1 package SQL::Abstract; # see doc at end of file
3 # LDNOTE : this code is heavy refactoring from original SQLA.
4 # Several design decisions will need discussion during
5 # the test / diffusion / acceptance phase; those are marked with flag
6 # 'LDNOTE' (note by laurent.dami AT free.fr)
11 use List::Util qw/first/;
12 use Scalar::Util qw/blessed/;
14 #======================================================================
16 #======================================================================
18 our $VERSION = '1.51';
20 # This would confuse some packagers
21 #$VERSION = eval $VERSION; # numify for warning-free dev releases
25 # special operators (-in, -between). May be extended/overridden by user.
26 # See section WHERE: BUILTIN SPECIAL OPERATORS below for implementation
27 my @BUILTIN_SPECIAL_OPS = (
28 {regex => qr/^(not )?between$/i, handler => \&_where_field_BETWEEN},
29 {regex => qr/^(not )?in$/i, handler => \&_where_field_IN},
32 #======================================================================
33 # DEBUGGING AND ERROR REPORTING
34 #======================================================================
37 return unless $_[0]->{debug}; shift; # a little faster
38 my $func = (caller(1))[3];
39 warn "[$func] ", @_, "\n";
43 my($func) = (caller(1))[3];
44 carp "[$func] Warning: ", @_;
48 my($func) = (caller(1))[3];
49 croak "[$func] Fatal: ", @_;
53 #======================================================================
55 #======================================================================
59 my $class = ref($self) || $self;
60 my %opt = (ref $_[0] eq 'HASH') ? %{$_[0]} : @_;
62 # choose our case by keeping an option around
63 delete $opt{case} if $opt{case} && $opt{case} ne 'lower';
65 # default logic for interpreting arrayrefs
66 $opt{logic} = $opt{logic} ? uc $opt{logic} : 'OR';
68 # how to return bind vars
69 # LDNOTE: changed nwiger code : why this 'delete' ??
70 # $opt{bindtype} ||= delete($opt{bind_type}) || 'normal';
71 $opt{bindtype} ||= 'normal';
73 # default comparison is "=", but can be overridden
76 # try to recognize which are the 'equality' and 'unequality' ops
77 # (temporary quickfix, should go through a more seasoned API)
78 $opt{equality_op} = qr/^(\Q$opt{cmp}\E|is|(is\s+)?like)$/i;
79 $opt{inequality_op} = qr/^(!=|<>|(is\s+)?not(\s+like)?)$/i;
82 $opt{sqltrue} ||= '1=1';
83 $opt{sqlfalse} ||= '0=1';
86 $opt{special_ops} ||= [];
87 push @{$opt{special_ops}}, @BUILTIN_SPECIAL_OPS;
89 return bless \%opt, $class;
94 #======================================================================
96 #======================================================================
100 my $table = $self->_table(shift);
101 my $data = shift || return;
103 my $method = $self->_METHOD_FOR_refkind("_insert", $data);
104 my ($sql, @bind) = $self->$method($data);
105 $sql = join " ", $self->_sqlcase('insert into'), $table, $sql;
106 return wantarray ? ($sql, @bind) : $sql;
109 sub _insert_HASHREF { # explicit list of fields and then values
110 my ($self, $data) = @_;
112 my @fields = sort keys %$data;
114 my ($sql, @bind) = $self->_insert_values($data);
117 $_ = $self->_quote($_) foreach @fields;
118 $sql = "( ".join(", ", @fields).") ".$sql;
120 return ($sql, @bind);
123 sub _insert_ARRAYREF { # just generate values(?,?) part (no list of fields)
124 my ($self, $data) = @_;
126 # no names (arrayref) so can't generate bindtype
127 $self->{bindtype} ne 'columns'
128 or belch "can't do 'columns' bindtype when called with arrayref";
130 # fold the list of values into a hash of column name - value pairs
131 # (where the column names are artificially generated, and their
132 # lexicographical ordering keep the ordering of the original list)
133 my $i = "a"; # incremented values will be in lexicographical order
134 my $data_in_hash = { map { ($i++ => $_) } @$data };
136 return $self->_insert_values($data_in_hash);
139 sub _insert_ARRAYREFREF { # literal SQL with bind
140 my ($self, $data) = @_;
142 my ($sql, @bind) = @${$data};
143 $self->_assert_bindval_matches_bindtype(@bind);
145 return ($sql, @bind);
149 sub _insert_SCALARREF { # literal SQL without bind
150 my ($self, $data) = @_;
156 my ($self, $data) = @_;
158 my (@values, @all_bind);
159 foreach my $column (sort keys %$data) {
160 my $v = $data->{$column};
162 $self->_SWITCH_refkind($v, {
165 if ($self->{array_datatypes}) { # if array datatype are activated
167 push @all_bind, $self->_bindtype($column, $v);
169 else { # else literal SQL with bind
170 my ($sql, @bind) = @$v;
171 $self->_assert_bindval_matches_bindtype(@bind);
173 push @all_bind, @bind;
177 ARRAYREFREF => sub { # literal SQL with bind
178 my ($sql, @bind) = @${$v};
179 $self->_assert_bindval_matches_bindtype(@bind);
181 push @all_bind, @bind;
184 # THINK : anything useful to do with a HASHREF ?
185 HASHREF => sub { # (nothing, but old SQLA passed it through)
186 #TODO in SQLA >= 2.0 it will die instead
187 belch "HASH ref as bind value in insert is not supported";
189 push @all_bind, $self->_bindtype($column, $v);
192 SCALARREF => sub { # literal SQL without bind
196 SCALAR_or_UNDEF => sub {
198 push @all_bind, $self->_bindtype($column, $v);
205 my $sql = $self->_sqlcase('values')." ( ".join(", ", @values)." )";
206 return ($sql, @all_bind);
211 #======================================================================
213 #======================================================================
218 my $table = $self->_table(shift);
219 my $data = shift || return;
222 # first build the 'SET' part of the sql statement
223 my (@set, @all_bind);
224 puke "Unsupported data type specified to \$sql->update"
225 unless ref $data eq 'HASH';
227 for my $k (sort keys %$data) {
230 my $label = $self->_quote($k);
232 $self->_SWITCH_refkind($v, {
234 if ($self->{array_datatypes}) { # array datatype
235 push @set, "$label = ?";
236 push @all_bind, $self->_bindtype($k, $v);
238 else { # literal SQL with bind
239 my ($sql, @bind) = @$v;
240 $self->_assert_bindval_matches_bindtype(@bind);
241 push @set, "$label = $sql";
242 push @all_bind, @bind;
245 ARRAYREFREF => sub { # literal SQL with bind
246 my ($sql, @bind) = @${$v};
247 $self->_assert_bindval_matches_bindtype(@bind);
248 push @set, "$label = $sql";
249 push @all_bind, @bind;
251 SCALARREF => sub { # literal SQL without bind
252 push @set, "$label = $$v";
254 SCALAR_or_UNDEF => sub {
255 push @set, "$label = ?";
256 push @all_bind, $self->_bindtype($k, $v);
262 my $sql = $self->_sqlcase('update') . " $table " . $self->_sqlcase('set ')
266 my($where_sql, @where_bind) = $self->where($where);
268 push @all_bind, @where_bind;
271 return wantarray ? ($sql, @all_bind) : $sql;
277 #======================================================================
279 #======================================================================
284 my $table = $self->_table(shift);
285 my $fields = shift || '*';
289 my($where_sql, @bind) = $self->where($where, $order);
291 my $f = (ref $fields eq 'ARRAY') ? join ', ', map { $self->_quote($_) } @$fields
293 my $sql = join(' ', $self->_sqlcase('select'), $f,
294 $self->_sqlcase('from'), $table)
297 return wantarray ? ($sql, @bind) : $sql;
300 #======================================================================
302 #======================================================================
307 my $table = $self->_table(shift);
311 my($where_sql, @bind) = $self->where($where);
312 my $sql = $self->_sqlcase('delete from') . " $table" . $where_sql;
314 return wantarray ? ($sql, @bind) : $sql;
318 #======================================================================
320 #======================================================================
324 # Finally, a separate routine just to handle WHERE clauses
326 my ($self, $where, $order) = @_;
329 my ($sql, @bind) = $self->_recurse_where($where);
330 $sql = $sql ? $self->_sqlcase(' where ') . "( $sql )" : '';
334 $sql .= $self->_order_by($order);
337 return wantarray ? ($sql, @bind) : $sql;
342 my ($self, $where, $logic) = @_;
344 # dispatch on appropriate method according to refkind of $where
345 my $method = $self->_METHOD_FOR_refkind("_where", $where);
348 my ($sql, @bind) = $self->$method($where, $logic);
350 # DBIx::Class directly calls _recurse_where in scalar context, so
351 # we must implement it, even if not in the official API
352 return wantarray ? ($sql, @bind) : $sql;
357 #======================================================================
358 # WHERE: top-level ARRAYREF
359 #======================================================================
362 sub _where_ARRAYREF {
363 my ($self, $where, $logic) = @_;
365 $logic = uc($logic || $self->{logic});
366 $logic eq 'AND' or $logic eq 'OR' or puke "unknown logic: $logic";
368 my @clauses = @$where;
370 my (@sql_clauses, @all_bind);
371 # need to use while() so can shift() for pairs
372 while (my $el = shift @clauses) {
374 # switch according to kind of $el and get corresponding ($sql, @bind)
375 my ($sql, @bind) = $self->_SWITCH_refkind($el, {
377 # skip empty elements, otherwise get invalid trailing AND stuff
378 ARRAYREF => sub {$self->_recurse_where($el) if @$el},
380 ARRAYREFREF => sub { @{${$el}} if @{${$el}}},
382 HASHREF => sub {$self->_recurse_where($el, 'and') if %$el},
383 # LDNOTE : previous SQLA code for hashrefs was creating a dirty
384 # side-effect: the first hashref within an array would change
385 # the global logic to 'AND'. So [ {cond1, cond2}, [cond3, cond4] ]
386 # was interpreted as "(cond1 AND cond2) OR (cond3 AND cond4)",
387 # whereas it should be "(cond1 AND cond2) OR (cond3 OR cond4)".
