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/^ ident $/ix, handler => sub { die "NOPE" }},
43 {regex => qr/^ value $/ix, handler => sub { die "NOPE" }},
44 {regex => qr/^ is (?: \s+ not )? $/ix, handler => sub { die "NOPE" }},
47 # unaryish operators - key maps to handler
48 my @BUILTIN_UNARY_OPS = (
49 # the digits are backcompat stuff
50 { regex => qr/^ and (?: [_\s]? \d+ )? $/xi, handler => '_where_op_ANDOR' },
51 { regex => qr/^ or (?: [_\s]? \d+ )? $/xi, handler => '_where_op_ANDOR' },
52 { regex => qr/^ nest (?: [_\s]? \d+ )? $/xi, handler => '_where_op_NEST' },
53 { regex => qr/^ (?: not \s )? bool $/xi, handler => '_where_op_BOOL' },
54 { regex => qr/^ ident $/xi, handler => '_where_op_IDENT' },
55 { regex => qr/^ value $/xi, handler => '_where_op_VALUE' },
56 { regex => qr/^ op $/xi, handler => '_where_op_OP' },
57 { regex => qr/^ bind $/xi, handler => '_where_op_BIND' },
58 { regex => qr/^ literal $/xi, handler => '_where_op_LITERAL' },
59 { regex => qr/^ func $/xi, handler => '_where_op_FUNC' },
62 #======================================================================
63 # DEBUGGING AND ERROR REPORTING
64 #======================================================================
67 return unless $_[0]->{debug}; shift; # a little faster
68 my $func = (caller(1))[3];
69 warn "[$func] ", @_, "\n";
73 my($func) = (caller(1))[3];
74 Carp::carp "[$func] Warning: ", @_;
78 my($func) = (caller(1))[3];
79 Carp::croak "[$func] Fatal: ", @_;
82 sub is_literal_value ($) {
83 ref $_[0] eq 'SCALAR' ? [ ${$_[0]} ]
84 : ( ref $_[0] eq 'REF' and ref ${$_[0]} eq 'ARRAY' ) ? [ @${ $_[0] } ]
88 # FIXME XSify - this can be done so much more efficiently
89 sub is_plain_value ($) {
91 ! length ref $_[0] ? \($_[0])
93 ref $_[0] eq 'HASH' and keys %{$_[0]} == 1
95 exists $_[0]->{-value}
96 ) ? \($_[0]->{-value})
98 # reuse @_ for even moar speedz
99 defined ( $_[1] = Scalar::Util::blessed $_[0] )
101 # deliberately not using Devel::OverloadInfo - the checks we are
102 # intersted in are much more limited than the fullblown thing, and
103 # this is a very hot piece of code
105 # simply using ->can('(""') can leave behind stub methods that
106 # break actually using the overload later (see L<perldiag/Stub
107 # found while resolving method "%s" overloading "%s" in package
108 # "%s"> and the source of overload::mycan())
110 # either has stringification which DBI SHOULD prefer out of the box
111 grep { *{ (qq[${_}::(""]) }{CODE} } @{ $_[2] = mro::get_linear_isa( $_[1] ) }
113 # has nummification or boolification, AND fallback is *not* disabled
115 SQL::Abstract::_ENV_::DETECT_AUTOGENERATED_STRINGIFICATION
118 grep { *{"${_}::(0+"}{CODE} } @{$_[2]}
120 grep { *{"${_}::(bool"}{CODE} } @{$_[2]}
124 # no fallback specified at all
125 ! ( ($_[3]) = grep { *{"${_}::()"}{CODE} } @{$_[2]} )
127 # fallback explicitly undef
128 ! defined ${"$_[3]::()"}
141 #======================================================================
143 #======================================================================
147 my $class = ref($self) || $self;
148 my %opt = (ref $_[0] eq 'HASH') ? %{$_[0]} : @_;
150 # choose our case by keeping an option around
151 delete $opt{case} if $opt{case} && $opt{case} ne 'lower';
153 # default logic for interpreting arrayrefs
154 $opt{logic} = $opt{logic} ? uc $opt{logic} : 'OR';
156 # how to return bind vars
157 $opt{bindtype} ||= 'normal';
159 # default comparison is "=", but can be overridden
162 # try to recognize which are the 'equality' and 'inequality' ops
163 # (temporary quickfix (in 2007), should go through a more seasoned API)
164 $opt{equality_op} = qr/^( \Q$opt{cmp}\E | \= )$/ix;
165 $opt{inequality_op} = qr/^( != | <> )$/ix;
167 $opt{like_op} = qr/^ (is\s+)? r?like $/xi;
168 $opt{not_like_op} = qr/^ (is\s+)? not \s+ r?like $/xi;
171 $opt{sqltrue} ||= '1=1';
172 $opt{sqlfalse} ||= '0=1';
175 $opt{user_special_ops} = [ @{$opt{special_ops} ||= []} ];
176 # regexes are applied in order, thus push after user-defines
177 push @{$opt{special_ops}}, @BUILTIN_SPECIAL_OPS;
180 $opt{unary_ops} ||= [];
181 push @{$opt{unary_ops}}, @BUILTIN_UNARY_OPS;
183 # rudimentary sanity-check for user supplied bits treated as functions/operators
184 # If a purported function matches this regular expression, an exception is thrown.
185 # Literal SQL is *NOT* subject to this check, only functions (and column names
186 # when quoting is not in effect)
189 # need to guard against ()'s in column names too, but this will break tons of
190 # hacks... ideas anyone?
191 $opt{injection_guard} ||= qr/
197 return bless \%opt, $class;
201 sub _assert_pass_injection_guard {
202 if ($_[1] =~ $_[0]->{injection_guard}) {
203 my $class = ref $_[0];
204 puke "Possible SQL injection attempt '$_[1]'. If this is indeed a part of the "
205 . "desired SQL use literal SQL ( \'...' or \[ '...' ] ) or supply your own "
206 . "{injection_guard} attribute to ${class}->new()"
211 #======================================================================
213 #======================================================================
217 my $table = $self->_table(shift);
218 my $data = shift || return;
221 my $method = $self->_METHOD_FOR_refkind("_insert", $data);
222 my ($sql, @bind) = $self->$method($data);
223 $sql = join " ", $self->_sqlcase('insert into'), $table, $sql;
225 if ($options->{returning}) {
226 my ($s, @b) = $self->_insert_returning($options);
231 return wantarray ? ($sql, @bind) : $sql;
234 # So that subclasses can override INSERT ... RETURNING separately from
235 # UPDATE and DELETE (e.g. DBIx::Class::SQLMaker::Oracle does this)
236 sub _insert_returning { shift->_returning(@_) }
239 my ($self, $options) = @_;
241 my $f = $options->{returning};
243 my $fieldlist = $self->_SWITCH_refkind($f, {
244 ARRAYREF => sub {join ', ', map { $self->_quote($_) } @$f;},
245 SCALAR => sub {$self->_quote($f)},
246 SCALARREF => sub {$$f},
248 return $self->_sqlcase(' returning ') . $fieldlist;
251 sub _insert_HASHREF { # explicit list of fields and then values
252 my ($self, $data) = @_;
254 my @fields = sort keys %$data;
256 my ($sql, @bind) = $self->_insert_values($data);
259 $_ = $self->_quote($_) foreach @fields;
260 $sql = "( ".join(", ", @fields).") ".$sql;
262 return ($sql, @bind);
265 sub _insert_ARRAYREF { # just generate values(?,?) part (no list of fields)
266 my ($self, $data) = @_;
268 # no names (arrayref) so can't generate bindtype
269 $self->{bindtype} ne 'columns'
270 or belch "can't do 'columns' bindtype when called with arrayref";
272 my (@values, @all_bind);
273 foreach my $value (@$data) {
274 my ($values, @bind) = $self->_insert_value(undef, $value);
275 push @values, $values;
276 push @all_bind, @bind;
278 my $sql = $self->_sqlcase('values')." ( ".join(", ", @values)." )";
279 return ($sql, @all_bind);
282 sub _insert_ARRAYREFREF { # literal SQL with bind
283 my ($self, $data) = @_;
285 my ($sql, @bind) = @${$data};
286 $self->_assert_bindval_matches_bindtype(@bind);
288 return ($sql, @bind);
292 sub _insert_SCALARREF { # literal SQL without bind
293 my ($self, $data) = @_;
299 my ($self, $data) = @_;
301 my (@values, @all_bind);
302 foreach my $column (sort keys %$data) {
303 my ($values, @bind) = $self->_insert_value($column, $data->{$column});
304 push @values, $values;
305 push @all_bind, @bind;
307 my $sql = $self->_sqlcase('values')." ( ".join(", ", @values)." )";
308 return ($sql, @all_bind);
312 my ($self, $column, $v) = @_;
314 my (@values, @all_bind);
315 $self->_SWITCH_refkind($v, {
318 if ($self->{array_datatypes}) { # if array datatype are activated
320 push @all_bind, $self->_bindtype($column, $v);
322 else { # else literal SQL with bind
323 my ($sql, @bind) = @$v;
324 $self->_assert_bindval_matches_bindtype(@bind);
326 push @all_bind, @bind;
330 ARRAYREFREF => sub { # literal SQL with bind
331 my ($sql, @bind) = @${$v};
332 $self->_assert_bindval_matches_bindtype(@bind);
334 push @all_bind, @bind;
337 # THINK: anything useful to do with a HASHREF ?
338 HASHREF => sub { # (nothing, but old SQLA passed it through)
339 #TODO in SQLA >= 2.0 it will die instead
340 belch "HASH ref as bind value in insert is not supported";
342 push @all_bind, $self->_bindtype($column, $v);
345 SCALARREF => sub { # literal SQL without bind
349 SCALAR_or_UNDEF => sub {
351 push @all_bind, $self->_bindtype($column, $v);
356 my $sql = join(", ", @values);
357 return ($sql, @all_bind);
362 #======================================================================
364 #======================================================================
369 my $table = $self->_table(shift);
370 my $data = shift || return;
374 # first build the 'SET' part of the sql statement
375 puke "Unsupported data type specified to \$sql->update"
376 unless ref $data eq 'HASH';
378 my ($sql, @all_bind) = $self->_update_set_values($data);
379 $sql = $self->_sqlcase('update ') . $table . $self->_sqlcase(' set ')
383 my($where_sql, @where_bind) = $self->where($where);
385 push @all_bind, @where_bind;
388 if ($options->{returning}) {
389 my ($returning_sql, @returning_bind) = $self->_update_returning($options);
390 $sql .= $returning_sql;
391 push @all_bind, @returning_bind;
394 return wantarray ? ($sql, @all_bind) : $sql;
397 sub _update_set_values {
398 my ($self, $data) = @_;
400 my (@set, @all_bind);
401 for my $k (sort keys %$data) {
404 my $label = $self->_quote($k);
406 $self->_SWITCH_refkind($v, {
408 if ($self->{array_datatypes}) { # array datatype
409 push @set, "$label = ?";
410 push @all_bind, $self->_bindtype($k, $v);
412 else { # literal SQL with bind
413 my ($sql, @bind) = @$v;
414 $self->_assert_bindval_matches_bindtype(@bind);
415 push @set, "$label = $sql";
416 push @all_bind, @bind;
419 ARRAYREFREF => sub { # literal SQL with bind
420 my ($sql, @bind) = @${$v};
421 $self->_assert_bindval_matches_bindtype(@bind);
422 push @set, "$label = $sql";
423 push @all_bind, @bind;
425 SCALARREF => sub { # literal SQL without bind
426 push @set, "$label = $$v";
429 my ($op, $arg, @rest) = %$v;
431 puke 'Operator calls in update must be in the form { -op => $arg }'
432 if (@rest or not $op =~ /^\-(.+)/);
434 local $self->{_nested_func_lhs} = $k;
435 my ($sql, @bind) = $self->_where_unary_op($1, $arg);
437 push @set, "$label = $sql";
438 push @all_bind, @bind;
440 SCALAR_or_UNDEF => sub {
441 push @set, "$label = ?";
442 push @all_bind, $self->_bindtype($k, $v);
448 my $sql = join ', ', @set;
450 return ($sql, @all_bind);
453 # So that subclasses can override UPDATE ... RETURNING separately from
455 sub _update_returning { shift->_returning(@_) }
459 #======================================================================
