Commit | Line | Data |
057db5ce |
1 | package DBIx::Class::Storage::DBI::Sybase::ASE; |
2 | |
3 | use strict; |
4 | use warnings; |
5 | |
6 | use base qw/ |
7 | DBIx::Class::Storage::DBI::Sybase |
8 | DBIx::Class::Storage::DBI::AutoCast |
9 | /; |
10 | use mro 'c3'; |
11 | use Carp::Clan qw/^DBIx::Class/; |
6298a324 |
12 | use Scalar::Util 'blessed'; |
13 | use List::Util 'first'; |
057db5ce |
14 | use Sub::Name(); |
6298a324 |
15 | use Data::Dumper::Concise 'Dumper'; |
ed7ab0f4 |
16 | use Try::Tiny; |
fd323bf1 |
17 | use namespace::clean; |
057db5ce |
18 | |
6a247f33 |
19 | __PACKAGE__->sql_limit_dialect ('RowCountOrGenericSubQ'); |
2b8cc2f2 |
20 | __PACKAGE__->sql_quote_char ([qw/[ ]/]); |
6f7a118e |
21 | __PACKAGE__->datetime_parser_type('DateTime::Format::Sybase'); |
6a247f33 |
22 | |
057db5ce |
23 | __PACKAGE__->mk_group_accessors('simple' => |
24 | qw/_identity _blob_log_on_update _writer_storage _is_extra_storage |
25 | _bulk_storage _is_bulk_storage _began_bulk_work |
26 | _bulk_disabled_due_to_coderef_connect_info_warned |
27 | _identity_method/ |
28 | ); |
29 | |
30 | my @also_proxy_to_extra_storages = qw/ |
31 | connect_call_set_auto_cast auto_cast connect_call_blob_setup |
32 | connect_call_datetime_setup |
33 | |
34 | disconnect _connect_info _sql_maker _sql_maker_opts disable_sth_caching |
35 | auto_savepoint unsafe cursor_class debug debugobj schema |
36 | /; |
37 | |
38 | =head1 NAME |
39 | |
40 | DBIx::Class::Storage::DBI::Sybase::ASE - Sybase ASE SQL Server support for |
41 | DBIx::Class |
42 | |
43 | =head1 SYNOPSIS |
44 | |
45 | This subclass supports L<DBD::Sybase> for real (non-Microsoft) Sybase databases. |
46 | |
47 | =head1 DESCRIPTION |
48 | |
49 | If your version of Sybase does not support placeholders, then your storage will |
50 | be reblessed to L<DBIx::Class::Storage::DBI::Sybase::ASE::NoBindVars>. |
51 | You can also enable that driver explicitly, see the documentation for more |
52 | details. |
53 | |
54 | With this driver there is unfortunately no way to get the C<last_insert_id> |
55 | without doing a C<SELECT MAX(col)>. This is done safely in a transaction |
56 | (locking the table.) See L</INSERTS WITH PLACEHOLDERS>. |
57 | |
8384a713 |
58 | A recommended L<connect_info|DBIx::Class::Storage::DBI/connect_info> setting: |
057db5ce |
59 | |
60 | on_connect_call => [['datetime_setup'], ['blob_setup', log_on_update => 0]] |
61 | |
62 | =head1 METHODS |
63 | |
64 | =cut |
65 | |
66 | sub _rebless { |
67 | my $self = shift; |
68 | |
69 | my $no_bind_vars = __PACKAGE__ . '::NoBindVars'; |
70 | |
71 | if ($self->using_freetds) { |
72 | carp <<'EOF' unless $ENV{DBIC_SYBASE_FREETDS_NOWARN}; |
73 | |
74 | You are using FreeTDS with Sybase. |
75 | |
76 | We will do our best to support this configuration, but please consider this |
77 | support experimental. |
78 | |
79 | TEXT/IMAGE columns will definitely not work. |
80 | |
81 | You are encouraged to recompile DBD::Sybase with the Sybase Open Client libraries |
82 | instead. |
83 | |
84 | See perldoc DBIx::Class::Storage::DBI::Sybase::ASE for more details. |
85 | |
86 | To turn off this warning set the DBIC_SYBASE_FREETDS_NOWARN environment |
87 | variable. |
88 | EOF |
89 | |
bbdda281 |
90 | if (not $self->_use_typeless_placeholders) { |
91 | if ($self->_use_placeholders) { |
057db5ce |
92 | $self->auto_cast(1); |
93 | } |
94 | else { |
95 | $self->ensure_class_loaded($no_bind_vars); |
96 | bless $self, $no_bind_vars; |
97 | $self->_rebless; |
98 | } |
99 | } |
100 | } |
101 | |
102 | elsif (not $self->_get_dbh->{syb_dynamic_supported}) { |
103 | # not necessarily FreeTDS, but no placeholders nevertheless |
104 | $self->ensure_class_loaded($no_bind_vars); |
105 | bless $self, $no_bind_vars; |
106 | $self->_rebless; |
107 | } |
108 | # this is highly unlikely, but we check just in case |
bbdda281 |
109 | elsif (not $self->_use_typeless_placeholders) { |
057db5ce |
110 | $self->auto_cast(1); |
111 | } |
112 | } |
113 | |
114 | sub _init { |
115 | my $self = shift; |
116 | $self->_set_max_connect(256); |
117 | |
118 | # create storage for insert/(update blob) transactions, |
119 | # unless this is that storage |
120 | return if $self->_is_extra_storage; |
121 | |
122 | my $writer_storage = (ref $self)->new; |
123 | |
124 | $writer_storage->_is_extra_storage(1); |
125 | $writer_storage->connect_info($self->connect_info); |
126 | $writer_storage->auto_cast($self->auto_cast); |
127 | |
128 | $self->_writer_storage($writer_storage); |
129 | |
130 | # create a bulk storage unless connect_info is a coderef |
131 | return if ref($self->_dbi_connect_info->[0]) eq 'CODE'; |
132 | |
133 | my $bulk_storage = (ref $self)->new; |
134 | |
135 | $bulk_storage->_is_extra_storage(1); |
136 | $bulk_storage->_is_bulk_storage(1); # for special ->disconnect acrobatics |
137 | $bulk_storage->connect_info($self->connect_info); |
138 | |
139 | # this is why |
140 | $bulk_storage->_dbi_connect_info->[0] .= ';bulkLogin=1'; |
141 | |
142 | $self->_bulk_storage($bulk_storage); |
