1 package SQL::Abstract; # see doc at end of file
3 # LDNOTE : this code is heavy refactoring from original SQLA.
4 # Several design decisions will need discussion during
5 # the test / diffusion / acceptance phase; those are marked with flag
6 # 'LDNOTE' (note by laurent.dami AT free.fr)
10 use warnings FATAL => 'all';
13 use Data::Query::Constants qw(
14 DQ_IDENTIFIER DQ_OPERATOR DQ_VALUE DQ_LITERAL DQ_JOIN DQ_SELECT DQ_ORDER
16 use Data::Query::ExprHelpers qw(perl_scalar_value);
18 #======================================================================
20 #======================================================================
22 our $VERSION = '1.72';
24 # This would confuse some packagers
25 $VERSION = eval $VERSION if $VERSION =~ /_/; # numify for warning-free dev releases
29 # special operators (-in, -between). May be extended/overridden by user.
30 # See section WHERE: BUILTIN SPECIAL OPERATORS below for implementation
31 my @BUILTIN_SPECIAL_OPS = (
32 {regex => qr/^ (?: not \s )? between $/ix, handler => '_where_field_BETWEEN'},
33 {regex => qr/^ (?: not \s )? in $/ix, handler => '_where_field_IN'},
34 {regex => qr/^ ident $/ix, handler => '_where_op_IDENT'},
35 {regex => qr/^ value $/ix, handler => '_where_op_VALUE'},
38 # unaryish operators - key maps to handler
39 my @BUILTIN_UNARY_OPS = (
40 # the digits are backcompat stuff
41 { regex => qr/^ and (?: [_\s]? \d+ )? $/xi, handler => '_where_op_ANDOR' },
42 { regex => qr/^ or (?: [_\s]? \d+ )? $/xi, handler => '_where_op_ANDOR' },
43 { regex => qr/^ nest (?: [_\s]? \d+ )? $/xi, handler => '_where_op_NEST' },
44 { regex => qr/^ (?: not \s )? bool $/xi, handler => '_where_op_BOOL' },
45 { regex => qr/^ ident $/xi, handler => '_where_op_IDENT' },
46 { regex => qr/^ value $/ix, handler => '_where_op_VALUE' },
49 #======================================================================
50 # DEBUGGING AND ERROR REPORTING
51 #======================================================================
54 return unless $_[0]->{debug}; shift; # a little faster
55 my $func = (caller(1))[3];
56 warn "[$func] ", @_, "\n";
60 my($func) = (caller(1))[3];
61 Carp::carp "[$func] Warning: ", @_;
65 my($func) = (caller(1))[3];
66 Carp::croak "[$func] Fatal: ", @_;
70 #======================================================================
72 #======================================================================
76 my $class = ref($self) || $self;
77 my %opt = (ref $_[0] eq 'HASH') ? %{$_[0]} : @_;
79 # choose our case by keeping an option around
80 delete $opt{case} if $opt{case} && $opt{case} ne 'lower';
82 # default logic for interpreting arrayrefs
83 $opt{logic} = $opt{logic} ? uc $opt{logic} : 'OR';
85 # how to return bind vars
86 # LDNOTE: changed nwiger code : why this 'delete' ??
87 # $opt{bindtype} ||= delete($opt{bind_type}) || 'normal';
88 $opt{bindtype} ||= 'normal';
90 # default comparison is "=", but can be overridden
93 # try to recognize which are the 'equality' and 'unequality' ops
94 # (temporary quickfix, should go through a more seasoned API)
95 $opt{equality_op} = qr/^(\Q$opt{cmp}\E|is|(is\s+)?like)$/i;
96 $opt{inequality_op} = qr/^(!=|<>|(is\s+)?not(\s+like)?)$/i;
99 $opt{sqltrue} ||= '1=1';
100 $opt{sqlfalse} ||= '0=1';
103 $opt{special_ops} ||= [];
104 # regexes are applied in order, thus push after user-defines
105 push @{$opt{special_ops}}, @BUILTIN_SPECIAL_OPS;
108 $opt{unary_ops} ||= [];
109 push @{$opt{unary_ops}}, @BUILTIN_UNARY_OPS;
111 # rudimentary saniy-check for user supplied bits treated as functions/operators
112 # If a purported function matches this regular expression, an exception is thrown.
113 # Literal SQL is *NOT* subject to this check, only functions (and column names
114 # when quoting is not in effect)
117 # need to guard against ()'s in column names too, but this will break tons of
118 # hacks... ideas anyone?
119 $opt{injection_guard} ||= qr/
125 $opt{name_sep} ||= '.';
127 $opt{renderer} ||= do {
128 require Data::Query::Renderer::SQL::Naive;
129 my ($always, $chars);
130 for ($opt{quote_char}) {
131 $chars = defined() ? (ref() ? $_ : [$_]) : ['',''];
134 Data::Query::Renderer::SQL::Naive->new({
135 quote_chars => $chars, always_quote => $always,
139 return bless \%opt, $class;
143 my ($self, $dq) = @_;
144 my ($sql, @bind) = @{$self->{renderer}->render($dq)};
146 ($self->{bindtype} eq 'normal'
147 ? ($sql, map $_->{value}, @bind)
148 : ($sql, map [ $_->{meta}, $_->{value} ], @bind)
154 my ($self, $literal) = @_;
156 ($literal, @bind) = @$literal if ref($literal) eq 'ARRAY';
161 (@bind ? (values => [ $self->_bind_to_dq(@bind) ]) : ()),
165 sub _literal_with_prepend_to_dq {
166 my ($self, $prepend, $literal) = @_;
168 $self->_literal_to_dq(
169 [ join(' ', $prepend, $literal->[0]), @{$literal}[1..$#$literal] ]
172 $self->_literal_to_dq(
173 join(' ', $prepend, $literal)
179 my ($self, @bind) = @_;
181 $self->{bindtype} eq 'normal'
182 ? map perl_scalar_value($_), @bind
184 $self->_assert_bindval_matches_bindtype(@bind);
185 map perl_scalar_value(reverse @$_), @bind
190 my ($self, $value) = @_;
191 perl_scalar_value($value);
195 my ($self, $ident) = @_;
197 type => DQ_IDENTIFIER,
198 elements => [ split /\Q$self->{name_sep}/, $ident ],
202 sub _assert_pass_injection_guard {
203 if ($_[1] =~ $_[0]->{injection_guard}) {
204 my $class = ref $_[0];
205 puke "Possible SQL injection attempt '$_[1]'. If this is indeed a part of the "
206 . "desired SQL use literal SQL ( \'...' or \[ '...' ] ) or supply your own "
207 . "{injection_guard} attribute to ${class}->new()"
212 #======================================================================
214 #======================================================================
218 my $table = $self->_table(shift);
219 my $data = shift || return;
222 my $method = $self->_METHOD_FOR_refkind("_insert", $data);
223 my ($sql, @bind) = $self->$method($data);
224 $sql = join " ", $self->_sqlcase('insert into'), $table, $sql;
226 if ($options->{returning}) {
227 my ($s, @b) = $self->_insert_returning ($options);
232 return wantarray ? ($sql, @bind) : $sql;
235 sub _insert_returning {
236 my ($self, $options) = @_;
238 my $f = $options->{returning};
240 my $fieldlist = $self->_SWITCH_refkind($f, {
241 ARRAYREF => sub {join ', ', map { $self->_quote($_) } @$f;},
242 SCALAR => sub {$self->_quote($f)},
243 SCALARREF => sub {$$f},
245 return $self->_sqlcase(' returning ') . $fieldlist;
248 sub _insert_HASHREF { # explicit list of fields and then values
249 my ($self, $data) = @_;
251 my @fields = sort keys %$data;
253 my ($sql, @bind) = $self->_insert_values($data);
256 $_ = $self->_quote($_) foreach @fields;
257 $sql = "( ".join(", ", @fields).") ".$sql;
259 return ($sql, @bind);
262 sub _insert_ARRAYREF { # just generate values(?,?) part (no list of fields)
263 my ($self, $data) = @_;
265 # no names (arrayref) so can't generate bindtype
266 $self->{bindtype} ne 'columns'
267 or belch "can't do 'columns' bindtype when called with arrayref";
269 # fold the list of values into a hash of column name - value pairs
270 # (where the column names are artificially generated, and their
271 # lexicographical ordering keep the ordering of the original list)
272 my $i = "a"; # incremented values will be in lexicographical order
273 my $data_in_hash = { map { ($i++ => $_) } @$data };
275 return $self->_insert_values($data_in_hash);
278 sub _insert_ARRAYREFREF { # literal SQL with bind
279 my ($self, $data) = @_;
281 my ($sql, @bind) = @${$data};
282 $self->_assert_bindval_matches_bindtype(@bind);
284 return ($sql, @bind);
288 sub _insert_SCALARREF { # literal SQL without bind
289 my ($self, $data) = @_;
295 my ($self, $data) = @_;
297 my (@values, @all_bind);
298 foreach my $column (sort keys %$data) {
299 my $v = $data->{$column};
301 $self->_SWITCH_refkind($v, {
304 if ($self->{array_datatypes}) { # if array datatype are activated
306 push @all_bind, $self->_bindtype($column, $v);
308 else { # else literal SQL with bind
309 my ($sql, @bind) = @$v;
310 $self->_assert_bindval_matches_bindtype(@bind);
312 push @all_bind, @bind;
316 ARRAYREFREF => sub { # literal SQL with bind
317 my ($sql, @bind) = @${$v};
318 $self->_assert_bindval_matches_bindtype(@bind);
320 push @all_bind, @bind;
323 # THINK : anything useful to do with a HASHREF ?
