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 { -op => [ $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') {
780 map $self->_expand_expr_hashpair($k, { $_ => $vv->{$_} }),
785 if (my $us = List::Util::first { $vk =~ $_->{regex} } @{$self->{user_special_ops}}) {
786 return { -op => [ $vk, { -ident => $k }, $vv ] };
788 if (ref($vv) eq 'ARRAY') {
789 my ($logic, @values) = (
790 (defined($vv->[0]) and $vv->[0] =~ /^-(and|or)$/i)
795 $vk =~ $self->{inequality_op}
796 or join(' ', split '_', $vk) =~ $self->{not_like_op}
798 if (lc($logic) eq '-or' and @values > 1) {
799 my $op = uc join ' ', split '_', $vk;
800 belch "A multi-element arrayref as an argument to the inequality op '$op' "
801 . 'is technically equivalent to an always-true 1=1 (you probably wanted '
802 . "to say ...{ \$inequality_op => [ -and => \@values ] }... instead)"
807 # try to DWIM on equality operators
808 my $op = join ' ', split '_', $vk;
810 $op =~ $self->{equality_op} ? $self->{sqlfalse}
811 : $op =~ $self->{like_op} ? belch("Supplying an empty arrayref to '@{[ uc $op]}' is deprecated") && $self->{sqlfalse}
812 : $op =~ $self->{inequality_op} ? $self->{sqltrue}
813 : $op =~ $self->{not_like_op} ? belch("Supplying an empty arrayref to '@{[ uc $op]}' is deprecated") && $self->{sqltrue}
814 : puke "operator '$op' applied on an empty array (field '$k')";
818 map $self->_expand_expr_hashpair($k => { $vk => $_ }),
826 and exists $vv->{-value}
827 and not defined $vv->{-value}
830 my $op = join ' ', split '_', $vk;
832 $op =~ /^not$/i ? 'is not' # legacy
833 : $op =~ $self->{equality_op} ? 'is'
834 : $op =~ $self->{like_op} ? belch("Supplying an undefined argument to '@{[ uc $op]}' is deprecated") && 'is'
835 : $op =~ $self->{inequality_op} ? 'is not'
836 : $op =~ $self->{not_like_op} ? belch("Supplying an undefined argument to '@{[ uc $op]}' is deprecated") && 'is not'
837 : puke "unexpected operator '$op' with undef operand";
838 return +{ -op => [ $is.' null', { -ident => $k } ] };
840 local our $Cur_Col_Meta = $k;
844 $self->_expand_expr($vv)
847 if (ref($v) eq 'ARRAY') {
848 return $self->{sqlfalse} unless @$v;
849 $self->_debug("ARRAY($k) means distribute over elements");
851 $v->[0] =~ /^-((?:and|or))$/i
852 ? ($v = [ @{$v}[1..$#$v] ], $1)
853 : ($self->{logic} || 'or')
855 return +{ "-${this_logic}" => [ map $self->_expand_expr({ $k => $_ }, $this_logic), @$v ] };
857 if (my $literal = is_literal_value($v)) {
859 belch 'Hash-pairs consisting of an empty string with a literal are deprecated, and will be removed in 2.0: use -and => [ $literal ] instead';
862 my ($sql, @bind) = @$literal;
863 if ($self->{bindtype} eq 'columns') {
865 if (!defined $_ || ref($_) ne 'ARRAY' || @$_ != 2) {
866 puke "bindtype 'columns' selected, you need to pass: [column_name => bind_value]"
870 return +{ -literal => [ $self->_quote($k).' '.$sql, @bind ] };
876 my ($self, $where, $logic) = @_;
878 #print STDERR Data::Dumper::Concise::Dumper([ $where, $logic ]);
880 my $where_exp = $self->_expand_expr($where, $logic);
882 #print STDERR Data::Dumper::Concise::Dumper([ EXP => $where_exp ]);
884 # dispatch on appropriate method according to refkind of $where
885 my $method = $self->_METHOD_FOR_refkind("_where", $where_exp);
887 my ($sql, @bind) = $self->$method($where_exp, $logic);
889 # DBIx::Class used to call _recurse_where in scalar context
890 # something else might too...
892 return ($sql, @bind);
895 belch "Calling _recurse_where in scalar context is deprecated and will go away before 2.0";
902 #======================================================================
903 # WHERE: top-level ARRAYREF
904 #======================================================================
907 sub _where_ARRAYREF {
908 my ($self, $where, $logic) = @_;
910 $logic = uc($logic || $self->{logic});
911 $logic eq 'AND' or $logic eq 'OR' or puke "unknown logic: $logic";
913 my @clauses = @$where;
915 my (@sql_clauses, @all_bind);
916 # need to use while() so can shift() for pairs
918 my $el = shift @clauses;
920 $el = undef if (defined $el and ! length $el);
922 # switch according to kind of $el and get corresponding ($sql, @bind)
923 my ($sql, @bind) = $self->_SWITCH_refkind($el, {
925 # skip empty elements, otherwise get invalid trailing AND stuff
926 ARRAYREF => sub {$self->_recurse_where($el) if @$el},
930 $self->_assert_bindval_matches_bindtype(@b);
934 HASHREF => sub {$self->_recurse_where($el, 'and') if %$el},
936 SCALARREF => sub { ($$el); },
939 # top-level arrayref with scalars, recurse in pairs
940 $self->_recurse_where({$el => shift(@clauses)})
943 UNDEF => sub {puke "Supplying an empty left hand side argument is not supported in array-pairs" },
947 push @sql_clauses, $sql;
948 push @all_bind, @bind;
952 return $self->_join_sql_clauses($logic, \@sql_clauses, \@all_bind);
955 #======================================================================
956 # WHERE: top-level ARRAYREFREF
957 #======================================================================
959 sub _where_ARRAYREFREF {
960 my ($self, $where) = @_;
961 my ($sql, @bind) = @$$where;
962 $self->_assert_bindval_matches_bindtype(@bind);
963 return ($sql, @bind);
966 #======================================================================
967 # WHERE: top-level HASHREF
968 #======================================================================
971 my ($self, $where) = @_;
972 my (@sql_clauses, @all_bind);
974 for my $k (sort keys %$where) {
975 my $v = $where->{$k};
977 # ($k => $v) is either a special unary op or a regular hashpair
978 my ($sql, @bind) = do {
980 # put the operator in canonical form
982 $op = substr $op, 1; # remove initial dash
983 $op =~ s/^\s+|\s+$//g;# remove leading/trailing space
984 $op =~ s/\s+/ /g; # compress whitespace
986 # so that -not_foo works correctly
987 $op =~ s/^not_/NOT /i;
989 $self->_debug("Unary OP(-$op) within hashref, recursing...");
990 my ($s, @b) = $self->_where_unary_op($op, $v);
992 # top level vs nested
993 # we assume that handled unary ops will take care of their ()s
995 List::Util::first {$op =~ $_->{regex}} @{$self->{unary_ops}}
997 ( defined $self->{_nested_func_lhs} and $self->{_nested_func_lhs} eq $k )
1003 if (is_literal_value ($v) ) {
1004 belch 'Hash-pairs consisting of an empty string with a literal are deprecated, and will be removed in 2.0: use -and => [ $literal ] instead';
1007 puke "Supplying an empty left hand side argument is not supported in hash-pairs";
1011 my $method = $self->_METHOD_FOR_refkind("_where_hashpair", $v);
1012 $self->$method($k, $v);
1016 push @sql_clauses, $sql;
1017 push @all_bind, @bind;
1020 return $self->_join_sql_clauses('and', \@sql_clauses, \@all_bind);
1023 sub _where_unary_op {
1024 my ($self, $op, $rhs) = @_;
1026 $op =~ s/^-// if length($op) > 1;
1028 # top level special ops are illegal in general
1029 puke "Illegal use of top-level '-$op'"
1030 if !(defined $self->{_nested_func_lhs})
1031 and List::Util::first { $op =~ $_->{regex} } @{$self->{special_ops}}
1032 and not List::Util::first { $op =~ $_->{regex} } @{$self->{unary_ops}};
1034 if (my $op_entry = List::Util::first { $op =~ $_->{regex} } @{$self->{unary_ops}}) {
1035 my $handler = $op_entry->{handler};
1037 if (not ref $handler) {
1038 if ($op =~ s/ [_\s]? \d+ $//x ) {
1039 belch 'Use of [and|or|nest]_N modifiers is deprecated and will be removed in SQLA v2.0. '
1040 . "You probably wanted ...-and => [ -$op => COND1, -$op => COND2 ... ]";
1042 return $self->$handler($op, $rhs);
1044 elsif (ref $handler eq 'CODE') {
1045 return $handler->($self, $op, $rhs);
1048 puke "Illegal handler for operator $op - expecting a method name or a coderef";
1052 $self->_debug("Generic unary OP: $op - recursing as function");
1054 $self->_assert_pass_injection_guard($op);
1056 my ($sql, @bind) = $self->_SWITCH_refkind($rhs, {
1058 puke "Illegal use of top-level '-$op'"
1059 unless defined $self->{_nested_func_lhs};
1062 $self->_convert('?'),
1063 $self->_bindtype($self->{_nested_func_lhs}, $rhs)
1067 $self->_recurse_where($rhs)
1071 $sql = sprintf('%s %s',
1072 $self->_sqlcase($op),
1076 return ($sql, @bind);
1079 sub _where_op_ANDOR {
1080 my ($self, $op, $v) = @_;
1082 $self->_SWITCH_refkind($v, {
1084 return $self->_where_ARRAYREF($v, $op);
1088 return ($op =~ /^or/i)
1089 ? $self->_where_ARRAYREF([ map { $_ => $v->{$_} } (sort keys %$v) ], $op)
1090 : $self->_where_HASHREF($v);
1094 puke "-$op => \\\$scalar makes little sense, use " .
1096 ? '[ \$scalar, \%rest_of_conditions ] instead'
1097 : '-and => [ \$scalar, \%rest_of_conditions ] instead'
1101 ARRAYREFREF => sub {
1102 puke "-$op => \\[...] makes little sense, use " .
