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'; |
70c28808 |
11 | use DBIx::Class::Carp; |
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 | $self->next::method; |
174 | } |
175 | |
c1e5a9ac |
176 | # This is only invoked for FreeTDS drivers by ::Storage::DBI::Sybase::FreeTDS |
177 | sub _set_autocommit_stmt { |
178 | my ($self, $on) = @_; |
179 | |
180 | return 'SET CHAINED ' . ($on ? 'OFF' : 'ON'); |
181 | } |
182 | |
057db5ce |
183 | # Set up session settings for Sybase databases for the connection. |
184 | # |
185 | # Make sure we have CHAINED mode turned on if AutoCommit is off in non-FreeTDS |
186 | # DBD::Sybase (since we don't know how DBD::Sybase was compiled.) If however |
187 | # we're using FreeTDS, CHAINED mode turns on an implicit transaction which we |
188 | # only want when AutoCommit is off. |
057db5ce |
189 | sub _run_connection_actions { |
190 | my $self = shift; |
191 | |
192 | if ($self->_is_bulk_storage) { |
c1e5a9ac |
193 | # this should be cleared on every reconnect |
057db5ce |
194 | $self->_began_bulk_work(0); |
195 | return; |
196 | } |
197 | |
c1e5a9ac |
198 | $self->_dbh->{syb_chained_txn} = 1 |
199 | unless $self->using_freetds; |
057db5ce |
200 | |
201 | $self->next::method(@_); |
202 | } |
203 | |
204 | =head2 connect_call_blob_setup |
205 | |
206 | Used as: |
207 | |
208 | on_connect_call => [ [ 'blob_setup', log_on_update => 0 ] ] |
209 | |
210 | Does C<< $dbh->{syb_binary_images} = 1; >> to return C<IMAGE> data as raw binary |
211 | instead of as a hex string. |
212 | |
213 | Recommended. |
214 | |
215 | Also sets the C<log_on_update> value for blob write operations. The default is |
216 | C<1>, but C<0> is better if your database is configured for it. |
217 | |
218 | See |
219 | L<DBD::Sybase/Handling_IMAGE/TEXT_data_with_syb_ct_get_data()/syb_ct_send_data()>. |
220 | |
221 | =cut |
222 | |
223 | sub connect_call_blob_setup { |
224 | my $self = shift; |
225 | my %args = @_; |
226 | my $dbh = $self->_dbh; |
227 | $dbh->{syb_binary_images} = 1; |
228 | |
229 | $self->_blob_log_on_update($args{log_on_update}) |
230 | if exists $args{log_on_update}; |
231 | } |
232 | |
057db5ce |
233 | sub _is_lob_column { |
234 | my ($self, $source, $column) = @_; |
235 | |
236 | return $self->_is_lob_type($source->column_info($column)->{data_type}); |
237 | } |
238 | |
239 | sub _prep_for_execute { |
240 | my $self = shift; |
0e773352 |
241 | my ($op, $ident, $args) = @_; |
057db5ce |
242 | |
243 | my ($sql, $bind) = $self->next::method (@_); |
244 | |
6298a324 |
245 | my $table = blessed $ident ? $ident->from : $ident; |
057db5ce |
246 | |
247 | my $bind_info = $self->_resolve_column_info( |
0e773352 |
248 | $ident, [map { $_->[0]{dbic_colname} || () } @{$bind}] |
057db5ce |
249 | ); |
6298a324 |
250 | my $bound_identity_col = |
251 | first { $bind_info->{$_}{is_auto_increment} } |
252 | keys %$bind_info |
057db5ce |
253 | ; |
e366f807 |
254 | |
255 | my $columns_info = blessed $ident && $ident->columns_info; |
256 | |
6298a324 |
257 | my $identity_col = |
e366f807 |
258 | $columns_info && |
259 | first { $columns_info->{$_}{is_auto_increment} } |
260 | keys %$columns_info |
057db5ce |
261 | ; |
262 | |
263 | if (($op eq 'insert' && $bound_identity_col) || |
264 | ($op eq 'update' && exists $args->[0]{$identity_col})) { |
265 | $sql = join ("\n", |
266 | $self->_set_table_identity_sql($op => $table, 'on'), |
267 | $sql, |
268 | $self->_set_table_identity_sql($op => $table, 'off'), |
269 | ); |
270 | } |
271 | |
272 | if ($op eq 'insert' && (not $bound_identity_col) && $identity_col && |
273 | (not $self->{insert_bulk})) { |
274 | $sql = |
275 | "$sql\n" . |
276 | $self->_fetch_identity_sql($ident, $identity_col); |
277 | } |
278 | |
279 | return ($sql, $bind); |
280 | } |
281 | |
282 | sub _set_table_identity_sql { |
283 | my ($self, $op, $table, $on_off) = @_; |
284 | |
285 | return sprintf 'SET IDENTITY_%s %s %s', |
286 | uc($op), $self->sql_maker->_quote($table), uc($on_off); |
287 | } |
288 | |
289 | # Stolen from SQLT, with some modifications. This is a makeshift |
290 | # solution before a sane type-mapping library is available, thus |
291 | # the 'our' for easy overrides. |