324 HASHREF => sub { # (nothing, but old SQLA passed it through)
325 #TODO in SQLA >= 2.0 it will die instead
326 belch "HASH ref as bind value in insert is not supported";
328 push @all_bind, $self->_bindtype($column, $v);
331 SCALARREF => sub { # literal SQL without bind
335 SCALAR_or_UNDEF => sub {
337 push @all_bind, $self->_bindtype($column, $v);
344 my $sql = $self->_sqlcase('values')." ( ".join(", ", @values)." )";
345 return ($sql, @all_bind);
350 #======================================================================
352 #======================================================================
357 my $table = $self->_table(shift);
358 my $data = shift || return;
361 # first build the 'SET' part of the sql statement
362 my (@set, @all_bind);
363 puke "Unsupported data type specified to \$sql->update"
364 unless ref $data eq 'HASH';
366 for my $k (sort keys %$data) {
369 my $label = $self->_quote($k);
371 $self->_SWITCH_refkind($v, {
373 if ($self->{array_datatypes}) { # array datatype
374 push @set, "$label = ?";
375 push @all_bind, $self->_bindtype($k, $v);
377 else { # literal SQL with bind
378 my ($sql, @bind) = @$v;
379 $self->_assert_bindval_matches_bindtype(@bind);
380 push @set, "$label = $sql";
381 push @all_bind, @bind;
384 ARRAYREFREF => sub { # literal SQL with bind
385 my ($sql, @bind) = @${$v};
386 $self->_assert_bindval_matches_bindtype(@bind);
387 push @set, "$label = $sql";
388 push @all_bind, @bind;
390 SCALARREF => sub { # literal SQL without bind
391 push @set, "$label = $$v";
394 my ($op, $arg, @rest) = %$v;
396 puke 'Operator calls in update must be in the form { -op => $arg }'
397 if (@rest or not $op =~ /^\-(.+)/);
399 local $self->{_nested_func_lhs} = $k;
400 my ($sql, @bind) = $self->_where_unary_op ($1, $arg);
402 push @set, "$label = $sql";
403 push @all_bind, @bind;
405 SCALAR_or_UNDEF => sub {
406 push @set, "$label = ?";
407 push @all_bind, $self->_bindtype($k, $v);
413 my $sql = $self->_sqlcase('update') . " $table " . $self->_sqlcase('set ')
417 my($where_sql, @where_bind) = $self->where($where);
419 push @all_bind, @where_bind;
422 return wantarray ? ($sql, @all_bind) : $sql;
428 #======================================================================
430 #======================================================================
436 my $fields = shift || '*';
440 my($where_sql, @bind) = $self->where($where, $order);
442 my $sql = $self->_render_dq({
445 map $self->_ident_to_dq($_),
446 ref($fields) eq 'ARRAY' ? @$fields : $fields
448 from => $self->_table_to_dq($table),
453 return wantarray ? ($sql, @bind) : $sql;
456 #======================================================================
458 #======================================================================
463 my $table = $self->_table(shift);
467 my($where_sql, @bind) = $self->where($where);
468 my $sql = $self->_sqlcase('delete from') . " $table" . $where_sql;
470 return wantarray ? ($sql, @bind) : $sql;
474 #======================================================================
476 #======================================================================
480 # Finally, a separate routine just to handle WHERE clauses
482 my ($self, $where, $order) = @_;
485 my ($sql, @bind) = $self->_recurse_where($where);
486 $sql = $sql ? $self->_sqlcase(' where ') . "( $sql )" : '';
490 $sql .= $self->_order_by($order);
493 return wantarray ? ($sql, @bind) : $sql;
498 my ($self, $where, $logic) = @_;
500 return $self->_render_dq($self->_where_to_dq($where, $logic));
504 my ($self, $where, $logic) = @_;
506 # dispatch on appropriate method according to refkind of $where
507 my $method = $self->_METHOD_FOR_refkind("_where_to_dq", $where);
509 return $self->$method($where, $logic);
512 sub _where_to_dq_ARRAYREF {
513 my ($self, $where, $logic) = @_;
515 $logic = uc($logic || 'OR');
516 $logic eq 'AND' or $logic eq 'OR' or puke "unknown logic: $logic";
518 return unless @$where;
520 my ($first, @rest) = @$where;
522 return $self->_where_to_dq($first) unless @rest;
526 $self->_where_hashpair_to_dq($first => shift(@rest));
528 $self->_where_to_dq($first);
532 return $self->_where_to_dq_ARRAYREF(\@rest, $logic) unless $first_dq;
536 operator => { 'SQL.Naive' => $logic },
537 args => [ $first_dq, $self->_where_to_dq_ARRAYREF(\@rest, $logic) ]
541 sub _where_to_dq_ARRAYREFREF {
542 my ($self, $where) = @_;
543 return $self->_literal_to_dq($$where);
546 sub _where_to_dq_SCALARREF {
547 my ($self, $where) = @_;
548 return $self->_literal_to_dq($$where);
551 sub _where_to_dq_HASHREF {
552 my ($self, $where, $logic) = @_;
554 $logic = uc($logic || 'AND');
557 $self->_where_hashpair_to_dq($_ => $where->{$_})
560 return $dq[0] unless @dq > 1;
562 my $final = pop(@dq);
564 foreach my $dq (reverse @dq) {
567 operator => { 'SQL.Naive' => $logic },
568 args => [ $dq, $final ]
575 sub _where_to_dq_SCALAR {
576 shift->_value_to_dq(@_);
579 sub _where_op_IDENT {
581 my ($op, $rhs) = splice @_, -2;
583 puke "-$op takes a single scalar argument (a quotable identifier)";
586 # in case we are called as a top level special op (no '=')
589 $_ = $self->_convert($self->_quote($_)) for ($lhs, $rhs);
597 sub _where_op_VALUE {
599 my ($op, $rhs) = splice @_, -2;
601 # in case we are called as a top level special op (no '=')
606 ($lhs || $self->{_nested_func_lhs}),
613 $self->_convert($self->_quote($lhs)) . ' = ' . $self->_convert('?'),
617 $self->_convert('?'),
623 sub _where_hashpair_to_dq {
624 my ($self, $k, $v) = @_;
628 if ($op eq 'AND' or $op eq 'OR') {
629 return $self->_where_to_dq($v, $op);
630 } elsif ($op eq 'NEST') {
631 return $self->_where_to_dq($v);
632 } elsif ($op eq 'NOT') {
635 operator => { 'SQL.Naive' => 'NOT' },
636 args => [ $self->_where_to_dq($v) ]
638 } elsif ($op eq 'BOOL') {
639 return ref($v) ? $self->_where_to_dq($v) : $self->_ident_to_dq($v);
640 } elsif ($op eq 'NOT_BOOL') {
643 operator => { 'SQL.Naive' => 'NOT' },
644 args => [ ref($v) ? $self->_where_to_dq($v) : $self->_ident_to_dq($v) ]
647 die "Not done this bit yet";
650 if (ref($v) eq 'ARRAY') {
652 return $self->_literal_to_dq($self->{sqlfalse});
653 } elsif (defined($v->[0]) && $v->[0] =~ /-(and|or)/i) {
654 return $self->_where_to_dq_ARRAYREF([
655 map +{ $k => $_ }, @{$v}[1..$#$v]
658 return $self->_where_to_dq_ARRAYREF([
659 map +{ $k => $_ }, @$v
661 } elsif (ref($v) eq 'SCALAR' or (ref($v) eq 'REF' and ref($$v) eq 'ARRAY')) {
662 return $self->_literal_with_prepend_to_dq($k, $$v);
664 my ($op, $rhs) = do {
665 if (ref($v) eq 'HASH') {
667 return $self->_where_to_dq_ARRAYREF([
668 map +{ $k => { $_ => $v->{$_} } }, keys %$v
671 (uc((keys %$v)[0]), (values %$v)[0]);
676 s/^-//, s/_/ /g for $op;
677 if ($op eq 'BETWEEN' or $op eq 'IN' or $op eq 'NOT IN' or $op eq 'NOT BETWEEN') {
678 if (ref($rhs) ne 'ARRAY') {
679 return $self->_literal_with_prepend_to_dq("$k $op", $$rhs);
683 operator => { 'SQL.Naive' => $op },
684 args => [ $self->_ident_to_dq($k), map $self->_where_to_dq($_), @$rhs ]
686 } elsif ($op =~ s/^NOT (?!LIKE)//) {
687 return $self->_where_hashpair_to_dq(-not => { $k => { $op => $rhs } });
688 } elsif (!defined($rhs)) {
690 if ($op eq '=' or $op eq 'LIKE') {
692 } elsif ($op eq '!=') {
695 die "Can't do undef -> NULL transform for operator ${op}";
700 operator => { 'SQL.Naive' => $null_op },
701 args => [ $self->_ident_to_dq($k) ]
704 if (ref($rhs) eq 'ARRAY') {
706 return $self->_literal_to_dq(
707 $op eq '!=' ? $self->{sqltrue} : $self->{sqlfalse}
709 } elsif (defined($rhs->[0]) and $rhs->[0] =~ /-(and|or)/i) {
710 return $self->_where_to_dq_ARRAYREF([
711 map +{ $k => { $op => $_ } }, @{$rhs}[1..$#$rhs]
714 return $self->_where_to_dq_ARRAYREF([
715 map +{ $k => { $op => $_ } }, @$rhs
720 operator => { 'SQL.Naive' => $op },
721 args => [ $self->_ident_to_dq($k), $self->_where_to_dq($rhs) ]
726 #======================================================================
728 #======================================================================
731 my ($self, $arg) = @_;
732 if (my $dq = $self->_order_by_to_dq($arg)) {
733 # SQLA generates ' ORDER BY foo'. The hilarity.
