3 $overload::hint_bits = 0x20000;
11 $ {$package . "::OVERLOAD"}{dummy}++; # Register with magic by touching.
12 *{$package . "::()"} = \&nil; # Make it findable via fetchmethod.
14 if ($_ eq 'fallback') {
18 if (not ref $sub and $sub !~ /::/) {
19 $ {$package . "::(" . $_} = $sub;
22 #print STDERR "Setting `$ {'package'}::\cO$_' to \\&`$sub'.\n";
23 *{$package . "::(" . $_} = \&{ $sub };
26 ${$package . "::()"} = $fb; # Make it findable too (fallback only).
30 $package = (caller())[0];
31 # *{$package . "::OVERLOAD"} = \&OVERLOAD;
33 $package->overload::OVERLOAD(@_);
37 $package = (caller())[0];
38 ${$package . "::OVERLOAD"}{dummy}++; # Upgrade the table
41 if ($_ eq 'fallback') {
42 undef $ {$package . "::()"};
44 delete $ {$package . "::"}{"(" . $_};
51 $package = ref $package if ref $package;
57 return undef unless $globref;
58 my $sub = \&{*$globref};
59 return $sub if $sub ne \&nil;
60 return shift->can($ {*$globref});
63 sub OverloadedStringify {
65 $package = ref $package if ref $package;
67 ov_method mycan($package, '(""'), $package
68 or ov_method mycan($package, '(0+'), $package
69 or ov_method mycan($package, '(bool'), $package
70 or ov_method mycan($package, '(nomethod'), $package;
75 $package = ref $package if ref $package;
76 #my $meth = $package->can('(' . shift);
77 ov_method mycan($package, '(' . shift), $package;
78 #return $meth if $meth ne \&nil;
83 my $package = ref $_[0];
84 return "$_[0]" unless $package;
85 bless $_[0], overload::Fake; # Non-overloaded package
87 bless $_[0], $package; # Back
88 $package . substr $str, index $str, '=';
92 (OverloadedStringify($_[0]) or ref($_[0]) eq 'Regexp') ?
97 sub mycan { # Real can would leave stubs.
98 my ($package, $meth) = @_;
99 return \*{$package . "::$meth"} if defined &{$package . "::$meth"};
101 foreach $p (@{$package . "::ISA"}) {
102 my $out = mycan($p, $meth);
116 %ops = ( with_assign => "+ - * / % ** << >> x .",
117 assign => "+= -= *= /= %= **= <<= >>= x= .=",
118 num_comparison => "< <= > >= == !=",
119 '3way_comparison'=> "<=> cmp",
120 str_comparison => "lt le gt ge eq ne",
124 func => "atan2 cos sin exp abs log sqrt",
125 conversion => 'bool "" 0+',
127 dereferencing => '${} @{} %{} &{} *{}',
128 special => 'nomethod fallback =');
130 use warnings::register;
132 # Arguments: what, sub
135 warnings::warnif ("Odd number of arguments for overload::constant");
138 elsif (!exists $constants {$_ [0]}) {
139 warnings::warnif ("`$_[0]' is not an overloadable type");
141 elsif (!ref $_ [1] || "$_[1]" !~ /CODE\(0x[\da-f]+\)$/) {
142 # Can't use C<ref $_[1] eq "CODE"> above as code references can be
143 # blessed, and C<ref> would return the package the ref is blessed into.
144 if (warnings::enabled) {
145 $_ [1] = "undef" unless defined $_ [1];
146 warnings::warn ("`$_[1]' is not a code reference");
151 $^H |= $constants{$_[0]} | $overload::hint_bits;
157 sub remove_constant {
158 # Arguments: what, sub
161 $^H &= ~ $constants{$_[0]};
172 overload - Package for overloading perl operations
185 $a = new SomeThing 57;
188 if (overload::Overloaded $b) {...}
190 $strval = overload::StrVal $b;
194 =head2 Declaration of overloaded functions
196 The compilation directive
203 declares function Number::add() for addition, and method muas() in
204 the "class" C<Number> (or one of its base classes)
205 for the assignment form C<*=> of multiplication.
207 Arguments of this directive come in (key, value) pairs. Legal values
208 are values legal inside a C<&{ ... }> call, so the name of a
209 subroutine, a reference to a subroutine, or an anonymous subroutine
210 will all work. Note that values specified as strings are
211 interpreted as methods, not subroutines. Legal keys are listed below.
213 The subroutine C<add> will be called to execute C<$a+$b> if $a
214 is a reference to an object blessed into the package C<Number>, or if $a is
215 not an object from a package with defined mathemagic addition, but $b is a
216 reference to a C<Number>. It can also be called in other situations, like
217 C<$a+=7>, or C<$a++>. See L<MAGIC AUTOGENERATION>. (Mathemagical
218 methods refer to methods triggered by an overloaded mathematical
221 Since overloading respects inheritance via the @ISA hierarchy, the
222 above declaration would also trigger overloading of C<+> and C<*=> in
223 all the packages which inherit from C<Number>.
