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/;
40 use Carp::Clan qw/^DBIx::Class/;
43 my $self = shift->SUPER::new(@_);
45 # This prevents the caching of $dbh in S::A::L, I believe
46 # If limit_dialect is a ref (like a $dbh), go ahead and replace
47 # it with what it resolves to:
48 $self->{limit_dialect} = $self->_find_syntax($self->{limit_dialect})
49 if ref $self->{limit_dialect};
56 # Some databases (sqlite) do not handle multiple parenthesis
57 # around in/between arguments. A tentative x IN ( ( 1, 2 ,3) )
58 # is interpreted as x IN 1 or something similar.
60 # Since we currently do not have access to the SQLA AST, resort
61 # to barbaric mutilation of any SQL supplied in literal form
63 sub _strip_outer_paren {
64 my ($self, $arg) = @_;
68 return $self->_SWITCH_refkind ($arg, {
70 $$arg->[0] = __strip_outer_paren ($$arg->[0]);
74 return \__strip_outer_paren( $$arg );
82 sub __strip_outer_paren {
85 if ($sql and not ref $sql) {
86 while ($sql =~ /^ \s* \( (.*) \) \s* $/x ) {
95 my ($self, $lhs, $op, $rhs) = @_;
96 $rhs = $self->_strip_outer_paren ($rhs);
97 return $self->SUPER::_where_field_IN ($lhs, $op, $rhs);
100 sub _where_field_BETWEEN {
101 my ($self, $lhs, $op, $rhs) = @_;
102 $rhs = $self->_strip_outer_paren ($rhs);
103 return $self->SUPER::_where_field_BETWEEN ($lhs, $op, $rhs);
108 # DB2 is the only remaining DB using this. Even though we are not sure if
109 # RowNumberOver is still needed here (should be part of SQLA) leave the
112 my ($self, $sql, $order, $rows, $offset ) = @_;
115 my $last = $rows + $offset;
116 my ( $order_by ) = $self->_order_by( $order );
121 SELECT Q1.*, ROW_NUMBER() OVER( ) AS ROW_NUM FROM (
126 WHERE ROW_NUM BETWEEN $offset AND $last
134 # While we're at it, this should make LIMIT queries more efficient,
135 # without digging into things too deeply
136 use Scalar::Util 'blessed';
138 my ($self, $syntax) = @_;
140 # DB2 is the only remaining DB using this. Even though we are not sure if
141 # RowNumberOver is still needed here (should be part of SQLA) leave the
143 my $dbhname = blessed($syntax) ? $syntax->{Driver}{Name} : $syntax;
144 if(ref($self) && $dbhname && $dbhname eq 'DB2') {
145 return 'RowNumberOver';
148 $self->{_cached_syntax} ||= $self->SUPER::_find_syntax($syntax);
152 my ($self, $table, $fields, $where, $order, @rest) = @_;
153 local $self->{having_bind} = [];
154 local $self->{from_bind} = [];
156 if (ref $table eq 'SCALAR') {
159 elsif (not ref $table) {
160 $table = $self->_quote($table);
162 local $self->{rownum_hack_count} = 1
163 if (defined $rest[0] && $self->{limit_dialect} eq 'RowNum');
164 @rest = (-1) unless defined $rest[0];
165 die "LIMIT 0 Does Not Compute" if $rest[0] == 0;
166 # and anyway, SQL::Abstract::Limit will cause a barf if we don't first
167 my ($sql, @where_bind) = $self->SUPER::select(
168 $table, $self->_recurse_fields($fields), $where, $order, @rest
173 $self->{for} eq 'update' ? ' FOR UPDATE' :
174 $self->{for} eq 'shared' ? ' FOR SHARE' :
179 return wantarray ? ($sql, @{$self->{from_bind}}, @where_bind, @{$self->{having_bind}}) : $sql;
185 $table = $self->_quote($table) unless ref($table);
186 $self->SUPER::insert($table, @_);
192 $table = $self->_quote($table) unless ref($table);
193 $self->SUPER::update($table, @_);
199 $table = $self->_quote($table) unless ref($table);
200 $self->SUPER::delete($table, @_);
206 return $_[1].$self->_order_by($_[2]);
208 return $self->SUPER::_emulate_limit(@_);
212 sub _recurse_fields {
213 my ($self, $fields, $params) = @_;
214 my $ref = ref $fields;
215 return $self->_quote($fields) unless $ref;
216 return $$fields if $ref eq 'SCALAR';
218 if ($ref eq 'ARRAY') {
219 return join(', ', map {
220 $self->_recurse_fields($_)
221 .(exists $self->{rownum_hack_count} && !($params && $params->{no_rownum_hack})
222 ? ' AS col'.$self->{rownum_hack_count}++
225 } elsif ($ref eq 'HASH') {
226 foreach my $func (keys %$fields) {
227 if ($func eq 'distinct') {
228 my $_fields = $fields->{$func};
229 if (ref $_fields eq 'ARRAY' && @{$_fields} > 1) {
230 die "Unsupported syntax, please use " .
231 "{ group_by => [ qw/" . (join ' ', @$_fields) . "/ ] }" .
233 "{ select => [ qw/" . (join ' ', @$_fields) . "/ ], distinct => 1 }";
236 carp "This syntax will be deprecated in 09, please use " .
237 "{ group_by => '${_fields}' }" .
239 "{ select => '${_fields}', distinct => 1 }";
243 return $self->_sqlcase($func)
244 .'( '.$self->_recurse_fields($fields->{$func}).' )';
247 # Is the second check absolutely necessary?
