X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=blobdiff_plain;f=lib%2FDBIx%2FClass%2FStorage%2FDBI%2FODBC%2FMicrosoft_SQL_Server.pm;h=bf01131c6a9a14590fc6420a63cf71b4687fb3e9;hb=1a58752c42ba9acf33e2943b678de4948cf3ee3f;hp=5c03762ab8a0754bafe72993b7fcb303813ab2fa;hpb=6f96f99f2a205c59cd53628a8d2b19f8a76bd7ab;p=dbsrgits%2FDBIx-Class.git

diff --git a/lib/DBIx/Class/Storage/DBI/ODBC/Microsoft_SQL_Server.pm b/lib/DBIx/Class/Storage/DBI/ODBC/Microsoft_SQL_Server.pm
index 5c03762..bf01131 100644
--- a/lib/DBIx/Class/Storage/DBI/ODBC/Microsoft_SQL_Server.pm
+++ b/lib/DBIx/Class/Storage/DBI/ODBC/Microsoft_SQL_Server.pm
@@ -2,82 +2,188 @@ package DBIx::Class::Storage::DBI::ODBC::Microsoft_SQL_Server;
 use strict;
 use warnings;
 
-use base qw/DBIx::Class::Storage::DBI/;
+use base qw/DBIx::Class::Storage::DBI::MSSQL/;
+use mro 'c3';
 
-sub _prep_for_execute {
-    my $self = shift;
-    my ($op, $extra_bind, $ident, $args) = @_;
+use List::Util();
+use Scalar::Util ();
 
-    my ($sql, $bind) = $self->SUPER::_prep_for_execute(@_);
-    $sql .= ';SELECT SCOPE_IDENTITY()' if $op eq 'insert';
+__PACKAGE__->mk_group_accessors(simple => qw/
+  _using_dynamic_cursors
+/);
 
-    return ($sql, $bind);
-}
+=head1 NAME
+
+DBIx::Class::Storage::DBI::ODBC::Microsoft_SQL_Server - Support specific
+to Microsoft SQL Server over ODBC
+
+=head1 DESCRIPTION
+
+This class implements support specific to Microsoft SQL Server over ODBC.  It is
+loaded automatically by by DBIx::Class::Storage::DBI::ODBC when it detects a
+MSSQL back-end.
+
+Most of the functionality is provided from the superclass
+L<DBIx::Class::Storage::DBI::MSSQL>.
+
+=head1 MULTIPLE ACTIVE STATEMENTS
+
+The following options are alternative ways to enable concurrent executing
+statement support. Each has its own advantages and drawbacks.
+
+=head2 connect_call_use_dynamic_cursors
+
+Use as:
+
+  on_connect_call => 'use_dynamic_cursors'
+
+in your L<DBIx::Class::Storage::DBI/connect_info> as one way to enable multiple
+concurrent statements.
+
+Will add C<< odbc_cursortype => 2 >> to your DBI connection attributes. See
+L<DBD::ODBC/odbc_cursortype> for more information.
+
+Alternatively, you can add it yourself and dynamic cursor support will be
+automatically enabled.
+
+If you're using FreeTDS, C<tds_version> must be set to at least C<8.0>.
+
+This will not work with CODE ref connect_info's.
+
+B<WARNING:> this will break C<SCOPE_IDENTITY()>, and C<SELECT @@IDENTITY> will
+be used instead, which on SQL Server 2005 and later will return erroneous
+results on tables which have an on insert trigger that inserts into another
+table with an C<IDENTITY> column.
+
+=cut
+
+sub connect_call_use_dynamic_cursors {
+  my $self = shift;
+
+  if (ref($self->_dbi_connect_info->[0]) eq 'CODE') {
+    $self->throw_exception ('cannot set DBI attributes on a CODE ref connect_info');
+  }
 
-sub insert {
-    my ($self, $source, $to_insert) = @_;
+  my $dbi_attrs = $self->_dbi_connect_info->[-1];
 
-    my $bind_attributes = $self->source_bind_attributes($source);
-    my (undef, $sth) = $self->_execute( 'insert' => [], $source, $bind_attributes, $to_insert);
-    $self->{_scope_identity} = $sth->fetchrow_array;
+  unless (ref($dbi_attrs) && Scalar::Util::reftype($dbi_attrs) eq 'HASH') {
+    $dbi_attrs = {};
+    push @{ $self->_dbi_connect_info }, $dbi_attrs;
+  }
 
-    return $to_insert;
+  if (not exists $dbi_attrs->{odbc_cursortype}) {
+    # turn on support for multiple concurrent statements, unless overridden
+    $dbi_attrs->{odbc_cursortype} = 2;
+    $self->disconnect; # resetting dbi attrs, so have to reconnect
+    $self->ensure_connected;
+    $self->_set_dynamic_cursors;
+  }
 }
 
