1 package # hide from the pauses
2 DBIx::Class::ResultSource::RowParser;
8 use List::Util 'first';
13 use base 'DBIx::Class';
15 # Accepts one or more relationships for the current source and returns an
16 # array of column names for each of those relationships. Column names are
17 # prefixed relative to the current source, in accordance with where they appear
18 # in the supplied relationships.
19 sub _resolve_prefetch {
20 my ($self, $pre, $alias, $alias_map, $order, $pref_path) = @_;
23 if (not defined $pre or not length $pre) {
26 elsif( ref $pre eq 'ARRAY' ) {
28 map { $self->_resolve_prefetch( $_, $alias, $alias_map, $order, [ @$pref_path ] ) }
31 elsif( ref $pre eq 'HASH' ) {
34 $self->_resolve_prefetch($_, $alias, $alias_map, $order, [ @$pref_path ] ),
35 $self->related_source($_)->_resolve_prefetch(
36 $pre->{$_}, "${alias}.$_", $alias_map, $order, [ @$pref_path, $_] )
41 $self->throw_exception(
42 "don't know how to resolve prefetch reftype ".ref($pre));
46 $p = $p->{$_} for (@$pref_path, $pre);
48 $self->throw_exception (
49 "Unable to resolve prefetch '$pre' - join alias map does not contain an entry for path: "
50 . join (' -> ', @$pref_path, $pre)
51 ) if (ref $p->{-join_aliases} ne 'ARRAY' or not @{$p->{-join_aliases}} );
53 my $as = shift @{$p->{-join_aliases}};
55 my $rel_info = $self->relationship_info( $pre );
56 $self->throw_exception( $self->source_name . " has no such relationship '$pre'" )
59 my $as_prefix = ($alias =~ /^.*?\.(.+)$/ ? $1.'.' : '');
61 return map { [ "${as}.$_", "${as_prefix}${pre}.$_", ] }
62 $self->related_source($pre)->columns;
66 # Takes an arrayref selection list and generates a collapse-map representing
67 # row-object fold-points. Every relationship is assigned a set of unique,
68 # non-nullable columns (which may *not even be* from the same resultset)
69 # and the collapser will use this information to correctly distinguish
70 # data of individual to-be-row-objects. See t/resultset/rowparser_internals.t
71 # for extensive RV examples
72 sub _resolve_collapse {
73 my ($self, $args, $common_args) = @_;
75 # for comprehensible error messages put ourselves at the head of the relationship chain
76 $args->{_rel_chain} ||= [ $self->source_name ];
78 # record top-level fully-qualified column index, start nodecount
80 _as_fq_idx => { %{$args->{as}} },
81 _node_idx => 1, # this is *deliberately* not 0-based
84 my ($my_cols, $rel_cols);
85 for (keys %{$args->{as}}) {
86 if ($_ =~ /^ ([^\.]+) \. (.+) /x) {
87 $rel_cols->{$1}{$2} = 1;
90 $my_cols->{$_} = {}; # important for ||= below
95 # run through relationships, collect metadata, inject non-left fk-bridges from
96 # *INNER-JOINED* children (if any)
97 for my $rel (keys %$rel_cols) {
98 my $rel_src = __get_related_source($self, $rel, $rel_cols->{$rel});
100 my $inf = $self->relationship_info ($rel);
102 $relinfo->{$rel}{is_single} = $inf->{attrs}{accessor} && $inf->{attrs}{accessor} ne 'multi';
103 $relinfo->{$rel}{is_inner} = ( $inf->{attrs}{join_type} || '' ) !~ /^left/i;
104 $relinfo->{$rel}{rsrc} = $rel_src;
106 my $cond = $inf->{cond};
113 ! first { $_ !~ /^foreign\./ } (keys %$cond)
115 ! first { $_ !~ /^self\./ } (values %$cond)
117 for my $f (keys %$cond) {
119 $_ =~ s/^ (?: foreign | self ) \.//x for ($f, $s);
120 $relinfo->{$rel}{fk_map}{$s} = $f;
122 # need to know source from *our* pov, hence $rel.