292 | our %TYPE_MAPPING = ( |
293 | number => 'numeric', |
294 | money => 'money', |
295 | varchar => 'varchar', |
296 | varchar2 => 'varchar', |
297 | timestamp => 'datetime', |
298 | text => 'varchar', |
299 | real => 'double precision', |
300 | comment => 'text', |
301 | bit => 'bit', |
302 | tinyint => 'smallint', |
303 | float => 'double precision', |
304 | serial => 'numeric', |
305 | bigserial => 'numeric', |
306 | boolean => 'varchar', |
307 | long => 'varchar', |
308 | ); |
309 | |
310 | sub _native_data_type { |
311 | my ($self, $type) = @_; |
312 | |
313 | $type = lc $type; |
314 | $type =~ s/\s* identity//x; |
315 | |
316 | return uc($TYPE_MAPPING{$type} || $type); |
317 | } |
318 | |
319 | sub _fetch_identity_sql { |
320 | my ($self, $source, $col) = @_; |
321 | |
322 | return sprintf ("SELECT MAX(%s) FROM %s", |
323 | map { $self->sql_maker->_quote ($_) } ($col, $source->from) |
324 | ); |
325 | } |
326 | |
327 | sub _execute { |
328 | my $self = shift; |
329 | my ($op) = @_; |
330 | |
0e773352 |
331 | my ($rv, $sth, @bind) = $self->next::method(@_); |
057db5ce |
332 | |
333 | if ($op eq 'insert') { |
334 | $self->_identity($sth->fetchrow_array); |
335 | $sth->finish; |
336 | } |
337 | |
338 | return wantarray ? ($rv, $sth, @bind) : $rv; |
339 | } |
340 | |
341 | sub last_insert_id { shift->_identity } |
342 | |
343 | # handles TEXT/IMAGE and transaction for last_insert_id |
344 | sub insert { |
345 | my $self = shift; |
346 | my ($source, $to_insert) = @_; |
347 | |
e366f807 |
348 | my $columns_info = $source->columns_info; |
349 | |
6298a324 |
350 | my $identity_col = |
e366f807 |
351 | (first { $columns_info->{$_}{is_auto_increment} } |
352 | keys %$columns_info ) |
6298a324 |
353 | || ''; |
057db5ce |
354 | |
355 | # check for empty insert |
356 | # INSERT INTO foo DEFAULT VALUES -- does not work with Sybase |
6469dabf |
357 | # try to insert explicit 'DEFAULT's instead (except for identity, timestamp |
358 | # and computed columns) |
057db5ce |
359 | if (not %$to_insert) { |
360 | for my $col ($source->columns) { |
361 | next if $col eq $identity_col; |
6469dabf |
362 | |
363 | my $info = $source->column_info($col); |
364 | |
365 | next if ref $info->{default_value} eq 'SCALAR' |
366 | || (exists $info->{data_type} && (not defined $info->{data_type})); |
367 | |
368 | next if $info->{data_type} && $info->{data_type} =~ /^timestamp\z/i; |
369 | |
057db5ce |
370 | $to_insert->{$col} = \'DEFAULT'; |
371 | } |
372 | } |
373 | |
374 | my $blob_cols = $self->_remove_blob_cols($source, $to_insert); |
375 | |
376 | # do we need the horrific SELECT MAX(COL) hack? |
377 | my $dumb_last_insert_id = |
378 | $identity_col |
379 | && (not exists $to_insert->{$identity_col}) |
380 | && ($self->_identity_method||'') ne '@@IDENTITY'; |
381 | |
382 | my $next = $self->next::can; |
383 | |
384 | # we are already in a transaction, or there are no blobs |
385 | # and we don't need the PK - just (try to) do it |
386 | if ($self->{transaction_depth} |
387 | || (!$blob_cols && !$dumb_last_insert_id) |
388 | ) { |
389 | return $self->_insert ( |
390 | $next, $source, $to_insert, $blob_cols, $identity_col |
391 | ); |
392 | } |
393 | |
394 | # otherwise use the _writer_storage to do the insert+transaction on another |
395 | # connection |
396 | my $guard = $self->_writer_storage->txn_scope_guard; |
397 | |
398 | my $updated_cols = $self->_writer_storage->_insert ( |
399 | $next, $source, $to_insert, $blob_cols, $identity_col |
400 | ); |
401 | |
402 | $self->_identity($self->_writer_storage->_identity); |
403 | |
404 | $guard->commit; |
405 | |
406 | return $updated_cols; |
407 | } |
408 | |
409 | sub _insert { |
410 | my ($self, $next, $source, $to_insert, $blob_cols, $identity_col) = @_; |
411 | |
412 | my $updated_cols = $self->$next ($source, $to_insert); |
413 | |
414 | my $final_row = { |
415 | ($identity_col ? |
416 | ($identity_col => $self->last_insert_id($source, $identity_col)) : ()), |
417 | %$to_insert, |
418 | %$updated_cols, |
419 | }; |
420 | |
421 | $self->_insert_blobs ($source, $blob_cols, $final_row) if $blob_cols; |
422 | |
423 | return $updated_cols; |
424 | } |
425 | |
426 | sub update { |
427 | my $self = shift; |