248 elsif ( $ref eq 'REF' and ref($$fields) eq 'ARRAY' ) {
249 return $self->_fold_sqlbind( $fields );
252 Carp::croak($ref . qq{ unexpected in _recurse_fields()})
260 if (ref $_[0] eq 'HASH') {
261 if (defined $_[0]->{group_by}) {
262 $ret = $self->_sqlcase(' group by ')
263 .$self->_recurse_fields($_[0]->{group_by}, { no_rownum_hack => 1 });
265 if (defined $_[0]->{having}) {
267 ($frag, @extra) = $self->_recurse_where($_[0]->{having});
268 push(@{$self->{having_bind}}, @extra);
269 $ret .= $self->_sqlcase(' having ').$frag;
271 if (defined $_[0]->{order_by}) {
272 $ret .= $self->_order_by($_[0]->{order_by});
274 if (grep { $_ =~ /^-(desc|asc)/i } keys %{$_[0]}) {
275 return $self->SUPER::_order_by($_[0]);
277 } elsif (ref $_[0] eq 'SCALAR') {
278 $ret = $self->_sqlcase(' order by ').${ $_[0] };
279 } elsif (ref $_[0] eq 'ARRAY' && @{$_[0]}) {
280 my @order = @{+shift};
281 $ret = $self->_sqlcase(' order by ')
283 my $r = $self->_order_by($_, @_);
284 $r =~ s/^ ?ORDER BY //i;
288 $ret = $self->SUPER::_order_by(@_);
293 sub _order_directions {
294 my ($self, $order) = @_;
295 $order = $order->{order_by} if ref $order eq 'HASH';
296 return $self->SUPER::_order_directions($order);
300 my ($self, $from) = @_;
301 if (ref $from eq 'ARRAY') {
302 return $self->_recurse_from(@$from);
303 } elsif (ref $from eq 'HASH') {
304 return $self->_make_as($from);
306 return $from; # would love to quote here but _table ends up getting called
307 # twice during an ->select without a limit clause due to
308 # the way S::A::Limit->select works. should maybe consider
309 # bypassing this and doing S::A::select($self, ...) in
310 # our select method above. meantime, quoting shims have
311 # been added to select/insert/update/delete here
316 my ($self, $from, @join) = @_;
318 push(@sqlf, $self->_make_as($from));
319 foreach my $j (@join) {
322 # check whether a join type exists
323 my $join_clause = '';
324 my $to_jt = ref($to) eq 'ARRAY' ? $to->[0] : $to;
325 if (ref($to_jt) eq 'HASH' and exists($to_jt->{-join_type})) {
326 $join_clause = ' '.uc($to_jt->{-join_type}).' JOIN ';
328 $join_clause = ' JOIN ';
330 push(@sqlf, $join_clause);
332 if (ref $to eq 'ARRAY') {
333 push(@sqlf, '(', $self->_recurse_from(@$to), ')');
335 push(@sqlf, $self->_make_as($to));
337 push(@sqlf, ' ON ', $self->_join_condition($on));
339 return join('', @sqlf);
343 my ($self, $sqlbind) = @_;
344 my $sql = shift @$$sqlbind;
345 push @{$self->{from_bind}}, @$$sqlbind;
350 my ($self, $from) = @_;
351 return join(' ', map { (ref $_ eq 'SCALAR' ? $$_
352 : ref $_ eq 'REF' ? $self->_fold_sqlbind($_)
354 } reverse each %{$self->_skip_options($from)});
358 my ($self, $hash) = @_;
360 $clean_hash->{$_} = $hash->{$_}
361 for grep {!/^-/} keys %$hash;
365 sub _join_condition {
366 my ($self, $cond) = @_;
367 if (ref $cond eq 'HASH') {
372 # XXX no throw_exception() in this package and croak() fails with strange results
373 Carp::croak(ref($v) . qq{ reference arguments are not supported in JOINS - try using \"..." instead'})
374 if ref($v) ne 'SCALAR';
378 my $x = '= '.$self->_quote($v); $j{$_} = \$x;
381 return scalar($self->_recurse_where(\%j));
382 } elsif (ref $cond eq 'ARRAY') {
383 return join(' OR ', map { $self->_join_condition($_) } @$cond);
385 die "Can't handle this yet!";
390 my ($self, $label) = @_;
391 return '' unless defined $label;
392 return "*" if $label eq '*';
393 return $label unless $self->{quote_char};
394 if(ref $self->{quote_char} eq "ARRAY"){
395 return $self->{quote_char}->[0] . $label . $self->{quote_char}->[1]
396 if !defined $self->{name_sep};
397 my $sep = $self->{name_sep};
398 return join($self->{name_sep},
399 map { $self->{quote_char}->[0] . $_ . $self->{quote_char}->[1] }
400 split(/\Q$sep\E/,$label));
402 return $self->SUPER::_quote($label);
407 $self->{limit_dialect} = shift if @_;
408 return $self->{limit_dialect};
413 $self->{quote_char} = shift if @_;
414 return $self->{quote_char};
419 $self->{name_sep} = shift if @_;
420 return $self->{name_sep};
423 } # End of BEGIN block
427 DBIx::Class::Storage::DBI - DBI storage handler
431 my $schema = MySchema->connect('dbi:SQLite:my.db');
433 $schema->storage->debug(1);
434 $schema->dbh_do("DROP TABLE authors");
436 $schema->resultset('Book')->search({
437 written_on => $schema->storage->datetime_parser(DateTime->now)
442 This class represents the connection to an RDBMS via L<DBI>. See
443 L<DBIx::Class::Storage> for general information. This pod only
444 documents DBI-specific methods and behaviors.
451 my $new = shift->next::method(@_);
453 $new->transaction_depth(0);
454 $new->_sql_maker_opts({});
455 $new->{savepoints} = [];
456 $new->{_in_dbh_do} = 0;
457 $new->{_dbh_gen} = 0;
464 This method is normally called by L<DBIx::Class::Schema/connection>, which
465 encapsulates its argument list in an arrayref before passing them here.
467 The argument list may contain:
473 The same 4-element argument set one would normally pass to
474 L<DBI/connect>, optionally followed by
475 L<extra attributes|/DBIx::Class specific connection attributes>
476 recognized by DBIx::Class:
478 $connect_info_args = [ $dsn, $user, $password, \%dbi_attributes?, \%extra_attributes? ];
482 A single code reference which returns a connected
483 L<DBI database handle|DBI/connect> optionally followed by
484 L<extra attributes|/DBIx::Class specific connection attributes> recognized
487 $connect_info_args = [ sub { DBI->connect (...) }, \%extra_attributes? ];
491 A single hashref with all the attributes and the dsn/user/password
494 $connect_info_args = [{
502 This is particularly useful for L<Catalyst> based applications, allowing the
503 following config (L<Config::General> style):
508 dsn dbi:mysql:database=test
517 Please note that the L<DBI> docs recommend that you always explicitly
518 set C<AutoCommit> to either I<0> or I<1>. L<DBIx::Class> further
519 recommends that it be set to I<1>, and that you perform transactions
520 via our L<DBIx::Class::Schema/txn_do> method. L<DBIx::Class> will set it
521 to I<1> if you do not do explicitly set it to zero. This is the default
522 for most DBDs. See L</DBIx::Class and AutoCommit> for details.
524 =head3 DBIx::Class specific connection attributes
526 In addition to the standard L<DBI|DBI/ATTRIBUTES_COMMON_TO_ALL_HANDLES>
527 L<connection|DBI/Database_Handle_Attributes> attributes, DBIx::Class recognizes
528 the following connection options. These options can be mixed in with your other
529 L<DBI> connection attributes, or placed in a seperate hashref
530 (C<\%extra_attributes>) as shown above.