735 ? do { my @r = $self->_render_dq($dq); $r[0] = ' '.$r[0]; @r }
736 : ' '.$self->_render_dq($dq);
742 sub _order_by_to_dq {
743 my ($self, $arg, $dir) = @_;
749 ($dir ? (direction => $dir) : ()),
753 $dq->{by} = $self->_ident_to_dq($arg);
754 } elsif (ref($arg) eq 'ARRAY') {
756 local our $Order_Inner unless our $Order_Recursing;
757 local $Order_Recursing = 1;
759 foreach my $member (@$arg) {
761 my $next = $self->_order_by_to_dq($member, $dir);
763 $inner->{from} = $next if $inner;
764 $inner = $Order_Inner || $next;
766 $Order_Inner = $inner;
768 } elsif (ref($arg) eq 'REF' and ref($$arg) eq 'ARRAY') {
769 $dq->{by} = $self->_literal_to_dq($$arg);
770 } elsif (ref($arg) eq 'SCALAR') {
771 $dq->{by} = $self->_literal_to_dq($$arg);
772 } elsif (ref($arg) eq 'HASH') {
773 my ($key, $val, @rest) = %$arg;
777 if (@rest or not $key =~ /^-(desc|asc)/i) {
778 puke "hash passed to _order_by must have exactly one key (-desc or -asc)";
781 return $self->_order_by_to_dq($val, $dir);
783 die "Can't handle $arg in _order_by_to_dq";
788 #======================================================================
789 # DATASOURCE (FOR NOW, JUST PLAIN TABLE OR LIST OF TABLES)
790 #======================================================================
793 my ($self, $from) = @_;
794 $self->_render_dq($self->_table_to_dq($from));
798 my ($self, $from) = @_;
799 $self->_SWITCH_refkind($from, {
801 die "Empty FROM list" unless my @f = @$from;
802 my $dq = $self->_ident_to_dq(shift @f);
803 while (my $x = shift @f) {
806 join => [ $dq, $self->_ident_to_dq($x) ]
811 SCALAR => sub { $self->_ident_to_dq($from) },
823 #======================================================================
825 #======================================================================
827 # highly optimized, as it's called way too often
829 # my ($self, $label) = @_;
831 return '' unless defined $_[1];
832 return ${$_[1]} if ref($_[1]) eq 'SCALAR';
834 unless ($_[0]->{quote_char}) {
835 $_[0]->_assert_pass_injection_guard($_[1]);
839 my $qref = ref $_[0]->{quote_char};
842 ($l, $r) = ( $_[0]->{quote_char}, $_[0]->{quote_char} );
844 elsif ($qref eq 'ARRAY') {
845 ($l, $r) = @{$_[0]->{quote_char}};
848 puke "Unsupported quote_char format: $_[0]->{quote_char}";
851 # parts containing * are naturally unquoted
852 return join( $_[0]->{name_sep}||'', map
853 { $_ eq '*' ? $_ : $l . $_ . $r }
854 ( $_[0]->{name_sep} ? split (/\Q$_[0]->{name_sep}\E/, $_[1] ) : $_[1] )
859 # Conversion, if applicable
861 #my ($self, $arg) = @_;
863 # LDNOTE : modified the previous implementation below because
864 # it was not consistent : the first "return" is always an array,
865 # the second "return" is context-dependent. Anyway, _convert
866 # seems always used with just a single argument, so make it a
868 # return @_ unless $self->{convert};
869 # my $conv = $self->_sqlcase($self->{convert});
870 # my @ret = map { $conv.'('.$_.')' } @_;
871 # return wantarray ? @ret : $ret[0];
872 if ($_[0]->{convert}) {
873 return $_[0]->_sqlcase($_[0]->{convert}) .'(' . $_[1] . ')';
880 #my ($self, $col, @vals) = @_;
882 #LDNOTE : changed original implementation below because it did not make
883 # sense when bindtype eq 'columns' and @vals > 1.
884 # return $self->{bindtype} eq 'columns' ? [ $col, @vals ] : @vals;
886 # called often - tighten code
887 return $_[0]->{bindtype} eq 'columns'
888 ? map {[$_[1], $_]} @_[2 .. $#_]
893 # Dies if any element of @bind is not in [colname => value] format
894 # if bindtype is 'columns'.
895 sub _assert_bindval_matches_bindtype {
896 # my ($self, @bind) = @_;
898 if ($self->{bindtype} eq 'columns') {
900 if (!defined $_ || ref($_) ne 'ARRAY' || @$_ != 2) {
901 puke "bindtype 'columns' selected, you need to pass: [column_name => bind_value]"
907 sub _join_sql_clauses {
908 my ($self, $logic, $clauses_aref, $bind_aref) = @_;
910 if (@$clauses_aref > 1) {
911 my $join = " " . $self->_sqlcase($logic) . " ";
912 my $sql = '( ' . join($join, @$clauses_aref) . ' )';
913 return ($sql, @$bind_aref);
915 elsif (@$clauses_aref) {
916 return ($clauses_aref->[0], @$bind_aref); # no parentheses
919 return (); # if no SQL, ignore @$bind_aref
924 # Fix SQL case, if so requested
926 # LDNOTE: if $self->{case} is true, then it contains 'lower', so we
927 # don't touch the argument ... crooked logic, but let's not change it!
928 return $_[0]->{case} ? $_[1] : uc($_[1]);
932 #======================================================================
933 # DISPATCHING FROM REFKIND
934 #======================================================================
937 my ($self, $data) = @_;
939 return 'UNDEF' unless defined $data;
941 # blessed objects are treated like scalars
942 my $ref = (Scalar::Util::blessed $data) ? '' : ref $data;
944 return 'SCALAR' unless $ref;
947 while ($ref eq 'REF') {
949 $ref = (Scalar::Util::blessed $data) ? '' : ref $data;
953 return ($ref||'SCALAR') . ('REF' x $n_steps);
957 my ($self, $data) = @_;
958 my @try = ($self->_refkind($data));
959 push @try, 'SCALAR_or_UNDEF' if $try[0] eq 'SCALAR' || $try[0] eq 'UNDEF';
960 push @try, 'FALLBACK';
964 sub _METHOD_FOR_refkind {
965 my ($self, $meth_prefix, $data) = @_;
968 for (@{$self->_try_refkind($data)}) {
969 $method = $self->can($meth_prefix."_".$_)
973 return $method || puke "cannot dispatch on '$meth_prefix' for ".$self->_refkind($data);
977 sub _SWITCH_refkind {
978 my ($self, $data, $dispatch_table) = @_;
981 for (@{$self->_try_refkind($data)}) {
982 $coderef = $dispatch_table->{$_}
986 puke "no dispatch entry for ".$self->_refkind($data)
995 #======================================================================
996 # VALUES, GENERATE, AUTOLOAD
997 #======================================================================
999 # LDNOTE: original code from nwiger, didn't touch code in that section
1000 # I feel the AUTOLOAD stuff should not be the default, it should
1001 # only be activated on explicit demand by user.