461 #======================================================================
466 my $table = $self->_table(shift);
467 my $fields = shift || '*';
471 my ($fields_sql, @bind) = $self->_select_fields($fields);
473 my ($where_sql, @where_bind) = $self->where($where, $order);
474 push @bind, @where_bind;
476 my $sql = join(' ', $self->_sqlcase('select'), $fields_sql,
477 $self->_sqlcase('from'), $table)
480 return wantarray ? ($sql, @bind) : $sql;
484 my ($self, $fields) = @_;
485 return ref $fields eq 'ARRAY' ? join ', ', map { $self->_quote($_) } @$fields
489 #======================================================================
491 #======================================================================
496 my $table = $self->_table(shift);
500 my($where_sql, @bind) = $self->where($where);
501 my $sql = $self->_sqlcase('delete from ') . $table . $where_sql;
503 if ($options->{returning}) {
504 my ($returning_sql, @returning_bind) = $self->_delete_returning($options);
505 $sql .= $returning_sql;
506 push @bind, @returning_bind;
509 return wantarray ? ($sql, @bind) : $sql;
512 # So that subclasses can override DELETE ... RETURNING separately from
514 sub _delete_returning { shift->_returning(@_) }
518 #======================================================================
520 #======================================================================
524 # Finally, a separate routine just to handle WHERE clauses
526 my ($self, $where, $order) = @_;
529 my ($sql, @bind) = $self->_recurse_where($where);
530 $sql = (defined $sql and length $sql) ? $self->_sqlcase(' where ') . "( $sql )" : '';
534 my ($order_sql, @order_bind) = $self->_order_by($order);
536 push @bind, @order_bind;
539 return wantarray ? ($sql, @bind) : $sql;
543 my ($self, $expr, $logic) = @_;
544 return undef unless defined($expr);
545 if (ref($expr) eq 'HASH') {
546 if (keys %$expr > 1) {
550 map $self->_expand_expr_hashpair($_ => $expr->{$_}, $logic),
554 return unless %$expr;
555 return $self->_expand_expr_hashpair(%$expr, $logic);
557 if (ref($expr) eq 'ARRAY') {
558 my $logic = lc($logic || $self->{logic});
559 $logic eq 'and' or $logic eq 'or' or puke "unknown logic: $logic";
565 while (my ($el) = splice @expr, 0, 1) {
566 puke "Supplying an empty left hand side argument is not supported in array-pairs"
567 unless defined($el) and length($el);
568 my $elref = ref($el);
570 push(@res, $self->_expand_expr({ $el, shift(@expr) }));
571 } elsif ($elref eq 'ARRAY') {
572 push(@res, $self->_expand_expr($el)) if @$el;
573 } elsif (is_literal_value($el)) {
575 } elsif ($elref eq 'HASH') {
576 push @res, $self->_expand_expr($el);
581 return { '-'.$logic => \@res };
583 if (my $literal = is_literal_value($expr)) {
584 return +{ -literal => $literal };
586 if (!ref($expr) or Scalar::Util::blessed($expr)) {
587 if (my $m = our $Cur_Col_Meta) {
588 return +{ -bind => [ $m, $expr ] };
590 return +{ -value => $expr };
595 sub _expand_expr_hashpair {
596 my ($self, $k, $v, $logic) = @_;
597 unless (defined($k) and length($k)) {
598 if (defined($k) and my $literal = is_literal_value($v)) {
599 belch 'Hash-pairs consisting of an empty string with a literal are deprecated, and will be removed in 2.0: use -and => [ $literal ] instead';
600 return { -literal => $literal };
602 puke "Supplying an empty left hand side argument is not supported";
605 $self->_assert_pass_injection_guard($k =~ /^-(.*)$/s);
606 if ($k =~ s/ [_\s]? \d+ $//x ) {
607 belch 'Use of [and|or|nest]_N modifiers is deprecated and will be removed in SQLA v2.0. '
608 . "You probably wanted ...-and => [ $k => COND1, $k => COND2 ... ]";
611 return $self->_expand_expr($v);
615 return $self->_expand_expr($v);
617 puke "-bool => undef not supported" unless defined($v);
618 return { -ident => $v };
621 return { -op => [ 'not', $self->_expand_expr($v) ] };
623 if (my ($rest) = $k =~/^-not[_ ](.*)$/) {
626 $self->_expand_expr_hashpair("-${rest}", $v, $logic)
629 if (my ($logic) = $k =~ /^-(and|or)$/i) {
630 if (ref($v) eq 'HASH') {
631 return $self->_expand_expr($v, $logic);
633 if (ref($v) eq 'ARRAY') {
634 return $self->_expand_expr($v, $logic);
639 $op =~ s/^-// if length($op) > 1;
641 # top level special ops are illegal in general
642 puke "Illegal use of top-level '-$op'"
643 if !(defined $self->{_nested_func_lhs})
644 and List::Util::first { $op =~ $_->{regex} } @{$self->{special_ops}}
645 and not List::Util::first { $op =~ $_->{regex} } @{$self->{unary_ops}};
647 if ($k eq '-value' and my $m = our $Cur_Col_Meta) {
648 return +{ -bind => [ $m, $v ] };
650 if ($k eq '-op' or $k eq '-ident' or $k eq '-value' or $k eq '-bind' or $k eq '-literal' or $k eq '-func') {
656 and (keys %$v)[0] =~ /^-/
658 my ($func) = $k =~ /^-(.*)$/;
659 return +{ -func => [ $func, $self->_expand_expr($v) ] };
661 if (!ref($v) or is_literal_value($v)) {
662 return +{ -op => [ $k =~ /^-(.*)$/, $self->_expand_expr($v) ] };
669 and exists $v->{-value}
670 and not defined $v->{-value}
673 return $self->_expand_expr_hashpair($k => { $self->{cmp} => undef });
675 if (!ref($v) or Scalar::Util::blessed($v)) {
680 { -bind => [ $k, $v ] }
684 if (ref($v) eq 'HASH') {
687 map $self->_expand_expr_hashpair($k => { $_ => $v->{$_} }),
694 $self->_assert_pass_injection_guard($vk);
695 if ($vk =~ s/ [_\s]? \d+ $//x ) {
696 belch 'Use of [and|or|nest]_N modifiers is deprecated and will be removed in SQLA v2.0. '
697 . "You probably wanted ...-and => [ -$vk => COND1, -$vk => COND2 ... ]";
699 if ($vk =~ /^(?:not[ _])?between$/) {
700 local our $Cur_Col_Meta = $k;
701 my @rhs = map $self->_expand_expr($_),
702 ref($vv) eq 'ARRAY' ? @$vv : $vv;
704 (@rhs == 1 and ref($rhs[0]) eq 'HASH' and $rhs[0]->{-literal})
706 (@rhs == 2 and defined($rhs[0]) and defined($rhs[1]))
708 puke "Operator '${\uc($vk)}' requires either an arrayref with two defined values or expressions, or a single literal scalarref/arrayref-ref";
711 join(' ', split '_', $vk),
716 if ($vk =~ /^(?:not[ _])?in$/) {
717 if (my $literal = is_literal_value($vv)) {
718 my ($sql, @bind) = @$literal;
719 my $opened_sql = $self->_open_outer_paren($sql);
721 $vk, { -ident => $k },
722 [ { -literal => [ $opened_sql, @bind ] } ]
726 'SQL::Abstract before v1.75 used to generate incorrect SQL when the '
727 . "-${\uc($vk)} operator was given an undef-containing list: !!!AUDIT YOUR CODE "
728 . 'AND DATA!!! (the upcoming Data::Query-based version of SQL::Abstract '
729 . 'will emit the logically correct SQL instead of raising this exception)'
731 puke("Argument passed to the '${\uc($vk)}' operator can not be undefined")
733 my @rhs = map $self->_expand_expr($_),
734 map { ref($_) ? $_ : { -bind => [ $k, $_ ] } }
735 map { defined($_) ? $_: puke($undef_err) }
736 (ref($vv) eq 'ARRAY' ? @$vv : $vv);
738 -literal => [ $self->{$vk =~ /^not/ ? 'sqltrue' : 'sqlfalse'} ]
742 join(' ', split '_', $vk),
747 if ($vk eq 'ident') {
748 if (! defined $vv or ref $vv) {
749 puke "-$vk requires a single plain scalar argument (a quotable identifier)";
757 if ($vk eq 'value') {
758 return $self->_expand_expr_hashpair($k, undef) unless defined($vv);
762 { -bind => [ $k, $vv ] }
765 if ($vk =~ /^is(?:[ _]not)?$/) {
766 puke "$vk can only take undef as argument"
770 and exists($vv->{-value})
771 and !defined($vv->{-value})
774 return +{ -op => [ $vk.' null', { -ident => $k } ] };
776 if ($vk =~ /^(and|or)$/) {
777 if (ref($vv) eq 'HASH') {
778 return +{ "-${vk}" => [
779 map $self->_expand_expr_hashpair($k, { $_ => $vv->{$_} }),
784 if (my $us = List::Util::first { $vk =~ $_->{regex} } @{$self->{user_special_ops}}) {
785 return { -op => [ $vk, { -ident => $k }, $vv ] };
787 if (ref($vv) eq 'ARRAY') {
788 my ($logic, @values) = (
789 (defined($vv->[0]) and $vv->[0] =~ /^-(and|or)$/i)
794 $vk =~ $self->{inequality_op}
795 or join(' ', split '_', $vk) =~ $self->{not_like_op}
797 if (lc($logic) eq '-or' and @values > 1) {
798 my $op = uc join ' ', split '_', $vk;
799 belch "A multi-element arrayref as an argument to the inequality op '$op' "
800 . 'is technically equivalent to an always-true 1=1 (you probably wanted '
801 . "to say ...{ \$inequality_op => [ -and => \@values ] }... instead)"
806 # try to DWIM on equality operators
807 my $op = join ' ', split '_', $vk;
809 $op =~ $self->{equality_op} ? $self->{sqlfalse}
810 : $op =~ $self->{like_op} ? belch("Supplying an empty arrayref to '@{[ uc $op]}' is deprecated") && $self->{sqlfalse}
811 : $op =~ $self->{inequality_op} ? $self->{sqltrue}
812 : $op =~ $self->{not_like_op} ? belch("Supplying an empty arrayref to '@{[ uc $op]}' is deprecated") && $self->{sqltrue}
813 : puke "operator '$op' applied on an empty array (field '$k')";
815 return +{ $logic => [
816 map $self->_expand_expr_hashpair($k => { $vk => $_ }),
824 and exists $vv->{-value}
825 and not defined $vv->{-value}
828 my $op = join ' ', split '_', $vk;
830 $op =~ /^not$/i ? 'is not' # legacy
831 : $op =~ $self->{equality_op} ? 'is'
832 : $op =~ $self->{like_op} ? belch("Supplying an undefined argument to '@{[ uc $op]}' is deprecated") && 'is'
833 : $op =~ $self->{inequality_op} ? 'is not'
834 : $op =~ $self->{not_like_op} ? belch("Supplying an undefined argument to '@{[ uc $op]}' is deprecated") && 'is not'
835 : puke "unexpected operator '$op' with undef operand";
836 return +{ -op => [ $is.' null', { -ident => $k } ] };
838 local our $Cur_Col_Meta = $k;
842 $self->_expand_expr($vv)
845 if (ref($v) eq 'ARRAY') {
846 return $self->{sqlfalse} unless @$v;
847 $self->_debug("ARRAY($k) means distribute over elements");
849 $v->[0] =~ /^-((?:and|or))$/i
850 ? ($v = [ @{$v}[1..$#$v] ], $1)
851 : ($self->{logic} || 'or')
853 return +{ "-${this_logic}" => [ map $self->_expand_expr({ $k => $_ }, $this_logic), @$v ] };
855 if (my $literal = is_literal_value($v)) {
857 belch 'Hash-pairs consisting of an empty string with a literal are deprecated, and will be removed in 2.0: use -and => [ $literal ] instead';
860 my ($sql, @bind) = @$literal;
861 if ($self->{bindtype} eq 'columns') {
863 if (!defined $_ || ref($_) ne 'ARRAY' || @$_ != 2) {
864 puke "bindtype 'columns' selected, you need to pass: [column_name => bind_value]"
868 return +{ -literal => [ $self->_quote($k).' '.$sql, @bind ] };
874 my ($self, $where, $logic) = @_;
876 #print STDERR Data::Dumper::Concise::Dumper([ $where, $logic ]);
878 my $where_exp = $self->_expand_expr($where, $logic);
880 #print STDERR Data::Dumper::Concise::Dumper([ EXP => $where_exp ]);
882 # dispatch on appropriate method according to refkind of $where
883 my $method = $self->_METHOD_FOR_refkind("_where", $where_exp);
885 my ($sql, @bind) = $self->$method($where_exp, $logic);
887 # DBIx::Class used to call _recurse_where in scalar context
888 # something else might too...