532 Every time C<connect_info> is invoked, any previous settings for
533 these options will be cleared before setting the new ones, regardless of
534 whether any options are specified in the new C<connect_info>.
541 Specifies things to do immediately after connecting or re-connecting to
542 the database. Its value may contain:
546 =item an array reference
548 This contains SQL statements to execute in order. Each element contains
549 a string or a code reference that returns a string.
551 =item a code reference
553 This contains some code to execute. Unlike code references within an
554 array reference, its return value is ignored.
558 =item on_disconnect_do
560 Takes arguments in the same form as L</on_connect_do> and executes them
561 immediately before disconnecting from the database.
563 Note, this only runs if you explicitly call L</disconnect> on the
566 =item disable_sth_caching
568 If set to a true value, this option will disable the caching of
569 statement handles via L<DBI/prepare_cached>.
573 Sets the limit dialect. This is useful for JDBC-bridge among others
574 where the remote SQL-dialect cannot be determined by the name of the
575 driver alone. See also L<SQL::Abstract::Limit>.
579 Specifies what characters to use to quote table and column names. If
580 you use this you will want to specify L</name_sep> as well.
582 C<quote_char> expects either a single character, in which case is it
583 is placed on either side of the table/column name, or an arrayref of length
584 2 in which case the table/column name is placed between the elements.
586 For example under MySQL you should use C<< quote_char => '`' >>, and for
587 SQL Server you should use C<< quote_char => [qw/[ ]/] >>.
591 This only needs to be used in conjunction with C<quote_char>, and is used to
592 specify the charecter that seperates elements (schemas, tables, columns) from
593 each other. In most cases this is simply a C<.>.
595 The consequences of not supplying this value is that L<SQL::Abstract>
596 will assume DBIx::Class' uses of aliases to be complete column
597 names. The output will look like I<"me.name"> when it should actually
602 This Storage driver normally installs its own C<HandleError>, sets
603 C<RaiseError> and C<ShowErrorStatement> on, and sets C<PrintError> off on
604 all database handles, including those supplied by a coderef. It does this
605 so that it can have consistent and useful error behavior.
607 If you set this option to a true value, Storage will not do its usual
608 modifications to the database handle's attributes, and instead relies on
609 the settings in your connect_info DBI options (or the values you set in
610 your connection coderef, in the case that you are connecting via coderef).
612 Note that your custom settings can cause Storage to malfunction,
613 especially if you set a C<HandleError> handler that suppresses exceptions
614 and/or disable C<RaiseError>.
618 If this option is true, L<DBIx::Class> will use savepoints when nesting
619 transactions, making it possible to recover from failure in the inner
620 transaction without having to abort all outer transactions.
624 Use this argument to supply a cursor class other than the default
625 L<DBIx::Class::Storage::DBI::Cursor>.
629 Some real-life examples of arguments to L</connect_info> and
630 L<DBIx::Class::Schema/connect>
632 # Simple SQLite connection
633 ->connect_info([ 'dbi:SQLite:./foo.db' ]);
636 ->connect_info([ sub { DBI->connect(...) } ]);
638 # A bit more complicated
645 { quote_char => q{"}, name_sep => q{.} },
649 # Equivalent to the previous example
655 { AutoCommit => 1, quote_char => q{"}, name_sep => q{.} },
659 # Same, but with hashref as argument
660 # See parse_connect_info for explanation
663 dsn => 'dbi:Pg:dbname=foo',
665 password => 'my_pg_password',
672 # Subref + DBIx::Class-specific connection options
675 sub { DBI->connect(...) },
679 on_connect_do => ['SET search_path TO myschema,otherschema,public'],
680 disable_sth_caching => 1,
690 my ($self, $info_arg) = @_;
692 return $self->_connect_info if !$info_arg;
694 my @args = @$info_arg; # take a shallow copy for further mutilation
695 $self->_connect_info([@args]); # copy for _connect_info
698 # combine/pre-parse arguments depending on invocation style
701 if (ref $args[0] eq 'CODE') { # coderef with optional \%extra_attributes
702 %attrs = %{ $args[1] || {} };
705 elsif (ref $args[0] eq 'HASH') { # single hashref (i.e. Catalyst config)
706 %attrs = %{$args[0]};
708 for (qw/password user dsn/) {
709 unshift @args, delete $attrs{$_};
712 else { # otherwise assume dsn/user/password + \%attrs + \%extra_attrs
714 % { $args[3] || {} },
715 % { $args[4] || {} },
717 @args = @args[0,1,2];
720 # Kill sql_maker/_sql_maker_opts, so we get a fresh one with only
721 # the new set of options
722 $self->_sql_maker(undef);
723 $self->_sql_maker_opts({});
726 for my $storage_opt (@storage_options, 'cursor_class') { # @storage_options is declared at the top of the module
727 if(my $value = delete $attrs{$storage_opt}) {
728 $self->$storage_opt($value);
731 for my $sql_maker_opt (qw/limit_dialect quote_char name_sep/) {
732 if(my $opt_val = delete $attrs{$sql_maker_opt}) {
733 $self->_sql_maker_opts->{$sql_maker_opt} = $opt_val;
738 %attrs = () if (ref $args[0] eq 'CODE'); # _connect() never looks past $args[0] in this case
740 $self->_dbi_connect_info([@args, keys %attrs ? \%attrs : ()]);
741 $self->_connect_info;
746 This method is deprecated in favour of setting via L</connect_info>.
751 Arguments: ($subref | $method_name), @extra_coderef_args?
753 Execute the given $subref or $method_name using the new exception-based
754 connection management.
756 The first two arguments will be the storage object that C<dbh_do> was called
757 on and a database handle to use. Any additional arguments will be passed
758 verbatim to the called subref as arguments 2 and onwards.
760 Using this (instead of $self->_dbh or $self->dbh) ensures correct
761 exception handling and reconnection (or failover in future subclasses).
763 Your subref should have no side-effects outside of the database, as
764 there is the potential for your subref to be partially double-executed
765 if the database connection was stale/dysfunctional.
