Horrible horrible rewrite of the aliastype scanner, but folks are starting to complai...
[dbsrgits/DBIx-Class.git] / lib / DBIx / Class / Storage / DBIHacks.pm
CommitLineData
c443438f 1package #hide from PAUSE
2 DBIx::Class::Storage::DBIHacks;
d28bb90d 3
4#
5# This module contains code that should never have seen the light of day,
6# does not belong in the Storage, or is otherwise unfit for public
7eb76996 7# display. The arrival of SQLA2 should immediately oboslete 90% of this
d28bb90d 8#
9
10use strict;
11use warnings;
12
13use base 'DBIx::Class::Storage';
14use mro 'c3';
15
16use Carp::Clan qw/^DBIx::Class/;
17
18#
052e8431 19# This code will remove non-selecting/non-restricting joins from
4b1b5ea3 20# {from} specs, aiding the RDBMS query optimizer
052e8431 21#
22sub _prune_unused_joins {
4b1b5ea3 23 my ($self) = shift;
24
25 my ($from, $select, $where, $attrs) = @_;
052e8431 26
052e8431 27 if (ref $from ne 'ARRAY' || ref $from->[0] ne 'HASH' || ref $from->[1] ne 'ARRAY') {
28 return $from; # only standard {from} specs are supported
29 }
30
4b1b5ea3 31 my $aliastypes = $self->_resolve_aliastypes_from_select_args(@_);
32
33 # a grouped set will not be affected by amount of rows. Thus any
34 # {multiplying} joins can go
35 delete $aliastypes->{multiplying} if $attrs->{group_by};
36
052e8431 37
38 my @newfrom = $from->[0]; # FROM head is always present
39
40 my %need_joins = (map { %{$_||{}} } (values %$aliastypes) );
41 for my $j (@{$from}[1..$#$from]) {
539ffe87 42 push @newfrom, $j if (
4b1b5ea3 43 (! $j->[0]{-alias}) # legacy crap
539ffe87 44 ||
45 $need_joins{$j->[0]{-alias}}
46 );
052e8431 47 }
48
49 return \@newfrom;
50}
51
052e8431 52#
d28bb90d 53# This is the code producing joined subqueries like:
54# SELECT me.*, other.* FROM ( SELECT me.* FROM ... ) JOIN other ON ...
55#
56sub _adjust_select_args_for_complex_prefetch {
57 my ($self, $from, $select, $where, $attrs) = @_;
58
59 $self->throw_exception ('Nothing to prefetch... how did we get here?!')
60 if not @{$attrs->{_prefetch_select}};
61
62 $self->throw_exception ('Complex prefetches are not supported on resultsets with a custom from attribute')
63 if (ref $from ne 'ARRAY' || ref $from->[0] ne 'HASH' || ref $from->[1] ne 'ARRAY');
64
65
66 # generate inner/outer attribute lists, remove stuff that doesn't apply
67 my $outer_attrs = { %$attrs };
68 delete $outer_attrs->{$_} for qw/where bind rows offset group_by having/;
69
70 my $inner_attrs = { %$attrs };
71 delete $inner_attrs->{$_} for qw/for collapse _prefetch_select _collapse_order_by select as/;
72
73
74 # bring over all non-collapse-induced order_by into the inner query (if any)
75 # the outer one will have to keep them all
76 delete $inner_attrs->{order_by};
77 if (my $ord_cnt = @{$outer_attrs->{order_by}} - @{$outer_attrs->{_collapse_order_by}} ) {
78 $inner_attrs->{order_by} = [
79 @{$outer_attrs->{order_by}}[ 0 .. $ord_cnt - 1]
80 ];
81 }
82
d28bb90d 83 # generate the inner/outer select lists
84 # for inside we consider only stuff *not* brought in by the prefetch
85 # on the outside we substitute any function for its alias
86 my $outer_select = [ @$select ];
87 my $inner_select = [];
88 for my $i (0 .. ( @$outer_select - @{$outer_attrs->{_prefetch_select}} - 1) ) {
89 my $sel = $outer_select->[$i];
90
91 if (ref $sel eq 'HASH' ) {
92 $sel->{-as} ||= $attrs->{as}[$i];
93 $outer_select->[$i] = join ('.', $attrs->{alias}, ($sel->{-as} || "inner_column_$i") );
94 }
95
96 push @$inner_select, $sel;
97 }
98
d28bb90d 99 # construct the inner $from for the subquery
48580715 100 # we need to prune first, because this will determine if we need a group_by below
052e8431 101 my $inner_from = $self->_prune_unused_joins ($from, $inner_select, $where, $inner_attrs);
ad630f4b 102
539ffe87 103 # if a multi-type join was needed in the subquery - add a group_by to simulate the
104 # collapse in the subq
105 $inner_attrs->{group_by} ||= $inner_select
106 if List::Util::first
107 { ! $_->[0]{-is_single} }
108 (@{$inner_from}[1 .. $#$inner_from])
109 ;
d28bb90d 110
d28bb90d 111 # generate the subquery
112 my $subq = $self->_select_args_to_query (
052e8431 113 $inner_from,
d28bb90d 114 $inner_select,
115 $where,
116 $inner_attrs,
117 );
118
119 my $subq_joinspec = {
120 -alias => $attrs->{alias},
052e8431 121 -source_handle => $inner_from->[0]{-source_handle},
d28bb90d 122 $attrs->{alias} => $subq,
123 };
124
125 # Generate the outer from - this is relatively easy (really just replace
126 # the join slot with the subquery), with a major caveat - we can not
127 # join anything that is non-selecting (not part of the prefetch), but at
128 # the same time is a multi-type relationship, as it will explode the result.