123 $my_cols->{$s} ||= { via_fk => "$rel.$f" } if (
124 defined $rel_cols->{$rel}{$f} # in fact selected
126 $relinfo->{$rel}{is_inner}
132 # if the parent is already defined, assume all of its related FKs are selected
133 # (even if they in fact are NOT in the select list). Keep a record of what we
134 # assumed, and if any such phantom-column becomes part of our own collapser,
135 # throw everything assumed-from-parent away and replace with the collapser of
136 # the parent (whatever it may be)
137 my $assumed_from_parent;
138 unless ($args->{_parent_info}{underdefined}) {
139 $assumed_from_parent->{columns} = { map
140 # only add to the list if we do not already select said columns
141 { ! exists $my_cols->{$_} ? ( $_ => 1 ) : () }
142 values %{$args->{_parent_info}{rel_condition} || {}}
145 $my_cols->{$_} = { via_collapse => $args->{_parent_info}{collapse_on_idcols} }
146 for keys %{$assumed_from_parent->{columns}};
149 # get colinfo for everything
151 my $ci = $self->columns_info;
152 $my_cols->{$_}{colinfo} = $ci->{$_} for keys %$my_cols;
157 # first try to reuse the parent's collapser (i.e. reuse collapser over 1:1)
158 # (makes for a leaner coderef later)
159 unless ($collapse_map->{-idcols_current_node}) {
160 $collapse_map->{-idcols_current_node} = $args->{_parent_info}{collapse_on_idcols}
161 if $args->{_parent_info}{collapser_reusable};
165 # Still dont know how to collapse - try to resolve based on our columns (plus already inserted FK bridges)
167 ! $collapse_map->{-idcols_current_node}
171 my $idset = $self->_identifying_column_set ({map { $_ => $my_cols->{$_}{colinfo} } keys %$my_cols})
173 # see if the resulting collapser relies on any implied columns,
174 # and fix stuff up if this is the case
175 my @reduced_set = grep { ! $assumed_from_parent->{columns}{$_} } @$idset;
177 $collapse_map->{-idcols_current_node} = [ __unique_numlist(
178 @{ $args->{_parent_info}{collapse_on_idcols}||[] },
183 @{$args->{_rel_chain}}[1 .. $#{$args->{_rel_chain}}],
184 ( $my_cols->{$_}{via_fk} || $_ ),
187 $common_args->{_as_fq_idx}->{$fqc};
194 # Stil don't know how to collapse - keep descending down 1:1 chains - if
195 # a related non-LEFT 1:1 is resolvable - its condition will collapse us
197 unless ($collapse_map->{-idcols_current_node}) {
200 for my $rel (keys %$relinfo) {
201 next unless ($relinfo->{$rel}{is_single} && $relinfo->{$rel}{is_inner});
203 if ( my $rel_collapse = $relinfo->{$rel}{rsrc}->_resolve_collapse ({
204 as => $rel_cols->{$rel},
205 _rel_chain => [ @{$args->{_rel_chain}}, $rel ],
206 _parent_info => { underdefined => 1 },
208 push @candidates, $rel_collapse->{-idcols_current_node};
212 # get the set with least amount of columns
213 # FIXME - maybe need to implement a data type order as well (i.e. prefer several ints
214 # to a single varchar)
216 ($collapse_map->{-idcols_current_node}) = sort { scalar @$a <=> scalar @$b } (@candidates);
220 # stop descending into children if we were called by a parent for first-pass
221 # and don't despair if nothing was found (there may be other parallel branches
223 if ($args->{_parent_info}{underdefined}) {
224 return $collapse_map->{-idcols_current_node} ? $collapse_map : undef
226 # nothing down the chain resolved - can't calculate a collapse-map
227 elsif (! $collapse_map->{-idcols_current_node}) {
228 $self->throw_exception ( sprintf
229 "Unable to calculate a definitive collapse column set for %s%s: fetch more unique non-nullable columns",