225 =head2 Calling Conventions for Binary Operations
227 The functions specified in the C<use overload ...> directive are called
228 with three (in one particular case with four, see L<Last Resort>)
229 arguments. If the corresponding operation is binary, then the first
230 two arguments are the two arguments of the operation. However, due to
231 general object calling conventions, the first argument should always be
232 an object in the package, so in the situation of C<7+$a>, the
233 order of the arguments is interchanged. It probably does not matter
234 when implementing the addition method, but whether the arguments
235 are reversed is vital to the subtraction method. The method can
236 query this information by examining the third argument, which can take
237 three different values:
243 the order of arguments is as in the current operation.
247 the arguments are reversed.
251 the current operation is an assignment variant (as in
252 C<$a+=7>), but the usual function is called instead. This additional
253 information can be used to generate some optimizations. Compare
254 L<Calling Conventions for Mutators>.
258 =head2 Calling Conventions for Unary Operations
260 Unary operation are considered binary operations with the second
261 argument being C<undef>. Thus the functions that overloads C<{"++"}>
262 is called with arguments C<($a,undef,'')> when $a++ is executed.
264 =head2 Calling Conventions for Mutators
266 Two types of mutators have different calling conventions:
270 =item C<++> and C<-->
272 The routines which implement these operators are expected to actually
273 I<mutate> their arguments. So, assuming that $obj is a reference to a
276 sub incr { my $n = $ {$_[0]}; ++$n; $_[0] = bless \$n}
278 is an appropriate implementation of overloaded C<++>. Note that
280 sub incr { ++$ {$_[0]} ; shift }
282 is OK if used with preincrement and with postincrement. (In the case
283 of postincrement a copying will be performed, see L<Copy Constructor>.)
285 =item C<x=> and other assignment versions
287 There is nothing special about these methods. They may change the
288 value of their arguments, and may leave it as is. The result is going
289 to be assigned to the value in the left-hand-side if different from
292 This allows for the same method to be used as overloaded C<+=> and
293 C<+>. Note that this is I<allowed>, but not recommended, since by the
294 semantic of L<"Fallback"> Perl will call the method for C<+> anyway,
295 if C<+=> is not overloaded.
299 B<Warning.> Due to the presense of assignment versions of operations,
300 routines which may be called in assignment context may create
301 self-referential structures. Currently Perl will not free self-referential
302 structures until cycles are C<explicitly> broken. You may get problems
303 when traversing your structures too.
307 use overload '+' => sub { bless [ \$_[0], \$_[1] ] };
309 is asking for trouble, since for code C<$obj += $foo> the subroutine
310 is called as C<$obj = add($obj, $foo, undef)>, or C<$obj = [\$obj,
311 \$foo]>. If using such a subroutine is an important optimization, one
312 can overload C<+=> explicitly by a non-"optimized" version, or switch
313 to non-optimized version if C<not defined $_[2]> (see
314 L<Calling Conventions for Binary Operations>).
316 Even if no I<explicit> assignment-variants of operators are present in
317 the script, they may be generated by the optimizer. Say, C<",$obj,"> or
318 C<',' . $obj . ','> may be both optimized to
320 my $tmp = ',' . $obj; $tmp .= ',';
322 =head2 Overloadable Operations
324 The following symbols can be specified in C<use overload> directive:
328 =item * I<Arithmetic operations>
330 "+", "+=", "-", "-=", "*", "*=", "/", "/=", "%", "%=",
331 "**", "**=", "<<", "<<=", ">>", ">>=", "x", "x=", ".", ".=",
333 For these operations a substituted non-assignment variant can be called if
334 the assignment variant is not available. Methods for operations "C<+>",
335 "C<->", "C<+=>", and "C<-=>" can be called to automatically generate
336 increment and decrement methods. The operation "C<->" can be used to
337 autogenerate missing methods for unary minus or C<abs>.
339 See L<"MAGIC AUTOGENERATION">, L<"Calling Conventions for Mutators"> and
340 L<"Calling Conventions for Binary Operations">) for details of these
343 =item * I<Comparison operations>
345 "<", "<=", ">", ">=", "==", "!=", "<=>",
346 "lt", "le", "gt", "ge", "eq", "ne", "cmp",
348 If the corresponding "spaceship" variant is available, it can be
349 used to substitute for the missing operation. During C<sort>ing
350 arrays, C<cmp> is used to compare values subject to C<use overload>.
352 =item * I<Bit operations>
354 "&", "^", "|", "neg", "!", "~",
356 "C<neg>" stands for unary minus. If the method for C<neg> is not
357 specified, it can be autogenerated using the method for
358 subtraction. If the method for "C<!>" is not specified, it can be
359 autogenerated using the methods for "C<bool>", or "C<\"\">", or "C<0+>".
361 =item * I<Increment and decrement>
365 If undefined, addition and subtraction methods can be
366 used instead. These operations are called both in prefix and
369 =item * I<Transcendental functions>
371 "atan2", "cos", "sin", "exp", "abs", "log", "sqrt",
373 If C<abs> is unavailable, it can be autogenerated using methods
374 for "E<lt>" or "E<lt>=E<gt>" combined with either unary minus or subtraction.
376 =item * I<Boolean, string and numeric conversion>
378 "bool", "\"\"", "0+",
380 If one or two of these operations are not overloaded, the remaining ones can
381 be used instead. C<bool> is used in the flow control operators
382 (like C<while>) and for the ternary "C<?:>" operation. These functions can
383 return any arbitrary Perl value. If the corresponding operation for this value
384 is overloaded too, that operation will be called again with this value.