143 | } |
144 | |
145 | for my $method (@also_proxy_to_extra_storages) { |
146 | no strict 'refs'; |
147 | no warnings 'redefine'; |
148 | |
149 | my $replaced = __PACKAGE__->can($method); |
150 | |
151 | *{$method} = Sub::Name::subname $method => sub { |
152 | my $self = shift; |
153 | $self->_writer_storage->$replaced(@_) if $self->_writer_storage; |
154 | $self->_bulk_storage->$replaced(@_) if $self->_bulk_storage; |
155 | return $self->$replaced(@_); |
156 | }; |
157 | } |
158 | |
159 | sub disconnect { |
160 | my $self = shift; |
161 | |
162 | # Even though we call $sth->finish for uses off the bulk API, there's still an |
163 | # "active statement" warning on disconnect, which we throw away here. |
164 | # This is due to the bug described in insert_bulk. |
165 | # Currently a noop because 'prepare' is used instead of 'prepare_cached'. |
166 | local $SIG{__WARN__} = sub { |
167 | warn $_[0] unless $_[0] =~ /active statement/i; |
168 | } if $self->_is_bulk_storage; |
169 | |
170 | # so that next transaction gets a dbh |
171 | $self->_began_bulk_work(0) if $self->_is_bulk_storage; |
172 | |
173 | $self->next::method; |
174 | } |
175 | |
176 | # Set up session settings for Sybase databases for the connection. |
177 | # |
178 | # Make sure we have CHAINED mode turned on if AutoCommit is off in non-FreeTDS |
179 | # DBD::Sybase (since we don't know how DBD::Sybase was compiled.) If however |
180 | # we're using FreeTDS, CHAINED mode turns on an implicit transaction which we |
181 | # only want when AutoCommit is off. |
182 | # |
183 | # Also SET TEXTSIZE for FreeTDS because LongReadLen doesn't work. |
184 | sub _run_connection_actions { |
185 | my $self = shift; |
186 | |
187 | if ($self->_is_bulk_storage) { |
188 | # this should be cleared on every reconnect |
189 | $self->_began_bulk_work(0); |
190 | return; |
191 | } |
192 | |
193 | if (not $self->using_freetds) { |
194 | $self->_dbh->{syb_chained_txn} = 1; |
195 | } else { |
196 | # based on LongReadLen in connect_info |
197 | $self->set_textsize; |
198 | |
199 | if ($self->_dbh_autocommit) { |
200 | $self->_dbh->do('SET CHAINED OFF'); |
201 | } else { |
202 | $self->_dbh->do('SET CHAINED ON'); |
203 | } |
204 | } |
205 | |
206 | $self->next::method(@_); |
207 | } |
208 | |
209 | =head2 connect_call_blob_setup |
210 | |
211 | Used as: |
212 | |
213 | on_connect_call => [ [ 'blob_setup', log_on_update => 0 ] ] |
214 | |
215 | Does C<< $dbh->{syb_binary_images} = 1; >> to return C<IMAGE> data as raw binary |
216 | instead of as a hex string. |
217 | |
218 | Recommended. |
219 | |
220 | Also sets the C<log_on_update> value for blob write operations. The default is |
221 | C<1>, but C<0> is better if your database is configured for it. |
222 | |
223 | See |
224 | L<DBD::Sybase/Handling_IMAGE/TEXT_data_with_syb_ct_get_data()/syb_ct_send_data()>. |
225 | |
226 | =cut |
227 | |
228 | sub connect_call_blob_setup { |
229 | my $self = shift; |
230 | my %args = @_; |
231 | my $dbh = $self->_dbh; |
232 | $dbh->{syb_binary_images} = 1; |
233 | |
234 | $self->_blob_log_on_update($args{log_on_update}) |
235 | if exists $args{log_on_update}; |
236 | } |
237 | |
057db5ce |
238 | sub _is_lob_column { |
239 | my ($self, $source, $column) = @_; |
240 | |
241 | return $self->_is_lob_type($source->column_info($column)->{data_type}); |
242 | } |
243 | |
244 | sub _prep_for_execute { |
245 | my $self = shift; |
246 | my ($op, $extra_bind, $ident, $args) = @_; |
247 | |
248 | my ($sql, $bind) = $self->next::method (@_); |
249 | |
6298a324 |
250 | my $table = blessed $ident ? $ident->from : $ident; |
057db5ce |
251 | |
252 | my $bind_info = $self->_resolve_column_info( |
253 | $ident, [map $_->[0], @{$bind}] |
254 | ); |
6298a324 |
255 | my $bound_identity_col = |
256 | first { $bind_info->{$_}{is_auto_increment} } |
257 | keys %$bind_info |
057db5ce |
258 | ; |
e366f807 |
259 | |
260 | my $columns_info = blessed $ident && $ident->columns_info; |
261 | |
6298a324 |
262 | my $identity_col = |
e366f807 |
263 | $columns_info && |
264 | first { $columns_info->{$_}{is_auto_increment} } |
265 | keys %$columns_info |
057db5ce |
266 | ; |
267 | |
268 | if (($op eq 'insert' && $bound_identity_col) || |
269 | ($op eq 'update' && exists $args->[0]{$identity_col})) { |
270 | $sql = join ("\n", |
271 | $self->_set_table_identity_sql($op => $table, 'on'), |
272 | $sql, |
273 | $self->_set_table_identity_sql($op => $table, 'off'), |
274 | ); |
275 | } |
276 | |
277 | if ($op eq 'insert' && (not $bound_identity_col) && $identity_col && |
278 | (not $self->{insert_bulk})) { |
279 | $sql = |
280 | "$sql\n" . |
281 | $self->_fetch_identity_sql($ident, $identity_col); |
282 | } |
283 | |
284 | return ($sql, $bind); |
285 | } |
286 | |
287 | sub _set_table_identity_sql { |
288 | my ($self, $op, $table, $on_off) = @_; |
289 | |
290 | return sprintf 'SET IDENTITY_%s %s %s', |
291 | uc($op), $self->sql_maker->_quote($table), uc($on_off); |
292 | } |
293 | |
294 | # Stolen from SQLT, with some modifications. This is a makeshift |
