1 package DBIx::Class::Storage::DBI;
2 # -*- mode: cperl; cperl-indent-level: 2 -*-
4 use base 'DBIx::Class::Storage';
8 use Carp::Clan qw/^DBIx::Class/;
10 use SQL::Abstract::Limit;
11 use DBIx::Class::Storage::DBI::Cursor;
12 use DBIx::Class::Storage::Statistics;
13 use Scalar::Util qw/blessed weaken/;
15 __PACKAGE__->mk_group_accessors('simple' =>
16 qw/_connect_info _dbi_connect_info _dbh _sql_maker _sql_maker_opts
17 _conn_pid _conn_tid transaction_depth _dbh_autocommit savepoints/
20 # the values for these accessors are picked out (and deleted) from
21 # the attribute hashref passed to connect_info
22 my @storage_options = qw/
23 on_connect_do on_disconnect_do disable_sth_caching unsafe auto_savepoint
25 __PACKAGE__->mk_group_accessors('simple' => @storage_options);
28 # default cursor class, overridable in connect_info attributes
29 __PACKAGE__->cursor_class('DBIx::Class::Storage::DBI::Cursor');
31 __PACKAGE__->mk_group_accessors('inherited' => qw/sql_maker_class/);
32 __PACKAGE__->sql_maker_class('DBIC::SQL::Abstract');
36 package # Hide from PAUSE
37 DBIC::SQL::Abstract; # Would merge upstream, but nate doesn't reply :(
39 use base qw/SQL::Abstract::Limit/;
41 # This prevents the caching of $dbh in S::A::L, I believe
43 my $self = shift->SUPER::new(@_);
45 # If limit_dialect is a ref (like a $dbh), go ahead and replace
46 # it with what it resolves to:
47 $self->{limit_dialect} = $self->_find_syntax($self->{limit_dialect})
48 if ref $self->{limit_dialect};
53 # DB2 is the only remaining DB using this. Even though we are not sure if
54 # RowNumberOver is still needed here (should be part of SQLA) leave the
57 my ($self, $sql, $order, $rows, $offset ) = @_;
60 my $last = $rows + $offset;
61 my ( $order_by ) = $self->_order_by( $order );
66 SELECT Q1.*, ROW_NUMBER() OVER( ) AS ROW_NUM FROM (
71 WHERE ROW_NUM BETWEEN $offset AND $last
79 # While we're at it, this should make LIMIT queries more efficient,
80 # without digging into things too deeply
81 use Scalar::Util 'blessed';
83 my ($self, $syntax) = @_;
85 # DB2 is the only remaining DB using this. Even though we are not sure if
86 # RowNumberOver is still needed here (should be part of SQLA) leave the
88 my $dbhname = blessed($syntax) ? $syntax->{Driver}{Name} : $syntax;
89 if(ref($self) && $dbhname && $dbhname eq 'DB2') {
90 return 'RowNumberOver';
93 $self->{_cached_syntax} ||= $self->SUPER::_find_syntax($syntax);
97 my ($self, $table, $fields, $where, $order, @rest) = @_;
98 local $self->{having_bind} = [];
100 # if (ref $table eq 'HASH') {
102 # ($alias, $table) = %$table;
105 if (ref $table eq 'SCALAR') {
108 elsif (ref $table eq 'REF') {
109 my ($sql, @bind) = @{${$table}};
110 push(@{$self->{having_bind}}, @bind);
113 elsif (not ref $table) {
114 $table = $self->_quote($table);
116 local $self->{rownum_hack_count} = 1
117 if (defined $rest[0] && $self->{limit_dialect} eq 'RowNum');
118 @rest = (-1) unless defined $rest[0];
119 die "LIMIT 0 Does Not Compute" if $rest[0] == 0;
120 # and anyway, SQL::Abstract::Limit will cause a barf if we don't first
121 my ($sql, @ret) = $self->SUPER::select(
122 $table, $self->_recurse_fields($fields), $where, $order, @rest
127 $self->{for} eq 'update' ? ' FOR UPDATE' :
128 $self->{for} eq 'shared' ? ' FOR SHARE' :
133 return wantarray ? ($sql, @ret, @{$self->{having_bind}}) : $sql;
139 $table = $self->_quote($table) unless ref($table);
140 $self->SUPER::insert($table, @_);
146 $table = $self->_quote($table) unless ref($table);
147 $self->SUPER::update($table, @_);
153 $table = $self->_quote($table) unless ref($table);
154 $self->SUPER::delete($table, @_);
160 return $_[1].$self->_order_by($_[2]);
162 return $self->SUPER::_emulate_limit(@_);
166 sub _recurse_fields {
167 my ($self, $fields, $params) = @_;
168 my $ref = ref $fields;
169 return $self->_quote($fields) unless $ref;
170 return $$fields if $ref eq 'SCALAR';
172 if ($ref eq 'ARRAY') {
173 return join(', ', map {
174 $self->_recurse_fields($_)
175 .(exists $self->{rownum_hack_count} && !($params && $params->{no_rownum_hack})
176 ? ' AS col'.$self->{rownum_hack_count}++
179 } elsif ($ref eq 'HASH') {
180 foreach my $func (keys %$fields) {
181 return $self->_sqlcase($func)
182 .'( '.$self->_recurse_fields($fields->{$func}).' )';
185 # Is the second check absolutely necessary?