1104 ? '[ \[...], \%rest_of_conditions ] instead'
1105 : '-and => [ \[...], \%rest_of_conditions ] instead'
1109 SCALAR => sub { # permissively interpreted as SQL
1110 puke "-$op => \$value makes little sense, use -bool => \$value instead";
1114 puke "-$op => undef not supported";
1119 sub _where_op_NEST {
1120 my ($self, $op, $v) = @_;
1122 $self->_SWITCH_refkind($v, {
1124 SCALAR => sub { # permissively interpreted as SQL
1125 belch "literal SQL should be -nest => \\'scalar' "
1126 . "instead of -nest => 'scalar' ";
1131 puke "-$op => undef not supported";
1135 $self->_recurse_where($v);
1142 sub _where_op_BOOL {
1143 my ($self, $op, $v) = @_;
1145 my ($s, @b) = $self->_SWITCH_refkind($v, {
1146 SCALAR => sub { # interpreted as SQL column
1147 $self->_convert($self->_quote($v));
1151 puke "-$op => undef not supported";
1155 $self->_recurse_where($v);
1159 $s = "(NOT $s)" if $op =~ /^not/i;
1164 sub _where_op_IDENT {
1166 my ($op, $rhs) = splice @_, -2;
1167 if (! defined $rhs or length ref $rhs) {
1168 puke "-$op requires a single plain scalar argument (a quotable identifier)";
1171 # in case we are called as a top level special op (no '=')
1172 my $has_lhs = my $lhs = shift;
1174 $_ = $self->_convert($self->_quote($_)) for ($lhs, $rhs);
1182 sub _where_op_VALUE {
1184 my ($op, $rhs) = splice @_, -2;
1186 # in case we are called as a top level special op (no '=')
1190 if (! defined $rhs) {
1192 ? $self->_where_hashpair_HASHREF($lhs, { -is => undef })
1199 (defined $lhs ? $lhs : $self->{_nested_func_lhs}),
1206 $self->_convert($self->_quote($lhs)) . ' = ' . $self->_convert('?'),
1210 $self->_convert('?'),
1217 my %unop_postfix = map +($_ => 1), 'is null', 'is not null';
1223 my ($self, $args) = @_;
1224 my ($left, $low, $high) = @$args;
1225 my ($rhsql, @rhbind) = do {
1227 puke "Single arg to between must be a literal"
1228 unless $low->{-literal};
1231 local $self->{_nested_func_lhs} = $left->{-ident}
1232 if ref($left) eq 'HASH' and $left->{-ident};
1233 my ($l, $h) = map [ $self->_where_unary_op(%$_) ], $low, $high;
1234 (join(' ', $l->[0], $self->_sqlcase('and'), $h->[0]),
1235 @{$l}[1..$#$l], @{$h}[1..$#$h])
1238 my ($lhsql, @lhbind) = $self->_recurse_where($left);
1240 join(' ', '(', $lhsql, $self->_sqlcase($op), $rhsql, ')'),
1244 }), 'between', 'not between'),
1248 my ($self, $args) = @_;
1249 my ($lhs, $rhs) = @$args;
1252 local $self->{_nested_func_lhs} = $lhs->{-ident}
1253 if ref($lhs) eq 'HASH' and $lhs->{-ident};
1254 my ($sql, @bind) = $self->_where_unary_op(%$_);
1255 push @in_bind, @bind;
1258 my ($lhsql, @lbind) = $self->_recurse_where($lhs);
1260 $lhsql.' '.$self->_sqlcase($op).' ( '
1261 .join(', ', @in_sql)
1266 }), 'in', 'not in'),
1270 my ($self, undef, $v) = @_;
1271 my ($op, @args) = @$v;
1272 $op =~ s/^-// if length($op) > 1;
1273 local $self->{_nested_func_lhs};
1274 if (my $h = $special{$op}) {
1275 return $self->$h(\@args);
1277 if (my $us = List::Util::first { $op =~ $_->{regex} } @{$self->{user_special_ops}}) {
1278 puke "Special op '${op}' requires first value to be identifier"
1279 unless my ($k) = map $_->{-ident}, grep ref($_) eq 'HASH', $args[0];
1280 return $self->${\($us->{handler})}($k, $op, $args[1]);
1282 my $final_op = $op =~ /^(?:is|not)_/ ? join(' ', split '_', $op) : $op;
1283 if (@args == 1 and $op !~ /^(and|or)$/) {
1284 my ($expr_sql, @bind) = $self->_recurse_where($args[0]);
1285 my $op_sql = $self->_sqlcase($final_op);
1287 $unop_postfix{lc($final_op)}
1288 ? "${expr_sql} ${op_sql}"
1289 : "${op_sql} ${expr_sql}"
1291 return (($op eq 'not' ? '('.$final_sql.')' : $final_sql), @bind);
1293 my @parts = map [ $self->_recurse_where($_) ], @args;
1294 my ($final_sql) = map +($op =~ /^(and|or)$/ ? "(${_})" : $_), join(
1295 ' '.$self->_sqlcase($final_op).' ',
1300 map @{$_}[1..$#$_], @parts
1306 sub _where_op_FUNC {
1307 my ($self, undef, $rest) = @_;
1308 my ($func, @args) = @$rest;
1312 push @arg_sql, shift @x;
1314 } map [ $self->_recurse_where($_) ], @args;
1315 return ($self->_sqlcase($func).'('.join(', ', @arg_sql).')', @bind);
1318 sub _where_op_BIND {
1319 my ($self, undef, $bind) = @_;
1320 return ($self->_convert('?'), $self->_bindtype(@$bind));
1323 sub _where_op_LITERAL {
1324 my ($self, undef, $literal) = @_;
1325 $self->_assert_bindval_matches_bindtype(@{$literal}[1..$#$literal]);
1329 sub _where_hashpair_ARRAYREF {
1330 my ($self, $k, $v) = @_;
1333 my @v = @$v; # need copy because of shift below
1334 $self->_debug("ARRAY($k) means distribute over elements");
1336 # put apart first element if it is an operator (-and, -or)
1338 (defined $v[0] && $v[0] =~ /^ - (?: AND|OR ) $/ix)
1342 my @distributed = map { {$k => $_} } @v;
1345 $self->_debug("OP($op) reinjected into the distributed array");
1346 unshift @distributed, $op;
1349 my $logic = $op ? substr($op, 1) : '';
1351 return $self->_recurse_where(\@distributed, $logic);
1354 $self->_debug("empty ARRAY($k) means 0=1");
1355 return ($self->{sqlfalse});
1359 sub _where_hashpair_HASHREF {
1360 my ($self, $k, $v, $logic) = @_;
1363 local $self->{_nested_func_lhs} = defined $self->{_nested_func_lhs}
1364 ? $self->{_nested_func_lhs}
1368 my ($all_sql, @all_bind);
1370 for my $orig_op (sort keys %$v) {
1371 my $val = $v->{$orig_op};
1373 # put the operator in canonical form
1376 # FIXME - we need to phase out dash-less ops
1377 $op =~ s/^-//; # remove possible initial dash
1378 $op =~ s/^\s+|\s+$//g;# remove leading/trailing space
1379 $op =~ s/\s+/ /g; # compress whitespace
1381 $self->_assert_pass_injection_guard($op);
1384 $op =~ s/^is_not/IS NOT/i;
1386 # so that -not_foo works correctly
1387 $op =~ s/^not_/NOT /i;
1389 # another retarded special case: foo => { $op => { -value => undef } }
1390 if (ref $val eq 'HASH' and keys %$val == 1 and exists $val->{-value} and ! defined $val->{-value} ) {
1396 # CASE: col-value logic modifiers
1397 if ($orig_op =~ /^ \- (and|or) $/xi) {
1398 ($sql, @bind) = $self->_where_hashpair_HASHREF($k, $val, $1);
1400 # CASE: special operators like -in or -between
1401 elsif (my $special_op = List::Util::first { $op =~ $_->{regex} } @{$self->{special_ops}}) {
1402 my $handler = $special_op->{handler};
1404 puke "No handler supplied for special operator $orig_op";
1406 elsif (not ref $handler) {
1407 ($sql, @bind) = $self->$handler($k, $op, $val);
1409 elsif (ref $handler eq 'CODE') {
1410 ($sql, @bind) = $handler->($self, $k, $op, $val);
1413 puke "Illegal handler for special operator $orig_op - expecting a method name or a coderef";
1417 $self->_SWITCH_refkind($val, {
1419 ARRAYREF => sub { # CASE: col => {op => \@vals}
1420 ($sql, @bind) = $self->_where_field_op_ARRAYREF($k, $op, $val);
1423 ARRAYREFREF => sub { # CASE: col => {op => \[$sql, @bind]} (literal SQL with bind)
1424 my ($sub_sql, @sub_bind) = @$$val;
1425 $self->_assert_bindval_matches_bindtype(@sub_bind);
1426 $sql = join ' ', $self->_convert($self->_quote($k)),
1427 $self->_sqlcase($op),
1432 UNDEF => sub { # CASE: col => {op => undef} : sql "IS (NOT)? NULL"
1434 $op =~ /^not$/i ? 'is not' # legacy
1435 : $op =~ $self->{equality_op} ? 'is'
1436 : $op =~ $self->{like_op} ? belch("Supplying an undefined argument to '@{[ uc $op]}' is deprecated") && 'is'
1437 : $op =~ $self->{inequality_op} ? 'is not'
1438 : $op =~ $self->{not_like_op} ? belch("Supplying an undefined argument to '@{[ uc $op]}' is deprecated") && 'is not'
1439 : puke "unexpected operator '$orig_op' with undef operand";
1441 $sql = $self->_quote($k) . $self->_sqlcase(" $is null");
1444 FALLBACK => sub { # CASE: col => {op/func => $stuff}
1445 ($sql, @bind) = $self->_where_unary_op($op, $val);
1448 $self->_convert($self->_quote($k)),
1449 $self->{_nested_func_lhs} eq $k ? $sql : "($sql)", # top level vs nested
1455 ($all_sql) = (defined $all_sql and $all_sql) ? $self->_join_sql_clauses($logic, [$all_sql, $sql], []) : $sql;
1456 push @all_bind, @bind;
1458 return ($all_sql, @all_bind);
1461 sub _where_field_IS {
1462 my ($self, $k, $op, $v) = @_;
1464 my ($s) = $self->_SWITCH_refkind($v, {
1467 $self->_convert($self->_quote($k)),
1468 map { $self->_sqlcase($_)} ($op, 'null')
1471 puke "$op can only take undef as argument";
1478 sub _where_field_op_ARRAYREF {
1479 my ($self, $k, $op, $vals) = @_;
1481 my @vals = @$vals; #always work on a copy
1484 $self->_debug(sprintf '%s means multiple elements: [ %s ]',
1486 join(', ', map { defined $_ ? "'$_'" : 'NULL' } @vals ),
1489 # see if the first element is an -and/-or op
1491 if (defined $vals[0] && $vals[0] =~ /^ - (AND|OR) $/ix) {
1496 # a long standing API wart - an attempt to change this behavior during
1497 # the 1.50 series failed *spectacularly*. Warn instead and leave the
1502 (!$logic or $logic eq 'OR')
1504 ($op =~ $self->{inequality_op} or $op =~ $self->{not_like_op})
1507 belch "A multi-element arrayref as an argument to the inequality op '$o' "
1508 . 'is technically equivalent to an always-true 1=1 (you probably wanted '
1509 . "to say ...{ \$inequality_op => [ -and => \@values ] }... instead)"
1513 # distribute $op over each remaining member of @vals, append logic if exists
1514 return $self->_recurse_where([map { {$k => {$op, $_}} } @vals], $logic);
1518 # try to DWIM on equality operators