129 #
130 # There are two possibilities here
131 # - either the join is non-restricting, in which case we simply throw it away
132 # - it is part of the restrictions, in which case we need to collapse the outer
133 # result by tackling yet another group_by to the outside of the query
134
052e8431 135 # normalize a copy of $from, so it will be easier to work with further
136 # down (i.e. promote the initial hashref to an AoH)
137 $from = [ @$from ];
138 $from->[0] = [ $from->[0] ];
139
d28bb90d 140 # so first generate the outer_from, up to the substitution point
141 my @outer_from;
142 while (my $j = shift @$from) {
143 if ($j->[0]{-alias} eq $attrs->{alias}) { # time to swap
144 push @outer_from, [
145 $subq_joinspec,
146 @{$j}[1 .. $#$j],
147 ];
148 last; # we'll take care of what's left in $from below
149 }
150 else {
151 push @outer_from, $j;
152 }
153 }
154
052e8431 155 # scan the from spec against different attributes, and see which joins are needed
156 # in what role
157 my $outer_aliastypes =
539ffe87 158 $self->_resolve_aliastypes_from_select_args( $from, $outer_select, $where, $outer_attrs );
052e8431 159
d28bb90d 160 # see what's left - throw away if not selecting/restricting
161 # also throw in a group_by if restricting to guard against
162 # cross-join explosions
163 #
164 while (my $j = shift @$from) {
165 my $alias = $j->[0]{-alias};
166
1a736efb 167 if ($outer_aliastypes->{selecting}{$alias}) {
d28bb90d 168 push @outer_from, $j;
169 }
1a736efb 170 elsif ($outer_aliastypes->{restricting}{$alias}) {
d28bb90d 171 push @outer_from, $j;
539ffe87 172 $outer_attrs->{group_by} ||= $outer_select unless $j->[0]{-is_single};
d28bb90d 173 }
174 }
175
176 # demote the outer_from head
177 $outer_from[0] = $outer_from[0][0];
178
179 # This is totally horrific - the $where ends up in both the inner and outer query
180 # Unfortunately not much can be done until SQLA2 introspection arrives, and even
181 # then if where conditions apply to the *right* side of the prefetch, you may have
182 # to both filter the inner select (e.g. to apply a limit) and then have to re-filter
183 # the outer select to exclude joins you didin't want in the first place
184 #
185 # OTOH it can be seen as a plus: <ash> (notes that this query would make a DBA cry ;)
186 return (\@outer_from, $outer_select, $where, $outer_attrs);
187}
188
1a736efb 189#
190# I KNOW THIS SUCKS! GET SQLA2 OUT THE DOOR SO THIS CAN DIE!
191#
ad630f4b 192# Due to a lack of SQLA2 we fall back to crude scans of all the
193# select/where/order/group attributes, in order to determine what
194# aliases are neded to fulfill the query. This information is used
195# throughout the code to prune unnecessary JOINs from the queries
196# in an attempt to reduce the execution time.
197# Although the method is pretty horrific, the worst thing that can
1a736efb 198# happen is for it to fail due to some scalar SQL, which in turn will
199# result in a vocal exception.