390 If not overloaded, the argument will be converted to a filehandle or
391 glob (which may require a stringification). The same overloading
392 happens both for the I<read-filehandle> syntax C<E<lt>$varE<gt>> and
393 I<globbing> syntax C<E<lt>${var}E<gt>>.
395 =item * I<Dereferencing>
397 '${}', '@{}', '%{}', '&{}', '*{}'.
399 If not overloaded, the argument will be dereferenced I<as is>, thus
400 should be of correct type. These functions should return a reference
401 of correct type, or another object with overloaded dereferencing.
405 "nomethod", "fallback", "=",
407 see L<SPECIAL SYMBOLS FOR C<use overload>>.
411 See L<"Fallback"> for an explanation of when a missing method can be
414 A computer-readable form of the above table is available in the hash
415 %overload::ops, with values being space-separated lists of names:
417 with_assign => '+ - * / % ** << >> x .',
418 assign => '+= -= *= /= %= **= <<= >>= x= .=',
419 num_comparison => '< <= > >= == !=',
420 '3way_comparison'=> '<=> cmp',
421 str_comparison => 'lt le gt ge eq ne',
425 func => 'atan2 cos sin exp abs log sqrt',
426 conversion => 'bool "" 0+',
428 dereferencing => '${} @{} %{} &{} *{}',
429 special => 'nomethod fallback ='
431 =head2 Inheritance and overloading
433 Inheritance interacts with overloading in two ways.
437 =item Strings as values of C<use overload> directive
441 use overload key => value;
443 is a string, it is interpreted as a method name.
445 =item Overloading of an operation is inherited by derived classes
447 Any class derived from an overloaded class is also overloaded. The
448 set of overloaded methods is the union of overloaded methods of all
449 the ancestors. If some method is overloaded in several ancestor, then
450 which description will be used is decided by the usual inheritance
453 If C<A> inherits from C<B> and C<C> (in this order), C<B> overloads
454 C<+> with C<\&D::plus_sub>, and C<C> overloads C<+> by C<"plus_meth">,
455 then the subroutine C<D::plus_sub> will be called to implement
456 operation C<+> for an object in package C<A>.
460 Note that since the value of the C<fallback> key is not a subroutine,
461 its inheritance is not governed by the above rules. In the current
462 implementation, the value of C<fallback> in the first overloaded
463 ancestor is used, but this is accidental and subject to change.
465 =head1 SPECIAL SYMBOLS FOR C<use overload>
467 Three keys are recognized by Perl that are not covered by the above
472 C<"nomethod"> should be followed by a reference to a function of four
473 parameters. If defined, it is called when the overloading mechanism
474 cannot find a method for some operation. The first three arguments of
475 this function coincide with the arguments for the corresponding method if
476 it were found, the fourth argument is the symbol
477 corresponding to the missing method. If several methods are tried,
478 the last one is used. Say, C<1-$a> can be equivalent to
480 &nomethodMethod($a,1,1,"-")
482 if the pair C<"nomethod" =E<gt> "nomethodMethod"> was specified in the
483 C<use overload> directive.
485 If some operation cannot be resolved, and there is no function
486 assigned to C<"nomethod">, then an exception will be raised via die()--
487 unless C<"fallback"> was specified as a key in C<use overload> directive.
491 The key C<"fallback"> governs what to do if a method for a particular
492 operation is not found. Three different cases are possible depending on
493 the value of C<"fallback">:
500 substituted method (see L<MAGIC AUTOGENERATION>). If this fails, it
501 then tries to calls C<"nomethod"> value; if missing, an exception
506 The same as for the C<undef> value, but no exception is raised. Instead,
507 it silently reverts to what it would have done were there no C<use overload>
510 =item * defined, but FALSE
512 No autogeneration is tried. Perl tries to call
513 C<"nomethod"> value, and if this is missing, raises an exception.
517 B<Note.> C<"fallback"> inheritance via @ISA is not carved in stone
518 yet, see L<"Inheritance and overloading">.
520 =head2 Copy Constructor
522 The value for C<"="> is a reference to a function with three
523 arguments, i.e., it looks like the other values in C<use
524 overload>. However, it does not overload the Perl assignment
525 operator. This would go against Camel hair.
527 This operation is called in the situations when a mutator is applied
528 to a reference that shares its object with some other reference, such
534 To make this change $a and not change $b, a copy of C<$$a> is made,
535 and $a is assigned a reference to this new object. This operation is
536 done during execution of the C<++$a>, and not during the assignment,
537 (so before the increment C<$$a> coincides with C<$$b>). This is only
538 done if C<++> is expressed via a method for C<'++'> or C<'+='> (or
539 C<nomethod>). Note that if this operation is expressed via C<'+'>
540 a nonmutator, i.e., as in
545 then C<$a> does not reference a new copy of C<$$a>, since $$a does not
546 appear as lvalue when the above code is executed.
548 If the copy constructor is required during the execution of some mutator,
549 but a method for C<'='> was not specified, it can be autogenerated as a
550 string copy if the object is a plain scalar.