389 SCALARREF => sub { ($$el); },
391 SCALAR => sub {# top-level arrayref with scalars, recurse in pairs
392 $self->_recurse_where({$el => shift(@clauses)})},
394 UNDEF => sub {puke "not supported : UNDEF in arrayref" },
398 push @sql_clauses, $sql;
399 push @all_bind, @bind;
403 return $self->_join_sql_clauses($logic, \@sql_clauses, \@all_bind);
406 #======================================================================
407 # WHERE: top-level ARRAYREFREF
408 #======================================================================
410 sub _where_ARRAYREFREF {
411 my ($self, $where) = @_;
412 my ($sql, @bind) = @{${$where}};
414 return ($sql, @bind);
417 #======================================================================
418 # WHERE: top-level HASHREF
419 #======================================================================
422 my ($self, $where) = @_;
423 my (@sql_clauses, @all_bind);
425 # LDNOTE : don't really know why we need to sort keys
426 for my $k (sort keys %$where) {
427 my $v = $where->{$k};
429 # ($k => $v) is either a special op or a regular hashpair
430 my ($sql, @bind) = ($k =~ /^-(.+)/) ? $self->_where_op_in_hash($1, $v)
432 my $method = $self->_METHOD_FOR_refkind("_where_hashpair", $v);
433 $self->$method($k, $v);
436 push @sql_clauses, $sql;
437 push @all_bind, @bind;
440 return $self->_join_sql_clauses('and', \@sql_clauses, \@all_bind);
444 sub _where_op_in_hash {
445 my ($self, $op_str, $v) = @_;
447 $op_str =~ /^ (AND|OR|NEST) ( \_? \d* ) $/xi
448 or puke "unknown operator: -$op_str";
450 my $op = uc($1); # uppercase, remove trailing digits
452 belch 'Use of [and|or|nest]_N modifiers is deprecated and will be removed in SQLA v2.0. '
453 . "You probably wanted ...-and => [ $op_str => COND1, $op_str => COND2 ... ]";
456 $self->_debug("OP(-$op) within hashref, recursing...");
458 $self->_SWITCH_refkind($v, {
461 return $self->_where_ARRAYREF($v, $op eq 'NEST' ? '' : $op);
466 return $self->_where_ARRAYREF([%$v], 'OR');
469 return $self->_where_HASHREF($v);
473 SCALARREF => sub { # literal SQL
475 or puke "-$op => \\\$scalar not supported, use -nest => ...";
479 ARRAYREFREF => sub { # literal SQL
481 or puke "-$op => \\[..] not supported, use -nest => ...";
485 SCALAR => sub { # permissively interpreted as SQL
487 or puke "-$op => 'scalar' not supported, use -nest => \\'scalar'";
488 belch "literal SQL should be -nest => \\'scalar' "
489 . "instead of -nest => 'scalar' ";
494 puke "-$op => undef not supported";
500 sub _where_hashpair_ARRAYREF {
501 my ($self, $k, $v) = @_;
504 my @v = @$v; # need copy because of shift below
505 $self->_debug("ARRAY($k) means distribute over elements");
507 # put apart first element if it is an operator (-and, -or)
509 (defined $v[0] && $v[0] =~ /^ - (?: AND|OR ) $/ix)
513 my @distributed = map { {$k => $_} } @v;
516 $self->_debug("OP($op) reinjected into the distributed array");
517 unshift @distributed, $op;
520 my $logic = $op ? substr($op, 1) : '';
522 return $self->_recurse_where(\@distributed, $logic);
525 # LDNOTE : not sure of this one. What does "distribute over nothing" mean?
526 $self->_debug("empty ARRAY($k) means 0=1");
527 return ($self->{sqlfalse});
531 sub _where_hashpair_HASHREF {
532 my ($self, $k, $v, $logic) = @_;
535 my ($all_sql, @all_bind);
537 for my $op (sort keys %$v) {
540 # put the operator in canonical form
541 $op =~ s/^-//; # remove initial dash
542 $op =~ tr/_/ /; # underscores become spaces
543 $op =~ s/^\s+//; # no initial space
544 $op =~ s/\s+$//; # no final space
545 $op =~ s/\s+/ /; # multiple spaces become one
549 # CASE: special operators like -in or -between
550 my $special_op = first {$op =~ $_->{regex}} @{$self->{special_ops}};
552 ($sql, @bind) = $special_op->{handler}->($self, $k, $op, $val);
555 $self->_SWITCH_refkind($val, {
557 ARRAYREF => sub { # CASE: col => {op => \@vals}
558 ($sql, @bind) = $self->_where_field_op_ARRAYREF($k, $op, $val);
561 SCALARREF => sub { # CASE: col => {op => \$scalar} (literal SQL without bind)
562 $sql = join ' ', $self->_convert($self->_quote($k)),
563 $self->_sqlcase($op),
567 ARRAYREFREF => sub { # CASE: col => {op => \[$sql, @bind]} (literal SQL with bind)
568 my ($sub_sql, @sub_bind) = @$$val;
569 $self->_assert_bindval_matches_bindtype(@sub_bind);
570 $sql = join ' ', $self->_convert($self->_quote($k)),
571 $self->_sqlcase($op),
577 ($sql, @bind) = $self->_where_hashpair_HASHREF($k, $val, $op);
580 UNDEF => sub { # CASE: col => {op => undef} : sql "IS (NOT)? NULL"
581 my $is = ($op =~ $self->{equality_op}) ? 'is' :
582 ($op =~ $self->{inequality_op}) ? 'is not' :
583 puke "unexpected operator '$op' with undef operand";
584 $sql = $self->_quote($k) . $self->_sqlcase(" $is null");
587 FALLBACK => sub { # CASE: col => {op => $scalar}
588 $sql = join ' ', $self->_convert($self->_quote($k)),
589 $self->_sqlcase($op),
590 $self->_convert('?');
591 @bind = $self->_bindtype($k, $val);
596 ($all_sql) = (defined $all_sql and $all_sql) ? $self->_join_sql_clauses($logic, [$all_sql, $sql], []) : $sql;
597 push @all_bind, @bind;
599 return ($all_sql, @all_bind);
604 sub _where_field_op_ARRAYREF {
605 my ($self, $k, $op, $vals) = @_;
608 $self->_debug("ARRAY($vals) means multiple elements: [ @$vals ]");
610 # LDNOTE : had planned to change the distribution logic when
611 # $op =~ $self->{inequality_op}, because of Morgan laws :
612 # with {field => {'!=' => [22, 33]}}, it would be ridiculous to generate
613 # WHERE field != 22 OR field != 33 : the user probably means
614 # WHERE field != 22 AND field != 33.
615 # To do this, replace the line below by :
616 # my $logic = ($op =~ $self->{inequality_op}) ? 'AND' : 'OR';
617 # return $self->_recurse_where([map { {$k => {$op, $_}} } @$vals], $logic);
619 # distribute $op over each member of @$vals
620 return $self->_recurse_where([map { {$k => {$op, $_}} } @$vals]);
623 # try to DWIM on equality operators
624 # LDNOTE : not 100% sure this is the correct thing to do ...