769 my @stuff = $schema->storage->dbh_do(
771 my ($storage, $dbh, @cols) = @_;
772 my $cols = join(q{, }, @cols);
773 $dbh->selectrow_array("SELECT $cols FROM foo");
784 my $dbh = $self->_dbh;
786 return $self->$code($dbh, @_) if $self->{_in_dbh_do}
787 || $self->{transaction_depth};
789 local $self->{_in_dbh_do} = 1;
792 my $want_array = wantarray;
795 $self->_verify_pid if $dbh;
797 $self->_populate_dbh;
802 @result = $self->$code($dbh, @_);
804 elsif(defined $want_array) {
805 $result[0] = $self->$code($dbh, @_);
808 $self->$code($dbh, @_);
813 if(!$exception) { return $want_array ? @result : $result[0] }
815 $self->throw_exception($exception) if $self->connected;
817 # We were not connected - reconnect and retry, but let any
818 # exception fall right through this time
819 $self->_populate_dbh;
820 $self->$code($self->_dbh, @_);
823 # This is basically a blend of dbh_do above and DBIx::Class::Storage::txn_do.
824 # It also informs dbh_do to bypass itself while under the direction of txn_do,
825 # via $self->{_in_dbh_do} (this saves some redundant eval and errorcheck, etc)
830 ref $coderef eq 'CODE' or $self->throw_exception
831 ('$coderef must be a CODE reference');
833 return $coderef->(@_) if $self->{transaction_depth} && ! $self->auto_savepoint;
835 local $self->{_in_dbh_do} = 1;
838 my $want_array = wantarray;
843 $self->_verify_pid if $self->_dbh;
844 $self->_populate_dbh if !$self->_dbh;
848 @result = $coderef->(@_);
850 elsif(defined $want_array) {
851 $result[0] = $coderef->(@_);
860 if(!$exception) { return $want_array ? @result : $result[0] }
862 if($tried++ > 0 || $self->connected) {
863 eval { $self->txn_rollback };
864 my $rollback_exception = $@;
865 if($rollback_exception) {
866 my $exception_class = "DBIx::Class::Storage::NESTED_ROLLBACK_EXCEPTION";
867 $self->throw_exception($exception) # propagate nested rollback
868 if $rollback_exception =~ /$exception_class/;
870 $self->throw_exception(
871 "Transaction aborted: ${exception}. "
872 . "Rollback failed: ${rollback_exception}"
875 $self->throw_exception($exception)
878 # We were not connected, and was first try - reconnect and retry
880 $self->_populate_dbh;
886 Our C<disconnect> method also performs a rollback first if the
887 database is not in C<AutoCommit> mode.
894 if( $self->connected ) {
895 my $connection_do = $self->on_disconnect_do;
896 $self->_do_connection_actions($connection_do) if ref($connection_do);
898 $self->_dbh->rollback unless $self->_dbh_autocommit;
899 $self->_dbh->disconnect;
905 =head2 with_deferred_fk_checks
909 =item Arguments: C<$coderef>
911 =item Return Value: The return value of $coderef
915 Storage specific method to run the code ref with FK checks deferred or
916 in MySQL's case disabled entirely.
920 # Storage subclasses should override this
921 sub with_deferred_fk_checks {
922 my ($self, $sub) = @_;
930 if(my $dbh = $self->_dbh) {
931 if(defined $self->_conn_tid && $self->_conn_tid != threads->tid) {
938 return 0 if !$self->_dbh;
940 return ($dbh->FETCH('Active') && $dbh->ping);
946 # handle pid changes correctly
947 # NOTE: assumes $self->_dbh is a valid $dbh
951 return if defined $self->_conn_pid && $self->_conn_pid == $$;
953 $self->_dbh->{InactiveDestroy} = 1;
960 sub ensure_connected {
963 unless ($self->connected) {
964 $self->_populate_dbh;
970 Returns the dbh - a data base handle of class L<DBI>.
977 $self->ensure_connected;
981 sub _sql_maker_args {
984 return ( bindtype=>'columns', array_datatypes => 1, limit_dialect => $self->dbh, %{$self->_sql_maker_opts} );
989 unless ($self->_sql_maker) {
990 my $sql_maker_class = $self->sql_maker_class;
991 $self->_sql_maker($sql_maker_class->new( $self->_sql_maker_args ));
993 return $self->_sql_maker;
1000 my @info = @{$self->_dbi_connect_info || []};
1001 $self->_dbh($self->_connect(@info));
1003 # Always set the transaction depth on connect, since
1004 # there is no transaction in progress by definition
1005 $self->{transaction_depth} = $self->_dbh_autocommit ? 0 : 1;
1007 if(ref $self eq 'DBIx::Class::Storage::DBI') {
1008 my $driver = $self->_dbh->{Driver}->{Name};
1009 if ($self->load_optional_class("DBIx::Class::Storage::DBI::${driver}")) {
1010 bless $self, "DBIx::Class::Storage::DBI::${driver}";
1015 $self->_conn_pid($$);
1016 $self->_conn_tid(threads->tid) if $INC{'threads.pm'};
1018 my $connection_do = $self->on_connect_do;
1019 $self->_do_connection_actions($connection_do) if ref($connection_do);
1022 sub _do_connection_actions {
1024 my $connection_do = shift;
1026 if (ref $connection_do eq 'ARRAY') {
1027 $self->_do_query($_) foreach @$connection_do;
1029 elsif (ref $connection_do eq 'CODE') {
1030 $connection_do->($self);
1037 my ($self, $action) = @_;
1039 if (ref $action eq 'CODE') {
1040 $action = $action->($self);
1041 $self->_do_query($_) foreach @$action;
1044 # Most debuggers expect ($sql, @bind), so we need to exclude
1045 # the attribute hash which is the second argument to $dbh->do
1046 # furthermore the bind values are usually to be presented
1047 # as named arrayref pairs, so wrap those here too
1048 my @do_args = (ref $action eq 'ARRAY') ? (@$action) : ($action);
1049 my $sql = shift @do_args;
1050 my $attrs = shift @do_args;
1051 my @bind = map { [ undef, $_ ] } @do_args;
1053 $self->_query_start($sql, @bind);
1054 $self->_dbh->do($sql, $attrs, @do_args);
1055 $self->_query_end($sql, @bind);
1062 my ($self, @info) = @_;
1064 $self->throw_exception("You failed to provide any connection info")
1067 my ($old_connect_via, $dbh);
1069 if ($INC{'Apache/DBI.pm'} && $ENV{MOD_PERL}) {
1070 $old_connect_via = $DBI::connect_via;
1071 $DBI::connect_via = 'connect';
1075 if(ref $info[0] eq 'CODE') {
1079 $dbh = DBI->connect(@info);
1082 if($dbh && !$self->unsafe) {
1083 my $weak_self = $self;
1085 $dbh->{HandleError} = sub {
1087 $weak_self->throw_exception("DBI Exception: $_[0]");
1090 croak ("DBI Exception: $_[0]");
1093 $dbh->{ShowErrorStatement} = 1;
1094 $dbh->{RaiseError} = 1;
1095 $dbh->{PrintError} = 0;
1099 $DBI::connect_via = $old_connect_via if $old_connect_via;
1101 $self->throw_exception("DBI Connection failed: " . ($@||$DBI::errstr))
1104 $self->_dbh_autocommit($dbh->{AutoCommit});
1110 my ($self, $name) = @_;
1112 $name = $self->_svp_generate_name
1113 unless defined $name;
1115 $self->throw_exception ("You can't use savepoints outside a transaction")
1116 if $self->{transaction_depth} == 0;
1118 $self->throw_exception ("Your Storage implementation doesn't support savepoints")
1119 unless $self->can('_svp_begin');
1121 push @{ $self->{savepoints} }, $name;
1123 $self->debugobj->svp_begin($name) if $self->debug;
1125 return $self->_svp_begin($name);
1129 my ($self, $name) = @_;
1131 $self->throw_exception ("You can't use savepoints outside a transaction")
1132 if $self->{transaction_depth} == 0;
1134 $self->throw_exception ("Your Storage implementation doesn't support savepoints")
1135 unless $self->can('_svp_release');
1137 if (defined $name) {
1138 $self->throw_exception ("Savepoint '$name' does not exist")
1139 unless grep { $_ eq $name } @{ $self->{savepoints} };
1141 # Dig through the stack until we find the one we are releasing. This keeps
1142 # the stack up to date.