1005 my $data = shift || return;
1006 puke "Argument to ", __PACKAGE__, "->values must be a \\%hash"
1007 unless ref $data eq 'HASH';
1010 foreach my $k ( sort keys %$data ) {
1011 my $v = $data->{$k};
1012 $self->_SWITCH_refkind($v, {
1014 if ($self->{array_datatypes}) { # array datatype
1015 push @all_bind, $self->_bindtype($k, $v);
1017 else { # literal SQL with bind
1018 my ($sql, @bind) = @$v;
1019 $self->_assert_bindval_matches_bindtype(@bind);
1020 push @all_bind, @bind;
1023 ARRAYREFREF => sub { # literal SQL with bind
1024 my ($sql, @bind) = @${$v};
1025 $self->_assert_bindval_matches_bindtype(@bind);
1026 push @all_bind, @bind;
1028 SCALARREF => sub { # literal SQL without bind
1030 SCALAR_or_UNDEF => sub {
1031 push @all_bind, $self->_bindtype($k, $v);
1042 my(@sql, @sqlq, @sqlv);
1046 if ($ref eq 'HASH') {
1047 for my $k (sort keys %$_) {
1050 my $label = $self->_quote($k);
1051 if ($r eq 'ARRAY') {
1052 # literal SQL with bind
1053 my ($sql, @bind) = @$v;
1054 $self->_assert_bindval_matches_bindtype(@bind);
1055 push @sqlq, "$label = $sql";
1057 } elsif ($r eq 'SCALAR') {
1058 # literal SQL without bind
1059 push @sqlq, "$label = $$v";
1061 push @sqlq, "$label = ?";
1062 push @sqlv, $self->_bindtype($k, $v);
1065 push @sql, $self->_sqlcase('set'), join ', ', @sqlq;
1066 } elsif ($ref eq 'ARRAY') {
1067 # unlike insert(), assume these are ONLY the column names, i.e. for SQL
1070 if ($r eq 'ARRAY') { # literal SQL with bind
1071 my ($sql, @bind) = @$v;
1072 $self->_assert_bindval_matches_bindtype(@bind);
1075 } elsif ($r eq 'SCALAR') { # literal SQL without bind
1076 # embedded literal SQL
1083 push @sql, '(' . join(', ', @sqlq) . ')';
1084 } elsif ($ref eq 'SCALAR') {
1088 # strings get case twiddled
1089 push @sql, $self->_sqlcase($_);
1093 my $sql = join ' ', @sql;
1095 # this is pretty tricky
1096 # if ask for an array, return ($stmt, @bind)
1097 # otherwise, s/?/shift @sqlv/ to put it inline
1099 return ($sql, @sqlv);
1101 1 while $sql =~ s/\?/my $d = shift(@sqlv);
1102 ref $d ? $d->[1] : $d/e;
1111 # This allows us to check for a local, then _form, attr
1113 my($name) = $AUTOLOAD =~ /.*::(.+)/;
1114 return $self->generate($name, @_);
1125 SQL::Abstract - Generate SQL from Perl data structures
1131 my $sql = SQL::Abstract->new;
1133 my($stmt, @bind) = $sql->select($table, \@fields, \%where, \@order);
1135 my($stmt, @bind) = $sql->insert($table, \%fieldvals || \@values);
1137 my($stmt, @bind) = $sql->update($table, \%fieldvals, \%where);
1139 my($stmt, @bind) = $sql->delete($table, \%where);
1141 # Then, use these in your DBI statements
1142 my $sth = $dbh->prepare($stmt);
1143 $sth->execute(@bind);
1145 # Just generate the WHERE clause
1146 my($stmt, @bind) = $sql->where(\%where, \@order);
1148 # Return values in the same order, for hashed queries
1149 # See PERFORMANCE section for more details
1150 my @bind = $sql->values(\%fieldvals);
1154 This module was inspired by the excellent L<DBIx::Abstract>.
1155 However, in using that module I found that what I really wanted
1156 to do was generate SQL, but still retain complete control over my
1157 statement handles and use the DBI interface. So, I set out to
1158 create an abstract SQL generation module.
1160 While based on the concepts used by L<DBIx::Abstract>, there are
1161 several important differences, especially when it comes to WHERE
1162 clauses. I have modified the concepts used to make the SQL easier
1163 to generate from Perl data structures and, IMO, more intuitive.
1164 The underlying idea is for this module to do what you mean, based
1165 on the data structures you provide it. The big advantage is that
1166 you don't have to modify your code every time your data changes,
1167 as this module figures it out.
1169 To begin with, an SQL INSERT is as easy as just specifying a hash
1170 of C<key=value> pairs:
1173 name => 'Jimbo Bobson',
1174 phone => '123-456-7890',
1175 address => '42 Sister Lane',
1176 city => 'St. Louis',
1177 state => 'Louisiana',
1180 The SQL can then be generated with this:
1182 my($stmt, @bind) = $sql->insert('people', \%data);
1184 Which would give you something like this:
1186 $stmt = "INSERT INTO people
1187 (address, city, name, phone, state)
1188 VALUES (?, ?, ?, ?, ?)";
1189 @bind = ('42 Sister Lane', 'St. Louis', 'Jimbo Bobson',
1190 '123-456-7890', 'Louisiana');
1192 These are then used directly in your DBI code:
1194 my $sth = $dbh->prepare($stmt);
1195 $sth->execute(@bind);
1197 =head2 Inserting and Updating Arrays
1199 If your database has array types (like for example Postgres),
1200 activate the special option C<< array_datatypes => 1 >>
1201 when creating the C<SQL::Abstract> object.
1202 Then you may use an arrayref to insert and update database array types:
1204 my $sql = SQL::Abstract->new(array_datatypes => 1);
1206 planets => [qw/Mercury Venus Earth Mars/]
1209 my($stmt, @bind) = $sql->insert('solar_system', \%data);
1213 $stmt = "INSERT INTO solar_system (planets) VALUES (?)"
1215 @bind = (['Mercury', 'Venus', 'Earth', 'Mars']);
1218 =head2 Inserting and Updating SQL
1220 In order to apply SQL functions to elements of your C<%data> you may
1221 specify a reference to an arrayref for the given hash value. For example,
1222 if you need to execute the Oracle C<to_date> function on a value, you can
1223 say something like this:
1227 date_entered => \["to_date(?,'MM/DD/YYYY')", "03/02/2003"],
1230 The first value in the array is the actual SQL. Any other values are
1231 optional and would be included in the bind values array. This gives
1234 my($stmt, @bind) = $sql->insert('people', \%data);
1236 $stmt = "INSERT INTO people (name, date_entered)
1237 VALUES (?, to_date(?,'MM/DD/YYYY'))";
1238 @bind = ('Bill', '03/02/2003');
1240 An UPDATE is just as easy, all you change is the name of the function:
1242 my($stmt, @bind) = $sql->update('people', \%data);
1244 Notice that your C<%data> isn't touched; the module will generate
1245 the appropriately quirky SQL for you automatically. Usually you'll
1246 want to specify a WHERE clause for your UPDATE, though, which is
1247 where handling C<%where> hashes comes in handy...
1249 =head2 Complex where statements
1251 This module can generate pretty complicated WHERE statements
1252 easily. For example, simple C<key=value> pairs are taken to mean
1253 equality, and if you want to see if a field is within a set
1254 of values, you can use an arrayref. Let's say we wanted to
1255 SELECT some data based on this criteria:
1258 requestor => 'inna',
1259 worker => ['nwiger', 'rcwe', 'sfz'],
1260 status => { '!=', 'completed' }
1263 my($stmt, @bind) = $sql->select('tickets', '*', \%where);
1265 The above would give you something like this:
1267 $stmt = "SELECT * FROM tickets WHERE
1268 ( requestor = ? ) AND ( status != ? )
1269 AND ( worker = ? OR worker = ? OR worker = ? )";
1270 @bind = ('inna', 'completed', 'nwiger', 'rcwe', 'sfz');
1272 Which you could then use in DBI code like so:
1274 my $sth = $dbh->prepare($stmt);
1275 $sth->execute(@bind);
1281 The functions are simple. There's one for each major SQL operation,
1282 and a constructor you use first. The arguments are specified in a
1283 similar order to each function (table, then fields, then a where
1284 clause) to try and simplify things.
1289 =head2 new(option => 'value')
1291 The C<new()> function takes a list of options and values, and returns
1292 a new B<SQL::Abstract> object which can then be used to generate SQL
1293 through the methods below. The options accepted are:
1299 If set to 'lower', then SQL will be generated in all lowercase. By
1300 default SQL is generated in "textbook" case meaning something like:
1302 SELECT a_field FROM a_table WHERE some_field LIKE '%someval%'
1304 Any setting other than 'lower' is ignored.
1308 This determines what the default comparison operator is. By default
1309 it is C<=>, meaning that a hash like this:
1311 %where = (name => 'nwiger', email => 'nate@wiger.org');
1313 Will generate SQL like this:
1315 WHERE name = 'nwiger' AND email = 'nate@wiger.org'
1317 However, you may want loose comparisons by default, so if you set
1318 C<cmp> to C<like> you would get SQL such as:
1320 WHERE name like 'nwiger' AND email like 'nate@wiger.org'
1322 You can also override the comparsion on an individual basis - see
1323 the huge section on L</"WHERE CLAUSES"> at the bottom.
1325 =item sqltrue, sqlfalse
1327 Expressions for inserting boolean values within SQL statements.
1328 By default these are C<1=1> and C<1=0>. They are used
1329 by the special operators C<-in> and C<-not_in> for generating
1330 correct SQL even when the argument is an empty array (see below).
