1 package SQL::Abstract; # see doc at end of file
10 our @EXPORT_OK = qw(is_plain_value is_literal_value);
20 *SQL::Abstract::_ENV_::DETECT_AUTOGENERATED_STRINGIFICATION = $ENV{SQLA_ISVALUE_IGNORE_AUTOGENERATED_STRINGIFICATION}
26 #======================================================================
28 #======================================================================
30 our $VERSION = '1.86';
32 # This would confuse some packagers
33 $VERSION = eval $VERSION if $VERSION =~ /_/; # numify for warning-free dev releases
37 # special operators (-in, -between). May be extended/overridden by user.
38 # See section WHERE: BUILTIN SPECIAL OPERATORS below for implementation
39 my @BUILTIN_SPECIAL_OPS = (
40 {regex => qr/^ (?: not \s )? between $/ix, handler => sub { die "NOPE" }},
41 {regex => qr/^ (?: not \s )? in $/ix, handler => sub { die "NOPE" }},
42 {regex => qr/^ is (?: \s+ not )? $/ix, handler => sub { die "NOPE" }},
45 #======================================================================
46 # DEBUGGING AND ERROR REPORTING
47 #======================================================================
50 return unless $_[0]->{debug}; shift; # a little faster
51 my $func = (caller(1))[3];
52 warn "[$func] ", @_, "\n";
56 my($func) = (caller(1))[3];
57 Carp::carp "[$func] Warning: ", @_;
61 my($func) = (caller(1))[3];
62 Carp::croak "[$func] Fatal: ", @_;
65 sub is_literal_value ($) {
66 ref $_[0] eq 'SCALAR' ? [ ${$_[0]} ]
67 : ( ref $_[0] eq 'REF' and ref ${$_[0]} eq 'ARRAY' ) ? [ @${ $_[0] } ]
71 # FIXME XSify - this can be done so much more efficiently
72 sub is_plain_value ($) {
74 ! length ref $_[0] ? \($_[0])
76 ref $_[0] eq 'HASH' and keys %{$_[0]} == 1
78 exists $_[0]->{-value}
79 ) ? \($_[0]->{-value})
81 # reuse @_ for even moar speedz
82 defined ( $_[1] = Scalar::Util::blessed $_[0] )
84 # deliberately not using Devel::OverloadInfo - the checks we are
85 # intersted in are much more limited than the fullblown thing, and
86 # this is a very hot piece of code
88 # simply using ->can('(""') can leave behind stub methods that
89 # break actually using the overload later (see L<perldiag/Stub
90 # found while resolving method "%s" overloading "%s" in package
91 # "%s"> and the source of overload::mycan())
93 # either has stringification which DBI SHOULD prefer out of the box
94 grep { *{ (qq[${_}::(""]) }{CODE} } @{ $_[2] = mro::get_linear_isa( $_[1] ) }
96 # has nummification or boolification, AND fallback is *not* disabled
98 SQL::Abstract::_ENV_::DETECT_AUTOGENERATED_STRINGIFICATION
101 grep { *{"${_}::(0+"}{CODE} } @{$_[2]}
103 grep { *{"${_}::(bool"}{CODE} } @{$_[2]}
107 # no fallback specified at all
108 ! ( ($_[3]) = grep { *{"${_}::()"}{CODE} } @{$_[2]} )
110 # fallback explicitly undef
111 ! defined ${"$_[3]::()"}
124 #======================================================================
126 #======================================================================
130 my $class = ref($self) || $self;
131 my %opt = (ref $_[0] eq 'HASH') ? %{$_[0]} : @_;
133 # choose our case by keeping an option around
134 delete $opt{case} if $opt{case} && $opt{case} ne 'lower';
136 # default logic for interpreting arrayrefs
137 $opt{logic} = $opt{logic} ? uc $opt{logic} : 'OR';
139 # how to return bind vars
140 $opt{bindtype} ||= 'normal';
142 # default comparison is "=", but can be overridden
145 # try to recognize which are the 'equality' and 'inequality' ops
146 # (temporary quickfix (in 2007), should go through a more seasoned API)
147 $opt{equality_op} = qr/^( \Q$opt{cmp}\E | \= )$/ix;
148 $opt{inequality_op} = qr/^( != | <> )$/ix;
150 $opt{like_op} = qr/^ (is\s+)? r?like $/xi;
151 $opt{not_like_op} = qr/^ (is\s+)? not \s+ r?like $/xi;
154 $opt{sqltrue} ||= '1=1';
155 $opt{sqlfalse} ||= '0=1';
158 $opt{special_ops} ||= [];
160 # regexes are applied in order, thus push after user-defines
161 push @{$opt{special_ops}}, @BUILTIN_SPECIAL_OPS;
164 $opt{unary_ops} ||= [];
166 # rudimentary sanity-check for user supplied bits treated as functions/operators
167 # If a purported function matches this regular expression, an exception is thrown.
168 # Literal SQL is *NOT* subject to this check, only functions (and column names
169 # when quoting is not in effect)
172 # need to guard against ()'s in column names too, but this will break tons of
173 # hacks... ideas anyone?
174 $opt{injection_guard} ||= qr/
180 $opt{node_types} = +{
181 map +("-$_" => '_render_'.$_),
182 qw(op func value bind ident literal)
185 $opt{expand_unary} = {};
187 return bless \%opt, $class;
190 sub sqltrue { +{ -literal => [ $_[0]->{sqltrue} ] } }
191 sub sqlfalse { +{ -literal => [ $_[0]->{sqlfalse} ] } }
193 sub _assert_pass_injection_guard {
194 if ($_[1] =~ $_[0]->{injection_guard}) {
195 my $class = ref $_[0];
196 puke "Possible SQL injection attempt '$_[1]'. If this is indeed a part of the "
197 . "desired SQL use literal SQL ( \'...' or \[ '...' ] ) or supply your own "
198 . "{injection_guard} attribute to ${class}->new()"
203 #======================================================================
205 #======================================================================
209 my $table = $self->_table(shift);
210 my $data = shift || return;
213 my $method = $self->_METHOD_FOR_refkind("_insert", $data);
214 my ($sql, @bind) = $self->$method($data);
215 $sql = join " ", $self->_sqlcase('insert into'), $table, $sql;
217 if ($options->{returning}) {
218 my ($s, @b) = $self->_insert_returning($options);
223 return wantarray ? ($sql, @bind) : $sql;
226 # So that subclasses can override INSERT ... RETURNING separately from
227 # UPDATE and DELETE (e.g. DBIx::Class::SQLMaker::Oracle does this)
228 sub _insert_returning { shift->_returning(@_) }
231 my ($self, $options) = @_;
233 my $f = $options->{returning};
235 my ($sql, @bind) = $self->_render_expr(
236 $self->_expand_maybe_list_expr($f, undef, -ident)
239 ? $self->_sqlcase(' returning ') . $sql
240 : ($self->_sqlcase(' returning ').$sql, @bind);
243 sub _insert_HASHREF { # explicit list of fields and then values
244 my ($self, $data) = @_;
246 my @fields = sort keys %$data;
248 my ($sql, @bind) = $self->_insert_values($data);
251 $_ = $self->_quote($_) foreach @fields;
252 $sql = "( ".join(", ", @fields).") ".$sql;
254 return ($sql, @bind);
257 sub _insert_ARRAYREF { # just generate values(?,?) part (no list of fields)
258 my ($self, $data) = @_;
260 # no names (arrayref) so can't generate bindtype
261 $self->{bindtype} ne 'columns'
262 or belch "can't do 'columns' bindtype when called with arrayref";
264 my (@values, @all_bind);
265 foreach my $value (@$data) {
266 my ($values, @bind) = $self->_insert_value(undef, $value);
267 push @values, $values;
268 push @all_bind, @bind;
270 my $sql = $self->_sqlcase('values')." ( ".join(", ", @values)." )";
271 return ($sql, @all_bind);
274 sub _insert_ARRAYREFREF { # literal SQL with bind
275 my ($self, $data) = @_;
277 my ($sql, @bind) = @${$data};
278 $self->_assert_bindval_matches_bindtype(@bind);
280 return ($sql, @bind);
284 sub _insert_SCALARREF { # literal SQL without bind
285 my ($self, $data) = @_;
291 my ($self, $data) = @_;
293 my (@values, @all_bind);
294 foreach my $column (sort keys %$data) {
295 my ($values, @bind) = $self->_insert_value($column, $data->{$column});
296 push @values, $values;
297 push @all_bind, @bind;
299 my $sql = $self->_sqlcase('values')." ( ".join(", ", @values)." )";
300 return ($sql, @all_bind);
304 my ($self, $column, $v) = @_;
306 return $self->_render_expr(
307 $self->_expand_insert_value($column, $v)
311 sub _expand_insert_value {
312 my ($self, $column, $v) = @_;
314 if (ref($v) eq 'ARRAY') {
315 if ($self->{array_datatypes}) {
316 return +{ -bind => [ $column, $v ] };
318 my ($sql, @bind) = @$v;
319 $self->_assert_bindval_matches_bindtype(@bind);
320 return +{ -literal => $v };
322 if (ref($v) eq 'HASH') {
323 if (grep !/^-/, keys %$v) {
324 belch "HASH ref as bind value in insert is not supported";
325 return +{ -bind => [ $column, $v ] };
329 return +{ -bind => [ $column, undef ] };
331 local our $Cur_Col_Meta = $column;
332 return $self->_expand_expr($v);
337 #======================================================================
339 #======================================================================
344 my $table = $self->_table(shift);
345 my $data = shift || return;
349 # first build the 'SET' part of the sql statement
350 puke "Unsupported data type specified to \$sql->update"
351 unless ref $data eq 'HASH';
353 my ($sql, @all_bind) = $self->_update_set_values($data);
354 $sql = $self->_sqlcase('update ') . $table . $self->_sqlcase(' set ')
358 my($where_sql, @where_bind) = $self->where($where);
360 push @all_bind, @where_bind;
363 if ($options->{returning}) {
364 my ($returning_sql, @returning_bind) = $self->_update_returning($options);
365 $sql .= $returning_sql;
366 push @all_bind, @returning_bind;
369 return wantarray ? ($sql, @all_bind) : $sql;
372 sub _update_set_values {
373 my ($self, $data) = @_;
375 return $self->_render_expr(
376 $self->_expand_update_set_values($data),