556 The actually executed code for
559 Something else which does not modify $a or $b....
565 Something else which does not modify $a or $b....
566 $a = $a->clone(undef,"");
569 if $b was mathemagical, and C<'++'> was overloaded with C<\&incr>,
570 C<'='> was overloaded with C<\&clone>.
574 Same behaviour is triggered by C<$b = $a++>, which is consider a synonym for
577 =head1 MAGIC AUTOGENERATION
579 If a method for an operation is not found, and the value for C<"fallback"> is
580 TRUE or undefined, Perl tries to autogenerate a substitute method for
581 the missing operation based on the defined operations. Autogenerated method
582 substitutions are possible for the following operations:
586 =item I<Assignment forms of arithmetic operations>
588 C<$a+=$b> can use the method for C<"+"> if the method for C<"+=">
591 =item I<Conversion operations>
593 String, numeric, and boolean conversion are calculated in terms of one
594 another if not all of them are defined.
596 =item I<Increment and decrement>
598 The C<++$a> operation can be expressed in terms of C<$a+=1> or C<$a+1>,
599 and C<$a--> in terms of C<$a-=1> and C<$a-1>.
603 can be expressed in terms of C<$aE<lt>0> and C<-$a> (or C<0-$a>).
607 can be expressed in terms of subtraction.
611 C<!> and C<not> can be expressed in terms of boolean conversion, or
612 string or numerical conversion.
614 =item I<Concatenation>
616 can be expressed in terms of string conversion.
618 =item I<Comparison operations>
620 can be expressed in terms of its "spaceship" counterpart: either
621 C<E<lt>=E<gt>> or C<cmp>:
623 <, >, <=, >=, ==, != in terms of <=>
624 lt, gt, le, ge, eq, ne in terms of cmp
628 <> in terms of builtin operations
630 =item I<Dereferencing>
632 ${} @{} %{} &{} *{} in terms of builtin operations
634 =item I<Copy operator>
636 can be expressed in terms of an assignment to the dereferenced value, if this
637 value is a scalar and not a reference.
641 =head1 Losing overloading
643 The restriction for the comparison operation is that even if, for example,
644 `C<cmp>' should return a blessed reference, the autogenerated `C<lt>'
645 function will produce only a standard logical value based on the
646 numerical value of the result of `C<cmp>'. In particular, a working
647 numeric conversion is needed in this case (possibly expressed in terms of
650 Similarly, C<.=> and C<x=> operators lose their mathemagical properties
651 if the string conversion substitution is applied.
653 When you chop() a mathemagical object it is promoted to a string and its
654 mathemagical properties are lost. The same can happen with other
657 =head1 Run-time Overloading
659 Since all C<use> directives are executed at compile-time, the only way to
660 change overloading during run-time is to
662 eval 'use overload "+" => \&addmethod';
666 eval 'no overload "+", "--", "<="';
668 though the use of these constructs during run-time is questionable.
670 =head1 Public functions
672 Package C<overload.pm> provides the following public functions:
676 =item overload::StrVal(arg)
678 Gives string value of C<arg> as in absence of stringify overloading.
680 =item overload::Overloaded(arg)
682 Returns true if C<arg> is subject to overloading of some operations.
684 =item overload::Method(obj,op)
686 Returns C<undef> or a reference to the method that implements C<op>.
690 =head1 Overloading constants
692 For some application Perl parser mangles constants too much. It is possible
693 to hook into this process via overload::constant() and overload::remove_constant()
696 These functions take a hash as an argument. The recognized keys of this hash
703 to overload integer constants,
707 to overload floating point constants,
711 to overload octal and hexadecimal constants,
715 to overload C<q>-quoted strings, constant pieces of C<qq>- and C<qx>-quoted
716 strings and here-documents,
720 to overload constant pieces of regular expressions.
724 The corresponding values are references to functions which take three arguments:
725 the first one is the I<initial> string form of the constant, the second one
726 is how Perl interprets this constant, the third one is how the constant is used.
727 Note that the initial string form does not
728 contain string delimiters, and has backslashes in backslash-delimiter
729 combinations stripped (thus the value of delimiter is not relevant for
730 processing of this string). The return value of this function is how this
731 constant is going to be interpreted by Perl. The third argument is undefined
732 unless for overloaded C<q>- and C<qr>- constants, it is C<q> in single-quote
733 context (comes from strings, regular expressions, and single-quote HERE
734 documents), it is C<tr> for arguments of C<tr>/C<y> operators,
735 it is C<s> for right-hand side of C<s>-operator, and it is C<qq> otherwise.
737 Since an expression C<"ab$cd,,"> is just a shortcut for C<'ab' . $cd . ',,'>,
738 it is expected that overloaded constant strings are equipped with reasonable
739 overloaded catenation operator, otherwise absurd results will result.
740 Similarly, negative numbers are considered as negations of positive constants.
742 Note that it is probably meaningless to call the functions overload::constant()
743 and overload::remove_constant() from anywhere but import() and unimport() methods.
744 From these methods they may be called as
749 die "unknown import: @_" unless @_ == 1 and $_[0] eq ':constant';
750 overload::constant integer => sub {Math::BigInt->new(shift)};
753 B<BUGS> Currently overloaded-ness of constants does not propagate
756 =head1 IMPLEMENTATION
758 What follows is subject to change RSN.
760 The table of methods for all operations is cached in magic for the
761 symbol table hash for the package. The cache is invalidated during
762 processing of C<use overload>, C<no overload>, new function
763 definitions, and changes in @ISA. However, this invalidation remains
764 unprocessed until the next C<bless>ing into the package. Hence if you
765 want to change overloading structure dynamically, you'll need an
766 additional (fake) C<bless>ing to update the table.
