6 SQL::Abstract - Generate SQL from Perl data structures
12 my $sql = SQL::Abstract->new;
14 my($stmt, @bind) = $sql->select($table, \@fields, \%where, \@order);
16 my($stmt, @bind) = $sql->insert($table, \%fieldvals || \@values);
18 my($stmt, @bind) = $sql->update($table, \%fieldvals, \%where);
20 my($stmt, @bind) = $sql->delete($table, \%where);
22 # Then, use these in your DBI statements
23 my $sth = $dbh->prepare($stmt);
26 # Just generate the WHERE clause
27 my($stmt, @bind) = $sql->where(\%where, \@order);
29 # Return values in the same order, for hashed queries
30 # See PERFORMANCE section for more details
31 my @bind = $sql->values(\%fieldvals);
35 This module was inspired by the excellent L<DBIx::Abstract>.
36 However, in using that module I found that what I really wanted
37 to do was generate SQL, but still retain complete control over my
38 statement handles and use the DBI interface. So, I set out to
39 create an abstract SQL generation module.
41 While based on the concepts used by L<DBIx::Abstract>, there are
42 several important differences, especially when it comes to WHERE
43 clauses. I have modified the concepts used to make the SQL easier
44 to generate from Perl data structures and, IMO, more intuitive.
45 The underlying idea is for this module to do what you mean, based
46 on the data structures you provide it. The big advantage is that
47 you don't have to modify your code every time your data changes,
48 as this module figures it out.
50 To begin with, an SQL INSERT is as easy as just specifying a hash
51 of C<key=value> pairs:
54 name => 'Jimbo Bobson',
55 phone => '123-456-7890',
56 address => '42 Sister Lane',
61 The SQL can then be generated with this:
63 my($stmt, @bind) = $sql->insert('people', \%data);
65 Which would give you something like this:
67 $stmt = "INSERT INTO people
68 (address, city, name, phone, state)
69 VALUES (?, ?, ?, ?, ?)";
70 @bind = ('42 Sister Lane', 'St. Louis', 'Jimbo Bobson',
71 '123-456-7890', 'Louisiana');
73 These are then used directly in your DBI code:
75 my $sth = $dbh->prepare($stmt);
78 In addition, you can apply SQL functions to elements of your C<%data>
79 by specifying an arrayref for the given hash value. For example, if
80 you need to execute the Oracle C<to_date> function on a value, you
81 can say something like this:
85 date_entered => ["to_date(?,'MM/DD/YYYY')", "03/02/2003"],
88 The first value in the array is the actual SQL. Any other values are
89 optional and would be included in the bind values array. This gives
92 my($stmt, @bind) = $sql->insert('people', \%data);
94 $stmt = "INSERT INTO people (name, date_entered)
95 VALUES (?, to_date(?,'MM/DD/YYYY'))";
96 @bind = ('Bill', '03/02/2003');
98 An UPDATE is just as easy, all you change is the name of the function:
100 my($stmt, @bind) = $sql->update('people', \%data);
102 Notice that your C<%data> isn't touched; the module will generate
103 the appropriately quirky SQL for you automatically. Usually you'll
104 want to specify a WHERE clause for your UPDATE, though, which is
105 where handling C<%where> hashes comes in handy...
107 This module can generate pretty complicated WHERE statements
108 easily. For example, simple C<key=value> pairs are taken to mean
109 equality, and if you want to see if a field is within a set
110 of values, you can use an arrayref. Let's say we wanted to
111 SELECT some data based on this criteria:
115 worker => ['nwiger', 'rcwe', 'sfz'],
116 status => { '!=', 'completed' }
119 my($stmt, @bind) = $sql->select('tickets', '*', \%where);
121 The above would give you something like this:
123 $stmt = "SELECT * FROM tickets WHERE
124 ( requestor = ? ) AND ( status != ? )
125 AND ( worker = ? OR worker = ? OR worker = ? )";
126 @bind = ('inna', 'completed', 'nwiger', 'rcwe', 'sfz');
128 Which you could then use in DBI code like so:
130 my $sth = $dbh->prepare($stmt);
131 $sth->execute(@bind);
137 The functions are simple. There's one for each major SQL operation,
138 and a constructor you use first. The arguments are specified in a
139 similar order to each function (table, then fields, then a where
140 clause) to try and simplify things.