428 | my ($source, $fields, $where, @rest) = @_; |
429 | |
057db5ce |
430 | my $blob_cols = $self->_remove_blob_cols($source, $fields); |
431 | |
432 | my $table = $source->name; |
433 | |
e366f807 |
434 | my $columns_info = $source->columns_info; |
435 | |
6298a324 |
436 | my $identity_col = |
e366f807 |
437 | first { $columns_info->{$_}{is_auto_increment} } |
438 | keys %$columns_info; |
057db5ce |
439 | |
440 | my $is_identity_update = $identity_col && defined $fields->{$identity_col}; |
441 | |
442 | return $self->next::method(@_) unless $blob_cols; |
443 | |
444 | # If there are any blobs in $where, Sybase will return a descriptive error |
445 | # message. |
446 | # XXX blobs can still be used with a LIKE query, and this should be handled. |
447 | |
448 | # update+blob update(s) done atomically on separate connection |
449 | $self = $self->_writer_storage; |
450 | |
451 | my $guard = $self->txn_scope_guard; |
452 | |
453 | # First update the blob columns to be updated to '' (taken from $fields, where |
454 | # it is originally put by _remove_blob_cols .) |
455 | my %blobs_to_empty = map { ($_ => delete $fields->{$_}) } keys %$blob_cols; |
456 | |
457 | # We can't only update NULL blobs, because blobs cannot be in the WHERE clause. |
458 | |
459 | $self->next::method($source, \%blobs_to_empty, $where, @rest); |
460 | |
461 | # Now update the blobs before the other columns in case the update of other |
462 | # columns makes the search condition invalid. |
463 | $self->_update_blobs($source, $blob_cols, $where); |
464 | |
465 | my @res; |
466 | if (%$fields) { |
cca282b6 |
467 | if (wantarray) { |
057db5ce |
468 | @res = $self->next::method(@_); |
469 | } |
cca282b6 |
470 | elsif (defined wantarray) { |
057db5ce |
471 | $res[0] = $self->next::method(@_); |
472 | } |
473 | else { |
474 | $self->next::method(@_); |
475 | } |
476 | } |
477 | |
478 | $guard->commit; |
479 | |
cca282b6 |
480 | return wantarray ? @res : $res[0]; |
057db5ce |
481 | } |
482 | |
483 | sub insert_bulk { |
484 | my $self = shift; |
485 | my ($source, $cols, $data) = @_; |
486 | |
e366f807 |
487 | my $columns_info = $source->columns_info; |
488 | |
6298a324 |
489 | my $identity_col = |
e366f807 |
490 | first { $columns_info->{$_}{is_auto_increment} } |
491 | keys %$columns_info; |
057db5ce |
492 | |
6298a324 |
493 | my $is_identity_insert = (first { $_ eq $identity_col } @{$cols}) ? 1 : 0; |
057db5ce |
494 | |
495 | my @source_columns = $source->columns; |
496 | |
497 | my $use_bulk_api = |
498 | $self->_bulk_storage && |
499 | $self->_get_dbh->{syb_has_blk}; |
500 | |
501 | if ((not $use_bulk_api) |
502 | && |
503 | (ref($self->_dbi_connect_info->[0]) eq 'CODE') |
504 | && |
505 | (not $self->_bulk_disabled_due_to_coderef_connect_info_warned)) { |
506 | carp <<'EOF'; |
507 | Bulk API support disabled due to use of a CODEREF connect_info. Reverting to |
508 | regular array inserts. |
509 | EOF |
510 | $self->_bulk_disabled_due_to_coderef_connect_info_warned(1); |
511 | } |
512 | |
513 | if (not $use_bulk_api) { |
514 | my $blob_cols = $self->_remove_blob_cols_array($source, $cols, $data); |
515 | |
516 | # _execute_array uses a txn anyway, but it ends too early in case we need to |
517 | # select max(col) to get the identity for inserting blobs. |
518 | ($self, my $guard) = $self->{transaction_depth} == 0 ? |
519 | ($self->_writer_storage, $self->_writer_storage->txn_scope_guard) |
520 | : |
521 | ($self, undef); |
522 | |
523 | local $self->{insert_bulk} = 1; |
524 | |
525 | $self->next::method(@_); |
526 | |
527 | if ($blob_cols) { |
528 | if ($is_identity_insert) { |
529 | $self->_insert_blobs_array ($source, $blob_cols, $cols, $data); |
530 | } |
531 | else { |
532 | my @cols_with_identities = (@$cols, $identity_col); |
533 | |
534 | ## calculate identities |
535 | # XXX This assumes identities always increase by 1, which may or may not |
536 | # be true. |
537 | my ($last_identity) = |
538 | $self->_dbh->selectrow_array ( |
539 | $self->_fetch_identity_sql($source, $identity_col) |
540 | ); |
541 | my @identities = (($last_identity - @$data + 1) .. $last_identity); |
542 | |
543 | my @data_with_identities = map [@$_, shift @identities], @$data; |