186 elsif ( $ref eq 'REF' and ref($$fields) eq 'ARRAY' ) {
187 return $self->_bind_to_sql( $fields );
190 Carp::croak($ref . qq{ unexpected in _recurse_fields()})
198 if (ref $_[0] eq 'HASH') {
199 if (defined $_[0]->{group_by}) {
200 $ret = $self->_sqlcase(' group by ')
201 .$self->_recurse_fields($_[0]->{group_by}, { no_rownum_hack => 1 });
203 if (defined $_[0]->{having}) {
205 ($frag, @extra) = $self->_recurse_where($_[0]->{having});
206 push(@{$self->{having_bind}}, @extra);
207 $ret .= $self->_sqlcase(' having ').$frag;
209 if (defined $_[0]->{order_by}) {
210 $ret .= $self->_order_by($_[0]->{order_by});
212 if (grep { $_ =~ /^-(desc|asc)/i } keys %{$_[0]}) {
213 return $self->SUPER::_order_by($_[0]);
215 } elsif (ref $_[0] eq 'SCALAR') {
216 $ret = $self->_sqlcase(' order by ').${ $_[0] };
217 } elsif (ref $_[0] eq 'ARRAY' && @{$_[0]}) {
218 my @order = @{+shift};
219 $ret = $self->_sqlcase(' order by ')
221 my $r = $self->_order_by($_, @_);
222 $r =~ s/^ ?ORDER BY //i;
226 $ret = $self->SUPER::_order_by(@_);
231 sub _order_directions {
232 my ($self, $order) = @_;
233 $order = $order->{order_by} if ref $order eq 'HASH';
234 return $self->SUPER::_order_directions($order);
238 my ($self, $from) = @_;
239 if (ref $from eq 'ARRAY') {
240 return $self->_recurse_from(@$from);
241 } elsif (ref $from eq 'HASH') {
242 return $self->_make_as($from);
244 return $from; # would love to quote here but _table ends up getting called
245 # twice during an ->select without a limit clause due to
246 # the way S::A::Limit->select works. should maybe consider
247 # bypassing this and doing S::A::select($self, ...) in
248 # our select method above. meantime, quoting shims have
249 # been added to select/insert/update/delete here
254 my ($self, $from, @join) = @_;
256 push(@sqlf, $self->_make_as($from));
257 foreach my $j (@join) {
260 # check whether a join type exists
261 my $join_clause = '';
262 my $to_jt = ref($to) eq 'ARRAY' ? $to->[0] : $to;
263 if (ref($to_jt) eq 'HASH' and exists($to_jt->{-join_type})) {
264 $join_clause = ' '.uc($to_jt->{-join_type}).' JOIN ';
266 $join_clause = ' JOIN ';
268 push(@sqlf, $join_clause);
270 if (ref $to eq 'ARRAY') {
271 push(@sqlf, '(', $self->_recurse_from(@$to), ')');
273 push(@sqlf, $self->_make_as($to));
275 push(@sqlf, ' ON ', $self->_join_condition($on));
277 return join('', @sqlf);
283 my $sql = shift @$$arr;
284 $sql =~ s/\?/$self->_quote((shift @$$arr)->[1])/eg;
289 my ($self, $from) = @_;
290 return join(' ', map { (ref $_ eq 'SCALAR' ? $$_
291 : ref $_ eq 'REF' ? $self->_bind_to_sql($_)
293 } reverse each %{$self->_skip_options($from)});
297 my ($self, $hash) = @_;
299 $clean_hash->{$_} = $hash->{$_}
300 for grep {!/^-/} keys %$hash;
304 sub _join_condition {
305 my ($self, $cond) = @_;
306 if (ref $cond eq 'HASH') {
311 # XXX no throw_exception() in this package and croak() fails with strange results
312 Carp::croak(ref($v) . qq{ reference arguments are not supported in JOINS - try using \"..." instead'})
313 if ref($v) ne 'SCALAR';
317 my $x = '= '.$self->_quote($v); $j{$_} = \$x;
320 return scalar($self->_recurse_where(\%j));
321 } elsif (ref $cond eq 'ARRAY') {
322 return join(' OR ', map { $self->_join_condition($_) } @$cond);
324 die "Can't handle this yet!";
329 my ($self, $label) = @_;
330 return '' unless defined $label;
331 return "*" if $label eq '*';
332 return $label unless $self->{quote_char};
333 if(ref $self->{quote_char} eq "ARRAY"){
334 return $self->{quote_char}->[0] . $label . $self->{quote_char}->[1]
335 if !defined $self->{name_sep};
336 my $sep = $self->{name_sep};
337 return join($self->{name_sep},
338 map { $self->{quote_char}->[0] . $_ . $self->{quote_char}->[1] }
339 split(/\Q$sep\E/,$label));
341 return $self->SUPER::_quote($label);
346 $self->{limit_dialect} = shift if @_;
347 return $self->{limit_dialect};
352 $self->{quote_char} = shift if @_;
353 return $self->{quote_char};
358 $self->{name_sep} = shift if @_;
359 return $self->{name_sep};
362 } # End of BEGIN block
366 DBIx::Class::Storage::DBI - DBI storage handler
370 my $schema = MySchema->connect('dbi:SQLite:my.db');
372 $schema->storage->debug(1);
373 $schema->dbh_do("DROP TABLE authors");
375 $schema->resultset('Book')->search({
376 written_on => $schema->storage->datetime_parser(DateTime->now)
381 This class represents the connection to an RDBMS via L<DBI>. See
382 L<DBIx::Class::Storage> for general information. This pod only
383 documents DBI-specific methods and behaviors.
390 my $new = shift->next::method(@_);
392 $new->transaction_depth(0);
393 $new->_sql_maker_opts({});
394 $new->{savepoints} = [];
395 $new->{_in_dbh_do} = 0;
396 $new->{_dbh_gen} = 0;
403 This method is normally called by L<DBIx::Class::Schema/connection>, which
404 encapsulates its argument list in an arrayref before passing them here.
406 The argument list may contain:
412 The same 4-element argument set one would normally pass to
413 L<DBI/connect>, optionally followed by
414 L<extra attributes|/DBIx::Class specific connection attributes>
415 recognized by DBIx::Class:
417 $connect_info_args = [ $dsn, $user, $password, \%dbi_attributes?, \%extra_attributes? ];
421 A single code reference which returns a connected
422 L<DBI database handle|DBI/connect> optionally followed by
423 L<extra attributes|/DBIx::Class specific connection attributes> recognized
426 $connect_info_args = [ sub { DBI->connect (...) }, \%extra_attributes? ];
430 A single hashref with all the attributes and the dsn/user/password
433 $connect_info_args = [{
441 This is particularly useful for L<Catalyst> based applications, allowing the
442 following config (L<Config::General> style):
447 dsn dbi:mysql:database=test
456 Please note that the L<DBI> docs recommend that you always explicitly
457 set C<AutoCommit> to either I<0> or I<1>. L<DBIx::Class> further
458 recommends that it be set to I<1>, and that you perform transactions
459 via our L<DBIx::Class::Schema/txn_do> method. L<DBIx::Class> will set it
460 to I<1> if you do not do explicitly set it to zero. This is the default
461 for most DBDs. See L</DBIx::Class and AutoCommit> for details.
463 =head3 DBIx::Class specific connection attributes
465 In addition to the standard L<DBI|DBI/ATTRIBUTES_COMMON_TO_ALL_HANDLES>
466 L<connection|DBI/Database_Handle_Attributes> attributes, DBIx::Class recognizes
467 the following connection options. These options can be mixed in with your other
468 L<DBI> connection attributes, or placed in a seperate hashref
469 (C<\%extra_attributes>) as shown above.