1145 do { $svp = pop @{ $self->{savepoints} } } while $svp ne $name;
1147 $name = pop @{ $self->{savepoints} };
1150 $self->debugobj->svp_release($name) if $self->debug;
1152 return $self->_svp_release($name);
1156 my ($self, $name) = @_;
1158 $self->throw_exception ("You can't use savepoints outside a transaction")
1159 if $self->{transaction_depth} == 0;
1161 $self->throw_exception ("Your Storage implementation doesn't support savepoints")
1162 unless $self->can('_svp_rollback');
1164 if (defined $name) {
1165 # If they passed us a name, verify that it exists in the stack
1166 unless(grep({ $_ eq $name } @{ $self->{savepoints} })) {
1167 $self->throw_exception("Savepoint '$name' does not exist!");
1170 # Dig through the stack until we find the one we are releasing. This keeps
1171 # the stack up to date.
1172 while(my $s = pop(@{ $self->{savepoints} })) {
1173 last if($s eq $name);
1175 # Add the savepoint back to the stack, as a rollback doesn't remove the
1176 # named savepoint, only everything after it.
1177 push(@{ $self->{savepoints} }, $name);
1179 # We'll assume they want to rollback to the last savepoint
1180 $name = $self->{savepoints}->[-1];
1183 $self->debugobj->svp_rollback($name) if $self->debug;
1185 return $self->_svp_rollback($name);
1188 sub _svp_generate_name {
1191 return 'savepoint_'.scalar(@{ $self->{'savepoints'} });
1196 $self->ensure_connected();
1197 if($self->{transaction_depth} == 0) {
1198 $self->debugobj->txn_begin()
1200 # this isn't ->_dbh-> because
1201 # we should reconnect on begin_work
1202 # for AutoCommit users
1203 $self->dbh->begin_work;
1204 } elsif ($self->auto_savepoint) {
1207 $self->{transaction_depth}++;
1212 if ($self->{transaction_depth} == 1) {
1213 my $dbh = $self->_dbh;
1214 $self->debugobj->txn_commit()
1217 $self->{transaction_depth} = 0
1218 if $self->_dbh_autocommit;
1220 elsif($self->{transaction_depth} > 1) {
1221 $self->{transaction_depth}--;
1223 if $self->auto_savepoint;
1229 my $dbh = $self->_dbh;
1231 if ($self->{transaction_depth} == 1) {
1232 $self->debugobj->txn_rollback()
1234 $self->{transaction_depth} = 0
1235 if $self->_dbh_autocommit;
1238 elsif($self->{transaction_depth} > 1) {
1239 $self->{transaction_depth}--;
1240 if ($self->auto_savepoint) {
1241 $self->svp_rollback;
1246 die DBIx::Class::Storage::NESTED_ROLLBACK_EXCEPTION->new;
1251 my $exception_class = "DBIx::Class::Storage::NESTED_ROLLBACK_EXCEPTION";
1252 $error =~ /$exception_class/ and $self->throw_exception($error);
1253 # ensure that a failed rollback resets the transaction depth
1254 $self->{transaction_depth} = $self->_dbh_autocommit ? 0 : 1;
1255 $self->throw_exception($error);
1259 # This used to be the top-half of _execute. It was split out to make it
1260 # easier to override in NoBindVars without duping the rest. It takes up
1261 # all of _execute's args, and emits $sql, @bind.
1262 sub _prep_for_execute {
1263 my ($self, $op, $extra_bind, $ident, $args) = @_;
1265 if( blessed($ident) && $ident->isa("DBIx::Class::ResultSource") ) {
1266 $ident = $ident->from();
1269 my ($sql, @bind) = $self->sql_maker->$op($ident, @$args);
1272 map { ref $_ eq 'ARRAY' ? $_ : [ '!!dummy', $_ ] } @$extra_bind)
1274 return ($sql, \@bind);
1277 sub _fix_bind_params {
1278 my ($self, @bind) = @_;
1280 ### Turn @bind from something like this:
1281 ### ( [ "artist", 1 ], [ "cdid", 1, 3 ] )
1283 ### ( "'1'", "'1'", "'3'" )
1286 if ( defined( $_ && $_->[1] ) ) {
1287 map { qq{'$_'}; } @{$_}[ 1 .. $#$_ ];
1294 my ( $self, $sql, @bind ) = @_;
1296 if ( $self->debug ) {
1297 @bind = $self->_fix_bind_params(@bind);
1299 $self->debugobj->query_start( $sql, @bind );
1304 my ( $self, $sql, @bind ) = @_;
1306 if ( $self->debug ) {
1307 @bind = $self->_fix_bind_params(@bind);
1308 $self->debugobj->query_end( $sql, @bind );
1313 my ($self, $dbh, $op, $extra_bind, $ident, $bind_attributes, @args) = @_;
1315 my ($sql, $bind) = $self->_prep_for_execute($op, $extra_bind, $ident, \@args);
1317 $self->_query_start( $sql, @$bind );
1319 my $sth = $self->sth($sql,$op);
1321 my $placeholder_index = 1;
1323 foreach my $bound (@$bind) {
1324 my $attributes = {};
1325 my($column_name, @data) = @$bound;
1327 if ($bind_attributes) {
1328 $attributes = $bind_attributes->{$column_name}
1329 if defined $bind_attributes->{$column_name};
1332 foreach my $data (@data) {
1333 my $ref = ref $data;
1334 $data = $ref && $ref ne 'ARRAY' ? ''.$data : $data; # stringify args (except arrayrefs)
1336 $sth->bind_param($placeholder_index, $data, $attributes);
1337 $placeholder_index++;
1341 # Can this fail without throwing an exception anyways???