380 sub _expand_update_set_values {
381 my ($self, $data) = @_;
382 $self->_expand_maybe_list_expr( [
385 $set = { -bind => $_ } unless defined $set;
386 +{ -op => [ '=', { -ident => $k }, $set ] };
392 ? ($self->{array_datatypes}
393 ? [ $k, +{ -bind => [ $k, $v ] } ]
394 : [ $k, +{ -literal => $v } ])
396 local our $Cur_Col_Meta = $k;
397 [ $k, $self->_expand_expr($v) ]
404 # So that subclasses can override UPDATE ... RETURNING separately from
406 sub _update_returning { shift->_returning(@_) }
410 #======================================================================
412 #======================================================================
417 my $table = $self->_table(shift);
418 my $fields = shift || '*';
422 my ($fields_sql, @bind) = $self->_select_fields($fields);
424 my ($where_sql, @where_bind) = $self->where($where, $order);
425 push @bind, @where_bind;
427 my $sql = join(' ', $self->_sqlcase('select'), $fields_sql,
428 $self->_sqlcase('from'), $table)
431 return wantarray ? ($sql, @bind) : $sql;
435 my ($self, $fields) = @_;
436 return $fields unless ref($fields);
437 return $self->_render_expr(
438 $self->_expand_maybe_list_expr($fields, undef, '-ident')
442 #======================================================================
444 #======================================================================
449 my $table = $self->_table(shift);
453 my($where_sql, @bind) = $self->where($where);
454 my $sql = $self->_sqlcase('delete from ') . $table . $where_sql;
456 if ($options->{returning}) {
457 my ($returning_sql, @returning_bind) = $self->_delete_returning($options);
458 $sql .= $returning_sql;
459 push @bind, @returning_bind;
462 return wantarray ? ($sql, @bind) : $sql;
465 # So that subclasses can override DELETE ... RETURNING separately from
467 sub _delete_returning { shift->_returning(@_) }
471 #======================================================================
473 #======================================================================
477 # Finally, a separate routine just to handle WHERE clauses
479 my ($self, $where, $order) = @_;
481 local $self->{convert_where} = $self->{convert};
484 my ($sql, @bind) = defined($where)
485 ? $self->_recurse_where($where)
487 $sql = (defined $sql and length $sql) ? $self->_sqlcase(' where ') . "( $sql )" : '';
491 my ($order_sql, @order_bind) = $self->_order_by($order);
493 push @bind, @order_bind;
496 return wantarray ? ($sql, @bind) : $sql;
500 my ($self, $expr, $logic, $default_scalar_to) = @_;
501 local our $Default_Scalar_To = $default_scalar_to if $default_scalar_to;
502 our $Expand_Depth ||= 0; local $Expand_Depth = $Expand_Depth + 1;
503 return undef unless defined($expr);
504 if (ref($expr) eq 'HASH') {
505 if (keys %$expr > 1) {
509 map $self->_expand_expr_hashpair($_ => $expr->{$_}, $logic),
513 return { -literal => [ '' ] } unless keys %$expr;
514 return $self->_expand_expr_hashpair(%$expr, $logic);
516 if (ref($expr) eq 'ARRAY') {
517 my $logic = lc($logic || $self->{logic});
518 $logic eq 'and' or $logic eq 'or' or puke "unknown logic: $logic";
522 (ref($_) eq 'ARRAY' and @$_)
523 or (ref($_) eq 'HASH' and %$_)
529 while (my ($el) = splice @expr, 0, 1) {
530 puke "Supplying an empty left hand side argument is not supported in array-pairs"
531 unless defined($el) and length($el);
532 my $elref = ref($el);
534 push(@res, $self->_expand_expr({ $el, shift(@expr) }));
535 } elsif ($elref eq 'ARRAY') {
536 push(@res, $self->_expand_expr($el)) if @$el;
537 } elsif (my $l = is_literal_value($el)) {
538 push @res, { -literal => $l };
539 } elsif ($elref eq 'HASH') {
540 push @res, $self->_expand_expr($el) if %$el;
545 return { -op => [ $logic, @res ] };
547 if (my $literal = is_literal_value($expr)) {
548 return +{ -literal => $literal };
550 if (!ref($expr) or Scalar::Util::blessed($expr)) {
551 if (my $d = $Default_Scalar_To) {
552 return +{ $d => $expr };
554 if (my $m = our $Cur_Col_Meta) {
555 return +{ -bind => [ $m, $expr ] };
557 return +{ -value => $expr };
564 sub _expand_expr_hashpair {
565 my ($self, $k, $v, $logic) = @_;
566 unless (defined($k) and length($k)) {
567 if (defined($k) and my $literal = is_literal_value($v)) {
568 belch 'Hash-pairs consisting of an empty string with a literal are deprecated, and will be removed in 2.0: use -and => [ $literal ] instead';
569 return { -literal => $literal };
571 puke "Supplying an empty left hand side argument is not supported";
574 $self->_assert_pass_injection_guard($k =~ /^-(.*)$/s);
575 if ($k =~ s/ [_\s]? \d+ $//x ) {
576 belch 'Use of [and|or|nest]_N modifiers is deprecated and will be removed in SQLA v2.0. '
577 . "You probably wanted ...-and => [ $k => COND1, $k => COND2 ... ]";
580 # DBIx::Class requires a nest warning to be emitted once but the private
581 # method it overrode to do so no longer exists
582 if (ref($self) =~ /^DBIx::Class::SQLMaker/) {
583 unless ($Nest_Warned) {
585 "-nest in search conditions is deprecated, you most probably wanted:\n"
586 .q|{..., -and => [ \%cond0, \@cond1, \'cond2', \[ 'cond3', [ col => bind ] ], etc. ], ... }|
591 return $self->_expand_expr($v);
595 return $self->_expand_expr($v);
597 puke "-bool => undef not supported" unless defined($v);
598 return { -ident => $v };
601 return { -op => [ 'not', $self->_expand_expr($v) ] };
603 if (my ($rest) = $k =~/^-not[_ ](.*)$/) {
606 $self->_expand_expr_hashpair("-${rest}", $v, $logic)
609 if (my ($logic) = $k =~ /^-(and|or)$/i) {
610 if (ref($v) eq 'HASH') {
611 return $self->_expand_expr($v, $logic);
613 if (ref($v) eq 'ARRAY') {
614 return $self->_expand_expr($v, $logic);
619 $op =~ s/^-// if length($op) > 1;
621 # top level special ops are illegal in general
622 # note that, arguably, if it makes no sense at top level, it also
623 # makes no sense on the other side of an = sign or similar but DBIC
624 # gets disappointingly upset if I disallow it
626 (our $Expand_Depth) == 1
627 and List::Util::first { $op =~ $_->{regex} } @{$self->{special_ops}}
629 puke "Illegal use of top-level '-$op'"
631 if (my $us = List::Util::first { $op =~ $_->{regex} } @{$self->{unary_ops}}) {
632 return { -op => [ $op, $v ] };
635 if ($k eq '-value' and my $m = our $Cur_Col_Meta) {
636 return +{ -bind => [ $m, $v ] };
638 if (my $custom = $self->{expand_unary}{$k}) {
639 return $self->$custom($v);
641 if ($self->{node_types}{$k}) {
647 and (keys %$v)[0] =~ /^-/
649 my ($func) = $k =~ /^-(.*)$/;
650 if (List::Util::first { $func =~ $_->{regex} } @{$self->{special_ops}}) {
651 return +{ -op => [ $func, $self->_expand_expr($v) ] };
653 return +{ -func => [ $func, $self->_expand_expr($v) ] };
655 if (!ref($v) or is_literal_value($v)) {
656 return +{ -op => [ $k =~ /^-(.*)$/, $self->_expand_expr($v) ] };
663 and exists $v->{-value}
664 and not defined $v->{-value}
667 return $self->_expand_expr_hashpair($k => { $self->{cmp} => undef });
669 if (!ref($v) or Scalar::Util::blessed($v)) {
670 my $d = our $Default_Scalar_To;
675 ($d ? { $d => $v } : { -bind => [ $k, $v ] })
679 if (ref($v) eq 'HASH') {
683 map $self->_expand_expr_hashpair($k => { $_ => $v->{$_} }),
690 $self->_assert_pass_injection_guard($vk);
691 if ($vk =~ s/ [_\s]? \d+ $//x ) {
692 belch 'Use of [and|or|nest]_N modifiers is deprecated and will be removed in SQLA v2.0. '
693 . "You probably wanted ...-and => [ -$vk => COND1, -$vk => COND2 ... ]";
695 if ($vk =~ /^(?:not[ _])?between$/) {
696 local our $Cur_Col_Meta = $k;
697 my @rhs = map $self->_expand_expr($_),
698 ref($vv) eq 'ARRAY' ? @$vv : $vv;
700 (@rhs == 1 and ref($rhs[0]) eq 'HASH' and $rhs[0]->{-literal})
702 (@rhs == 2 and defined($rhs[0]) and defined($rhs[1]))
704 puke "Operator '${\uc($vk)}' requires either an arrayref with two defined values or expressions, or a single literal scalarref/arrayref-ref";
707 join(' ', split '_', $vk),
712 if ($vk =~ /^(?:not[ _])?in$/) {
713 if (my $literal = is_literal_value($vv)) {
714 my ($sql, @bind) = @$literal;
715 my $opened_sql = $self->_open_outer_paren($sql);
717 $vk, { -ident => $k },
718 [ { -literal => [ $opened_sql, @bind ] } ]
722 'SQL::Abstract before v1.75 used to generate incorrect SQL when the '
723 . "-${\uc($vk)} operator was given an undef-containing list: !!!AUDIT YOUR CODE "
724 . 'AND DATA!!! (the upcoming Data::Query-based version of SQL::Abstract '
725 . 'will emit the logically correct SQL instead of raising this exception)'
727 puke("Argument passed to the '${\uc($vk)}' operator can not be undefined")
729 my @rhs = map $self->_expand_expr($_),
730 map { ref($_) ? $_ : { -bind => [ $k, $_ ] } }
731 map { defined($_) ? $_: puke($undef_err) }
732 (ref($vv) eq 'ARRAY' ? @$vv : $vv);
733 return $self->${\($vk =~ /^not/ ? 'sqltrue' : 'sqlfalse')} unless @rhs;
736 join(' ', split '_', $vk),
741 if ($vk eq 'ident') {
742 if (! defined $vv or (ref($vv) and ref($vv) eq 'ARRAY')) {
743 puke "-$vk requires a single plain scalar argument (a quotable identifier) or an arrayref of identifier parts";
751 if ($vk eq 'value') {
752 return $self->_expand_expr_hashpair($k, undef) unless defined($vv);