544 | |
545 | $self->_insert_blobs_array ( |
546 | $source, $blob_cols, \@cols_with_identities, \@data_with_identities |
547 | ); |
548 | } |
549 | } |
550 | |
551 | $guard->commit if $guard; |
552 | |
553 | return; |
554 | } |
555 | |
556 | # otherwise, use the bulk API |
557 | |
558 | # rearrange @$data so that columns are in database order |
559 | my %orig_idx; |
560 | @orig_idx{@$cols} = 0..$#$cols; |
561 | |
562 | my %new_idx; |
563 | @new_idx{@source_columns} = 0..$#source_columns; |
564 | |
565 | my @new_data; |
566 | for my $datum (@$data) { |
567 | my $new_datum = []; |
568 | for my $col (@source_columns) { |
569 | # identity data will be 'undef' if not $is_identity_insert |
570 | # columns with defaults will also be 'undef' |
571 | $new_datum->[ $new_idx{$col} ] = |
572 | exists $orig_idx{$col} ? $datum->[ $orig_idx{$col} ] : undef; |
573 | } |
574 | push @new_data, $new_datum; |
575 | } |
576 | |
577 | # bcp identity index is 1-based |
578 | my $identity_idx = exists $new_idx{$identity_col} ? |
579 | $new_idx{$identity_col} + 1 : 0; |
580 | |
581 | ## Set a client-side conversion error handler, straight from DBD::Sybase docs. |
582 | # This ignores any data conversion errors detected by the client side libs, as |
583 | # they are usually harmless. |
584 | my $orig_cslib_cb = DBD::Sybase::set_cslib_cb( |
585 | Sub::Name::subname insert_bulk => sub { |
586 | my ($layer, $origin, $severity, $errno, $errmsg, $osmsg, $blkmsg) = @_; |
587 | |
588 | return 1 if $errno == 36; |
589 | |
590 | carp |
591 | "Layer: $layer, Origin: $origin, Severity: $severity, Error: $errno" . |
592 | ($errmsg ? "\n$errmsg" : '') . |
593 | ($osmsg ? "\n$osmsg" : '') . |
594 | ($blkmsg ? "\n$blkmsg" : ''); |
595 | |
596 | return 0; |
597 | }); |
598 | |
4edfce2f |
599 | my $exception = ''; |
ed7ab0f4 |
600 | try { |
057db5ce |
601 | my $bulk = $self->_bulk_storage; |
602 | |
603 | my $guard = $bulk->txn_scope_guard; |
604 | |
605 | ## XXX get this to work instead of our own $sth |
606 | ## will require SQLA or *Hacks changes for ordered columns |
607 | # $bulk->next::method($source, \@source_columns, \@new_data, { |
608 | # syb_bcp_attribs => { |
609 | # identity_flag => $is_identity_insert, |
610 | # identity_column => $identity_idx, |
611 | # } |
612 | # }); |
613 | my $sql = 'INSERT INTO ' . |
614 | $bulk->sql_maker->_quote($source->name) . ' (' . |
615 | # colname list is ignored for BCP, but does no harm |
616 | (join ', ', map $bulk->sql_maker->_quote($_), @source_columns) . ') '. |
617 | ' VALUES ('. (join ', ', ('?') x @source_columns) . ')'; |
618 | |
619 | ## XXX there's a bug in the DBD::Sybase bulk support that makes $sth->finish for |
620 | ## a prepare_cached statement ineffective. Replace with ->sth when fixed, or |
621 | ## better yet the version above. Should be fixed in DBD::Sybase . |
622 | my $sth = $bulk->_get_dbh->prepare($sql, |
623 | # 'insert', # op |
624 | { |
625 | syb_bcp_attribs => { |
626 | identity_flag => $is_identity_insert, |
627 | identity_column => $identity_idx, |
628 | } |
629 | } |
630 | ); |
631 | |
0e773352 |
632 | my @bind = map { [ $source_columns[$_] => $_ ] } (0 .. $#source_columns); |
057db5ce |
633 | |
634 | $self->_execute_array( |
635 | $source, $sth, \@bind, \@source_columns, \@new_data, sub { |
636 | $guard->commit |
637 | } |
638 | ); |
639 | |
640 | $bulk->_query_end($sql); |
ed7ab0f4 |
641 | } catch { |
642 | $exception = shift; |
057db5ce |
643 | }; |
644 | |
057db5ce |
645 | DBD::Sybase::set_cslib_cb($orig_cslib_cb); |
646 | |
647 | if ($exception =~ /-Y option/) { |
f32e99f9 |
648 | my $w = 'Sybase bulk API operation failed due to character set incompatibility, ' |
649 | . 'reverting to regular array inserts. Try unsetting the LANG environment variable' |
650 | ; |
651 | $w .= "\n$exception" if $self->debug; |
652 | carp $w; |
057db5ce |
653 | |
057db5ce |
654 | $self->_bulk_storage(undef); |
655 | unshift @_, $self; |
656 | goto \&insert_bulk; |
657 | } |
658 | elsif ($exception) { |
659 | # rollback makes the bulkLogin connection unusable |
660 | $self->_bulk_storage->disconnect; |
661 | $self->throw_exception($exception); |
662 | } |
663 | } |
664 | |
665 | sub _dbh_execute_array { |