768 (Every SVish thing has a magic queue, and magic is an entry in that
769 queue. This is how a single variable may participate in multiple
770 forms of magic simultaneously. For instance, environment variables
771 regularly have two forms at once: their %ENV magic and their taint
772 magic. However, the magic which implements overloading is applied to
773 the stashes, which are rarely used directly, thus should not slow down
776 If an object belongs to a package using overload, it carries a special
777 flag. Thus the only speed penalty during arithmetic operations without
778 overloading is the checking of this flag.
780 In fact, if C<use overload> is not present, there is almost no overhead
781 for overloadable operations, so most programs should not suffer
782 measurable performance penalties. A considerable effort was made to
783 minimize the overhead when overload is used in some package, but the
784 arguments in question do not belong to packages using overload. When
785 in doubt, test your speed with C<use overload> and without it. So far
786 there have been no reports of substantial speed degradation if Perl is
787 compiled with optimization turned on.
789 There is no size penalty for data if overload is not used. The only
790 size penalty if overload is used in some package is that I<all> the
791 packages acquire a magic during the next C<bless>ing into the
792 package. This magic is three-words-long for packages without
793 overloading, and carries the cache table if the package is overloaded.
795 Copying (C<$a=$b>) is shallow; however, a one-level-deep copying is
796 carried out before any operation that can imply an assignment to the
797 object $a (or $b) refers to, like C<$a++>. You can override this
798 behavior by defining your own copy constructor (see L<"Copy Constructor">).
800 It is expected that arguments to methods that are not explicitly supposed
801 to be changed are constant (but this is not enforced).
803 =head1 Metaphor clash
805 One may wonder why the semantic of overloaded C<=> is so counter intuitive.
806 If it I<looks> counter intuitive to you, you are subject to a metaphor
809 Here is a Perl object metaphor:
811 I< object is a reference to blessed data>
813 and an arithmetic metaphor:
815 I< object is a thing by itself>.
817 The I<main> problem of overloading C<=> is the fact that these metaphors
818 imply different actions on the assignment C<$a = $b> if $a and $b are
819 objects. Perl-think implies that $a becomes a reference to whatever
820 $b was referencing. Arithmetic-think implies that the value of "object"
821 $a is changed to become the value of the object $b, preserving the fact
822 that $a and $b are separate entities.
824 The difference is not relevant in the absence of mutators. After
825 a Perl-way assignment an operation which mutates the data referenced by $a
826 would change the data referenced by $b too. Effectively, after
827 C<$a = $b> values of $a and $b become I<indistinguishable>.
829 On the other hand, anyone who has used algebraic notation knows the
830 expressive power of the arithmetic metaphor. Overloading works hard
831 to enable this metaphor while preserving the Perlian way as far as
832 possible. Since it is not not possible to freely mix two contradicting
833 metaphors, overloading allows the arithmetic way to write things I<as
834 far as all the mutators are called via overloaded access only>. The
835 way it is done is described in L<Copy Constructor>.
837 If some mutator methods are directly applied to the overloaded values,
838 one may need to I<explicitly unlink> other values which references the
843 $b = $a; # $b is "linked" to $a
845 $a = $a->clone; # Unlink $b from $a
848 Note that overloaded access makes this transparent:
851 $b = $a; # $b is "linked" to $a
852 $a += 4; # would unlink $b automagically
854 However, it would not make
857 $a = 4; # Now $a is a plain 4, not 'Data'
859 preserve "objectness" of $a. But Perl I<has> a way to make assignments
860 to an object do whatever you want. It is just not the overload, but
861 tie()ing interface (see L<perlfunc/tie>). Adding a FETCH() method
862 which returns the object itself, and STORE() method which changes the
863 value of the object, one can reproduce the arithmetic metaphor in its
864 completeness, at least for variables which were tie()d from the start.
866 (Note that a workaround for a bug may be needed, see L<"BUGS">.)
870 Please add examples to what follows!
872 =head2 Two-face scalars
874 Put this in F<two_face.pm> in your Perl library directory:
876 package two_face; # Scalars with separate string and
878 sub new { my $p = shift; bless [@_], $p }
879 use overload '""' => \&str, '0+' => \&num, fallback => 1;
886 my $seven = new two_face ("vii", 7);
887 printf "seven=$seven, seven=%d, eight=%d\n", $seven, $seven+1;
888 print "seven contains `i'\n" if $seven =~ /i/;
890 (The second line creates a scalar which has both a string value, and a
891 numeric value.) This prints:
893 seven=vii, seven=7, eight=8
896 =head2 Two-face references
898 Suppose you want to create an object which is accessible as both an
899 array reference and a hash reference, similar to the
900 L<pseudo-hash|perlref/"Pseudo-hashes: Using an array as a hash">
901 builtin Perl type. Let's make it better than a pseudo-hash by
902 allowing index 0 to be treated as a normal element.