1342 my $rv = $sth->execute();
1343 $self->throw_exception($sth->errstr) if !$rv;
1345 $self->_query_end( $sql, @$bind );
1347 return (wantarray ? ($rv, $sth, @$bind) : $rv);
1352 $self->dbh_do('_dbh_execute', @_)
1356 my ($self, $source, $to_insert) = @_;
1358 my $ident = $source->from;
1359 my $bind_attributes = $self->source_bind_attributes($source);
1361 my $updated_cols = {};
1363 $self->ensure_connected;
1364 foreach my $col ( $source->columns ) {
1365 if ( !defined $to_insert->{$col} ) {
1366 my $col_info = $source->column_info($col);
1368 if ( $col_info->{auto_nextval} ) {
1369 $updated_cols->{$col} = $to_insert->{$col} = $self->_sequence_fetch( 'nextval', $col_info->{sequence} || $self->_dbh_get_autoinc_seq($self->dbh, $source) );
1374 $self->_execute('insert' => [], $source, $bind_attributes, $to_insert);
1376 return $updated_cols;
1379 ## Still not quite perfect, and EXPERIMENTAL
1380 ## Currently it is assumed that all values passed will be "normal", i.e. not
1381 ## scalar refs, or at least, all the same type as the first set, the statement is
1382 ## only prepped once.
1384 my ($self, $source, $cols, $data) = @_;
1386 my $table = $source->from;
1387 @colvalues{@$cols} = (0..$#$cols);
1388 my ($sql, @bind) = $self->sql_maker->insert($table, \%colvalues);
1390 $self->_query_start( $sql, @bind );
1391 my $sth = $self->sth($sql);
1393 # @bind = map { ref $_ ? ''.$_ : $_ } @bind; # stringify args
1395 ## This must be an arrayref, else nothing works!
1397 my $tuple_status = [];
1400 ##print STDERR Dumper( $data, $sql, [@bind] );
1404 ## Get the bind_attributes, if any exist
1405 my $bind_attributes = $self->source_bind_attributes($source);
1407 ## Bind the values and execute
1408 my $placeholder_index = 1;
1410 foreach my $bound (@bind) {
1412 my $attributes = {};
1413 my ($column_name, $data_index) = @$bound;
1415 if( $bind_attributes ) {
1416 $attributes = $bind_attributes->{$column_name}
1417 if defined $bind_attributes->{$column_name};
1420 my @data = map { $_->[$data_index] } @$data;
1422 $sth->bind_param_array( $placeholder_index, [@data], $attributes );
1423 $placeholder_index++;
1425 my $rv = $sth->execute_array({ArrayTupleStatus => $tuple_status});
1426 $self->throw_exception($sth->errstr) if !$rv;
1428 $self->_query_end( $sql, @bind );
1429 return (wantarray ? ($rv, $sth, @bind) : $rv);
1433 my $self = shift @_;
1434 my $source = shift @_;
1435 my $bind_attributes = $self->source_bind_attributes($source);
1437 return $self->_execute('update' => [], $source, $bind_attributes, @_);
1442 my $self = shift @_;
1443 my $source = shift @_;
1445 my $bind_attrs = {}; ## If ever it's needed...
1447 return $self->_execute('delete' => [], $source, $bind_attrs, @_);
1452 my $sql_maker = $self->sql_maker;
1453 local $sql_maker->{for};
1454 return $self->_execute($self->_select_args(@_));
1458 my ($self, $ident, $select, $condition, $attrs) = @_;
1459 my $order = $attrs->{order_by};
1461 my $for = delete $attrs->{for};
1462 my $sql_maker = $self->sql_maker;
1463 $sql_maker->{for} = $for;
1465 if (exists $attrs->{group_by} || $attrs->{having}) {
1467 group_by => $attrs->{group_by},
1468 having => $attrs->{having},
1469 ($order ? (order_by => $order) : ())
1472 my $bind_attrs = {}; ## Future support
1473 my @args = ('select', $attrs->{bind}, $ident, $bind_attrs, $select, $condition, $order);
1474 if ($attrs->{software_limit} ||
1475 $self->sql_maker->_default_limit_syntax eq "GenericSubQ") {
1476 $attrs->{software_limit} = 1;
1478 $self->throw_exception("rows attribute must be positive if present")
1479 if (defined($attrs->{rows}) && !($attrs->{rows} > 0));
1481 # MySQL actually recommends this approach. I cringe.
1482 $attrs->{rows} = 2**48 if not defined $attrs->{rows} and defined $attrs->{offset};
1483 push @args, $attrs->{rows}, $attrs->{offset};
1488 sub source_bind_attributes {
1489 my ($self, $source) = @_;
1491 my $bind_attributes;
1492 foreach my $column ($source->columns) {
1494 my $data_type = $source->column_info($column)->{data_type} || '';
1495 $bind_attributes->{$column} = $self->bind_attribute_by_data_type($data_type)
1499 return $bind_attributes;
1506 =item Arguments: $ident, $select, $condition, $attrs
1510 Handle a SQL select statement.
1516 my ($ident, $select, $condition, $attrs) = @_;
1517 return $self->cursor_class->new($self, \@_, $attrs);
1522 my ($rv, $sth, @bind) = $self->_select(@_);
1523 my @row = $sth->fetchrow_array;
1524 my @nextrow = $sth->fetchrow_array if @row;
1525 if(@row && @nextrow) {
1526 carp "Query returned more than one row. SQL that returns multiple rows is DEPRECATED for ->find and ->single";