147 our $VERSION = '1.22';
148 our $REVISION = '$Id$';
151 # Fix SQL case, if so requested
154 return $self->{case} ? $_[0] : uc($_[0]);
157 # Anon copies of arrays/hashes
158 # Based on deep_copy example by merlyn
159 # http://www.stonehenge.com/merlyn/UnixReview/col30.html
162 return (ref $orig eq 'HASH') ? +{map { $_ => _anoncopy($orig->{$_}) } keys %$orig}
163 : (ref $orig eq 'ARRAY') ? [map _anoncopy($_), @$orig]
169 return unless $_[0]->{debug}; shift; # a little faster
170 my $func = (caller(1))[3];
171 warn "[$func] ", @_, "\n";
175 my($func) = (caller(1))[3];
176 carp "[$func] Warning: ", @_;
180 my($func) = (caller(1))[3];
181 croak "[$func] Fatal: ", @_;
188 if (ref $from eq 'ARRAY') {
189 return $self->_recurse_from(@$from);
190 } elsif (ref $from eq 'HASH') {
191 return $self->_make_as($from);
193 return $self->_quote($from);
198 my ($self, $from, @join) = @_;
200 push(@sqlf, $self->_make_as($from));
201 foreach my $j (@join) {
202 push @sqlf, ', ' . $self->_quote($j) and next unless ref $j;
203 push @sqlf, ', ' . $$j and next if ref $j eq 'SCALAR';
206 # check whether a join type exists
207 my $join_clause = '';
208 my $to_jt = ref($to) eq 'ARRAY' ? $to->[0] : $to;
209 if (ref($to_jt) eq 'HASH' and exists($to_jt->{-join_type})) {
210 $join_clause = $self->_sqlcase(' '.($to_jt->{-join_type}).' JOIN ');
212 $join_clause = $self->_sqlcase(' JOIN ');
214 push(@sqlf, $join_clause);
216 if (ref $to eq 'ARRAY') {
217 push(@sqlf, '(', $self->_recurse_from(@$to), ')');
219 push(@sqlf, $self->_make_as($to));
221 push(@sqlf, $self->_sqlcase(' ON '), $self->_join_condition($on));
223 return join('', @sqlf);
227 my ($self, $from) = @_;
228 return $self->_quote($from) unless ref $from;
229 return $$from if ref $from eq 'SCALAR';
230 return join(' ', map { (ref $_ eq 'SCALAR' ? $$_ : $self->_quote($_)) }
231 reverse each %{$self->_skip_options($from)});
235 my ($self, $hash) = @_;
237 $clean_hash->{$_} = $hash->{$_}
238 for grep {!/^-/} keys %$hash;
242 sub _join_condition {
243 my ($self, $cond) = @_;
244 if (ref $cond eq 'HASH') {
247 my $x = '= '.$self->_quote($cond->{$_}); $j{$_} = \$x;
249 return $self->_recurse_where(\%j);
250 } elsif (ref $cond eq 'ARRAY') {
251 return join(' OR ', map { $self->_join_condition($_) } @$cond);
253 die "Can't handle this yet!";
262 return '' unless defined $label;
267 return $label unless $self->{quote_char};
269 if (ref $self->{quote_char} eq "ARRAY") {
271 return $self->{quote_char}->[0] . $label . $self->{quote_char}->[1]
272 if !defined $self->{name_sep};
274 my $sep = $self->{name_sep};
275 return join($self->{name_sep},
276 map { $self->{quote_char}->[0] . $_ . $self->{quote_char}->[1] }
277 split( /\Q$sep\E/, $label ) );
281 return $self->{quote_char} . $label . $self->{quote_char}
282 if !defined $self->{name_sep};
284 return join $self->{name_sep},
285 map { $self->{quote_char} . $_ . $self->{quote_char} }
286 split /\Q$self->{name_sep}\E/, $label;
289 # Conversion, if applicable
292 return @_ unless $self->{convert};
293 my $conv = $self->_sqlcase($self->{convert});
294 my @ret = map { $conv.'('.$_.')' } @_;
295 return wantarray ? @ret : $ret[0];
302 return $self->{bindtype} eq 'columns' ? [ @_ ] : @val;
305 # Modified -logic or -nest
308 my $sym = @_ ? lc(shift) : $self->{logic};
310 $sym = $self->{logic} if $sym eq 'nest';
311 return $self->_sqlcase($sym); # override join
314 =head2 new(option => 'value')
316 The C<new()> function takes a list of options and values, and returns
317 a new B<SQL::Abstract> object which can then be used to generate SQL
318 through the methods below. The options accepted are:
324 If set to 'lower', then SQL will be generated in all lowercase. By
325 default SQL is generated in "textbook" case meaning something like:
327 SELECT a_field FROM a_table WHERE some_field LIKE '%someval%'
331 This determines what the default comparison operator is. By default
332 it is C<=>, meaning that a hash like this:
334 %where = (name => 'nwiger', email => 'nate@wiger.org');
336 Will generate SQL like this:
338 WHERE name = 'nwiger' AND email = 'nate@wiger.org'
340 However, you may want loose comparisons by default, so if you set
341 C<cmp> to C<like> you would get SQL such as:
343 WHERE name like 'nwiger' AND email like 'nate@wiger.org'
345 You can also override the comparsion on an individual basis - see
346 the huge section on L</"WHERE CLAUSES"> at the bottom.