756 { -bind => [ $k, $vv ] }
759 if ($vk =~ /^is(?:[ _]not)?$/) {
760 puke "$vk can only take undef as argument"
764 and exists($vv->{-value})
765 and !defined($vv->{-value})
768 return +{ -op => [ $vk.' null', { -ident => $k } ] };
770 if ($vk =~ /^(and|or)$/) {
771 if (ref($vv) eq 'HASH') {
774 map $self->_expand_expr_hashpair($k, { $_ => $vv->{$_} }),
779 if (my $us = List::Util::first { $vk =~ $_->{regex} } @{$self->{special_ops}}) {
780 return { -op => [ $vk, { -ident => $k }, $vv ] };
782 if (my $us = List::Util::first { $vk =~ $_->{regex} } @{$self->{unary_ops}}) {
786 { -op => [ $vk, $vv ] }
789 if (ref($vv) eq 'ARRAY') {
790 my ($logic, @values) = (
791 (defined($vv->[0]) and $vv->[0] =~ /^-(and|or)$/i)
796 $vk =~ $self->{inequality_op}
797 or join(' ', split '_', $vk) =~ $self->{not_like_op}
799 if (lc($logic) eq '-or' and @values > 1) {
800 my $op = uc join ' ', split '_', $vk;
801 belch "A multi-element arrayref as an argument to the inequality op '$op' "
802 . 'is technically equivalent to an always-true 1=1 (you probably wanted '
803 . "to say ...{ \$inequality_op => [ -and => \@values ] }... instead)"
808 # try to DWIM on equality operators
809 my $op = join ' ', split '_', $vk;
811 $op =~ $self->{equality_op} ? $self->sqlfalse
812 : $op =~ $self->{like_op} ? belch("Supplying an empty arrayref to '@{[ uc $op]}' is deprecated") && $self->sqlfalse
813 : $op =~ $self->{inequality_op} ? $self->sqltrue
814 : $op =~ $self->{not_like_op} ? belch("Supplying an empty arrayref to '@{[ uc $op]}' is deprecated") && $self->sqltrue
815 : puke "operator '$op' applied on an empty array (field '$k')";
819 map $self->_expand_expr_hashpair($k => { $vk => $_ }),
827 and exists $vv->{-value}
828 and not defined $vv->{-value}
831 my $op = join ' ', split '_', $vk;
833 $op =~ /^not$/i ? 'is not' # legacy
834 : $op =~ $self->{equality_op} ? 'is'
835 : $op =~ $self->{like_op} ? belch("Supplying an undefined argument to '@{[ uc $op]}' is deprecated") && 'is'
836 : $op =~ $self->{inequality_op} ? 'is not'
837 : $op =~ $self->{not_like_op} ? belch("Supplying an undefined argument to '@{[ uc $op]}' is deprecated") && 'is not'
838 : puke "unexpected operator '$op' with undef operand";
839 return +{ -op => [ $is.' null', { -ident => $k } ] };
841 local our $Cur_Col_Meta = $k;
845 $self->_expand_expr($vv)
848 if (ref($v) eq 'ARRAY') {
849 return $self->sqlfalse unless @$v;
850 $self->_debug("ARRAY($k) means distribute over elements");
852 $v->[0] =~ /^-((?:and|or))$/i
853 ? ($v = [ @{$v}[1..$#$v] ], $1)
854 : ($self->{logic} || 'or')
858 map $self->_expand_expr({ $k => $_ }, $this_logic), @$v
861 if (my $literal = is_literal_value($v)) {
863 belch 'Hash-pairs consisting of an empty string with a literal are deprecated, and will be removed in 2.0: use -and => [ $literal ] instead';
866 my ($sql, @bind) = @$literal;
867 if ($self->{bindtype} eq 'columns') {
869 $self->_assert_bindval_matches_bindtype($_);
872 return +{ -literal => [ $self->_quote($k).' '.$sql, @bind ] };
878 my ($self, $expr) = @_;
879 my ($k, $v, @rest) = %$expr;
881 if (my $meth = $self->{node_types}{$k}) {
882 return $self->$meth($v);
884 die "notreached: $k";
888 my ($self, $where, $logic) = @_;
890 #print STDERR Data::Dumper::Concise::Dumper([ $where, $logic ]);
892 # Special case: top level simple string treated as literal
894 my $where_exp = (ref($where)
895 ? $self->_expand_expr($where, $logic)
896 : { -literal => [ $where ] });
897 #::Dwarn([ EXPANDED => $where_exp ]);
899 #print STDERR Data::Dumper::Concise::Dumper([ EXP => $where_exp ]);
901 # dispatch on appropriate method according to refkind of $where
902 # my $method = $self->_METHOD_FOR_refkind("_where", $where_exp);
904 # my ($sql, @bind) = $self->$method($where_exp, $logic);
906 my ($sql, @bind) = defined($where_exp) ? $self->_render_expr($where_exp) : (undef);
907 # DBIx::Class used to call _recurse_where in scalar context
908 # something else might too...
910 return ($sql, @bind);
913 belch "Calling _recurse_where in scalar context is deprecated and will go away before 2.0";
919 my ($self, $ident) = @_;
921 return $self->_convert($self->_quote($ident));
925 my ($self, $value) = @_;
927 return ($self->_convert('?'), $self->_bindtype(undef, $value));
930 my %unop_postfix = map +($_ => 1),
931 'is null', 'is not null',
939 my ($self, $args) = @_;
940 my ($left, $low, $high) = @$args;
941 my ($rhsql, @rhbind) = do {
943 puke "Single arg to between must be a literal"
944 unless $low->{-literal};
947 my ($l, $h) = map [ $self->_render_expr($_) ], $low, $high;
948 (join(' ', $l->[0], $self->_sqlcase('and'), $h->[0]),
949 @{$l}[1..$#$l], @{$h}[1..$#$h])
952 my ($lhsql, @lhbind) = $self->_render_expr($left);
954 join(' ', '(', $lhsql, $self->_sqlcase($op), $rhsql, ')'),
958 }), 'between', 'not between'),
962 my ($self, $args) = @_;
963 my ($lhs, $rhs) = @$args;
966 my ($sql, @bind) = $self->_render_expr($_);
967 push @in_bind, @bind;
970 my ($lhsql, @lbind) = $self->_render_expr($lhs);
972 $lhsql.' '.$self->_sqlcase($op).' ( '
983 my ($op, @args) = @$v;
984 $op =~ s/^-// if length($op) > 1;
986 if (my $h = $special{$op}) {
987 return $self->$h(\@args);
989 my $us = List::Util::first { $op =~ $_->{regex} } @{$self->{special_ops}};
990 if ($us and @args > 1) {
991 puke "Special op '${op}' requires first value to be identifier"
992 unless my ($k) = map $_->{-ident}, grep ref($_) eq 'HASH', $args[0];
993 return $self->${\($us->{handler})}($k, $op, $args[1]);
995 if (my $us = List::Util::first { $op =~ $_->{regex} } @{$self->{unary_ops}}) {
996 return $self->${\($us->{handler})}($op, $args[0]);
998 my $final_op = $op =~ /^(?:is|not)_/ ? join(' ', split '_', $op) : $op;
999 if (@args == 1 and $op !~ /^(and|or)$/) {
1000 my ($expr_sql, @bind) = $self->_render_expr($args[0]);
1001 my $op_sql = $self->_sqlcase($final_op);
1003 $unop_postfix{lc($final_op)}
1004 ? "${expr_sql} ${op_sql}"
1005 : "${op_sql} ${expr_sql}"
1007 return (($op eq 'not' || $us ? '('.$final_sql.')' : $final_sql), @bind);
1008 #} elsif (@args == 0) {
1011 my @parts = grep length($_->[0]), map [ $self->_render_expr($_) ], @args;
1012 return '' unless @parts;
1013 my $is_andor = !!($op =~ /^(and|or)$/);
1014 return @{$parts[0]} if $is_andor and @parts == 1;
1015 my ($final_sql) = map +($is_andor ? "( ${_} )" : $_), join(
1016 ($final_op eq ',' ? '' : ' ').$self->_sqlcase($final_op).' ',
1021 map @{$_}[1..$#$_], @parts
1028 my ($self, $rest) = @_;
1029 my ($func, @args) = @$rest;
1033 push @arg_sql, shift @x;
1035 } map [ $self->_render_expr($_) ], @args;
1036 return ($self->_sqlcase($func).'('.join(', ', @arg_sql).')', @bind);
1040 my ($self, $bind) = @_;
1041 return ($self->_convert('?'), $self->_bindtype(@$bind));
1044 sub _render_literal {
1045 my ($self, $literal) = @_;
1046 $self->_assert_bindval_matches_bindtype(@{$literal}[1..$#$literal]);
1050 # Some databases (SQLite) treat col IN (1, 2) different from
1051 # col IN ( (1, 2) ). Use this to strip all outer parens while
1052 # adding them back in the corresponding method
1053 sub _open_outer_paren {
1054 my ($self, $sql) = @_;
1056 while (my ($inner) = $sql =~ /^ \s* \( (.*) \) \s* $/xs) {
1058 # there are closing parens inside, need the heavy duty machinery
1059 # to reevaluate the extraction starting from $sql (full reevaluation)
1060 if ($inner =~ /\)/) {
1061 require Text::Balanced;
1063 my (undef, $remainder) = do {
1064 # idiotic design - writes to $@ but *DOES NOT* throw exceptions
1066 Text::Balanced::extract_bracketed($sql, '()', qr/\s*/);
1069 # the entire expression needs to be a balanced bracketed thing
1070 # (after an extract no remainder sans trailing space)
1071 last if defined $remainder and $remainder =~ /\S/;
1081 #======================================================================
1083 #======================================================================
1085 sub _expand_order_by {
1086 my ($self, $arg) = @_;
1088 return unless defined($arg) and not (ref($arg) eq 'ARRAY' and !@$arg);
1090 my $expander = sub {
1091 my ($self, $dir, $expr) = @_;
1092 my @to_expand = ref($expr) eq 'ARRAY' ? @$expr : $expr;
1093 foreach my $arg (@to_expand) {
1097 and grep /^-(asc|desc)$/, keys %$arg
1099 puke "ordering direction hash passed to order by must have exactly one key (-asc or -desc)";
1102 my @exp = map +(defined($dir) ? { -op => [ $dir => $_ ] } : $_),
1103 map $self->_expand_expr($_, undef, -ident),
1104 map ref($_) eq 'ARRAY' ? @$_ : $_, @to_expand;
1105 return (@exp > 1 ? { -op => [ ',', @exp ] } : $exp[0]);
1108 local @{$self->{expand_unary}}{qw(-asc -desc)} = (
1109 sub { shift->$expander(asc => @_) },
1110 sub { shift->$expander(desc => @_) },
1113 return $self->$expander(undef, $arg);
1117 my ($self, $arg) = @_;
1119 return '' unless defined(my $expanded = $self->_expand_order_by($arg));
1121 my ($sql, @bind) = $self->_render_expr($expanded);
1123 my $final_sql = $self->_sqlcase(' order by ').$sql;
1125 return wantarray ? ($final_sql, @bind) : $final_sql;
1128 # _order_by no longer needs to call this so doesn't but DBIC uses it.