1520 $op =~ $self->{equality_op} ? $self->{sqlfalse}
1521 : $op =~ $self->{like_op} ? belch("Supplying an empty arrayref to '@{[ uc $op]}' is deprecated") && $self->{sqlfalse}
1522 : $op =~ $self->{inequality_op} ? $self->{sqltrue}
1523 : $op =~ $self->{not_like_op} ? belch("Supplying an empty arrayref to '@{[ uc $op]}' is deprecated") && $self->{sqltrue}
1524 : puke "operator '$op' applied on an empty array (field '$k')";
1529 sub _where_hashpair_SCALARREF {
1530 my ($self, $k, $v) = @_;
1531 $self->_debug("SCALAR($k) means literal SQL: $$v");
1532 my $sql = $self->_quote($k) . " " . $$v;
1536 # literal SQL with bind
1537 sub _where_hashpair_ARRAYREFREF {
1538 my ($self, $k, $v) = @_;
1539 $self->_debug("REF($k) means literal SQL: @${$v}");
1540 my ($sql, @bind) = @$$v;
1541 $self->_assert_bindval_matches_bindtype(@bind);
1542 $sql = $self->_quote($k) . " " . $sql;
1543 return ($sql, @bind );
1546 # literal SQL without bind
1547 sub _where_hashpair_SCALAR {
1548 my ($self, $k, $v) = @_;
1549 $self->_debug("NOREF($k) means simple key=val: $k $self->{cmp} $v");
1550 return ($self->_where_hashpair_HASHREF($k, { $self->{cmp} => $v }));
1554 sub _where_hashpair_UNDEF {
1555 my ($self, $k, $v) = @_;
1556 $self->_debug("UNDEF($k) means IS NULL");
1557 return $self->_where_hashpair_HASHREF($k, { -is => undef });
1560 #======================================================================
1561 # WHERE: TOP-LEVEL OTHERS (SCALARREF, SCALAR, UNDEF)
1562 #======================================================================
1565 sub _where_SCALARREF {
1566 my ($self, $where) = @_;
1569 $self->_debug("SCALAR(*top) means literal SQL: $$where");
1575 my ($self, $where) = @_;
1578 $self->_debug("NOREF(*top) means literal SQL: $where");
1589 #======================================================================
1590 # WHERE: BUILTIN SPECIAL OPERATORS (-in, -between)
1591 #======================================================================
1594 sub _where_field_BETWEEN {
1595 my ($self, $k, $op, $vals) = @_;
1597 my ($label, $and, $placeholder);
1598 $label = $self->_convert($self->_quote($k));
1599 $and = ' ' . $self->_sqlcase('and') . ' ';
1600 $placeholder = $self->_convert('?');
1601 $op = $self->_sqlcase($op);
1603 my $invalid_args = "Operator '$op' requires either an arrayref with two defined values or expressions, or a single literal scalarref/arrayref-ref";
1605 my ($clause, @bind) = $self->_SWITCH_refkind($vals, {
1606 ARRAYREFREF => sub {
1607 my ($s, @b) = @$$vals;
1608 $self->_assert_bindval_matches_bindtype(@b);
1615 puke $invalid_args if @$vals != 2;
1617 my (@all_sql, @all_bind);
1618 foreach my $val (@$vals) {
1619 my ($sql, @bind) = $self->_SWITCH_refkind($val, {
1621 return ($placeholder, $self->_bindtype($k, $val) );
1626 ARRAYREFREF => sub {
1627 my ($sql, @bind) = @$$val;
1628 $self->_assert_bindval_matches_bindtype(@bind);
1629 return ($sql, @bind);
1632 my ($func, $arg, @rest) = %$val;
1633 puke "Only simple { -func => arg } functions accepted as sub-arguments to BETWEEN"
1634 if (@rest or $func !~ /^ \- (.+)/x);
1635 $self->_where_unary_op($1 => $arg);
1641 push @all_sql, $sql;
1642 push @all_bind, @bind;
1646 (join $and, @all_sql),
1655 my $sql = "( $label $op $clause )";
1656 return ($sql, @bind)
1660 sub _where_field_IN {
1661 my ($self, $k, $op, $vals) = @_;
1663 # backwards compatibility: if scalar, force into an arrayref
1664 $vals = [$vals] if defined $vals && ! ref $vals;
1666 my ($label) = $self->_convert($self->_quote($k));
1667 my ($placeholder) = $self->_convert('?');
1668 $op = $self->_sqlcase($op);
1670 my ($sql, @bind) = $self->_SWITCH_refkind($vals, {
1671 ARRAYREF => sub { # list of choices
1672 if (@$vals) { # nonempty list
1673 my (@all_sql, @all_bind);
1675 for my $val (@$vals) {
1676 my ($sql, @bind) = $self->_SWITCH_refkind($val, {
1678 return ($placeholder, $val);
1683 ARRAYREFREF => sub {
1684 my ($sql, @bind) = @$$val;
1685 $self->_assert_bindval_matches_bindtype(@bind);
1686 return ($sql, @bind);
1689 my ($func, $arg, @rest) = %$val;
1690 puke "Only simple { -func => arg } functions accepted as sub-arguments to IN"
1691 if (@rest or $func !~ /^ \- (.+)/x);
1692 $self->_where_unary_op($1 => $arg);
1696 'SQL::Abstract before v1.75 used to generate incorrect SQL when the '
1697 . "-$op operator was given an undef-containing list: !!!AUDIT YOUR CODE "
1698 . 'AND DATA!!! (the upcoming Data::Query-based version of SQL::Abstract '
1699 . 'will emit the logically correct SQL instead of raising this exception)'
1703 push @all_sql, $sql;
1704 push @all_bind, @bind;
1708 sprintf('%s %s ( %s )',
1711 join(', ', @all_sql)
1713 $self->_bindtype($k, @all_bind),
1716 else { # empty list: some databases won't understand "IN ()", so DWIM
1717 my $sql = ($op =~ /\bnot\b/i) ? $self->{sqltrue} : $self->{sqlfalse};
1722 SCALARREF => sub { # literal SQL
1723 my $sql = $self->_open_outer_paren($$vals);
1724 return ("$label $op ( $sql )");
1726 ARRAYREFREF => sub { # literal SQL with bind
1727 my ($sql, @bind) = @$$vals;
1728 $self->_assert_bindval_matches_bindtype(@bind);
1729 $sql = $self->_open_outer_paren($sql);
1730 return ("$label $op ( $sql )", @bind);
1734 puke "Argument passed to the '$op' operator can not be undefined";
1738 puke "special op $op requires an arrayref (or scalarref/arrayref-ref)";
1742 return ($sql, @bind);
1745 # Some databases (SQLite) treat col IN (1, 2) different from
1746 # col IN ( (1, 2) ). Use this to strip all outer parens while
1747 # adding them back in the corresponding method
1748 sub _open_outer_paren {
1749 my ($self, $sql) = @_;
1751 while (my ($inner) = $sql =~ /^ \s* \( (.*) \) \s* $/xs) {
1753 # there are closing parens inside, need the heavy duty machinery
1754 # to reevaluate the extraction starting from $sql (full reevaluation)
1755 if ($inner =~ /\)/) {
1756 require Text::Balanced;
1758 my (undef, $remainder) = do {
1759 # idiotic design - writes to $@ but *DOES NOT* throw exceptions
1761 Text::Balanced::extract_bracketed($sql, '()', qr/\s*/);
1764 # the entire expression needs to be a balanced bracketed thing
1765 # (after an extract no remainder sans trailing space)
1766 last if defined $remainder and $remainder =~ /\S/;
1776 #======================================================================
1778 #======================================================================
1781 my ($self, $arg) = @_;
1784 for my $c ($self->_order_by_chunks($arg) ) {
1785 $self->_SWITCH_refkind($c, {
1786 SCALAR => sub { push @sql, $c },
1787 ARRAYREF => sub { push @sql, shift @$c; push @bind, @$c },
1793 $self->_sqlcase(' order by'),
1799 return wantarray ? ($sql, @bind) : $sql;
1802 sub _order_by_chunks {
1803 my ($self, $arg) = @_;
1805 return $self->_SWITCH_refkind($arg, {
1808 map { $self->_order_by_chunks($_ ) } @$arg;
1811 ARRAYREFREF => sub {
1812 my ($s, @b) = @$$arg;
1813 $self->_assert_bindval_matches_bindtype(@b);
1817 SCALAR => sub {$self->_quote($arg)},
1819 UNDEF => sub {return () },
1821 SCALARREF => sub {$$arg}, # literal SQL, no quoting
1824 # get first pair in hash
1825 my ($key, $val, @rest) = %$arg;
1827 return () unless $key;
1829 if (@rest or not $key =~ /^-(desc|asc)/i) {
1830 puke "hash passed to _order_by must have exactly one key (-desc or -asc)";
1836 for my $c ($self->_order_by_chunks($val)) {
1839 $self->_SWITCH_refkind($c, {
1844 ($sql, @bind) = @$c;
1848 $sql = $sql . ' ' . $self->_sqlcase($direction);
1850 push @ret, [ $sql, @bind];
1859 #======================================================================
1860 # DATASOURCE (FOR NOW, JUST PLAIN TABLE OR LIST OF TABLES)
1861 #======================================================================
1866 $self->_SWITCH_refkind($from, {
1867 ARRAYREF => sub {join ', ', map { $self->_quote($_) } @$from;},
1868 SCALAR => sub {$self->_quote($from)},
1869 SCALARREF => sub {$$from},
1874 #======================================================================
1876 #======================================================================
1878 # highly optimized, as it's called way too often
1880 # my ($self, $label) = @_;
1882 return '' unless defined $_[1];
1883 return ${$_[1]} if ref($_[1]) eq 'SCALAR';
1885 $_[0]->{quote_char} or
1886 ($_[0]->_assert_pass_injection_guard($_[1]), return $_[1]);
1888 my $qref = ref $_[0]->{quote_char};
1890 !$qref ? ($_[0]->{quote_char}, $_[0]->{quote_char})
1891 : ($qref eq 'ARRAY') ? @{$_[0]->{quote_char}}
1892 : puke "Unsupported quote_char format: $_[0]->{quote_char}";
1894 my $esc = $_[0]->{escape_char} || $r;
1896 # parts containing * are naturally unquoted
1897 return join($_[0]->{name_sep}||'', map
1898 +( $_ eq '*' ? $_ : do { (my $n = $_) =~ s/(\Q$esc\E|\Q$r\E)/$esc$1/g; $l . $n . $r } ),
1899 ( $_[0]->{name_sep} ? split (/\Q$_[0]->{name_sep}\E/, $_[1] ) : $_[1] )
1904 # Conversion, if applicable
1906 #my ($self, $arg) = @_;
1907 if ($_[0]->{convert}) {
1908 return $_[0]->_sqlcase($_[0]->{convert}) .'(' . $_[1] . ')';
1915 #my ($self, $col, @vals) = @_;
1916 # called often - tighten code
1917 return $_[0]->{bindtype} eq 'columns'
1918 ? map {[$_[1], $_]} @_[2 .. $#_]
1923 # Dies if any element of @bind is not in [colname => value] format
1924 # if bindtype is 'columns'.