890 return ($sql, @bind);
893 belch "Calling _recurse_where in scalar context is deprecated and will go away before 2.0";
900 #======================================================================
901 # WHERE: top-level ARRAYREF
902 #======================================================================
905 sub _where_ARRAYREF {
906 my ($self, $where, $logic) = @_;
908 $logic = uc($logic || $self->{logic});
909 $logic eq 'AND' or $logic eq 'OR' or puke "unknown logic: $logic";
911 my @clauses = @$where;
913 my (@sql_clauses, @all_bind);
914 # need to use while() so can shift() for pairs
916 my $el = shift @clauses;
918 $el = undef if (defined $el and ! length $el);
920 # switch according to kind of $el and get corresponding ($sql, @bind)
921 my ($sql, @bind) = $self->_SWITCH_refkind($el, {
923 # skip empty elements, otherwise get invalid trailing AND stuff
924 ARRAYREF => sub {$self->_recurse_where($el) if @$el},
928 $self->_assert_bindval_matches_bindtype(@b);
932 HASHREF => sub {$self->_recurse_where($el, 'and') if %$el},
934 SCALARREF => sub { ($$el); },
937 # top-level arrayref with scalars, recurse in pairs
938 $self->_recurse_where({$el => shift(@clauses)})
941 UNDEF => sub {puke "Supplying an empty left hand side argument is not supported in array-pairs" },
945 push @sql_clauses, $sql;
946 push @all_bind, @bind;
950 return $self->_join_sql_clauses($logic, \@sql_clauses, \@all_bind);
953 #======================================================================
954 # WHERE: top-level ARRAYREFREF
955 #======================================================================
957 sub _where_ARRAYREFREF {
958 my ($self, $where) = @_;
959 my ($sql, @bind) = @$$where;
960 $self->_assert_bindval_matches_bindtype(@bind);
961 return ($sql, @bind);
964 #======================================================================
965 # WHERE: top-level HASHREF
966 #======================================================================
969 my ($self, $where) = @_;
970 my (@sql_clauses, @all_bind);
972 for my $k (sort keys %$where) {
973 my $v = $where->{$k};
975 # ($k => $v) is either a special unary op or a regular hashpair
976 my ($sql, @bind) = do {
978 # put the operator in canonical form
980 $op = substr $op, 1; # remove initial dash
981 $op =~ s/^\s+|\s+$//g;# remove leading/trailing space
982 $op =~ s/\s+/ /g; # compress whitespace
984 # so that -not_foo works correctly
985 $op =~ s/^not_/NOT /i;
987 $self->_debug("Unary OP(-$op) within hashref, recursing...");
988 my ($s, @b) = $self->_where_unary_op($op, $v);
990 # top level vs nested
991 # we assume that handled unary ops will take care of their ()s
993 List::Util::first {$op =~ $_->{regex}} @{$self->{unary_ops}}
995 ( defined $self->{_nested_func_lhs} and $self->{_nested_func_lhs} eq $k )
1001 if (is_literal_value ($v) ) {
1002 belch 'Hash-pairs consisting of an empty string with a literal are deprecated, and will be removed in 2.0: use -and => [ $literal ] instead';
1005 puke "Supplying an empty left hand side argument is not supported in hash-pairs";
1009 my $method = $self->_METHOD_FOR_refkind("_where_hashpair", $v);
1010 $self->$method($k, $v);
1014 push @sql_clauses, $sql;
1015 push @all_bind, @bind;
1018 return $self->_join_sql_clauses('and', \@sql_clauses, \@all_bind);
1021 sub _where_unary_op {
1022 my ($self, $op, $rhs) = @_;
1024 $op =~ s/^-// if length($op) > 1;
1026 # top level special ops are illegal in general
1027 puke "Illegal use of top-level '-$op'"
1028 if !(defined $self->{_nested_func_lhs})
1029 and List::Util::first { $op =~ $_->{regex} } @{$self->{special_ops}}
1030 and not List::Util::first { $op =~ $_->{regex} } @{$self->{unary_ops}};
1032 if (my $op_entry = List::Util::first { $op =~ $_->{regex} } @{$self->{unary_ops}}) {
1033 my $handler = $op_entry->{handler};
1035 if (not ref $handler) {
1036 if ($op =~ s/ [_\s]? \d+ $//x ) {
1037 belch 'Use of [and|or|nest]_N modifiers is deprecated and will be removed in SQLA v2.0. '
1038 . "You probably wanted ...-and => [ -$op => COND1, -$op => COND2 ... ]";
1040 return $self->$handler($op, $rhs);
1042 elsif (ref $handler eq 'CODE') {
1043 return $handler->($self, $op, $rhs);
1046 puke "Illegal handler for operator $op - expecting a method name or a coderef";
1050 $self->_debug("Generic unary OP: $op - recursing as function");
1052 $self->_assert_pass_injection_guard($op);
1054 my ($sql, @bind) = $self->_SWITCH_refkind($rhs, {
1056 puke "Illegal use of top-level '-$op'"
1057 unless defined $self->{_nested_func_lhs};
1060 $self->_convert('?'),
1061 $self->_bindtype($self->{_nested_func_lhs}, $rhs)
1065 $self->_recurse_where($rhs)
1069 $sql = sprintf('%s %s',
1070 $self->_sqlcase($op),
1074 return ($sql, @bind);
1077 sub _where_op_ANDOR {
1078 my ($self, $op, $v) = @_;
1080 $self->_SWITCH_refkind($v, {
1082 return $self->_where_ARRAYREF($v, $op);
1086 return ($op =~ /^or/i)
1087 ? $self->_where_ARRAYREF([ map { $_ => $v->{$_} } (sort keys %$v) ], $op)
1088 : $self->_where_HASHREF($v);
1092 puke "-$op => \\\$scalar makes little sense, use " .
1094 ? '[ \$scalar, \%rest_of_conditions ] instead'
1095 : '-and => [ \$scalar, \%rest_of_conditions ] instead'
1099 ARRAYREFREF => sub {
1100 puke "-$op => \\[...] makes little sense, use " .
1102 ? '[ \[...], \%rest_of_conditions ] instead'
1103 : '-and => [ \[...], \%rest_of_conditions ] instead'
1107 SCALAR => sub { # permissively interpreted as SQL
1108 puke "-$op => \$value makes little sense, use -bool => \$value instead";
1112 puke "-$op => undef not supported";
1117 sub _where_op_NEST {
1118 my ($self, $op, $v) = @_;
1120 $self->_SWITCH_refkind($v, {
1122 SCALAR => sub { # permissively interpreted as SQL
1123 belch "literal SQL should be -nest => \\'scalar' "
1124 . "instead of -nest => 'scalar' ";
1129 puke "-$op => undef not supported";
1133 $self->_recurse_where($v);
1140 sub _where_op_BOOL {
1141 my ($self, $op, $v) = @_;
1143 my ($s, @b) = $self->_SWITCH_refkind($v, {
1144 SCALAR => sub { # interpreted as SQL column
1145 $self->_convert($self->_quote($v));
1149 puke "-$op => undef not supported";
1153 $self->_recurse_where($v);
1157 $s = "(NOT $s)" if $op =~ /^not/i;
1162 sub _where_op_IDENT {
1164 my ($op, $rhs) = splice @_, -2;
1165 if (! defined $rhs or length ref $rhs) {
1166 puke "-$op requires a single plain scalar argument (a quotable identifier)";
1169 # in case we are called as a top level special op (no '=')
1170 my $has_lhs = my $lhs = shift;
1172 $_ = $self->_convert($self->_quote($_)) for ($lhs, $rhs);
1180 sub _where_op_VALUE {
1182 my ($op, $rhs) = splice @_, -2;
1184 # in case we are called as a top level special op (no '=')
1188 if (! defined $rhs) {
1190 ? $self->_where_hashpair_HASHREF($lhs, { -is => undef })
1197 (defined $lhs ? $lhs : $self->{_nested_func_lhs}),
1204 $self->_convert($self->_quote($lhs)) . ' = ' . $self->_convert('?'),
1208 $self->_convert('?'),
1215 my %unop_postfix = map +($_ => 1), 'is null', 'is not null';
1221 my ($self, $args) = @_;
1222 my ($left, $low, $high) = @$args;
1223 my ($rhsql, @rhbind) = do {
1225 puke "Single arg to between must be a literal"
1226 unless $low->{-literal};
1229 local $self->{_nested_func_lhs} = $left->{-ident}
1230 if ref($left) eq 'HASH' and $left->{-ident};
1231 my ($l, $h) = map [ $self->_where_unary_op(%$_) ], $low, $high;
1232 (join(' ', $l->[0], $self->_sqlcase('and'), $h->[0]),
1233 @{$l}[1..$#$l], @{$h}[1..$#$h])
1236 my ($lhsql, @lhbind) = $self->_recurse_where($left);
1238 join(' ', '(', $lhsql, $self->_sqlcase($op), $rhsql, ')'),
1242 }), 'between', 'not between'),
1246 my ($self, $args) = @_;
1247 my ($lhs, $rhs) = @$args;
1250 local $self->{_nested_func_lhs} = $lhs->{-ident}
1251 if ref($lhs) eq 'HASH' and $lhs->{-ident};
1252 my ($sql, @bind) = $self->_where_unary_op(%$_);
1253 push @in_bind, @bind;
1256 my ($lhsql, @lbind) = $self->_recurse_where($lhs);
1258 $lhsql.' '.$self->_sqlcase($op).' ( '
1259 .join(', ', @in_sql)
1264 }), 'in', 'not in'),
1268 my ($self, undef, $v) = @_;
1269 my ($op, @args) = @$v;
1270 $op =~ s/^-// if length($op) > 1;
1271 local $self->{_nested_func_lhs};
1272 if (my $h = $special{$op}) {
1273 return $self->$h(\@args);
1275 if (my $us = List::Util::first { $op =~ $_->{regex} } @{$self->{user_special_ops}}) {
1276 puke "Special op '${op}' requires first value to be identifier"
1277 unless my ($k) = map $_->{-ident}, grep ref($_) eq 'HASH', $args[0];
1278 return $self->${\($us->{handler})}($k, $op, $args[1]);
1280 my $final_op = $op =~ /^(?:is|not)_/ ? join(' ', split '_', $op) : $op;
1282 my ($expr_sql, @bind) = $self->_recurse_where($args[0]);
1283 my $op_sql = $self->_sqlcase($final_op);
1285 $unop_postfix{lc($final_op)}
1286 ? "${expr_sql} ${op_sql}"
1287 : "${op_sql} ${expr_sql}"
1289 return (($op eq 'not' ? '('.$final_sql.')' : $final_sql), @bind);
1291 my @parts = map [ $self->_recurse_where($_) ], @args;
1292 my ($final_sql) = map +($op =~ /^(and|or)$/ ? "(${_})" : $_), join(
1293 ' '.$self->_sqlcase($final_op).' ',
1298 map @{$_}[1..$#$_], @parts
1304 sub _where_op_FUNC {
1305 my ($self, undef, $rest) = @_;
1306 my ($func, @args) = @$rest;
1310 push @arg_sql, shift @x;
1312 } map [ $self->_recurse_where($_) ], @args;
1313 return ($self->_sqlcase($func).'('.join(', ', @arg_sql).')', @bind);
1316 sub _where_op_BIND {
1317 my ($self, undef, $bind) = @_;
1318 return ($self->_convert('?'), $self->_bindtype(@$bind));
1321 sub _where_op_LITERAL {
1322 my ($self, undef, $literal) = @_;
1323 $self->_assert_bindval_matches_bindtype(@{$literal}[1..$#$literal]);
1327 sub _where_hashpair_ARRAYREF {
1328 my ($self, $k, $v) = @_;
1331 my @v = @$v; # need copy because of shift below
1332 $self->_debug("ARRAY($k) means distribute over elements");
1334 # put apart first element if it is an operator (-and, -or)
1336 (defined $v[0] && $v[0] =~ /^ - (?: AND|OR ) $/ix)
1340 my @distributed = map { {$k => $_} } @v;
1343 $self->_debug("OP($op) reinjected into the distributed array");
1344 unshift @distributed, $op;
1347 my $logic = $op ? substr($op, 1) : '';
1349 return $self->_recurse_where(\@distributed, $logic);
1352 $self->_debug("empty ARRAY($k) means 0=1");
1353 return ($self->{sqlfalse});
1357 sub _where_hashpair_HASHREF {
1358 my ($self, $k, $v, $logic) = @_;
1361 local $self->{_nested_func_lhs} = defined $self->{_nested_func_lhs}
1362 ? $self->{_nested_func_lhs}
1366 my ($all_sql, @all_bind);
1368 for my $orig_op (sort keys %$v) {
1369 my $val = $v->{$orig_op};
1371 # put the operator in canonical form
1374 # FIXME - we need to phase out dash-less ops
1375 $op =~ s/^-//; # remove possible initial dash
1376 $op =~ s/^\s+|\s+$//g;# remove leading/trailing space
1377 $op =~ s/\s+/ /g; # compress whitespace
1379 $self->_assert_pass_injection_guard($op);
1382 $op =~ s/^is_not/IS NOT/i;
1384 # so that -not_foo works correctly
1385 $op =~ s/^not_/NOT /i;
1387 # another retarded special case: foo => { $op => { -value => undef } }
1388 if (ref $val eq 'HASH' and keys %$val == 1 and exists $val->{-value} and ! defined $val->{-value} ) {
1394 # CASE: col-value logic modifiers
1395 if ($orig_op =~ /^ \- (and|or) $/xi) {
1396 ($sql, @bind) = $self->_where_hashpair_HASHREF($k, $val, $1);
1398 # CASE: special operators like -in or -between
1399 elsif (my $special_op = List::Util::first { $op =~ $_->{regex} } @{$self->{special_ops}}) {
1400 my $handler = $special_op->{handler};
1402 puke "No handler supplied for special operator $orig_op";
1404 elsif (not ref $handler) {
1405 ($sql, @bind) = $self->$handler($k, $op, $val);
1407 elsif (ref $handler eq 'CODE') {
1408 ($sql, @bind) = $handler->($self, $k, $op, $val);
1411 puke "Illegal handler for special operator $orig_op - expecting a method name or a coderef";
1415 $self->_SWITCH_refkind($val, {
1417 ARRAYREF => sub { # CASE: col => {op => \@vals}