350 This determines the default logical operator for multiple WHERE
351 statements in arrays. By default it is "or", meaning that a WHERE
355 event_date => {'>=', '2/13/99'},
356 event_date => {'<=', '4/24/03'},
359 Will generate SQL like this:
361 WHERE event_date >= '2/13/99' OR event_date <= '4/24/03'
363 This is probably not what you want given this query, though (look
364 at the dates). To change the "OR" to an "AND", simply specify:
366 my $sql = SQL::Abstract->new(logic => 'and');
368 Which will change the above C<WHERE> to:
370 WHERE event_date >= '2/13/99' AND event_date <= '4/24/03'
374 This will automatically convert comparisons using the specified SQL
375 function for both column and value. This is mostly used with an argument
376 of C<upper> or C<lower>, so that the SQL will have the effect of
377 case-insensitive "searches". For example, this:
379 $sql = SQL::Abstract->new(convert => 'upper');
380 %where = (keywords => 'MaKe iT CAse inSeNSItive');
382 Will turn out the following SQL:
384 WHERE upper(keywords) like upper('MaKe iT CAse inSeNSItive')
386 The conversion can be C<upper()>, C<lower()>, or any other SQL function
387 that can be applied symmetrically to fields (actually B<SQL::Abstract> does
388 not validate this option; it will just pass through what you specify verbatim).
392 This is a kludge because many databases suck. For example, you can't
393 just bind values using DBI's C<execute()> for Oracle C<CLOB> or C<BLOB> fields.
394 Instead, you have to use C<bind_param()>:
396 $sth->bind_param(1, 'reg data');
397 $sth->bind_param(2, $lots, {ora_type => ORA_CLOB});
399 The problem is, B<SQL::Abstract> will normally just return a C<@bind> array,
400 which loses track of which field each slot refers to. Fear not.
402 If you specify C<bindtype> in new, you can determine how C<@bind> is returned.
403 Currently, you can specify either C<normal> (default) or C<columns>. If you
404 specify C<columns>, you will get an array that looks like this:
406 my $sql = SQL::Abstract->new(bindtype => 'columns');
407 my($stmt, @bind) = $sql->insert(...);
410 [ 'column1', 'value1' ],
411 [ 'column2', 'value2' ],
412 [ 'column3', 'value3' ],
415 You can then iterate through this manually, using DBI's C<bind_param()>.
417 $sth->prepare($stmt);
420 my($col, $data) = @$_;
421 if ($col eq 'details' || $col eq 'comments') {
422 $sth->bind_param($i, $data, {ora_type => ORA_CLOB});
423 } elsif ($col eq 'image') {
424 $sth->bind_param($i, $data, {ora_type => ORA_BLOB});
426 $sth->bind_param($i, $data);
430 $sth->execute; # execute without @bind now
432 Now, why would you still use B<SQL::Abstract> if you have to do this crap?
433 Basically, the advantage is still that you don't have to care which fields
434 are or are not included. You could wrap that above C<for> loop in a simple
435 sub called C<bind_fields()> or something and reuse it repeatedly. You still
436 get a layer of abstraction over manual SQL specification.
440 This is the character that a table or column name will be quoted
441 with. By default this is an empty string, but you could set it to
442 the character C<`>, to generate SQL like this:
444 SELECT `a_field` FROM `a_table` WHERE `some_field` LIKE '%someval%'
446 This is useful if you have tables or columns that are reserved words
447 in your database's SQL dialect.
451 This is the character that separates a table and column name. It is
452 necessary to specify this when the C<quote_char> option is selected,
453 so that tables and column names can be individually quoted like this:
455 SELECT `table`.`one_field` FROM `table` WHERE `table`.`other_field` = 1
463 my $class = ref($self) || $self;
464 my %opt = (ref $_[0] eq 'HASH') ? %{$_[0]} : @_;
466 # choose our case by keeping an option around
467 delete $opt{case} if $opt{case} && $opt{case} ne 'lower';
469 # override logical operator
470 $opt{logic} = uc $opt{logic} if $opt{logic};
472 # how to return bind vars
473 $opt{bindtype} ||= delete($opt{bind_type}) || 'normal';
475 # default comparison is "=", but can be overridden
478 # default quotation character around tables/columns
479 $opt{quote_char} ||= '';
481 return bless \%opt, $class;
484 =head2 insert($table, \@values || \%fieldvals)
486 This is the simplest function. You simply give it a table name
487 and either an arrayref of values or hashref of field/value pairs.