295 | # solution before a sane type-mapping library is available, thus |
296 | # the 'our' for easy overrides. |
297 | our %TYPE_MAPPING = ( |
298 | number => 'numeric', |
299 | money => 'money', |
300 | varchar => 'varchar', |
301 | varchar2 => 'varchar', |
302 | timestamp => 'datetime', |
303 | text => 'varchar', |
304 | real => 'double precision', |
305 | comment => 'text', |
306 | bit => 'bit', |
307 | tinyint => 'smallint', |
308 | float => 'double precision', |
309 | serial => 'numeric', |
310 | bigserial => 'numeric', |
311 | boolean => 'varchar', |
312 | long => 'varchar', |
313 | ); |
314 | |
315 | sub _native_data_type { |
316 | my ($self, $type) = @_; |
317 | |
318 | $type = lc $type; |
319 | $type =~ s/\s* identity//x; |
320 | |
321 | return uc($TYPE_MAPPING{$type} || $type); |
322 | } |
323 | |
324 | sub _fetch_identity_sql { |
325 | my ($self, $source, $col) = @_; |
326 | |
327 | return sprintf ("SELECT MAX(%s) FROM %s", |
328 | map { $self->sql_maker->_quote ($_) } ($col, $source->from) |
329 | ); |
330 | } |
331 | |
332 | sub _execute { |
333 | my $self = shift; |
334 | my ($op) = @_; |
335 | |
336 | my ($rv, $sth, @bind) = $self->dbh_do($self->can('_dbh_execute'), @_); |
337 | |
338 | if ($op eq 'insert') { |
339 | $self->_identity($sth->fetchrow_array); |
340 | $sth->finish; |
341 | } |
342 | |
343 | return wantarray ? ($rv, $sth, @bind) : $rv; |
344 | } |
345 | |
346 | sub last_insert_id { shift->_identity } |
347 | |
348 | # handles TEXT/IMAGE and transaction for last_insert_id |
349 | sub insert { |
350 | my $self = shift; |
351 | my ($source, $to_insert) = @_; |
352 | |
e366f807 |
353 | my $columns_info = $source->columns_info; |
354 | |
6298a324 |
355 | my $identity_col = |
e366f807 |
356 | (first { $columns_info->{$_}{is_auto_increment} } |
357 | keys %$columns_info ) |
6298a324 |
358 | || ''; |
057db5ce |
359 | |
360 | # check for empty insert |
361 | # INSERT INTO foo DEFAULT VALUES -- does not work with Sybase |
6469dabf |
362 | # try to insert explicit 'DEFAULT's instead (except for identity, timestamp |
363 | # and computed columns) |
057db5ce |
364 | if (not %$to_insert) { |
365 | for my $col ($source->columns) { |
366 | next if $col eq $identity_col; |
6469dabf |
367 | |
368 | my $info = $source->column_info($col); |
369 | |
370 | next if ref $info->{default_value} eq 'SCALAR' |
371 | || (exists $info->{data_type} && (not defined $info->{data_type})); |
372 | |
373 | next if $info->{data_type} && $info->{data_type} =~ /^timestamp\z/i; |
374 | |
057db5ce |
375 | $to_insert->{$col} = \'DEFAULT'; |
376 | } |
377 | } |
378 | |
379 | my $blob_cols = $self->_remove_blob_cols($source, $to_insert); |
380 | |
381 | # do we need the horrific SELECT MAX(COL) hack? |
382 | my $dumb_last_insert_id = |
383 | $identity_col |
384 | && (not exists $to_insert->{$identity_col}) |
385 | && ($self->_identity_method||'') ne '@@IDENTITY'; |
386 | |
387 | my $next = $self->next::can; |
388 | |
389 | # we are already in a transaction, or there are no blobs |
390 | # and we don't need the PK - just (try to) do it |
391 | if ($self->{transaction_depth} |
392 | || (!$blob_cols && !$dumb_last_insert_id) |
393 | ) { |
394 | return $self->_insert ( |
395 | $next, $source, $to_insert, $blob_cols, $identity_col |
396 | ); |
397 | } |
398 | |
399 | # otherwise use the _writer_storage to do the insert+transaction on another |
400 | # connection |
401 | my $guard = $self->_writer_storage->txn_scope_guard; |
402 | |
403 | my $updated_cols = $self->_writer_storage->_insert ( |
404 | $next, $source, $to_insert, $blob_cols, $identity_col |
405 | ); |
406 | |
407 | $self->_identity($self->_writer_storage->_identity); |
408 | |
409 | $guard->commit; |
410 | |
411 | return $updated_cols; |
412 | } |
413 | |
414 | sub _insert { |
415 | my ($self, $next, $source, $to_insert, $blob_cols, $identity_col) = @_; |
416 | |
417 | my $updated_cols = $self->$next ($source, $to_insert); |
418 | |
419 | my $final_row = { |
420 | ($identity_col ? |
421 | ($identity_col => $self->last_insert_id($source, $identity_col)) : ()), |
422 | %$to_insert, |
423 | %$updated_cols, |
424 | }; |
425 | |
426 | $self->_insert_blobs ($source, $blob_cols, $final_row) if $blob_cols; |
427 | |
428 | return $updated_cols; |
429 | } |
430 | |
431 | sub update { |
432 | my $self = shift; |
433 | my ($source, $fields, $where, @rest) = @_; |
434 | |
057db5ce |
435 | my $blob_cols = $self->_remove_blob_cols($source, $fields); |
436 | |
437 | my $table = $source->name; |
438 | |
e366f807 |
439 | my $columns_info = $source->columns_info; |
440 | |
6298a324 |
441 | my $identity_col = |
e366f807 |
442 | first { $columns_info->{$_}{is_auto_increment} } |
443 | keys %$columns_info; |
057db5ce |
444 | |
445 | my $is_identity_update = $identity_col && defined $fields->{$identity_col}; |
446 | |
447 | return $self->next::method(@_) unless $blob_cols; |
448 | |
449 | # If there are any blobs in $where, Sybase will return a descriptive error |
450 | # message. |