625 return ($self->{sqlfalse}) if $op =~ $self->{equality_op};
626 return ($self->{sqltrue}) if $op =~ $self->{inequality_op};
629 puke "operator '$op' applied on an empty array (field '$k')";
634 sub _where_hashpair_SCALARREF {
635 my ($self, $k, $v) = @_;
636 $self->_debug("SCALAR($k) means literal SQL: $$v");
637 my $sql = $self->_quote($k) . " " . $$v;
641 # literal SQL with bind
642 sub _where_hashpair_ARRAYREFREF {
643 my ($self, $k, $v) = @_;
644 $self->_debug("REF($k) means literal SQL: @${$v}");
645 my ($sql, @bind) = @${$v};
646 $self->_assert_bindval_matches_bindtype(@bind);
647 $sql = $self->_quote($k) . " " . $sql;
648 return ($sql, @bind );
651 # literal SQL without bind
652 sub _where_hashpair_SCALAR {
653 my ($self, $k, $v) = @_;
654 $self->_debug("NOREF($k) means simple key=val: $k $self->{cmp} $v");
655 my $sql = join ' ', $self->_convert($self->_quote($k)),
656 $self->_sqlcase($self->{cmp}),
657 $self->_convert('?');
658 my @bind = $self->_bindtype($k, $v);
659 return ( $sql, @bind);
663 sub _where_hashpair_UNDEF {
664 my ($self, $k, $v) = @_;
665 $self->_debug("UNDEF($k) means IS NULL");
666 my $sql = $self->_quote($k) . $self->_sqlcase(' is null');
670 #======================================================================
671 # WHERE: TOP-LEVEL OTHERS (SCALARREF, SCALAR, UNDEF)
672 #======================================================================
675 sub _where_SCALARREF {
676 my ($self, $where) = @_;
679 $self->_debug("SCALAR(*top) means literal SQL: $$where");
685 my ($self, $where) = @_;
688 $self->_debug("NOREF(*top) means literal SQL: $where");
699 #======================================================================
700 # WHERE: BUILTIN SPECIAL OPERATORS (-in, -between)
701 #======================================================================
704 sub _where_field_BETWEEN {
705 my ($self, $k, $op, $vals) = @_;
707 (ref $vals eq 'ARRAY' && @$vals == 2) or
708 (ref $vals eq 'REF' && (@$$vals == 1 || @$$vals == 2 || @$$vals == 3))
709 or puke "special op 'between' requires an arrayref of two values (or a scalarref or arrayrefref for literal SQL)";
711 my ($clause, @bind, $label, $and, $placeholder);
712 $label = $self->_convert($self->_quote($k));
713 $and = ' ' . $self->_sqlcase('and') . ' ';
714 $placeholder = $self->_convert('?');
715 $op = $self->_sqlcase($op);
717 if (ref $vals eq 'REF') {
718 ($clause, @bind) = @$$vals;
721 my (@all_sql, @all_bind);
723 foreach my $val (@$vals) {
724 my ($sql, @bind) = $self->_SWITCH_refkind($val, {
726 return ($placeholder, ($val));
729 return ($self->_convert($$val), ());
733 push @all_bind, @bind;
736 $clause = (join $and, @all_sql);
737 @bind = $self->_bindtype($k, @all_bind);
739 my $sql = "( $label $op $clause )";
744 sub _where_field_IN {
745 my ($self, $k, $op, $vals) = @_;
747 # backwards compatibility : if scalar, force into an arrayref
748 $vals = [$vals] if defined $vals && ! ref $vals;
750 my ($label) = $self->_convert($self->_quote($k));
751 my ($placeholder) = $self->_convert('?');
752 $op = $self->_sqlcase($op);
754 my ($sql, @bind) = $self->_SWITCH_refkind($vals, {
755 ARRAYREF => sub { # list of choices
756 if (@$vals) { # nonempty list
757 my $placeholders = join ", ", (($placeholder) x @$vals);
758 my $sql = "$label $op ( $placeholders )";
759 my @bind = $self->_bindtype($k, @$vals);
761 return ($sql, @bind);
763 else { # empty list : some databases won't understand "IN ()", so DWIM
764 my $sql = ($op =~ /\bnot\b/i) ? $self->{sqltrue} : $self->{sqlfalse};
769 ARRAYREFREF => sub { # literal SQL with bind
770 my ($sql, @bind) = @$$vals;
771 $self->_assert_bindval_matches_bindtype(@bind);
772 return ("$label $op ( $sql )", @bind);
776 puke "special op 'in' requires an arrayref (or arrayref-ref)";
780 return ($sql, @bind);
788 #======================================================================
790 #======================================================================
793 my ($self, $arg) = @_;
795 # construct list of ordering instructions
796 my @order = $self->_SWITCH_refkind($arg, {
799 map {$self->_SWITCH_refkind($_, {
800 SCALAR => sub {$self->_quote($_)},
802 SCALARREF => sub {$$_}, # literal SQL, no quoting
803 HASHREF => sub {$self->_order_by_hash($_)}
807 SCALAR => sub {$self->_quote($arg)},
809 SCALARREF => sub {$$arg}, # literal SQL, no quoting
810 HASHREF => sub {$self->_order_by_hash($arg)},
815 my $order = join ', ', @order;
816 return $order ? $self->_sqlcase(' order by')." $order" : '';
821 my ($self, $hash) = @_;
823 # get first pair in hash
824 my ($key, $val) = each %$hash;
826 # check if one pair was found and no other pair in hash
827 $key && !(each %$hash)
828 or puke "hash passed to _order_by must have exactly one key (-desc or -asc)";
830 my ($order) = ($key =~ /^-(desc|asc)/i)
831 or puke "invalid key in _order_by hash : $key";
833 $val = ref $val eq 'ARRAY' ? $val : [$val];
834 return join ', ', map { $self->_quote($_) . ' ' . $self->_sqlcase($order) } @$val;
839 #======================================================================
840 # DATASOURCE (FOR NOW, JUST PLAIN TABLE OR LIST OF TABLES)
841 #======================================================================
846 $self->_SWITCH_refkind($from, {
847 ARRAYREF => sub {join ', ', map { $self->_quote($_) } @$from;},
848 SCALAR => sub {$self->_quote($from)},
849 SCALARREF => sub {$$from},
850 ARRAYREFREF => sub {join ', ', @$from;},
855 #======================================================================
857 #======================================================================
863 $label or puke "can't quote an empty label";
865 # left and right quote characters
866 my ($ql, $qr, @other) = $self->_SWITCH_refkind($self->{quote_char}, {
867 SCALAR => sub {($self->{quote_char}, $self->{quote_char})},
868 ARRAYREF => sub {@{$self->{quote_char}}},
872 or puke "quote_char must be an arrayref of 2 values";
874 # no quoting if no quoting chars
875 $ql or return $label;
877 # no quoting for literal SQL
878 return $$label if ref($label) eq 'SCALAR';
880 # separate table / column (if applicable)
881 my $sep = $self->{name_sep} || '';
882 my @to_quote = $sep ? split /\Q$sep\E/, $label : ($label);
884 # do the quoting, except for "*" or for `table`.*
885 my @quoted = map { $_ eq '*' ? $_: $ql.$_.$qr} @to_quote;
887 # reassemble and return.
888 return join $sep, @quoted;
892 # Conversion, if applicable
894 my ($self, $arg) = @_;
896 # LDNOTE : modified the previous implementation below because
897 # it was not consistent : the first "return" is always an array,
898 # the second "return" is context-dependent. Anyway, _convert
899 # seems always used with just a single argument, so make it a
901 # return @_ unless $self->{convert};
902 # my $conv = $self->_sqlcase($self->{convert});
903 # my @ret = map { $conv.'('.$_.')' } @_;
904 # return wantarray ? @ret : $ret[0];
905 if ($self->{convert}) {
906 my $conv = $self->_sqlcase($self->{convert});
907 $arg = $conv.'('.$arg.')';
915 my($col, @vals) = @_;
917 #LDNOTE : changed original implementation below because it did not make
918 # sense when bindtype eq 'columns' and @vals > 1.
919 # return $self->{bindtype} eq 'columns' ? [ $col, @vals ] : @vals;
921 return $self->{bindtype} eq 'columns' ? map {[$col, $_]} @vals : @vals;
924 # Dies if any element of @bind is not in [colname => value] format
925 # if bindtype is 'columns'.