471 Every time C<connect_info> is invoked, any previous settings for
472 these options will be cleared before setting the new ones, regardless of
473 whether any options are specified in the new C<connect_info>.
480 Specifies things to do immediately after connecting or re-connecting to
481 the database. Its value may contain:
485 =item an array reference
487 This contains SQL statements to execute in order. Each element contains
488 a string or a code reference that returns a string.
490 =item a code reference
492 This contains some code to execute. Unlike code references within an
493 array reference, its return value is ignored.
497 =item on_disconnect_do
499 Takes arguments in the same form as L</on_connect_do> and executes them
500 immediately before disconnecting from the database.
502 Note, this only runs if you explicitly call L</disconnect> on the
505 =item disable_sth_caching
507 If set to a true value, this option will disable the caching of
508 statement handles via L<DBI/prepare_cached>.
512 Sets the limit dialect. This is useful for JDBC-bridge among others
513 where the remote SQL-dialect cannot be determined by the name of the
514 driver alone. See also L<SQL::Abstract::Limit>.
518 Specifies what characters to use to quote table and column names. If
519 you use this you will want to specify L</name_sep> as well.
521 C<quote_char> expects either a single character, in which case is it
522 is placed on either side of the table/column name, or an arrayref of length
523 2 in which case the table/column name is placed between the elements.
525 For example under MySQL you should use C<< quote_char => '`' >>, and for
526 SQL Server you should use C<< quote_char => [qw/[ ]/] >>.
530 This only needs to be used in conjunction with C<quote_char>, and is used to
531 specify the charecter that seperates elements (schemas, tables, columns) from
532 each other. In most cases this is simply a C<.>.
534 The consequences of not supplying this value is that L<SQL::Abstract>
535 will assume DBIx::Class' uses of aliases to be complete column
536 names. The output will look like I<"me.name"> when it should actually
541 This Storage driver normally installs its own C<HandleError>, sets
542 C<RaiseError> and C<ShowErrorStatement> on, and sets C<PrintError> off on
543 all database handles, including those supplied by a coderef. It does this
544 so that it can have consistent and useful error behavior.
546 If you set this option to a true value, Storage will not do its usual
547 modifications to the database handle's attributes, and instead relies on
548 the settings in your connect_info DBI options (or the values you set in
549 your connection coderef, in the case that you are connecting via coderef).
551 Note that your custom settings can cause Storage to malfunction,
552 especially if you set a C<HandleError> handler that suppresses exceptions
553 and/or disable C<RaiseError>.
557 If this option is true, L<DBIx::Class> will use savepoints when nesting
558 transactions, making it possible to recover from failure in the inner
559 transaction without having to abort all outer transactions.
563 Use this argument to supply a cursor class other than the default
564 L<DBIx::Class::Storage::DBI::Cursor>.
568 Some real-life examples of arguments to L</connect_info> and
569 L<DBIx::Class::Schema/connect>
571 # Simple SQLite connection
572 ->connect_info([ 'dbi:SQLite:./foo.db' ]);
575 ->connect_info([ sub { DBI->connect(...) } ]);
577 # A bit more complicated
584 { quote_char => q{"}, name_sep => q{.} },
588 # Equivalent to the previous example
594 { AutoCommit => 1, quote_char => q{"}, name_sep => q{.} },
598 # Same, but with hashref as argument
599 # See parse_connect_info for explanation
602 dsn => 'dbi:Pg:dbname=foo',
604 password => 'my_pg_password',
611 # Subref + DBIx::Class-specific connection options
614 sub { DBI->connect(...) },
618 on_connect_do => ['SET search_path TO myschema,otherschema,public'],
619 disable_sth_caching => 1,
629 my ($self, $info_arg) = @_;
631 return $self->_connect_info if !$info_arg;
633 my @args = @$info_arg; # take a shallow copy for further mutilation
634 $self->_connect_info([@args]); # copy for _connect_info
637 # combine/pre-parse arguments depending on invocation style
640 if (ref $args[0] eq 'CODE') { # coderef with optional \%extra_attributes
641 %attrs = %{ $args[1] || {} };
644 elsif (ref $args[0] eq 'HASH') { # single hashref (i.e. Catalyst config)
645 %attrs = %{$args[0]};
647 for (qw/password user dsn/) {
648 unshift @args, delete $attrs{$_};
651 else { # otherwise assume dsn/user/password + \%attrs + \%extra_attrs
653 % { $args[3] || {} },
654 % { $args[4] || {} },
656 @args = @args[0,1,2];
659 # Kill sql_maker/_sql_maker_opts, so we get a fresh one with only
660 # the new set of options
661 $self->_sql_maker(undef);
662 $self->_sql_maker_opts({});
665 for my $storage_opt (@storage_options, 'cursor_class') { # @storage_options is declared at the top of the module
666 if(my $value = delete $attrs{$storage_opt}) {
667 $self->$storage_opt($value);
670 for my $sql_maker_opt (qw/limit_dialect quote_char name_sep/) {
671 if(my $opt_val = delete $attrs{$sql_maker_opt}) {
672 $self->_sql_maker_opts->{$sql_maker_opt} = $opt_val;
677 %attrs = () if (ref $args[0] eq 'CODE'); # _connect() never looks past $args[0] in this case
679 $self->_dbi_connect_info([@args, keys %attrs ? \%attrs : ()]);
680 $self->_connect_info;
685 This method is deprecated in favour of setting via L</connect_info>.
690 Arguments: ($subref | $method_name), @extra_coderef_args?
692 Execute the given $subref or $method_name using the new exception-based
693 connection management.
695 The first two arguments will be the storage object that C<dbh_do> was called
696 on and a database handle to use. Any additional arguments will be passed
697 verbatim to the called subref as arguments 2 and onwards.
699 Using this (instead of $self->_dbh or $self->dbh) ensures correct
700 exception handling and reconnection (or failover in future subclasses).
702 Your subref should have no side-effects outside of the database, as
703 there is the potential for your subref to be partially double-executed
704 if the database connection was stale/dysfunctional.
708 my @stuff = $schema->storage->dbh_do(
710 my ($storage, $dbh, @cols) = @_;
711 my $cols = join(q{, }, @cols);
712 $dbh->selectrow_array("SELECT $cols FROM foo");
723 my $dbh = $self->_dbh;
725 return $self->$code($dbh, @_) if $self->{_in_dbh_do}
726 || $self->{transaction_depth};
728 local $self->{_in_dbh_do} = 1;
731 my $want_array = wantarray;
734 $self->_verify_pid if $dbh;
736 $self->_populate_dbh;
741 @result = $self->$code($dbh, @_);
743 elsif(defined $want_array) {
744 $result[0] = $self->$code($dbh, @_);
747 $self->$code($dbh, @_);
752 if(!$exception) { return $want_array ? @result : $result[0] }
754 $self->throw_exception($exception) if $self->connected;
756 # We were not connected - reconnect and retry, but let any
757 # exception fall right through this time
758 $self->_populate_dbh;
759 $self->$code($self->_dbh, @_);
762 # This is basically a blend of dbh_do above and DBIx::Class::Storage::txn_do.