1130 sub _order_by_chunks {
1131 my ($self, $arg) = @_;
1133 return () unless defined(my $expanded = $self->_expand_order_by($arg));
1135 return $self->_chunkify_order_by($expanded);
1138 sub _chunkify_order_by {
1139 my ($self, $expanded) = @_;
1141 return $self->_render_expr($expanded)
1142 if $expanded->{-ident} or @{$expanded->{-literal}||[]} == 1;
1145 if (ref() eq 'HASH' and my $op = $_->{-op}) {
1146 if ($op->[0] eq ',') {
1147 return map $self->_chunkify_order_by($_), @{$op}[1..$#$op];
1150 return [ $self->_render_expr($_) ];
1154 #======================================================================
1155 # DATASOURCE (FOR NOW, JUST PLAIN TABLE OR LIST OF TABLES)
1156 #======================================================================
1161 ($self->_render_expr(
1162 $self->_expand_maybe_list_expr($from, undef, -ident)
1167 #======================================================================
1169 #======================================================================
1171 sub _expand_maybe_list_expr {
1172 my ($self, $expr, $logic, $default) = @_;
1174 if (ref($expr) eq 'ARRAY') {
1176 ',', map $self->_expand_expr($_, $logic, $default), @$expr
1183 return $self->_expand_expr($e, $logic, $default);
1186 # highly optimized, as it's called way too often
1188 # my ($self, $label) = @_;
1190 return '' unless defined $_[1];
1191 return ${$_[1]} if ref($_[1]) eq 'SCALAR';
1192 puke 'Identifier cannot be hashref' if ref($_[1]) eq 'HASH';
1194 unless ($_[0]->{quote_char}) {
1195 if (ref($_[1]) eq 'ARRAY') {
1196 return join($_[0]->{name_sep}||'.', @{$_[1]});
1198 $_[0]->_assert_pass_injection_guard($_[1]);
1203 my $qref = ref $_[0]->{quote_char};
1205 !$qref ? ($_[0]->{quote_char}, $_[0]->{quote_char})
1206 : ($qref eq 'ARRAY') ? @{$_[0]->{quote_char}}
1207 : puke "Unsupported quote_char format: $_[0]->{quote_char}";
1209 my $esc = $_[0]->{escape_char} || $r;
1211 # parts containing * are naturally unquoted
1213 $_[0]->{name_sep}||'',
1217 : do { (my $n = $_) =~ s/(\Q$esc\E|\Q$r\E)/$esc$1/g; $l . $n . $r }
1219 (ref($_[1]) eq 'ARRAY'
1223 ? split (/\Q$_[0]->{name_sep}\E/, $_[1] )
1231 # Conversion, if applicable
1233 #my ($self, $arg) = @_;
1234 if ($_[0]->{convert_where}) {
1235 return $_[0]->_sqlcase($_[0]->{convert_where}) .'(' . $_[1] . ')';
1242 #my ($self, $col, @vals) = @_;
1243 # called often - tighten code
1244 return $_[0]->{bindtype} eq 'columns'
1245 ? map {[$_[1], $_]} @_[2 .. $#_]
1250 # Dies if any element of @bind is not in [colname => value] format
1251 # if bindtype is 'columns'.
1252 sub _assert_bindval_matches_bindtype {
1253 # my ($self, @bind) = @_;
1255 if ($self->{bindtype} eq 'columns') {
1257 if (!defined $_ || ref($_) ne 'ARRAY' || @$_ != 2) {
1258 puke "bindtype 'columns' selected, you need to pass: [column_name => bind_value]"
1264 sub _join_sql_clauses {
1265 my ($self, $logic, $clauses_aref, $bind_aref) = @_;
1267 if (@$clauses_aref > 1) {
1268 my $join = " " . $self->_sqlcase($logic) . " ";
1269 my $sql = '( ' . join($join, @$clauses_aref) . ' )';
1270 return ($sql, @$bind_aref);
1272 elsif (@$clauses_aref) {
1273 return ($clauses_aref->[0], @$bind_aref); # no parentheses
1276 return (); # if no SQL, ignore @$bind_aref
1281 # Fix SQL case, if so requested
1283 # LDNOTE: if $self->{case} is true, then it contains 'lower', so we
1284 # don't touch the argument ... crooked logic, but let's not change it!
1285 return $_[0]->{case} ? $_[1] : uc($_[1]);
1289 #======================================================================
1290 # DISPATCHING FROM REFKIND
1291 #======================================================================
1294 my ($self, $data) = @_;
1296 return 'UNDEF' unless defined $data;
1298 # blessed objects are treated like scalars
1299 my $ref = (Scalar::Util::blessed $data) ? '' : ref $data;
1301 return 'SCALAR' unless $ref;
1304 while ($ref eq 'REF') {
1306 $ref = (Scalar::Util::blessed $data) ? '' : ref $data;
1310 return ($ref||'SCALAR') . ('REF' x $n_steps);
1314 my ($self, $data) = @_;
1315 my @try = ($self->_refkind($data));
1316 push @try, 'SCALAR_or_UNDEF' if $try[0] eq 'SCALAR' || $try[0] eq 'UNDEF';
1317 push @try, 'FALLBACK';
1321 sub _METHOD_FOR_refkind {
1322 my ($self, $meth_prefix, $data) = @_;
1325 for (@{$self->_try_refkind($data)}) {
1326 $method = $self->can($meth_prefix."_".$_)
1330 return $method || puke "cannot dispatch on '$meth_prefix' for ".$self->_refkind($data);
1334 sub _SWITCH_refkind {
1335 my ($self, $data, $dispatch_table) = @_;
1338 for (@{$self->_try_refkind($data)}) {
1339 $coderef = $dispatch_table->{$_}
1343 puke "no dispatch entry for ".$self->_refkind($data)
1352 #======================================================================
1353 # VALUES, GENERATE, AUTOLOAD
1354 #======================================================================
1356 # LDNOTE: original code from nwiger, didn't touch code in that section
1357 # I feel the AUTOLOAD stuff should not be the default, it should
1358 # only be activated on explicit demand by user.
1362 my $data = shift || return;
1363 puke "Argument to ", __PACKAGE__, "->values must be a \\%hash"
1364 unless ref $data eq 'HASH';
1367 foreach my $k (sort keys %$data) {
1368 my $v = $data->{$k};
1369 $self->_SWITCH_refkind($v, {
1371 if ($self->{array_datatypes}) { # array datatype
1372 push @all_bind, $self->_bindtype($k, $v);
1374 else { # literal SQL with bind
1375 my ($sql, @bind) = @$v;
1376 $self->_assert_bindval_matches_bindtype(@bind);
1377 push @all_bind, @bind;
1380 ARRAYREFREF => sub { # literal SQL with bind
1381 my ($sql, @bind) = @${$v};
1382 $self->_assert_bindval_matches_bindtype(@bind);
1383 push @all_bind, @bind;
1385 SCALARREF => sub { # literal SQL without bind
1387 SCALAR_or_UNDEF => sub {
1388 push @all_bind, $self->_bindtype($k, $v);
1399 my(@sql, @sqlq, @sqlv);
1403 if ($ref eq 'HASH') {
1404 for my $k (sort keys %$_) {
1407 my $label = $self->_quote($k);
1408 if ($r eq 'ARRAY') {
1409 # literal SQL with bind
1410 my ($sql, @bind) = @$v;
1411 $self->_assert_bindval_matches_bindtype(@bind);
1412 push @sqlq, "$label = $sql";
1414 } elsif ($r eq 'SCALAR') {
1415 # literal SQL without bind
1416 push @sqlq, "$label = $$v";
1418 push @sqlq, "$label = ?";
1419 push @sqlv, $self->_bindtype($k, $v);
1422 push @sql, $self->_sqlcase('set'), join ', ', @sqlq;
1423 } elsif ($ref eq 'ARRAY') {
1424 # unlike insert(), assume these are ONLY the column names, i.e. for SQL
1427 if ($r eq 'ARRAY') { # literal SQL with bind
1428 my ($sql, @bind) = @$v;
1429 $self->_assert_bindval_matches_bindtype(@bind);
1432 } elsif ($r eq 'SCALAR') { # literal SQL without bind
1433 # embedded literal SQL
1440 push @sql, '(' . join(', ', @sqlq) . ')';
1441 } elsif ($ref eq 'SCALAR') {
1445 # strings get case twiddled
1446 push @sql, $self->_sqlcase($_);
1450 my $sql = join ' ', @sql;
1452 # this is pretty tricky
1453 # if ask for an array, return ($stmt, @bind)
1454 # otherwise, s/?/shift @sqlv/ to put it inline
1456 return ($sql, @sqlv);
1458 1 while $sql =~ s/\?/my $d = shift(@sqlv);
1459 ref $d ? $d->[1] : $d/e;
1468 # This allows us to check for a local, then _form, attr
1470 my($name) = $AUTOLOAD =~ /.*::(.+)/;
1471 return $self->generate($name, @_);
1482 SQL::Abstract - Generate SQL from Perl data structures
1488 my $sql = SQL::Abstract->new;
1490 my($stmt, @bind) = $sql->select($source, \@fields, \%where, $order);
1492 my($stmt, @bind) = $sql->insert($table, \%fieldvals || \@values);
1494 my($stmt, @bind) = $sql->update($table, \%fieldvals, \%where);
1496 my($stmt, @bind) = $sql->delete($table, \%where);
1498 # Then, use these in your DBI statements
1499 my $sth = $dbh->prepare($stmt);
1500 $sth->execute(@bind);
1502 # Just generate the WHERE clause
1503 my($stmt, @bind) = $sql->where(\%where, $order);
1505 # Return values in the same order, for hashed queries
1506 # See PERFORMANCE section for more details
1507 my @bind = $sql->values(\%fieldvals);
1511 This module was inspired by the excellent L<DBIx::Abstract>.
1512 However, in using that module I found that what I really wanted
1513 to do was generate SQL, but still retain complete control over my
1514 statement handles and use the DBI interface. So, I set out to
1515 create an abstract SQL generation module.
1517 While based on the concepts used by L<DBIx::Abstract>, there are
1518 several important differences, especially when it comes to WHERE
1519 clauses. I have modified the concepts used to make the SQL easier
1520 to generate from Perl data structures and, IMO, more intuitive.
1521 The underlying idea is for this module to do what you mean, based
1522 on the data structures you provide it. The big advantage is that
1523 you don't have to modify your code every time your data changes,
1524 as this module figures it out.
1526 To begin with, an SQL INSERT is as easy as just specifying a hash
1527 of C<key=value> pairs:
1530 name => 'Jimbo Bobson',
1531 phone => '123-456-7890',
1532 address => '42 Sister Lane',
1533 city => 'St. Louis',
1534 state => 'Louisiana',
1537 The SQL can then be generated with this:
1539 my($stmt, @bind) = $sql->insert('people', \%data);
1541 Which would give you something like this:
1543 $stmt = "INSERT INTO people
1544 (address, city, name, phone, state)
1545 VALUES (?, ?, ?, ?, ?)";
1546 @bind = ('42 Sister Lane', 'St. Louis', 'Jimbo Bobson',
1547 '123-456-7890', 'Louisiana');
1549 These are then used directly in your DBI code:
1551 my $sth = $dbh->prepare($stmt);
1552 $sth->execute(@bind);
1554 =head2 Inserting and Updating Arrays
1556 If your database has array types (like for example Postgres),
1557 activate the special option C<< array_datatypes => 1 >>
1558 when creating the C<SQL::Abstract> object.
1559 Then you may use an arrayref to insert and update database array types:
1561 my $sql = SQL::Abstract->new(array_datatypes => 1);
1563 planets => [qw/Mercury Venus Earth Mars/]
1566 my($stmt, @bind) = $sql->insert('solar_system', \%data);
1570 $stmt = "INSERT INTO solar_system (planets) VALUES (?)"
1572 @bind = (['Mercury', 'Venus', 'Earth', 'Mars']);
1575 =head2 Inserting and Updating SQL
1577 In order to apply SQL functions to elements of your C<%data> you may
1578 specify a reference to an arrayref for the given hash value. For example,
1579 if you need to execute the Oracle C<to_date> function on a value, you can
1580 say something like this:
1584 date_entered => \[ "to_date(?,'MM/DD/YYYY')", "03/02/2003" ],
1587 The first value in the array is the actual SQL. Any other values are
1588 optional and would be included in the bind values array. This gives
1591 my($stmt, @bind) = $sql->insert('people', \%data);
1593 $stmt = "INSERT INTO people (name, date_entered)
1594 VALUES (?, to_date(?,'MM/DD/YYYY'))";
1595 @bind = ('Bill', '03/02/2003');
1597 An UPDATE is just as easy, all you change is the name of the function:
1599 my($stmt, @bind) = $sql->update('people', \%data);
1601 Notice that your C<%data> isn't touched; the module will generate
1602 the appropriately quirky SQL for you automatically. Usually you'll
1603 want to specify a WHERE clause for your UPDATE, though, which is
1604 where handling C<%where> hashes comes in handy...