231 @{$args->{_rel_chain}} > 1
232 ? sprintf (' (last member of the %s chain)', join ' -> ', @{$args->{_rel_chain}} )
238 # If we got that far - we are collapsable - GREAT! Now go down all children
239 # a second time, and fill in the rest
241 $collapse_map->{-is_optional} = 1 if $args->{_parent_info}{is_optional};
242 $collapse_map->{-node_index} = $common_args->{_node_idx}++;
246 for my $rel (sort keys %$relinfo) {
248 $collapse_map->{$rel} = $relinfo->{$rel}{rsrc}->_resolve_collapse ({
249 as => { map { $_ => 1 } ( keys %{$rel_cols->{$rel}} ) },
250 _rel_chain => [ @{$args->{_rel_chain}}, $rel],
253 collapse_on_idcols => [ @{$collapse_map->{-idcols_current_node}} ],
255 rel_condition => $relinfo->{$rel}{fk_map},
257 is_optional => $collapse_map->{-is_optional},
259 # if this is a 1:1 our own collapser can be used as a collapse-map
260 # (regardless of left or not)
261 collapser_reusable => $relinfo->{$rel}{is_single},
265 $collapse_map->{$rel}{-is_single} = 1 if $relinfo->{$rel}{is_single};
266 $collapse_map->{$rel}{-is_optional} ||= 1 unless $relinfo->{$rel}{is_inner};
267 push @id_sets, ( map { @$_ } (
268 $collapse_map->{$rel}{-idcols_current_node},
269 $collapse_map->{$rel}{-idcols_extra_from_children} || (),
274 my $cur_nodeid_hash = { map { $_ => 1 } @{$collapse_map->{-idcols_current_node}} };
275 $collapse_map->{-idcols_extra_from_children} = [ grep
276 { ! $cur_nodeid_hash->{$_} }
277 __unique_numlist( @id_sets )
281 return $collapse_map;
284 # Takes an arrayref of {as} dbic column aliases and the collapse and select
285 # attributes from the same $rs (the selector requirement is a temporary
286 # workaround... I hope), and returns a coderef capable of:
287 # my $me_pref_clps = $coderef->([$rs->cursor->next/all])
288 # Where the $me_pref_clps arrayref is the future argument to inflate_result()
290 # For an example of this coderef in action (and to see its guts) look at
291 # t/resultset/rowparser_internals.t
293 # This is a huge performance win, as we call the same code for # every row
294 # returned from the db, thus avoiding repeated method lookups when traversing
297 # Also since the coderef is completely stateless (the returned structure is
298 # always fresh on every new invocation) this is a very good opportunity for
299 # memoization if further speed improvements are needed
301 # The way we construct this coderef is somewhat fugly, although the result is
302 # really worth it. The final coderef does not perform any kind of recursion -
303 # the entire nested structure constructor is rolled out into a single scope.
305 # In any case - the output of this thing is meticulously micro-tested, so
306 # any sort of adjustment/rewrite should be relatively easy (fsvo relatively)
309 my ($self, $args) = @_;
311 my $inflate_index = { map
312 { $args->{inflate_map}[$_] => $_ }
313 ( 0 .. $#{$args->{inflate_map}} )
318 # the non-collapsing assembler is easy
319 # FIXME SUBOPTIMAL there could be a yet faster way to do things here, but
320 # need to try an actual implementation and benchmark it:
322 # <timbunce_> First setup the nested data structure you want for each row
323 # Then call bind_col() to alias the row fields into the right place in
324 # the data structure, then to fetch the data do:
325 # push @rows, dclone($row_data_struct) while ($sth->fetchrow);
327 if (!$args->{collapse}) {
328 $parser_src = sprintf('$_ = %s for @{$_[0]}', __visit_infmap_simple(
330 { rsrc => $self }, # need the $rsrc to sanity-check inflation map once