1528 # Need to call finish() to work round broken DBDs
1537 =item Arguments: $sql
1541 Returns a L<DBI> sth (statement handle) for the supplied SQL.
1546 my ($self, $dbh, $sql) = @_;
1548 # 3 is the if_active parameter which avoids active sth re-use
1549 my $sth = $self->disable_sth_caching
1550 ? $dbh->prepare($sql)
1551 : $dbh->prepare_cached($sql, {}, 3);
1553 # XXX You would think RaiseError would make this impossible,
1554 # but apparently that's not true :(
1555 $self->throw_exception($dbh->errstr) if !$sth;
1561 my ($self, $sql) = @_;
1562 $self->dbh_do('_dbh_sth', $sql);
1565 sub _dbh_columns_info_for {
1566 my ($self, $dbh, $table) = @_;
1568 if ($dbh->can('column_info')) {
1571 my ($schema,$tab) = $table =~ /^(.+?)\.(.+)$/ ? ($1,$2) : (undef,$table);
1572 my $sth = $dbh->column_info( undef,$schema, $tab, '%' );
1574 while ( my $info = $sth->fetchrow_hashref() ){
1576 $column_info{data_type} = $info->{TYPE_NAME};
1577 $column_info{size} = $info->{COLUMN_SIZE};
1578 $column_info{is_nullable} = $info->{NULLABLE} ? 1 : 0;
1579 $column_info{default_value} = $info->{COLUMN_DEF};
1580 my $col_name = $info->{COLUMN_NAME};
1581 $col_name =~ s/^\"(.*)\"$/$1/;
1583 $result{$col_name} = \%column_info;
1586 return \%result if !$@ && scalar keys %result;
1590 my $sth = $dbh->prepare($self->sql_maker->select($table, undef, \'1 = 0'));
1592 my @columns = @{$sth->{NAME_lc}};
1593 for my $i ( 0 .. $#columns ){
1595 $column_info{data_type} = $sth->{TYPE}->[$i];
1596 $column_info{size} = $sth->{PRECISION}->[$i];
1597 $column_info{is_nullable} = $sth->{NULLABLE}->[$i] ? 1 : 0;
1599 if ($column_info{data_type} =~ m/^(.*?)\((.*?)\)$/) {
1600 $column_info{data_type} = $1;
1601 $column_info{size} = $2;
1604 $result{$columns[$i]} = \%column_info;
1608 foreach my $col (keys %result) {
1609 my $colinfo = $result{$col};
1610 my $type_num = $colinfo->{data_type};
1612 if(defined $type_num && $dbh->can('type_info')) {
1613 my $type_info = $dbh->type_info($type_num);
1614 $type_name = $type_info->{TYPE_NAME} if $type_info;
1615 $colinfo->{data_type} = $type_name if $type_name;
1622 sub columns_info_for {
1623 my ($self, $table) = @_;
1624 $self->dbh_do('_dbh_columns_info_for', $table);
1627 =head2 last_insert_id
1629 Return the row id of the last insert.
1633 sub _dbh_last_insert_id {
1634 # All Storage's need to register their own _dbh_last_insert_id
1635 # the old SQLite-based method was highly inappropriate
1638 my $class = ref $self;
1639 $self->throw_exception (<<EOE);
1641 No _dbh_last_insert_id() method found in $class.
1642 Since the method of obtaining the autoincrement id of the last insert
1643 operation varies greatly between different databases, this method must be
1644 individually implemented for every storage class.
1648 sub last_insert_id {
1650 $self->dbh_do('_dbh_last_insert_id', @_);
1655 Returns the database driver name.
1659 sub sqlt_type { shift->dbh->{Driver}->{Name} }
1661 =head2 bind_attribute_by_data_type
1663 Given a datatype from column info, returns a database specific bind
1664 attribute for C<< $dbh->bind_param($val,$attribute) >> or nothing if we will
1665 let the database planner just handle it.
1667 Generally only needed for special case column types, like bytea in postgres.
1671 sub bind_attribute_by_data_type {
1675 =head2 create_ddl_dir
1679 =item Arguments: $schema \@databases, $version, $directory, $preversion, \%sqlt_args
1683 Creates a SQL file based on the Schema, for each of the specified
1684 database types, in the given directory.
1686 By default, C<\%sqlt_args> will have
1688 { add_drop_table => 1, ignore_constraint_names => 1, ignore_index_names => 1 }
1690 merged with the hash passed in. To disable any of those features, pass in a
1691 hashref like the following
1693 { ignore_constraint_names => 0, # ... other options }
1697 sub create_ddl_dir {
1698 my ($self, $schema, $databases, $version, $dir, $preversion, $sqltargs) = @_;
1700 if(!$dir || !-d $dir) {
1701 warn "No directory given, using ./\n";
1704 $databases ||= ['MySQL', 'SQLite', 'PostgreSQL'];
1705 $databases = [ $databases ] if(ref($databases) ne 'ARRAY');
1707 my $schema_version = $schema->schema_version || '1.x';
1708 $version ||= $schema_version;
1711 add_drop_table => 1,
1712 ignore_constraint_names => 1,
1713 ignore_index_names => 1,
1717 $self->throw_exception(q{Can't create a ddl file without SQL::Translator 0.09003: '}
1718 . $self->_check_sqlt_message . q{'})
1719 if !$self->_check_sqlt_version;
1721 my $sqlt = SQL::Translator->new( $sqltargs );
1723 $sqlt->parser('SQL::Translator::Parser::DBIx::Class');
1724 my $sqlt_schema = $sqlt->translate({ data => $schema }) or die $sqlt->error;
1726 foreach my $db (@$databases) {
1728 $sqlt->{schema} = $sqlt_schema;
1729 $sqlt->producer($db);
1732 my $filename = $schema->ddl_filename($db, $version, $dir);
1733 if (-e $filename && ($version eq $schema_version )) {
1734 # if we are dumping the current version, overwrite the DDL
1735 warn "Overwriting existing DDL file - $filename";
1739 my $output = $sqlt->translate;
1741 warn("Failed to translate to $db, skipping. (" . $sqlt->error . ")");
1744 if(!open($file, ">$filename")) {
1745 $self->throw_exception("Can't open $filename for writing ($!)");
1748 print $file $output;
1751 next unless ($preversion);
1753 require SQL::Translator::Diff;
1755 my $prefilename = $schema->ddl_filename($db, $preversion, $dir);
1756 if(!-e $prefilename) {
1757 warn("No previous schema file found ($prefilename)");
1761 my $difffile = $schema->ddl_filename($db, $version, $dir, $preversion);
1763 warn("Overwriting existing diff file - $difffile");
1769 my $t = SQL::Translator->new($sqltargs);
1772 $t->parser( $db ) or die $t->error;
1773 my $out = $t->translate( $prefilename ) or die $t->error;
1774 $source_schema = $t->schema;
1775 unless ( $source_schema->name ) {
1776 $source_schema->name( $prefilename );
1780 # The "new" style of producers have sane normalization and can support
1781 # diffing a SQL file against a DBIC->SQLT schema. Old style ones don't
1782 # And we have to diff parsed SQL against parsed SQL.