488 It returns an SQL INSERT statement and a list of bind values.
494 my $table = $self->_table(shift);
495 my $data = shift || return;
497 my $sql = $self->_sqlcase('insert into') . " $table ";
498 my(@sqlf, @sqlv, @sqlq) = ();
501 if ($ref eq 'HASH') {
502 for my $k (sort keys %$data) {
505 # named fields, so must save names in order
506 push @sqlf, $self->_quote($k);
508 # SQL included for values
510 push @sqlq, shift @val;
511 push @sqlv, $self->_bindtype($k, @val);
512 } elsif ($r eq 'SCALAR') {
513 # embedded literal SQL
517 push @sqlv, $self->_bindtype($k, $v);
520 $sql .= '(' . join(', ', @sqlf) .') '. $self->_sqlcase('values') . ' ('. join(', ', @sqlq) .')';
521 } elsif ($ref eq 'ARRAY') {
522 # just generate values(?,?) part
523 # no names (arrayref) so can't generate bindtype
524 carp "Warning: ",__PACKAGE__,"->insert called with arrayref when bindtype set"
525 if $self->{bindtype} ne 'normal';
530 push @sqlq, shift @val;
532 } elsif ($r eq 'SCALAR') {
533 # embedded literal SQL
540 $sql .= $self->_sqlcase('values') . ' ('. join(', ', @sqlq) .')';
541 } elsif ($ref eq 'SCALAR') {
545 puke "Unsupported data type specified to \$sql->insert";
548 return wantarray ? ($sql, @sqlv) : $sql;
551 =head2 update($table, \%fieldvals, \%where)
553 This takes a table, hashref of field/value pairs, and an optional
554 hashref WHERE clause. It returns an SQL UPDATE function and a list
561 my $table = $self->_table(shift);
562 my $data = shift || return;
565 my $sql = $self->_sqlcase('update') . " $table " . $self->_sqlcase('set ');
566 my(@sqlf, @sqlv) = ();
568 puke "Unsupported data type specified to \$sql->update"
569 unless ref $data eq 'HASH';
571 for my $k (sort keys %$data) {
574 my $label = $self->_quote($k);
576 # SQL included for values
578 my $sql = shift @bind;
579 push @sqlf, "$label = $sql";
580 push @sqlv, $self->_bindtype($k, @bind);
581 } elsif ($r eq 'SCALAR') {
582 # embedded literal SQL
583 push @sqlf, "$label = $$v";
585 push @sqlf, "$label = ?";
586 push @sqlv, $self->_bindtype($k, $v);
590 $sql .= join ', ', @sqlf;
593 my($wsql, @wval) = $self->where($where);
598 return wantarray ? ($sql, @sqlv) : $sql;
601 =head2 select($table, \@fields, \%where, \@order)
603 This takes a table, arrayref of fields (or '*'), optional hashref
604 WHERE clause, and optional arrayref order by, and returns the
605 corresponding SQL SELECT statement and list of bind values.
611 my $table = $self->_table(shift);
612 my $fields = shift || '*';
616 my $f = (ref $fields eq 'ARRAY') ? join ', ', map { $self->_quote($_) } @$fields : $fields;
617 my $sql = join ' ', $self->_sqlcase('select'), $f, $self->_sqlcase('from'), $table;
619 my(@sqlf, @sqlv) = ();
620 my($wsql, @wval) = $self->where($where, $order);
624 return wantarray ? ($sql, @sqlv) : $sql;
627 =head2 delete($table, \%where)
629 This takes a table name and optional hashref WHERE clause.
630 It returns an SQL DELETE statement and list of bind values.
636 my $table = $self->_table(shift);
639 my $sql = $self->_sqlcase('delete from') . " $table";
640 my(@sqlf, @sqlv) = ();
643 my($wsql, @wval) = $self->where($where);
648 return wantarray ? ($sql, @sqlv) : $sql;
651 =head2 where(\%where, \@order)
653 This is used to generate just the WHERE clause. For example,
654 if you have an arbitrary data structure and know what the
655 rest of your SQL is going to look like, but want an easy way
656 to produce a WHERE clause, use this. It returns an SQL WHERE
657 clause and list of bind values.
661 # Finally, a separate routine just to handle WHERE clauses
667 # Need a separate routine to properly wrap w/ "where"
669 my @ret = $self->_recurse_where($where);
672 $sql .= $self->_sqlcase(' where ') . $wh if $wh;
677 $sql .= $self->_order_by($order);
680 return wantarray ? ($sql, @ret) : $sql;
685 local $^W = 0; # really, you've gotta be fucking kidding me
687 my $where = _anoncopy(shift); # prevent destroying original
688 my $ref = ref $where || '';
689 my $join = shift || $self->{logic} ||
690 ($ref eq 'ARRAY' ? $self->_sqlcase('or') : $self->_sqlcase('and'));
692 # For assembling SQL fields and values
693 my(@sqlf, @sqlv) = ();
695 # If an arrayref, then we join each element
696 if ($ref eq 'ARRAY') {
697 # need to use while() so can shift() for arrays
699 while (my $el = shift @$where) {
701 # skip empty elements, otherwise get invalid trailing AND stuff
702 if (my $ref2 = ref $el) {
703 if ($ref2 eq 'ARRAY') {
705 } elsif ($ref2 eq 'HASH') {
707 $subjoin ||= $self->_sqlcase('and');
708 } elsif ($ref2 eq 'SCALAR') {
713 $self->_debug("$ref2(*top) means join with $subjoin");
715 # top-level arrayref with scalars, recurse in pairs
716 $self->_debug("NOREF(*top) means join with $subjoin");
717 $el = {$el => shift(@$where)};
719 my @ret = $self->_recurse_where($el, $subjoin);
720 push @sqlf, shift @ret;
724 elsif ($ref eq 'HASH') {
725 # Note: during recursion, the last element will always be a hashref,
726 # since it needs to point a column => value. So this be the end.