666 | my ($self, $sth, $tuple_status, $cb) = @_; |
667 | |
668 | my $rv = $self->next::method($sth, $tuple_status); |
669 | $cb->() if $cb; |
670 | |
671 | return $rv; |
672 | } |
673 | |
674 | # Make sure blobs are not bound as placeholders, and return any non-empty ones |
675 | # as a hash. |
676 | sub _remove_blob_cols { |
677 | my ($self, $source, $fields) = @_; |
678 | |
679 | my %blob_cols; |
680 | |
681 | for my $col (keys %$fields) { |
682 | if ($self->_is_lob_column($source, $col)) { |
683 | my $blob_val = delete $fields->{$col}; |
684 | if (not defined $blob_val) { |
685 | $fields->{$col} = \'NULL'; |
686 | } |
687 | else { |
688 | $fields->{$col} = \"''"; |
689 | $blob_cols{$col} = $blob_val unless $blob_val eq ''; |
690 | } |
691 | } |
692 | } |
693 | |
694 | return %blob_cols ? \%blob_cols : undef; |
695 | } |
696 | |
697 | # same for insert_bulk |
698 | sub _remove_blob_cols_array { |
699 | my ($self, $source, $cols, $data) = @_; |
700 | |
701 | my @blob_cols; |
702 | |
703 | for my $i (0..$#$cols) { |
704 | my $col = $cols->[$i]; |
705 | |
706 | if ($self->_is_lob_column($source, $col)) { |
707 | for my $j (0..$#$data) { |
708 | my $blob_val = delete $data->[$j][$i]; |
709 | if (not defined $blob_val) { |
710 | $data->[$j][$i] = \'NULL'; |
711 | } |
712 | else { |
713 | $data->[$j][$i] = \"''"; |
714 | $blob_cols[$j][$i] = $blob_val |
715 | unless $blob_val eq ''; |
716 | } |
717 | } |
718 | } |
719 | } |
720 | |
721 | return @blob_cols ? \@blob_cols : undef; |
722 | } |
723 | |
724 | sub _update_blobs { |
725 | my ($self, $source, $blob_cols, $where) = @_; |
726 | |
9780718f |
727 | my @primary_cols = try |
728 | { $source->_pri_cols } |
729 | catch { |
730 | $self->throw_exception("Cannot update TEXT/IMAGE column(s): $_") |
731 | }; |
057db5ce |
732 | |
733 | # check if we're updating a single row by PK |
734 | my $pk_cols_in_where = 0; |
735 | for my $col (@primary_cols) { |
736 | $pk_cols_in_where++ if defined $where->{$col}; |
737 | } |
738 | my @rows; |
739 | |
740 | if ($pk_cols_in_where == @primary_cols) { |
741 | my %row_to_update; |
742 | @row_to_update{@primary_cols} = @{$where}{@primary_cols}; |
743 | @rows = \%row_to_update; |
744 | } else { |
745 | my $cursor = $self->select ($source, \@primary_cols, $where, {}); |
746 | @rows = map { |
747 | my %row; @row{@primary_cols} = @$_; \%row |
748 | } $cursor->all; |
749 | } |
750 | |
751 | for my $row (@rows) { |
752 | $self->_insert_blobs($source, $blob_cols, $row); |
753 | } |
754 | } |
755 | |
756 | sub _insert_blobs { |
757 | my ($self, $source, $blob_cols, $row) = @_; |
758 | my $dbh = $self->_get_dbh; |
759 | |
760 | my $table = $source->name; |
761 | |
762 | my %row = %$row; |
9780718f |
763 | my @primary_cols = try |
764 | { $source->_pri_cols } |
765 | catch { |
766 | $self->throw_exception("Cannot update TEXT/IMAGE column(s): $_") |
767 | }; |
057db5ce |
768 | |
769 | $self->throw_exception('Cannot update TEXT/IMAGE column(s) without primary key values') |
770 | if ((grep { defined $row{$_} } @primary_cols) != @primary_cols); |
771 | |
772 | for my $col (keys %$blob_cols) { |
773 | my $blob = $blob_cols->{$col}; |
774 | |
775 | my %where = map { ($_, $row{$_}) } @primary_cols; |
776 | |
777 | my $cursor = $self->select ($source, [$col], \%where, {}); |
778 | $cursor->next; |
779 | my $sth = $cursor->sth; |
780 | |
781 | if (not $sth) { |
057db5ce |
782 | $self->throw_exception( |
783 | "Could not find row in table '$table' for blob update:\n" |
6298a324 |
784 | . (Dumper \%where) |
057db5ce |
785 | ); |
786 | } |
787 | |
9780718f |
788 | try { |
057db5ce |
789 | do { |
790 | $sth->func('CS_GET', 1, 'ct_data_info') or die $sth->errstr; |
791 | } while $sth->fetch; |
792 | |
793 | $sth->func('ct_prepare_send') or die $sth->errstr; |
794 | |
795 | my $log_on_update = $self->_blob_log_on_update; |
796 | $log_on_update = 1 if not defined $log_on_update; |
797 | |
798 | $sth->func('CS_SET', 1, { |
799 | total_txtlen => length($blob), |
800 | log_on_update => $log_on_update |
801 | }, 'ct_data_info') or die $sth->errstr; |
802 | |
803 | $sth->func($blob, length($blob), 'ct_send_data') or die $sth->errstr; |
804 | |