333 # change the quoted placeholders to unquoted alias-references
334 $parser_src =~ s/ \' \xFF__VALPOS__(\d+)__\xFF \' /"\$_->[$1]"/gex;
337 # the collapsing parser is more complicated - it needs to keep a lot of state
340 my $collapse_map = $self->_resolve_collapse ({
342 # only consider real columns (not functions) during collapse resolution
343 # this check shouldn't really be here, as fucktards are not supposed to
344 # alias random crap to existing column names anyway, but still - just in
346 # FIXME !!!! - this does not yet deal with unbalanced selectors correctly
347 # (it is now trivial as the attrs specify where things go out of sync
350 { ref $args->{selection}[$inflate_index->{$_}] ? () : ( $_ => $inflate_index->{$_} ) }
355 my $all_idcols_as_list = join ', ', sort map { @$_ } (
356 $collapse_map->{-idcols_current_node},
357 $collapse_map->{-idcols_extra_from_children} || (),
360 my $top_node_id_path = join ('', map
361 { "{'\xFF__IDVALPOS__${_}__\xFF'}" }
362 @{$collapse_map->{-idcols_current_node}}
365 my $rel_assemblers = __visit_infmap_collapse (
366 $inflate_index, $collapse_map
369 $parser_src = sprintf (<<'EOS', $all_idcols_as_list, $top_node_id_path, $rel_assemblers);
370 ### BEGIN LITERAL STRING EVAL
372 my ($rows_pos, $result_pos, $cur_row, @cur_row_ids, @collapse_idx, $is_new_res) = (0,0);
374 # this loop is a bit arcane - the rationale is that the passed in
375 # $_[0] will either have only one row (->next) or will have all
376 # rows already pulled in (->all and/or unordered). Given that the
377 # result can be rather large - we reuse the same already allocated
378 # array, since the collapsed prefetch is smaller by definition.
379 # At the end we cut the leftovers away and move on.
381 ( ( $rows_pos >= 0 and $_[0][$rows_pos++] ) or do { $rows_pos = -1; undef } )
383 ($_[1] and $_[1]->())
386 # due to left joins some of the ids may be NULL/undef, and
387 # won't play well when used as hash lookups
388 $cur_row_ids[$_] = defined $cur_row->[$_] ? $cur_row->[$_] : "\xFF\xFFN\xFFU\xFFL\xFFL\xFF\xFF"
391 $is_new_res = ! $collapse_idx[1]%2$s and (
392 $_[1] and $result_pos and (unshift @{$_[2]}, $cur_row) and last
397 $_[0][$result_pos++] = $collapse_idx[1]%2$s
401 splice @{$_[0]}, $result_pos; # truncate the passed in array for cases of collapsing ->all()
402 ### END LITERAL STRING EVAL
405 # !!! note - different var than the one above
406 # change the quoted placeholders to unquoted alias-references
407 $parser_src =~ s/ \' \xFF__VALPOS__(\d+)__\xFF \' /"\$cur_row->[$1]"/gex;
408 $parser_src =~ s/ \' \xFF__IDVALPOS__(\d+)__\xFF \' /"\$cur_row_ids[$1]"/gex;
414 # the simple non-collapsing nested structure recursor
415 sub __visit_infmap_simple {
416 my ($val_idx, $args) = @_;
420 for (keys %$val_idx) {
421 if ($_ =~ /^ ([^\.]+) \. (.+) /x) {
422 $rel_cols->{$1}{$2} = $val_idx->{$_};
425 $my_cols->{$_} = $val_idx->{$_};
429 for my $rel (sort keys %$rel_cols) {
432 #my $optional = $args->{is_optional};
433 #$optional ||= ($args->{rsrc}->relationship_info($rel)->{attrs}{join_type} || '') =~ /^left/i;
435 push @relperl, join ' => ', perlstring($rel), __visit_infmap_simple($rel_cols->{$rel}, {
436 rsrc => __get_related_source($args->{rsrc}, $rel, $rel_cols->{$rel}),
439 #is_optional => $optional,
442 # FIXME SUBOPTIMAL DISABLEPRUNE - disabled to satisfy t/resultset/inflate_result_api.t
443 #if ($optional and my @branch_null_checks = map
444 # { "(! defined '\xFF__VALPOS__${_}__\xFF')" }