1925 sub _assert_bindval_matches_bindtype {
1926 # my ($self, @bind) = @_;
1928 if ($self->{bindtype} eq 'columns') {
1930 if (!defined $_ || ref($_) ne 'ARRAY' || @$_ != 2) {
1931 puke "bindtype 'columns' selected, you need to pass: [column_name => bind_value]"
1937 sub _join_sql_clauses {
1938 my ($self, $logic, $clauses_aref, $bind_aref) = @_;
1940 if (@$clauses_aref > 1) {
1941 my $join = " " . $self->_sqlcase($logic) . " ";
1942 my $sql = '( ' . join($join, @$clauses_aref) . ' )';
1943 return ($sql, @$bind_aref);
1945 elsif (@$clauses_aref) {
1946 return ($clauses_aref->[0], @$bind_aref); # no parentheses
1949 return (); # if no SQL, ignore @$bind_aref
1954 # Fix SQL case, if so requested
1956 # LDNOTE: if $self->{case} is true, then it contains 'lower', so we
1957 # don't touch the argument ... crooked logic, but let's not change it!
1958 return $_[0]->{case} ? $_[1] : uc($_[1]);
1962 #======================================================================
1963 # DISPATCHING FROM REFKIND
1964 #======================================================================
1967 my ($self, $data) = @_;
1969 return 'UNDEF' unless defined $data;
1971 # blessed objects are treated like scalars
1972 my $ref = (Scalar::Util::blessed $data) ? '' : ref $data;
1974 return 'SCALAR' unless $ref;
1977 while ($ref eq 'REF') {
1979 $ref = (Scalar::Util::blessed $data) ? '' : ref $data;
1983 return ($ref||'SCALAR') . ('REF' x $n_steps);
1987 my ($self, $data) = @_;
1988 my @try = ($self->_refkind($data));
1989 push @try, 'SCALAR_or_UNDEF' if $try[0] eq 'SCALAR' || $try[0] eq 'UNDEF';
1990 push @try, 'FALLBACK';
1994 sub _METHOD_FOR_refkind {
1995 my ($self, $meth_prefix, $data) = @_;
1998 for (@{$self->_try_refkind($data)}) {
1999 $method = $self->can($meth_prefix."_".$_)
2003 return $method || puke "cannot dispatch on '$meth_prefix' for ".$self->_refkind($data);
2007 sub _SWITCH_refkind {
2008 my ($self, $data, $dispatch_table) = @_;
2011 for (@{$self->_try_refkind($data)}) {
2012 $coderef = $dispatch_table->{$_}
2016 puke "no dispatch entry for ".$self->_refkind($data)
2025 #======================================================================
2026 # VALUES, GENERATE, AUTOLOAD
2027 #======================================================================
2029 # LDNOTE: original code from nwiger, didn't touch code in that section
2030 # I feel the AUTOLOAD stuff should not be the default, it should
2031 # only be activated on explicit demand by user.
2035 my $data = shift || return;
2036 puke "Argument to ", __PACKAGE__, "->values must be a \\%hash"
2037 unless ref $data eq 'HASH';
2040 foreach my $k (sort keys %$data) {
2041 my $v = $data->{$k};
2042 $self->_SWITCH_refkind($v, {
2044 if ($self->{array_datatypes}) { # array datatype
2045 push @all_bind, $self->_bindtype($k, $v);
2047 else { # literal SQL with bind
2048 my ($sql, @bind) = @$v;
2049 $self->_assert_bindval_matches_bindtype(@bind);
2050 push @all_bind, @bind;
2053 ARRAYREFREF => sub { # literal SQL with bind
2054 my ($sql, @bind) = @${$v};
2055 $self->_assert_bindval_matches_bindtype(@bind);
2056 push @all_bind, @bind;
2058 SCALARREF => sub { # literal SQL without bind
2060 SCALAR_or_UNDEF => sub {
2061 push @all_bind, $self->_bindtype($k, $v);
2072 my(@sql, @sqlq, @sqlv);
2076 if ($ref eq 'HASH') {
2077 for my $k (sort keys %$_) {
2080 my $label = $self->_quote($k);
2081 if ($r eq 'ARRAY') {
2082 # literal SQL with bind
2083 my ($sql, @bind) = @$v;
2084 $self->_assert_bindval_matches_bindtype(@bind);
2085 push @sqlq, "$label = $sql";
2087 } elsif ($r eq 'SCALAR') {
2088 # literal SQL without bind
2089 push @sqlq, "$label = $$v";
2091 push @sqlq, "$label = ?";
2092 push @sqlv, $self->_bindtype($k, $v);
2095 push @sql, $self->_sqlcase('set'), join ', ', @sqlq;
2096 } elsif ($ref eq 'ARRAY') {
2097 # unlike insert(), assume these are ONLY the column names, i.e. for SQL
2100 if ($r eq 'ARRAY') { # literal SQL with bind
2101 my ($sql, @bind) = @$v;
2102 $self->_assert_bindval_matches_bindtype(@bind);
2105 } elsif ($r eq 'SCALAR') { # literal SQL without bind
2106 # embedded literal SQL
2113 push @sql, '(' . join(', ', @sqlq) . ')';
2114 } elsif ($ref eq 'SCALAR') {
2118 # strings get case twiddled
2119 push @sql, $self->_sqlcase($_);
2123 my $sql = join ' ', @sql;
2125 # this is pretty tricky
2126 # if ask for an array, return ($stmt, @bind)
2127 # otherwise, s/?/shift @sqlv/ to put it inline
2129 return ($sql, @sqlv);
2131 1 while $sql =~ s/\?/my $d = shift(@sqlv);
2132 ref $d ? $d->[1] : $d/e;
2141 # This allows us to check for a local, then _form, attr
2143 my($name) = $AUTOLOAD =~ /.*::(.+)/;
2144 return $self->generate($name, @_);
2155 SQL::Abstract - Generate SQL from Perl data structures
2161 my $sql = SQL::Abstract->new;
2163 my($stmt, @bind) = $sql->select($source, \@fields, \%where, $order);
2165 my($stmt, @bind) = $sql->insert($table, \%fieldvals || \@values);
2167 my($stmt, @bind) = $sql->update($table, \%fieldvals, \%where);
2169 my($stmt, @bind) = $sql->delete($table, \%where);
2171 # Then, use these in your DBI statements
2172 my $sth = $dbh->prepare($stmt);
2173 $sth->execute(@bind);
2175 # Just generate the WHERE clause
2176 my($stmt, @bind) = $sql->where(\%where, $order);
2178 # Return values in the same order, for hashed queries
2179 # See PERFORMANCE section for more details
2180 my @bind = $sql->values(\%fieldvals);
2184 This module was inspired by the excellent L<DBIx::Abstract>.
2185 However, in using that module I found that what I really wanted
2186 to do was generate SQL, but still retain complete control over my
2187 statement handles and use the DBI interface. So, I set out to
2188 create an abstract SQL generation module.
2190 While based on the concepts used by L<DBIx::Abstract>, there are
2191 several important differences, especially when it comes to WHERE
2192 clauses. I have modified the concepts used to make the SQL easier
2193 to generate from Perl data structures and, IMO, more intuitive.
2194 The underlying idea is for this module to do what you mean, based
2195 on the data structures you provide it. The big advantage is that
2196 you don't have to modify your code every time your data changes,
2197 as this module figures it out.
2199 To begin with, an SQL INSERT is as easy as just specifying a hash
2200 of C<key=value> pairs:
2203 name => 'Jimbo Bobson',
2204 phone => '123-456-7890',
2205 address => '42 Sister Lane',
2206 city => 'St. Louis',
2207 state => 'Louisiana',
2210 The SQL can then be generated with this:
2212 my($stmt, @bind) = $sql->insert('people', \%data);
2214 Which would give you something like this:
2216 $stmt = "INSERT INTO people
2217 (address, city, name, phone, state)
2218 VALUES (?, ?, ?, ?, ?)";
2219 @bind = ('42 Sister Lane', 'St. Louis', 'Jimbo Bobson',
2220 '123-456-7890', 'Louisiana');
2222 These are then used directly in your DBI code:
2224 my $sth = $dbh->prepare($stmt);
2225 $sth->execute(@bind);
2227 =head2 Inserting and Updating Arrays
2229 If your database has array types (like for example Postgres),
2230 activate the special option C<< array_datatypes => 1 >>
2231 when creating the C<SQL::Abstract> object.
2232 Then you may use an arrayref to insert and update database array types:
2234 my $sql = SQL::Abstract->new(array_datatypes => 1);
2236 planets => [qw/Mercury Venus Earth Mars/]
2239 my($stmt, @bind) = $sql->insert('solar_system', \%data);
2243 $stmt = "INSERT INTO solar_system (planets) VALUES (?)"
2245 @bind = (['Mercury', 'Venus', 'Earth', 'Mars']);
2248 =head2 Inserting and Updating SQL
2250 In order to apply SQL functions to elements of your C<%data> you may
2251 specify a reference to an arrayref for the given hash value. For example,
2252 if you need to execute the Oracle C<to_date> function on a value, you can
2253 say something like this:
2257 date_entered => \[ "to_date(?,'MM/DD/YYYY')", "03/02/2003" ],
2260 The first value in the array is the actual SQL. Any other values are
2261 optional and would be included in the bind values array. This gives
2264 my($stmt, @bind) = $sql->insert('people', \%data);
2266 $stmt = "INSERT INTO people (name, date_entered)
2267 VALUES (?, to_date(?,'MM/DD/YYYY'))";
2268 @bind = ('Bill', '03/02/2003');
2270 An UPDATE is just as easy, all you change is the name of the function:
2272 my($stmt, @bind) = $sql->update('people', \%data);
2274 Notice that your C<%data> isn't touched; the module will generate
2275 the appropriately quirky SQL for you automatically. Usually you'll
2276 want to specify a WHERE clause for your UPDATE, though, which is
2277 where handling C<%where> hashes comes in handy...
2279 =head2 Complex where statements
2281 This module can generate pretty complicated WHERE statements
2282 easily. For example, simple C<key=value> pairs are taken to mean
2283 equality, and if you want to see if a field is within a set
2284 of values, you can use an arrayref. Let's say we wanted to
2285 SELECT some data based on this criteria:
2288 requestor => 'inna',
2289 worker => ['nwiger', 'rcwe', 'sfz'],
2290 status => { '!=', 'completed' }
2293 my($stmt, @bind) = $sql->select('tickets', '*', \%where);
2295 The above would give you something like this:
2297 $stmt = "SELECT * FROM tickets WHERE
2298 ( requestor = ? ) AND ( status != ? )
2299 AND ( worker = ? OR worker = ? OR worker = ? )";
2300 @bind = ('inna', 'completed', 'nwiger', 'rcwe', 'sfz');
2302 Which you could then use in DBI code like so:
2304 my $sth = $dbh->prepare($stmt);
2305 $sth->execute(@bind);
2311 The methods are simple. There's one for every major SQL operation,
2312 and a constructor you use first. The arguments are specified in a
2313 similar order for each method (table, then fields, then a where
2314 clause) to try and simplify things.