1783 my $dest_schema = $sqlt_schema;
1785 unless ( "SQL::Translator::Producer::$db"->can('preprocess_schema') ) {
1786 my $t = SQL::Translator->new($sqltargs);
1789 $t->parser( $db ) or die $t->error;
1790 my $out = $t->translate( $filename ) or die $t->error;
1791 $dest_schema = $t->schema;
1792 $dest_schema->name( $filename )
1793 unless $dest_schema->name;
1796 my $diff = SQL::Translator::Diff::schema_diff($source_schema, $db,
1800 if(!open $file, ">$difffile") {
1801 $self->throw_exception("Can't write to $difffile ($!)");
1809 =head2 deployment_statements
1813 =item Arguments: $schema, $type, $version, $directory, $sqlt_args
1817 Returns the statements used by L</deploy> and L<DBIx::Class::Schema/deploy>.
1818 The database driver name is given by C<$type>, though the value from
1819 L</sqlt_type> is used if it is not specified.
1821 C<$directory> is used to return statements from files in a previously created
1822 L</create_ddl_dir> directory and is optional. The filenames are constructed
1823 from L<DBIx::Class::Schema/ddl_filename>, the schema name and the C<$version>.
1825 If no C<$directory> is specified then the statements are constructed on the
1826 fly using L<SQL::Translator> and C<$version> is ignored.
1828 See L<SQL::Translator/METHODS> for a list of values for C<$sqlt_args>.
1832 sub deployment_statements {
1833 my ($self, $schema, $type, $version, $dir, $sqltargs) = @_;
1834 # Need to be connected to get the correct sqlt_type
1835 $self->ensure_connected() unless $type;
1836 $type ||= $self->sqlt_type;
1837 $version ||= $schema->schema_version || '1.x';
1839 my $filename = $schema->ddl_filename($type, $version, $dir);
1843 open($file, "<$filename")
1844 or $self->throw_exception("Can't open $filename ($!)");
1847 return join('', @rows);
1850 $self->throw_exception(q{Can't deploy without SQL::Translator 0.09003: '}
1851 . $self->_check_sqlt_message . q{'})
1852 if !$self->_check_sqlt_version;
1854 require SQL::Translator::Parser::DBIx::Class;
1855 eval qq{use SQL::Translator::Producer::${type}};
1856 $self->throw_exception($@) if $@;
1858 # sources needs to be a parser arg, but for simplicty allow at top level
1860 $sqltargs->{parser_args}{sources} = delete $sqltargs->{sources}
1861 if exists $sqltargs->{sources};
1863 my $tr = SQL::Translator->new(%$sqltargs);
1864 SQL::Translator::Parser::DBIx::Class::parse( $tr, $schema );
1865 return "SQL::Translator::Producer::${type}"->can('produce')->($tr);
1869 my ($self, $schema, $type, $sqltargs, $dir) = @_;
1872 return if($line =~ /^--/);
1874 # next if($line =~ /^DROP/m);
1875 return if($line =~ /^BEGIN TRANSACTION/m);
1876 return if($line =~ /^COMMIT/m);
1877 return if $line =~ /^\s+$/; # skip whitespace only
1878 $self->_query_start($line);
1880 $self->dbh->do($line); # shouldn't be using ->dbh ?
1883 warn qq{$@ (running "${line}")};
1885 $self->_query_end($line);
1887 my @statements = $self->deployment_statements($schema, $type, undef, $dir, { no_comments => 1, %{ $sqltargs || {} } } );
1888 if (@statements > 1) {
1889 foreach my $statement (@statements) {
1890 $deploy->( $statement );
1893 elsif (@statements == 1) {
1894 foreach my $line ( split(";\n", $statements[0])) {
1900 =head2 datetime_parser
1902 Returns the datetime parser class
1906 sub datetime_parser {
1908 return $self->{datetime_parser} ||= do {
1909 $self->ensure_connected;
1910 $self->build_datetime_parser(@_);
1914 =head2 datetime_parser_type
1916 Defines (returns) the datetime parser class - currently hardwired to
1917 L<DateTime::Format::MySQL>
1921 sub datetime_parser_type { "DateTime::Format::MySQL"; }
1923 =head2 build_datetime_parser
1925 See L</datetime_parser>
1929 sub build_datetime_parser {
1931 my $type = $self->datetime_parser_type(@_);
1933 $self->throw_exception("Couldn't load ${type}: $@") if $@;
1938 my $_check_sqlt_version; # private
1939 my $_check_sqlt_message; # private
1940 sub _check_sqlt_version {
1941 return $_check_sqlt_version if defined $_check_sqlt_version;
1942 eval 'use SQL::Translator "0.09003"';
1943 $_check_sqlt_message = $@ || '';
1944 $_check_sqlt_version = !$@;
1947 sub _check_sqlt_message {
1948 _check_sqlt_version if !defined $_check_sqlt_message;
1949 $_check_sqlt_message;
1953 =head2 is_replicating
1955 A boolean that reports if a particular L<DBIx::Class::Storage::DBI> is set to
1956 replicate from a master database. Default is undef, which is the result
1957 returned by databases that don't support replication.
1961 sub is_replicating {
1966 =head2 lag_behind_master
1968 Returns a number that represents a certain amount of lag behind a master db
1969 when a given storage is replicating. The number is database dependent, but
1970 starts at zero and increases with the amount of lag. Default in undef
1974 sub lag_behind_master {
1980 return if !$self->_dbh;
1989 =head2 DBIx::Class and AutoCommit
1991 DBIx::Class can do some wonderful magic with handling exceptions,
1992 disconnections, and transactions when you use C<< AutoCommit => 1 >>
1993 combined with C<txn_do> for transaction support.
1995 If you set C<< AutoCommit => 0 >> in your connect info, then you are always
1996 in an assumed transaction between commits, and you're telling us you'd
1997 like to manage that manually. A lot of the magic protections offered by
1998 this module will go away. We can't protect you from exceptions due to database
1999 disconnects because we don't know anything about how to restart your
2000 transactions. You're on your own for handling all sorts of exceptional
2001 cases if you choose the C<< AutoCommit => 0 >> path, just as you would
2007 The module defines a set of methods within the DBIC::SQL::Abstract
2008 namespace. These build on L<SQL::Abstract::Limit> to provide the
2009 SQL query functions.
2011 The following methods are extended:-
2025 See L</connect_info> for details.
2029 See L</connect_info> for details.
2033 See L</connect_info> for details.
2039 Matt S. Trout <mst@shadowcatsystems.co.uk>
2041 Andy Grundman <andy@hybridized.org>
2045 You may distribute this code under the same terms as Perl itself.