445 # sort { $a <=> $b } values %{$rel_cols->{$rel}}
447 # $relperl[-1] = sprintf ( '(%s) ? ( %s => [] ) : ( %s )',
448 # join (' && ', @branch_null_checks ),
455 my $me_struct = keys %$my_cols
456 ? __visit_dump({ map { $_ => "\xFF__VALPOS__$my_cols->{$_}__\xFF" } (keys %$my_cols) })
460 return sprintf '[%s]', join (',',
462 @relperl ? sprintf ('{ %s }', join (',', @relperl)) : (),
466 # the collapsing nested structure recursor
467 sub __visit_infmap_collapse {
469 my ($val_idx, $collapse_map, $parent_info) = @_;
473 for (keys %$val_idx) {
474 if ($_ =~ /^ ([^\.]+) \. (.+) /x) {
475 $rel_cols->{$1}{$2} = $val_idx->{$_};
478 $my_cols->{$_} = $val_idx->{$_};
482 my $sequenced_node_id = join ('', map
483 { "{'\xFF__IDVALPOS__${_}__\xFF'}" }
484 @{$collapse_map->{-idcols_current_node}}
487 my $me_struct = keys %$my_cols
488 ? __visit_dump([{ map { $_ => "\xFF__VALPOS__$my_cols->{$_}__\xFF" } (keys %$my_cols) }])
491 my $node_idx_ref = sprintf '$collapse_idx[%d]%s', $collapse_map->{-node_index}, $sequenced_node_id;
493 my $parent_idx_ref = sprintf( '$collapse_idx[%d]%s[1]{%s}',
494 @{$parent_info}{qw/node_idx sequenced_node_id/},
495 perlstring($parent_info->{relname}),
499 if ($collapse_map->{-node_index} == 1) {
500 push @src, sprintf( '%s ||= %s;',
505 elsif ($collapse_map->{-is_single}) {
506 push @src, sprintf ( '%s ||= %s%s;',
509 $me_struct ? " ||= $me_struct" : '',
513 push @src, sprintf('push @{%s}, %s%s unless %s;',
516 $me_struct ? " ||= $me_struct" : '',
522 #my $known_defined = { %{ $parent_info->{known_defined} || {} } };
523 #$known_defined->{$_}++ for @{$collapse_map->{-idcols_current_node}};
525 for my $rel (sort keys %$rel_cols) {
527 # push @src, sprintf(
528 # '%s[1]{%s} ||= [];', $node_idx_ref, perlstring($rel)
529 # ) unless $collapse_map->{$rel}{-is_single};
531 push @src, __visit_infmap_collapse($rel_cols->{$rel}, $collapse_map->{$rel}, {
532 node_idx => $collapse_map->{-node_index},
533 sequenced_node_id => $sequenced_node_id,
536 #known_defined => $known_defined,
539 # FIXME SUBOPTIMAL DISABLEPRUNE - disabled to satisfy t/resultset/inflate_result_api.t
540 #if ($collapse_map->{$rel}{-is_optional} and my @null_checks = map
541 # { "(! defined '\xFF__IDVALPOS__${_}__\xFF')" }
542 # sort { $a <=> $b } grep
543 # { ! $known_defined->{$_} }
544 # @{$collapse_map->{$rel}{-idcols_current_node}}
546 # $src[-1] = sprintf( '(%s) or %s',
547 # join (' || ', @null_checks ),
556 # adding a dep on MoreUtils *just* for this is retarded
557 sub __unique_numlist {
558 sort { $a <=> $b } keys %{ {map { $_ => 1 } @_ }}
561 # This error must be thrown from two distinct codepaths, joining them is
562 # rather hard. Go for this hack instead.
563 sub __get_related_source {
564 my ($rsrc, $rel, $relcols) = @_;
566 $rsrc->related_source ($rel)
568 $rsrc->throw_exception(sprintf(
569 "Can't inflate prefetch into non-existent relationship '%s' from '%s', "
570 . "check the inflation specification (columns/as) ending in '...%s.%s'.",
574 (sort { length($a) <=> length ($b) } keys %$relcols)[0],
578 # keep our own DD object around so we don't have to fitz with quoting
581 # we actually will be producing functional perl code here,
582 # thus no second-guessing of what these globals might have
583 # been set to. DO NOT CHANGE!
584 ($dumper_obj ||= do {
585 require Data::Dumper;
586 Data::Dumper->new([])
596 ->Indent (0) # faster but harder to read, perhaps leave at 1 ?
597 })->Values ([$_[0]])->Dump;