1606 =head2 Complex where statements
1608 This module can generate pretty complicated WHERE statements
1609 easily. For example, simple C<key=value> pairs are taken to mean
1610 equality, and if you want to see if a field is within a set
1611 of values, you can use an arrayref. Let's say we wanted to
1612 SELECT some data based on this criteria:
1615 requestor => 'inna',
1616 worker => ['nwiger', 'rcwe', 'sfz'],
1617 status => { '!=', 'completed' }
1620 my($stmt, @bind) = $sql->select('tickets', '*', \%where);
1622 The above would give you something like this:
1624 $stmt = "SELECT * FROM tickets WHERE
1625 ( requestor = ? ) AND ( status != ? )
1626 AND ( worker = ? OR worker = ? OR worker = ? )";
1627 @bind = ('inna', 'completed', 'nwiger', 'rcwe', 'sfz');
1629 Which you could then use in DBI code like so:
1631 my $sth = $dbh->prepare($stmt);
1632 $sth->execute(@bind);
1638 The methods are simple. There's one for every major SQL operation,
1639 and a constructor you use first. The arguments are specified in a
1640 similar order for each method (table, then fields, then a where
1641 clause) to try and simplify things.
1643 =head2 new(option => 'value')
1645 The C<new()> function takes a list of options and values, and returns
1646 a new B<SQL::Abstract> object which can then be used to generate SQL
1647 through the methods below. The options accepted are:
1653 If set to 'lower', then SQL will be generated in all lowercase. By
1654 default SQL is generated in "textbook" case meaning something like:
1656 SELECT a_field FROM a_table WHERE some_field LIKE '%someval%'
1658 Any setting other than 'lower' is ignored.
1662 This determines what the default comparison operator is. By default
1663 it is C<=>, meaning that a hash like this:
1665 %where = (name => 'nwiger', email => 'nate@wiger.org');
1667 Will generate SQL like this:
1669 WHERE name = 'nwiger' AND email = 'nate@wiger.org'
1671 However, you may want loose comparisons by default, so if you set
1672 C<cmp> to C<like> you would get SQL such as:
1674 WHERE name like 'nwiger' AND email like 'nate@wiger.org'
1676 You can also override the comparison on an individual basis - see
1677 the huge section on L</"WHERE CLAUSES"> at the bottom.
1679 =item sqltrue, sqlfalse
1681 Expressions for inserting boolean values within SQL statements.
1682 By default these are C<1=1> and C<1=0>. They are used
1683 by the special operators C<-in> and C<-not_in> for generating
1684 correct SQL even when the argument is an empty array (see below).
1688 This determines the default logical operator for multiple WHERE
1689 statements in arrays or hashes. If absent, the default logic is "or"
1690 for arrays, and "and" for hashes. This means that a WHERE
1694 event_date => {'>=', '2/13/99'},
1695 event_date => {'<=', '4/24/03'},
1698 will generate SQL like this:
1700 WHERE event_date >= '2/13/99' OR event_date <= '4/24/03'
1702 This is probably not what you want given this query, though (look
1703 at the dates). To change the "OR" to an "AND", simply specify:
1705 my $sql = SQL::Abstract->new(logic => 'and');
1707 Which will change the above C<WHERE> to:
1709 WHERE event_date >= '2/13/99' AND event_date <= '4/24/03'
1711 The logic can also be changed locally by inserting
1712 a modifier in front of an arrayref:
1714 @where = (-and => [event_date => {'>=', '2/13/99'},
1715 event_date => {'<=', '4/24/03'} ]);
1717 See the L</"WHERE CLAUSES"> section for explanations.
1721 This will automatically convert comparisons using the specified SQL
1722 function for both column and value. This is mostly used with an argument
1723 of C<upper> or C<lower>, so that the SQL will have the effect of
1724 case-insensitive "searches". For example, this:
1726 $sql = SQL::Abstract->new(convert => 'upper');
1727 %where = (keywords => 'MaKe iT CAse inSeNSItive');
1729 Will turn out the following SQL:
1731 WHERE upper(keywords) like upper('MaKe iT CAse inSeNSItive')
1733 The conversion can be C<upper()>, C<lower()>, or any other SQL function
1734 that can be applied symmetrically to fields (actually B<SQL::Abstract> does
1735 not validate this option; it will just pass through what you specify verbatim).
1739 This is a kludge because many databases suck. For example, you can't
1740 just bind values using DBI's C<execute()> for Oracle C<CLOB> or C<BLOB> fields.
1741 Instead, you have to use C<bind_param()>:
1743 $sth->bind_param(1, 'reg data');
1744 $sth->bind_param(2, $lots, {ora_type => ORA_CLOB});
1746 The problem is, B<SQL::Abstract> will normally just return a C<@bind> array,
1747 which loses track of which field each slot refers to. Fear not.
1749 If you specify C<bindtype> in new, you can determine how C<@bind> is returned.
1750 Currently, you can specify either C<normal> (default) or C<columns>. If you
1751 specify C<columns>, you will get an array that looks like this:
1753 my $sql = SQL::Abstract->new(bindtype => 'columns');
1754 my($stmt, @bind) = $sql->insert(...);
1757 [ 'column1', 'value1' ],
1758 [ 'column2', 'value2' ],
1759 [ 'column3', 'value3' ],
1762 You can then iterate through this manually, using DBI's C<bind_param()>.
1764 $sth->prepare($stmt);
1767 my($col, $data) = @$_;
1768 if ($col eq 'details' || $col eq 'comments') {
1769 $sth->bind_param($i, $data, {ora_type => ORA_CLOB});
1770 } elsif ($col eq 'image') {
1771 $sth->bind_param($i, $data, {ora_type => ORA_BLOB});
1773 $sth->bind_param($i, $data);
1777 $sth->execute; # execute without @bind now
1779 Now, why would you still use B<SQL::Abstract> if you have to do this crap?
1780 Basically, the advantage is still that you don't have to care which fields
1781 are or are not included. You could wrap that above C<for> loop in a simple
1782 sub called C<bind_fields()> or something and reuse it repeatedly. You still
1783 get a layer of abstraction over manual SQL specification.
1785 Note that if you set L</bindtype> to C<columns>, the C<\[ $sql, @bind ]>
1786 construct (see L</Literal SQL with placeholders and bind values (subqueries)>)
1787 will expect the bind values in this format.
1791 This is the character that a table or column name will be quoted
1792 with. By default this is an empty string, but you could set it to
1793 the character C<`>, to generate SQL like this:
1795 SELECT `a_field` FROM `a_table` WHERE `some_field` LIKE '%someval%'
1797 Alternatively, you can supply an array ref of two items, the first being the left
1798 hand quote character, and the second the right hand quote character. For
1799 example, you could supply C<['[',']']> for SQL Server 2000 compliant quotes
1800 that generates SQL like this:
1802 SELECT [a_field] FROM [a_table] WHERE [some_field] LIKE '%someval%'
1804 Quoting is useful if you have tables or columns names that are reserved
1805 words in your database's SQL dialect.
1809 This is the character that will be used to escape L</quote_char>s appearing
1810 in an identifier before it has been quoted.
1812 The parameter default in case of a single L</quote_char> character is the quote
1815 When opening-closing-style quoting is used (L</quote_char> is an arrayref)
1816 this parameter defaults to the B<closing (right)> L</quote_char>. Occurrences
1817 of the B<opening (left)> L</quote_char> within the identifier are currently left
1818 untouched. The default for opening-closing-style quotes may change in future
1819 versions, thus you are B<strongly encouraged> to specify the escape character
1824 This is the character that separates a table and column name. It is
1825 necessary to specify this when the C<quote_char> option is selected,
1826 so that tables and column names can be individually quoted like this:
1828 SELECT `table`.`one_field` FROM `table` WHERE `table`.`other_field` = 1
1830 =item injection_guard
1832 A regular expression C<qr/.../> that is applied to any C<-function> and unquoted
1833 column name specified in a query structure. This is a safety mechanism to avoid
1834 injection attacks when mishandling user input e.g.:
1836 my %condition_as_column_value_pairs = get_values_from_user();
1837 $sqla->select( ... , \%condition_as_column_value_pairs );
1839 If the expression matches an exception is thrown. Note that literal SQL
1840 supplied via C<\'...'> or C<\['...']> is B<not> checked in any way.
1842 Defaults to checking for C<;> and the C<GO> keyword (TransactSQL)
1844 =item array_datatypes
1846 When this option is true, arrayrefs in INSERT or UPDATE are
1847 interpreted as array datatypes and are passed directly
1849 When this option is false, arrayrefs are interpreted
1850 as literal SQL, just like refs to arrayrefs
1851 (but this behavior is for backwards compatibility; when writing
1852 new queries, use the "reference to arrayref" syntax
1858 Takes a reference to a list of "special operators"
1859 to extend the syntax understood by L<SQL::Abstract>.
1860 See section L</"SPECIAL OPERATORS"> for details.
1864 Takes a reference to a list of "unary operators"
1865 to extend the syntax understood by L<SQL::Abstract>.
1866 See section L</"UNARY OPERATORS"> for details.
1872 =head2 insert($table, \@values || \%fieldvals, \%options)
1874 This is the simplest function. You simply give it a table name
1875 and either an arrayref of values or hashref of field/value pairs.
1876 It returns an SQL INSERT statement and a list of bind values.
1877 See the sections on L</"Inserting and Updating Arrays"> and
1878 L</"Inserting and Updating SQL"> for information on how to insert
1879 with those data types.
1881 The optional C<\%options> hash reference may contain additional
1882 options to generate the insert SQL. Currently supported options
1889 Takes either a scalar of raw SQL fields, or an array reference of
1890 field names, and adds on an SQL C<RETURNING> statement at the end.
1891 This allows you to return data generated by the insert statement
1892 (such as row IDs) without performing another C<SELECT> statement.
1893 Note, however, this is not part of the SQL standard and may not
1894 be supported by all database engines.
1898 =head2 update($table, \%fieldvals, \%where, \%options)
1900 This takes a table, hashref of field/value pairs, and an optional
1901 hashref L<WHERE clause|/WHERE CLAUSES>. It returns an SQL UPDATE function and a list
1903 See the sections on L</"Inserting and Updating Arrays"> and
1904 L</"Inserting and Updating SQL"> for information on how to insert
1905 with those data types.
1907 The optional C<\%options> hash reference may contain additional
1908 options to generate the update SQL. Currently supported options
1915 See the C<returning> option to
1916 L<insert|/insert($table, \@values || \%fieldvals, \%options)>.
1920 =head2 select($source, $fields, $where, $order)
1922 This returns a SQL SELECT statement and associated list of bind values, as
1923 specified by the arguments:
1929 Specification of the 'FROM' part of the statement.
1930 The argument can be either a plain scalar (interpreted as a table
1931 name, will be quoted), or an arrayref (interpreted as a list
1932 of table names, joined by commas, quoted), or a scalarref
1933 (literal SQL, not quoted).