727 for my $k (sort keys %$where) {
728 my $v = $where->{$k};
729 my $label = $self->_quote($k);
731 if ($k =~ /^-(\D+)/) {
732 # special nesting, like -and, -or, -nest, so shift over
733 my $subjoin = $self->_modlogic($1);
734 $self->_debug("OP(-$1) means special logic ($subjoin), recursing...");
735 my @ret = $self->_recurse_where($v, $subjoin);
736 push @sqlf, shift @ret;
738 } elsif (! defined($v)) {
740 $self->_debug("UNDEF($k) means IS NULL");
741 push @sqlf, $label . $self->_sqlcase(' is null');
742 } elsif (ref $v eq 'ARRAY') {
745 # multiple elements: multiple options
746 $self->_debug("ARRAY($k) means multiple elements: [ @v ]");
748 # special nesting, like -and, -or, -nest, so shift over
749 my $subjoin = $self->_sqlcase('or');
750 if ($v[0] =~ /^-(\D+)/) {
751 $subjoin = $self->_modlogic($1); # override subjoin
752 $self->_debug("OP(-$1) means special logic ($subjoin), shifting...");
756 # map into an array of hashrefs and recurse
757 my @ret = $self->_recurse_where([map { {$k => $_} } @v], $subjoin);
759 # push results into our structure
760 push @sqlf, shift @ret;
762 } elsif (ref $v eq 'HASH') {
763 # modified operator { '!=', 'completed' }
764 for my $f (sort keys %$v) {
767 # do the right thing for single -in values
768 $x = [$x] if ($f =~ /^-?\s*(not[\s_]+)?in\s*$/i && ref $x ne 'ARRAY');
770 $self->_debug("HASH($k) means modified operator: { $f }");
772 # check for the operator being "IN" or "BETWEEN" or whatever
773 if (ref $x eq 'ARRAY') {
774 if ($f =~ /^-?\s*(not[\s_]+)?(in|between)\s*$/i) {
775 my $u = $self->_modlogic($1 . $2);
776 $self->_debug("HASH($f => $x) uses special operator: [ $u ]");
777 if ($u =~ /between/i) {
779 # Throw an exception if you try to use between with
780 # anything other than 2 values
781 $self->puke("You need two values to use between") unless @$x == 2;
782 push @sqlf, join ' ', $self->_convert($label), $u, $self->_convert('?'),
783 $self->_sqlcase('and'), $self->_convert('?');
785 # DWIM for empty arrayrefs
786 push @sqlf, join ' ', $self->_convert($label), $u, '(',
787 join(', ', map { $self->_convert('?') } @$x),
790 # Empty IN defaults to 0=1 and empty NOT IN to 1=1
791 push(@sqlf, ($u =~ /not/i ? "1=1" : "0=1"));
793 push @sqlv, $self->_bindtype($k, @$x);
795 # multiple elements: multiple options
796 $self->_debug("ARRAY($x) means multiple elements: [ @$x ]");
798 # map into an array of hashrefs and recurse
799 my @ret = $self->_recurse_where([map { {$k => {$f, $_}} } @$x]);
801 # push results into our structure
802 push @sqlf, shift @ret;
805 } elsif (! defined($x)) {
807 my $not = ($f eq '!=' || $f eq 'not like') ? ' not' : '';
808 push @sqlf, $label . $self->_sqlcase(" is$not null");
811 $f =~ s/^-//; # strip leading -like =>
812 $f =~ s/_/ /; # _ => " "
813 push @sqlf, join ' ', $self->_convert($label), $self->_sqlcase($f), $self->_convert('?');
814 push @sqlv, $self->_bindtype($k, $x);
817 } elsif (ref $v eq 'SCALAR') {
819 $self->_debug("SCALAR($k) means literal SQL: $$v");
820 push @sqlf, "$label $$v";
822 # standard key => val
823 $self->_debug("NOREF($k) means simple key=val: $k $self->{cmp} $v");
824 push @sqlf, join ' ', $self->_convert($label), $self->_sqlcase($self->{cmp}), $self->_convert('?');
825 push @sqlv, $self->_bindtype($k, $v);
829 elsif ($ref eq 'SCALAR') {
831 $self->_debug("SCALAR(*top) means literal SQL: $$where");
834 elsif (defined $where) {
836 $self->_debug("NOREF(*top) means literal SQL: $where");
840 # assemble and return sql
841 my $wsql = @sqlf ? '( ' . join(" $join ", @sqlf) . ' )' : '';
842 return wantarray ? ($wsql, @sqlv) : $wsql;
849 my @vals = $ref eq 'ARRAY' ? @{$_[0]} :
850 $ref eq 'SCALAR' ? ${$_[0]} :
852 puke "Unsupported data struct $ref for ORDER BY";
854 my $val = join ', ', map { $self->_quote($_) } @vals;
855 return $val ? $self->_sqlcase(' order by')." $val" : '';
858 =head2 values(\%data)
860 This just returns the values from the hash C<%data>, in the same
861 order that would be returned from any of the other above queries.
