Commit | Line | Data |
---|---|---|
86530b38 AT |
1 | package overload; |
2 | ||
3 | our $VERSION = '1.00'; | |
4 | ||
5 | $overload::hint_bits = 0x20000; | |
6 | ||
7 | sub nil {} | |
8 | ||
9 | sub OVERLOAD { | |
10 | $package = shift; | |
11 | my %arg = @_; | |
12 | my ($sub, $fb); | |
13 | $ {$package . "::OVERLOAD"}{dummy}++; # Register with magic by touching. | |
14 | *{$package . "::()"} = \&nil; # Make it findable via fetchmethod. | |
15 | for (keys %arg) { | |
16 | if ($_ eq 'fallback') { | |
17 | $fb = $arg{$_}; | |
18 | } else { | |
19 | $sub = $arg{$_}; | |
20 | if (not ref $sub and $sub !~ /::/) { | |
21 | $ {$package . "::(" . $_} = $sub; | |
22 | $sub = \&nil; | |
23 | } | |
24 | #print STDERR "Setting `$ {'package'}::\cO$_' to \\&`$sub'.\n"; | |
25 | *{$package . "::(" . $_} = \&{ $sub }; | |
26 | } | |
27 | } | |
28 | ${$package . "::()"} = $fb; # Make it findable too (fallback only). | |
29 | } | |
30 | ||
31 | sub import { | |
32 | $package = (caller())[0]; | |
33 | # *{$package . "::OVERLOAD"} = \&OVERLOAD; | |
34 | shift; | |
35 | $package->overload::OVERLOAD(@_); | |
36 | } | |
37 | ||
38 | sub unimport { | |
39 | $package = (caller())[0]; | |
40 | ${$package . "::OVERLOAD"}{dummy}++; # Upgrade the table | |
41 | shift; | |
42 | for (@_) { | |
43 | if ($_ eq 'fallback') { | |
44 | undef $ {$package . "::()"}; | |
45 | } else { | |
46 | delete $ {$package . "::"}{"(" . $_}; | |
47 | } | |
48 | } | |
49 | } | |
50 | ||
51 | sub Overloaded { | |
52 | my $package = shift; | |
53 | $package = ref $package if ref $package; | |
54 | $package->can('()'); | |
55 | } | |
56 | ||
57 | sub ov_method { | |
58 | my $globref = shift; | |
59 | return undef unless $globref; | |
60 | my $sub = \&{*$globref}; | |
61 | return $sub if $sub ne \&nil; | |
62 | return shift->can($ {*$globref}); | |
63 | } | |
64 | ||
65 | sub OverloadedStringify { | |
66 | my $package = shift; | |
67 | $package = ref $package if ref $package; | |
68 | #$package->can('(""') | |
69 | ov_method mycan($package, '(""'), $package | |
70 | or ov_method mycan($package, '(0+'), $package | |
71 | or ov_method mycan($package, '(bool'), $package | |
72 | or ov_method mycan($package, '(nomethod'), $package; | |
73 | } | |
74 | ||
75 | sub Method { | |
76 | my $package = shift; | |
77 | $package = ref $package if ref $package; | |
78 | #my $meth = $package->can('(' . shift); | |
79 | ov_method mycan($package, '(' . shift), $package; | |
80 | #return $meth if $meth ne \&nil; | |
81 | #return $ {*{$meth}}; | |
82 | } | |
83 | ||
84 | sub AddrRef { | |
85 | my $package = ref $_[0]; | |
86 | return "$_[0]" unless $package; | |
87 | bless $_[0], overload::Fake; # Non-overloaded package | |
88 | my $str = "$_[0]"; | |
89 | bless $_[0], $package; # Back | |
90 | $package . substr $str, index $str, '='; | |
91 | } | |
92 | ||
93 | sub StrVal { | |
94 | (ref $_[0] && OverloadedStringify($_[0]) or ref($_[0]) eq 'Regexp') ? | |
95 | (AddrRef(shift)) : | |
96 | "$_[0]"; | |
97 | } | |
98 | ||
99 | sub mycan { # Real can would leave stubs. | |
100 | my ($package, $meth) = @_; | |
101 | return \*{$package . "::$meth"} if defined &{$package . "::$meth"}; | |
102 | my $p; | |
103 | foreach $p (@{$package . "::ISA"}) { | |
104 | my $out = mycan($p, $meth); | |
105 | return $out if $out; | |
106 | } | |
107 | return undef; | |
108 | } | |
109 | ||
110 | %constants = ( | |
111 | 'integer' => 0x1000, | |
112 | 'float' => 0x2000, | |
113 | 'binary' => 0x4000, | |
114 | 'q' => 0x8000, | |
115 | 'qr' => 0x10000, | |