1334 This determines the default logical operator for multiple WHERE
1335 statements in arrays or hashes. If absent, the default logic is "or"
1336 for arrays, and "and" for hashes. This means that a WHERE
1340 event_date => {'>=', '2/13/99'},
1341 event_date => {'<=', '4/24/03'},
1344 will generate SQL like this:
1346 WHERE event_date >= '2/13/99' OR event_date <= '4/24/03'
1348 This is probably not what you want given this query, though (look
1349 at the dates). To change the "OR" to an "AND", simply specify:
1351 my $sql = SQL::Abstract->new(logic => 'and');
1353 Which will change the above C<WHERE> to:
1355 WHERE event_date >= '2/13/99' AND event_date <= '4/24/03'
1357 The logic can also be changed locally by inserting
1358 a modifier in front of an arrayref :
1360 @where = (-and => [event_date => {'>=', '2/13/99'},
1361 event_date => {'<=', '4/24/03'} ]);
1363 See the L</"WHERE CLAUSES"> section for explanations.
1367 This will automatically convert comparisons using the specified SQL
1368 function for both column and value. This is mostly used with an argument
1369 of C<upper> or C<lower>, so that the SQL will have the effect of
1370 case-insensitive "searches". For example, this:
1372 $sql = SQL::Abstract->new(convert => 'upper');
1373 %where = (keywords => 'MaKe iT CAse inSeNSItive');
1375 Will turn out the following SQL:
1377 WHERE upper(keywords) like upper('MaKe iT CAse inSeNSItive')
1379 The conversion can be C<upper()>, C<lower()>, or any other SQL function
1380 that can be applied symmetrically to fields (actually B<SQL::Abstract> does
1381 not validate this option; it will just pass through what you specify verbatim).
1385 This is a kludge because many databases suck. For example, you can't
1386 just bind values using DBI's C<execute()> for Oracle C<CLOB> or C<BLOB> fields.
1387 Instead, you have to use C<bind_param()>:
1389 $sth->bind_param(1, 'reg data');
1390 $sth->bind_param(2, $lots, {ora_type => ORA_CLOB});
1392 The problem is, B<SQL::Abstract> will normally just return a C<@bind> array,
1393 which loses track of which field each slot refers to. Fear not.
1395 If you specify C<bindtype> in new, you can determine how C<@bind> is returned.
1396 Currently, you can specify either C<normal> (default) or C<columns>. If you
1397 specify C<columns>, you will get an array that looks like this:
1399 my $sql = SQL::Abstract->new(bindtype => 'columns');
1400 my($stmt, @bind) = $sql->insert(...);
1403 [ 'column1', 'value1' ],
1404 [ 'column2', 'value2' ],
1405 [ 'column3', 'value3' ],
1408 You can then iterate through this manually, using DBI's C<bind_param()>.
1410 $sth->prepare($stmt);
1413 my($col, $data) = @$_;
1414 if ($col eq 'details' || $col eq 'comments') {
1415 $sth->bind_param($i, $data, {ora_type => ORA_CLOB});
1416 } elsif ($col eq 'image') {
1417 $sth->bind_param($i, $data, {ora_type => ORA_BLOB});
1419 $sth->bind_param($i, $data);
1423 $sth->execute; # execute without @bind now
1425 Now, why would you still use B<SQL::Abstract> if you have to do this crap?
1426 Basically, the advantage is still that you don't have to care which fields
1427 are or are not included. You could wrap that above C<for> loop in a simple
1428 sub called C<bind_fields()> or something and reuse it repeatedly. You still
1429 get a layer of abstraction over manual SQL specification.
1431 Note that if you set L</bindtype> to C<columns>, the C<\[$sql, @bind]>
1432 construct (see L</Literal SQL with placeholders and bind values (subqueries)>)
1433 will expect the bind values in this format.
1437 This is the character that a table or column name will be quoted
1438 with. By default this is an empty string, but you could set it to
1439 the character C<`>, to generate SQL like this:
1441 SELECT `a_field` FROM `a_table` WHERE `some_field` LIKE '%someval%'
1443 Alternatively, you can supply an array ref of two items, the first being the left
1444 hand quote character, and the second the right hand quote character. For
1445 example, you could supply C<['[',']']> for SQL Server 2000 compliant quotes
1446 that generates SQL like this:
1448 SELECT [a_field] FROM [a_table] WHERE [some_field] LIKE '%someval%'
1450 Quoting is useful if you have tables or columns names that are reserved
1451 words in your database's SQL dialect.
1455 This is the character that separates a table and column name. It is
1456 necessary to specify this when the C<quote_char> option is selected,
1457 so that tables and column names can be individually quoted like this:
1459 SELECT `table`.`one_field` FROM `table` WHERE `table`.`other_field` = 1
1461 =item injection_guard
1463 A regular expression C<qr/.../> that is applied to any C<-function> and unquoted
1464 column name specified in a query structure. This is a safety mechanism to avoid
1465 injection attacks when mishandling user input e.g.:
1467 my %condition_as_column_value_pairs = get_values_from_user();
1468 $sqla->select( ... , \%condition_as_column_value_pairs );
1470 If the expression matches an exception is thrown. Note that literal SQL
1471 supplied via C<\'...'> or C<\['...']> is B<not> checked in any way.
1473 Defaults to checking for C<;> and the C<GO> keyword (TransactSQL)
1475 =item array_datatypes
1477 When this option is true, arrayrefs in INSERT or UPDATE are
1478 interpreted as array datatypes and are passed directly
1480 When this option is false, arrayrefs are interpreted
1481 as literal SQL, just like refs to arrayrefs
1482 (but this behavior is for backwards compatibility; when writing
1483 new queries, use the "reference to arrayref" syntax
1489 Takes a reference to a list of "special operators"
1490 to extend the syntax understood by L<SQL::Abstract>.
1491 See section L</"SPECIAL OPERATORS"> for details.
1495 Takes a reference to a list of "unary operators"
1496 to extend the syntax understood by L<SQL::Abstract>.
1497 See section L</"UNARY OPERATORS"> for details.
1503 =head2 insert($table, \@values || \%fieldvals, \%options)
1505 This is the simplest function. You simply give it a table name
1506 and either an arrayref of values or hashref of field/value pairs.
1507 It returns an SQL INSERT statement and a list of bind values.
1508 See the sections on L</"Inserting and Updating Arrays"> and
1509 L</"Inserting and Updating SQL"> for information on how to insert
1510 with those data types.
1512 The optional C<\%options> hash reference may contain additional
1513 options to generate the insert SQL. Currently supported options
1520 Takes either a scalar of raw SQL fields, or an array reference of
1521 field names, and adds on an SQL C<RETURNING> statement at the end.
1522 This allows you to return data generated by the insert statement
1523 (such as row IDs) without performing another C<SELECT> statement.
1524 Note, however, this is not part of the SQL standard and may not
1525 be supported by all database engines.
1529 =head2 update($table, \%fieldvals, \%where)
1531 This takes a table, hashref of field/value pairs, and an optional
1532 hashref L<WHERE clause|/WHERE CLAUSES>. It returns an SQL UPDATE function and a list
1534 See the sections on L</"Inserting and Updating Arrays"> and
1535 L</"Inserting and Updating SQL"> for information on how to insert
1536 with those data types.
1538 =head2 select($source, $fields, $where, $order)
1540 This returns a SQL SELECT statement and associated list of bind values, as
1541 specified by the arguments :
1547 Specification of the 'FROM' part of the statement.
1548 The argument can be either a plain scalar (interpreted as a table
1549 name, will be quoted), or an arrayref (interpreted as a list
1550 of table names, joined by commas, quoted), or a scalarref
1551 (literal table name, not quoted), or a ref to an arrayref
1552 (list of literal table names, joined by commas, not quoted).
1556 Specification of the list of fields to retrieve from
1558 The argument can be either an arrayref (interpreted as a list
1559 of field names, will be joined by commas and quoted), or a
1560 plain scalar (literal SQL, not quoted).
1561 Please observe that this API is not as flexible as for
1562 the first argument C<$table>, for backwards compatibility reasons.
1566 Optional argument to specify the WHERE part of the query.
1567 The argument is most often a hashref, but can also be
1568 an arrayref or plain scalar --
1569 see section L<WHERE clause|/"WHERE CLAUSES"> for details.
1573 Optional argument to specify the ORDER BY part of the query.
1574 The argument can be a scalar, a hashref or an arrayref
1575 -- see section L<ORDER BY clause|/"ORDER BY CLAUSES">
1581 =head2 delete($table, \%where)
1583 This takes a table name and optional hashref L<WHERE clause|/WHERE CLAUSES>.
1584 It returns an SQL DELETE statement and list of bind values.
1586 =head2 where(\%where, \@order)
1588 This is used to generate just the WHERE clause. For example,
1589 if you have an arbitrary data structure and know what the
1590 rest of your SQL is going to look like, but want an easy way
1591 to produce a WHERE clause, use this. It returns an SQL WHERE
1592 clause and list of bind values.
1595 =head2 values(\%data)
1597 This just returns the values from the hash C<%data>, in the same
1598 order that would be returned from any of the other above queries.
1599 Using this allows you to markedly speed up your queries if you
1600 are affecting lots of rows. See below under the L</"PERFORMANCE"> section.
1602 =head2 generate($any, 'number', $of, \@data, $struct, \%types)
1604 Warning: This is an experimental method and subject to change.
1606 This returns arbitrarily generated SQL. It's a really basic shortcut.
1607 It will return two different things, depending on return context:
1609 my($stmt, @bind) = $sql->generate('create table', \$table, \@fields);
1610 my $stmt_and_val = $sql->generate('create table', \$table, \@fields);
1612 These would return the following:
1614 # First calling form
1615 $stmt = "CREATE TABLE test (?, ?)";
1616 @bind = (field1, field2);
1618 # Second calling form
1619 $stmt_and_val = "CREATE TABLE test (field1, field2)";
1621 Depending on what you're trying to do, it's up to you to choose the correct
1622 format. In this example, the second form is what you would want.