862 Using this allows you to markedly speed up your queries if you
863 are affecting lots of rows. See below under the L</"PERFORMANCE"> section.
869 my $data = shift || return;
870 puke "Argument to ", __PACKAGE__, "->values must be a \\%hash"
871 unless ref $data eq 'HASH';
872 return map { $self->_bindtype($_, $data->{$_}) } sort keys %$data;
875 =head2 generate($any, 'number', $of, \@data, $struct, \%types)
877 Warning: This is an experimental method and subject to change.
879 This returns arbitrarily generated SQL. It's a really basic shortcut.
880 It will return two different things, depending on return context:
882 my($stmt, @bind) = $sql->generate('create table', \$table, \@fields);
883 my $stmt_and_val = $sql->generate('create table', \$table, \@fields);
885 These would return the following:
888 $stmt = "CREATE TABLE test (?, ?)";
889 @bind = (field1, field2);
891 # Second calling form
892 $stmt_and_val = "CREATE TABLE test (field1, field2)";
894 Depending on what you're trying to do, it's up to you to choose the correct
895 format. In this example, the second form is what you would want.
899 $sql->generate('alter session', { nls_date_format => 'MM/YY' });
903 ALTER SESSION SET nls_date_format = 'MM/YY'
905 You get the idea. Strings get their case twiddled, but everything
906 else remains verbatim.
913 my(@sql, @sqlq, @sqlv);
917 if ($ref eq 'HASH') {
918 for my $k (sort keys %$_) {
921 my $label = $self->_quote($k);
923 # SQL included for values
925 my $sql = shift @bind;
926 push @sqlq, "$label = $sql";
927 push @sqlv, $self->_bindtype($k, @bind);
928 } elsif ($r eq 'SCALAR') {
929 # embedded literal SQL
930 push @sqlq, "$label = $$v";
932 push @sqlq, "$label = ?";
933 push @sqlv, $self->_bindtype($k, $v);
936 push @sql, $self->_sqlcase('set'), join ', ', @sqlq;
937 } elsif ($ref eq 'ARRAY') {
938 # unlike insert(), assume these are ONLY the column names, i.e. for SQL
943 push @sqlq, shift @val;
945 } elsif ($r eq 'SCALAR') {
946 # embedded literal SQL
953 push @sql, '(' . join(', ', @sqlq) . ')';
954 } elsif ($ref eq 'SCALAR') {
958 # strings get case twiddled
959 push @sql, $self->_sqlcase($_);
963 my $sql = join ' ', @sql;
965 # this is pretty tricky
966 # if ask for an array, return ($stmt, @bind)
967 # otherwise, s/?/shift @sqlv/ to put it inline
969 return ($sql, @sqlv);
971 1 while $sql =~ s/\?/my $d = shift(@sqlv);
972 ref $d ? $d->[1] : $d/e;
979 # This allows us to check for a local, then _form, attr
981 my($name) = $AUTOLOAD =~ /.*::(.+)/;
982 return $self->generate($name, @_);
991 This module uses a variation on the idea from L<DBIx::Abstract>. It
992 is B<NOT>, repeat I<not> 100% compatible. B<The main logic of this
993 module is that things in arrays are OR'ed, and things in hashes
996 The easiest way to explain is to show lots of examples. After
997 each C<%where> hash shown, it is assumed you used:
999 my($stmt, @bind) = $sql->where(\%where);
1001 However, note that the C<%where> hash can be used directly in any
1002 of the other functions as well, as described above.
1004 So, let's get started. To begin, a simple hash:
1008 status => 'completed'
1011 Is converted to SQL C<key = val> statements:
1013 $stmt = "WHERE user = ? AND status = ?";
1014 @bind = ('nwiger', 'completed');
1016 One common thing I end up doing is having a list of values that
1017 a field can be in. To do this, simply specify a list inside of
1022 status => ['assigned', 'in-progress', 'pending'];
1025 This simple code will create the following:
1027 $stmt = "WHERE user = ? AND ( status = ? OR status = ? OR status = ? )";
1028 @bind = ('nwiger', 'assigned', 'in-progress', 'pending');
1030 If you want to specify a different type of operator for your comparison,
1031 you can use a hashref for a given column:
1035 status => { '!=', 'completed' }
1038 Which would generate:
1040 $stmt = "WHERE user = ? AND status != ?";
1041 @bind = ('nwiger', 'completed');
1043 To test against multiple values, just enclose the values in an arrayref:
1045 status => { '!=', ['assigned', 'in-progress', 'pending'] };
1047 Which would give you:
1049 "WHERE status != ? OR status != ? OR status != ?"