926 sub _assert_bindval_matches_bindtype {
927 my ($self, @bind) = @_;
929 if ($self->{bindtype} eq 'columns') {
930 foreach my $val (@bind) {
931 if (!defined $val || ref($val) ne 'ARRAY' || @$val != 2) {
932 die "bindtype 'columns' selected, you need to pass: [column_name => bind_value]"
938 sub _join_sql_clauses {
939 my ($self, $logic, $clauses_aref, $bind_aref) = @_;
941 if (@$clauses_aref > 1) {
942 my $join = " " . $self->_sqlcase($logic) . " ";
943 my $sql = '( ' . join($join, @$clauses_aref) . ' )';
944 return ($sql, @$bind_aref);
946 elsif (@$clauses_aref) {
947 return ($clauses_aref->[0], @$bind_aref); # no parentheses
950 return (); # if no SQL, ignore @$bind_aref
955 # Fix SQL case, if so requested
959 # LDNOTE: if $self->{case} is true, then it contains 'lower', so we
960 # don't touch the argument ... crooked logic, but let's not change it!
961 return $self->{case} ? $_[0] : uc($_[0]);
965 #======================================================================
966 # DISPATCHING FROM REFKIND
967 #======================================================================
970 my ($self, $data) = @_;
976 # blessed objects are treated like scalars
977 $ref = (blessed $data) ? '' : ref $data;
978 $n_steps += 1 if $ref;
979 last if $ref ne 'REF';
983 my $base = $ref || (defined $data ? 'SCALAR' : 'UNDEF');
985 return $base . ('REF' x $n_steps);
991 my ($self, $data) = @_;
992 my @try = ($self->_refkind($data));
993 push @try, 'SCALAR_or_UNDEF' if $try[0] eq 'SCALAR' || $try[0] eq 'UNDEF';
994 push @try, 'FALLBACK';
998 sub _METHOD_FOR_refkind {
999 my ($self, $meth_prefix, $data) = @_;
1000 my $method = first {$_} map {$self->can($meth_prefix."_".$_)}
1001 $self->_try_refkind($data)
1002 or puke "cannot dispatch on '$meth_prefix' for ".$self->_refkind($data);
1007 sub _SWITCH_refkind {
1008 my ($self, $data, $dispatch_table) = @_;
1010 my $coderef = first {$_} map {$dispatch_table->{$_}}
1011 $self->_try_refkind($data)
1012 or puke "no dispatch entry for ".$self->_refkind($data);
1019 #======================================================================
1020 # VALUES, GENERATE, AUTOLOAD
1021 #======================================================================
1023 # LDNOTE: original code from nwiger, didn't touch code in that section
1024 # I feel the AUTOLOAD stuff should not be the default, it should
1025 # only be activated on explicit demand by user.
1029 my $data = shift || return;
1030 puke "Argument to ", __PACKAGE__, "->values must be a \\%hash"
1031 unless ref $data eq 'HASH';
1034 foreach my $k ( sort keys %$data ) {
1035 my $v = $data->{$k};
1036 $self->_SWITCH_refkind($v, {
1038 if ($self->{array_datatypes}) { # array datatype
1039 push @all_bind, $self->_bindtype($k, $v);
1041 else { # literal SQL with bind
1042 my ($sql, @bind) = @$v;
1043 $self->_assert_bindval_matches_bindtype(@bind);
1044 push @all_bind, @bind;
1047 ARRAYREFREF => sub { # literal SQL with bind
1048 my ($sql, @bind) = @${$v};
1049 $self->_assert_bindval_matches_bindtype(@bind);
1050 push @all_bind, @bind;
1052 SCALARREF => sub { # literal SQL without bind
1054 SCALAR_or_UNDEF => sub {
1055 push @all_bind, $self->_bindtype($k, $v);
1066 my(@sql, @sqlq, @sqlv);
1070 if ($ref eq 'HASH') {
1071 for my $k (sort keys %$_) {
1074 my $label = $self->_quote($k);
1075 if ($r eq 'ARRAY') {
1076 # literal SQL with bind
1077 my ($sql, @bind) = @$v;
1078 $self->_assert_bindval_matches_bindtype(@bind);
1079 push @sqlq, "$label = $sql";
1081 } elsif ($r eq 'SCALAR') {
1082 # literal SQL without bind
1083 push @sqlq, "$label = $$v";
1085 push @sqlq, "$label = ?";
1086 push @sqlv, $self->_bindtype($k, $v);
1089 push @sql, $self->_sqlcase('set'), join ', ', @sqlq;
1090 } elsif ($ref eq 'ARRAY') {
1091 # unlike insert(), assume these are ONLY the column names, i.e. for SQL
1094 if ($r eq 'ARRAY') { # literal SQL with bind
1095 my ($sql, @bind) = @$v;
1096 $self->_assert_bindval_matches_bindtype(@bind);
1099 } elsif ($r eq 'SCALAR') { # literal SQL without bind
1100 # embedded literal SQL
1107 push @sql, '(' . join(', ', @sqlq) . ')';
1108 } elsif ($ref eq 'SCALAR') {
1112 # strings get case twiddled
1113 push @sql, $self->_sqlcase($_);
1117 my $sql = join ' ', @sql;
1119 # this is pretty tricky
1120 # if ask for an array, return ($stmt, @bind)
1121 # otherwise, s/?/shift @sqlv/ to put it inline
1123 return ($sql, @sqlv);
1125 1 while $sql =~ s/\?/my $d = shift(@sqlv);
1126 ref $d ? $d->[1] : $d/e;
1135 # This allows us to check for a local, then _form, attr
1137 my($name) = $AUTOLOAD =~ /.*::(.+)/;
1138 return $self->generate($name, @_);
1149 SQL::Abstract - Generate SQL from Perl data structures
1155 my $sql = SQL::Abstract->new;
1157 my($stmt, @bind) = $sql->select($table, \@fields, \%where, \@order);
1159 my($stmt, @bind) = $sql->insert($table, \%fieldvals || \@values);
1161 my($stmt, @bind) = $sql->update($table, \%fieldvals, \%where);
1163 my($stmt, @bind) = $sql->delete($table, \%where);
1165 # Then, use these in your DBI statements
1166 my $sth = $dbh->prepare($stmt);
1167 $sth->execute(@bind);
1169 # Just generate the WHERE clause
1170 my($stmt, @bind) = $sql->where(\%where, \@order);
1172 # Return values in the same order, for hashed queries
1173 # See PERFORMANCE section for more details
1174 my @bind = $sql->values(\%fieldvals);
1178 This module was inspired by the excellent L<DBIx::Abstract>.
1179 However, in using that module I found that what I really wanted
1180 to do was generate SQL, but still retain complete control over my
1181 statement handles and use the DBI interface. So, I set out to
1182 create an abstract SQL generation module.
1184 While based on the concepts used by L<DBIx::Abstract>, there are
1185 several important differences, especially when it comes to WHERE
1186 clauses. I have modified the concepts used to make the SQL easier
1187 to generate from Perl data structures and, IMO, more intuitive.
1188 The underlying idea is for this module to do what you mean, based
1189 on the data structures you provide it. The big advantage is that
1190 you don't have to modify your code every time your data changes,
1191 as this module figures it out.
1193 To begin with, an SQL INSERT is as easy as just specifying a hash
1194 of C<key=value> pairs:
1197 name => 'Jimbo Bobson',
1198 phone => '123-456-7890',
1199 address => '42 Sister Lane',
1200 city => 'St. Louis',
1201 state => 'Louisiana',
1204 The SQL can then be generated with this:
1206 my($stmt, @bind) = $sql->insert('people', \%data);
1208 Which would give you something like this:
1210 $stmt = "INSERT INTO people
1211 (address, city, name, phone, state)
1212 VALUES (?, ?, ?, ?, ?)";
1213 @bind = ('42 Sister Lane', 'St. Louis', 'Jimbo Bobson',
1214 '123-456-7890', 'Louisiana');
1216 These are then used directly in your DBI code:
1218 my $sth = $dbh->prepare($stmt);
1219 $sth->execute(@bind);
1221 =head2 Inserting and Updating Arrays
1223 If your database has array types (like for example Postgres),
1224 activate the special option C<< array_datatypes => 1 >>
1225 when creating the C<SQL::Abstract> object.
1226 Then you may use an arrayref to insert and update database array types:
1228 my $sql = SQL::Abstract->new(array_datatypes => 1);
1230 planets => [qw/Mercury Venus Earth Mars/]
1233 my($stmt, @bind) = $sql->insert('solar_system', \%data);
1237 $stmt = "INSERT INTO solar_system (planets) VALUES (?)"