1418 ($sql, @bind) = $self->_where_field_op_ARRAYREF($k, $op, $val);
1421 ARRAYREFREF => sub { # CASE: col => {op => \[$sql, @bind]} (literal SQL with bind)
1422 my ($sub_sql, @sub_bind) = @$$val;
1423 $self->_assert_bindval_matches_bindtype(@sub_bind);
1424 $sql = join ' ', $self->_convert($self->_quote($k)),
1425 $self->_sqlcase($op),
1430 UNDEF => sub { # CASE: col => {op => undef} : sql "IS (NOT)? NULL"
1432 $op =~ /^not$/i ? 'is not' # legacy
1433 : $op =~ $self->{equality_op} ? 'is'
1434 : $op =~ $self->{like_op} ? belch("Supplying an undefined argument to '@{[ uc $op]}' is deprecated") && 'is'
1435 : $op =~ $self->{inequality_op} ? 'is not'
1436 : $op =~ $self->{not_like_op} ? belch("Supplying an undefined argument to '@{[ uc $op]}' is deprecated") && 'is not'
1437 : puke "unexpected operator '$orig_op' with undef operand";
1439 $sql = $self->_quote($k) . $self->_sqlcase(" $is null");
1442 FALLBACK => sub { # CASE: col => {op/func => $stuff}
1443 ($sql, @bind) = $self->_where_unary_op($op, $val);
1446 $self->_convert($self->_quote($k)),
1447 $self->{_nested_func_lhs} eq $k ? $sql : "($sql)", # top level vs nested
1453 ($all_sql) = (defined $all_sql and $all_sql) ? $self->_join_sql_clauses($logic, [$all_sql, $sql], []) : $sql;
1454 push @all_bind, @bind;
1456 return ($all_sql, @all_bind);
1459 sub _where_field_IS {
1460 my ($self, $k, $op, $v) = @_;
1462 my ($s) = $self->_SWITCH_refkind($v, {
1465 $self->_convert($self->_quote($k)),
1466 map { $self->_sqlcase($_)} ($op, 'null')
1469 puke "$op can only take undef as argument";
1476 sub _where_field_op_ARRAYREF {
1477 my ($self, $k, $op, $vals) = @_;
1479 my @vals = @$vals; #always work on a copy
1482 $self->_debug(sprintf '%s means multiple elements: [ %s ]',
1484 join(', ', map { defined $_ ? "'$_'" : 'NULL' } @vals ),
1487 # see if the first element is an -and/-or op
1489 if (defined $vals[0] && $vals[0] =~ /^ - (AND|OR) $/ix) {
1494 # a long standing API wart - an attempt to change this behavior during
1495 # the 1.50 series failed *spectacularly*. Warn instead and leave the
1500 (!$logic or $logic eq 'OR')
1502 ($op =~ $self->{inequality_op} or $op =~ $self->{not_like_op})
1505 belch "A multi-element arrayref as an argument to the inequality op '$o' "
1506 . 'is technically equivalent to an always-true 1=1 (you probably wanted '
1507 . "to say ...{ \$inequality_op => [ -and => \@values ] }... instead)"
1511 # distribute $op over each remaining member of @vals, append logic if exists
1512 return $self->_recurse_where([map { {$k => {$op, $_}} } @vals], $logic);
1516 # try to DWIM on equality operators
1518 $op =~ $self->{equality_op} ? $self->{sqlfalse}
1519 : $op =~ $self->{like_op} ? belch("Supplying an empty arrayref to '@{[ uc $op]}' is deprecated") && $self->{sqlfalse}
1520 : $op =~ $self->{inequality_op} ? $self->{sqltrue}
1521 : $op =~ $self->{not_like_op} ? belch("Supplying an empty arrayref to '@{[ uc $op]}' is deprecated") && $self->{sqltrue}
1522 : puke "operator '$op' applied on an empty array (field '$k')";
1527 sub _where_hashpair_SCALARREF {
1528 my ($self, $k, $v) = @_;
1529 $self->_debug("SCALAR($k) means literal SQL: $$v");
1530 my $sql = $self->_quote($k) . " " . $$v;
1534 # literal SQL with bind
1535 sub _where_hashpair_ARRAYREFREF {
1536 my ($self, $k, $v) = @_;
1537 $self->_debug("REF($k) means literal SQL: @${$v}");
1538 my ($sql, @bind) = @$$v;
1539 $self->_assert_bindval_matches_bindtype(@bind);
1540 $sql = $self->_quote($k) . " " . $sql;
1541 return ($sql, @bind );
1544 # literal SQL without bind
1545 sub _where_hashpair_SCALAR {
1546 my ($self, $k, $v) = @_;
1547 $self->_debug("NOREF($k) means simple key=val: $k $self->{cmp} $v");
1548 return ($self->_where_hashpair_HASHREF($k, { $self->{cmp} => $v }));
1552 sub _where_hashpair_UNDEF {
1553 my ($self, $k, $v) = @_;
1554 $self->_debug("UNDEF($k) means IS NULL");
1555 return $self->_where_hashpair_HASHREF($k, { -is => undef });
1558 #======================================================================
1559 # WHERE: TOP-LEVEL OTHERS (SCALARREF, SCALAR, UNDEF)
1560 #======================================================================
1563 sub _where_SCALARREF {
1564 my ($self, $where) = @_;
1567 $self->_debug("SCALAR(*top) means literal SQL: $$where");
1573 my ($self, $where) = @_;
1576 $self->_debug("NOREF(*top) means literal SQL: $where");
1587 #======================================================================
1588 # WHERE: BUILTIN SPECIAL OPERATORS (-in, -between)
1589 #======================================================================
1592 sub _where_field_BETWEEN {
1593 my ($self, $k, $op, $vals) = @_;
1595 my ($label, $and, $placeholder);
1596 $label = $self->_convert($self->_quote($k));
1597 $and = ' ' . $self->_sqlcase('and') . ' ';
1598 $placeholder = $self->_convert('?');
1599 $op = $self->_sqlcase($op);
1601 my $invalid_args = "Operator '$op' requires either an arrayref with two defined values or expressions, or a single literal scalarref/arrayref-ref";
1603 my ($clause, @bind) = $self->_SWITCH_refkind($vals, {
1604 ARRAYREFREF => sub {
1605 my ($s, @b) = @$$vals;
1606 $self->_assert_bindval_matches_bindtype(@b);
1613 puke $invalid_args if @$vals != 2;
1615 my (@all_sql, @all_bind);
1616 foreach my $val (@$vals) {
1617 my ($sql, @bind) = $self->_SWITCH_refkind($val, {
1619 return ($placeholder, $self->_bindtype($k, $val) );
1624 ARRAYREFREF => sub {
1625 my ($sql, @bind) = @$$val;
1626 $self->_assert_bindval_matches_bindtype(@bind);
1627 return ($sql, @bind);
1630 my ($func, $arg, @rest) = %$val;
1631 puke "Only simple { -func => arg } functions accepted as sub-arguments to BETWEEN"
1632 if (@rest or $func !~ /^ \- (.+)/x);
1633 $self->_where_unary_op($1 => $arg);
1639 push @all_sql, $sql;
1640 push @all_bind, @bind;
1644 (join $and, @all_sql),
1653 my $sql = "( $label $op $clause )";
1654 return ($sql, @bind)
1658 sub _where_field_IN {
1659 my ($self, $k, $op, $vals) = @_;
1661 # backwards compatibility: if scalar, force into an arrayref
1662 $vals = [$vals] if defined $vals && ! ref $vals;
1664 my ($label) = $self->_convert($self->_quote($k));
1665 my ($placeholder) = $self->_convert('?');
1666 $op = $self->_sqlcase($op);
1668 my ($sql, @bind) = $self->_SWITCH_refkind($vals, {
1669 ARRAYREF => sub { # list of choices
1670 if (@$vals) { # nonempty list
1671 my (@all_sql, @all_bind);
1673 for my $val (@$vals) {
1674 my ($sql, @bind) = $self->_SWITCH_refkind($val, {
1676 return ($placeholder, $val);
1681 ARRAYREFREF => sub {
1682 my ($sql, @bind) = @$$val;
1683 $self->_assert_bindval_matches_bindtype(@bind);
1684 return ($sql, @bind);
1687 my ($func, $arg, @rest) = %$val;
1688 puke "Only simple { -func => arg } functions accepted as sub-arguments to IN"
1689 if (@rest or $func !~ /^ \- (.+)/x);
1690 $self->_where_unary_op($1 => $arg);
1694 'SQL::Abstract before v1.75 used to generate incorrect SQL when the '
1695 . "-$op operator was given an undef-containing list: !!!AUDIT YOUR CODE "
1696 . 'AND DATA!!! (the upcoming Data::Query-based version of SQL::Abstract '
1697 . 'will emit the logically correct SQL instead of raising this exception)'
1701 push @all_sql, $sql;
1702 push @all_bind, @bind;
1706 sprintf('%s %s ( %s )',
1709 join(', ', @all_sql)
1711 $self->_bindtype($k, @all_bind),
1714 else { # empty list: some databases won't understand "IN ()", so DWIM
1715 my $sql = ($op =~ /\bnot\b/i) ? $self->{sqltrue} : $self->{sqlfalse};
1720 SCALARREF => sub { # literal SQL
1721 my $sql = $self->_open_outer_paren($$vals);
1722 return ("$label $op ( $sql )");
1724 ARRAYREFREF => sub { # literal SQL with bind
1725 my ($sql, @bind) = @$$vals;
1726 $self->_assert_bindval_matches_bindtype(@bind);
1727 $sql = $self->_open_outer_paren($sql);
1728 return ("$label $op ( $sql )", @bind);
1732 puke "Argument passed to the '$op' operator can not be undefined";
1736 puke "special op $op requires an arrayref (or scalarref/arrayref-ref)";
1740 return ($sql, @bind);
1743 # Some databases (SQLite) treat col IN (1, 2) different from
1744 # col IN ( (1, 2) ). Use this to strip all outer parens while
1745 # adding them back in the corresponding method
1746 sub _open_outer_paren {
1747 my ($self, $sql) = @_;
1749 while (my ($inner) = $sql =~ /^ \s* \( (.*) \) \s* $/xs) {
1751 # there are closing parens inside, need the heavy duty machinery
1752 # to reevaluate the extraction starting from $sql (full reevaluation)
1753 if ($inner =~ /\)/) {
1754 require Text::Balanced;
1756 my (undef, $remainder) = do {
1757 # idiotic design - writes to $@ but *DOES NOT* throw exceptions
1759 Text::Balanced::extract_bracketed($sql, '()', qr/\s*/);
1762 # the entire expression needs to be a balanced bracketed thing
1763 # (after an extract no remainder sans trailing space)
1764 last if defined $remainder and $remainder =~ /\S/;
1774 #======================================================================
1776 #======================================================================
1779 my ($self, $arg) = @_;
1782 for my $c ($self->_order_by_chunks($arg) ) {
1783 $self->_SWITCH_refkind($c, {
1784 SCALAR => sub { push @sql, $c },
1785 ARRAYREF => sub { push @sql, shift @$c; push @bind, @$c },
1791 $self->_sqlcase(' order by'),
1797 return wantarray ? ($sql, @bind) : $sql;
1800 sub _order_by_chunks {
1801 my ($self, $arg) = @_;
1803 return $self->_SWITCH_refkind($arg, {
1806 map { $self->_order_by_chunks($_ ) } @$arg;
1809 ARRAYREFREF => sub {
1810 my ($s, @b) = @$$arg;
1811 $self->_assert_bindval_matches_bindtype(@b);
1815 SCALAR => sub {$self->_quote($arg)},
1817 UNDEF => sub {return () },
1819 SCALARREF => sub {$$arg}, # literal SQL, no quoting
1822 # get first pair in hash
1823 my ($key, $val, @rest) = %$arg;
1825 return () unless $key;
1827 if (@rest or not $key =~ /^-(desc|asc)/i) {
1828 puke "hash passed to _order_by must have exactly one key (-desc or -asc)";
1834 for my $c ($self->_order_by_chunks($val)) {
1837 $self->_SWITCH_refkind($c, {
1842 ($sql, @bind) = @$c;
1846 $sql = $sql . ' ' . $self->_sqlcase($direction);
1848 push @ret, [ $sql, @bind];
1857 #======================================================================
1858 # DATASOURCE (FOR NOW, JUST PLAIN TABLE OR LIST OF TABLES)
1859 #======================================================================
1864 $self->_SWITCH_refkind($from, {
1865 ARRAYREF => sub {join ', ', map { $self->_quote($_) } @$from;},
1866 SCALAR => sub {$self->_quote($from)},
1867 SCALARREF => sub {$$from},
1872 #======================================================================
1874 #======================================================================
1876 # highly optimized, as it's called way too often
1878 # my ($self, $label) = @_;
1880 return '' unless defined $_[1];
1881 return ${$_[1]} if ref($_[1]) eq 'SCALAR';
1883 $_[0]->{quote_char} or
1884 ($_[0]->_assert_pass_injection_guard($_[1]), return $_[1]);
1886 my $qref = ref $_[0]->{quote_char};
1888 !$qref ? ($_[0]->{quote_char}, $_[0]->{quote_char})
1889 : ($qref eq 'ARRAY') ? @{$_[0]->{quote_char}}
1890 : puke "Unsupported quote_char format: $_[0]->{quote_char}";
1892 my $esc = $_[0]->{escape_char} || $r;
1894 # parts containing * are naturally unquoted
1895 return join($_[0]->{name_sep}||'', map
1896 +( $_ eq '*' ? $_ : do { (my $n = $_) =~ s/(\Q$esc\E|\Q$r\E)/$esc$1/g; $l . $n . $r } ),
1897 ( $_[0]->{name_sep} ? split (/\Q$_[0]->{name_sep}\E/, $_[1] ) : $_[1] )
1902 # Conversion, if applicable
1904 #my ($self, $arg) = @_;
1905 if ($_[0]->{convert}) {
1906 return $_[0]->_sqlcase($_[0]->{convert}) .'(' . $_[1] . ')';
1913 #my ($self, $col, @vals) = @_;
1914 # called often - tighten code
1915 return $_[0]->{bindtype} eq 'columns'
1916 ? map {[$_[1], $_]} @_[2 .. $#_]
1921 # Dies if any element of @bind is not in [colname => value] format
1922 # if bindtype is 'columns'.