451 | # XXX blobs can still be used with a LIKE query, and this should be handled. |
452 | |
453 | # update+blob update(s) done atomically on separate connection |
454 | $self = $self->_writer_storage; |
455 | |
456 | my $guard = $self->txn_scope_guard; |
457 | |
458 | # First update the blob columns to be updated to '' (taken from $fields, where |
459 | # it is originally put by _remove_blob_cols .) |
460 | my %blobs_to_empty = map { ($_ => delete $fields->{$_}) } keys %$blob_cols; |
461 | |
462 | # We can't only update NULL blobs, because blobs cannot be in the WHERE clause. |
463 | |
464 | $self->next::method($source, \%blobs_to_empty, $where, @rest); |
465 | |
466 | # Now update the blobs before the other columns in case the update of other |
467 | # columns makes the search condition invalid. |
468 | $self->_update_blobs($source, $blob_cols, $where); |
469 | |
470 | my @res; |
471 | if (%$fields) { |
cca282b6 |
472 | if (wantarray) { |
057db5ce |
473 | @res = $self->next::method(@_); |
474 | } |
cca282b6 |
475 | elsif (defined wantarray) { |
057db5ce |
476 | $res[0] = $self->next::method(@_); |
477 | } |
478 | else { |
479 | $self->next::method(@_); |
480 | } |
481 | } |
482 | |
483 | $guard->commit; |
484 | |
cca282b6 |
485 | return wantarray ? @res : $res[0]; |
057db5ce |
486 | } |
487 | |
488 | sub insert_bulk { |
489 | my $self = shift; |
490 | my ($source, $cols, $data) = @_; |
491 | |
e366f807 |
492 | my $columns_info = $source->columns_info; |
493 | |
6298a324 |
494 | my $identity_col = |
e366f807 |
495 | first { $columns_info->{$_}{is_auto_increment} } |
496 | keys %$columns_info; |
057db5ce |
497 | |
6298a324 |
498 | my $is_identity_insert = (first { $_ eq $identity_col } @{$cols}) ? 1 : 0; |
057db5ce |
499 | |
500 | my @source_columns = $source->columns; |
501 | |
502 | my $use_bulk_api = |
503 | $self->_bulk_storage && |
504 | $self->_get_dbh->{syb_has_blk}; |
505 | |
506 | if ((not $use_bulk_api) |
507 | && |
508 | (ref($self->_dbi_connect_info->[0]) eq 'CODE') |
509 | && |
510 | (not $self->_bulk_disabled_due_to_coderef_connect_info_warned)) { |
511 | carp <<'EOF'; |
512 | Bulk API support disabled due to use of a CODEREF connect_info. Reverting to |
513 | regular array inserts. |
514 | EOF |
515 | $self->_bulk_disabled_due_to_coderef_connect_info_warned(1); |
516 | } |
517 | |
518 | if (not $use_bulk_api) { |
519 | my $blob_cols = $self->_remove_blob_cols_array($source, $cols, $data); |
520 | |
521 | # _execute_array uses a txn anyway, but it ends too early in case we need to |
522 | # select max(col) to get the identity for inserting blobs. |
523 | ($self, my $guard) = $self->{transaction_depth} == 0 ? |
524 | ($self->_writer_storage, $self->_writer_storage->txn_scope_guard) |
525 | : |
526 | ($self, undef); |
527 | |
528 | local $self->{insert_bulk} = 1; |
529 | |
530 | $self->next::method(@_); |
531 | |
532 | if ($blob_cols) { |
533 | if ($is_identity_insert) { |
534 | $self->_insert_blobs_array ($source, $blob_cols, $cols, $data); |
535 | } |
536 | else { |
537 | my @cols_with_identities = (@$cols, $identity_col); |
538 | |
539 | ## calculate identities |
540 | # XXX This assumes identities always increase by 1, which may or may not |
541 | # be true. |
542 | my ($last_identity) = |
543 | $self->_dbh->selectrow_array ( |
544 | $self->_fetch_identity_sql($source, $identity_col) |
545 | ); |
546 | my @identities = (($last_identity - @$data + 1) .. $last_identity); |
547 | |
548 | my @data_with_identities = map [@$_, shift @identities], @$data; |
549 | |
550 | $self->_insert_blobs_array ( |
551 | $source, $blob_cols, \@cols_with_identities, \@data_with_identities |
552 | ); |
553 | } |
554 | } |
555 | |
556 | $guard->commit if $guard; |
557 | |
558 | return; |
559 | } |
560 | |
561 | # otherwise, use the bulk API |
562 | |
563 | # rearrange @$data so that columns are in database order |
564 | my %orig_idx; |
565 | @orig_idx{@$cols} = 0..$#$cols; |
566 | |
567 | my %new_idx; |
568 | @new_idx{@source_columns} = 0..$#source_columns; |
569 | |
570 | my @new_data; |
571 | for my $datum (@$data) { |
572 | my $new_datum = []; |
573 | for my $col (@source_columns) { |
574 | # identity data will be 'undef' if not $is_identity_insert |
575 | # columns with defaults will also be 'undef' |
576 | $new_datum->[ $new_idx{$col} ] = |
577 | exists $orig_idx{$col} ? $datum->[ $orig_idx{$col} ] : undef; |
578 | } |
579 | push @new_data, $new_datum; |
580 | } |
581 | |
582 | # bcp identity index is 1-based |
583 | my $identity_idx = exists $new_idx{$identity_col} ? |
584 | $new_idx{$identity_col} + 1 : 0; |
585 | |
586 | ## Set a client-side conversion error handler, straight from DBD::Sybase docs. |
587 | # This ignores any data conversion errors detected by the client side libs, as |
588 | # they are usually harmless. |
589 | my $orig_cslib_cb = DBD::Sybase::set_cslib_cb( |
590 | Sub::Name::subname insert_bulk => sub { |
591 | my ($layer, $origin, $severity, $errno, $errmsg, $osmsg, $blkmsg) = @_; |
592 | |
