2 execute_reliably ($coderef, ?@args)
Given a coderef, saves the current state of the L</read_handler>, forces it to
-use reliable storage (ie sets it to the master), executes a coderef and then
+use reliable storage (e.g. sets it to the master), executes a coderef and then
restores the original state.
Example:
my $reliably = sub {
my $name = shift @_;
$schema->resultset('User')->create({name=>$name});
- my $user_rs = $schema->resultset('User')->find({name=>$name});
+ my $user_rs = $schema->resultset('User')->find({name=>$name});
return $user_rs;
};
=cut
sub execute_reliably {
- my ($self, $coderef, @args) = @_;
+ my $self = shift;
+ my $coderef = shift;
unless( ref $coderef eq 'CODE') {
$self->throw_exception('Second argument must be a coderef');
}
- ##Get copy of master storage
- my $master = $self->master;
+ ## replace the current read handler for the remainder of the scope
+ local $self->{read_handler} = $self->master;
- ##Get whatever the current read hander is
- my $current = $self->read_handler;
-
- ##Set the read handler to master
- $self->read_handler($master);
-
- ## do whatever the caller needs
- my @result;
- my $want_array = wantarray;
-
- eval {
- if($want_array) {
- @result = $coderef->(@args);
- } elsif(defined $want_array) {
- ($result[0]) = ($coderef->(@args));
- } else {
- $coderef->(@args);
- }
+ my $args = \@_;
+ return try {
+ $coderef->(@$args);
+ } catch {
+ $self->throw_exception("coderef returned an error: $_");
};
-
- ##Reset to the original state
- $self->read_handler($current);
-
- ##Exception testing has to come last, otherwise you might leave the
- ##read_handler set to master.
-
- if($@) {
- $self->throw_exception("coderef returned an error: $@");
- } else {
- return $want_array ? @result : $result[0];
- }
}
=head2 set_reliable_storage
=head2 set_balanced_storage
Sets the current $schema to be use the </balancer> for all reads, while all
-writea are sent to the master only
+writes are sent to the master only
=cut
foreach my $source ($self->all_storages) {
$source->limit_dialect(@_);
}
- return $self->master->quote_char;
+ return $self->master->limit_dialect;
}
=head2 quote_char
my $self = shift;
return max map $_->lag_behind_master, $self->replicants;
-}
+}
=head2 is_replicating
sub _driver_determined {
my $self = shift;
-
+
if (@_) {
$_->_driver_determined(@_) for $self->all_storages;
}
sub _init {
my $self = shift;
-
+
$_->_init for $self->all_storages;
}
sub _run_connection_actions {
my $self = shift;
-
+
$_->_run_connection_actions for $self->all_storages;
}
sub _do_connection_actions {
my $self = shift;
-
+
if (@_) {
$_->_do_connection_actions(@_) for $self->all_storages;
}
return min map $_->_ping, $self->all_storages;
}
+# not using the normalized_version, because we want to preserve
+# version numbers much longer than the conventional xxx.yyyzzz
+my $numify_ver = sub {
+ my $ver = shift;
+ my @numparts = split /\D+/, $ver;
+ my $format = '%d.' . (join '', ('%06d') x (@numparts - 1));
+
+ return sprintf $format, @numparts;
+};
+sub _server_info {
+ my $self = shift;
+
+ if (not $self->_dbh_details->{info}) {
+ $self->_dbh_details->{info} = (
+ reduce { $a->[0] < $b->[0] ? $a : $b }
+ map [ $numify_ver->($_->{dbms_version}), $_ ],
+ map $_->_server_info, $self->all_storages
+ )->[1];
+ }
+
+ return $self->next::method;
+}
+
+sub _get_server_version {
+ my $self = shift;
+
+ return $self->_server_info->{dbms_version};
+}
+
=head1 GOTCHAS
Due to the fact that replicants can lag behind a master, you must take care to
make sure you use one of the methods to force read queries to a master should
you need realtime data integrity. For example, if you insert a row, and then
-immediately re-read it from the database (say, by doing $row->discard_changes)
+immediately re-read it from the database (say, by doing
+L<< $result->discard_changes|DBIx::Class::Row/discard_changes >>)
or you insert a row and then immediately build a query that expects that row
to be an item, you should force the master to handle reads. Otherwise, due to
the lag, there is no certainty your data will be in the expected state.
Otherwise, you can force a single query to use the master with the 'force_pool'
attribute:
- my $row = $resultset->search(undef, {force_pool=>'master'})->find($pk);
+ my $result = $resultset->search(undef, {force_pool=>'master'})->find($pk);
-This attribute will safely be ignore by non replicated storages, so you can use
+This attribute will safely be ignored by non replicated storages, so you can use
the same code for both types of systems.
Lastly, you can use the L</execute_reliably> method, which works very much like
Based on code originated by:
- Norbert Csongrádi <bert@cpan.org>
- Peter Siklósi <einon@einon.hu>
+ Norbert Csongrádi <bert@cpan.org>
+ Peter Siklósi <einon@einon.hu>
=head1 LICENSE