2316 =head2 new(option => 'value')
2318 The C<new()> function takes a list of options and values, and returns
2319 a new B<SQL::Abstract> object which can then be used to generate SQL
2320 through the methods below. The options accepted are:
2326 If set to 'lower', then SQL will be generated in all lowercase. By
2327 default SQL is generated in "textbook" case meaning something like:
2329 SELECT a_field FROM a_table WHERE some_field LIKE '%someval%'
2331 Any setting other than 'lower' is ignored.
2335 This determines what the default comparison operator is. By default
2336 it is C<=>, meaning that a hash like this:
2338 %where = (name => 'nwiger', email => 'nate@wiger.org');
2340 Will generate SQL like this:
2342 WHERE name = 'nwiger' AND email = 'nate@wiger.org'
2344 However, you may want loose comparisons by default, so if you set
2345 C<cmp> to C<like> you would get SQL such as:
2347 WHERE name like 'nwiger' AND email like 'nate@wiger.org'
2349 You can also override the comparison on an individual basis - see
2350 the huge section on L</"WHERE CLAUSES"> at the bottom.
2352 =item sqltrue, sqlfalse
2354 Expressions for inserting boolean values within SQL statements.
2355 By default these are C<1=1> and C<1=0>. They are used
2356 by the special operators C<-in> and C<-not_in> for generating
2357 correct SQL even when the argument is an empty array (see below).
2361 This determines the default logical operator for multiple WHERE
2362 statements in arrays or hashes. If absent, the default logic is "or"
2363 for arrays, and "and" for hashes. This means that a WHERE
2367 event_date => {'>=', '2/13/99'},
2368 event_date => {'<=', '4/24/03'},
2371 will generate SQL like this:
2373 WHERE event_date >= '2/13/99' OR event_date <= '4/24/03'
2375 This is probably not what you want given this query, though (look
2376 at the dates). To change the "OR" to an "AND", simply specify:
2378 my $sql = SQL::Abstract->new(logic => 'and');
2380 Which will change the above C<WHERE> to:
2382 WHERE event_date >= '2/13/99' AND event_date <= '4/24/03'
2384 The logic can also be changed locally by inserting
2385 a modifier in front of an arrayref:
2387 @where = (-and => [event_date => {'>=', '2/13/99'},
2388 event_date => {'<=', '4/24/03'} ]);
2390 See the L</"WHERE CLAUSES"> section for explanations.
2394 This will automatically convert comparisons using the specified SQL
2395 function for both column and value. This is mostly used with an argument
2396 of C<upper> or C<lower>, so that the SQL will have the effect of
2397 case-insensitive "searches". For example, this:
2399 $sql = SQL::Abstract->new(convert => 'upper');
2400 %where = (keywords => 'MaKe iT CAse inSeNSItive');
2402 Will turn out the following SQL:
2404 WHERE upper(keywords) like upper('MaKe iT CAse inSeNSItive')
2406 The conversion can be C<upper()>, C<lower()>, or any other SQL function
2407 that can be applied symmetrically to fields (actually B<SQL::Abstract> does
2408 not validate this option; it will just pass through what you specify verbatim).
2412 This is a kludge because many databases suck. For example, you can't
2413 just bind values using DBI's C<execute()> for Oracle C<CLOB> or C<BLOB> fields.
2414 Instead, you have to use C<bind_param()>:
2416 $sth->bind_param(1, 'reg data');
2417 $sth->bind_param(2, $lots, {ora_type => ORA_CLOB});
2419 The problem is, B<SQL::Abstract> will normally just return a C<@bind> array,
2420 which loses track of which field each slot refers to. Fear not.
2422 If you specify C<bindtype> in new, you can determine how C<@bind> is returned.
2423 Currently, you can specify either C<normal> (default) or C<columns>. If you
2424 specify C<columns>, you will get an array that looks like this:
2426 my $sql = SQL::Abstract->new(bindtype => 'columns');
2427 my($stmt, @bind) = $sql->insert(...);
2430 [ 'column1', 'value1' ],
2431 [ 'column2', 'value2' ],
2432 [ 'column3', 'value3' ],
2435 You can then iterate through this manually, using DBI's C<bind_param()>.
2437 $sth->prepare($stmt);
2440 my($col, $data) = @$_;
2441 if ($col eq 'details' || $col eq 'comments') {
2442 $sth->bind_param($i, $data, {ora_type => ORA_CLOB});
2443 } elsif ($col eq 'image') {
2444 $sth->bind_param($i, $data, {ora_type => ORA_BLOB});
2446 $sth->bind_param($i, $data);
2450 $sth->execute; # execute without @bind now
2452 Now, why would you still use B<SQL::Abstract> if you have to do this crap?
2453 Basically, the advantage is still that you don't have to care which fields
2454 are or are not included. You could wrap that above C<for> loop in a simple
2455 sub called C<bind_fields()> or something and reuse it repeatedly. You still
2456 get a layer of abstraction over manual SQL specification.
2458 Note that if you set L</bindtype> to C<columns>, the C<\[ $sql, @bind ]>
2459 construct (see L</Literal SQL with placeholders and bind values (subqueries)>)
2460 will expect the bind values in this format.
2464 This is the character that a table or column name will be quoted
2465 with. By default this is an empty string, but you could set it to
2466 the character C<`>, to generate SQL like this:
2468 SELECT `a_field` FROM `a_table` WHERE `some_field` LIKE '%someval%'
2470 Alternatively, you can supply an array ref of two items, the first being the left
2471 hand quote character, and the second the right hand quote character. For
2472 example, you could supply C<['[',']']> for SQL Server 2000 compliant quotes
2473 that generates SQL like this:
2475 SELECT [a_field] FROM [a_table] WHERE [some_field] LIKE '%someval%'
2477 Quoting is useful if you have tables or columns names that are reserved
2478 words in your database's SQL dialect.
2482 This is the character that will be used to escape L</quote_char>s appearing
2483 in an identifier before it has been quoted.
2485 The parameter default in case of a single L</quote_char> character is the quote
2488 When opening-closing-style quoting is used (L</quote_char> is an arrayref)
2489 this parameter defaults to the B<closing (right)> L</quote_char>. Occurrences
2490 of the B<opening (left)> L</quote_char> within the identifier are currently left
2491 untouched. The default for opening-closing-style quotes may change in future
2492 versions, thus you are B<strongly encouraged> to specify the escape character
2497 This is the character that separates a table and column name. It is
2498 necessary to specify this when the C<quote_char> option is selected,
2499 so that tables and column names can be individually quoted like this:
2501 SELECT `table`.`one_field` FROM `table` WHERE `table`.`other_field` = 1
2503 =item injection_guard
2505 A regular expression C<qr/.../> that is applied to any C<-function> and unquoted
2506 column name specified in a query structure. This is a safety mechanism to avoid
2507 injection attacks when mishandling user input e.g.:
2509 my %condition_as_column_value_pairs = get_values_from_user();
2510 $sqla->select( ... , \%condition_as_column_value_pairs );
2512 If the expression matches an exception is thrown. Note that literal SQL
2513 supplied via C<\'...'> or C<\['...']> is B<not> checked in any way.
2515 Defaults to checking for C<;> and the C<GO> keyword (TransactSQL)
2517 =item array_datatypes
2519 When this option is true, arrayrefs in INSERT or UPDATE are
2520 interpreted as array datatypes and are passed directly
2522 When this option is false, arrayrefs are interpreted
2523 as literal SQL, just like refs to arrayrefs
2524 (but this behavior is for backwards compatibility; when writing
2525 new queries, use the "reference to arrayref" syntax
2531 Takes a reference to a list of "special operators"
2532 to extend the syntax understood by L<SQL::Abstract>.
2533 See section L</"SPECIAL OPERATORS"> for details.
2537 Takes a reference to a list of "unary operators"
2538 to extend the syntax understood by L<SQL::Abstract>.
2539 See section L</"UNARY OPERATORS"> for details.
2545 =head2 insert($table, \@values || \%fieldvals, \%options)
2547 This is the simplest function. You simply give it a table name
2548 and either an arrayref of values or hashref of field/value pairs.
2549 It returns an SQL INSERT statement and a list of bind values.
2550 See the sections on L</"Inserting and Updating Arrays"> and
2551 L</"Inserting and Updating SQL"> for information on how to insert
2552 with those data types.
2554 The optional C<\%options> hash reference may contain additional
2555 options to generate the insert SQL. Currently supported options
2562 Takes either a scalar of raw SQL fields, or an array reference of
2563 field names, and adds on an SQL C<RETURNING> statement at the end.
2564 This allows you to return data generated by the insert statement
2565 (such as row IDs) without performing another C<SELECT> statement.
2566 Note, however, this is not part of the SQL standard and may not
2567 be supported by all database engines.
2571 =head2 update($table, \%fieldvals, \%where, \%options)
2573 This takes a table, hashref of field/value pairs, and an optional
2574 hashref L<WHERE clause|/WHERE CLAUSES>. It returns an SQL UPDATE function and a list
2576 See the sections on L</"Inserting and Updating Arrays"> and
2577 L</"Inserting and Updating SQL"> for information on how to insert
2578 with those data types.
2580 The optional C<\%options> hash reference may contain additional
2581 options to generate the update SQL. Currently supported options
2588 See the C<returning> option to
2589 L<insert|/insert($table, \@values || \%fieldvals, \%options)>.
2593 =head2 select($source, $fields, $where, $order)
2595 This returns a SQL SELECT statement and associated list of bind values, as
2596 specified by the arguments:
2602 Specification of the 'FROM' part of the statement.
2603 The argument can be either a plain scalar (interpreted as a table
2604 name, will be quoted), or an arrayref (interpreted as a list
2605 of table names, joined by commas, quoted), or a scalarref
2606 (literal SQL, not quoted).
2610 Specification of the list of fields to retrieve from
2612 The argument can be either an arrayref (interpreted as a list
2613 of field names, will be joined by commas and quoted), or a
2614 plain scalar (literal SQL, not quoted).
2615 Please observe that this API is not as flexible as that of
2616 the first argument C<$source>, for backwards compatibility reasons.
2620 Optional argument to specify the WHERE part of the query.
2621 The argument is most often a hashref, but can also be
2622 an arrayref or plain scalar --
2623 see section L<WHERE clause|/"WHERE CLAUSES"> for details.