1923 sub _assert_bindval_matches_bindtype {
1924 # my ($self, @bind) = @_;
1926 if ($self->{bindtype} eq 'columns') {
1928 if (!defined $_ || ref($_) ne 'ARRAY' || @$_ != 2) {
1929 puke "bindtype 'columns' selected, you need to pass: [column_name => bind_value]"
1935 sub _join_sql_clauses {
1936 my ($self, $logic, $clauses_aref, $bind_aref) = @_;
1938 if (@$clauses_aref > 1) {
1939 my $join = " " . $self->_sqlcase($logic) . " ";
1940 my $sql = '( ' . join($join, @$clauses_aref) . ' )';
1941 return ($sql, @$bind_aref);
1943 elsif (@$clauses_aref) {
1944 return ($clauses_aref->[0], @$bind_aref); # no parentheses
1947 return (); # if no SQL, ignore @$bind_aref
1952 # Fix SQL case, if so requested
1954 # LDNOTE: if $self->{case} is true, then it contains 'lower', so we
1955 # don't touch the argument ... crooked logic, but let's not change it!
1956 return $_[0]->{case} ? $_[1] : uc($_[1]);
1960 #======================================================================
1961 # DISPATCHING FROM REFKIND
1962 #======================================================================
1965 my ($self, $data) = @_;
1967 return 'UNDEF' unless defined $data;
1969 # blessed objects are treated like scalars
1970 my $ref = (Scalar::Util::blessed $data) ? '' : ref $data;
1972 return 'SCALAR' unless $ref;
1975 while ($ref eq 'REF') {
1977 $ref = (Scalar::Util::blessed $data) ? '' : ref $data;
1981 return ($ref||'SCALAR') . ('REF' x $n_steps);
1985 my ($self, $data) = @_;
1986 my @try = ($self->_refkind($data));
1987 push @try, 'SCALAR_or_UNDEF' if $try[0] eq 'SCALAR' || $try[0] eq 'UNDEF';
1988 push @try, 'FALLBACK';
1992 sub _METHOD_FOR_refkind {
1993 my ($self, $meth_prefix, $data) = @_;
1996 for (@{$self->_try_refkind($data)}) {
1997 $method = $self->can($meth_prefix."_".$_)
2001 return $method || puke "cannot dispatch on '$meth_prefix' for ".$self->_refkind($data);
2005 sub _SWITCH_refkind {
2006 my ($self, $data, $dispatch_table) = @_;
2009 for (@{$self->_try_refkind($data)}) {
2010 $coderef = $dispatch_table->{$_}
2014 puke "no dispatch entry for ".$self->_refkind($data)
2023 #======================================================================
2024 # VALUES, GENERATE, AUTOLOAD
2025 #======================================================================
2027 # LDNOTE: original code from nwiger, didn't touch code in that section
2028 # I feel the AUTOLOAD stuff should not be the default, it should
2029 # only be activated on explicit demand by user.
2033 my $data = shift || return;
2034 puke "Argument to ", __PACKAGE__, "->values must be a \\%hash"
2035 unless ref $data eq 'HASH';
2038 foreach my $k (sort keys %$data) {
2039 my $v = $data->{$k};
2040 $self->_SWITCH_refkind($v, {
2042 if ($self->{array_datatypes}) { # array datatype
2043 push @all_bind, $self->_bindtype($k, $v);
2045 else { # literal SQL with bind
2046 my ($sql, @bind) = @$v;
2047 $self->_assert_bindval_matches_bindtype(@bind);
2048 push @all_bind, @bind;
2051 ARRAYREFREF => sub { # literal SQL with bind
2052 my ($sql, @bind) = @${$v};
2053 $self->_assert_bindval_matches_bindtype(@bind);
2054 push @all_bind, @bind;
2056 SCALARREF => sub { # literal SQL without bind
2058 SCALAR_or_UNDEF => sub {
2059 push @all_bind, $self->_bindtype($k, $v);
2070 my(@sql, @sqlq, @sqlv);
2074 if ($ref eq 'HASH') {
2075 for my $k (sort keys %$_) {
2078 my $label = $self->_quote($k);
2079 if ($r eq 'ARRAY') {
2080 # literal SQL with bind
2081 my ($sql, @bind) = @$v;
2082 $self->_assert_bindval_matches_bindtype(@bind);
2083 push @sqlq, "$label = $sql";
2085 } elsif ($r eq 'SCALAR') {
2086 # literal SQL without bind
2087 push @sqlq, "$label = $$v";
2089 push @sqlq, "$label = ?";
2090 push @sqlv, $self->_bindtype($k, $v);
2093 push @sql, $self->_sqlcase('set'), join ', ', @sqlq;
2094 } elsif ($ref eq 'ARRAY') {
2095 # unlike insert(), assume these are ONLY the column names, i.e. for SQL
2098 if ($r eq 'ARRAY') { # literal SQL with bind
2099 my ($sql, @bind) = @$v;
2100 $self->_assert_bindval_matches_bindtype(@bind);
2103 } elsif ($r eq 'SCALAR') { # literal SQL without bind
2104 # embedded literal SQL
2111 push @sql, '(' . join(', ', @sqlq) . ')';
2112 } elsif ($ref eq 'SCALAR') {
2116 # strings get case twiddled
2117 push @sql, $self->_sqlcase($_);
2121 my $sql = join ' ', @sql;
2123 # this is pretty tricky
2124 # if ask for an array, return ($stmt, @bind)
2125 # otherwise, s/?/shift @sqlv/ to put it inline
2127 return ($sql, @sqlv);
2129 1 while $sql =~ s/\?/my $d = shift(@sqlv);
2130 ref $d ? $d->[1] : $d/e;
2139 # This allows us to check for a local, then _form, attr
2141 my($name) = $AUTOLOAD =~ /.*::(.+)/;
2142 return $self->generate($name, @_);
2153 SQL::Abstract - Generate SQL from Perl data structures
2159 my $sql = SQL::Abstract->new;
2161 my($stmt, @bind) = $sql->select($source, \@fields, \%where, $order);
2163 my($stmt, @bind) = $sql->insert($table, \%fieldvals || \@values);
2165 my($stmt, @bind) = $sql->update($table, \%fieldvals, \%where);
2167 my($stmt, @bind) = $sql->delete($table, \%where);
2169 # Then, use these in your DBI statements
2170 my $sth = $dbh->prepare($stmt);
2171 $sth->execute(@bind);
2173 # Just generate the WHERE clause
2174 my($stmt, @bind) = $sql->where(\%where, $order);
2176 # Return values in the same order, for hashed queries
2177 # See PERFORMANCE section for more details
2178 my @bind = $sql->values(\%fieldvals);
2182 This module was inspired by the excellent L<DBIx::Abstract>.
2183 However, in using that module I found that what I really wanted
2184 to do was generate SQL, but still retain complete control over my
2185 statement handles and use the DBI interface. So, I set out to
2186 create an abstract SQL generation module.
2188 While based on the concepts used by L<DBIx::Abstract>, there are
2189 several important differences, especially when it comes to WHERE
2190 clauses. I have modified the concepts used to make the SQL easier
2191 to generate from Perl data structures and, IMO, more intuitive.
2192 The underlying idea is for this module to do what you mean, based
2193 on the data structures you provide it. The big advantage is that
2194 you don't have to modify your code every time your data changes,
2195 as this module figures it out.
2197 To begin with, an SQL INSERT is as easy as just specifying a hash
2198 of C<key=value> pairs:
2201 name => 'Jimbo Bobson',
2202 phone => '123-456-7890',
2203 address => '42 Sister Lane',
2204 city => 'St. Louis',
2205 state => 'Louisiana',
2208 The SQL can then be generated with this:
2210 my($stmt, @bind) = $sql->insert('people', \%data);
2212 Which would give you something like this:
2214 $stmt = "INSERT INTO people
2215 (address, city, name, phone, state)
2216 VALUES (?, ?, ?, ?, ?)";
2217 @bind = ('42 Sister Lane', 'St. Louis', 'Jimbo Bobson',
2218 '123-456-7890', 'Louisiana');
2220 These are then used directly in your DBI code:
2222 my $sth = $dbh->prepare($stmt);
2223 $sth->execute(@bind);
2225 =head2 Inserting and Updating Arrays
2227 If your database has array types (like for example Postgres),
2228 activate the special option C<< array_datatypes => 1 >>
2229 when creating the C<SQL::Abstract> object.
2230 Then you may use an arrayref to insert and update database array types:
2232 my $sql = SQL::Abstract->new(array_datatypes => 1);
2234 planets => [qw/Mercury Venus Earth Mars/]
2237 my($stmt, @bind) = $sql->insert('solar_system', \%data);
2241 $stmt = "INSERT INTO solar_system (planets) VALUES (?)"
2243 @bind = (['Mercury', 'Venus', 'Earth', 'Mars']);
2246 =head2 Inserting and Updating SQL
2248 In order to apply SQL functions to elements of your C<%data> you may
2249 specify a reference to an arrayref for the given hash value. For example,
2250 if you need to execute the Oracle C<to_date> function on a value, you can
2251 say something like this:
2255 date_entered => \[ "to_date(?,'MM/DD/YYYY')", "03/02/2003" ],
2258 The first value in the array is the actual SQL. Any other values are
2259 optional and would be included in the bind values array. This gives
2262 my($stmt, @bind) = $sql->insert('people', \%data);
2264 $stmt = "INSERT INTO people (name, date_entered)
2265 VALUES (?, to_date(?,'MM/DD/YYYY'))";
2266 @bind = ('Bill', '03/02/2003');
2268 An UPDATE is just as easy, all you change is the name of the function:
2270 my($stmt, @bind) = $sql->update('people', \%data);
2272 Notice that your C<%data> isn't touched; the module will generate
2273 the appropriately quirky SQL for you automatically. Usually you'll
2274 want to specify a WHERE clause for your UPDATE, though, which is
2275 where handling C<%where> hashes comes in handy...
2277 =head2 Complex where statements
2279 This module can generate pretty complicated WHERE statements
2280 easily. For example, simple C<key=value> pairs are taken to mean
2281 equality, and if you want to see if a field is within a set
2282 of values, you can use an arrayref. Let's say we wanted to
2283 SELECT some data based on this criteria:
2286 requestor => 'inna',
2287 worker => ['nwiger', 'rcwe', 'sfz'],
2288 status => { '!=', 'completed' }
2291 my($stmt, @bind) = $sql->select('tickets', '*', \%where);
2293 The above would give you something like this:
2295 $stmt = "SELECT * FROM tickets WHERE
2296 ( requestor = ? ) AND ( status != ? )
2297 AND ( worker = ? OR worker = ? OR worker = ? )";
2298 @bind = ('inna', 'completed', 'nwiger', 'rcwe', 'sfz');
2300 Which you could then use in DBI code like so:
2302 my $sth = $dbh->prepare($stmt);
2303 $sth->execute(@bind);
2309 The methods are simple. There's one for every major SQL operation,
2310 and a constructor you use first. The arguments are specified in a
2311 similar order for each method (table, then fields, then a where
2312 clause) to try and simplify things.
2314 =head2 new(option => 'value')
2316 The C<new()> function takes a list of options and values, and returns
2317 a new B<SQL::Abstract> object which can then be used to generate SQL
2318 through the methods below. The options accepted are:
2324 If set to 'lower', then SQL will be generated in all lowercase. By
2325 default SQL is generated in "textbook" case meaning something like:
2327 SELECT a_field FROM a_table WHERE some_field LIKE '%someval%'
2329 Any setting other than 'lower' is ignored.
2333 This determines what the default comparison operator is. By default
2334 it is C<=>, meaning that a hash like this:
2336 %where = (name => 'nwiger', email => 'nate@wiger.org');
2338 Will generate SQL like this:
2340 WHERE name = 'nwiger' AND email = 'nate@wiger.org'
2342 However, you may want loose comparisons by default, so if you set
2343 C<cmp> to C<like> you would get SQL such as:
2345 WHERE name like 'nwiger' AND email like 'nate@wiger.org'
2347 You can also override the comparison on an individual basis - see
2348 the huge section on L</"WHERE CLAUSES"> at the bottom.
2350 =item sqltrue, sqlfalse
2352 Expressions for inserting boolean values within SQL statements.
2353 By default these are C<1=1> and C<1=0>. They are used
2354 by the special operators C<-in> and C<-not_in> for generating
2355 correct SQL even when the argument is an empty array (see below).
2359 This determines the default logical operator for multiple WHERE
2360 statements in arrays or hashes. If absent, the default logic is "or"
2361 for arrays, and "and" for hashes. This means that a WHERE
2365 event_date => {'>=', '2/13/99'},
2366 event_date => {'<=', '4/24/03'},
2369 will generate SQL like this:
2371 WHERE event_date >= '2/13/99' OR event_date <= '4/24/03'
2373 This is probably not what you want given this query, though (look
2374 at the dates). To change the "OR" to an "AND", simply specify:
2376 my $sql = SQL::Abstract->new(logic => 'and');
2378 Which will change the above C<WHERE> to:
2380 WHERE event_date >= '2/13/99' AND event_date <= '4/24/03'
2382 The logic can also be changed locally by inserting
2383 a modifier in front of an arrayref:
2385 @where = (-and => [event_date => {'>=', '2/13/99'},
2386 event_date => {'<=', '4/24/03'} ]);
2388 See the L</"WHERE CLAUSES"> section for explanations.
2392 This will automatically convert comparisons using the specified SQL
2393 function for both column and value. This is mostly used with an argument
2394 of C<upper> or C<lower>, so that the SQL will have the effect of
2395 case-insensitive "searches". For example, this:
2397 $sql = SQL::Abstract->new(convert => 'upper');
2398 %where = (keywords => 'MaKe iT CAse inSeNSItive');
2400 Will turn out the following SQL:
2402 WHERE upper(keywords) like upper('MaKe iT CAse inSeNSItive')
2404 The conversion can be C<upper()>, C<lower()>, or any other SQL function
2405 that can be applied symmetrically to fields (actually B<SQL::Abstract> does
2406 not validate this option; it will just pass through what you specify verbatim).