1937 Specification of the list of fields to retrieve from
1939 The argument can be either an arrayref (interpreted as a list
1940 of field names, will be joined by commas and quoted), or a
1941 plain scalar (literal SQL, not quoted).
1942 Please observe that this API is not as flexible as that of
1943 the first argument C<$source>, for backwards compatibility reasons.
1947 Optional argument to specify the WHERE part of the query.
1948 The argument is most often a hashref, but can also be
1949 an arrayref or plain scalar --
1950 see section L<WHERE clause|/"WHERE CLAUSES"> for details.
1954 Optional argument to specify the ORDER BY part of the query.
1955 The argument can be a scalar, a hashref or an arrayref
1956 -- see section L<ORDER BY clause|/"ORDER BY CLAUSES">
1962 =head2 delete($table, \%where, \%options)
1964 This takes a table name and optional hashref L<WHERE clause|/WHERE CLAUSES>.
1965 It returns an SQL DELETE statement and list of bind values.
1967 The optional C<\%options> hash reference may contain additional
1968 options to generate the delete SQL. Currently supported options
1975 See the C<returning> option to
1976 L<insert|/insert($table, \@values || \%fieldvals, \%options)>.
1980 =head2 where(\%where, $order)
1982 This is used to generate just the WHERE clause. For example,
1983 if you have an arbitrary data structure and know what the
1984 rest of your SQL is going to look like, but want an easy way
1985 to produce a WHERE clause, use this. It returns an SQL WHERE
1986 clause and list of bind values.
1989 =head2 values(\%data)
1991 This just returns the values from the hash C<%data>, in the same
1992 order that would be returned from any of the other above queries.
1993 Using this allows you to markedly speed up your queries if you
1994 are affecting lots of rows. See below under the L</"PERFORMANCE"> section.
1996 =head2 generate($any, 'number', $of, \@data, $struct, \%types)
1998 Warning: This is an experimental method and subject to change.
2000 This returns arbitrarily generated SQL. It's a really basic shortcut.
2001 It will return two different things, depending on return context:
2003 my($stmt, @bind) = $sql->generate('create table', \$table, \@fields);
2004 my $stmt_and_val = $sql->generate('create table', \$table, \@fields);
2006 These would return the following:
2008 # First calling form
2009 $stmt = "CREATE TABLE test (?, ?)";
2010 @bind = (field1, field2);
2012 # Second calling form
2013 $stmt_and_val = "CREATE TABLE test (field1, field2)";
2015 Depending on what you're trying to do, it's up to you to choose the correct
2016 format. In this example, the second form is what you would want.
2020 $sql->generate('alter session', { nls_date_format => 'MM/YY' });
2024 ALTER SESSION SET nls_date_format = 'MM/YY'
2026 You get the idea. Strings get their case twiddled, but everything
2027 else remains verbatim.
2029 =head1 EXPORTABLE FUNCTIONS
2031 =head2 is_plain_value
2033 Determines if the supplied argument is a plain value as understood by this
2038 =item * The value is C<undef>
2040 =item * The value is a non-reference
2042 =item * The value is an object with stringification overloading
2044 =item * The value is of the form C<< { -value => $anything } >>
2048 On failure returns C<undef>, on success returns a B<scalar> reference
2049 to the original supplied argument.
2055 The stringification overloading detection is rather advanced: it takes
2056 into consideration not only the presence of a C<""> overload, but if that
2057 fails also checks for enabled
2058 L<autogenerated versions of C<"">|overload/Magic Autogeneration>, based
2059 on either C<0+> or C<bool>.
2061 Unfortunately testing in the field indicates that this
2062 detection B<< may tickle a latent bug in perl versions before 5.018 >>,
2063 but only when very large numbers of stringifying objects are involved.
2064 At the time of writing ( Sep 2014 ) there is no clear explanation of
2065 the direct cause, nor is there a manageably small test case that reliably
2066 reproduces the problem.
2068 If you encounter any of the following exceptions in B<random places within
2069 your application stack> - this module may be to blame:
2071 Operation "ne": no method found,
2072 left argument in overloaded package <something>,
2073 right argument in overloaded package <something>
2077 Stub found while resolving method "???" overloading """" in package <something>
2079 If you fall victim to the above - please attempt to reduce the problem
2080 to something that could be sent to the L<SQL::Abstract developers
2081 |DBIx::Class/GETTING HELP/SUPPORT>
2082 (either publicly or privately). As a workaround in the meantime you can
2083 set C<$ENV{SQLA_ISVALUE_IGNORE_AUTOGENERATED_STRINGIFICATION}> to a true
2084 value, which will most likely eliminate your problem (at the expense of
2085 not being able to properly detect exotic forms of stringification).
2087 This notice and environment variable will be removed in a future version,
2088 as soon as the underlying problem is found and a reliable workaround is
2093 =head2 is_literal_value
2095 Determines if the supplied argument is a literal value as understood by this
2100 =item * C<\$sql_string>
2102 =item * C<\[ $sql_string, @bind_values ]>
2106 On failure returns C<undef>, on success returns an B<array> reference
2107 containing the unpacked version of the supplied literal SQL and bind values.
2109 =head1 WHERE CLAUSES
2113 This module uses a variation on the idea from L<DBIx::Abstract>. It
2114 is B<NOT>, repeat I<not> 100% compatible. B<The main logic of this
2115 module is that things in arrays are OR'ed, and things in hashes
2118 The easiest way to explain is to show lots of examples. After
2119 each C<%where> hash shown, it is assumed you used:
2121 my($stmt, @bind) = $sql->where(\%where);
2123 However, note that the C<%where> hash can be used directly in any
2124 of the other functions as well, as described above.
2126 =head2 Key-value pairs
2128 So, let's get started. To begin, a simple hash:
2132 status => 'completed'
2135 Is converted to SQL C<key = val> statements:
2137 $stmt = "WHERE user = ? AND status = ?";
2138 @bind = ('nwiger', 'completed');
2140 One common thing I end up doing is having a list of values that
2141 a field can be in. To do this, simply specify a list inside of
2146 status => ['assigned', 'in-progress', 'pending'];
2149 This simple code will create the following:
2151 $stmt = "WHERE user = ? AND ( status = ? OR status = ? OR status = ? )";
2152 @bind = ('nwiger', 'assigned', 'in-progress', 'pending');
2154 A field associated to an empty arrayref will be considered a
2155 logical false and will generate 0=1.
2157 =head2 Tests for NULL values
2159 If the value part is C<undef> then this is converted to SQL <IS NULL>
2168 $stmt = "WHERE user = ? AND status IS NULL";
2171 To test if a column IS NOT NULL:
2175 status => { '!=', undef },
2178 =head2 Specific comparison operators
2180 If you want to specify a different type of operator for your comparison,
2181 you can use a hashref for a given column:
2185 status => { '!=', 'completed' }
2188 Which would generate:
2190 $stmt = "WHERE user = ? AND status != ?";
2191 @bind = ('nwiger', 'completed');
2193 To test against multiple values, just enclose the values in an arrayref:
2195 status => { '=', ['assigned', 'in-progress', 'pending'] };
2197 Which would give you:
2199 "WHERE status = ? OR status = ? OR status = ?"
2202 The hashref can also contain multiple pairs, in which case it is expanded
2203 into an C<AND> of its elements:
2207 status => { '!=', 'completed', -not_like => 'pending%' }
2210 # Or more dynamically, like from a form
2211 $where{user} = 'nwiger';
2212 $where{status}{'!='} = 'completed';
2213 $where{status}{'-not_like'} = 'pending%';
2215 # Both generate this
2216 $stmt = "WHERE user = ? AND status != ? AND status NOT LIKE ?";
2217 @bind = ('nwiger', 'completed', 'pending%');
2220 To get an OR instead, you can combine it with the arrayref idea:
2224 priority => [ { '=', 2 }, { '>', 5 } ]
2227 Which would generate:
2229 $stmt = "WHERE ( priority = ? OR priority > ? ) AND user = ?";
2230 @bind = ('2', '5', 'nwiger');
2232 If you want to include literal SQL (with or without bind values), just use a
2233 scalar reference or reference to an arrayref as the value:
2236 date_entered => { '>' => \["to_date(?, 'MM/DD/YYYY')", "11/26/2008"] },
2237 date_expires => { '<' => \"now()" }
2240 Which would generate:
2242 $stmt = "WHERE date_entered > to_date(?, 'MM/DD/YYYY') AND date_expires < now()";
2243 @bind = ('11/26/2008');
2246 =head2 Logic and nesting operators
2248 In the example above,
2249 there is a subtle trap if you want to say something like
2250 this (notice the C<AND>):
2252 WHERE priority != ? AND priority != ?
2254 Because, in Perl you I<can't> do this:
2256 priority => { '!=' => 2, '!=' => 1 }
2258 As the second C<!=> key will obliterate the first. The solution
2259 is to use the special C<-modifier> form inside an arrayref:
2261 priority => [ -and => {'!=', 2},
2265 Normally, these would be joined by C<OR>, but the modifier tells it
2266 to use C<AND> instead. (Hint: You can use this in conjunction with the
2267 C<logic> option to C<new()> in order to change the way your queries
2268 work by default.) B<Important:> Note that the C<-modifier> goes
2269 B<INSIDE> the arrayref, as an extra first element. This will
2270 B<NOT> do what you think it might:
2272 priority => -and => [{'!=', 2}, {'!=', 1}] # WRONG!
2274 Here is a quick list of equivalencies, since there is some overlap:
2277 status => {'!=', 'completed', 'not like', 'pending%' }
2278 status => [ -and => {'!=', 'completed'}, {'not like', 'pending%'}]
2281 status => {'=', ['assigned', 'in-progress']}
2282 status => [ -or => {'=', 'assigned'}, {'=', 'in-progress'}]
2283 status => [ {'=', 'assigned'}, {'=', 'in-progress'} ]
2287 =head2 Special operators: IN, BETWEEN, etc.
2289 You can also use the hashref format to compare a list of fields using the
2290 C<IN> comparison operator, by specifying the list as an arrayref:
2293 status => 'completed',
2294 reportid => { -in => [567, 2335, 2] }
2297 Which would generate:
2299 $stmt = "WHERE status = ? AND reportid IN (?,?,?)";
2300 @bind = ('completed', '567', '2335', '2');
2302 The reverse operator C<-not_in> generates SQL C<NOT IN> and is used in
2305 If the argument to C<-in> is an empty array, 'sqlfalse' is generated
2306 (by default: C<1=0>). Similarly, C<< -not_in => [] >> generates
2307 'sqltrue' (by default: C<1=1>).
2309 In addition to the array you can supply a chunk of literal sql or
2310 literal sql with bind:
2313 customer => { -in => \[
2314 'SELECT cust_id FROM cust WHERE balance > ?',
2317 status => { -in => \'SELECT status_codes FROM states' },
2323 customer IN ( SELECT cust_id FROM cust WHERE balance > ? )
2324 AND status IN ( SELECT status_codes FROM states )
2328 Finally, if the argument to C<-in> is not a reference, it will be
2329 treated as a single-element array.