905 use overload '%{}' => \&gethash, '@{}' => sub { $ {shift()} };
913 tie %h, ref $self, $self;
917 sub TIEHASH { my $p = shift; bless \ shift, $p }
920 $fields{$_} = $i++ foreach qw{zero one two three};
922 my $self = ${shift()};
923 my $key = $fields{shift()};
924 defined $key or die "Out of band access";
925 $$self->[$key] = shift;
928 my $self = ${shift()};
929 my $key = $fields{shift()};
930 defined $key or die "Out of band access";
934 Now one can access an object using both the array and hash syntax:
936 my $bar = new two_refs 3,4,5,6;
938 $bar->{two} == 11 or die 'bad hash fetch';
940 Note several important features of this example. First of all, the
941 I<actual> type of $bar is a scalar reference, and we do not overload
942 the scalar dereference. Thus we can get the I<actual> non-overloaded
943 contents of $bar by just using C<$$bar> (what we do in functions which
944 overload dereference). Similarly, the object returned by the
945 TIEHASH() method is a scalar reference.
947 Second, we create a new tied hash each time the hash syntax is used.
948 This allows us not to worry about a possibility of a reference loop,
949 would would lead to a memory leak.
951 Both these problems can be cured. Say, if we want to overload hash
952 dereference on a reference to an object which is I<implemented> as a
953 hash itself, the only problem one has to circumvent is how to access
954 this I<actual> hash (as opposed to the I<virtual> exhibited by
955 overloaded dereference operator). Here is one possible fetching routine:
958 my ($self, $key) = (shift, shift);
959 my $class = ref $self;
960 bless $self, 'overload::dummy'; # Disable overloading of %{}
961 my $out = $self->{$key};
962 bless $self, $class; # Restore overloading
966 To move creation of the tied hash on each access, one may an extra
967 level of indirection which allows a non-circular structure of references:
970 use overload '%{}' => sub { ${shift()}->[1] },
971 '@{}' => sub { ${shift()}->[0] };
977 bless \ [$a, \%h], $p;
982 tie %h, ref $self, $self;
986 sub TIEHASH { my $p = shift; bless \ shift, $p }
989 $fields{$_} = $i++ foreach qw{zero one two three};
992 my $key = $fields{shift()};
993 defined $key or die "Out of band access";
998 my $key = $fields{shift()};
999 defined $key or die "Out of band access";
1003 Now if $baz is overloaded like this, then C<$bar> is a reference to a
1004 reference to the intermediate array, which keeps a reference to an
1005 actual array, and the access hash. The tie()ing object for the access
1006 hash is also a reference to a reference to the actual array, so
1012 There are no loops of references.
1016 Both "objects" which are blessed into the class C<two_refs1> are
1017 references to a reference to an array, thus references to a I<scalar>.
1018 Thus the accessor expression C<$$foo-E<gt>[$ind]> involves no
1019 overloaded operations.
1023 =head2 Symbolic calculator
1025 Put this in F<symbolic.pm> in your Perl library directory:
1027 package symbolic; # Primitive symbolic calculator
1028 use overload nomethod => \&wrap;
1030 sub new { shift; bless ['n', @_] }
1032 my ($obj, $other, $inv, $meth) = @_;
1033 ($obj, $other) = ($other, $obj) if $inv;
1034 bless [$meth, $obj, $other];
1037 This module is very unusual as overloaded modules go: it does not
1038 provide any usual overloaded operators, instead it provides the L<Last
1039 Resort> operator C<nomethod>. In this example the corresponding
1040 subroutine returns an object which encapsulates operations done over
1041 the objects: C<new symbolic 3> contains C<['n', 3]>, C<2 + new
1042 symbolic 3> contains C<['+', 2, ['n', 3]]>.
1044 Here is an example of the script which "calculates" the side of
1045 circumscribed octagon using the above package:
1048 my $iter = 1; # 2**($iter+2) = 8
1049 my $side = new symbolic 1;
1053 $side = (sqrt(1 + $side**2) - 1)/$side;
1057 The value of $side is
1059 ['/', ['-', ['sqrt', ['+', 1, ['**', ['n', 1], 2]],
1060 undef], 1], ['n', 1]]
1062 Note that while we obtained this value using a nice little script,
1063 there is no simple way to I<use> this value. In fact this value may
1064 be inspected in debugger (see L<perldebug>), but ony if
1065 C<bareStringify> B<O>ption is set, and not via C<p> command.
1067 If one attempts to print this value, then the overloaded operator
1068 C<""> will be called, which will call C<nomethod> operator. The
1069 result of this operator will be stringified again, but this result is
1070 again of type C<symbolic>, which will lead to an infinite loop.
1072 Add a pretty-printer method to the module F<symbolic.pm>:
1075 my ($meth, $a, $b) = @{+shift};
1076 $a = 'u' unless defined $a;
1077 $b = 'u' unless defined $b;
1078 $a = $a->pretty if ref $a;
1079 $b = $b->pretty if ref $b;
1083 Now one can finish the script by
1085 print "side = ", $side->pretty, "\n";
1087 The method C<pretty> is doing object-to-string conversion, so it
1088 is natural to overload the operator C<""> using this method. However,
1089 inside such a method it is not necessary to pretty-print the
1090 I<components> $a and $b of an object. In the above subroutine
1091 C<"[$meth $a $b]"> is a catenation of some strings and components $a
1092 and $b. If these components use overloading, the catenation operator
1093 will look for an overloaded operator C<.>, if not present, it will
1094 look for an overloaded operator C<"">. Thus it is enough to use
1096 use overload nomethod => \&wrap, '""' => \&str;
1098 my ($meth, $a, $b) = @{+shift};
1099 $a = 'u' unless defined $a;
1100 $b = 'u' unless defined $b;
1104 Now one can change the last line of the script to
1106 print "side = $side\n";
1110 side = [/ [- [sqrt [+ 1 [** [n 1 u] 2]] u] 1] [n 1 u]]
1112 and one can inspect the value in debugger using all the possible
1115 Something is is still amiss: consider the loop variable $cnt of the
1116 script. It was a number, not an object. We cannot make this value of
1117 type C<symbolic>, since then the loop will not terminate.