805 | $sth->func('ct_finish_send') or die $sth->errstr; |
9780718f |
806 | } |
807 | catch { |
057db5ce |
808 | if ($self->using_freetds) { |
809 | $self->throw_exception ( |
9780718f |
810 | "TEXT/IMAGE operation failed, probably because you are using FreeTDS: $_" |
057db5ce |
811 | ); |
9780718f |
812 | } |
813 | else { |
814 | $self->throw_exception($_); |
057db5ce |
815 | } |
816 | } |
9780718f |
817 | finally { |
818 | $sth->finish if $sth; |
819 | }; |
057db5ce |
820 | } |
821 | } |
822 | |
823 | sub _insert_blobs_array { |
824 | my ($self, $source, $blob_cols, $cols, $data) = @_; |
825 | |
826 | for my $i (0..$#$data) { |
827 | my $datum = $data->[$i]; |
828 | |
829 | my %row; |
830 | @row{ @$cols } = @$datum; |
831 | |
832 | my %blob_vals; |
833 | for my $j (0..$#$cols) { |
834 | if (exists $blob_cols->[$i][$j]) { |
835 | $blob_vals{ $cols->[$j] } = $blob_cols->[$i][$j]; |
836 | } |
837 | } |
838 | |
839 | $self->_insert_blobs ($source, \%blob_vals, \%row); |
840 | } |
841 | } |
842 | |
843 | =head2 connect_call_datetime_setup |
844 | |
845 | Used as: |
846 | |
847 | on_connect_call => 'datetime_setup' |
848 | |
8384a713 |
849 | In L<connect_info|DBIx::Class::Storage::DBI/connect_info> to set: |
057db5ce |
850 | |
851 | $dbh->syb_date_fmt('ISO_strict'); # output fmt: 2004-08-21T14:36:48.080Z |
852 | $dbh->do('set dateformat mdy'); # input fmt: 08/13/1979 18:08:55.080 |
853 | |
854 | On connection for use with L<DBIx::Class::InflateColumn::DateTime>, using |
855 | L<DateTime::Format::Sybase>, which you will need to install. |
856 | |
c1e5a9ac |
857 | This works for both C<DATETIME> and C<SMALLDATETIME> columns, note that |
057db5ce |
858 | C<SMALLDATETIME> columns only have minute precision. |
859 | |
860 | =cut |
861 | |
70c28808 |
862 | sub connect_call_datetime_setup { |
863 | my $self = shift; |
864 | my $dbh = $self->_get_dbh; |
057db5ce |
865 | |
70c28808 |
866 | if ($dbh->can('syb_date_fmt')) { |
867 | # amazingly, this works with FreeTDS |
868 | $dbh->syb_date_fmt('ISO_strict'); |
869 | } |
870 | else { |
871 | carp_once |
872 | 'Your DBD::Sybase is too old to support ' |
873 | .'DBIx::Class::InflateColumn::DateTime, please upgrade!'; |
057db5ce |
874 | |
875 | $dbh->do('SET DATEFORMAT mdy'); |
057db5ce |
876 | 1; |
877 | } |
878 | } |
879 | |
057db5ce |
880 | |
057db5ce |
881 | sub _dbh_begin_work { |
882 | my $self = shift; |
883 | |
884 | # bulkLogin=1 connections are always in a transaction, and can only call BEGIN |
885 | # TRAN once. However, we need to make sure there's a $dbh. |
886 | return if $self->_is_bulk_storage && $self->_dbh && $self->_began_bulk_work; |
887 | |
888 | $self->next::method(@_); |
889 | |
057db5ce |
890 | $self->_began_bulk_work(1) if $self->_is_bulk_storage; |
891 | } |
892 | |
057db5ce |
893 | # savepoint support using ASE syntax |
894 | |
895 | sub _svp_begin { |
896 | my ($self, $name) = @_; |
897 | |
898 | $self->_get_dbh->do("SAVE TRANSACTION $name"); |
899 | } |
900 | |
901 | # A new SAVE TRANSACTION with the same name releases the previous one. |
902 | sub _svp_release { 1 } |
903 | |
904 | sub _svp_rollback { |
905 | my ($self, $name) = @_; |
906 | |
907 | $self->_get_dbh->do("ROLLBACK TRANSACTION $name"); |
908 | } |
909 | |
910 | 1; |
911 | |
912 | =head1 Schema::Loader Support |
913 | |
290da7d6 |
914 | As of version C<0.05000>, L<DBIx::Class::Schema::Loader> should work well with |
c1e5a9ac |
915 | most versions of Sybase ASE. |
057db5ce |
916 | |
917 | =head1 FreeTDS |
918 | |
919 | This driver supports L<DBD::Sybase> compiled against FreeTDS |
920 | (L<http://www.freetds.org/>) to the best of our ability, however it is |
921 | recommended that you recompile L<DBD::Sybase> against the Sybase Open Client |
922 | libraries. They are a part of the Sybase ASE distribution: |
923 | |
924 | The Open Client FAQ is here: |
925 | L<http://www.isug.com/Sybase_FAQ/ASE/section7.html>. |
926 | |
927 | Sybase ASE for Linux (which comes with the Open Client libraries) may be |
928 | downloaded here: L<http://response.sybase.com/forms/ASE_Linux_Download>. |
929 | |
930 | To see if you're using FreeTDS check C<< $schema->storage->using_freetds >>, or run: |
931 | |