763 # It also informs dbh_do to bypass itself while under the direction of txn_do,
764 # via $self->{_in_dbh_do} (this saves some redundant eval and errorcheck, etc)
769 ref $coderef eq 'CODE' or $self->throw_exception
770 ('$coderef must be a CODE reference');
772 return $coderef->(@_) if $self->{transaction_depth} && ! $self->auto_savepoint;
774 local $self->{_in_dbh_do} = 1;
777 my $want_array = wantarray;
782 $self->_verify_pid if $self->_dbh;
783 $self->_populate_dbh if !$self->_dbh;
787 @result = $coderef->(@_);
789 elsif(defined $want_array) {
790 $result[0] = $coderef->(@_);
799 if(!$exception) { return $want_array ? @result : $result[0] }
801 if($tried++ > 0 || $self->connected) {
802 eval { $self->txn_rollback };
803 my $rollback_exception = $@;
804 if($rollback_exception) {
805 my $exception_class = "DBIx::Class::Storage::NESTED_ROLLBACK_EXCEPTION";
806 $self->throw_exception($exception) # propagate nested rollback
807 if $rollback_exception =~ /$exception_class/;
809 $self->throw_exception(
810 "Transaction aborted: ${exception}. "
811 . "Rollback failed: ${rollback_exception}"
814 $self->throw_exception($exception)
817 # We were not connected, and was first try - reconnect and retry
819 $self->_populate_dbh;
825 Our C<disconnect> method also performs a rollback first if the
826 database is not in C<AutoCommit> mode.
833 if( $self->connected ) {
834 my $connection_do = $self->on_disconnect_do;
835 $self->_do_connection_actions($connection_do) if ref($connection_do);
837 $self->_dbh->rollback unless $self->_dbh_autocommit;
838 $self->_dbh->disconnect;
844 =head2 with_deferred_fk_checks
848 =item Arguments: C<$coderef>
850 =item Return Value: The return value of $coderef
854 Storage specific method to run the code ref with FK checks deferred or
855 in MySQL's case disabled entirely.
859 # Storage subclasses should override this
860 sub with_deferred_fk_checks {
861 my ($self, $sub) = @_;
869 if(my $dbh = $self->_dbh) {
870 if(defined $self->_conn_tid && $self->_conn_tid != threads->tid) {
877 return 0 if !$self->_dbh;
879 return ($dbh->FETCH('Active') && $dbh->ping);
885 # handle pid changes correctly
886 # NOTE: assumes $self->_dbh is a valid $dbh
890 return if defined $self->_conn_pid && $self->_conn_pid == $$;
892 $self->_dbh->{InactiveDestroy} = 1;
899 sub ensure_connected {
902 unless ($self->connected) {
903 $self->_populate_dbh;
909 Returns the dbh - a data base handle of class L<DBI>.
916 $self->ensure_connected;
920 sub _sql_maker_args {
923 return ( bindtype=>'columns', array_datatypes => 1, limit_dialect => $self->dbh, %{$self->_sql_maker_opts} );
928 unless ($self->_sql_maker) {
929 my $sql_maker_class = $self->sql_maker_class;
930 $self->_sql_maker($sql_maker_class->new( $self->_sql_maker_args ));
932 return $self->_sql_maker;
939 my @info = @{$self->_dbi_connect_info || []};
940 $self->_dbh($self->_connect(@info));
942 # Always set the transaction depth on connect, since
943 # there is no transaction in progress by definition
944 $self->{transaction_depth} = $self->_dbh_autocommit ? 0 : 1;
946 if(ref $self eq 'DBIx::Class::Storage::DBI') {
947 my $driver = $self->_dbh->{Driver}->{Name};
948 if ($self->load_optional_class("DBIx::Class::Storage::DBI::${driver}")) {
949 bless $self, "DBIx::Class::Storage::DBI::${driver}";
954 $self->_conn_pid($$);
955 $self->_conn_tid(threads->tid) if $INC{'threads.pm'};
957 my $connection_do = $self->on_connect_do;
958 $self->_do_connection_actions($connection_do) if ref($connection_do);
961 sub _do_connection_actions {
963 my $connection_do = shift;
965 if (ref $connection_do eq 'ARRAY') {
966 $self->_do_query($_) foreach @$connection_do;
968 elsif (ref $connection_do eq 'CODE') {
969 $connection_do->($self);
976 my ($self, $action) = @_;
978 if (ref $action eq 'CODE') {
979 $action = $action->($self);
980 $self->_do_query($_) foreach @$action;
983 # Most debuggers expect ($sql, @bind), so we need to exclude
984 # the attribute hash which is the second argument to $dbh->do
985 # furthermore the bind values are usually to be presented
986 # as named arrayref pairs, so wrap those here too
987 my @do_args = (ref $action eq 'ARRAY') ? (@$action) : ($action);
988 my $sql = shift @do_args;
989 my $attrs = shift @do_args;
990 my @bind = map { [ undef, $_ ] } @do_args;
992 $self->_query_start($sql, @bind);
993 $self->_dbh->do($sql, $attrs, @do_args);
994 $self->_query_end($sql, @bind);
1001 my ($self, @info) = @_;
1003 $self->throw_exception("You failed to provide any connection info")
1006 my ($old_connect_via, $dbh);
1008 if ($INC{'Apache/DBI.pm'} && $ENV{MOD_PERL}) {
1009 $old_connect_via = $DBI::connect_via;
1010 $DBI::connect_via = 'connect';
1014 if(ref $info[0] eq 'CODE') {
1018 $dbh = DBI->connect(@info);
1021 if($dbh && !$self->unsafe) {
1022 my $weak_self = $self;
1024 $dbh->{HandleError} = sub {
1026 $weak_self->throw_exception("DBI Exception: $_[0]");
1029 croak ("DBI Exception: $_[0]");
1032 $dbh->{ShowErrorStatement} = 1;
1033 $dbh->{RaiseError} = 1;
1034 $dbh->{PrintError} = 0;
1038 $DBI::connect_via = $old_connect_via if $old_connect_via;
1040 $self->throw_exception("DBI Connection failed: " . ($@||$DBI::errstr))
1043 $self->_dbh_autocommit($dbh->{AutoCommit});
1049 my ($self, $name) = @_;
1051 $name = $self->_svp_generate_name
1052 unless defined $name;
1054 $self->throw_exception ("You can't use savepoints outside a transaction")
1055 if $self->{transaction_depth} == 0;
1057 $self->throw_exception ("Your Storage implementation doesn't support savepoints")
1058 unless $self->can('_svp_begin');
1060 push @{ $self->{savepoints} }, $name;
1062 $self->debugobj->svp_begin($name) if $self->debug;
1064 return $self->_svp_begin($name);
1068 my ($self, $name) = @_;
1070 $self->throw_exception ("You can't use savepoints outside a transaction")
1071 if $self->{transaction_depth} == 0;
1073 $self->throw_exception ("Your Storage implementation doesn't support savepoints")
1074 unless $self->can('_svp_release');
1076 if (defined $name) {
1077 $self->throw_exception ("Savepoint '$name' does not exist")
1078 unless grep { $_ eq $name } @{ $self->{savepoints} };
1080 # Dig through the stack until we find the one we are releasing. This keeps
1081 # the stack up to date.