1051 But, this is probably not what you want in this case (look at it). So
1052 the hashref can also contain multiple pairs, in which case it is expanded
1053 into an C<AND> of its elements:
1057 status => { '!=', 'completed', -not_like => 'pending%' }
1060 # Or more dynamically, like from a form
1061 $where{user} = 'nwiger';
1062 $where{status}{'!='} = 'completed';
1063 $where{status}{'-not_like'} = 'pending%';
1065 # Both generate this
1066 $stmt = "WHERE user = ? AND status != ? AND status NOT LIKE ?";
1067 @bind = ('nwiger', 'completed', 'pending%');
1069 To get an OR instead, you can combine it with the arrayref idea:
1073 priority => [ {'=', 2}, {'!=', 1} ]
1076 Which would generate:
1078 $stmt = "WHERE user = ? AND priority = ? OR priority != ?";
1079 @bind = ('nwiger', '2', '1');
1081 However, there is a subtle trap if you want to say something like
1082 this (notice the C<AND>):
1084 WHERE priority != ? AND priority != ?
1086 Because, in Perl you I<can't> do this:
1088 priority => { '!=', 2, '!=', 1 }
1090 As the second C<!=> key will obliterate the first. The solution
1091 is to use the special C<-modifier> form inside an arrayref:
1093 priority => [ -and => {'!=', 2}, {'!=', 1} ]
1095 Normally, these would be joined by C<OR>, but the modifier tells it
1096 to use C<AND> instead. (Hint: You can use this in conjunction with the
1097 C<logic> option to C<new()> in order to change the way your queries
1098 work by default.) B<Important:> Note that the C<-modifier> goes
1099 B<INSIDE> the arrayref, as an extra first element. This will
1100 B<NOT> do what you think it might:
1102 priority => -and => [{'!=', 2}, {'!=', 1}] # WRONG!
1104 Here is a quick list of equivalencies, since there is some overlap:
1107 status => {'!=', 'completed', 'not like', 'pending%' }
1108 status => [ -and => {'!=', 'completed'}, {'not like', 'pending%'}]
1111 status => {'=', ['assigned', 'in-progress']}
1112 status => [ -or => {'=', 'assigned'}, {'=', 'in-progress'}]
1113 status => [ {'=', 'assigned'}, {'=', 'in-progress'} ]
1115 In addition to C<-and> and C<-or>, there is also a special C<-nest>
1116 operator which adds an additional set of parens, to create a subquery.
1117 For example, to get something like this:
1119 $stmt = WHERE user = ? AND ( workhrs > ? OR geo = ? )
1120 @bind = ('nwiger', '20', 'ASIA');
1126 -nest => [ workhrs => {'>', 20}, geo => 'ASIA' ],
1129 You can also use the hashref format to compare a list of fields using the
1130 C<IN> comparison operator, by specifying the list as an arrayref:
1133 status => 'completed',
1134 reportid => { -in => [567, 2335, 2] }
1137 Which would generate:
1139 $stmt = "WHERE status = ? AND reportid IN (?,?,?)";
1140 @bind = ('completed', '567', '2335', '2');
1142 You can use this same format to use other grouping functions, such
1143 as C<BETWEEN>, C<SOME>, and so forth. For example:
1147 completion_date => {
1148 -not_between => ['2002-10-01', '2003-02-06']
1154 WHERE user = ? AND completion_date NOT BETWEEN ( ? AND ? )
1156 So far, we've seen how multiple conditions are joined with a top-level
1157 C<AND>. We can change this by putting the different conditions we want in
1158 hashes and then putting those hashes in an array. For example:
1163 status => { -like => ['pending%', 'dispatched'] },
1167 status => 'unassigned',
1171 This data structure would create the following:
1173 $stmt = "WHERE ( user = ? AND ( status LIKE ? OR status LIKE ? ) )
1174 OR ( user = ? AND status = ? ) )";
1175 @bind = ('nwiger', 'pending', 'dispatched', 'robot', 'unassigned');
1177 This can be combined with the C<-nest> operator to properly group
1184 -and => [workhrs => {'>', 20}, geo => 'ASIA' ],
1185 -and => [workhrs => {'<', 50}, geo => 'EURO' ]
1192 WHERE ( user = ? AND
1193 ( ( workhrs > ? AND geo = ? )
1194 OR ( workhrs < ? AND geo = ? ) ) )
1196 Finally, sometimes only literal SQL will do. If you want to include
1197 literal SQL verbatim, you can specify it as a scalar reference, namely:
1199 my $inn = 'is Not Null';
1201 priority => { '<', 2 },
1207 $stmt = "WHERE priority < ? AND requestor is Not Null";
1210 Note that in this example, you only get one bind parameter back, since
1211 the verbatim SQL is passed as part of the statement.