539ffe87 200sub _resolve_aliastypes_from_select_args {
052e8431 201 my ( $self, $from, $select, $where, $attrs ) = @_;
546f1cd9 202
ad630f4b 203 $self->throw_exception ('Unable to analyze custom {from}')
204 if ref $from ne 'ARRAY';
546f1cd9 205
ad630f4b 206 # what we will return
964a3c71 207 my $aliases_by_type;
546f1cd9 208
ad630f4b 209 # see what aliases are there to work with
210 my $alias_list;
539ffe87 211 for (@$from) {
212 my $j = $_;
ad630f4b 213 $j = $j->[0] if ref $j eq 'ARRAY';
539ffe87 214 my $al = $j->{-alias}
215 or next;
216
217 $alias_list->{$al} = $j;
218 $aliases_by_type->{multiplying}{$al} = 1
219 unless $j->{-is_single};
546f1cd9 220 }
546f1cd9 221
1a736efb 222 # get a column to source/alias map (including unqualified ones)
223 my $colinfo = $self->_resolve_column_info ($from);
224
ad630f4b 225 # set up a botched SQLA
226 my $sql_maker = $self->sql_maker;
227 my $sep = quotemeta ($self->_sql_maker_opts->{name_sep} || '.');
07f31d19 228
1a736efb 229 my ($orig_lquote, $orig_rquote) = map { quotemeta $_ } (do {
230 if (ref $sql_maker->{quote_char} eq 'ARRAY') {
231 @{$sql_maker->{quote_char}}
232 }
233 else {
234 ($sql_maker->{quote_char} || '') x 2;
235 }
ad630f4b 236 });
07f31d19 237
1a736efb 238 local $sql_maker->{quote_char} = "\x00"; # so that we can regex away
239
240 # generate sql chunks
241 my $to_scan = {
242 restricting => [
243 $sql_maker->_recurse_where ($where),
244 $sql_maker->_order_by({
245 map { $_ => $attrs->{$_} } (qw/group_by having/)
246 }),
247 ],
248 selecting => [
249 $self->_parse_order_by ($attrs->{order_by}, $sql_maker),
250 $sql_maker->_recurse_fields ($select),
251 ],
252 };
253
254 # throw away empty chunks
255 $_ = [ map { $_ || () } @$_ ] for values %$to_scan;
256
257 # first loop through all fully qualified columns and get the corresponding
258 # alias (should work even if they are in scalarrefs)
ad630f4b 259 for my $alias (keys %$alias_list) {
1a736efb 260 my $al_re = qr/
261 \x00 $alias \x00 $sep
262 |
263 \b $alias $sep
264 /x;
265
266 # add matching for possible quoted literal sql
267 $al_re = qr/ $al_re | $orig_lquote $alias $orig_rquote /x
268 if ($orig_lquote && $orig_rquote);
07f31d19 269
1a736efb 270
271 for my $type (keys %$to_scan) {
272 for my $piece (@{$to_scan->{$type}}) {
273 $aliases_by_type->{$type}{$alias} = 1 if ($piece =~ $al_re);
274 }
ad630f4b 275 }
07f31d19 276
1a736efb 277 }
278
279 # now loop through unqualified column names, and try to locate them within
280 # the chunks
281 for my $col (keys %$colinfo) {
282 next if $col =~ $sep; # if column is qualified it was caught by the above
283
284 my $col_re = qr/ \x00 $col \x00 /x;
285
286 $col_re = qr/ $col_re | $orig_lquote $col $orig_rquote /x
287 if ($orig_lquote && $orig_rquote);
288
289 for my $type (keys %$to_scan) {
290 for my $piece (@{$to_scan->{$type}}) {
291 $aliases_by_type->{$type}{$colinfo->{$col}{-source_alias}} = 1 if ($piece =~ $col_re);
292 }
07f31d19 293 }
294 }
295
296 # Add any non-left joins to the restriction list (such joins are indeed restrictions)
ad630f4b 297 for my $j (values %$alias_list) {
07f31d19 298 my $alias = $j->{-alias} or next;
1a736efb 299 $aliases_by_type->{restricting}{$alias} = 1 if (
07f31d19 300 (not $j->{-join_type})
301 or
302 ($j->{-join_type} !~ /^left (?: \s+ outer)? $/xi)
303 );
304 }
305
306 # mark all join parents as mentioned
307 # (e.g. join => { cds => 'tracks' } - tracks will need to bring cds too )
964a3c71 308 for my $type (keys %$aliases_by_type) {
309 for my $alias (keys %{$aliases_by_type->{$type}}) {
310 $aliases_by_type->{$type}{$_} = 1
38f42d85 311 for (map { keys %$_ } @{ $alias_list->{$alias}{-join_path} || [] });