1084 do { $svp = pop @{ $self->{savepoints} } } while $svp ne $name;
1086 $name = pop @{ $self->{savepoints} };
1089 $self->debugobj->svp_release($name) if $self->debug;
1091 return $self->_svp_release($name);
1095 my ($self, $name) = @_;
1097 $self->throw_exception ("You can't use savepoints outside a transaction")
1098 if $self->{transaction_depth} == 0;
1100 $self->throw_exception ("Your Storage implementation doesn't support savepoints")
1101 unless $self->can('_svp_rollback');
1103 if (defined $name) {
1104 # If they passed us a name, verify that it exists in the stack
1105 unless(grep({ $_ eq $name } @{ $self->{savepoints} })) {
1106 $self->throw_exception("Savepoint '$name' does not exist!");
1109 # Dig through the stack until we find the one we are releasing. This keeps
1110 # the stack up to date.
1111 while(my $s = pop(@{ $self->{savepoints} })) {
1112 last if($s eq $name);
1114 # Add the savepoint back to the stack, as a rollback doesn't remove the
1115 # named savepoint, only everything after it.
1116 push(@{ $self->{savepoints} }, $name);
1118 # We'll assume they want to rollback to the last savepoint
1119 $name = $self->{savepoints}->[-1];
1122 $self->debugobj->svp_rollback($name) if $self->debug;
1124 return $self->_svp_rollback($name);
1127 sub _svp_generate_name {
1130 return 'savepoint_'.scalar(@{ $self->{'savepoints'} });
1135 $self->ensure_connected();
1136 if($self->{transaction_depth} == 0) {
1137 $self->debugobj->txn_begin()
1139 # this isn't ->_dbh-> because
1140 # we should reconnect on begin_work
1141 # for AutoCommit users
1142 $self->dbh->begin_work;
1143 } elsif ($self->auto_savepoint) {
1146 $self->{transaction_depth}++;
1151 if ($self->{transaction_depth} == 1) {
1152 my $dbh = $self->_dbh;
1153 $self->debugobj->txn_commit()
1156 $self->{transaction_depth} = 0
1157 if $self->_dbh_autocommit;
1159 elsif($self->{transaction_depth} > 1) {
1160 $self->{transaction_depth}--;
1162 if $self->auto_savepoint;
1168 my $dbh = $self->_dbh;
1170 if ($self->{transaction_depth} == 1) {
1171 $self->debugobj->txn_rollback()
1173 $self->{transaction_depth} = 0
1174 if $self->_dbh_autocommit;
1177 elsif($self->{transaction_depth} > 1) {
1178 $self->{transaction_depth}--;
1179 if ($self->auto_savepoint) {
1180 $self->svp_rollback;
1185 die DBIx::Class::Storage::NESTED_ROLLBACK_EXCEPTION->new;
1190 my $exception_class = "DBIx::Class::Storage::NESTED_ROLLBACK_EXCEPTION";
1191 $error =~ /$exception_class/ and $self->throw_exception($error);
1192 # ensure that a failed rollback resets the transaction depth
1193 $self->{transaction_depth} = $self->_dbh_autocommit ? 0 : 1;
1194 $self->throw_exception($error);
1198 # This used to be the top-half of _execute. It was split out to make it
1199 # easier to override in NoBindVars without duping the rest. It takes up
1200 # all of _execute's args, and emits $sql, @bind.
1201 sub _prep_for_execute {
1202 my ($self, $op, $extra_bind, $ident, $args) = @_;
1204 if( blessed($ident) && $ident->isa("DBIx::Class::ResultSource") ) {
1205 $ident = $ident->from();
1208 my ($sql, @bind) = $self->sql_maker->$op($ident, @$args);
1211 map { ref $_ eq 'ARRAY' ? $_ : [ '!!dummy', $_ ] } @$extra_bind)
1213 return ($sql, \@bind);
1216 sub _fix_bind_params {
1217 my ($self, @bind) = @_;
1219 ### Turn @bind from something like this:
1220 ### ( [ "artist", 1 ], [ "cdid", 1, 3 ] )
1222 ### ( "'1'", "'1'", "'3'" )
1225 if ( defined( $_ && $_->[1] ) ) {
1226 map { qq{'$_'}; } @{$_}[ 1 .. $#$_ ];
1233 my ( $self, $sql, @bind ) = @_;
1235 if ( $self->debug ) {
1236 @bind = $self->_fix_bind_params(@bind);
1238 $self->debugobj->query_start( $sql, @bind );
1243 my ( $self, $sql, @bind ) = @_;
1245 if ( $self->debug ) {
1246 @bind = $self->_fix_bind_params(@bind);
1247 $self->debugobj->query_end( $sql, @bind );
1252 my ($self, $dbh, $op, $extra_bind, $ident, $bind_attributes, @args) = @_;
1254 my ($sql, $bind) = $self->_prep_for_execute($op, $extra_bind, $ident, \@args);
1256 $self->_query_start( $sql, @$bind );
1258 my $sth = $self->sth($sql,$op);
1260 my $placeholder_index = 1;
1262 foreach my $bound (@$bind) {
1263 my $attributes = {};
1264 my($column_name, @data) = @$bound;
1266 if ($bind_attributes) {
1267 $attributes = $bind_attributes->{$column_name}
1268 if defined $bind_attributes->{$column_name};
1271 foreach my $data (@data) {
1272 my $ref = ref $data;
1273 $data = $ref && $ref ne 'ARRAY' ? ''.$data : $data; # stringify args (except arrayrefs)
1275 $sth->bind_param($placeholder_index, $data, $attributes);
1276 $placeholder_index++;
1280 # Can this fail without throwing an exception anyways???