2627 Optional argument to specify the ORDER BY part of the query.
2628 The argument can be a scalar, a hashref or an arrayref
2629 -- see section L<ORDER BY clause|/"ORDER BY CLAUSES">
2635 =head2 delete($table, \%where, \%options)
2637 This takes a table name and optional hashref L<WHERE clause|/WHERE CLAUSES>.
2638 It returns an SQL DELETE statement and list of bind values.
2640 The optional C<\%options> hash reference may contain additional
2641 options to generate the delete SQL. Currently supported options
2648 See the C<returning> option to
2649 L<insert|/insert($table, \@values || \%fieldvals, \%options)>.
2653 =head2 where(\%where, $order)
2655 This is used to generate just the WHERE clause. For example,
2656 if you have an arbitrary data structure and know what the
2657 rest of your SQL is going to look like, but want an easy way
2658 to produce a WHERE clause, use this. It returns an SQL WHERE
2659 clause and list of bind values.
2662 =head2 values(\%data)
2664 This just returns the values from the hash C<%data>, in the same
2665 order that would be returned from any of the other above queries.
2666 Using this allows you to markedly speed up your queries if you
2667 are affecting lots of rows. See below under the L</"PERFORMANCE"> section.
2669 =head2 generate($any, 'number', $of, \@data, $struct, \%types)
2671 Warning: This is an experimental method and subject to change.
2673 This returns arbitrarily generated SQL. It's a really basic shortcut.
2674 It will return two different things, depending on return context:
2676 my($stmt, @bind) = $sql->generate('create table', \$table, \@fields);
2677 my $stmt_and_val = $sql->generate('create table', \$table, \@fields);
2679 These would return the following:
2681 # First calling form
2682 $stmt = "CREATE TABLE test (?, ?)";
2683 @bind = (field1, field2);
2685 # Second calling form
2686 $stmt_and_val = "CREATE TABLE test (field1, field2)";
2688 Depending on what you're trying to do, it's up to you to choose the correct
2689 format. In this example, the second form is what you would want.
2693 $sql->generate('alter session', { nls_date_format => 'MM/YY' });
2697 ALTER SESSION SET nls_date_format = 'MM/YY'
2699 You get the idea. Strings get their case twiddled, but everything
2700 else remains verbatim.
2702 =head1 EXPORTABLE FUNCTIONS
2704 =head2 is_plain_value
2706 Determines if the supplied argument is a plain value as understood by this
2711 =item * The value is C<undef>
2713 =item * The value is a non-reference
2715 =item * The value is an object with stringification overloading
2717 =item * The value is of the form C<< { -value => $anything } >>
2721 On failure returns C<undef>, on success returns a B<scalar> reference
2722 to the original supplied argument.
2728 The stringification overloading detection is rather advanced: it takes
2729 into consideration not only the presence of a C<""> overload, but if that
2730 fails also checks for enabled
2731 L<autogenerated versions of C<"">|overload/Magic Autogeneration>, based
2732 on either C<0+> or C<bool>.
2734 Unfortunately testing in the field indicates that this
2735 detection B<< may tickle a latent bug in perl versions before 5.018 >>,
2736 but only when very large numbers of stringifying objects are involved.
2737 At the time of writing ( Sep 2014 ) there is no clear explanation of
2738 the direct cause, nor is there a manageably small test case that reliably
2739 reproduces the problem.
2741 If you encounter any of the following exceptions in B<random places within
2742 your application stack> - this module may be to blame:
2744 Operation "ne": no method found,
2745 left argument in overloaded package <something>,
2746 right argument in overloaded package <something>
2750 Stub found while resolving method "???" overloading """" in package <something>
2752 If you fall victim to the above - please attempt to reduce the problem
2753 to something that could be sent to the L<SQL::Abstract developers
2754 |DBIx::Class/GETTING HELP/SUPPORT>
2755 (either publicly or privately). As a workaround in the meantime you can
2756 set C<$ENV{SQLA_ISVALUE_IGNORE_AUTOGENERATED_STRINGIFICATION}> to a true
2757 value, which will most likely eliminate your problem (at the expense of
2758 not being able to properly detect exotic forms of stringification).
2760 This notice and environment variable will be removed in a future version,
2761 as soon as the underlying problem is found and a reliable workaround is
2766 =head2 is_literal_value
2768 Determines if the supplied argument is a literal value as understood by this
2773 =item * C<\$sql_string>
2775 =item * C<\[ $sql_string, @bind_values ]>
2779 On failure returns C<undef>, on success returns an B<array> reference
2780 containing the unpacked version of the supplied literal SQL and bind values.
2782 =head1 WHERE CLAUSES
2786 This module uses a variation on the idea from L<DBIx::Abstract>. It
2787 is B<NOT>, repeat I<not> 100% compatible. B<The main logic of this
2788 module is that things in arrays are OR'ed, and things in hashes
2791 The easiest way to explain is to show lots of examples. After
2792 each C<%where> hash shown, it is assumed you used:
2794 my($stmt, @bind) = $sql->where(\%where);
2796 However, note that the C<%where> hash can be used directly in any
2797 of the other functions as well, as described above.
2799 =head2 Key-value pairs
2801 So, let's get started. To begin, a simple hash:
2805 status => 'completed'
2808 Is converted to SQL C<key = val> statements:
2810 $stmt = "WHERE user = ? AND status = ?";
2811 @bind = ('nwiger', 'completed');
2813 One common thing I end up doing is having a list of values that
2814 a field can be in. To do this, simply specify a list inside of
2819 status => ['assigned', 'in-progress', 'pending'];
2822 This simple code will create the following:
2824 $stmt = "WHERE user = ? AND ( status = ? OR status = ? OR status = ? )";
2825 @bind = ('nwiger', 'assigned', 'in-progress', 'pending');
2827 A field associated to an empty arrayref will be considered a
2828 logical false and will generate 0=1.
2830 =head2 Tests for NULL values
2832 If the value part is C<undef> then this is converted to SQL <IS NULL>
2841 $stmt = "WHERE user = ? AND status IS NULL";
2844 To test if a column IS NOT NULL:
2848 status => { '!=', undef },
2851 =head2 Specific comparison operators
2853 If you want to specify a different type of operator for your comparison,
2854 you can use a hashref for a given column:
2858 status => { '!=', 'completed' }
2861 Which would generate:
2863 $stmt = "WHERE user = ? AND status != ?";
2864 @bind = ('nwiger', 'completed');
2866 To test against multiple values, just enclose the values in an arrayref:
2868 status => { '=', ['assigned', 'in-progress', 'pending'] };
2870 Which would give you:
2872 "WHERE status = ? OR status = ? OR status = ?"
2875 The hashref can also contain multiple pairs, in which case it is expanded
2876 into an C<AND> of its elements:
2880 status => { '!=', 'completed', -not_like => 'pending%' }
2883 # Or more dynamically, like from a form
2884 $where{user} = 'nwiger';
2885 $where{status}{'!='} = 'completed';
2886 $where{status}{'-not_like'} = 'pending%';
2888 # Both generate this
2889 $stmt = "WHERE user = ? AND status != ? AND status NOT LIKE ?";
2890 @bind = ('nwiger', 'completed', 'pending%');
2893 To get an OR instead, you can combine it with the arrayref idea:
2897 priority => [ { '=', 2 }, { '>', 5 } ]
2900 Which would generate:
2902 $stmt = "WHERE ( priority = ? OR priority > ? ) AND user = ?";
2903 @bind = ('2', '5', 'nwiger');
2905 If you want to include literal SQL (with or without bind values), just use a
2906 scalar reference or reference to an arrayref as the value:
2909 date_entered => { '>' => \["to_date(?, 'MM/DD/YYYY')", "11/26/2008"] },
2910 date_expires => { '<' => \"now()" }
2913 Which would generate:
2915 $stmt = "WHERE date_entered > to_date(?, 'MM/DD/YYYY') AND date_expires < now()";
2916 @bind = ('11/26/2008');
2919 =head2 Logic and nesting operators
2921 In the example above,
2922 there is a subtle trap if you want to say something like
2923 this (notice the C<AND>):
2925 WHERE priority != ? AND priority != ?
2927 Because, in Perl you I<can't> do this:
2929 priority => { '!=' => 2, '!=' => 1 }
2931 As the second C<!=> key will obliterate the first. The solution
2932 is to use the special C<-modifier> form inside an arrayref:
2934 priority => [ -and => {'!=', 2},
2938 Normally, these would be joined by C<OR>, but the modifier tells it
2939 to use C<AND> instead. (Hint: You can use this in conjunction with the
2940 C<logic> option to C<new()> in order to change the way your queries
2941 work by default.) B<Important:> Note that the C<-modifier> goes
2942 B<INSIDE> the arrayref, as an extra first element. This will
2943 B<NOT> do what you think it might:
2945 priority => -and => [{'!=', 2}, {'!=', 1}] # WRONG!
2947 Here is a quick list of equivalencies, since there is some overlap:
2950 status => {'!=', 'completed', 'not like', 'pending%' }
2951 status => [ -and => {'!=', 'completed'}, {'not like', 'pending%'}]
2954 status => {'=', ['assigned', 'in-progress']}
2955 status => [ -or => {'=', 'assigned'}, {'=', 'in-progress'}]
2956 status => [ {'=', 'assigned'}, {'=', 'in-progress'} ]
2960 =head2 Special operators: IN, BETWEEN, etc.
2962 You can also use the hashref format to compare a list of fields using the
2963 C<IN> comparison operator, by specifying the list as an arrayref:
2966 status => 'completed',
2967 reportid => { -in => [567, 2335, 2] }
2970 Which would generate:
2972 $stmt = "WHERE status = ? AND reportid IN (?,?,?)";
2973 @bind = ('completed', '567', '2335', '2');
2975 The reverse operator C<-not_in> generates SQL C<NOT IN> and is used in
2978 If the argument to C<-in> is an empty array, 'sqlfalse' is generated
2979 (by default: C<1=0>). Similarly, C<< -not_in => [] >> generates
2980 'sqltrue' (by default: C<1=1>).
2982 In addition to the array you can supply a chunk of literal sql or
2983 literal sql with bind:
2986 customer => { -in => \[
2987 'SELECT cust_id FROM cust WHERE balance > ?',
2990 status => { -in => \'SELECT status_codes FROM states' },
2996 customer IN ( SELECT cust_id FROM cust WHERE balance > ? )
2997 AND status IN ( SELECT status_codes FROM states )
3001 Finally, if the argument to C<-in> is not a reference, it will be
3002 treated as a single-element array.
3004 Another pair of operators is C<-between> and C<-not_between>,
3005 used with an arrayref of two values:
3009 completion_date => {
3010 -not_between => ['2002-10-01', '2003-02-06']
3016 WHERE user = ? AND completion_date NOT BETWEEN ( ? AND ? )
3018 Just like with C<-in> all plausible combinations of literal SQL
3022 start0 => { -between => [ 1, 2 ] },
3023 start1 => { -between => \["? AND ?", 1, 2] },
3024 start2 => { -between => \"lower(x) AND upper(y)" },
3025 start3 => { -between => [
3027 \["upper(?)", 'stuff' ],
3034 ( start0 BETWEEN ? AND ? )
3035 AND ( start1 BETWEEN ? AND ? )
3036 AND ( start2 BETWEEN lower(x) AND upper(y) )
3037 AND ( start3 BETWEEN lower(x) AND upper(?) )
3039 @bind = (1, 2, 1, 2, 'stuff');
3042 These are the two builtin "special operators"; but the
3043 list can be expanded: see section L</"SPECIAL OPERATORS"> below.