1626 $sql->generate('alter session', { nls_date_format => 'MM/YY' });
1630 ALTER SESSION SET nls_date_format = 'MM/YY'
1632 You get the idea. Strings get their case twiddled, but everything
1633 else remains verbatim.
1635 =head1 WHERE CLAUSES
1639 This module uses a variation on the idea from L<DBIx::Abstract>. It
1640 is B<NOT>, repeat I<not> 100% compatible. B<The main logic of this
1641 module is that things in arrays are OR'ed, and things in hashes
1644 The easiest way to explain is to show lots of examples. After
1645 each C<%where> hash shown, it is assumed you used:
1647 my($stmt, @bind) = $sql->where(\%where);
1649 However, note that the C<%where> hash can be used directly in any
1650 of the other functions as well, as described above.
1652 =head2 Key-value pairs
1654 So, let's get started. To begin, a simple hash:
1658 status => 'completed'
1661 Is converted to SQL C<key = val> statements:
1663 $stmt = "WHERE user = ? AND status = ?";
1664 @bind = ('nwiger', 'completed');
1666 One common thing I end up doing is having a list of values that
1667 a field can be in. To do this, simply specify a list inside of
1672 status => ['assigned', 'in-progress', 'pending'];
1675 This simple code will create the following:
1677 $stmt = "WHERE user = ? AND ( status = ? OR status = ? OR status = ? )";
1678 @bind = ('nwiger', 'assigned', 'in-progress', 'pending');
1680 A field associated to an empty arrayref will be considered a
1681 logical false and will generate 0=1.
1683 =head2 Tests for NULL values
1685 If the value part is C<undef> then this is converted to SQL <IS NULL>
1694 $stmt = "WHERE user = ? AND status IS NULL";
1697 To test if a column IS NOT NULL:
1701 status => { '!=', undef },
1704 =head2 Specific comparison operators
1706 If you want to specify a different type of operator for your comparison,
1707 you can use a hashref for a given column:
1711 status => { '!=', 'completed' }
1714 Which would generate:
1716 $stmt = "WHERE user = ? AND status != ?";
1717 @bind = ('nwiger', 'completed');
1719 To test against multiple values, just enclose the values in an arrayref:
1721 status => { '=', ['assigned', 'in-progress', 'pending'] };
1723 Which would give you:
1725 "WHERE status = ? OR status = ? OR status = ?"
1728 The hashref can also contain multiple pairs, in which case it is expanded
1729 into an C<AND> of its elements:
1733 status => { '!=', 'completed', -not_like => 'pending%' }
1736 # Or more dynamically, like from a form
1737 $where{user} = 'nwiger';
1738 $where{status}{'!='} = 'completed';
1739 $where{status}{'-not_like'} = 'pending%';
1741 # Both generate this
1742 $stmt = "WHERE user = ? AND status != ? AND status NOT LIKE ?";
1743 @bind = ('nwiger', 'completed', 'pending%');
1746 To get an OR instead, you can combine it with the arrayref idea:
1750 priority => [ { '=', 2 }, { '>', 5 } ]
1753 Which would generate:
1755 $stmt = "WHERE ( priority = ? OR priority > ? ) AND user = ?";
1756 @bind = ('2', '5', 'nwiger');
1758 If you want to include literal SQL (with or without bind values), just use a
1759 scalar reference or array reference as the value:
1762 date_entered => { '>' => \["to_date(?, 'MM/DD/YYYY')", "11/26/2008"] },
1763 date_expires => { '<' => \"now()" }
1766 Which would generate:
1768 $stmt = "WHERE date_entered > "to_date(?, 'MM/DD/YYYY') AND date_expires < now()";
1769 @bind = ('11/26/2008');
1772 =head2 Logic and nesting operators
1774 In the example above,
1775 there is a subtle trap if you want to say something like
1776 this (notice the C<AND>):
1778 WHERE priority != ? AND priority != ?
1780 Because, in Perl you I<can't> do this:
1782 priority => { '!=', 2, '!=', 1 }
1784 As the second C<!=> key will obliterate the first. The solution
1785 is to use the special C<-modifier> form inside an arrayref:
1787 priority => [ -and => {'!=', 2},
1791 Normally, these would be joined by C<OR>, but the modifier tells it
1792 to use C<AND> instead. (Hint: You can use this in conjunction with the
1793 C<logic> option to C<new()> in order to change the way your queries
1794 work by default.) B<Important:> Note that the C<-modifier> goes
1795 B<INSIDE> the arrayref, as an extra first element. This will
1796 B<NOT> do what you think it might:
1798 priority => -and => [{'!=', 2}, {'!=', 1}] # WRONG!
1800 Here is a quick list of equivalencies, since there is some overlap:
1803 status => {'!=', 'completed', 'not like', 'pending%' }
1804 status => [ -and => {'!=', 'completed'}, {'not like', 'pending%'}]
1807 status => {'=', ['assigned', 'in-progress']}
1808 status => [ -or => {'=', 'assigned'}, {'=', 'in-progress'}]
1809 status => [ {'=', 'assigned'}, {'=', 'in-progress'} ]
1813 =head2 Special operators : IN, BETWEEN, etc.
1815 You can also use the hashref format to compare a list of fields using the
1816 C<IN> comparison operator, by specifying the list as an arrayref:
1819 status => 'completed',
1820 reportid => { -in => [567, 2335, 2] }
1823 Which would generate:
1825 $stmt = "WHERE status = ? AND reportid IN (?,?,?)";
1826 @bind = ('completed', '567', '2335', '2');
1828 The reverse operator C<-not_in> generates SQL C<NOT IN> and is used in
1831 If the argument to C<-in> is an empty array, 'sqlfalse' is generated
1832 (by default : C<1=0>). Similarly, C<< -not_in => [] >> generates
1833 'sqltrue' (by default : C<1=1>).
1835 In addition to the array you can supply a chunk of literal sql or
1836 literal sql with bind:
1839 customer => { -in => \[
1840 'SELECT cust_id FROM cust WHERE balance > ?',
1843 status => { -in => \'SELECT status_codes FROM states' },
1849 customer IN ( SELECT cust_id FROM cust WHERE balance > ? )
1850 AND status IN ( SELECT status_codes FROM states )
1856 Another pair of operators is C<-between> and C<-not_between>,
1857 used with an arrayref of two values:
1861 completion_date => {
1862 -not_between => ['2002-10-01', '2003-02-06']
1868 WHERE user = ? AND completion_date NOT BETWEEN ( ? AND ? )
1870 Just like with C<-in> all plausible combinations of literal SQL
1874 start0 => { -between => [ 1, 2 ] },
1875 start1 => { -between => \["? AND ?", 1, 2] },
1876 start2 => { -between => \"lower(x) AND upper(y)" },
1877 start3 => { -between => [
1879 \["upper(?)", 'stuff' ],
1886 ( start0 BETWEEN ? AND ? )
1887 AND ( start1 BETWEEN ? AND ? )
1888 AND ( start2 BETWEEN lower(x) AND upper(y) )
1889 AND ( start3 BETWEEN lower(x) AND upper(?) )
1891 @bind = (1, 2, 1, 2, 'stuff');
1894 These are the two builtin "special operators"; but the
1895 list can be expanded : see section L</"SPECIAL OPERATORS"> below.
1897 =head2 Unary operators: bool
1899 If you wish to test against boolean columns or functions within your
1900 database you can use the C<-bool> and C<-not_bool> operators. For
1901 example to test the column C<is_user> being true and the column
1902 C<is_enabled> being false you would use:-
1906 -not_bool => 'is_enabled',
1911 WHERE is_user AND NOT is_enabled
1913 If a more complex combination is required, testing more conditions,
1914 then you should use the and/or operators:-
1921 -not_bool => 'four',
1927 WHERE one AND two AND three AND NOT four
1930 =head2 Nested conditions, -and/-or prefixes
1932 So far, we've seen how multiple conditions are joined with a top-level
1933 C<AND>. We can change this by putting the different conditions we want in
1934 hashes and then putting those hashes in an array. For example:
1939 status => { -like => ['pending%', 'dispatched'] },
1943 status => 'unassigned',
1947 This data structure would create the following:
1949 $stmt = "WHERE ( user = ? AND ( status LIKE ? OR status LIKE ? ) )
1950 OR ( user = ? AND status = ? ) )";
1951 @bind = ('nwiger', 'pending', 'dispatched', 'robot', 'unassigned');
1954 Clauses in hashrefs or arrayrefs can be prefixed with an C<-and> or C<-or>
1955 to change the logic inside :
1961 -and => [ workhrs => {'>', 20}, geo => 'ASIA' ],
1962 -or => { workhrs => {'<', 50}, geo => 'EURO' },
1969 WHERE ( user = ? AND (
1970 ( workhrs > ? AND geo = ? )
1971 OR ( workhrs < ? OR geo = ? )
1974 =head3 Algebraic inconsistency, for historical reasons
1976 C<Important note>: when connecting several conditions, the C<-and->|C<-or>
1977 operator goes C<outside> of the nested structure; whereas when connecting
1978 several constraints on one column, the C<-and> operator goes
1979 C<inside> the arrayref. Here is an example combining both features :
1982 -and => [a => 1, b => 2],
1983 -or => [c => 3, d => 4],
1984 e => [-and => {-like => 'foo%'}, {-like => '%bar'} ]
1989 WHERE ( ( ( a = ? AND b = ? )
1990 OR ( c = ? OR d = ? )
1991 OR ( e LIKE ? AND e LIKE ? ) ) )
1993 This difference in syntax is unfortunate but must be preserved for
1994 historical reasons. So be careful : the two examples below would
1995 seem algebraically equivalent, but they are not
1997 {col => [-and => {-like => 'foo%'}, {-like => '%bar'}]}
1998 # yields : WHERE ( ( col LIKE ? AND col LIKE ? ) )
2000 [-and => {col => {-like => 'foo%'}, {col => {-like => '%bar'}}]]
2001 # yields : WHERE ( ( col LIKE ? OR col LIKE ? ) )
2004 =head2 Literal SQL and value type operators
2006 The basic premise of SQL::Abstract is that in WHERE specifications the "left
2007 side" is a column name and the "right side" is a value (normally rendered as
2008 a placeholder). This holds true for both hashrefs and arrayref pairs as you
2009 see in the L</WHERE CLAUSES> examples above. Sometimes it is necessary to
2010 alter this behavior. There are several ways of doing so.