07f31d19 312 }
313 }
ad630f4b 314
964a3c71 315 return $aliases_by_type;
07f31d19 316}
317
d28bb90d 318sub _resolve_ident_sources {
319 my ($self, $ident) = @_;
320
321 my $alias2source = {};
322 my $rs_alias;
323
324 # the reason this is so contrived is that $ident may be a {from}
325 # structure, specifying multiple tables to join
326 if ( Scalar::Util::blessed($ident) && $ident->isa("DBIx::Class::ResultSource") ) {
327 # this is compat mode for insert/update/delete which do not deal with aliases
328 $alias2source->{me} = $ident;
329 $rs_alias = 'me';
330 }
331 elsif (ref $ident eq 'ARRAY') {
332
333 for (@$ident) {
334 my $tabinfo;
335 if (ref $_ eq 'HASH') {
336 $tabinfo = $_;
337 $rs_alias = $tabinfo->{-alias};
338 }
339 if (ref $_ eq 'ARRAY' and ref $_->[0] eq 'HASH') {
340 $tabinfo = $_->[0];
341 }
342
343 $alias2source->{$tabinfo->{-alias}} = $tabinfo->{-source_handle}->resolve
344 if ($tabinfo->{-source_handle});
345 }
346 }
347
348 return ($alias2source, $rs_alias);
349}
350
351# Takes $ident, \@column_names
352#
353# returns { $column_name => \%column_info, ... }
354# also note: this adds -result_source => $rsrc to the column info
355#
09e14fdc 356# If no columns_names are supplied returns info about *all* columns
357# for all sources
d28bb90d 358sub _resolve_column_info {
359 my ($self, $ident, $colnames) = @_;
360 my ($alias2src, $root_alias) = $self->_resolve_ident_sources($ident);
361
362 my $sep = $self->_sql_maker_opts->{name_sep} || '.';
09e14fdc 363 my $qsep = quotemeta $sep;
d28bb90d 364
09e14fdc 365 my (%return, %seen_cols, @auto_colnames);
d28bb90d 366
367 # compile a global list of column names, to be able to properly
368 # disambiguate unqualified column names (if at all possible)
369 for my $alias (keys %$alias2src) {
370 my $rsrc = $alias2src->{$alias};
371 for my $colname ($rsrc->columns) {
372 push @{$seen_cols{$colname}}, $alias;
09e14fdc 373 push @auto_colnames, "$alias$sep$colname" unless $colnames;
d28bb90d 374 }
375 }
376
09e14fdc 377 $colnames ||= [
378 @auto_colnames,
379 grep { @{$seen_cols{$_}} == 1 } (keys %seen_cols),
380 ];
381
d28bb90d 382 COLUMN:
383 foreach my $col (@$colnames) {
09e14fdc 384 my ($alias, $colname) = $col =~ m/^ (?: ([^$qsep]+) $qsep)? (.+) $/x;
d28bb90d 385
386 unless ($alias) {
387 # see if the column was seen exactly once (so we know which rsrc it came from)
388 if ($seen_cols{$colname} and @{$seen_cols{$colname}} == 1) {
389 $alias = $seen_cols{$colname}[0];
390 }
391 else {
392 next COLUMN;
393 }
394 }
395
396 my $rsrc = $alias2src->{$alias};
397 $return{$col} = $rsrc && {
398 %{$rsrc->column_info($colname)},
399 -result_source => $rsrc,
400 -source_alias => $alias,
401 };
402 }
403
404 return \%return;
405}
406
289ac713 407# The DBIC relationship chaining implementation is pretty simple - every
408# new related_relationship is pushed onto the {from} stack, and the {select}
409# window simply slides further in. This means that when we count somewhere
410# in the middle, we got to make sure that everything in the join chain is an
411# actual inner join, otherwise the count will come back with unpredictable
412# results (a resultset may be generated with _some_ rows regardless of if
413# the relation which the $rs currently selects has rows or not). E.g.
414# $artist_rs->cds->count - normally generates:
415# SELECT COUNT( * ) FROM artist me LEFT JOIN cd cds ON cds.artist = me.artistid
416# which actually returns the number of artists * (number of cds || 1)
417#
418# So what we do here is crawl {from}, determine if the current alias is at
419# the top of the stack, and if not - make sure the chain is inner-joined down
420# to the root.