593 | return 1 if $errno == 36; |
594 | |
595 | carp |
596 | "Layer: $layer, Origin: $origin, Severity: $severity, Error: $errno" . |
597 | ($errmsg ? "\n$errmsg" : '') . |
598 | ($osmsg ? "\n$osmsg" : '') . |
599 | ($blkmsg ? "\n$blkmsg" : ''); |
600 | |
601 | return 0; |
602 | }); |
603 | |
4edfce2f |
604 | my $exception = ''; |
ed7ab0f4 |
605 | try { |
057db5ce |
606 | my $bulk = $self->_bulk_storage; |
607 | |
608 | my $guard = $bulk->txn_scope_guard; |
609 | |
610 | ## XXX get this to work instead of our own $sth |
611 | ## will require SQLA or *Hacks changes for ordered columns |
612 | # $bulk->next::method($source, \@source_columns, \@new_data, { |
613 | # syb_bcp_attribs => { |
614 | # identity_flag => $is_identity_insert, |
615 | # identity_column => $identity_idx, |
616 | # } |
617 | # }); |
618 | my $sql = 'INSERT INTO ' . |
619 | $bulk->sql_maker->_quote($source->name) . ' (' . |
620 | # colname list is ignored for BCP, but does no harm |
621 | (join ', ', map $bulk->sql_maker->_quote($_), @source_columns) . ') '. |
622 | ' VALUES ('. (join ', ', ('?') x @source_columns) . ')'; |
623 | |
624 | ## XXX there's a bug in the DBD::Sybase bulk support that makes $sth->finish for |
625 | ## a prepare_cached statement ineffective. Replace with ->sth when fixed, or |
626 | ## better yet the version above. Should be fixed in DBD::Sybase . |
627 | my $sth = $bulk->_get_dbh->prepare($sql, |
628 | # 'insert', # op |
629 | { |
630 | syb_bcp_attribs => { |
631 | identity_flag => $is_identity_insert, |
632 | identity_column => $identity_idx, |
633 | } |
634 | } |
635 | ); |
636 | |
637 | my @bind = do { |
638 | my $idx = 0; |
639 | map [ $_, $idx++ ], @source_columns; |
640 | }; |
641 | |
642 | $self->_execute_array( |
643 | $source, $sth, \@bind, \@source_columns, \@new_data, sub { |
644 | $guard->commit |
645 | } |
646 | ); |
647 | |
648 | $bulk->_query_end($sql); |
ed7ab0f4 |
649 | } catch { |
650 | $exception = shift; |
057db5ce |
651 | }; |
652 | |
057db5ce |
653 | DBD::Sybase::set_cslib_cb($orig_cslib_cb); |
654 | |
655 | if ($exception =~ /-Y option/) { |
f32e99f9 |
656 | my $w = 'Sybase bulk API operation failed due to character set incompatibility, ' |
657 | . 'reverting to regular array inserts. Try unsetting the LANG environment variable' |
658 | ; |
659 | $w .= "\n$exception" if $self->debug; |
660 | carp $w; |
057db5ce |
661 | |
057db5ce |
662 | $self->_bulk_storage(undef); |
663 | unshift @_, $self; |
664 | goto \&insert_bulk; |
665 | } |
666 | elsif ($exception) { |
667 | # rollback makes the bulkLogin connection unusable |
668 | $self->_bulk_storage->disconnect; |
669 | $self->throw_exception($exception); |
670 | } |
671 | } |
672 | |
673 | sub _dbh_execute_array { |
674 | my ($self, $sth, $tuple_status, $cb) = @_; |
675 | |
676 | my $rv = $self->next::method($sth, $tuple_status); |
677 | $cb->() if $cb; |
678 | |
679 | return $rv; |
680 | } |
681 | |
682 | # Make sure blobs are not bound as placeholders, and return any non-empty ones |
683 | # as a hash. |
684 | sub _remove_blob_cols { |
685 | my ($self, $source, $fields) = @_; |
686 | |
687 | my %blob_cols; |
688 | |
689 | for my $col (keys %$fields) { |
690 | if ($self->_is_lob_column($source, $col)) { |
691 | my $blob_val = delete $fields->{$col}; |
692 | if (not defined $blob_val) { |
693 | $fields->{$col} = \'NULL'; |
694 | } |
695 | else { |
696 | $fields->{$col} = \"''"; |
697 | $blob_cols{$col} = $blob_val unless $blob_val eq ''; |
698 | } |
699 | } |
700 | } |
701 | |
702 | return %blob_cols ? \%blob_cols : undef; |
703 | } |
704 | |
705 | # same for insert_bulk |
706 | sub _remove_blob_cols_array { |
707 | my ($self, $source, $cols, $data) = @_; |
708 | |
709 | my @blob_cols; |
710 | |
711 | for my $i (0..$#$cols) { |
712 | my $col = $cols->[$i]; |
713 | |
714 | if ($self->_is_lob_column($source, $col)) { |
715 | for my $j (0..$#$data) { |
716 | my $blob_val = delete $data->[$j][$i]; |
717 | if (not defined $blob_val) { |
718 | $data->[$j][$i] = \'NULL'; |
719 | } |
720 | else { |
721 | $data->[$j][$i] = \"''"; |
722 | $blob_cols[$j][$i] = $blob_val |
723 | unless $blob_val eq ''; |
724 | } |
725 | } |
726 | } |
727 | } |
728 | |
729 | return @blob_cols ? \@blob_cols : undef; |
730 | } |
731 | |
732 | sub _update_blobs { |
733 | my ($self, $source, $blob_cols, $where) = @_; |
734 | |
9780718f |
735 | my @primary_cols = try |
736 | { $source->_pri_cols } |
737 | catch { |
738 | $self->throw_exception("Cannot update TEXT/IMAGE column(s): $_") |
739 | }; |
057db5ce |
740 | |
741 | # check if we're updating a single row by PK |
742 | my $pk_cols_in_where = 0; |
743 | for my $col (@primary_cols) { |
744 | $pk_cols_in_where++ if defined $where->{$col}; |
745 | } |
746 | my @rows; |
747 | |
748 | if ($pk_cols_in_where == @primary_cols) { |
749 | my %row_to_update; |
750 | @row_to_update{@primary_cols} = @{$where}{@primary_cols}; |
751 | @rows = \%row_to_update; |
752 | } else { |
753 | my $cursor = $self->select ($source, \@primary_cols, $where, {}); |
754 | @rows = map { |