932 | perl -MDBI -le 'my $dbh = DBI->connect($dsn, $user, $pass); print $dbh->{syb_oc_version}' |
933 | |
c1e5a9ac |
934 | It is recommended to set C<tds version> for your ASE server to C<5.0> in |
935 | C</etc/freetds/freetds.conf>. |
936 | |
937 | Some versions or configurations of the libraries involved will not support |
938 | placeholders, in which case the storage will be reblessed to |
057db5ce |
939 | L<DBIx::Class::Storage::DBI::Sybase::ASE::NoBindVars>. |
940 | |
941 | In some configurations, placeholders will work but will throw implicit type |
942 | conversion errors for anything that's not expecting a string. In such a case, |
943 | the C<auto_cast> option from L<DBIx::Class::Storage::DBI::AutoCast> is |
944 | automatically set, which you may enable on connection with |
c1e5a9ac |
945 | L<connect_call_set_auto_cast|DBIx::Class::Storage::DBI::AutoCast/connect_call_set_auto_cast>. |
946 | The type info for the C<CAST>s is taken from the |
947 | L<DBIx::Class::ResultSource/data_type> definitions in your Result classes, and |
948 | are mapped to a Sybase type (if it isn't already) using a mapping based on |
949 | L<SQL::Translator>. |
057db5ce |
950 | |
48580715 |
951 | In other configurations, placeholders will work just as they do with the Sybase |
057db5ce |
952 | Open Client libraries. |
953 | |
954 | Inserts or updates of TEXT/IMAGE columns will B<NOT> work with FreeTDS. |
955 | |
956 | =head1 INSERTS WITH PLACEHOLDERS |
957 | |
958 | With placeholders enabled, inserts are done in a transaction so that there are |
959 | no concurrency issues with getting the inserted identity value using |
960 | C<SELECT MAX(col)>, which is the only way to get the C<IDENTITY> value in this |
961 | mode. |
962 | |
963 | In addition, they are done on a separate connection so that it's possible to |
964 | have active cursors when doing an insert. |
965 | |
966 | When using C<DBIx::Class::Storage::DBI::Sybase::ASE::NoBindVars> transactions |
c1e5a9ac |
967 | are unnecessary and not used, as there are no concurrency issues with C<SELECT |
968 | @@IDENTITY> which is a session variable. |
057db5ce |
969 | |
970 | =head1 TRANSACTIONS |
971 | |
c1e5a9ac |
972 | Due to limitations of the TDS protocol and L<DBD::Sybase>, you cannot begin a |
973 | transaction while there are active cursors, nor can you use multiple active |
974 | cursors within a transaction. An active cursor is, for example, a |
057db5ce |
975 | L<ResultSet|DBIx::Class::ResultSet> that has been executed using C<next> or |
976 | C<first> but has not been exhausted or L<reset|DBIx::Class::ResultSet/reset>. |
977 | |
978 | For example, this will not work: |
979 | |
980 | $schema->txn_do(sub { |
981 | my $rs = $schema->resultset('Book'); |
982 | while (my $row = $rs->next) { |
983 | $schema->resultset('MetaData')->create({ |
984 | book_id => $row->id, |
985 | ... |
986 | }); |
987 | } |
988 | }); |
989 | |
990 | This won't either: |
991 | |
992 | my $first_row = $large_rs->first; |
993 | $schema->txn_do(sub { ... }); |
994 | |
995 | Transactions done for inserts in C<AutoCommit> mode when placeholders are in use |
996 | are not affected, as they are done on an extra database handle. |
997 | |
998 | Some workarounds: |
999 | |
1000 | =over 4 |
1001 | |
1002 | =item * use L<DBIx::Class::Storage::DBI::Replicated> |
1003 | |
1004 | =item * L<connect|DBIx::Class::Schema/connect> another L<Schema|DBIx::Class::Schema> |
1005 | |
1006 | =item * load the data from your cursor with L<DBIx::Class::ResultSet/all> |
1007 | |
1008 | =back |
1009 | |
1010 | =head1 MAXIMUM CONNECTIONS |
1011 | |
1012 | The TDS protocol makes separate connections to the server for active statements |
1013 | in the background. By default the number of such connections is limited to 25, |
1014 | on both the client side and the server side. |
1015 | |
1016 | This is a bit too low for a complex L<DBIx::Class> application, so on connection |
1017 | the client side setting is set to C<256> (see L<DBD::Sybase/maxConnect>.) You |
1018 | can override it to whatever setting you like in the DSN. |
1019 | |
1020 | See |
1021 | L<http://infocenter.sybase.com/help/index.jsp?topic=/com.sybase.help.ase_15.0.sag1/html/sag1/sag1272.htm> |