1281 my $rv = $sth->execute();
1282 $self->throw_exception($sth->errstr) if !$rv;
1284 $self->_query_end( $sql, @$bind );
1286 return (wantarray ? ($rv, $sth, @$bind) : $rv);
1291 $self->dbh_do('_dbh_execute', @_)
1295 my ($self, $source, $to_insert) = @_;
1297 my $ident = $source->from;
1298 my $bind_attributes = $self->source_bind_attributes($source);
1300 $self->ensure_connected;
1301 foreach my $col ( $source->columns ) {
1302 if ( !defined $to_insert->{$col} ) {
1303 my $col_info = $source->column_info($col);
1305 if ( $col_info->{auto_nextval} ) {
1306 $to_insert->{$col} = $self->_sequence_fetch( 'nextval', $col_info->{sequence} || $self->_dbh_get_autoinc_seq($self->dbh, $source) );
1311 $self->_execute('insert' => [], $source, $bind_attributes, $to_insert);
1316 ## Still not quite perfect, and EXPERIMENTAL
1317 ## Currently it is assumed that all values passed will be "normal", i.e. not
1318 ## scalar refs, or at least, all the same type as the first set, the statement is
1319 ## only prepped once.
1321 my ($self, $source, $cols, $data) = @_;
1323 my $table = $source->from;
1324 @colvalues{@$cols} = (0..$#$cols);
1325 my ($sql, @bind) = $self->sql_maker->insert($table, \%colvalues);
1327 $self->_query_start( $sql, @bind );
1328 my $sth = $self->sth($sql);
1330 # @bind = map { ref $_ ? ''.$_ : $_ } @bind; # stringify args
1332 ## This must be an arrayref, else nothing works!
1334 my $tuple_status = [];
1337 ##print STDERR Dumper( $data, $sql, [@bind] );
1341 ## Get the bind_attributes, if any exist
1342 my $bind_attributes = $self->source_bind_attributes($source);
1344 ## Bind the values and execute
1345 my $placeholder_index = 1;
1347 foreach my $bound (@bind) {
1349 my $attributes = {};
1350 my ($column_name, $data_index) = @$bound;
1352 if( $bind_attributes ) {
1353 $attributes = $bind_attributes->{$column_name}
1354 if defined $bind_attributes->{$column_name};
1357 my @data = map { $_->[$data_index] } @$data;
1359 $sth->bind_param_array( $placeholder_index, [@data], $attributes );
1360 $placeholder_index++;
1362 my $rv = $sth->execute_array({ArrayTupleStatus => $tuple_status});
1363 $self->throw_exception($sth->errstr) if !$rv;
1365 $self->_query_end( $sql, @bind );
1366 return (wantarray ? ($rv, $sth, @bind) : $rv);
1370 my $self = shift @_;
1371 my $source = shift @_;
1372 my $bind_attributes = $self->source_bind_attributes($source);
1374 return $self->_execute('update' => [], $source, $bind_attributes, @_);
1379 my $self = shift @_;
1380 my $source = shift @_;
1382 my $bind_attrs = {}; ## If ever it's needed...
1384 return $self->_execute('delete' => [], $source, $bind_attrs, @_);
1389 my $sql_maker = $self->sql_maker;
1390 local $sql_maker->{for};
1391 return $self->_execute($self->_select_args(@_));
1395 my ($self, $ident, $select, $condition, $attrs) = @_;
1396 my $order = $attrs->{order_by};
1398 if (ref $condition eq 'SCALAR') {
1399 my $unwrap = ${$condition};
1400 if ($unwrap =~ s/ORDER BY (.*)$//i) {
1402 $condition = \$unwrap;
1406 my $for = delete $attrs->{for};
1407 my $sql_maker = $self->sql_maker;
1408 $sql_maker->{for} = $for;
1410 if (exists $attrs->{group_by} || $attrs->{having}) {
1412 group_by => $attrs->{group_by},
1413 having => $attrs->{having},
1414 ($order ? (order_by => $order) : ())
1417 my $bind_attrs = {}; ## Future support
1418 my @args = ('select', $attrs->{bind}, $ident, $bind_attrs, $select, $condition, $order);
1419 if ($attrs->{software_limit} ||
1420 $self->sql_maker->_default_limit_syntax eq "GenericSubQ") {
1421 $attrs->{software_limit} = 1;
1423 $self->throw_exception("rows attribute must be positive if present")
1424 if (defined($attrs->{rows}) && !($attrs->{rows} > 0));
1426 # MySQL actually recommends this approach. I cringe.
1427 $attrs->{rows} = 2**48 if not defined $attrs->{rows} and defined $attrs->{offset};
1428 push @args, $attrs->{rows}, $attrs->{offset};
1433 sub source_bind_attributes {
1434 my ($self, $source) = @_;
1436 my $bind_attributes;
1437 foreach my $column ($source->columns) {
1439 my $data_type = $source->column_info($column)->{data_type} || '';
1440 $bind_attributes->{$column} = $self->bind_attribute_by_data_type($data_type)
1444 return $bind_attributes;
1451 =item Arguments: $ident, $select, $condition, $attrs
1455 Handle a SQL select statement.
1461 my ($ident, $select, $condition, $attrs) = @_;
1462 return $self->cursor_class->new($self, \@_, $attrs);
1467 my ($rv, $sth, @bind) = $self->_select(@_);
1468 my @row = $sth->fetchrow_array;
1469 my @nextrow = $sth->fetchrow_array if @row;
1470 if(@row && @nextrow) {
1471 carp "Query returned more than one row. SQL that returns multiple rows is DEPRECATED for ->find and ->single";