2331 Another pair of operators is C<-between> and C<-not_between>,
2332 used with an arrayref of two values:
2336 completion_date => {
2337 -not_between => ['2002-10-01', '2003-02-06']
2343 WHERE user = ? AND completion_date NOT BETWEEN ( ? AND ? )
2345 Just like with C<-in> all plausible combinations of literal SQL
2349 start0 => { -between => [ 1, 2 ] },
2350 start1 => { -between => \["? AND ?", 1, 2] },
2351 start2 => { -between => \"lower(x) AND upper(y)" },
2352 start3 => { -between => [
2354 \["upper(?)", 'stuff' ],
2361 ( start0 BETWEEN ? AND ? )
2362 AND ( start1 BETWEEN ? AND ? )
2363 AND ( start2 BETWEEN lower(x) AND upper(y) )
2364 AND ( start3 BETWEEN lower(x) AND upper(?) )
2366 @bind = (1, 2, 1, 2, 'stuff');
2369 These are the two builtin "special operators"; but the
2370 list can be expanded: see section L</"SPECIAL OPERATORS"> below.
2372 =head2 Unary operators: bool
2374 If you wish to test against boolean columns or functions within your
2375 database you can use the C<-bool> and C<-not_bool> operators. For
2376 example to test the column C<is_user> being true and the column
2377 C<is_enabled> being false you would use:-
2381 -not_bool => 'is_enabled',
2386 WHERE is_user AND NOT is_enabled
2388 If a more complex combination is required, testing more conditions,
2389 then you should use the and/or operators:-
2394 -not_bool => { two=> { -rlike => 'bar' } },
2395 -not_bool => { three => [ { '=', 2 }, { '>', 5 } ] },
2406 (NOT ( three = ? OR three > ? ))
2409 =head2 Nested conditions, -and/-or prefixes
2411 So far, we've seen how multiple conditions are joined with a top-level
2412 C<AND>. We can change this by putting the different conditions we want in
2413 hashes and then putting those hashes in an array. For example:
2418 status => { -like => ['pending%', 'dispatched'] },
2422 status => 'unassigned',
2426 This data structure would create the following:
2428 $stmt = "WHERE ( user = ? AND ( status LIKE ? OR status LIKE ? ) )
2429 OR ( user = ? AND status = ? ) )";
2430 @bind = ('nwiger', 'pending', 'dispatched', 'robot', 'unassigned');
2433 Clauses in hashrefs or arrayrefs can be prefixed with an C<-and> or C<-or>
2434 to change the logic inside:
2440 -and => [ workhrs => {'>', 20}, geo => 'ASIA' ],
2441 -or => { workhrs => {'<', 50}, geo => 'EURO' },
2448 $stmt = "WHERE ( user = ?
2449 AND ( ( workhrs > ? AND geo = ? )
2450 OR ( workhrs < ? OR geo = ? ) ) )";
2451 @bind = ('nwiger', '20', 'ASIA', '50', 'EURO');
2453 =head3 Algebraic inconsistency, for historical reasons
2455 C<Important note>: when connecting several conditions, the C<-and->|C<-or>
2456 operator goes C<outside> of the nested structure; whereas when connecting
2457 several constraints on one column, the C<-and> operator goes
2458 C<inside> the arrayref. Here is an example combining both features:
2461 -and => [a => 1, b => 2],
2462 -or => [c => 3, d => 4],
2463 e => [-and => {-like => 'foo%'}, {-like => '%bar'} ]
2468 WHERE ( ( ( a = ? AND b = ? )
2469 OR ( c = ? OR d = ? )
2470 OR ( e LIKE ? AND e LIKE ? ) ) )
2472 This difference in syntax is unfortunate but must be preserved for
2473 historical reasons. So be careful: the two examples below would
2474 seem algebraically equivalent, but they are not
2477 { -like => 'foo%' },
2478 { -like => '%bar' },
2480 # yields: WHERE ( ( col LIKE ? AND col LIKE ? ) )
2483 { col => { -like => 'foo%' } },
2484 { col => { -like => '%bar' } },
2486 # yields: WHERE ( ( col LIKE ? OR col LIKE ? ) )
2489 =head2 Literal SQL and value type operators
2491 The basic premise of SQL::Abstract is that in WHERE specifications the "left
2492 side" is a column name and the "right side" is a value (normally rendered as
2493 a placeholder). This holds true for both hashrefs and arrayref pairs as you
2494 see in the L</WHERE CLAUSES> examples above. Sometimes it is necessary to
2495 alter this behavior. There are several ways of doing so.
2499 This is a virtual operator that signals the string to its right side is an
2500 identifier (a column name) and not a value. For example to compare two
2501 columns you would write:
2504 priority => { '<', 2 },
2505 requestor => { -ident => 'submitter' },
2510 $stmt = "WHERE priority < ? AND requestor = submitter";
2513 If you are maintaining legacy code you may see a different construct as
2514 described in L</Deprecated usage of Literal SQL>, please use C<-ident> in new
2519 This is a virtual operator that signals that the construct to its right side
2520 is a value to be passed to DBI. This is for example necessary when you want
2521 to write a where clause against an array (for RDBMS that support such
2522 datatypes). For example:
2525 array => { -value => [1, 2, 3] }
2530 $stmt = 'WHERE array = ?';
2531 @bind = ([1, 2, 3]);
2533 Note that if you were to simply say:
2539 the result would probably not be what you wanted:
2541 $stmt = 'WHERE array = ? OR array = ? OR array = ?';
2546 Finally, sometimes only literal SQL will do. To include a random snippet
2547 of SQL verbatim, you specify it as a scalar reference. Consider this only
2548 as a last resort. Usually there is a better way. For example:
2551 priority => { '<', 2 },
2552 requestor => { -in => \'(SELECT name FROM hitmen)' },
2557 $stmt = "WHERE priority < ? AND requestor IN (SELECT name FROM hitmen)"
2560 Note that in this example, you only get one bind parameter back, since
2561 the verbatim SQL is passed as part of the statement.
2565 Never use untrusted input as a literal SQL argument - this is a massive
2566 security risk (there is no way to check literal snippets for SQL
2567 injections and other nastyness). If you need to deal with untrusted input
2568 use literal SQL with placeholders as described next.
2570 =head3 Literal SQL with placeholders and bind values (subqueries)
2572 If the literal SQL to be inserted has placeholders and bind values,
2573 use a reference to an arrayref (yes this is a double reference --
2574 not so common, but perfectly legal Perl). For example, to find a date
2575 in Postgres you can use something like this:
2578 date_column => \[ "= date '2008-09-30' - ?::integer", 10 ]
2583 $stmt = "WHERE ( date_column = date '2008-09-30' - ?::integer )"
2586 Note that you must pass the bind values in the same format as they are returned
2587 by L<where|/where(\%where, $order)>. This means that if you set L</bindtype>
2588 to C<columns>, you must provide the bind values in the
2589 C<< [ column_meta => value ] >> format, where C<column_meta> is an opaque
2590 scalar value; most commonly the column name, but you can use any scalar value
2591 (including references and blessed references), L<SQL::Abstract> will simply
2592 pass it through intact. So if C<bindtype> is set to C<columns> the above
2593 example will look like:
2596 date_column => \[ "= date '2008-09-30' - ?::integer", [ {} => 10 ] ]
2599 Literal SQL is especially useful for nesting parenthesized clauses in the
2600 main SQL query. Here is a first example:
2602 my ($sub_stmt, @sub_bind) = ("SELECT c1 FROM t1 WHERE c2 < ? AND c3 LIKE ?",
2606 bar => \["IN ($sub_stmt)" => @sub_bind],
2611 $stmt = "WHERE (foo = ? AND bar IN (SELECT c1 FROM t1
2612 WHERE c2 < ? AND c3 LIKE ?))";
2613 @bind = (1234, 100, "foo%");
2615 Other subquery operators, like for example C<"E<gt> ALL"> or C<"NOT IN">,
2616 are expressed in the same way. Of course the C<$sub_stmt> and
2617 its associated bind values can be generated through a former call
2620 my ($sub_stmt, @sub_bind)
2621 = $sql->select("t1", "c1", {c2 => {"<" => 100},
2622 c3 => {-like => "foo%"}});
2625 bar => \["> ALL ($sub_stmt)" => @sub_bind],
2628 In the examples above, the subquery was used as an operator on a column;
2629 but the same principle also applies for a clause within the main C<%where>
2630 hash, like an EXISTS subquery:
2632 my ($sub_stmt, @sub_bind)
2633 = $sql->select("t1", "*", {c1 => 1, c2 => \"> t0.c0"});
2634 my %where = ( -and => [
2636 \["EXISTS ($sub_stmt)" => @sub_bind],
2641 $stmt = "WHERE (foo = ? AND EXISTS (SELECT * FROM t1
2642 WHERE c1 = ? AND c2 > t0.c0))";
2646 Observe that the condition on C<c2> in the subquery refers to
2647 column C<t0.c0> of the main query: this is I<not> a bind
2648 value, so we have to express it through a scalar ref.
2649 Writing C<< c2 => {">" => "t0.c0"} >> would have generated
2650 C<< c2 > ? >> with bind value C<"t0.c0"> ... not exactly
2651 what we wanted here.
2653 Finally, here is an example where a subquery is used
2654 for expressing unary negation:
2656 my ($sub_stmt, @sub_bind)
2657 = $sql->where({age => [{"<" => 10}, {">" => 20}]});
2658 $sub_stmt =~ s/^ where //i; # don't want "WHERE" in the subclause
2660 lname => {like => '%son%'},
2661 \["NOT ($sub_stmt)" => @sub_bind],
2666 $stmt = "lname LIKE ? AND NOT ( age < ? OR age > ? )"
2667 @bind = ('%son%', 10, 20)
2669 =head3 Deprecated usage of Literal SQL
2671 Below are some examples of archaic use of literal SQL. It is shown only as
2672 reference for those who deal with legacy code. Each example has a much
2673 better, cleaner and safer alternative that users should opt for in new code.
2679 my %where = ( requestor => \'IS NOT NULL' )
2681 $stmt = "WHERE requestor IS NOT NULL"
2683 This used to be the way of generating NULL comparisons, before the handling
2684 of C<undef> got formalized. For new code please use the superior syntax as
2685 described in L</Tests for NULL values>.
2689 my %where = ( requestor => \'= submitter' )
2691 $stmt = "WHERE requestor = submitter"
2693 This used to be the only way to compare columns. Use the superior L</-ident>
2694 method for all new code. For example an identifier declared in such a way
2695 will be properly quoted if L</quote_char> is properly set, while the legacy
2696 form will remain as supplied.
2700 my %where = ( is_ready => \"", completed => { '>', '2012-12-21' } )
2702 $stmt = "WHERE completed > ? AND is_ready"
2703 @bind = ('2012-12-21')
2705 Using an empty string literal used to be the only way to express a boolean.
2706 For all new code please use the much more readable
2707 L<-bool|/Unary operators: bool> operator.
2713 These pages could go on for a while, since the nesting of the data
2714 structures this module can handle are pretty much unlimited (the
2715 module implements the C<WHERE> expansion as a recursive function
2716 internally). Your best bet is to "play around" with the module a
2717 little to see how the data structures behave, and choose the best
2718 format for your data based on that.