3045 =head2 Unary operators: bool
3047 If you wish to test against boolean columns or functions within your
3048 database you can use the C<-bool> and C<-not_bool> operators. For
3049 example to test the column C<is_user> being true and the column
3050 C<is_enabled> being false you would use:-
3054 -not_bool => 'is_enabled',
3059 WHERE is_user AND NOT is_enabled
3061 If a more complex combination is required, testing more conditions,
3062 then you should use the and/or operators:-
3067 -not_bool => { two=> { -rlike => 'bar' } },
3068 -not_bool => { three => [ { '=', 2 }, { '>', 5 } ] },
3079 (NOT ( three = ? OR three > ? ))
3082 =head2 Nested conditions, -and/-or prefixes
3084 So far, we've seen how multiple conditions are joined with a top-level
3085 C<AND>. We can change this by putting the different conditions we want in
3086 hashes and then putting those hashes in an array. For example:
3091 status => { -like => ['pending%', 'dispatched'] },
3095 status => 'unassigned',
3099 This data structure would create the following:
3101 $stmt = "WHERE ( user = ? AND ( status LIKE ? OR status LIKE ? ) )
3102 OR ( user = ? AND status = ? ) )";
3103 @bind = ('nwiger', 'pending', 'dispatched', 'robot', 'unassigned');
3106 Clauses in hashrefs or arrayrefs can be prefixed with an C<-and> or C<-or>
3107 to change the logic inside:
3113 -and => [ workhrs => {'>', 20}, geo => 'ASIA' ],
3114 -or => { workhrs => {'<', 50}, geo => 'EURO' },
3121 $stmt = "WHERE ( user = ?
3122 AND ( ( workhrs > ? AND geo = ? )
3123 OR ( workhrs < ? OR geo = ? ) ) )";
3124 @bind = ('nwiger', '20', 'ASIA', '50', 'EURO');
3126 =head3 Algebraic inconsistency, for historical reasons
3128 C<Important note>: when connecting several conditions, the C<-and->|C<-or>
3129 operator goes C<outside> of the nested structure; whereas when connecting
3130 several constraints on one column, the C<-and> operator goes
3131 C<inside> the arrayref. Here is an example combining both features:
3134 -and => [a => 1, b => 2],
3135 -or => [c => 3, d => 4],
3136 e => [-and => {-like => 'foo%'}, {-like => '%bar'} ]
3141 WHERE ( ( ( a = ? AND b = ? )
3142 OR ( c = ? OR d = ? )
3143 OR ( e LIKE ? AND e LIKE ? ) ) )
3145 This difference in syntax is unfortunate but must be preserved for
3146 historical reasons. So be careful: the two examples below would
3147 seem algebraically equivalent, but they are not
3150 { -like => 'foo%' },
3151 { -like => '%bar' },
3153 # yields: WHERE ( ( col LIKE ? AND col LIKE ? ) )
3156 { col => { -like => 'foo%' } },
3157 { col => { -like => '%bar' } },
3159 # yields: WHERE ( ( col LIKE ? OR col LIKE ? ) )
3162 =head2 Literal SQL and value type operators
3164 The basic premise of SQL::Abstract is that in WHERE specifications the "left
3165 side" is a column name and the "right side" is a value (normally rendered as
3166 a placeholder). This holds true for both hashrefs and arrayref pairs as you
3167 see in the L</WHERE CLAUSES> examples above. Sometimes it is necessary to
3168 alter this behavior. There are several ways of doing so.
3172 This is a virtual operator that signals the string to its right side is an
3173 identifier (a column name) and not a value. For example to compare two
3174 columns you would write:
3177 priority => { '<', 2 },
3178 requestor => { -ident => 'submitter' },
3183 $stmt = "WHERE priority < ? AND requestor = submitter";
3186 If you are maintaining legacy code you may see a different construct as
3187 described in L</Deprecated usage of Literal SQL>, please use C<-ident> in new
3192 This is a virtual operator that signals that the construct to its right side
3193 is a value to be passed to DBI. This is for example necessary when you want
3194 to write a where clause against an array (for RDBMS that support such
3195 datatypes). For example:
3198 array => { -value => [1, 2, 3] }
3203 $stmt = 'WHERE array = ?';
3204 @bind = ([1, 2, 3]);
3206 Note that if you were to simply say:
3212 the result would probably not be what you wanted:
3214 $stmt = 'WHERE array = ? OR array = ? OR array = ?';
3219 Finally, sometimes only literal SQL will do. To include a random snippet
3220 of SQL verbatim, you specify it as a scalar reference. Consider this only
3221 as a last resort. Usually there is a better way. For example:
3224 priority => { '<', 2 },
3225 requestor => { -in => \'(SELECT name FROM hitmen)' },
3230 $stmt = "WHERE priority < ? AND requestor IN (SELECT name FROM hitmen)"
3233 Note that in this example, you only get one bind parameter back, since
3234 the verbatim SQL is passed as part of the statement.
3238 Never use untrusted input as a literal SQL argument - this is a massive
3239 security risk (there is no way to check literal snippets for SQL
3240 injections and other nastyness). If you need to deal with untrusted input
3241 use literal SQL with placeholders as described next.
3243 =head3 Literal SQL with placeholders and bind values (subqueries)
3245 If the literal SQL to be inserted has placeholders and bind values,
3246 use a reference to an arrayref (yes this is a double reference --
3247 not so common, but perfectly legal Perl). For example, to find a date
3248 in Postgres you can use something like this:
3251 date_column => \[ "= date '2008-09-30' - ?::integer", 10 ]
3256 $stmt = "WHERE ( date_column = date '2008-09-30' - ?::integer )"
3259 Note that you must pass the bind values in the same format as they are returned
3260 by L<where|/where(\%where, $order)>. This means that if you set L</bindtype>
3261 to C<columns>, you must provide the bind values in the
3262 C<< [ column_meta => value ] >> format, where C<column_meta> is an opaque
3263 scalar value; most commonly the column name, but you can use any scalar value
3264 (including references and blessed references), L<SQL::Abstract> will simply
3265 pass it through intact. So if C<bindtype> is set to C<columns> the above
3266 example will look like:
3269 date_column => \[ "= date '2008-09-30' - ?::integer", [ {} => 10 ] ]
3272 Literal SQL is especially useful for nesting parenthesized clauses in the
3273 main SQL query. Here is a first example:
3275 my ($sub_stmt, @sub_bind) = ("SELECT c1 FROM t1 WHERE c2 < ? AND c3 LIKE ?",
3279 bar => \["IN ($sub_stmt)" => @sub_bind],
3284 $stmt = "WHERE (foo = ? AND bar IN (SELECT c1 FROM t1
3285 WHERE c2 < ? AND c3 LIKE ?))";
3286 @bind = (1234, 100, "foo%");
3288 Other subquery operators, like for example C<"E<gt> ALL"> or C<"NOT IN">,
3289 are expressed in the same way. Of course the C<$sub_stmt> and
3290 its associated bind values can be generated through a former call
3293 my ($sub_stmt, @sub_bind)
3294 = $sql->select("t1", "c1", {c2 => {"<" => 100},
3295 c3 => {-like => "foo%"}});
3298 bar => \["> ALL ($sub_stmt)" => @sub_bind],
3301 In the examples above, the subquery was used as an operator on a column;
3302 but the same principle also applies for a clause within the main C<%where>
3303 hash, like an EXISTS subquery:
3305 my ($sub_stmt, @sub_bind)
3306 = $sql->select("t1", "*", {c1 => 1, c2 => \"> t0.c0"});
3307 my %where = ( -and => [
3309 \["EXISTS ($sub_stmt)" => @sub_bind],
3314 $stmt = "WHERE (foo = ? AND EXISTS (SELECT * FROM t1
3315 WHERE c1 = ? AND c2 > t0.c0))";
3319 Observe that the condition on C<c2> in the subquery refers to
3320 column C<t0.c0> of the main query: this is I<not> a bind
3321 value, so we have to express it through a scalar ref.
3322 Writing C<< c2 => {">" => "t0.c0"} >> would have generated
3323 C<< c2 > ? >> with bind value C<"t0.c0"> ... not exactly
3324 what we wanted here.
3326 Finally, here is an example where a subquery is used
3327 for expressing unary negation:
3329 my ($sub_stmt, @sub_bind)
3330 = $sql->where({age => [{"<" => 10}, {">" => 20}]});
3331 $sub_stmt =~ s/^ where //i; # don't want "WHERE" in the subclause
3333 lname => {like => '%son%'},
3334 \["NOT ($sub_stmt)" => @sub_bind],
3339 $stmt = "lname LIKE ? AND NOT ( age < ? OR age > ? )"
3340 @bind = ('%son%', 10, 20)
3342 =head3 Deprecated usage of Literal SQL
3344 Below are some examples of archaic use of literal SQL. It is shown only as
3345 reference for those who deal with legacy code. Each example has a much
3346 better, cleaner and safer alternative that users should opt for in new code.
3352 my %where = ( requestor => \'IS NOT NULL' )
3354 $stmt = "WHERE requestor IS NOT NULL"
3356 This used to be the way of generating NULL comparisons, before the handling
3357 of C<undef> got formalized. For new code please use the superior syntax as
3358 described in L</Tests for NULL values>.
3362 my %where = ( requestor => \'= submitter' )
3364 $stmt = "WHERE requestor = submitter"
3366 This used to be the only way to compare columns. Use the superior L</-ident>
3367 method for all new code. For example an identifier declared in such a way
3368 will be properly quoted if L</quote_char> is properly set, while the legacy
3369 form will remain as supplied.
3373 my %where = ( is_ready => \"", completed => { '>', '2012-12-21' } )
3375 $stmt = "WHERE completed > ? AND is_ready"
3376 @bind = ('2012-12-21')
3378 Using an empty string literal used to be the only way to express a boolean.
3379 For all new code please use the much more readable
3380 L<-bool|/Unary operators: bool> operator.
3386 These pages could go on for a while, since the nesting of the data
3387 structures this module can handle are pretty much unlimited (the
3388 module implements the C<WHERE> expansion as a recursive function
3389 internally). Your best bet is to "play around" with the module a
3390 little to see how the data structures behave, and choose the best
3391 format for your data based on that.