755 | my %row; @row{@primary_cols} = @$_; \%row |
756 | } $cursor->all; |
757 | } |
758 | |
759 | for my $row (@rows) { |
760 | $self->_insert_blobs($source, $blob_cols, $row); |
761 | } |
762 | } |
763 | |
764 | sub _insert_blobs { |
765 | my ($self, $source, $blob_cols, $row) = @_; |
766 | my $dbh = $self->_get_dbh; |
767 | |
768 | my $table = $source->name; |
769 | |
770 | my %row = %$row; |
9780718f |
771 | my @primary_cols = try |
772 | { $source->_pri_cols } |
773 | catch { |
774 | $self->throw_exception("Cannot update TEXT/IMAGE column(s): $_") |
775 | }; |
057db5ce |
776 | |
777 | $self->throw_exception('Cannot update TEXT/IMAGE column(s) without primary key values') |
778 | if ((grep { defined $row{$_} } @primary_cols) != @primary_cols); |
779 | |
780 | for my $col (keys %$blob_cols) { |
781 | my $blob = $blob_cols->{$col}; |
782 | |
783 | my %where = map { ($_, $row{$_}) } @primary_cols; |
784 | |
785 | my $cursor = $self->select ($source, [$col], \%where, {}); |
786 | $cursor->next; |
787 | my $sth = $cursor->sth; |
788 | |
789 | if (not $sth) { |
057db5ce |
790 | $self->throw_exception( |
791 | "Could not find row in table '$table' for blob update:\n" |
6298a324 |
792 | . (Dumper \%where) |
057db5ce |
793 | ); |
794 | } |
795 | |
9780718f |
796 | try { |
057db5ce |
797 | do { |
798 | $sth->func('CS_GET', 1, 'ct_data_info') or die $sth->errstr; |
799 | } while $sth->fetch; |
800 | |
801 | $sth->func('ct_prepare_send') or die $sth->errstr; |
802 | |
803 | my $log_on_update = $self->_blob_log_on_update; |
804 | $log_on_update = 1 if not defined $log_on_update; |
805 | |
806 | $sth->func('CS_SET', 1, { |
807 | total_txtlen => length($blob), |
808 | log_on_update => $log_on_update |
809 | }, 'ct_data_info') or die $sth->errstr; |
810 | |
811 | $sth->func($blob, length($blob), 'ct_send_data') or die $sth->errstr; |
812 | |
813 | $sth->func('ct_finish_send') or die $sth->errstr; |
9780718f |
814 | } |
815 | catch { |
057db5ce |
816 | if ($self->using_freetds) { |
817 | $self->throw_exception ( |
9780718f |
818 | "TEXT/IMAGE operation failed, probably because you are using FreeTDS: $_" |
057db5ce |
819 | ); |
9780718f |
820 | } |
821 | else { |
822 | $self->throw_exception($_); |
057db5ce |
823 | } |
824 | } |
9780718f |
825 | finally { |
826 | $sth->finish if $sth; |
827 | }; |
057db5ce |
828 | } |
829 | } |
830 | |
831 | sub _insert_blobs_array { |
832 | my ($self, $source, $blob_cols, $cols, $data) = @_; |
833 | |
834 | for my $i (0..$#$data) { |
835 | my $datum = $data->[$i]; |
836 | |
837 | my %row; |
838 | @row{ @$cols } = @$datum; |
839 | |
840 | my %blob_vals; |
841 | for my $j (0..$#$cols) { |
842 | if (exists $blob_cols->[$i][$j]) { |
843 | $blob_vals{ $cols->[$j] } = $blob_cols->[$i][$j]; |
844 | } |
845 | } |
846 | |
847 | $self->_insert_blobs ($source, \%blob_vals, \%row); |
848 | } |
849 | } |
850 | |
851 | =head2 connect_call_datetime_setup |
852 | |
853 | Used as: |
854 | |
855 | on_connect_call => 'datetime_setup' |
856 | |
8384a713 |
857 | In L<connect_info|DBIx::Class::Storage::DBI/connect_info> to set: |
057db5ce |
858 | |
859 | $dbh->syb_date_fmt('ISO_strict'); # output fmt: 2004-08-21T14:36:48.080Z |
860 | $dbh->do('set dateformat mdy'); # input fmt: 08/13/1979 18:08:55.080 |
861 | |
862 | On connection for use with L<DBIx::Class::InflateColumn::DateTime>, using |
863 | L<DateTime::Format::Sybase>, which you will need to install. |
864 | |
865 | This works for both C<DATETIME> and C<SMALLDATETIME> columns, although |
866 | C<SMALLDATETIME> columns only have minute precision. |
867 | |
868 | =cut |
869 | |
870 | { |
871 | my $old_dbd_warned = 0; |
872 | |
873 | sub connect_call_datetime_setup { |
874 | my $self = shift; |
875 | my $dbh = $self->_get_dbh; |
876 | |
877 | if ($dbh->can('syb_date_fmt')) { |
878 | # amazingly, this works with FreeTDS |
879 | $dbh->syb_date_fmt('ISO_strict'); |
880 | } elsif (not $old_dbd_warned) { |
881 | carp "Your DBD::Sybase is too old to support ". |
882 | "DBIx::Class::InflateColumn::DateTime, please upgrade!"; |
883 | $old_dbd_warned = 1; |
884 | } |
885 | |
886 | $dbh->do('SET DATEFORMAT mdy'); |
887 | |
888 | 1; |
889 | } |
890 | } |
891 | |
057db5ce |
892 | |
893 | # ->begin_work and such have no effect with FreeTDS but we run them anyway to |
894 | # let the DBD keep any state it needs to. |
895 | # |
896 | # If they ever do start working, the extra statements will do no harm (because |
897 | # Sybase supports nested transactions.) |
898 | |
899 | sub _dbh_begin_work { |
900 | my $self = shift; |
901 | |
902 | # bulkLogin=1 connections are always in a transaction, and can only call BEGIN |
903 | # TRAN once. However, we need to make sure there's a $dbh. |
904 | return if $self->_is_bulk_storage && $self->_dbh && $self->_began_bulk_work; |
905 | |
906 | $self->next::method(@_); |
907 | |
908 | if ($self->using_freetds) { |
909 | $self->_get_dbh->do('BEGIN TRAN'); |
910 | } |
911 | |
912 | $self->_began_bulk_work(1) if $self->_is_bulk_storage; |
913 | } |