2720 And of course, all the values above will probably be replaced with
2721 variables gotten from forms or the command line. After all, if you
2722 knew everything ahead of time, you wouldn't have to worry about
2723 dynamically-generating SQL and could just hardwire it into your
2726 =head1 ORDER BY CLAUSES
2728 Some functions take an order by clause. This can either be a scalar (just a
2729 column name), a hashref of C<< { -desc => 'col' } >> or C<< { -asc => 'col' }
2730 >>, a scalarref, an arrayref-ref, or an arrayref of any of the previous
2733 Given | Will Generate
2734 ---------------------------------------------------------------
2736 'colA' | ORDER BY colA
2738 [qw/colA colB/] | ORDER BY colA, colB
2740 {-asc => 'colA'} | ORDER BY colA ASC
2742 {-desc => 'colB'} | ORDER BY colB DESC
2744 ['colA', {-asc => 'colB'}] | ORDER BY colA, colB ASC
2746 { -asc => [qw/colA colB/] } | ORDER BY colA ASC, colB ASC
2748 \'colA DESC' | ORDER BY colA DESC
2750 \[ 'FUNC(colA, ?)', $x ] | ORDER BY FUNC(colA, ?)
2751 | /* ...with $x bound to ? */
2754 { -asc => 'colA' }, | colA ASC,
2755 { -desc => [qw/colB/] }, | colB DESC,
2756 { -asc => [qw/colC colD/] },| colC ASC, colD ASC,
2757 \'colE DESC', | colE DESC,
2758 \[ 'FUNC(colF, ?)', $x ], | FUNC(colF, ?)
2759 ] | /* ...with $x bound to ? */
2760 ===============================================================
2764 =head1 SPECIAL OPERATORS
2766 my $sqlmaker = SQL::Abstract->new(special_ops => [
2770 my ($self, $field, $op, $arg) = @_;
2776 handler => 'method_name',
2780 A "special operator" is a SQL syntactic clause that can be
2781 applied to a field, instead of a usual binary operator.
2784 WHERE field IN (?, ?, ?)
2785 WHERE field BETWEEN ? AND ?
2786 WHERE MATCH(field) AGAINST (?, ?)
2788 Special operators IN and BETWEEN are fairly standard and therefore
2789 are builtin within C<SQL::Abstract> (as the overridable methods
2790 C<_where_field_IN> and C<_where_field_BETWEEN>). For other operators,
2791 like the MATCH .. AGAINST example above which is specific to MySQL,
2792 you can write your own operator handlers - supply a C<special_ops>
2793 argument to the C<new> method. That argument takes an arrayref of
2794 operator definitions; each operator definition is a hashref with two
2801 the regular expression to match the operator
2805 Either a coderef or a plain scalar method name. In both cases
2806 the expected return is C<< ($sql, @bind) >>.
2808 When supplied with a method name, it is simply called on the
2809 L<SQL::Abstract> object as:
2811 $self->$method_name($field, $op, $arg)
2815 $field is the LHS of the operator
2816 $op is the part that matched the handler regex
2819 When supplied with a coderef, it is called as:
2821 $coderef->($self, $field, $op, $arg)
2826 For example, here is an implementation
2827 of the MATCH .. AGAINST syntax for MySQL
2829 my $sqlmaker = SQL::Abstract->new(special_ops => [
2831 # special op for MySql MATCH (field) AGAINST(word1, word2, ...)
2832 {regex => qr/^match$/i,
2834 my ($self, $field, $op, $arg) = @_;
2835 $arg = [$arg] if not ref $arg;
2836 my $label = $self->_quote($field);
2837 my ($placeholder) = $self->_convert('?');
2838 my $placeholders = join ", ", (($placeholder) x @$arg);
2839 my $sql = $self->_sqlcase('match') . " ($label) "
2840 . $self->_sqlcase('against') . " ($placeholders) ";
2841 my @bind = $self->_bindtype($field, @$arg);
2842 return ($sql, @bind);
2849 =head1 UNARY OPERATORS
2851 my $sqlmaker = SQL::Abstract->new(unary_ops => [
2855 my ($self, $op, $arg) = @_;
2861 handler => 'method_name',
2865 A "unary operator" is a SQL syntactic clause that can be
2866 applied to a field - the operator goes before the field
2868 You can write your own operator handlers - supply a C<unary_ops>
2869 argument to the C<new> method. That argument takes an arrayref of
2870 operator definitions; each operator definition is a hashref with two
2877 the regular expression to match the operator
2881 Either a coderef or a plain scalar method name. In both cases
2882 the expected return is C<< $sql >>.
2884 When supplied with a method name, it is simply called on the
2885 L<SQL::Abstract> object as:
2887 $self->$method_name($op, $arg)
2891 $op is the part that matched the handler regex
2892 $arg is the RHS or argument of the operator
2894 When supplied with a coderef, it is called as:
2896 $coderef->($self, $op, $arg)
2904 Thanks to some benchmarking by Mark Stosberg, it turns out that
2905 this module is many orders of magnitude faster than using C<DBIx::Abstract>.
2906 I must admit this wasn't an intentional design issue, but it's a
2907 byproduct of the fact that you get to control your C<DBI> handles
2910 To maximize performance, use a code snippet like the following:
2912 # prepare a statement handle using the first row
2913 # and then reuse it for the rest of the rows
2915 for my $href (@array_of_hashrefs) {
2916 $stmt ||= $sql->insert('table', $href);
2917 $sth ||= $dbh->prepare($stmt);
2918 $sth->execute($sql->values($href));
2921 The reason this works is because the keys in your C<$href> are sorted
2922 internally by B<SQL::Abstract>. Thus, as long as your data retains
2923 the same structure, you only have to generate the SQL the first time
2924 around. On subsequent queries, simply use the C<values> function provided
2925 by this module to return your values in the correct order.
2927 However this depends on the values having the same type - if, for
2928 example, the values of a where clause may either have values
2929 (resulting in sql of the form C<column = ?> with a single bind
2930 value), or alternatively the values might be C<undef> (resulting in
2931 sql of the form C<column IS NULL> with no bind value) then the
2932 caching technique suggested will not work.
2936 If you use my C<CGI::FormBuilder> module at all, you'll hopefully
2937 really like this part (I do, at least). Building up a complex query
2938 can be as simple as the following:
2945 use CGI::FormBuilder;
2948 my $form = CGI::FormBuilder->new(...);
2949 my $sql = SQL::Abstract->new;
2951 if ($form->submitted) {
2952 my $field = $form->field;
2953 my $id = delete $field->{id};
2954 my($stmt, @bind) = $sql->update('table', $field, {id => $id});
2957 Of course, you would still have to connect using C<DBI> to run the
2958 query, but the point is that if you make your form look like your
2959 table, the actual query script can be extremely simplistic.
2961 If you're B<REALLY> lazy (I am), check out C<HTML::QuickTable> for
2962 a fast interface to returning and formatting data. I frequently
2963 use these three modules together to write complex database query
2964 apps in under 50 lines.
2966 =head1 HOW TO CONTRIBUTE
2968 Contributions are always welcome, in all usable forms (we especially
2969 welcome documentation improvements). The delivery methods include git-
2970 or unified-diff formatted patches, GitHub pull requests, or plain bug
2971 reports either via RT or the Mailing list. Contributors are generally
2972 granted full access to the official repository after their first several
2973 patches pass successful review.
2975 This project is maintained in a git repository. The code and related tools are
2976 accessible at the following locations:
2980 =item * Official repo: L<git://git.shadowcat.co.uk/dbsrgits/SQL-Abstract.git>
2982 =item * Official gitweb: L<http://git.shadowcat.co.uk/gitweb/gitweb.cgi?p=dbsrgits/SQL-Abstract.git>
2984 =item * GitHub mirror: L<https://github.com/dbsrgits/sql-abstract>
2986 =item * Authorized committers: L<ssh://dbsrgits@git.shadowcat.co.uk/SQL-Abstract.git>
2992 Version 1.50 was a major internal refactoring of C<SQL::Abstract>.
2993 Great care has been taken to preserve the I<published> behavior
2994 documented in previous versions in the 1.* family; however,
2995 some features that were previously undocumented, or behaved
2996 differently from the documentation, had to be changed in order
2997 to clarify the semantics. Hence, client code that was relying
2998 on some dark areas of C<SQL::Abstract> v1.*
2999 B<might behave differently> in v1.50.
3001 The main changes are:
3007 support for literal SQL through the C<< \ [ $sql, @bind ] >> syntax.
3011 support for the { operator => \"..." } construct (to embed literal SQL)
3015 support for the { operator => \["...", @bind] } construct (to embed literal SQL with bind values)
3019 optional support for L<array datatypes|/"Inserting and Updating Arrays">
3023 defensive programming: check arguments
3027 fixed bug with global logic, which was previously implemented
3028 through global variables yielding side-effects. Prior versions would
3029 interpret C<< [ {cond1, cond2}, [cond3, cond4] ] >>
3030 as C<< "(cond1 AND cond2) OR (cond3 AND cond4)" >>.
3031 Now this is interpreted
3032 as C<< "(cond1 AND cond2) OR (cond3 OR cond4)" >>.
3037 fixed semantics of _bindtype on array args
3041 dropped the C<_anoncopy> of the %where tree. No longer necessary,
3042 we just avoid shifting arrays within that tree.
3046 dropped the C<_modlogic> function
3050 =head1 ACKNOWLEDGEMENTS
3052 There are a number of individuals that have really helped out with
3053 this module. Unfortunately, most of them submitted bugs via CPAN
3054 so I have no idea who they are! But the people I do know are:
3056 Ash Berlin (order_by hash term support)
3057 Matt Trout (DBIx::Class support)
3058 Mark Stosberg (benchmarking)
3059 Chas Owens (initial "IN" operator support)
3060 Philip Collins (per-field SQL functions)
3061 Eric Kolve (hashref "AND" support)
3062 Mike Fragassi (enhancements to "BETWEEN" and "LIKE")
3063 Dan Kubb (support for "quote_char" and "name_sep")
3064 Guillermo Roditi (patch to cleanup "IN" and "BETWEEN", fix and tests for _order_by)
3065 Laurent Dami (internal refactoring, extensible list of special operators, literal SQL)
3066 Norbert Buchmuller (support for literal SQL in hashpair, misc. fixes & tests)
3067 Peter Rabbitson (rewrite of SQLA::Test, misc. fixes & tests)
3068 Oliver Charles (support for "RETURNING" after "INSERT")
3074 L<DBIx::Class>, L<DBIx::Abstract>, L<CGI::FormBuilder>, L<HTML::QuickTable>.
3078 Copyright (c) 2001-2007 Nathan Wiger <nwiger@cpan.org>. All Rights Reserved.
3080 This module is actively maintained by Matt Trout <mst@shadowcatsystems.co.uk>
3082 For support, your best bet is to try the C<DBIx::Class> users mailing list.
3083 While not an official support venue, C<DBIx::Class> makes heavy use of
3084 C<SQL::Abstract>, and as such list members there are very familiar with
3085 how to create queries.
3089 This module is free software; you may copy this under the same
3090 terms as perl itself (either the GNU General Public License or
3091 the Artistic License)