-sub last_insert_id { shift->{_scope_identity} }
+sub _set_dynamic_cursors {
+  my $self = shift;
+  my $dbh  = $self->_get_dbh;
+
+  eval {
+    local $dbh->{RaiseError} = 1;
+    local $dbh->{PrintError} = 0;
+    $dbh->do('SELECT @@IDENTITY');
+  };
+  if ($@) {
+    $self->throw_exception (<<'EOF');
+
+Your drivers do not seem to support dynamic cursors (odbc_cursortype => 2),
+if you're using FreeTDS, make sure to set tds_version to 8.0 or greater.
+EOF
+  }
+
+  $self->_using_dynamic_cursors(1);
+  $self->_identity_method('@@identity');
+}
 
-sub sqlt_type { 'SQLServer' }
+sub _init {
+  my $self = shift;
 
-sub _sql_maker_opts {
-    my ( $self, $opts ) = @_;
+  no warnings qw/uninitialized/;
 
-    if ( $opts ) {
-        $self->{_sql_maker_opts} = { %$opts };
-    }
+  if (
+    ref($self->_dbi_connect_info->[0]) ne 'CODE'
+      &&
+    ref ($self->_dbi_connect_info->[-1]) eq 'HASH'
+      &&
+    $self->_dbi_connect_info->[-1]{odbc_cursortype} == 2
+  ) {
+    $self->_set_dynamic_cursors;
+    return;
+  }
 
-    return { limit_dialect => 'Top', %{$self->{_sql_maker_opts}||{}} };
+  $self->_using_dynamic_cursors(0);
 }
 
-1;
+=head2 connect_call_use_server_cursors
 
-__END__
+Use as:
 
-=head1 NAME
+  on_connect_call => 'use_server_cursors'
 
-DBIx::Class::Storage::ODBC::Microsoft_SQL_Server - Support specific to
-Microsoft SQL Server over ODBC
+May allow multiple active select statements. See
+L<DBD::ODBC/odbc_SQL_ROWSET_SIZE> for more information.
 
-=head1 DESCRIPTION
+Takes an optional parameter for the value to set the attribute to, default is
+C<2>.
 
-This class implements support specific to Microsoft SQL Server over ODBC,
-including auto-increment primary keys and SQL::Abstract::Limit dialect.  It
-is loaded automatically by by DBIx::Class::Storage::DBI::ODBC when it
-detects a MSSQL back-end.
+B<WARNING>: this does not work on all versions of SQL Server, and may lock up
+your database!
 
-=head1 IMPLEMENTATION NOTES
+=cut
 
-Microsoft SQL Server supports three methods of retrieving the IDENTITY
-value for inserted row: IDENT_CURRENT, @@IDENTITY, and SCOPE_IDENTITY().
-SCOPE_IDENTITY is used here because it is the safest.  However, it must
-be called is the same execute statement, not just the same connection.
+sub connect_call_use_server_cursors {
+  my $self            = shift;
+  my $sql_rowset_size = shift || 2;
+
+  $self->_get_dbh->{odbc_SQL_ROWSET_SIZE} = $sql_rowset_size;
+}
 
-So, this implementation appends a SELECT SCOPE_IDENTITY() statement
-onto each INSERT to accommodate that requirement.
+=head2 connect_call_use_MARS
 
-=head1 METHODS
+Use as:
 
-=head2 insert
+  on_connect_call => 'use_MARS'
 
-=head2 last_insert_id
+Use to enable a feature of SQL Server 2005 and later, "Multiple Active Result
+Sets". See L<DBD::ODBC::FAQ/Does DBD::ODBC support Multiple Active Statements?>
+for more information.
 
-=head2 sqlt_type
+B<WARNING>: This has implications for the way transactions are handled.
+
+=cut
+
+sub connect_call_use_MARS {
+  my $self = shift;
+
+  my $dsn = $self->_dbi_connect_info->[0];
+
+  if (ref($dsn) eq 'CODE') {
+    $self->throw_exception('cannot change the DBI DSN on a CODE ref connect_info');
+  }
+
+  if ($dsn !~ /MARS_Connection=/) {
+    $self->_dbi_connect_info->[0] = "$dsn;MARS_Connection=Yes";
+    my $was_connected = defined $self->_dbh;
+    $self->disconnect;
+    $self->ensure_connected if $was_connected;
+  }
+}
+
+1;
 
-=head1 AUTHORS
+=head1 AUTHOR
 
-Marc Mims C<< <marc@questright.com> >>
+See L<DBIx::Class/CONTRIBUTORS>.
 
 =head1 LICENSE
 
 You may distribute this code under the same terms as Perl itself.
 
 =cut
+# vim: sw=2 sts=2