914 | |
915 | sub _dbh_commit { |
916 | my $self = shift; |
917 | if ($self->using_freetds) { |
918 | $self->_dbh->do('COMMIT'); |
919 | } |
920 | return $self->next::method(@_); |
921 | } |
922 | |
923 | sub _dbh_rollback { |
924 | my $self = shift; |
925 | if ($self->using_freetds) { |
926 | $self->_dbh->do('ROLLBACK'); |
927 | } |
928 | return $self->next::method(@_); |
929 | } |
930 | |
931 | # savepoint support using ASE syntax |
932 | |
933 | sub _svp_begin { |
934 | my ($self, $name) = @_; |
935 | |
936 | $self->_get_dbh->do("SAVE TRANSACTION $name"); |
937 | } |
938 | |
939 | # A new SAVE TRANSACTION with the same name releases the previous one. |
940 | sub _svp_release { 1 } |
941 | |
942 | sub _svp_rollback { |
943 | my ($self, $name) = @_; |
944 | |
945 | $self->_get_dbh->do("ROLLBACK TRANSACTION $name"); |
946 | } |
947 | |
948 | 1; |
949 | |
950 | =head1 Schema::Loader Support |
951 | |
290da7d6 |
952 | As of version C<0.05000>, L<DBIx::Class::Schema::Loader> should work well with |
953 | most (if not all) versions of Sybase ASE. |
057db5ce |
954 | |
955 | =head1 FreeTDS |
956 | |
957 | This driver supports L<DBD::Sybase> compiled against FreeTDS |
958 | (L<http://www.freetds.org/>) to the best of our ability, however it is |
959 | recommended that you recompile L<DBD::Sybase> against the Sybase Open Client |
960 | libraries. They are a part of the Sybase ASE distribution: |
961 | |
962 | The Open Client FAQ is here: |
963 | L<http://www.isug.com/Sybase_FAQ/ASE/section7.html>. |
964 | |
965 | Sybase ASE for Linux (which comes with the Open Client libraries) may be |
966 | downloaded here: L<http://response.sybase.com/forms/ASE_Linux_Download>. |
967 | |
968 | To see if you're using FreeTDS check C<< $schema->storage->using_freetds >>, or run: |
969 | |
970 | perl -MDBI -le 'my $dbh = DBI->connect($dsn, $user, $pass); print $dbh->{syb_oc_version}' |
971 | |
972 | Some versions of the libraries involved will not support placeholders, in which |
973 | case the storage will be reblessed to |
974 | L<DBIx::Class::Storage::DBI::Sybase::ASE::NoBindVars>. |
975 | |
976 | In some configurations, placeholders will work but will throw implicit type |
977 | conversion errors for anything that's not expecting a string. In such a case, |
978 | the C<auto_cast> option from L<DBIx::Class::Storage::DBI::AutoCast> is |
979 | automatically set, which you may enable on connection with |
980 | L<DBIx::Class::Storage::DBI::AutoCast/connect_call_set_auto_cast>. The type info |
981 | for the C<CAST>s is taken from the L<DBIx::Class::ResultSource/data_type> |
982 | definitions in your Result classes, and are mapped to a Sybase type (if it isn't |
983 | already) using a mapping based on L<SQL::Translator>. |
984 | |
48580715 |
985 | In other configurations, placeholders will work just as they do with the Sybase |
057db5ce |
986 | Open Client libraries. |
987 | |
988 | Inserts or updates of TEXT/IMAGE columns will B<NOT> work with FreeTDS. |
989 | |
990 | =head1 INSERTS WITH PLACEHOLDERS |
991 | |
992 | With placeholders enabled, inserts are done in a transaction so that there are |
993 | no concurrency issues with getting the inserted identity value using |
994 | C<SELECT MAX(col)>, which is the only way to get the C<IDENTITY> value in this |
995 | mode. |
996 | |
997 | In addition, they are done on a separate connection so that it's possible to |
998 | have active cursors when doing an insert. |
999 | |
1000 | When using C<DBIx::Class::Storage::DBI::Sybase::ASE::NoBindVars> transactions |
1001 | are disabled, as there are no concurrency issues with C<SELECT @@IDENTITY> as |
1002 | it's a session variable. |
1003 | |
1004 | =head1 TRANSACTIONS |
1005 | |
48580715 |
1006 | Due to limitations of the TDS protocol, L<DBD::Sybase>, or both, you cannot |
1007 | begin a transaction while there are active cursors, nor can you use multiple |
057db5ce |
1008 | active cursors within a transaction. An active cursor is, for example, a |
1009 | L<ResultSet|DBIx::Class::ResultSet> that has been executed using C<next> or |
1010 | C<first> but has not been exhausted or L<reset|DBIx::Class::ResultSet/reset>. |
1011 | |
1012 | For example, this will not work: |
1013 | |
1014 | $schema->txn_do(sub { |
1015 | my $rs = $schema->resultset('Book'); |
1016 | while (my $row = $rs->next) { |
1017 | $schema->resultset('MetaData')->create({ |
1018 | book_id => $row->id, |
1019 | ... |
1020 | }); |
1021 | } |
1022 | }); |
1023 | |
1024 | This won't either: |
1025 | |
1026 | my $first_row = $large_rs->first; |
1027 | $schema->txn_do(sub { ... }); |
1028 | |
1029 | Transactions done for inserts in C<AutoCommit> mode when placeholders are in use |
1030 | are not affected, as they are done on an extra database handle. |
1031 | |
1032 | Some workarounds: |
1033 | |
1034 | =over 4 |
1035 | |
1036 | =item * use L<DBIx::Class::Storage::DBI::Replicated> |
1037 | |
1038 | =item * L<connect|DBIx::Class::Schema/connect> another L<Schema|DBIx::Class::Schema> |
1039 | |