1119 Indeed, to terminate the cycle, the $cnt should become false.
1120 However, the operator C<bool> for checking falsity is overloaded (this
1121 time via overloaded C<"">), and returns a long string, thus any object
1122 of type C<symbolic> is true. To overcome this, we need a way to
1123 compare an object to 0. In fact, it is easier to write a numeric
1126 Here is the text of F<symbolic.pm> with such a routine added (and
1127 slightly modified str()):
1129 package symbolic; # Primitive symbolic calculator
1131 nomethod => \&wrap, '""' => \&str, '0+' => \#
1133 sub new { shift; bless ['n', @_] }
1135 my ($obj, $other, $inv, $meth) = @_;
1136 ($obj, $other) = ($other, $obj) if $inv;
1137 bless [$meth, $obj, $other];
1140 my ($meth, $a, $b) = @{+shift};
1141 $a = 'u' unless defined $a;
1148 my %subr = ( n => sub {$_[0]},
1149 sqrt => sub {sqrt $_[0]},
1150 '-' => sub {shift() - shift()},
1151 '+' => sub {shift() + shift()},
1152 '/' => sub {shift() / shift()},
1153 '*' => sub {shift() * shift()},
1154 '**' => sub {shift() ** shift()},
1157 my ($meth, $a, $b) = @{+shift};
1158 my $subr = $subr{$meth}
1159 or die "Do not know how to ($meth) in symbolic";
1160 $a = $a->num if ref $a eq __PACKAGE__;
1161 $b = $b->num if ref $b eq __PACKAGE__;
1165 All the work of numeric conversion is done in %subr and num(). Of
1166 course, %subr is not complete, it contains only operators used in the
1167 example below. Here is the extra-credit question: why do we need an
1168 explicit recursion in num()? (Answer is at the end of this section.)
1170 Use this module like this:
1173 my $iter = new symbolic 2; # 16-gon
1174 my $side = new symbolic 1;
1178 $cnt = $cnt - 1; # Mutator `--' not implemented
1179 $side = (sqrt(1 + $side**2) - 1)/$side;
1181 printf "%s=%f\n", $side, $side;
1182 printf "pi=%f\n", $side*(2**($iter+2));
1184 It prints (without so many line breaks)
1186 [/ [- [sqrt [+ 1 [** [/ [- [sqrt [+ 1 [** [n 1] 2]]] 1]
1188 [/ [- [sqrt [+ 1 [** [n 1] 2]]] 1] [n 1]]]=0.198912
1191 The above module is very primitive. It does not implement
1192 mutator methods (C<++>, C<-=> and so on), does not do deep copying
1193 (not required without mutators!), and implements only those arithmetic
1194 operations which are used in the example.
1196 To implement most arithmetic operations is easy, one should just use
1197 the tables of operations, and change the code which fills %subr to
1199 my %subr = ( 'n' => sub {$_[0]} );
1200 foreach my $op (split " ", $overload::ops{with_assign}) {
1201 $subr{$op} = $subr{"$op="} = eval "sub {shift() $op shift()}";
1203 my @bins = qw(binary 3way_comparison num_comparison str_comparison);
1204 foreach my $op (split " ", "@overload::ops{ @bins }") {
1205 $subr{$op} = eval "sub {shift() $op shift()}";
1207 foreach my $op (split " ", "@overload::ops{qw(unary func)}") {
1208 print "defining `$op'\n";
1209 $subr{$op} = eval "sub {$op shift()}";
1212 Due to L<Calling Conventions for Mutators>, we do not need anything
1213 special to make C<+=> and friends work, except filling C<+=> entry of
1214 %subr, and defining a copy constructor (needed since Perl has no
1215 way to know that the implementation of C<'+='> does not mutate
1216 the argument, compare L<Copy Constructor>).
1218 To implement a copy constructor, add C<'=' => \&cpy> to C<use overload>
1219 line, and code (this code assumes that mutators change things one level
1220 deep only, so recursive copying is not needed):
1224 bless [@$self], ref $self;
1227 To make C<++> and C<--> work, we need to implement actual mutators,
1228 either directly, or in C<nomethod>. We continue to do things inside
1229 C<nomethod>, thus add
1231 if ($meth eq '++' or $meth eq '--') {
1232 @$obj = ($meth, (bless [@$obj]), 1); # Avoid circular reference
1236 after the first line of wrap(). This is not a most effective
1237 implementation, one may consider
1239 sub inc { $_[0] = bless ['++', shift, 1]; }
1243 As a final remark, note that one can fill %subr by
1245 my %subr = ( 'n' => sub {$_[0]} );
1246 foreach my $op (split " ", $overload::ops{with_assign}) {
1247 $subr{$op} = $subr{"$op="} = eval "sub {shift() $op shift()}";
1249 my @bins = qw(binary 3way_comparison num_comparison str_comparison);
1250 foreach my $op (split " ", "@overload::ops{ @bins }") {
1251 $subr{$op} = eval "sub {shift() $op shift()}";
1253 foreach my $op (split " ", "@overload::ops{qw(unary func)}") {
1254 $subr{$op} = eval "sub {$op shift()}";
1256 $subr{'++'} = $subr{'+'};
1257 $subr{'--'} = $subr{'-'};
1259 This finishes implementation of a primitive symbolic calculator in
1260 50 lines of Perl code. Since the numeric values of subexpressions
1261 are not cached, the calculator is very slow.