1022 | for information on changing the setting on the server side. |
1023 | |
1024 | =head1 DATES |
1025 | |
1026 | See L</connect_call_datetime_setup> to setup date formats |
1027 | for L<DBIx::Class::InflateColumn::DateTime>. |
1028 | |
1029 | =head1 TEXT/IMAGE COLUMNS |
1030 | |
1031 | L<DBD::Sybase> compiled with FreeTDS will B<NOT> allow you to insert or update |
1032 | C<TEXT/IMAGE> columns. |
1033 | |
1034 | Setting C<< $dbh->{LongReadLen} >> will also not work with FreeTDS use either: |
1035 | |
1036 | $schema->storage->dbh->do("SET TEXTSIZE $bytes"); |
1037 | |
1038 | or |
1039 | |
1040 | $schema->storage->set_textsize($bytes); |
1041 | |
1042 | instead. |
1043 | |
1044 | However, the C<LongReadLen> you pass in |
8384a713 |
1045 | L<connect_info|DBIx::Class::Storage::DBI/connect_info> is used to execute the |
1046 | equivalent C<SET TEXTSIZE> command on connection. |
057db5ce |
1047 | |
8384a713 |
1048 | See L</connect_call_blob_setup> for a |
1049 | L<connect_info|DBIx::Class::Storage::DBI/connect_info> setting you need to work |
1050 | with C<IMAGE> columns. |
057db5ce |
1051 | |
1052 | =head1 BULK API |
1053 | |
1054 | The experimental L<DBD::Sybase> Bulk API support is used for |
1055 | L<populate|DBIx::Class::ResultSet/populate> in B<void> context, in a transaction |
1056 | on a separate connection. |
1057 | |
1058 | To use this feature effectively, use a large number of rows for each |
1059 | L<populate|DBIx::Class::ResultSet/populate> call, eg.: |
1060 | |
1061 | while (my $rows = $data_source->get_100_rows()) { |
1062 | $rs->populate($rows); |
1063 | } |
1064 | |
1065 | B<NOTE:> the L<add_columns|DBIx::Class::ResultSource/add_columns> |
1066 | calls in your C<Result> classes B<must> list columns in database order for this |
1067 | to work. Also, you may have to unset the C<LANG> environment variable before |
c1e5a9ac |
1068 | loading your app, as C<BCP -Y> is not yet supported in DBD::Sybase . |
057db5ce |
1069 | |
1070 | When inserting IMAGE columns using this method, you'll need to use |
1071 | L</connect_call_blob_setup> as well. |
1072 | |
6476fd66 |
1073 | =head1 COMPUTED COLUMNS |
1074 | |
1075 | If you have columns such as: |
1076 | |
1077 | created_dtm AS getdate() |
1078 | |
1079 | represent them in your Result classes as: |
1080 | |
1081 | created_dtm => { |
1082 | data_type => undef, |
1083 | default_value => \'getdate()', |
1084 | is_nullable => 0, |
c1e5a9ac |
1085 | inflate_datetime => 1, |
6476fd66 |
1086 | } |
1087 | |
1088 | The C<data_type> must exist and must be C<undef>. Then empty inserts will work |
1089 | on tables with such columns. |
1090 | |
1091 | =head1 TIMESTAMP COLUMNS |
1092 | |
1093 | C<timestamp> columns in Sybase ASE are not really timestamps, see: |
1094 | L<http://dba.fyicenter.com/Interview-Questions/SYBASE/The_timestamp_datatype_in_Sybase_.html>. |
1095 | |
1096 | They should be defined in your Result classes as: |
1097 | |
1098 | ts => { |
1099 | data_type => 'timestamp', |
1100 | is_nullable => 0, |
1101 | inflate_datetime => 0, |
1102 | } |
1103 | |
1104 | The C<<inflate_datetime => 0>> is necessary if you use |
1105 | L<DBIx::Class::InflateColumn::DateTime>, and most people do, and still want to |
1106 | be able to read these values. |
1107 | |
1108 | The values will come back as hexadecimal. |
1109 | |
057db5ce |
1110 | =head1 TODO |
1111 | |
1112 | =over |
1113 | |
1114 | =item * |
1115 | |
1116 | Transitions to AutoCommit=0 (starting a transaction) mode by exhausting |
1117 | any active cursors, using eager cursors. |
1118 | |
1119 | =item * |
1120 | |
1121 | Real limits and limited counts using stored procedures deployed on startup. |
1122 | |
1123 | =item * |
1124 | |
057db5ce |
1125 | Blob update with a LIKE query on a blob, without invalidating the WHERE condition. |
1126 | |
1127 | =item * |
1128 | |
1129 | bulk_insert using prepare_cached (see comments.) |
1130 | |
1131 | =back |
1132 | |
1133 | =head1 AUTHOR |
1134 | |
8138f96a |
1135 | See L<DBIx::Class/AUTHOR> and L<DBIx::Class/CONTRIBUTORS>. |
057db5ce |
1136 | |
1137 | =head1 LICENSE |
1138 | |
1139 | You may distribute this code under the same terms as Perl itself. |
1140 | |
1141 | =cut |
1142 | # vim:sts=2 sw=2: |