1473 # Need to call finish() to work round broken DBDs
1482 =item Arguments: $sql
1486 Returns a L<DBI> sth (statement handle) for the supplied SQL.
1491 my ($self, $dbh, $sql) = @_;
1493 # 3 is the if_active parameter which avoids active sth re-use
1494 my $sth = $self->disable_sth_caching
1495 ? $dbh->prepare($sql)
1496 : $dbh->prepare_cached($sql, {}, 3);
1498 # XXX You would think RaiseError would make this impossible,
1499 # but apparently that's not true :(
1500 $self->throw_exception($dbh->errstr) if !$sth;
1506 my ($self, $sql) = @_;
1507 $self->dbh_do('_dbh_sth', $sql);
1510 sub _dbh_columns_info_for {
1511 my ($self, $dbh, $table) = @_;
1513 if ($dbh->can('column_info')) {
1516 my ($schema,$tab) = $table =~ /^(.+?)\.(.+)$/ ? ($1,$2) : (undef,$table);
1517 my $sth = $dbh->column_info( undef,$schema, $tab, '%' );
1519 while ( my $info = $sth->fetchrow_hashref() ){
1521 $column_info{data_type} = $info->{TYPE_NAME};
1522 $column_info{size} = $info->{COLUMN_SIZE};
1523 $column_info{is_nullable} = $info->{NULLABLE} ? 1 : 0;
1524 $column_info{default_value} = $info->{COLUMN_DEF};
1525 my $col_name = $info->{COLUMN_NAME};
1526 $col_name =~ s/^\"(.*)\"$/$1/;
1528 $result{$col_name} = \%column_info;
1531 return \%result if !$@ && scalar keys %result;
1535 my $sth = $dbh->prepare($self->sql_maker->select($table, undef, \'1 = 0'));
1537 my @columns = @{$sth->{NAME_lc}};
1538 for my $i ( 0 .. $#columns ){
1540 $column_info{data_type} = $sth->{TYPE}->[$i];
1541 $column_info{size} = $sth->{PRECISION}->[$i];
1542 $column_info{is_nullable} = $sth->{NULLABLE}->[$i] ? 1 : 0;
1544 if ($column_info{data_type} =~ m/^(.*?)\((.*?)\)$/) {
1545 $column_info{data_type} = $1;
1546 $column_info{size} = $2;
1549 $result{$columns[$i]} = \%column_info;
1553 foreach my $col (keys %result) {
1554 my $colinfo = $result{$col};
1555 my $type_num = $colinfo->{data_type};
1557 if(defined $type_num && $dbh->can('type_info')) {
1558 my $type_info = $dbh->type_info($type_num);
1559 $type_name = $type_info->{TYPE_NAME} if $type_info;
1560 $colinfo->{data_type} = $type_name if $type_name;
1567 sub columns_info_for {
1568 my ($self, $table) = @_;
1569 $self->dbh_do('_dbh_columns_info_for', $table);
1572 =head2 last_insert_id
1574 Return the row id of the last insert.
1578 sub _dbh_last_insert_id {
1579 # All Storage's need to register their own _dbh_last_insert_id
1580 # the old SQLite-based method was highly inappropriate
1583 my $class = ref $self;
1584 $self->throw_exception (<<EOE);
1586 No _dbh_last_insert_id() method found in $class.
1587 Since the method of obtaining the autoincrement id of the last insert
1588 operation varies greatly between different databases, this method must be
1589 individually implemented for every storage class.
1593 sub last_insert_id {
1595 $self->dbh_do('_dbh_last_insert_id', @_);
1600 Returns the database driver name.
1604 sub sqlt_type { shift->dbh->{Driver}->{Name} }
1606 =head2 bind_attribute_by_data_type
1608 Given a datatype from column info, returns a database specific bind
1609 attribute for C<< $dbh->bind_param($val,$attribute) >> or nothing if we will
1610 let the database planner just handle it.
1612 Generally only needed for special case column types, like bytea in postgres.
1616 sub bind_attribute_by_data_type {
1620 =head2 create_ddl_dir
1624 =item Arguments: $schema \@databases, $version, $directory, $preversion, \%sqlt_args
1628 Creates a SQL file based on the Schema, for each of the specified
1629 database types, in the given directory.
1631 By default, C<\%sqlt_args> will have
1633 { add_drop_table => 1, ignore_constraint_names => 1, ignore_index_names => 1 }
1635 merged with the hash passed in. To disable any of those features, pass in a
1636 hashref like the following
1638 { ignore_constraint_names => 0, # ... other options }
1642 sub create_ddl_dir {
1643 my ($self, $schema, $databases, $version, $dir, $preversion, $sqltargs) = @_;
1645 if(!$dir || !-d $dir) {
1646 warn "No directory given, using ./\n";
1649 $databases ||= ['MySQL', 'SQLite', 'PostgreSQL'];
1650 $databases = [ $databases ] if(ref($databases) ne 'ARRAY');
1652 my $schema_version = $schema->schema_version || '1.x';
1653 $version ||= $schema_version;
1656 add_drop_table => 1,
1657 ignore_constraint_names => 1,
1658 ignore_index_names => 1,
1662 $self->throw_exception(q{Can't create a ddl file without SQL::Translator 0.09003: '}
1663 . $self->_check_sqlt_message . q{'})
1664 if !$self->_check_sqlt_version;
1666 my $sqlt = SQL::Translator->new( $sqltargs );
1668 $sqlt->parser('SQL::Translator::Parser::DBIx::Class');
1669 my $sqlt_schema = $sqlt->translate({ data => $schema }) or die $sqlt->error;
1671 foreach my $db (@$databases) {
1673 $sqlt->{schema} = $sqlt_schema;
1674 $sqlt->producer($db);
1677 my $filename = $schema->ddl_filename($db, $version, $dir);
1678 if (-e $filename && ($version eq $schema_version )) {
1679 # if we are dumping the current version, overwrite the DDL
1680 warn "Overwriting existing DDL file - $filename";
1684 my $output = $sqlt->translate;
1686 warn("Failed to translate to $db, skipping. (" . $sqlt->error . ")");
1689 if(!open($file, ">$filename")) {
1690 $self->throw_exception("Can't open $filename for writing ($!)");
1693 print $file $output;
1696 next unless ($preversion);
1698 require SQL::Translator::Diff;
1700 my $prefilename = $schema->ddl_filename($db, $preversion, $dir);
1701 if(!-e $prefilename) {
1702 warn("No previous schema file found ($prefilename)");
1706 my $difffile = $schema->ddl_filename($db, $version, $dir, $preversion);
1708 warn("Overwriting existing diff file - $difffile");
1714 my $t = SQL::Translator->new($sqltargs);
1717 $t->parser( $db ) or die $t->error;
1718 my $out = $t->translate( $prefilename ) or die $t->error;
1719 $source_schema = $t->schema;
1720 unless ( $source_schema->name ) {
1721 $source_schema->name( $prefilename );
1725 # The "new" style of producers have sane normalization and can support
1726 # diffing a SQL file against a DBIC->SQLT schema. Old style ones don't
1727 # And we have to diff parsed SQL against parsed SQL.