2014 This is a virtual operator that signals the string to its right side is an
2015 identifier (a column name) and not a value. For example to compare two
2016 columns you would write:
2019 priority => { '<', 2 },
2020 requestor => { -ident => 'submitter' },
2025 $stmt = "WHERE priority < ? AND requestor = submitter";
2028 If you are maintaining legacy code you may see a different construct as
2029 described in L</Deprecated usage of Literal SQL>, please use C<-ident> in new
2034 This is a virtual operator that signals that the construct to its right side
2035 is a value to be passed to DBI. This is for example necessary when you want
2036 to write a where clause against an array (for RDBMS that support such
2037 datatypes). For example:
2040 array => { -value => [1, 2, 3] }
2045 $stmt = 'WHERE array = ?';
2046 @bind = ([1, 2, 3]);
2048 Note that if you were to simply say:
2054 the result would porbably be not what you wanted:
2056 $stmt = 'WHERE array = ? OR array = ? OR array = ?';
2061 Finally, sometimes only literal SQL will do. To include a random snippet
2062 of SQL verbatim, you specify it as a scalar reference. Consider this only
2063 as a last resort. Usually there is a better way. For example:
2066 priority => { '<', 2 },
2067 requestor => { -in => \'(SELECT name FROM hitmen)' },
2072 $stmt = "WHERE priority < ? AND requestor IN (SELECT name FROM hitmen)"
2075 Note that in this example, you only get one bind parameter back, since
2076 the verbatim SQL is passed as part of the statement.
2080 Never use untrusted input as a literal SQL argument - this is a massive
2081 security risk (there is no way to check literal snippets for SQL
2082 injections and other nastyness). If you need to deal with untrusted input
2083 use literal SQL with placeholders as described next.
2085 =head3 Literal SQL with placeholders and bind values (subqueries)
2087 If the literal SQL to be inserted has placeholders and bind values,
2088 use a reference to an arrayref (yes this is a double reference --
2089 not so common, but perfectly legal Perl). For example, to find a date
2090 in Postgres you can use something like this:
2093 date_column => \[q/= date '2008-09-30' - ?::integer/, 10/]
2098 $stmt = "WHERE ( date_column = date '2008-09-30' - ?::integer )"
2101 Note that you must pass the bind values in the same format as they are returned
2102 by L</where>. That means that if you set L</bindtype> to C<columns>, you must
2103 provide the bind values in the C<< [ column_meta => value ] >> format, where
2104 C<column_meta> is an opaque scalar value; most commonly the column name, but
2105 you can use any scalar value (including references and blessed references),
2106 L<SQL::Abstract> will simply pass it through intact. So if C<bindtype> is set
2107 to C<columns> the above example will look like:
2110 date_column => \[q/= date '2008-09-30' - ?::integer/, [ dummy => 10 ]/]
2113 Literal SQL is especially useful for nesting parenthesized clauses in the
2114 main SQL query. Here is a first example :
2116 my ($sub_stmt, @sub_bind) = ("SELECT c1 FROM t1 WHERE c2 < ? AND c3 LIKE ?",
2120 bar => \["IN ($sub_stmt)" => @sub_bind],
2125 $stmt = "WHERE (foo = ? AND bar IN (SELECT c1 FROM t1
2126 WHERE c2 < ? AND c3 LIKE ?))";
2127 @bind = (1234, 100, "foo%");
2129 Other subquery operators, like for example C<"E<gt> ALL"> or C<"NOT IN">,
2130 are expressed in the same way. Of course the C<$sub_stmt> and
2131 its associated bind values can be generated through a former call
2134 my ($sub_stmt, @sub_bind)
2135 = $sql->select("t1", "c1", {c2 => {"<" => 100},
2136 c3 => {-like => "foo%"}});
2139 bar => \["> ALL ($sub_stmt)" => @sub_bind],
2142 In the examples above, the subquery was used as an operator on a column;
2143 but the same principle also applies for a clause within the main C<%where>
2144 hash, like an EXISTS subquery :
2146 my ($sub_stmt, @sub_bind)
2147 = $sql->select("t1", "*", {c1 => 1, c2 => \"> t0.c0"});
2148 my %where = ( -and => [
2150 \["EXISTS ($sub_stmt)" => @sub_bind],
2155 $stmt = "WHERE (foo = ? AND EXISTS (SELECT * FROM t1
2156 WHERE c1 = ? AND c2 > t0.c0))";
2160 Observe that the condition on C<c2> in the subquery refers to
2161 column C<t0.c0> of the main query : this is I<not> a bind
2162 value, so we have to express it through a scalar ref.
2163 Writing C<< c2 => {">" => "t0.c0"} >> would have generated
2164 C<< c2 > ? >> with bind value C<"t0.c0"> ... not exactly
2165 what we wanted here.
2167 Finally, here is an example where a subquery is used
2168 for expressing unary negation:
2170 my ($sub_stmt, @sub_bind)
2171 = $sql->where({age => [{"<" => 10}, {">" => 20}]});
2172 $sub_stmt =~ s/^ where //i; # don't want "WHERE" in the subclause
2174 lname => {like => '%son%'},
2175 \["NOT ($sub_stmt)" => @sub_bind],
2180 $stmt = "lname LIKE ? AND NOT ( age < ? OR age > ? )"
2181 @bind = ('%son%', 10, 20)
2183 =head3 Deprecated usage of Literal SQL
2185 Below are some examples of archaic use of literal SQL. It is shown only as
2186 reference for those who deal with legacy code. Each example has a much
2187 better, cleaner and safer alternative that users should opt for in new code.
2193 my %where = ( requestor => \'IS NOT NULL' )
2195 $stmt = "WHERE requestor IS NOT NULL"
2197 This used to be the way of generating NULL comparisons, before the handling
2198 of C<undef> got formalized. For new code please use the superior syntax as
2199 described in L</Tests for NULL values>.
2203 my %where = ( requestor => \'= submitter' )
2205 $stmt = "WHERE requestor = submitter"
2207 This used to be the only way to compare columns. Use the superior L</-ident>
2208 method for all new code. For example an identifier declared in such a way
2209 will be properly quoted if L</quote_char> is properly set, while the legacy
2210 form will remain as supplied.
2214 my %where = ( is_ready => \"", completed => { '>', '2012-12-21' } )
2216 $stmt = "WHERE completed > ? AND is_ready"
2217 @bind = ('2012-12-21')
2219 Using an empty string literal used to be the only way to express a boolean.
2220 For all new code please use the much more readable
2221 L<-bool|/Unary operators: bool> operator.
2227 These pages could go on for a while, since the nesting of the data
2228 structures this module can handle are pretty much unlimited (the
2229 module implements the C<WHERE> expansion as a recursive function
2230 internally). Your best bet is to "play around" with the module a
2231 little to see how the data structures behave, and choose the best
2232 format for your data based on that.
2234 And of course, all the values above will probably be replaced with
2235 variables gotten from forms or the command line. After all, if you
2236 knew everything ahead of time, you wouldn't have to worry about
2237 dynamically-generating SQL and could just hardwire it into your
2240 =head1 ORDER BY CLAUSES
2242 Some functions take an order by clause. This can either be a scalar (just a
2243 column name,) a hash of C<< { -desc => 'col' } >> or C<< { -asc => 'col' } >>,
2244 or an array of either of the two previous forms. Examples:
2246 Given | Will Generate
2247 ----------------------------------------------------------
2249 \'colA DESC' | ORDER BY colA DESC
2251 'colA' | ORDER BY colA
2253 [qw/colA colB/] | ORDER BY colA, colB
2255 {-asc => 'colA'} | ORDER BY colA ASC
2257 {-desc => 'colB'} | ORDER BY colB DESC
2259 ['colA', {-asc => 'colB'}] | ORDER BY colA, colB ASC
2261 { -asc => [qw/colA colB/] } | ORDER BY colA ASC, colB ASC
2264 { -asc => 'colA' }, | ORDER BY colA ASC, colB DESC,
2265 { -desc => [qw/colB/], | colC ASC, colD ASC
2266 { -asc => [qw/colC colD/],|
2268 ===========================================================
2272 =head1 SPECIAL OPERATORS
2274 my $sqlmaker = SQL::Abstract->new(special_ops => [
2278 my ($self, $field, $op, $arg) = @_;
2284 handler => 'method_name',
2288 A "special operator" is a SQL syntactic clause that can be
2289 applied to a field, instead of a usual binary operator.