3393 And of course, all the values above will probably be replaced with
3394 variables gotten from forms or the command line. After all, if you
3395 knew everything ahead of time, you wouldn't have to worry about
3396 dynamically-generating SQL and could just hardwire it into your
3399 =head1 ORDER BY CLAUSES
3401 Some functions take an order by clause. This can either be a scalar (just a
3402 column name), a hashref of C<< { -desc => 'col' } >> or C<< { -asc => 'col' }
3403 >>, a scalarref, an arrayref-ref, or an arrayref of any of the previous
3406 Given | Will Generate
3407 ---------------------------------------------------------------
3409 'colA' | ORDER BY colA
3411 [qw/colA colB/] | ORDER BY colA, colB
3413 {-asc => 'colA'} | ORDER BY colA ASC
3415 {-desc => 'colB'} | ORDER BY colB DESC
3417 ['colA', {-asc => 'colB'}] | ORDER BY colA, colB ASC
3419 { -asc => [qw/colA colB/] } | ORDER BY colA ASC, colB ASC
3421 \'colA DESC' | ORDER BY colA DESC
3423 \[ 'FUNC(colA, ?)', $x ] | ORDER BY FUNC(colA, ?)
3424 | /* ...with $x bound to ? */
3427 { -asc => 'colA' }, | colA ASC,
3428 { -desc => [qw/colB/] }, | colB DESC,
3429 { -asc => [qw/colC colD/] },| colC ASC, colD ASC,
3430 \'colE DESC', | colE DESC,
3431 \[ 'FUNC(colF, ?)', $x ], | FUNC(colF, ?)
3432 ] | /* ...with $x bound to ? */
3433 ===============================================================
3437 =head1 SPECIAL OPERATORS
3439 my $sqlmaker = SQL::Abstract->new(special_ops => [
3443 my ($self, $field, $op, $arg) = @_;
3449 handler => 'method_name',
3453 A "special operator" is a SQL syntactic clause that can be
3454 applied to a field, instead of a usual binary operator.
3457 WHERE field IN (?, ?, ?)
3458 WHERE field BETWEEN ? AND ?
3459 WHERE MATCH(field) AGAINST (?, ?)
3461 Special operators IN and BETWEEN are fairly standard and therefore
3462 are builtin within C<SQL::Abstract> (as the overridable methods
3463 C<_where_field_IN> and C<_where_field_BETWEEN>). For other operators,
3464 like the MATCH .. AGAINST example above which is specific to MySQL,
3465 you can write your own operator handlers - supply a C<special_ops>
3466 argument to the C<new> method. That argument takes an arrayref of
3467 operator definitions; each operator definition is a hashref with two
3474 the regular expression to match the operator
3478 Either a coderef or a plain scalar method name. In both cases
3479 the expected return is C<< ($sql, @bind) >>.
3481 When supplied with a method name, it is simply called on the
3482 L<SQL::Abstract> object as:
3484 $self->$method_name($field, $op, $arg)
3488 $field is the LHS of the operator
3489 $op is the part that matched the handler regex
3492 When supplied with a coderef, it is called as:
3494 $coderef->($self, $field, $op, $arg)
3499 For example, here is an implementation
3500 of the MATCH .. AGAINST syntax for MySQL
3502 my $sqlmaker = SQL::Abstract->new(special_ops => [
3504 # special op for MySql MATCH (field) AGAINST(word1, word2, ...)
3505 {regex => qr/^match$/i,
3507 my ($self, $field, $op, $arg) = @_;
3508 $arg = [$arg] if not ref $arg;
3509 my $label = $self->_quote($field);
3510 my ($placeholder) = $self->_convert('?');
3511 my $placeholders = join ", ", (($placeholder) x @$arg);
3512 my $sql = $self->_sqlcase('match') . " ($label) "
3513 . $self->_sqlcase('against') . " ($placeholders) ";
3514 my @bind = $self->_bindtype($field, @$arg);
3515 return ($sql, @bind);
3522 =head1 UNARY OPERATORS
3524 my $sqlmaker = SQL::Abstract->new(unary_ops => [
3528 my ($self, $op, $arg) = @_;
3534 handler => 'method_name',
3538 A "unary operator" is a SQL syntactic clause that can be
3539 applied to a field - the operator goes before the field
3541 You can write your own operator handlers - supply a C<unary_ops>
3542 argument to the C<new> method. That argument takes an arrayref of
3543 operator definitions; each operator definition is a hashref with two
3550 the regular expression to match the operator
3554 Either a coderef or a plain scalar method name. In both cases
3555 the expected return is C<< $sql >>.
3557 When supplied with a method name, it is simply called on the
3558 L<SQL::Abstract> object as:
3560 $self->$method_name($op, $arg)
3564 $op is the part that matched the handler regex
3565 $arg is the RHS or argument of the operator
3567 When supplied with a coderef, it is called as:
3569 $coderef->($self, $op, $arg)
3577 Thanks to some benchmarking by Mark Stosberg, it turns out that
3578 this module is many orders of magnitude faster than using C<DBIx::Abstract>.
3579 I must admit this wasn't an intentional design issue, but it's a
3580 byproduct of the fact that you get to control your C<DBI> handles
3583 To maximize performance, use a code snippet like the following:
3585 # prepare a statement handle using the first row
3586 # and then reuse it for the rest of the rows
3588 for my $href (@array_of_hashrefs) {
3589 $stmt ||= $sql->insert('table', $href);
3590 $sth ||= $dbh->prepare($stmt);
3591 $sth->execute($sql->values($href));
3594 The reason this works is because the keys in your C<$href> are sorted
3595 internally by B<SQL::Abstract>. Thus, as long as your data retains
3596 the same structure, you only have to generate the SQL the first time
3597 around. On subsequent queries, simply use the C<values> function provided
3598 by this module to return your values in the correct order.
3600 However this depends on the values having the same type - if, for
3601 example, the values of a where clause may either have values
3602 (resulting in sql of the form C<column = ?> with a single bind
3603 value), or alternatively the values might be C<undef> (resulting in
3604 sql of the form C<column IS NULL> with no bind value) then the
3605 caching technique suggested will not work.
3609 If you use my C<CGI::FormBuilder> module at all, you'll hopefully
3610 really like this part (I do, at least). Building up a complex query
3611 can be as simple as the following:
3618 use CGI::FormBuilder;
3621 my $form = CGI::FormBuilder->new(...);
3622 my $sql = SQL::Abstract->new;
3624 if ($form->submitted) {
3625 my $field = $form->field;
3626 my $id = delete $field->{id};
3627 my($stmt, @bind) = $sql->update('table', $field, {id => $id});
3630 Of course, you would still have to connect using C<DBI> to run the
3631 query, but the point is that if you make your form look like your
3632 table, the actual query script can be extremely simplistic.
3634 If you're B<REALLY> lazy (I am), check out C<HTML::QuickTable> for
3635 a fast interface to returning and formatting data. I frequently
3636 use these three modules together to write complex database query
3637 apps in under 50 lines.
3639 =head1 HOW TO CONTRIBUTE
3641 Contributions are always welcome, in all usable forms (we especially
3642 welcome documentation improvements). The delivery methods include git-
3643 or unified-diff formatted patches, GitHub pull requests, or plain bug
3644 reports either via RT or the Mailing list. Contributors are generally
3645 granted full access to the official repository after their first several
3646 patches pass successful review.
3648 This project is maintained in a git repository. The code and related tools are
3649 accessible at the following locations:
3653 =item * Official repo: L<git://git.shadowcat.co.uk/dbsrgits/SQL-Abstract.git>
3655 =item * Official gitweb: L<http://git.shadowcat.co.uk/gitweb/gitweb.cgi?p=dbsrgits/SQL-Abstract.git>
3657 =item * GitHub mirror: L<https://github.com/dbsrgits/sql-abstract>
3659 =item * Authorized committers: L<ssh://dbsrgits@git.shadowcat.co.uk/SQL-Abstract.git>
3665 Version 1.50 was a major internal refactoring of C<SQL::Abstract>.
3666 Great care has been taken to preserve the I<published> behavior
3667 documented in previous versions in the 1.* family; however,
3668 some features that were previously undocumented, or behaved
3669 differently from the documentation, had to be changed in order
3670 to clarify the semantics. Hence, client code that was relying
3671 on some dark areas of C<SQL::Abstract> v1.*
3672 B<might behave differently> in v1.50.
3674 The main changes are:
3680 support for literal SQL through the C<< \ [ $sql, @bind ] >> syntax.
3684 support for the { operator => \"..." } construct (to embed literal SQL)
3688 support for the { operator => \["...", @bind] } construct (to embed literal SQL with bind values)
3692 optional support for L<array datatypes|/"Inserting and Updating Arrays">
3696 defensive programming: check arguments
3700 fixed bug with global logic, which was previously implemented
3701 through global variables yielding side-effects. Prior versions would
3702 interpret C<< [ {cond1, cond2}, [cond3, cond4] ] >>
3703 as C<< "(cond1 AND cond2) OR (cond3 AND cond4)" >>.
3704 Now this is interpreted
3705 as C<< "(cond1 AND cond2) OR (cond3 OR cond4)" >>.
3710 fixed semantics of _bindtype on array args
3714 dropped the C<_anoncopy> of the %where tree. No longer necessary,
3715 we just avoid shifting arrays within that tree.
3719 dropped the C<_modlogic> function
3723 =head1 ACKNOWLEDGEMENTS
3725 There are a number of individuals that have really helped out with
3726 this module. Unfortunately, most of them submitted bugs via CPAN
3727 so I have no idea who they are! But the people I do know are:
3729 Ash Berlin (order_by hash term support)
3730 Matt Trout (DBIx::Class support)
3731 Mark Stosberg (benchmarking)
3732 Chas Owens (initial "IN" operator support)
3733 Philip Collins (per-field SQL functions)
3734 Eric Kolve (hashref "AND" support)
3735 Mike Fragassi (enhancements to "BETWEEN" and "LIKE")
3736 Dan Kubb (support for "quote_char" and "name_sep")
3737 Guillermo Roditi (patch to cleanup "IN" and "BETWEEN", fix and tests for _order_by)
3738 Laurent Dami (internal refactoring, extensible list of special operators, literal SQL)
3739 Norbert Buchmuller (support for literal SQL in hashpair, misc. fixes & tests)
3740 Peter Rabbitson (rewrite of SQLA::Test, misc. fixes & tests)
3741 Oliver Charles (support for "RETURNING" after "INSERT")
3747 L<DBIx::Class>, L<DBIx::Abstract>, L<CGI::FormBuilder>, L<HTML::QuickTable>.
3751 Copyright (c) 2001-2007 Nathan Wiger <nwiger@cpan.org>. All Rights Reserved.
3753 This module is actively maintained by Matt Trout <mst@shadowcatsystems.co.uk>
3755 For support, your best bet is to try the C<DBIx::Class> users mailing list.
3756 While not an official support venue, C<DBIx::Class> makes heavy use of
3757 C<SQL::Abstract>, and as such list members there are very familiar with
3758 how to create queries.
3762 This module is free software; you may copy this under the same
3763 terms as perl itself (either the GNU General Public License or
3764 the Artistic License)