2410 This is a kludge because many databases suck. For example, you can't
2411 just bind values using DBI's C<execute()> for Oracle C<CLOB> or C<BLOB> fields.
2412 Instead, you have to use C<bind_param()>:
2414 $sth->bind_param(1, 'reg data');
2415 $sth->bind_param(2, $lots, {ora_type => ORA_CLOB});
2417 The problem is, B<SQL::Abstract> will normally just return a C<@bind> array,
2418 which loses track of which field each slot refers to. Fear not.
2420 If you specify C<bindtype> in new, you can determine how C<@bind> is returned.
2421 Currently, you can specify either C<normal> (default) or C<columns>. If you
2422 specify C<columns>, you will get an array that looks like this:
2424 my $sql = SQL::Abstract->new(bindtype => 'columns');
2425 my($stmt, @bind) = $sql->insert(...);
2428 [ 'column1', 'value1' ],
2429 [ 'column2', 'value2' ],
2430 [ 'column3', 'value3' ],
2433 You can then iterate through this manually, using DBI's C<bind_param()>.
2435 $sth->prepare($stmt);
2438 my($col, $data) = @$_;
2439 if ($col eq 'details' || $col eq 'comments') {
2440 $sth->bind_param($i, $data, {ora_type => ORA_CLOB});
2441 } elsif ($col eq 'image') {
2442 $sth->bind_param($i, $data, {ora_type => ORA_BLOB});
2444 $sth->bind_param($i, $data);
2448 $sth->execute; # execute without @bind now
2450 Now, why would you still use B<SQL::Abstract> if you have to do this crap?
2451 Basically, the advantage is still that you don't have to care which fields
2452 are or are not included. You could wrap that above C<for> loop in a simple
2453 sub called C<bind_fields()> or something and reuse it repeatedly. You still
2454 get a layer of abstraction over manual SQL specification.
2456 Note that if you set L</bindtype> to C<columns>, the C<\[ $sql, @bind ]>
2457 construct (see L</Literal SQL with placeholders and bind values (subqueries)>)
2458 will expect the bind values in this format.
2462 This is the character that a table or column name will be quoted
2463 with. By default this is an empty string, but you could set it to
2464 the character C<`>, to generate SQL like this:
2466 SELECT `a_field` FROM `a_table` WHERE `some_field` LIKE '%someval%'
2468 Alternatively, you can supply an array ref of two items, the first being the left
2469 hand quote character, and the second the right hand quote character. For
2470 example, you could supply C<['[',']']> for SQL Server 2000 compliant quotes
2471 that generates SQL like this:
2473 SELECT [a_field] FROM [a_table] WHERE [some_field] LIKE '%someval%'
2475 Quoting is useful if you have tables or columns names that are reserved
2476 words in your database's SQL dialect.
2480 This is the character that will be used to escape L</quote_char>s appearing
2481 in an identifier before it has been quoted.
2483 The parameter default in case of a single L</quote_char> character is the quote
2486 When opening-closing-style quoting is used (L</quote_char> is an arrayref)
2487 this parameter defaults to the B<closing (right)> L</quote_char>. Occurrences
2488 of the B<opening (left)> L</quote_char> within the identifier are currently left
2489 untouched. The default for opening-closing-style quotes may change in future
2490 versions, thus you are B<strongly encouraged> to specify the escape character
2495 This is the character that separates a table and column name. It is
2496 necessary to specify this when the C<quote_char> option is selected,
2497 so that tables and column names can be individually quoted like this:
2499 SELECT `table`.`one_field` FROM `table` WHERE `table`.`other_field` = 1
2501 =item injection_guard
2503 A regular expression C<qr/.../> that is applied to any C<-function> and unquoted
2504 column name specified in a query structure. This is a safety mechanism to avoid
2505 injection attacks when mishandling user input e.g.:
2507 my %condition_as_column_value_pairs = get_values_from_user();
2508 $sqla->select( ... , \%condition_as_column_value_pairs );
2510 If the expression matches an exception is thrown. Note that literal SQL
2511 supplied via C<\'...'> or C<\['...']> is B<not> checked in any way.
2513 Defaults to checking for C<;> and the C<GO> keyword (TransactSQL)
2515 =item array_datatypes
2517 When this option is true, arrayrefs in INSERT or UPDATE are
2518 interpreted as array datatypes and are passed directly
2520 When this option is false, arrayrefs are interpreted
2521 as literal SQL, just like refs to arrayrefs
2522 (but this behavior is for backwards compatibility; when writing
2523 new queries, use the "reference to arrayref" syntax
2529 Takes a reference to a list of "special operators"
2530 to extend the syntax understood by L<SQL::Abstract>.
2531 See section L</"SPECIAL OPERATORS"> for details.
2535 Takes a reference to a list of "unary operators"
2536 to extend the syntax understood by L<SQL::Abstract>.
2537 See section L</"UNARY OPERATORS"> for details.
2543 =head2 insert($table, \@values || \%fieldvals, \%options)
2545 This is the simplest function. You simply give it a table name
2546 and either an arrayref of values or hashref of field/value pairs.
2547 It returns an SQL INSERT statement and a list of bind values.
2548 See the sections on L</"Inserting and Updating Arrays"> and
2549 L</"Inserting and Updating SQL"> for information on how to insert
2550 with those data types.
2552 The optional C<\%options> hash reference may contain additional
2553 options to generate the insert SQL. Currently supported options
2560 Takes either a scalar of raw SQL fields, or an array reference of
2561 field names, and adds on an SQL C<RETURNING> statement at the end.
2562 This allows you to return data generated by the insert statement
2563 (such as row IDs) without performing another C<SELECT> statement.
2564 Note, however, this is not part of the SQL standard and may not
2565 be supported by all database engines.
2569 =head2 update($table, \%fieldvals, \%where, \%options)
2571 This takes a table, hashref of field/value pairs, and an optional
2572 hashref L<WHERE clause|/WHERE CLAUSES>. It returns an SQL UPDATE function and a list
2574 See the sections on L</"Inserting and Updating Arrays"> and
2575 L</"Inserting and Updating SQL"> for information on how to insert
2576 with those data types.
2578 The optional C<\%options> hash reference may contain additional
2579 options to generate the update SQL. Currently supported options
2586 See the C<returning> option to
2587 L<insert|/insert($table, \@values || \%fieldvals, \%options)>.
2591 =head2 select($source, $fields, $where, $order)
2593 This returns a SQL SELECT statement and associated list of bind values, as
2594 specified by the arguments:
2600 Specification of the 'FROM' part of the statement.
2601 The argument can be either a plain scalar (interpreted as a table
2602 name, will be quoted), or an arrayref (interpreted as a list
2603 of table names, joined by commas, quoted), or a scalarref
2604 (literal SQL, not quoted).
2608 Specification of the list of fields to retrieve from
2610 The argument can be either an arrayref (interpreted as a list
2611 of field names, will be joined by commas and quoted), or a
2612 plain scalar (literal SQL, not quoted).
2613 Please observe that this API is not as flexible as that of
2614 the first argument C<$source>, for backwards compatibility reasons.
2618 Optional argument to specify the WHERE part of the query.
2619 The argument is most often a hashref, but can also be
2620 an arrayref or plain scalar --
2621 see section L<WHERE clause|/"WHERE CLAUSES"> for details.
2625 Optional argument to specify the ORDER BY part of the query.
2626 The argument can be a scalar, a hashref or an arrayref
2627 -- see section L<ORDER BY clause|/"ORDER BY CLAUSES">
2633 =head2 delete($table, \%where, \%options)
2635 This takes a table name and optional hashref L<WHERE clause|/WHERE CLAUSES>.
2636 It returns an SQL DELETE statement and list of bind values.
2638 The optional C<\%options> hash reference may contain additional
2639 options to generate the delete SQL. Currently supported options
2646 See the C<returning> option to
2647 L<insert|/insert($table, \@values || \%fieldvals, \%options)>.
2651 =head2 where(\%where, $order)
2653 This is used to generate just the WHERE clause. For example,
2654 if you have an arbitrary data structure and know what the
2655 rest of your SQL is going to look like, but want an easy way
2656 to produce a WHERE clause, use this. It returns an SQL WHERE
2657 clause and list of bind values.
2660 =head2 values(\%data)
2662 This just returns the values from the hash C<%data>, in the same
2663 order that would be returned from any of the other above queries.
2664 Using this allows you to markedly speed up your queries if you
2665 are affecting lots of rows. See below under the L</"PERFORMANCE"> section.
2667 =head2 generate($any, 'number', $of, \@data, $struct, \%types)
2669 Warning: This is an experimental method and subject to change.
2671 This returns arbitrarily generated SQL. It's a really basic shortcut.
2672 It will return two different things, depending on return context:
2674 my($stmt, @bind) = $sql->generate('create table', \$table, \@fields);
2675 my $stmt_and_val = $sql->generate('create table', \$table, \@fields);
2677 These would return the following:
2679 # First calling form
2680 $stmt = "CREATE TABLE test (?, ?)";
2681 @bind = (field1, field2);
2683 # Second calling form
2684 $stmt_and_val = "CREATE TABLE test (field1, field2)";
2686 Depending on what you're trying to do, it's up to you to choose the correct
2687 format. In this example, the second form is what you would want.
2691 $sql->generate('alter session', { nls_date_format => 'MM/YY' });
2695 ALTER SESSION SET nls_date_format = 'MM/YY'
2697 You get the idea. Strings get their case twiddled, but everything
2698 else remains verbatim.
2700 =head1 EXPORTABLE FUNCTIONS
2702 =head2 is_plain_value
2704 Determines if the supplied argument is a plain value as understood by this
2709 =item * The value is C<undef>
2711 =item * The value is a non-reference
2713 =item * The value is an object with stringification overloading
2715 =item * The value is of the form C<< { -value => $anything } >>
2719 On failure returns C<undef>, on success returns a B<scalar> reference
2720 to the original supplied argument.
2726 The stringification overloading detection is rather advanced: it takes
2727 into consideration not only the presence of a C<""> overload, but if that
2728 fails also checks for enabled
2729 L<autogenerated versions of C<"">|overload/Magic Autogeneration>, based
2730 on either C<0+> or C<bool>.
2732 Unfortunately testing in the field indicates that this
2733 detection B<< may tickle a latent bug in perl versions before 5.018 >>,
2734 but only when very large numbers of stringifying objects are involved.
2735 At the time of writing ( Sep 2014 ) there is no clear explanation of
2736 the direct cause, nor is there a manageably small test case that reliably
2737 reproduces the problem.
2739 If you encounter any of the following exceptions in B<random places within
2740 your application stack> - this module may be to blame:
2742 Operation "ne": no method found,
2743 left argument in overloaded package <something>,
2744 right argument in overloaded package <something>
2748 Stub found while resolving method "???" overloading """" in package <something>
2750 If you fall victim to the above - please attempt to reduce the problem
2751 to something that could be sent to the L<SQL::Abstract developers
2752 |DBIx::Class/GETTING HELP/SUPPORT>
2753 (either publicly or privately). As a workaround in the meantime you can
2754 set C<$ENV{SQLA_ISVALUE_IGNORE_AUTOGENERATED_STRINGIFICATION}> to a true
2755 value, which will most likely eliminate your problem (at the expense of
2756 not being able to properly detect exotic forms of stringification).
2758 This notice and environment variable will be removed in a future version,
2759 as soon as the underlying problem is found and a reliable workaround is
2764 =head2 is_literal_value
2766 Determines if the supplied argument is a literal value as understood by this
2771 =item * C<\$sql_string>
2773 =item * C<\[ $sql_string, @bind_values ]>
2777 On failure returns C<undef>, on success returns an B<array> reference
2778 containing the unpacked version of the supplied literal SQL and bind values.
2780 =head1 WHERE CLAUSES
2784 This module uses a variation on the idea from L<DBIx::Abstract>. It
2785 is B<NOT>, repeat I<not> 100% compatible. B<The main logic of this
2786 module is that things in arrays are OR'ed, and things in hashes
2789 The easiest way to explain is to show lots of examples. After
2790 each C<%where> hash shown, it is assumed you used:
2792 my($stmt, @bind) = $sql->where(\%where);
2794 However, note that the C<%where> hash can be used directly in any
2795 of the other functions as well, as described above.
2797 =head2 Key-value pairs
2799 So, let's get started. To begin, a simple hash:
2803 status => 'completed'
2806 Is converted to SQL C<key = val> statements:
2808 $stmt = "WHERE user = ? AND status = ?";
2809 @bind = ('nwiger', 'completed');
2811 One common thing I end up doing is having a list of values that
2812 a field can be in. To do this, simply specify a list inside of
2817 status => ['assigned', 'in-progress', 'pending'];
2820 This simple code will create the following:
2822 $stmt = "WHERE user = ? AND ( status = ? OR status = ? OR status = ? )";
2823 @bind = ('nwiger', 'assigned', 'in-progress', 'pending');
2825 A field associated to an empty arrayref will be considered a
2826 logical false and will generate 0=1.