1239 @bind = (['Mercury', 'Venus', 'Earth', 'Mars']);
1242 =head2 Inserting and Updating SQL
1244 In order to apply SQL functions to elements of your C<%data> you may
1245 specify a reference to an arrayref for the given hash value. For example,
1246 if you need to execute the Oracle C<to_date> function on a value, you can
1247 say something like this:
1251 date_entered => \["to_date(?,'MM/DD/YYYY')", "03/02/2003"],
1254 The first value in the array is the actual SQL. Any other values are
1255 optional and would be included in the bind values array. This gives
1258 my($stmt, @bind) = $sql->insert('people', \%data);
1260 $stmt = "INSERT INTO people (name, date_entered)
1261 VALUES (?, to_date(?,'MM/DD/YYYY'))";
1262 @bind = ('Bill', '03/02/2003');
1264 An UPDATE is just as easy, all you change is the name of the function:
1266 my($stmt, @bind) = $sql->update('people', \%data);
1268 Notice that your C<%data> isn't touched; the module will generate
1269 the appropriately quirky SQL for you automatically. Usually you'll
1270 want to specify a WHERE clause for your UPDATE, though, which is
1271 where handling C<%where> hashes comes in handy...
1273 =head2 Complex where statements
1275 This module can generate pretty complicated WHERE statements
1276 easily. For example, simple C<key=value> pairs are taken to mean
1277 equality, and if you want to see if a field is within a set
1278 of values, you can use an arrayref. Let's say we wanted to
1279 SELECT some data based on this criteria:
1282 requestor => 'inna',
1283 worker => ['nwiger', 'rcwe', 'sfz'],
1284 status => { '!=', 'completed' }
1287 my($stmt, @bind) = $sql->select('tickets', '*', \%where);
1289 The above would give you something like this:
1291 $stmt = "SELECT * FROM tickets WHERE
1292 ( requestor = ? ) AND ( status != ? )
1293 AND ( worker = ? OR worker = ? OR worker = ? )";
1294 @bind = ('inna', 'completed', 'nwiger', 'rcwe', 'sfz');
1296 Which you could then use in DBI code like so:
1298 my $sth = $dbh->prepare($stmt);
1299 $sth->execute(@bind);
1305 The functions are simple. There's one for each major SQL operation,
1306 and a constructor you use first. The arguments are specified in a
1307 similar order to each function (table, then fields, then a where
1308 clause) to try and simplify things.
1313 =head2 new(option => 'value')
1315 The C<new()> function takes a list of options and values, and returns
1316 a new B<SQL::Abstract> object which can then be used to generate SQL
1317 through the methods below. The options accepted are:
1323 If set to 'lower', then SQL will be generated in all lowercase. By
1324 default SQL is generated in "textbook" case meaning something like:
1326 SELECT a_field FROM a_table WHERE some_field LIKE '%someval%'
1328 Any setting other than 'lower' is ignored.
1332 This determines what the default comparison operator is. By default
1333 it is C<=>, meaning that a hash like this:
1335 %where = (name => 'nwiger', email => 'nate@wiger.org');
1337 Will generate SQL like this:
1339 WHERE name = 'nwiger' AND email = 'nate@wiger.org'
1341 However, you may want loose comparisons by default, so if you set
1342 C<cmp> to C<like> you would get SQL such as:
1344 WHERE name like 'nwiger' AND email like 'nate@wiger.org'
1346 You can also override the comparsion on an individual basis - see
1347 the huge section on L</"WHERE CLAUSES"> at the bottom.
1349 =item sqltrue, sqlfalse
1351 Expressions for inserting boolean values within SQL statements.
1352 By default these are C<1=1> and C<1=0>. They are used
1353 by the special operators C<-in> and C<-not_in> for generating
1354 correct SQL even when the argument is an empty array (see below).
1358 This determines the default logical operator for multiple WHERE
1359 statements in arrays or hashes. If absent, the default logic is "or"
1360 for arrays, and "and" for hashes. This means that a WHERE
1364 event_date => {'>=', '2/13/99'},
1365 event_date => {'<=', '4/24/03'},
1368 will generate SQL like this:
1370 WHERE event_date >= '2/13/99' OR event_date <= '4/24/03'
1372 This is probably not what you want given this query, though (look
1373 at the dates). To change the "OR" to an "AND", simply specify:
1375 my $sql = SQL::Abstract->new(logic => 'and');
1377 Which will change the above C<WHERE> to:
1379 WHERE event_date >= '2/13/99' AND event_date <= '4/24/03'
1381 The logic can also be changed locally by inserting
1382 a modifier in front of an arrayref :
1384 @where = (-and => [event_date => {'>=', '2/13/99'},
1385 event_date => {'<=', '4/24/03'} ]);
1387 See the L</"WHERE CLAUSES"> section for explanations.
1391 This will automatically convert comparisons using the specified SQL
1392 function for both column and value. This is mostly used with an argument
1393 of C<upper> or C<lower>, so that the SQL will have the effect of
1394 case-insensitive "searches". For example, this:
1396 $sql = SQL::Abstract->new(convert => 'upper');
1397 %where = (keywords => 'MaKe iT CAse inSeNSItive');
1399 Will turn out the following SQL:
1401 WHERE upper(keywords) like upper('MaKe iT CAse inSeNSItive')
1403 The conversion can be C<upper()>, C<lower()>, or any other SQL function
1404 that can be applied symmetrically to fields (actually B<SQL::Abstract> does
1405 not validate this option; it will just pass through what you specify verbatim).
1409 This is a kludge because many databases suck. For example, you can't
1410 just bind values using DBI's C<execute()> for Oracle C<CLOB> or C<BLOB> fields.
1411 Instead, you have to use C<bind_param()>:
1413 $sth->bind_param(1, 'reg data');
1414 $sth->bind_param(2, $lots, {ora_type => ORA_CLOB});
1416 The problem is, B<SQL::Abstract> will normally just return a C<@bind> array,
1417 which loses track of which field each slot refers to. Fear not.
1419 If you specify C<bindtype> in new, you can determine how C<@bind> is returned.
1420 Currently, you can specify either C<normal> (default) or C<columns>. If you
1421 specify C<columns>, you will get an array that looks like this:
1423 my $sql = SQL::Abstract->new(bindtype => 'columns');
1424 my($stmt, @bind) = $sql->insert(...);
1427 [ 'column1', 'value1' ],
1428 [ 'column2', 'value2' ],
1429 [ 'column3', 'value3' ],
1432 You can then iterate through this manually, using DBI's C<bind_param()>.
1434 $sth->prepare($stmt);
1437 my($col, $data) = @$_;
1438 if ($col eq 'details' || $col eq 'comments') {
1439 $sth->bind_param($i, $data, {ora_type => ORA_CLOB});
1440 } elsif ($col eq 'image') {
1441 $sth->bind_param($i, $data, {ora_type => ORA_BLOB});
1443 $sth->bind_param($i, $data);
1447 $sth->execute; # execute without @bind now
1449 Now, why would you still use B<SQL::Abstract> if you have to do this crap?
1450 Basically, the advantage is still that you don't have to care which fields
1451 are or are not included. You could wrap that above C<for> loop in a simple
1452 sub called C<bind_fields()> or something and reuse it repeatedly. You still
1453 get a layer of abstraction over manual SQL specification.
1455 Note that if you set L</bindtype> to C<columns>, the C<\[$sql, @bind]>
1456 construct (see L</Literal SQL with placeholders and bind values (subqueries)>)
1457 will expect the bind values in this format.
1461 This is the character that a table or column name will be quoted
1462 with. By default this is an empty string, but you could set it to
1463 the character C<`>, to generate SQL like this:
1465 SELECT `a_field` FROM `a_table` WHERE `some_field` LIKE '%someval%'
1467 Alternatively, you can supply an array ref of two items, the first being the left
1468 hand quote character, and the second the right hand quote character. For
1469 example, you could supply C<['[',']']> for SQL Server 2000 compliant quotes
1470 that generates SQL like this:
1472 SELECT [a_field] FROM [a_table] WHERE [some_field] LIKE '%someval%'
1474 Quoting is useful if you have tables or columns names that are reserved
1475 words in your database's SQL dialect.
1479 This is the character that separates a table and column name. It is
1480 necessary to specify this when the C<quote_char> option is selected,
1481 so that tables and column names can be individually quoted like this:
1483 SELECT `table`.`one_field` FROM `table` WHERE `table`.`other_field` = 1
1485 =item array_datatypes
1487 When this option is true, arrayrefs in INSERT or UPDATE are
1488 interpreted as array datatypes and are passed directly
1490 When this option is false, arrayrefs are interpreted
1491 as literal SQL, just like refs to arrayrefs
1492 (but this behavior is for backwards compatibility; when writing
1493 new queries, use the "reference to arrayref" syntax
1499 Takes a reference to a list of "special operators"
1500 to extend the syntax understood by L<SQL::Abstract>.