1263 Here is the answer for the exercise: In the case of str(), we need no
1264 explicit recursion since the overloaded C<.>-operator will fall back
1265 to an existing overloaded operator C<"">. Overloaded arithmetic
1266 operators I<do not> fall back to numeric conversion if C<fallback> is
1267 not explicitly requested. Thus without an explicit recursion num()
1268 would convert C<['+', $a, $b]> to C<$a + $b>, which would just rebuild
1269 the argument of num().
1271 If you wonder why defaults for conversion are different for str() and
1272 num(), note how easy it was to write the symbolic calculator. This
1273 simplicity is due to an appropriate choice of defaults. One extra
1274 note: due to the explicit recursion num() is more fragile than sym():
1275 we need to explicitly check for the type of $a and $b. If components
1276 $a and $b happen to be of some related type, this may lead to problems.
1278 =head2 I<Really> symbolic calculator
1280 One may wonder why we call the above calculator symbolic. The reason
1281 is that the actual calculation of the value of expression is postponed
1282 until the value is I<used>.
1284 To see it in action, add a method
1289 @$obj->[0,1] = ('=', shift);
1292 to the package C<symbolic>. After this change one can do
1294 my $a = new symbolic 3;
1295 my $b = new symbolic 4;
1296 my $c = sqrt($a**2 + $b**2);
1298 and the numeric value of $c becomes 5. However, after calling
1300 $a->STORE(12); $b->STORE(5);
1302 the numeric value of $c becomes 13. There is no doubt now that the module
1303 symbolic provides a I<symbolic> calculator indeed.
1305 To hide the rough edges under the hood, provide a tie()d interface to the
1306 package C<symbolic> (compare with L<Metaphor clash>). Add methods
1308 sub TIESCALAR { my $pack = shift; $pack->new(@_) }
1310 sub nop { } # Around a bug
1312 (the bug is described in L<"BUGS">). One can use this new interface as
1314 tie $a, 'symbolic', 3;
1315 tie $b, 'symbolic', 4;
1316 $a->nop; $b->nop; # Around a bug
1318 my $c = sqrt($a**2 + $b**2);
1320 Now numeric value of $c is 5. After C<$a = 12; $b = 5> the numeric value
1321 of $c becomes 13. To insulate the user of the module add a method
1323 sub vars { my $p = shift; tie($_, $p), $_->nop foreach @_; }
1328 symbolic->vars($a, $b);
1329 my $c = sqrt($a**2 + $b**2);
1332 printf "c5 %s=%f\n", $c, $c;
1335 printf "c13 %s=%f\n", $c, $c;
1337 shows that the numeric value of $c follows changes to the values of $a
1342 Ilya Zakharevich E<lt>F<ilya@math.mps.ohio-state.edu>E<gt>.
1346 When Perl is run with the B<-Do> switch or its equivalent, overloading
1347 induces diagnostic messages.
1349 Using the C<m> command of Perl debugger (see L<perldebug>) one can
1350 deduce which operations are overloaded (and which ancestor triggers
1351 this overloading). Say, if C<eq> is overloaded, then the method C<(eq>
1352 is shown by debugger. The method C<()> corresponds to the C<fallback>
1353 key (in fact a presence of this method shows that this package has
1354 overloading enabled, and it is what is used by the C<Overloaded>
1355 function of module C<overload>).
1357 The module might issue the following warnings:
1361 =item Odd number of arguments for overload::constant
1363 (W) The call to overload::constant contained an odd number of arguments.
1364 The arguments should come in pairs.
1366 =item `%s' is not an overloadable type
1368 (W) You tried to overload a constant type the overload package is unaware of.
1370 =item `%s' is not a code reference
1372 (W) The second (fourth, sixth, ...) argument of overload::constant needs
1373 to be a code reference. Either an anonymous subroutine, or a reference
1380 Because it is used for overloading, the per-package hash %OVERLOAD now
1381 has a special meaning in Perl. The symbol table is filled with names
1382 looking like line-noise.
1384 For the purpose of inheritance every overloaded package behaves as if
1385 C<fallback> is present (possibly undefined). This may create
1386 interesting effects if some package is not overloaded, but inherits
1387 from two overloaded packages.
1389 Relation between overloading and tie()ing is broken. Overloading is
1390 triggered or not basing on the I<previous> class of tie()d value.
1392 This happens because the presence of overloading is checked too early,
1393 before any tie()d access is attempted. If the FETCH()ed class of the
1394 tie()d value does not change, a simple workaround is to access the value
1395 immediately after tie()ing, so that after this call the I<previous> class
1396 coincides with the current one.
1398 B<Needed:> a way to fix this without a speed penalty.
1400 Barewords are not covered by overloaded string constants.
1402 This document is confusing. There are grammos and misleading language
1403 used in places. It would seem a total rewrite is needed.