421#
422sub _straight_join_to_node {
423 my ($self, $from, $alias) = @_;
424
425 # subqueries and other oddness are naturally not supported
426 return $from if (
427 ref $from ne 'ARRAY'
428 ||
429 @$from <= 1
430 ||
431 ref $from->[0] ne 'HASH'
432 ||
433 ! $from->[0]{-alias}
434 ||
7eb76996 435 $from->[0]{-alias} eq $alias # this last bit means $alias is the head of $from - nothing to do
289ac713 436 );
437
438 # find the current $alias in the $from structure
439 my $switch_branch;
440 JOINSCAN:
441 for my $j (@{$from}[1 .. $#$from]) {
442 if ($j->[0]{-alias} eq $alias) {
443 $switch_branch = $j->[0]{-join_path};
444 last JOINSCAN;
445 }
446 }
447
7eb76996 448 # something else went quite wrong
289ac713 449 return $from unless $switch_branch;
450
451 # So it looks like we will have to switch some stuff around.
452 # local() is useless here as we will be leaving the scope
453 # anyway, and deep cloning is just too fucking expensive
7eb76996 454 # So replace the first hashref in the node arrayref manually
289ac713 455 my @new_from = ($from->[0]);
38f42d85 456 my $sw_idx = { map { values %$_ => 1 } @$switch_branch };
289ac713 457
458 for my $j (@{$from}[1 .. $#$from]) {
459 my $jalias = $j->[0]{-alias};
460
461 if ($sw_idx->{$jalias}) {
462 my %attrs = %{$j->[0]};
463 delete $attrs{-join_type};
464 push @new_from, [
465 \%attrs,
466 @{$j}[ 1 .. $#$j ],
467 ];
468 }
469 else {
470 push @new_from, $j;
471 }
472 }
473
474 return \@new_from;
475}
476
bac6c4fb 477# Most databases do not allow aliasing of tables in UPDATE/DELETE. Thus
478# a condition containing 'me' or other table prefixes will not work
479# at all. What this code tries to do (badly) is introspect the condition
480# and remove all column qualifiers. If it bails out early (returns undef)
481# the calling code should try another approach (e.g. a subquery)
482sub _strip_cond_qualifiers {
483 my ($self, $where) = @_;
484
485 my $cond = {};
486
487 # No-op. No condition, we're updating/deleting everything
488 return $cond unless $where;
489
490 if (ref $where eq 'ARRAY') {
491 $cond = [
492 map {
493 my %hash;
494 foreach my $key (keys %{$_}) {
495 $key =~ /([^.]+)$/;
496 $hash{$1} = $_->{$key};
497 }
498 \%hash;
499 } @$where
500 ];
501 }
502 elsif (ref $where eq 'HASH') {
503 if ( (keys %$where) == 1 && ( (keys %{$where})[0] eq '-and' )) {
504 $cond->{-and} = [];
505 my @cond = @{$where->{-and}};
506 for (my $i = 0; $i < @cond; $i++) {
507 my $entry = $cond[$i];
508 my $hash;
037e8dca 509 my $ref = ref $entry;
510 if ($ref eq 'HASH' or $ref eq 'ARRAY') {
bac6c4fb 511 $hash = $self->_strip_cond_qualifiers($entry);
512 }
037e8dca 513 elsif (! $ref) {
bac6c4fb 514 $entry =~ /([^.]+)$/;
515 $hash->{$1} = $cond[++$i];
516 }
037e8dca 517 else {
518 $self->throw_exception ("_strip_cond_qualifiers() is unable to handle a condition reftype $ref");
519 }
bac6c4fb 520 push @{$cond->{-and}}, $hash;
521 }
522 }
523 else {
524 foreach my $key (keys %$where) {
525 $key =~ /([^.]+)$/;
526 $cond->{$1} = $where->{$key};
527 }
528 }
529 }
530 else {
531 return undef;
532 }
533
534 return $cond;
535}
536
c0748280 537sub _parse_order_by {
1a736efb 538 my ($self, $order_by, $sql_maker) = @_;
c0748280 539
1a736efb 540 my $parser = sub {
541 my ($sql_maker, $order_by) = @_;
c0748280 542
1a736efb 543 return scalar $sql_maker->_order_by_chunks ($order_by)
544 unless wantarray;
c0748280 545
1a736efb 546 my @chunks;
547 for my $chunk (map { ref $_ ? @$_ : $_ } ($sql_maker->_order_by_chunks ($order_by) ) ) {
548 $chunk =~ s/\s+ (?: ASC|DESC ) \s* $//ix;
549 push @chunks, $chunk;
550 }
551
552 return @chunks;
553 };
554
555 if ($sql_maker) {
556 return $parser->($sql_maker, $order_by);
557 }
558 else {
559 $sql_maker = $self->sql_maker;
560 local $sql_maker->{quote_char};
561 return $parser->($sql_maker, $order_by);
562 }
c0748280 563}
bac6c4fb 564
d28bb90d 5651;