1501 See section L</"SPECIAL OPERATORS"> for details.
1507 =head2 insert($table, \@values || \%fieldvals)
1509 This is the simplest function. You simply give it a table name
1510 and either an arrayref of values or hashref of field/value pairs.
1511 It returns an SQL INSERT statement and a list of bind values.
1512 See the sections on L</"Inserting and Updating Arrays"> and
1513 L</"Inserting and Updating SQL"> for information on how to insert
1514 with those data types.
1516 =head2 update($table, \%fieldvals, \%where)
1518 This takes a table, hashref of field/value pairs, and an optional
1519 hashref L<WHERE clause|/WHERE CLAUSES>. It returns an SQL UPDATE function and a list
1521 See the sections on L</"Inserting and Updating Arrays"> and
1522 L</"Inserting and Updating SQL"> for information on how to insert
1523 with those data types.
1525 =head2 select($source, $fields, $where, $order)
1527 This returns a SQL SELECT statement and associated list of bind values, as
1528 specified by the arguments :
1534 Specification of the 'FROM' part of the statement.
1535 The argument can be either a plain scalar (interpreted as a table
1536 name, will be quoted), or an arrayref (interpreted as a list
1537 of table names, joined by commas, quoted), or a scalarref
1538 (literal table name, not quoted), or a ref to an arrayref
1539 (list of literal table names, joined by commas, not quoted).
1543 Specification of the list of fields to retrieve from
1545 The argument can be either an arrayref (interpreted as a list
1546 of field names, will be joined by commas and quoted), or a
1547 plain scalar (literal SQL, not quoted).
1548 Please observe that this API is not as flexible as for
1549 the first argument C<$table>, for backwards compatibility reasons.
1553 Optional argument to specify the WHERE part of the query.
1554 The argument is most often a hashref, but can also be
1555 an arrayref or plain scalar --
1556 see section L<WHERE clause|/"WHERE CLAUSES"> for details.
1560 Optional argument to specify the ORDER BY part of the query.
1561 The argument can be a scalar, a hashref or an arrayref
1562 -- see section L<ORDER BY clause|/"ORDER BY CLAUSES">
1568 =head2 delete($table, \%where)
1570 This takes a table name and optional hashref L<WHERE clause|/WHERE CLAUSES>.
1571 It returns an SQL DELETE statement and list of bind values.
1573 =head2 where(\%where, \@order)
1575 This is used to generate just the WHERE clause. For example,
1576 if you have an arbitrary data structure and know what the
1577 rest of your SQL is going to look like, but want an easy way
1578 to produce a WHERE clause, use this. It returns an SQL WHERE
1579 clause and list of bind values.
1582 =head2 values(\%data)
1584 This just returns the values from the hash C<%data>, in the same
1585 order that would be returned from any of the other above queries.
1586 Using this allows you to markedly speed up your queries if you
1587 are affecting lots of rows. See below under the L</"PERFORMANCE"> section.
1589 =head2 generate($any, 'number', $of, \@data, $struct, \%types)
1591 Warning: This is an experimental method and subject to change.
1593 This returns arbitrarily generated SQL. It's a really basic shortcut.
1594 It will return two different things, depending on return context:
1596 my($stmt, @bind) = $sql->generate('create table', \$table, \@fields);
1597 my $stmt_and_val = $sql->generate('create table', \$table, \@fields);
1599 These would return the following:
1601 # First calling form
1602 $stmt = "CREATE TABLE test (?, ?)";
1603 @bind = (field1, field2);
1605 # Second calling form
1606 $stmt_and_val = "CREATE TABLE test (field1, field2)";
1608 Depending on what you're trying to do, it's up to you to choose the correct
1609 format. In this example, the second form is what you would want.
1613 $sql->generate('alter session', { nls_date_format => 'MM/YY' });
1617 ALTER SESSION SET nls_date_format = 'MM/YY'
1619 You get the idea. Strings get their case twiddled, but everything
1620 else remains verbatim.
1625 =head1 WHERE CLAUSES
1629 This module uses a variation on the idea from L<DBIx::Abstract>. It
1630 is B<NOT>, repeat I<not> 100% compatible. B<The main logic of this
1631 module is that things in arrays are OR'ed, and things in hashes
1634 The easiest way to explain is to show lots of examples. After
1635 each C<%where> hash shown, it is assumed you used:
1637 my($stmt, @bind) = $sql->where(\%where);
1639 However, note that the C<%where> hash can be used directly in any
1640 of the other functions as well, as described above.
1642 =head2 Key-value pairs
1644 So, let's get started. To begin, a simple hash:
1648 status => 'completed'
1651 Is converted to SQL C<key = val> statements:
1653 $stmt = "WHERE user = ? AND status = ?";
1654 @bind = ('nwiger', 'completed');
1656 One common thing I end up doing is having a list of values that
1657 a field can be in. To do this, simply specify a list inside of
1662 status => ['assigned', 'in-progress', 'pending'];
1665 This simple code will create the following:
1667 $stmt = "WHERE user = ? AND ( status = ? OR status = ? OR status = ? )";
1668 @bind = ('nwiger', 'assigned', 'in-progress', 'pending');
1670 A field associated to an empty arrayref will be considered a
1671 logical false and will generate 0=1.
1673 =head2 Specific comparison operators
1675 If you want to specify a different type of operator for your comparison,
1676 you can use a hashref for a given column:
1680 status => { '!=', 'completed' }
1683 Which would generate:
1685 $stmt = "WHERE user = ? AND status != ?";
1686 @bind = ('nwiger', 'completed');
1688 To test against multiple values, just enclose the values in an arrayref:
1690 status => { '=', ['assigned', 'in-progress', 'pending'] };
1692 Which would give you:
1694 "WHERE status = ? OR status = ? OR status = ?"
1697 The hashref can also contain multiple pairs, in which case it is expanded
1698 into an C<AND> of its elements:
1702 status => { '!=', 'completed', -not_like => 'pending%' }
1705 # Or more dynamically, like from a form
1706 $where{user} = 'nwiger';
1707 $where{status}{'!='} = 'completed';
1708 $where{status}{'-not_like'} = 'pending%';
1710 # Both generate this
1711 $stmt = "WHERE user = ? AND status != ? AND status NOT LIKE ?";
1712 @bind = ('nwiger', 'completed', 'pending%');
1715 To get an OR instead, you can combine it with the arrayref idea:
1719 priority => [ {'=', 2}, {'!=', 1} ]
1722 Which would generate:
1724 $stmt = "WHERE user = ? AND priority = ? OR priority != ?";
1725 @bind = ('nwiger', '2', '1');
1727 If you want to include literal SQL (with or without bind values), just use a
1728 scalar reference or array reference as the value:
1731 date_entered => { '>' => \["to_date(?, 'MM/DD/YYYY')", "11/26/2008"] },
1732 date_expires => { '<' => \"now()" }
1735 Which would generate:
1737 $stmt = "WHERE date_entered > "to_date(?, 'MM/DD/YYYY') AND date_expires < now()";
1738 @bind = ('11/26/2008');
1741 =head2 Logic and nesting operators
1743 In the example above,
1744 there is a subtle trap if you want to say something like
1745 this (notice the C<AND>):
1747 WHERE priority != ? AND priority != ?
1749 Because, in Perl you I<can't> do this:
1751 priority => { '!=', 2, '!=', 1 }
1753 As the second C<!=> key will obliterate the first. The solution
1754 is to use the special C<-modifier> form inside an arrayref:
1756 priority => [ -and => {'!=', 2},
1760 Normally, these would be joined by C<OR>, but the modifier tells it
1761 to use C<AND> instead. (Hint: You can use this in conjunction with the
1762 C<logic> option to C<new()> in order to change the way your queries
1763 work by default.) B<Important:> Note that the C<-modifier> goes
1764 B<INSIDE> the arrayref, as an extra first element. This will
1765 B<NOT> do what you think it might:
1767 priority => -and => [{'!=', 2}, {'!=', 1}] # WRONG!