2292 WHERE field IN (?, ?, ?)
2293 WHERE field BETWEEN ? AND ?
2294 WHERE MATCH(field) AGAINST (?, ?)
2296 Special operators IN and BETWEEN are fairly standard and therefore
2297 are builtin within C<SQL::Abstract> (as the overridable methods
2298 C<_where_field_IN> and C<_where_field_BETWEEN>). For other operators,
2299 like the MATCH .. AGAINST example above which is specific to MySQL,
2300 you can write your own operator handlers - supply a C<special_ops>
2301 argument to the C<new> method. That argument takes an arrayref of
2302 operator definitions; each operator definition is a hashref with two
2309 the regular expression to match the operator
2313 Either a coderef or a plain scalar method name. In both cases
2314 the expected return is C<< ($sql, @bind) >>.
2316 When supplied with a method name, it is simply called on the
2317 L<SQL::Abstract/> object as:
2319 $self->$method_name ($field, $op, $arg)
2323 $op is the part that matched the handler regex
2324 $field is the LHS of the operator
2327 When supplied with a coderef, it is called as:
2329 $coderef->($self, $field, $op, $arg)
2334 For example, here is an implementation
2335 of the MATCH .. AGAINST syntax for MySQL
2337 my $sqlmaker = SQL::Abstract->new(special_ops => [
2339 # special op for MySql MATCH (field) AGAINST(word1, word2, ...)
2340 {regex => qr/^match$/i,
2342 my ($self, $field, $op, $arg) = @_;
2343 $arg = [$arg] if not ref $arg;
2344 my $label = $self->_quote($field);
2345 my ($placeholder) = $self->_convert('?');
2346 my $placeholders = join ", ", (($placeholder) x @$arg);
2347 my $sql = $self->_sqlcase('match') . " ($label) "
2348 . $self->_sqlcase('against') . " ($placeholders) ";
2349 my @bind = $self->_bindtype($field, @$arg);
2350 return ($sql, @bind);
2357 =head1 UNARY OPERATORS
2359 my $sqlmaker = SQL::Abstract->new(unary_ops => [
2363 my ($self, $op, $arg) = @_;
2369 handler => 'method_name',
2373 A "unary operator" is a SQL syntactic clause that can be
2374 applied to a field - the operator goes before the field
2376 You can write your own operator handlers - supply a C<unary_ops>
2377 argument to the C<new> method. That argument takes an arrayref of
2378 operator definitions; each operator definition is a hashref with two
2385 the regular expression to match the operator
2389 Either a coderef or a plain scalar method name. In both cases
2390 the expected return is C<< $sql >>.
2392 When supplied with a method name, it is simply called on the
2393 L<SQL::Abstract/> object as:
2395 $self->$method_name ($op, $arg)
2399 $op is the part that matched the handler regex
2400 $arg is the RHS or argument of the operator
2402 When supplied with a coderef, it is called as:
2404 $coderef->($self, $op, $arg)
2412 Thanks to some benchmarking by Mark Stosberg, it turns out that
2413 this module is many orders of magnitude faster than using C<DBIx::Abstract>.
2414 I must admit this wasn't an intentional design issue, but it's a
2415 byproduct of the fact that you get to control your C<DBI> handles
2418 To maximize performance, use a code snippet like the following:
2420 # prepare a statement handle using the first row
2421 # and then reuse it for the rest of the rows
2423 for my $href (@array_of_hashrefs) {
2424 $stmt ||= $sql->insert('table', $href);
2425 $sth ||= $dbh->prepare($stmt);
2426 $sth->execute($sql->values($href));
2429 The reason this works is because the keys in your C<$href> are sorted
2430 internally by B<SQL::Abstract>. Thus, as long as your data retains
2431 the same structure, you only have to generate the SQL the first time
2432 around. On subsequent queries, simply use the C<values> function provided
2433 by this module to return your values in the correct order.
2435 However this depends on the values having the same type - if, for
2436 example, the values of a where clause may either have values
2437 (resulting in sql of the form C<column = ?> with a single bind
2438 value), or alternatively the values might be C<undef> (resulting in
2439 sql of the form C<column IS NULL> with no bind value) then the
2440 caching technique suggested will not work.
2444 If you use my C<CGI::FormBuilder> module at all, you'll hopefully
2445 really like this part (I do, at least). Building up a complex query
2446 can be as simple as the following:
2450 use CGI::FormBuilder;
2453 my $form = CGI::FormBuilder->new(...);
2454 my $sql = SQL::Abstract->new;
2456 if ($form->submitted) {
2457 my $field = $form->field;
2458 my $id = delete $field->{id};
2459 my($stmt, @bind) = $sql->update('table', $field, {id => $id});
2462 Of course, you would still have to connect using C<DBI> to run the
2463 query, but the point is that if you make your form look like your
2464 table, the actual query script can be extremely simplistic.
2466 If you're B<REALLY> lazy (I am), check out C<HTML::QuickTable> for
2467 a fast interface to returning and formatting data. I frequently
2468 use these three modules together to write complex database query
2469 apps in under 50 lines.
2475 =item * gitweb: L<http://git.shadowcat.co.uk/gitweb/gitweb.cgi?p=dbsrgits/SQL-Abstract.git>
2477 =item * git: L<git://git.shadowcat.co.uk/dbsrgits/SQL-Abstract.git>
2483 Version 1.50 was a major internal refactoring of C<SQL::Abstract>.
2484 Great care has been taken to preserve the I<published> behavior
2485 documented in previous versions in the 1.* family; however,
2486 some features that were previously undocumented, or behaved
2487 differently from the documentation, had to be changed in order
2488 to clarify the semantics. Hence, client code that was relying
2489 on some dark areas of C<SQL::Abstract> v1.*
2490 B<might behave differently> in v1.50.
2492 The main changes are :
2498 support for literal SQL through the C<< \ [$sql, bind] >> syntax.
2502 support for the { operator => \"..." } construct (to embed literal SQL)
2506 support for the { operator => \["...", @bind] } construct (to embed literal SQL with bind values)
2510 optional support for L<array datatypes|/"Inserting and Updating Arrays">
2514 defensive programming : check arguments
2518 fixed bug with global logic, which was previously implemented
2519 through global variables yielding side-effects. Prior versions would
2520 interpret C<< [ {cond1, cond2}, [cond3, cond4] ] >>
2521 as C<< "(cond1 AND cond2) OR (cond3 AND cond4)" >>.
2522 Now this is interpreted
2523 as C<< "(cond1 AND cond2) OR (cond3 OR cond4)" >>.
2528 fixed semantics of _bindtype on array args
2532 dropped the C<_anoncopy> of the %where tree. No longer necessary,
2533 we just avoid shifting arrays within that tree.
2537 dropped the C<_modlogic> function
2541 =head1 ACKNOWLEDGEMENTS
2543 There are a number of individuals that have really helped out with
2544 this module. Unfortunately, most of them submitted bugs via CPAN
2545 so I have no idea who they are! But the people I do know are:
2547 Ash Berlin (order_by hash term support)
2548 Matt Trout (DBIx::Class support)
2549 Mark Stosberg (benchmarking)
2550 Chas Owens (initial "IN" operator support)
2551 Philip Collins (per-field SQL functions)
2552 Eric Kolve (hashref "AND" support)
2553 Mike Fragassi (enhancements to "BETWEEN" and "LIKE")
2554 Dan Kubb (support for "quote_char" and "name_sep")
2555 Guillermo Roditi (patch to cleanup "IN" and "BETWEEN", fix and tests for _order_by)
2556 Laurent Dami (internal refactoring, extensible list of special operators, literal SQL)
2557 Norbert Buchmuller (support for literal SQL in hashpair, misc. fixes & tests)
2558 Peter Rabbitson (rewrite of SQLA::Test, misc. fixes & tests)
2559 Oliver Charles (support for "RETURNING" after "INSERT")
2565 L<DBIx::Class>, L<DBIx::Abstract>, L<CGI::FormBuilder>, L<HTML::QuickTable>.
2569 Copyright (c) 2001-2007 Nathan Wiger <nwiger@cpan.org>. All Rights Reserved.
2571 This module is actively maintained by Matt Trout <mst@shadowcatsystems.co.uk>
2573 For support, your best bet is to try the C<DBIx::Class> users mailing list.
2574 While not an official support venue, C<DBIx::Class> makes heavy use of
2575 C<SQL::Abstract>, and as such list members there are very familiar with
2576 how to create queries.
2580 This module is free software; you may copy this under the same
2581 terms as perl itself (either the GNU General Public License or
2582 the Artistic License)