1728 my $dest_schema = $sqlt_schema;
1730 unless ( "SQL::Translator::Producer::$db"->can('preprocess_schema') ) {
1731 my $t = SQL::Translator->new($sqltargs);
1734 $t->parser( $db ) or die $t->error;
1735 my $out = $t->translate( $filename ) or die $t->error;
1736 $dest_schema = $t->schema;
1737 $dest_schema->name( $filename )
1738 unless $dest_schema->name;
1741 my $diff = SQL::Translator::Diff::schema_diff($source_schema, $db,
1745 if(!open $file, ">$difffile") {
1746 $self->throw_exception("Can't write to $difffile ($!)");
1754 =head2 deployment_statements
1758 =item Arguments: $schema, $type, $version, $directory, $sqlt_args
1762 Returns the statements used by L</deploy> and L<DBIx::Class::Schema/deploy>.
1763 The database driver name is given by C<$type>, though the value from
1764 L</sqlt_type> is used if it is not specified.
1766 C<$directory> is used to return statements from files in a previously created
1767 L</create_ddl_dir> directory and is optional. The filenames are constructed
1768 from L<DBIx::Class::Schema/ddl_filename>, the schema name and the C<$version>.
1770 If no C<$directory> is specified then the statements are constructed on the
1771 fly using L<SQL::Translator> and C<$version> is ignored.
1773 See L<SQL::Translator/METHODS> for a list of values for C<$sqlt_args>.
1777 sub deployment_statements {
1778 my ($self, $schema, $type, $version, $dir, $sqltargs) = @_;
1779 # Need to be connected to get the correct sqlt_type
1780 $self->ensure_connected() unless $type;
1781 $type ||= $self->sqlt_type;
1782 $version ||= $schema->schema_version || '1.x';
1784 my $filename = $schema->ddl_filename($type, $version, $dir);
1788 open($file, "<$filename")
1789 or $self->throw_exception("Can't open $filename ($!)");
1792 return join('', @rows);
1795 $self->throw_exception(q{Can't deploy without SQL::Translator 0.09003: '}
1796 . $self->_check_sqlt_message . q{'})
1797 if !$self->_check_sqlt_version;
1799 require SQL::Translator::Parser::DBIx::Class;
1800 eval qq{use SQL::Translator::Producer::${type}};
1801 $self->throw_exception($@) if $@;
1803 # sources needs to be a parser arg, but for simplicty allow at top level
1805 $sqltargs->{parser_args}{sources} = delete $sqltargs->{sources}
1806 if exists $sqltargs->{sources};
1808 my $tr = SQL::Translator->new(%$sqltargs);
1809 SQL::Translator::Parser::DBIx::Class::parse( $tr, $schema );
1810 return "SQL::Translator::Producer::${type}"->can('produce')->($tr);
1814 my ($self, $schema, $type, $sqltargs, $dir) = @_;
1817 return if($line =~ /^--/);
1819 # next if($line =~ /^DROP/m);
1820 return if($line =~ /^BEGIN TRANSACTION/m);
1821 return if($line =~ /^COMMIT/m);
1822 return if $line =~ /^\s+$/; # skip whitespace only
1823 $self->_query_start($line);
1825 $self->dbh->do($line); # shouldn't be using ->dbh ?
1828 warn qq{$@ (running "${line}")};
1830 $self->_query_end($line);
1832 my @statements = $self->deployment_statements($schema, $type, undef, $dir, { no_comments => 1, %{ $sqltargs || {} } } );
1833 if (@statements > 1) {
1834 foreach my $statement (@statements) {
1835 $deploy->( $statement );
1838 elsif (@statements == 1) {
1839 foreach my $line ( split(";\n", $statements[0])) {
1845 =head2 datetime_parser
1847 Returns the datetime parser class
1851 sub datetime_parser {
1853 return $self->{datetime_parser} ||= do {
1854 $self->ensure_connected;
1855 $self->build_datetime_parser(@_);
1859 =head2 datetime_parser_type
1861 Defines (returns) the datetime parser class - currently hardwired to
1862 L<DateTime::Format::MySQL>
1866 sub datetime_parser_type { "DateTime::Format::MySQL"; }
1868 =head2 build_datetime_parser
1870 See L</datetime_parser>
1874 sub build_datetime_parser {
1876 my $type = $self->datetime_parser_type(@_);
1878 $self->throw_exception("Couldn't load ${type}: $@") if $@;
1883 my $_check_sqlt_version; # private
1884 my $_check_sqlt_message; # private
1885 sub _check_sqlt_version {
1886 return $_check_sqlt_version if defined $_check_sqlt_version;
1887 eval 'use SQL::Translator "0.09003"';
1888 $_check_sqlt_message = $@ || '';
1889 $_check_sqlt_version = !$@;
1892 sub _check_sqlt_message {
1893 _check_sqlt_version if !defined $_check_sqlt_message;
1894 $_check_sqlt_message;
1898 =head2 is_replicating
1900 A boolean that reports if a particular L<DBIx::Class::Storage::DBI> is set to
1901 replicate from a master database. Default is undef, which is the result
1902 returned by databases that don't support replication.
1906 sub is_replicating {
1911 =head2 lag_behind_master
1913 Returns a number that represents a certain amount of lag behind a master db
1914 when a given storage is replicating. The number is database dependent, but
1915 starts at zero and increases with the amount of lag. Default in undef
1919 sub lag_behind_master {
1925 return if !$self->_dbh;
1934 =head2 DBIx::Class and AutoCommit
1936 DBIx::Class can do some wonderful magic with handling exceptions,
1937 disconnections, and transactions when you use C<< AutoCommit => 1 >>
1938 combined with C<txn_do> for transaction support.
1940 If you set C<< AutoCommit => 0 >> in your connect info, then you are always
1941 in an assumed transaction between commits, and you're telling us you'd
1942 like to manage that manually. A lot of the magic protections offered by
1943 this module will go away. We can't protect you from exceptions due to database
1944 disconnects because we don't know anything about how to restart your
1945 transactions. You're on your own for handling all sorts of exceptional
1946 cases if you choose the C<< AutoCommit => 0 >> path, just as you would
1952 The module defines a set of methods within the DBIC::SQL::Abstract
1953 namespace. These build on L<SQL::Abstract::Limit> to provide the
1954 SQL query functions.
1956 The following methods are extended:-
1970 See L</connect_info> for details.
1974 See L</connect_info> for details.
1978 See L</connect_info> for details.
1984 Matt S. Trout <mst@shadowcatsystems.co.uk>
1986 Andy Grundman <andy@hybridized.org>
1990 You may distribute this code under the same terms as Perl itself.