1769 Here is a quick list of equivalencies, since there is some overlap:
1772 status => {'!=', 'completed', 'not like', 'pending%' }
1773 status => [ -and => {'!=', 'completed'}, {'not like', 'pending%'}]
1776 status => {'=', ['assigned', 'in-progress']}
1777 status => [ -or => {'=', 'assigned'}, {'=', 'in-progress'}]
1778 status => [ {'=', 'assigned'}, {'=', 'in-progress'} ]
1782 =head2 Special operators : IN, BETWEEN, etc.
1784 You can also use the hashref format to compare a list of fields using the
1785 C<IN> comparison operator, by specifying the list as an arrayref:
1788 status => 'completed',
1789 reportid => { -in => [567, 2335, 2] }
1792 Which would generate:
1794 $stmt = "WHERE status = ? AND reportid IN (?,?,?)";
1795 @bind = ('completed', '567', '2335', '2');
1797 The reverse operator C<-not_in> generates SQL C<NOT IN> and is used in
1800 If the argument to C<-in> is an empty array, 'sqlfalse' is generated
1801 (by default : C<1=0>). Similarly, C<< -not_in => [] >> generates
1802 'sqltrue' (by default : C<1=1>).
1806 Another pair of operators is C<-between> and C<-not_between>,
1807 used with an arrayref of two values:
1811 completion_date => {
1812 -not_between => ['2002-10-01', '2003-02-06']
1818 WHERE user = ? AND completion_date NOT BETWEEN ( ? AND ? )
1820 These are the two builtin "special operators"; but the
1821 list can be expanded : see section L</"SPECIAL OPERATORS"> below.
1823 =head2 Nested conditions, -and/-or prefixes
1825 So far, we've seen how multiple conditions are joined with a top-level
1826 C<AND>. We can change this by putting the different conditions we want in
1827 hashes and then putting those hashes in an array. For example:
1832 status => { -like => ['pending%', 'dispatched'] },
1836 status => 'unassigned',
1840 This data structure would create the following:
1842 $stmt = "WHERE ( user = ? AND ( status LIKE ? OR status LIKE ? ) )
1843 OR ( user = ? AND status = ? ) )";
1844 @bind = ('nwiger', 'pending', 'dispatched', 'robot', 'unassigned');
1847 There is also a special C<-nest>
1848 operator which adds an additional set of parens, to create a subquery.
1849 For example, to get something like this:
1851 $stmt = "WHERE user = ? AND ( workhrs > ? OR geo = ? )";
1852 @bind = ('nwiger', '20', 'ASIA');
1858 -nest => [ workhrs => {'>', 20}, geo => 'ASIA' ],
1862 Finally, clauses in hashrefs or arrayrefs can be
1863 prefixed with an C<-and> or C<-or> to change the logic
1870 -and => [workhrs => {'>', 20}, geo => 'ASIA' ],
1871 -and => [workhrs => {'<', 50}, geo => 'EURO' ]
1878 WHERE ( user = ? AND
1879 ( ( workhrs > ? AND geo = ? )
1880 OR ( workhrs < ? AND geo = ? ) ) )
1883 =head2 Algebraic inconsistency, for historical reasons
1885 C<Important note>: when connecting several conditions, the C<-and->|C<-or>
1886 operator goes C<outside> of the nested structure; whereas when connecting
1887 several constraints on one column, the C<-and> operator goes
1888 C<inside> the arrayref. Here is an example combining both features :
1891 -and => [a => 1, b => 2],
1892 -or => [c => 3, d => 4],
1893 e => [-and => {-like => 'foo%'}, {-like => '%bar'} ]
1898 WHERE ( ( ( a = ? AND b = ? )
1899 OR ( c = ? OR d = ? )
1900 OR ( e LIKE ? AND e LIKE ? ) ) )
1902 This difference in syntax is unfortunate but must be preserved for
1903 historical reasons. So be careful : the two examples below would
1904 seem algebraically equivalent, but they are not
1906 {col => [-and => {-like => 'foo%'}, {-like => '%bar'}]}
1907 # yields : WHERE ( ( col LIKE ? AND col LIKE ? ) )
1909 [-and => {col => {-like => 'foo%'}, {col => {-like => '%bar'}}]]
1910 # yields : WHERE ( ( col LIKE ? OR col LIKE ? ) )
1915 Finally, sometimes only literal SQL will do. If you want to include
1916 literal SQL verbatim, you can specify it as a scalar reference, namely:
1918 my $inn = 'is Not Null';
1920 priority => { '<', 2 },
1926 $stmt = "WHERE priority < ? AND requestor is Not Null";
1929 Note that in this example, you only get one bind parameter back, since
1930 the verbatim SQL is passed as part of the statement.
1932 Of course, just to prove a point, the above can also be accomplished
1936 priority => { '<', 2 },
1937 requestor => { '!=', undef },
1943 Conditions on boolean columns can be expressed in the
1944 same way, passing a reference to an empty string :
1947 priority => { '<', 2 },
1953 $stmt = "WHERE priority < ? AND is_ready";
1957 =head2 Literal SQL with placeholders and bind values (subqueries)
1959 If the literal SQL to be inserted has placeholders and bind values,
1960 use a reference to an arrayref (yes this is a double reference --
1961 not so common, but perfectly legal Perl). For example, to find a date
1962 in Postgres you can use something like this:
1965 date_column => \[q/= date '2008-09-30' - ?::integer/, 10/]
1970 $stmt = "WHERE ( date_column = date '2008-09-30' - ?::integer )"
1973 Note that you must pass the bind values in the same format as they are returned
1974 by L</where>. That means that if you set L</bindtype> to C<columns>, you must
1975 provide the bind values in the C<< [ column_meta => value ] >> format, where
1976 C<column_meta> is an opaque scalar value; most commonly the column name, but
1977 you can use any scalar value (including references and blessed references),
1978 L<SQL::Abstract> will simply pass it through intact. So if C<bindtype> is set
1979 to C<columns> the above example will look like:
1982 date_column => \[q/= date '2008-09-30' - ?::integer/, [ dummy => 10 ]/]
1985 Literal SQL is especially useful for nesting parenthesized clauses in the
1986 main SQL query. Here is a first example :
1988 my ($sub_stmt, @sub_bind) = ("SELECT c1 FROM t1 WHERE c2 < ? AND c3 LIKE ?",
1992 bar => \["IN ($sub_stmt)" => @sub_bind],
1997 $stmt = "WHERE (foo = ? AND bar IN (SELECT c1 FROM t1
1998 WHERE c2 < ? AND c3 LIKE ?))";
1999 @bind = (1234, 100, "foo%");
2001 Other subquery operators, like for example C<"E<gt> ALL"> or C<"NOT IN">,
2002 are expressed in the same way. Of course the C<$sub_stmt> and
2003 its associated bind values can be generated through a former call
2006 my ($sub_stmt, @sub_bind)
2007 = $sql->select("t1", "c1", {c2 => {"<" => 100},
2008 c3 => {-like => "foo%"}});
2011 bar => \["> ALL ($sub_stmt)" => @sub_bind],
2014 In the examples above, the subquery was used as an operator on a column;
2015 but the same principle also applies for a clause within the main C<%where>
2016 hash, like an EXISTS subquery :
2018 my ($sub_stmt, @sub_bind)
2019 = $sql->select("t1", "*", {c1 => 1, c2 => \"> t0.c0"});
2022 -nest => \["EXISTS ($sub_stmt)" => @sub_bind],
2027 $stmt = "WHERE (foo = ? AND EXISTS (SELECT * FROM t1
2028 WHERE c1 = ? AND c2 > t0.c0))";
2032 Observe that the condition on C<c2> in the subquery refers to
2033 column C<t0.c0> of the main query : this is I<not> a bind
2034 value, so we have to express it through a scalar ref.
2035 Writing C<< c2 => {">" => "t0.c0"} >> would have generated
2036 C<< c2 > ? >> with bind value C<"t0.c0"> ... not exactly
2037 what we wanted here.
2039 Another use of the subquery technique is when some SQL clauses need
2040 parentheses, as it often occurs with some proprietary SQL extensions
2041 like for example fulltext expressions, geospatial expressions,
2042 NATIVE clauses, etc. Here is an example of a fulltext query in MySQL :
2045 -nest => \["MATCH (col1, col2) AGAINST (?)" => qw/apples/]
2048 Finally, here is an example where a subquery is used
2049 for expressing unary negation:
2051 my ($sub_stmt, @sub_bind)
2052 = $sql->where({age => [{"<" => 10}, {">" => 20}]});
2053 $sub_stmt =~ s/^ where //i; # don't want "WHERE" in the subclause
2055 lname => {like => '%son%'},
2056 -nest => \["NOT ($sub_stmt)" => @sub_bind],
2061 $stmt = "lname LIKE ? AND NOT ( age < ? OR age > ? )"
2062 @bind = ('%son%', 10, 20)
2068 These pages could go on for a while, since the nesting of the data
2069 structures this module can handle are pretty much unlimited (the
2070 module implements the C<WHERE> expansion as a recursive function
2071 internally). Your best bet is to "play around" with the module a
2072 little to see how the data structures behave, and choose the best
2073 format for your data based on that.