2828 =head2 Tests for NULL values
2830 If the value part is C<undef> then this is converted to SQL <IS NULL>
2839 $stmt = "WHERE user = ? AND status IS NULL";
2842 To test if a column IS NOT NULL:
2846 status => { '!=', undef },
2849 =head2 Specific comparison operators
2851 If you want to specify a different type of operator for your comparison,
2852 you can use a hashref for a given column:
2856 status => { '!=', 'completed' }
2859 Which would generate:
2861 $stmt = "WHERE user = ? AND status != ?";
2862 @bind = ('nwiger', 'completed');
2864 To test against multiple values, just enclose the values in an arrayref:
2866 status => { '=', ['assigned', 'in-progress', 'pending'] };
2868 Which would give you:
2870 "WHERE status = ? OR status = ? OR status = ?"
2873 The hashref can also contain multiple pairs, in which case it is expanded
2874 into an C<AND> of its elements:
2878 status => { '!=', 'completed', -not_like => 'pending%' }
2881 # Or more dynamically, like from a form
2882 $where{user} = 'nwiger';
2883 $where{status}{'!='} = 'completed';
2884 $where{status}{'-not_like'} = 'pending%';
2886 # Both generate this
2887 $stmt = "WHERE user = ? AND status != ? AND status NOT LIKE ?";
2888 @bind = ('nwiger', 'completed', 'pending%');
2891 To get an OR instead, you can combine it with the arrayref idea:
2895 priority => [ { '=', 2 }, { '>', 5 } ]
2898 Which would generate:
2900 $stmt = "WHERE ( priority = ? OR priority > ? ) AND user = ?";
2901 @bind = ('2', '5', 'nwiger');
2903 If you want to include literal SQL (with or without bind values), just use a
2904 scalar reference or reference to an arrayref as the value:
2907 date_entered => { '>' => \["to_date(?, 'MM/DD/YYYY')", "11/26/2008"] },
2908 date_expires => { '<' => \"now()" }
2911 Which would generate:
2913 $stmt = "WHERE date_entered > to_date(?, 'MM/DD/YYYY') AND date_expires < now()";
2914 @bind = ('11/26/2008');
2917 =head2 Logic and nesting operators
2919 In the example above,
2920 there is a subtle trap if you want to say something like
2921 this (notice the C<AND>):
2923 WHERE priority != ? AND priority != ?
2925 Because, in Perl you I<can't> do this:
2927 priority => { '!=' => 2, '!=' => 1 }
2929 As the second C<!=> key will obliterate the first. The solution
2930 is to use the special C<-modifier> form inside an arrayref:
2932 priority => [ -and => {'!=', 2},
2936 Normally, these would be joined by C<OR>, but the modifier tells it
2937 to use C<AND> instead. (Hint: You can use this in conjunction with the
2938 C<logic> option to C<new()> in order to change the way your queries
2939 work by default.) B<Important:> Note that the C<-modifier> goes
2940 B<INSIDE> the arrayref, as an extra first element. This will
2941 B<NOT> do what you think it might:
2943 priority => -and => [{'!=', 2}, {'!=', 1}] # WRONG!
2945 Here is a quick list of equivalencies, since there is some overlap:
2948 status => {'!=', 'completed', 'not like', 'pending%' }
2949 status => [ -and => {'!=', 'completed'}, {'not like', 'pending%'}]
2952 status => {'=', ['assigned', 'in-progress']}
2953 status => [ -or => {'=', 'assigned'}, {'=', 'in-progress'}]
2954 status => [ {'=', 'assigned'}, {'=', 'in-progress'} ]
2958 =head2 Special operators: IN, BETWEEN, etc.
2960 You can also use the hashref format to compare a list of fields using the
2961 C<IN> comparison operator, by specifying the list as an arrayref:
2964 status => 'completed',
2965 reportid => { -in => [567, 2335, 2] }
2968 Which would generate:
2970 $stmt = "WHERE status = ? AND reportid IN (?,?,?)";
2971 @bind = ('completed', '567', '2335', '2');
2973 The reverse operator C<-not_in> generates SQL C<NOT IN> and is used in
2976 If the argument to C<-in> is an empty array, 'sqlfalse' is generated
2977 (by default: C<1=0>). Similarly, C<< -not_in => [] >> generates
2978 'sqltrue' (by default: C<1=1>).
2980 In addition to the array you can supply a chunk of literal sql or
2981 literal sql with bind:
2984 customer => { -in => \[
2985 'SELECT cust_id FROM cust WHERE balance > ?',
2988 status => { -in => \'SELECT status_codes FROM states' },
2994 customer IN ( SELECT cust_id FROM cust WHERE balance > ? )
2995 AND status IN ( SELECT status_codes FROM states )
2999 Finally, if the argument to C<-in> is not a reference, it will be
3000 treated as a single-element array.
3002 Another pair of operators is C<-between> and C<-not_between>,
3003 used with an arrayref of two values:
3007 completion_date => {
3008 -not_between => ['2002-10-01', '2003-02-06']
3014 WHERE user = ? AND completion_date NOT BETWEEN ( ? AND ? )
3016 Just like with C<-in> all plausible combinations of literal SQL
3020 start0 => { -between => [ 1, 2 ] },
3021 start1 => { -between => \["? AND ?", 1, 2] },
3022 start2 => { -between => \"lower(x) AND upper(y)" },
3023 start3 => { -between => [
3025 \["upper(?)", 'stuff' ],
3032 ( start0 BETWEEN ? AND ? )
3033 AND ( start1 BETWEEN ? AND ? )
3034 AND ( start2 BETWEEN lower(x) AND upper(y) )
3035 AND ( start3 BETWEEN lower(x) AND upper(?) )
3037 @bind = (1, 2, 1, 2, 'stuff');
3040 These are the two builtin "special operators"; but the
3041 list can be expanded: see section L</"SPECIAL OPERATORS"> below.
3043 =head2 Unary operators: bool
3045 If you wish to test against boolean columns or functions within your
3046 database you can use the C<-bool> and C<-not_bool> operators. For
3047 example to test the column C<is_user> being true and the column
3048 C<is_enabled> being false you would use:-
3052 -not_bool => 'is_enabled',
3057 WHERE is_user AND NOT is_enabled
3059 If a more complex combination is required, testing more conditions,
3060 then you should use the and/or operators:-
3065 -not_bool => { two=> { -rlike => 'bar' } },
3066 -not_bool => { three => [ { '=', 2 }, { '>', 5 } ] },
3077 (NOT ( three = ? OR three > ? ))
3080 =head2 Nested conditions, -and/-or prefixes
3082 So far, we've seen how multiple conditions are joined with a top-level
3083 C<AND>. We can change this by putting the different conditions we want in
3084 hashes and then putting those hashes in an array. For example:
3089 status => { -like => ['pending%', 'dispatched'] },
3093 status => 'unassigned',
3097 This data structure would create the following:
3099 $stmt = "WHERE ( user = ? AND ( status LIKE ? OR status LIKE ? ) )
3100 OR ( user = ? AND status = ? ) )";
3101 @bind = ('nwiger', 'pending', 'dispatched', 'robot', 'unassigned');
3104 Clauses in hashrefs or arrayrefs can be prefixed with an C<-and> or C<-or>
3105 to change the logic inside:
3111 -and => [ workhrs => {'>', 20}, geo => 'ASIA' ],
3112 -or => { workhrs => {'<', 50}, geo => 'EURO' },
3119 $stmt = "WHERE ( user = ?
3120 AND ( ( workhrs > ? AND geo = ? )
3121 OR ( workhrs < ? OR geo = ? ) ) )";
3122 @bind = ('nwiger', '20', 'ASIA', '50', 'EURO');
3124 =head3 Algebraic inconsistency, for historical reasons
3126 C<Important note>: when connecting several conditions, the C<-and->|C<-or>
3127 operator goes C<outside> of the nested structure; whereas when connecting
3128 several constraints on one column, the C<-and> operator goes
3129 C<inside> the arrayref. Here is an example combining both features:
3132 -and => [a => 1, b => 2],
3133 -or => [c => 3, d => 4],
3134 e => [-and => {-like => 'foo%'}, {-like => '%bar'} ]
3139 WHERE ( ( ( a = ? AND b = ? )
3140 OR ( c = ? OR d = ? )
3141 OR ( e LIKE ? AND e LIKE ? ) ) )
3143 This difference in syntax is unfortunate but must be preserved for
3144 historical reasons. So be careful: the two examples below would
3145 seem algebraically equivalent, but they are not
3148 { -like => 'foo%' },
3149 { -like => '%bar' },
3151 # yields: WHERE ( ( col LIKE ? AND col LIKE ? ) )
3154 { col => { -like => 'foo%' } },
3155 { col => { -like => '%bar' } },
3157 # yields: WHERE ( ( col LIKE ? OR col LIKE ? ) )
3160 =head2 Literal SQL and value type operators
3162 The basic premise of SQL::Abstract is that in WHERE specifications the "left
3163 side" is a column name and the "right side" is a value (normally rendered as
3164 a placeholder). This holds true for both hashrefs and arrayref pairs as you
3165 see in the L</WHERE CLAUSES> examples above. Sometimes it is necessary to
3166 alter this behavior. There are several ways of doing so.
3170 This is a virtual operator that signals the string to its right side is an
3171 identifier (a column name) and not a value. For example to compare two
3172 columns you would write:
3175 priority => { '<', 2 },
3176 requestor => { -ident => 'submitter' },
3181 $stmt = "WHERE priority < ? AND requestor = submitter";
3184 If you are maintaining legacy code you may see a different construct as
3185 described in L</Deprecated usage of Literal SQL>, please use C<-ident> in new
3190 This is a virtual operator that signals that the construct to its right side
3191 is a value to be passed to DBI. This is for example necessary when you want
3192 to write a where clause against an array (for RDBMS that support such
3193 datatypes). For example:
3196 array => { -value => [1, 2, 3] }
3201 $stmt = 'WHERE array = ?';
3202 @bind = ([1, 2, 3]);
3204 Note that if you were to simply say:
3210 the result would probably not be what you wanted:
3212 $stmt = 'WHERE array = ? OR array = ? OR array = ?';
3217 Finally, sometimes only literal SQL will do. To include a random snippet
3218 of SQL verbatim, you specify it as a scalar reference. Consider this only
3219 as a last resort. Usually there is a better way. For example:
3222 priority => { '<', 2 },
3223 requestor => { -in => \'(SELECT name FROM hitmen)' },
3228 $stmt = "WHERE priority < ? AND requestor IN (SELECT name FROM hitmen)"
3231 Note that in this example, you only get one bind parameter back, since
3232 the verbatim SQL is passed as part of the statement.
3236 Never use untrusted input as a literal SQL argument - this is a massive
3237 security risk (there is no way to check literal snippets for SQL
3238 injections and other nastyness). If you need to deal with untrusted input
3239 use literal SQL with placeholders as described next.
3241 =head3 Literal SQL with placeholders and bind values (subqueries)
3243 If the literal SQL to be inserted has placeholders and bind values,
3244 use a reference to an arrayref (yes this is a double reference --
3245 not so common, but perfectly legal Perl). For example, to find a date
3246 in Postgres you can use something like this:
3249 date_column => \[ "= date '2008-09-30' - ?::integer", 10 ]
3254 $stmt = "WHERE ( date_column = date '2008-09-30' - ?::integer )"
3257 Note that you must pass the bind values in the same format as they are returned
3258 by L<where|/where(\%where, $order)>. This means that if you set L</bindtype>
3259 to C<columns>, you must provide the bind values in the
3260 C<< [ column_meta => value ] >> format, where C<column_meta> is an opaque
3261 scalar value; most commonly the column name, but you can use any scalar value
3262 (including references and blessed references), L<SQL::Abstract> will simply
3263 pass it through intact. So if C<bindtype> is set to C<columns> the above
3264 example will look like:
3267 date_column => \[ "= date '2008-09-30' - ?::integer", [ {} => 10 ] ]
3270 Literal SQL is especially useful for nesting parenthesized clauses in the
3271 main SQL query. Here is a first example:
3273 my ($sub_stmt, @sub_bind) = ("SELECT c1 FROM t1 WHERE c2 < ? AND c3 LIKE ?",
3277 bar => \["IN ($sub_stmt)" => @sub_bind],
3282 $stmt = "WHERE (foo = ? AND bar IN (SELECT c1 FROM t1
3283 WHERE c2 < ? AND c3 LIKE ?))";
3284 @bind = (1234, 100, "foo%");
3286 Other subquery operators, like for example C<"E<gt> ALL"> or C<"NOT IN">,
3287 are expressed in the same way. Of course the C<$sub_stmt> and
3288 its associated bind values can be generated through a former call
3291 my ($sub_stmt, @sub_bind)
3292 = $sql->select("t1", "c1", {c2 => {"<" => 100},
3293 c3 => {-like => "foo%"}});
3296 bar => \["> ALL ($sub_stmt)" => @sub_bind],
3299 In the examples above, the subquery was used as an operator on a column;
3300 but the same principle also applies for a clause within the main C<%where>
3301 hash, like an EXISTS subquery:
3303 my ($sub_stmt, @sub_bind)
3304 = $sql->select("t1", "*", {c1 => 1, c2 => \"> t0.c0"});
3305 my %where = ( -and => [
3307 \["EXISTS ($sub_stmt)" => @sub_bind],
3312 $stmt = "WHERE (foo = ? AND EXISTS (SELECT * FROM t1
3313 WHERE c1 = ? AND c2 > t0.c0))";
3317 Observe that the condition on C<c2> in the subquery refers to
3318 column C<t0.c0> of the main query: this is I<not> a bind
3319 value, so we have to express it through a scalar ref.