1213 Of course, just to prove a point, the above can also be accomplished
1217 priority => { '<', 2 },
1218 requestor => { '!=', undef },
1223 These pages could go on for a while, since the nesting of the data
1224 structures this module can handle are pretty much unlimited (the
1225 module implements the C<WHERE> expansion as a recursive function
1226 internally). Your best bet is to "play around" with the module a
1227 little to see how the data structures behave, and choose the best
1228 format for your data based on that.
1230 And of course, all the values above will probably be replaced with
1231 variables gotten from forms or the command line. After all, if you
1232 knew everything ahead of time, you wouldn't have to worry about
1233 dynamically-generating SQL and could just hardwire it into your
1238 Thanks to some benchmarking by Mark Stosberg, it turns out that
1239 this module is many orders of magnitude faster than using C<DBIx::Abstract>.
1240 I must admit this wasn't an intentional design issue, but it's a
1241 byproduct of the fact that you get to control your C<DBI> handles
1244 To maximize performance, use a code snippet like the following:
1246 # prepare a statement handle using the first row
1247 # and then reuse it for the rest of the rows
1249 for my $href (@array_of_hashrefs) {
1250 $stmt ||= $sql->insert('table', $href);
1251 $sth ||= $dbh->prepare($stmt);
1252 $sth->execute($sql->values($href));
1255 The reason this works is because the keys in your C<$href> are sorted
1256 internally by B<SQL::Abstract>. Thus, as long as your data retains
1257 the same structure, you only have to generate the SQL the first time
1258 around. On subsequent queries, simply use the C<values> function provided
1259 by this module to return your values in the correct order.
1263 If you use my C<CGI::FormBuilder> module at all, you'll hopefully
1264 really like this part (I do, at least). Building up a complex query
1265 can be as simple as the following:
1269 use CGI::FormBuilder;
1272 my $form = CGI::FormBuilder->new(...);
1273 my $sql = SQL::Abstract->new;
1275 if ($form->submitted) {
1276 my $field = $form->field;
1277 my $id = delete $field->{id};
1278 my($stmt, @bind) = $sql->update('table', $field, {id => $id});
1281 Of course, you would still have to connect using C<DBI> to run the
1282 query, but the point is that if you make your form look like your
1283 table, the actual query script can be extremely simplistic.
1285 If you're B<REALLY> lazy (I am), check out C<HTML::QuickTable> for
1286 a fast interface to returning and formatting data. I frequently
1287 use these three modules together to write complex database query
1288 apps in under 50 lines.
1292 There is not (yet) any explicit support for SQL compound logic
1293 statements like "AND NOT". Instead, just do the de Morgan's
1294 law transformations yourself. For example, this:
1296 "lname LIKE '%son%' AND NOT ( age < 10 OR age > 20 )"
1300 "lname LIKE '%son%' AND ( age >= 10 AND age <= 20 )"
1302 With the corresponding C<%where> hash:
1305 lname => {like => '%son%'},
1306 age => [-and => {'>=', 10}, {'<=', 20}],
1309 Again, remember that the C<-and> goes I<inside> the arrayref.
1311 =head1 ACKNOWLEDGEMENTS
1313 There are a number of individuals that have really helped out with
1314 this module. Unfortunately, most of them submitted bugs via CPAN
1315 so I have no idea who they are! But the people I do know are:
1317 Mark Stosberg (benchmarking)
1318 Chas Owens (initial "IN" operator support)
1319 Philip Collins (per-field SQL functions)
1320 Eric Kolve (hashref "AND" support)
1321 Mike Fragassi (enhancements to "BETWEEN" and "LIKE")
1322 Dan Kubb (support for "quote_char" and "name_sep")
1323 Matt Trout (DBIx::Class support)
1329 If found, please DO NOT submit anything via C<rt.cpan.org> - that
1330 just causes me a ton of work. Email me a patch (or script demonstrating
1331 the problem) to the below address, and include the VERSION you're using.
1335 L<DBIx::Abstract>, L<DBI|DBI>, L<CGI::FormBuilder>, L<HTML::QuickTable>
1339 Copyright (c) 2001-2006 Nathan Wiger <nwiger@cpan.org>. All Rights Reserved.
1341 For support, your best bet is to try the C<DBIx::Class> users mailing list.
1342 While not an official support venue, C<DBIx::Class> makes heavy use of
1343 C<SQL::Abstract>, and as such list members there are very familiar with
1344 how to create queries.
1346 This module is free software; you may copy this under the terms of
1347 the GNU General Public License, or the Artistic License, copies of
1348 which should have accompanied your Perl kit.