2075 And of course, all the values above will probably be replaced with
2076 variables gotten from forms or the command line. After all, if you
2077 knew everything ahead of time, you wouldn't have to worry about
2078 dynamically-generating SQL and could just hardwire it into your
2084 =head1 ORDER BY CLAUSES
2086 Some functions take an order by clause. This can either be a scalar (just a
2087 column name,) a hash of C<< { -desc => 'col' } >> or C<< { -asc => 'col' } >>,
2088 or an array of either of the two previous forms. Examples:
2090 Given | Will Generate
2091 ----------------------------------------------------------
2093 \'colA DESC' | ORDER BY colA DESC
2095 'colA' | ORDER BY colA
2097 [qw/colA colB/] | ORDER BY colA, colB
2099 {-asc => 'colA'} | ORDER BY colA ASC
2101 {-desc => 'colB'} | ORDER BY colB DESC
2103 ['colA', {-asc => 'colB'}] | ORDER BY colA, colB ASC
2105 { -asc => [qw/colA colB] } | ORDER BY colA ASC, colB ASC
2108 { -asc => 'colA' }, | ORDER BY colA ASC, colB DESC,
2109 { -desc => [qw/colB/], | colC ASC, colD ASC
2110 { -asc => [qw/colC colD/],|
2112 ===========================================================
2116 =head1 SPECIAL OPERATORS
2118 my $sqlmaker = SQL::Abstract->new(special_ops => [
2121 my ($self, $field, $op, $arg) = @_;
2127 A "special operator" is a SQL syntactic clause that can be
2128 applied to a field, instead of a usual binary operator.
2131 WHERE field IN (?, ?, ?)
2132 WHERE field BETWEEN ? AND ?
2133 WHERE MATCH(field) AGAINST (?, ?)
2135 Special operators IN and BETWEEN are fairly standard and therefore
2136 are builtin within C<SQL::Abstract>. For other operators,
2137 like the MATCH .. AGAINST example above which is
2138 specific to MySQL, you can write your own operator handlers :
2139 supply a C<special_ops> argument to the C<new> method.
2140 That argument takes an arrayref of operator definitions;
2141 each operator definition is a hashref with two entries
2147 the regular expression to match the operator
2151 coderef that will be called when meeting that operator
2152 in the input tree. The coderef will be called with
2153 arguments C<< ($self, $field, $op, $arg) >>, and
2154 should return a C<< ($sql, @bind) >> structure.
2158 For example, here is an implementation
2159 of the MATCH .. AGAINST syntax for MySQL
2161 my $sqlmaker = SQL::Abstract->new(special_ops => [
2163 # special op for MySql MATCH (field) AGAINST(word1, word2, ...)
2164 {regex => qr/^match$/i,
2166 my ($self, $field, $op, $arg) = @_;
2167 $arg = [$arg] if not ref $arg;
2168 my $label = $self->_quote($field);
2169 my ($placeholder) = $self->_convert('?');
2170 my $placeholders = join ", ", (($placeholder) x @$arg);
2171 my $sql = $self->_sqlcase('match') . " ($label) "
2172 . $self->_sqlcase('against') . " ($placeholders) ";
2173 my @bind = $self->_bindtype($field, @$arg);
2174 return ($sql, @bind);
2183 Thanks to some benchmarking by Mark Stosberg, it turns out that
2184 this module is many orders of magnitude faster than using C<DBIx::Abstract>.
2185 I must admit this wasn't an intentional design issue, but it's a
2186 byproduct of the fact that you get to control your C<DBI> handles
2189 To maximize performance, use a code snippet like the following:
2191 # prepare a statement handle using the first row
2192 # and then reuse it for the rest of the rows
2194 for my $href (@array_of_hashrefs) {
2195 $stmt ||= $sql->insert('table', $href);
2196 $sth ||= $dbh->prepare($stmt);
2197 $sth->execute($sql->values($href));
2200 The reason this works is because the keys in your C<$href> are sorted
2201 internally by B<SQL::Abstract>. Thus, as long as your data retains
2202 the same structure, you only have to generate the SQL the first time
2203 around. On subsequent queries, simply use the C<values> function provided
2204 by this module to return your values in the correct order.
2209 If you use my C<CGI::FormBuilder> module at all, you'll hopefully
2210 really like this part (I do, at least). Building up a complex query
2211 can be as simple as the following:
2215 use CGI::FormBuilder;
2218 my $form = CGI::FormBuilder->new(...);
2219 my $sql = SQL::Abstract->new;
2221 if ($form->submitted) {
2222 my $field = $form->field;
2223 my $id = delete $field->{id};
2224 my($stmt, @bind) = $sql->update('table', $field, {id => $id});
2227 Of course, you would still have to connect using C<DBI> to run the
2228 query, but the point is that if you make your form look like your
2229 table, the actual query script can be extremely simplistic.
2231 If you're B<REALLY> lazy (I am), check out C<HTML::QuickTable> for
2232 a fast interface to returning and formatting data. I frequently
2233 use these three modules together to write complex database query
2234 apps in under 50 lines.
2239 Version 1.50 was a major internal refactoring of C<SQL::Abstract>.
2240 Great care has been taken to preserve the I<published> behavior
2241 documented in previous versions in the 1.* family; however,
2242 some features that were previously undocumented, or behaved
2243 differently from the documentation, had to be changed in order
2244 to clarify the semantics. Hence, client code that was relying
2245 on some dark areas of C<SQL::Abstract> v1.*
2246 B<might behave differently> in v1.50.
2248 The main changes are :
2254 support for literal SQL through the C<< \ [$sql, bind] >> syntax.
2258 support for the { operator => \"..." } construct (to embed literal SQL)
2262 support for the { operator => \["...", @bind] } construct (to embed literal SQL with bind values)
2266 optional support for L<array datatypes|/"Inserting and Updating Arrays">
2270 defensive programming : check arguments
2274 fixed bug with global logic, which was previously implemented
2275 through global variables yielding side-effects. Prior versions would
2276 interpret C<< [ {cond1, cond2}, [cond3, cond4] ] >>
2277 as C<< "(cond1 AND cond2) OR (cond3 AND cond4)" >>.
2278 Now this is interpreted
2279 as C<< "(cond1 AND cond2) OR (cond3 OR cond4)" >>.
2284 fixed semantics of _bindtype on array args
2288 dropped the C<_anoncopy> of the %where tree. No longer necessary,
2289 we just avoid shifting arrays within that tree.
2293 dropped the C<_modlogic> function
2299 =head1 ACKNOWLEDGEMENTS
2301 There are a number of individuals that have really helped out with
2302 this module. Unfortunately, most of them submitted bugs via CPAN
2303 so I have no idea who they are! But the people I do know are:
2305 Ash Berlin (order_by hash term support)
2306 Matt Trout (DBIx::Class support)
2307 Mark Stosberg (benchmarking)
2308 Chas Owens (initial "IN" operator support)
2309 Philip Collins (per-field SQL functions)
2310 Eric Kolve (hashref "AND" support)
2311 Mike Fragassi (enhancements to "BETWEEN" and "LIKE")
2312 Dan Kubb (support for "quote_char" and "name_sep")
2313 Guillermo Roditi (patch to cleanup "IN" and "BETWEEN", fix and tests for _order_by)
2314 Laurent Dami (internal refactoring, multiple -nest, extensible list of special operators, literal SQL)
2315 Norbert Buchmuller (support for literal SQL in hashpair, misc. fixes & tests)
2321 L<DBIx::Class>, L<DBIx::Abstract>, L<CGI::FormBuilder>, L<HTML::QuickTable>.
2325 Copyright (c) 2001-2007 Nathan Wiger <nwiger@cpan.org>. All Rights Reserved.
2327 This module is actively maintained by Matt Trout <mst@shadowcatsystems.co.uk>
2329 For support, your best bet is to try the C<DBIx::Class> users mailing list.
2330 While not an official support venue, C<DBIx::Class> makes heavy use of
2331 C<SQL::Abstract>, and as such list members there are very familiar with
2332 how to create queries.
2334 This module is free software; you may copy this under the terms of
2335 the GNU General Public License, or the Artistic License, copies of
2336 which should have accompanied your Perl kit.