1040 | =item * load the data from your cursor with L<DBIx::Class::ResultSet/all> |
1041 | |
1042 | =back |
1043 | |
1044 | =head1 MAXIMUM CONNECTIONS |
1045 | |
1046 | The TDS protocol makes separate connections to the server for active statements |
1047 | in the background. By default the number of such connections is limited to 25, |
1048 | on both the client side and the server side. |
1049 | |
1050 | This is a bit too low for a complex L<DBIx::Class> application, so on connection |
1051 | the client side setting is set to C<256> (see L<DBD::Sybase/maxConnect>.) You |
1052 | can override it to whatever setting you like in the DSN. |
1053 | |
1054 | See |
1055 | L<http://infocenter.sybase.com/help/index.jsp?topic=/com.sybase.help.ase_15.0.sag1/html/sag1/sag1272.htm> |
1056 | for information on changing the setting on the server side. |
1057 | |
1058 | =head1 DATES |
1059 | |
1060 | See L</connect_call_datetime_setup> to setup date formats |
1061 | for L<DBIx::Class::InflateColumn::DateTime>. |
1062 | |
1063 | =head1 TEXT/IMAGE COLUMNS |
1064 | |
1065 | L<DBD::Sybase> compiled with FreeTDS will B<NOT> allow you to insert or update |
1066 | C<TEXT/IMAGE> columns. |
1067 | |
1068 | Setting C<< $dbh->{LongReadLen} >> will also not work with FreeTDS use either: |
1069 | |
1070 | $schema->storage->dbh->do("SET TEXTSIZE $bytes"); |
1071 | |
1072 | or |
1073 | |
1074 | $schema->storage->set_textsize($bytes); |
1075 | |
1076 | instead. |
1077 | |
1078 | However, the C<LongReadLen> you pass in |
8384a713 |
1079 | L<connect_info|DBIx::Class::Storage::DBI/connect_info> is used to execute the |
1080 | equivalent C<SET TEXTSIZE> command on connection. |
057db5ce |
1081 | |
8384a713 |
1082 | See L</connect_call_blob_setup> for a |
1083 | L<connect_info|DBIx::Class::Storage::DBI/connect_info> setting you need to work |
1084 | with C<IMAGE> columns. |
057db5ce |
1085 | |
1086 | =head1 BULK API |
1087 | |
1088 | The experimental L<DBD::Sybase> Bulk API support is used for |
1089 | L<populate|DBIx::Class::ResultSet/populate> in B<void> context, in a transaction |
1090 | on a separate connection. |
1091 | |
1092 | To use this feature effectively, use a large number of rows for each |
1093 | L<populate|DBIx::Class::ResultSet/populate> call, eg.: |
1094 | |
1095 | while (my $rows = $data_source->get_100_rows()) { |
1096 | $rs->populate($rows); |
1097 | } |
1098 | |
1099 | B<NOTE:> the L<add_columns|DBIx::Class::ResultSource/add_columns> |
1100 | calls in your C<Result> classes B<must> list columns in database order for this |
1101 | to work. Also, you may have to unset the C<LANG> environment variable before |
1102 | loading your app, if it doesn't match the character set of your database. |
1103 | |
1104 | When inserting IMAGE columns using this method, you'll need to use |
1105 | L</connect_call_blob_setup> as well. |
1106 | |
6476fd66 |
1107 | =head1 COMPUTED COLUMNS |
1108 | |
1109 | If you have columns such as: |
1110 | |
1111 | created_dtm AS getdate() |
1112 | |
1113 | represent them in your Result classes as: |
1114 | |
1115 | created_dtm => { |
1116 | data_type => undef, |
1117 | default_value => \'getdate()', |
1118 | is_nullable => 0, |
1119 | } |
1120 | |
1121 | The C<data_type> must exist and must be C<undef>. Then empty inserts will work |
1122 | on tables with such columns. |
1123 | |
1124 | =head1 TIMESTAMP COLUMNS |
1125 | |
1126 | C<timestamp> columns in Sybase ASE are not really timestamps, see: |
1127 | L<http://dba.fyicenter.com/Interview-Questions/SYBASE/The_timestamp_datatype_in_Sybase_.html>. |
1128 | |
1129 | They should be defined in your Result classes as: |
1130 | |
1131 | ts => { |
1132 | data_type => 'timestamp', |
1133 | is_nullable => 0, |
1134 | inflate_datetime => 0, |
1135 | } |
1136 | |
1137 | The C<<inflate_datetime => 0>> is necessary if you use |
1138 | L<DBIx::Class::InflateColumn::DateTime>, and most people do, and still want to |
1139 | be able to read these values. |
1140 | |
1141 | The values will come back as hexadecimal. |
1142 | |
057db5ce |
1143 | =head1 TODO |
1144 | |
1145 | =over |
1146 | |
1147 | =item * |
1148 | |
1149 | Transitions to AutoCommit=0 (starting a transaction) mode by exhausting |
1150 | any active cursors, using eager cursors. |
1151 | |
1152 | =item * |
1153 | |
1154 | Real limits and limited counts using stored procedures deployed on startup. |
1155 | |
1156 | =item * |
1157 | |
d5dedbd6 |
1158 | Adaptive Server Anywhere (ASA) support |
057db5ce |
1159 | |
1160 | =item * |
1161 | |
1162 | Blob update with a LIKE query on a blob, without invalidating the WHERE condition. |
1163 | |
1164 | =item * |
1165 | |
1166 | bulk_insert using prepare_cached (see comments.) |
1167 | |
1168 | =back |
1169 | |
1170 | =head1 AUTHOR |
1171 | |
8138f96a |
1172 | See L<DBIx::Class/AUTHOR> and L<DBIx::Class/CONTRIBUTORS>. |
057db5ce |
1173 | |
1174 | =head1 LICENSE |
1175 | |
1176 | You may distribute this code under the same terms as Perl itself. |
1177 | |
1178 | =cut |
1179 | # vim:sts=2 sw=2: |