3320 Writing C<< c2 => {">" => "t0.c0"} >> would have generated
3321 C<< c2 > ? >> with bind value C<"t0.c0"> ... not exactly
3322 what we wanted here.
3324 Finally, here is an example where a subquery is used
3325 for expressing unary negation:
3327 my ($sub_stmt, @sub_bind)
3328 = $sql->where({age => [{"<" => 10}, {">" => 20}]});
3329 $sub_stmt =~ s/^ where //i; # don't want "WHERE" in the subclause
3331 lname => {like => '%son%'},
3332 \["NOT ($sub_stmt)" => @sub_bind],
3337 $stmt = "lname LIKE ? AND NOT ( age < ? OR age > ? )"
3338 @bind = ('%son%', 10, 20)
3340 =head3 Deprecated usage of Literal SQL
3342 Below are some examples of archaic use of literal SQL. It is shown only as
3343 reference for those who deal with legacy code. Each example has a much
3344 better, cleaner and safer alternative that users should opt for in new code.
3350 my %where = ( requestor => \'IS NOT NULL' )
3352 $stmt = "WHERE requestor IS NOT NULL"
3354 This used to be the way of generating NULL comparisons, before the handling
3355 of C<undef> got formalized. For new code please use the superior syntax as
3356 described in L</Tests for NULL values>.
3360 my %where = ( requestor => \'= submitter' )
3362 $stmt = "WHERE requestor = submitter"
3364 This used to be the only way to compare columns. Use the superior L</-ident>
3365 method for all new code. For example an identifier declared in such a way
3366 will be properly quoted if L</quote_char> is properly set, while the legacy
3367 form will remain as supplied.
3371 my %where = ( is_ready => \"", completed => { '>', '2012-12-21' } )
3373 $stmt = "WHERE completed > ? AND is_ready"
3374 @bind = ('2012-12-21')
3376 Using an empty string literal used to be the only way to express a boolean.
3377 For all new code please use the much more readable
3378 L<-bool|/Unary operators: bool> operator.
3384 These pages could go on for a while, since the nesting of the data
3385 structures this module can handle are pretty much unlimited (the
3386 module implements the C<WHERE> expansion as a recursive function
3387 internally). Your best bet is to "play around" with the module a
3388 little to see how the data structures behave, and choose the best
3389 format for your data based on that.
3391 And of course, all the values above will probably be replaced with
3392 variables gotten from forms or the command line. After all, if you
3393 knew everything ahead of time, you wouldn't have to worry about
3394 dynamically-generating SQL and could just hardwire it into your
3397 =head1 ORDER BY CLAUSES
3399 Some functions take an order by clause. This can either be a scalar (just a
3400 column name), a hashref of C<< { -desc => 'col' } >> or C<< { -asc => 'col' }
3401 >>, a scalarref, an arrayref-ref, or an arrayref of any of the previous
3404 Given | Will Generate
3405 ---------------------------------------------------------------
3407 'colA' | ORDER BY colA
3409 [qw/colA colB/] | ORDER BY colA, colB
3411 {-asc => 'colA'} | ORDER BY colA ASC
3413 {-desc => 'colB'} | ORDER BY colB DESC
3415 ['colA', {-asc => 'colB'}] | ORDER BY colA, colB ASC
3417 { -asc => [qw/colA colB/] } | ORDER BY colA ASC, colB ASC
3419 \'colA DESC' | ORDER BY colA DESC
3421 \[ 'FUNC(colA, ?)', $x ] | ORDER BY FUNC(colA, ?)
3422 | /* ...with $x bound to ? */
3425 { -asc => 'colA' }, | colA ASC,
3426 { -desc => [qw/colB/] }, | colB DESC,
3427 { -asc => [qw/colC colD/] },| colC ASC, colD ASC,
3428 \'colE DESC', | colE DESC,
3429 \[ 'FUNC(colF, ?)', $x ], | FUNC(colF, ?)
3430 ] | /* ...with $x bound to ? */
3431 ===============================================================
3435 =head1 SPECIAL OPERATORS
3437 my $sqlmaker = SQL::Abstract->new(special_ops => [
3441 my ($self, $field, $op, $arg) = @_;
3447 handler => 'method_name',
3451 A "special operator" is a SQL syntactic clause that can be
3452 applied to a field, instead of a usual binary operator.
3455 WHERE field IN (?, ?, ?)
3456 WHERE field BETWEEN ? AND ?
3457 WHERE MATCH(field) AGAINST (?, ?)
3459 Special operators IN and BETWEEN are fairly standard and therefore
3460 are builtin within C<SQL::Abstract> (as the overridable methods
3461 C<_where_field_IN> and C<_where_field_BETWEEN>). For other operators,
3462 like the MATCH .. AGAINST example above which is specific to MySQL,
3463 you can write your own operator handlers - supply a C<special_ops>
3464 argument to the C<new> method. That argument takes an arrayref of
3465 operator definitions; each operator definition is a hashref with two
3472 the regular expression to match the operator
3476 Either a coderef or a plain scalar method name. In both cases
3477 the expected return is C<< ($sql, @bind) >>.
3479 When supplied with a method name, it is simply called on the
3480 L<SQL::Abstract> object as:
3482 $self->$method_name($field, $op, $arg)
3486 $field is the LHS of the operator
3487 $op is the part that matched the handler regex
3490 When supplied with a coderef, it is called as:
3492 $coderef->($self, $field, $op, $arg)
3497 For example, here is an implementation
3498 of the MATCH .. AGAINST syntax for MySQL
3500 my $sqlmaker = SQL::Abstract->new(special_ops => [
3502 # special op for MySql MATCH (field) AGAINST(word1, word2, ...)
3503 {regex => qr/^match$/i,
3505 my ($self, $field, $op, $arg) = @_;
3506 $arg = [$arg] if not ref $arg;
3507 my $label = $self->_quote($field);
3508 my ($placeholder) = $self->_convert('?');
3509 my $placeholders = join ", ", (($placeholder) x @$arg);
3510 my $sql = $self->_sqlcase('match') . " ($label) "
3511 . $self->_sqlcase('against') . " ($placeholders) ";
3512 my @bind = $self->_bindtype($field, @$arg);
3513 return ($sql, @bind);
3520 =head1 UNARY OPERATORS
3522 my $sqlmaker = SQL::Abstract->new(unary_ops => [
3526 my ($self, $op, $arg) = @_;
3532 handler => 'method_name',
3536 A "unary operator" is a SQL syntactic clause that can be
3537 applied to a field - the operator goes before the field
3539 You can write your own operator handlers - supply a C<unary_ops>
3540 argument to the C<new> method. That argument takes an arrayref of
3541 operator definitions; each operator definition is a hashref with two
3548 the regular expression to match the operator
3552 Either a coderef or a plain scalar method name. In both cases
3553 the expected return is C<< $sql >>.
3555 When supplied with a method name, it is simply called on the
3556 L<SQL::Abstract> object as:
3558 $self->$method_name($op, $arg)
3562 $op is the part that matched the handler regex
3563 $arg is the RHS or argument of the operator
3565 When supplied with a coderef, it is called as:
3567 $coderef->($self, $op, $arg)
3575 Thanks to some benchmarking by Mark Stosberg, it turns out that
3576 this module is many orders of magnitude faster than using C<DBIx::Abstract>.
3577 I must admit this wasn't an intentional design issue, but it's a
3578 byproduct of the fact that you get to control your C<DBI> handles
3581 To maximize performance, use a code snippet like the following:
3583 # prepare a statement handle using the first row
3584 # and then reuse it for the rest of the rows
3586 for my $href (@array_of_hashrefs) {
3587 $stmt ||= $sql->insert('table', $href);
3588 $sth ||= $dbh->prepare($stmt);
3589 $sth->execute($sql->values($href));
3592 The reason this works is because the keys in your C<$href> are sorted
3593 internally by B<SQL::Abstract>. Thus, as long as your data retains
3594 the same structure, you only have to generate the SQL the first time
3595 around. On subsequent queries, simply use the C<values> function provided
3596 by this module to return your values in the correct order.
3598 However this depends on the values having the same type - if, for
3599 example, the values of a where clause may either have values
3600 (resulting in sql of the form C<column = ?> with a single bind
3601 value), or alternatively the values might be C<undef> (resulting in
3602 sql of the form C<column IS NULL> with no bind value) then the
3603 caching technique suggested will not work.
3607 If you use my C<CGI::FormBuilder> module at all, you'll hopefully
3608 really like this part (I do, at least). Building up a complex query
3609 can be as simple as the following:
3616 use CGI::FormBuilder;
3619 my $form = CGI::FormBuilder->new(...);
3620 my $sql = SQL::Abstract->new;
3622 if ($form->submitted) {
3623 my $field = $form->field;
3624 my $id = delete $field->{id};
3625 my($stmt, @bind) = $sql->update('table', $field, {id => $id});
3628 Of course, you would still have to connect using C<DBI> to run the
3629 query, but the point is that if you make your form look like your
3630 table, the actual query script can be extremely simplistic.
3632 If you're B<REALLY> lazy (I am), check out C<HTML::QuickTable> for
3633 a fast interface to returning and formatting data. I frequently
3634 use these three modules together to write complex database query
3635 apps in under 50 lines.
3637 =head1 HOW TO CONTRIBUTE
3639 Contributions are always welcome, in all usable forms (we especially
3640 welcome documentation improvements). The delivery methods include git-
3641 or unified-diff formatted patches, GitHub pull requests, or plain bug
3642 reports either via RT or the Mailing list. Contributors are generally
3643 granted full access to the official repository after their first several
3644 patches pass successful review.
3646 This project is maintained in a git repository. The code and related tools are
3647 accessible at the following locations:
3651 =item * Official repo: L<git://git.shadowcat.co.uk/dbsrgits/SQL-Abstract.git>
3653 =item * Official gitweb: L<http://git.shadowcat.co.uk/gitweb/gitweb.cgi?p=dbsrgits/SQL-Abstract.git>
3655 =item * GitHub mirror: L<https://github.com/dbsrgits/sql-abstract>
3657 =item * Authorized committers: L<ssh://dbsrgits@git.shadowcat.co.uk/SQL-Abstract.git>
3663 Version 1.50 was a major internal refactoring of C<SQL::Abstract>.
3664 Great care has been taken to preserve the I<published> behavior
3665 documented in previous versions in the 1.* family; however,
3666 some features that were previously undocumented, or behaved
3667 differently from the documentation, had to be changed in order
3668 to clarify the semantics. Hence, client code that was relying
3669 on some dark areas of C<SQL::Abstract> v1.*
3670 B<might behave differently> in v1.50.
3672 The main changes are:
3678 support for literal SQL through the C<< \ [ $sql, @bind ] >> syntax.
3682 support for the { operator => \"..." } construct (to embed literal SQL)
3686 support for the { operator => \["...", @bind] } construct (to embed literal SQL with bind values)
3690 optional support for L<array datatypes|/"Inserting and Updating Arrays">
3694 defensive programming: check arguments
3698 fixed bug with global logic, which was previously implemented
3699 through global variables yielding side-effects. Prior versions would
3700 interpret C<< [ {cond1, cond2}, [cond3, cond4] ] >>
3701 as C<< "(cond1 AND cond2) OR (cond3 AND cond4)" >>.
3702 Now this is interpreted
3703 as C<< "(cond1 AND cond2) OR (cond3 OR cond4)" >>.
3708 fixed semantics of _bindtype on array args
3712 dropped the C<_anoncopy> of the %where tree. No longer necessary,
3713 we just avoid shifting arrays within that tree.
3717 dropped the C<_modlogic> function
3721 =head1 ACKNOWLEDGEMENTS
3723 There are a number of individuals that have really helped out with
3724 this module. Unfortunately, most of them submitted bugs via CPAN
3725 so I have no idea who they are! But the people I do know are:
3727 Ash Berlin (order_by hash term support)
3728 Matt Trout (DBIx::Class support)
3729 Mark Stosberg (benchmarking)
3730 Chas Owens (initial "IN" operator support)
3731 Philip Collins (per-field SQL functions)
3732 Eric Kolve (hashref "AND" support)
3733 Mike Fragassi (enhancements to "BETWEEN" and "LIKE")
3734 Dan Kubb (support for "quote_char" and "name_sep")
3735 Guillermo Roditi (patch to cleanup "IN" and "BETWEEN", fix and tests for _order_by)
3736 Laurent Dami (internal refactoring, extensible list of special operators, literal SQL)
3737 Norbert Buchmuller (support for literal SQL in hashpair, misc. fixes & tests)
3738 Peter Rabbitson (rewrite of SQLA::Test, misc. fixes & tests)
3739 Oliver Charles (support for "RETURNING" after "INSERT")
3745 L<DBIx::Class>, L<DBIx::Abstract>, L<CGI::FormBuilder>, L<HTML::QuickTable>.
3749 Copyright (c) 2001-2007 Nathan Wiger <nwiger@cpan.org>. All Rights Reserved.
3751 This module is actively maintained by Matt Trout <mst@shadowcatsystems.co.uk>
3753 For support, your best bet is to try the C<DBIx::Class> users mailing list.
3754 While not an official support venue, C<DBIx::Class> makes heavy use of
3755 C<SQL::Abstract>, and as such list members there are very familiar with
3756 how to create queries.
3760 This module is free software; you may copy this under the same
3761 terms as perl itself (either the GNU General Public License or
3762 the Artistic License)