116 | ); | |
117 | ||
118 | %ops = ( with_assign => "+ - * / % ** << >> x .", | |
119 | assign => "+= -= *= /= %= **= <<= >>= x= .=", | |
120 | num_comparison => "< <= > >= == !=", | |
121 | '3way_comparison'=> "<=> cmp", | |
122 | str_comparison => "lt le gt ge eq ne", | |
123 | binary => "& | ^", | |
124 | unary => "neg ! ~", | |
125 | mutators => '++ --', | |
126 | func => "atan2 cos sin exp abs log sqrt int", | |
127 | conversion => 'bool "" 0+', | |
128 | iterators => '<>', | |
129 | dereferencing => '${} @{} %{} &{} *{}', | |
130 | special => 'nomethod fallback ='); | |
131 | ||
132 | use warnings::register; | |
133 | sub constant { | |
134 | # Arguments: what, sub | |
135 | while (@_) { | |
136 | if (@_ == 1) { | |
137 | warnings::warnif ("Odd number of arguments for overload::constant"); | |
138 | last; | |
139 | } | |
140 | elsif (!exists $constants {$_ [0]}) { | |
141 | warnings::warnif ("`$_[0]' is not an overloadable type"); | |
142 | } | |
143 | elsif (!ref $_ [1] || "$_[1]" !~ /CODE\(0x[\da-f]+\)$/) { | |
144 | # Can't use C<ref $_[1] eq "CODE"> above as code references can be | |
145 | # blessed, and C<ref> would return the package the ref is blessed into. | |
146 | if (warnings::enabled) { | |
147 | $_ [1] = "undef" unless defined $_ [1]; | |
148 | warnings::warn ("`$_[1]' is not a code reference"); | |
149 | } | |
150 | } | |
151 | else { | |
152 | $^H{$_[0]} = $_[1]; | |
153 | $^H |= $constants{$_[0]} | $overload::hint_bits; | |
154 | } | |
155 | shift, shift; | |
156 | } | |
157 | } | |
158 | ||
159 | sub remove_constant { | |
160 | # Arguments: what, sub | |
161 | while (@_) { | |
162 | delete $^H{$_[0]}; | |
163 | $^H &= ~ $constants{$_[0]}; | |
164 | shift, shift; | |
165 | } | |
166 | } | |
167 | ||
168 | 1; | |
169 | ||
170 | __END__ | |
171 | ||
172 | =head1 NAME | |
173 | ||
174 | overload - Package for overloading perl operations | |
175 | ||
176 | =head1 SYNOPSIS | |
177 | ||
178 | package SomeThing; | |
179 | ||
180 | use overload | |
181 | '+' => \&myadd, | |
182 | '-' => \&mysub; | |
183 | # etc | |
184 | ... | |
185 | ||
186 | package main; | |
187 | $a = new SomeThing 57; | |
188 | $b=5+$a; | |
189 | ... | |
190 | if (overload::Overloaded $b) {...} | |
191 | ... | |
192 | $strval = overload::StrVal $b; | |
193 | ||
194 | =head1 DESCRIPTION | |
195 | ||
196 | =head2 Declaration of overloaded functions | |
197 | ||
198 | The compilation directive | |
199 | ||
200 | package Number; | |
201 | use overload | |
202 | "+" => \&add, | |
203 | "*=" => "muas"; | |
204 | ||
205 | declares function Number::add() for addition, and method muas() in | |
206 | the "class" C<Number> (or one of its base classes) | |
207 | for the assignment form C<*=> of multiplication. | |
208 | ||
209 | Arguments of this directive come in (key, value) pairs. Legal values | |
210 | are values legal inside a C<&{ ... }> call, so the name of a | |
211 | subroutine, a reference to a subroutine, or an anonymous subroutine | |
212 | will all work. Note that values specified as strings are | |
213 | interpreted as methods, not subroutines. Legal keys are listed below. | |
214 | ||
215 | The subroutine C<add> will be called to execute C<$a+$b> if $a | |
216 | is a reference to an object blessed into the package C<Number>, or if $a is | |
217 | not an object from a package with defined mathemagic addition, but $b is a | |
218 | reference to a C<Number>. It can also be called in other situations, like | |
219 | C<$a+=7>, or C<$a++>. See L<MAGIC AUTOGENERATION>. (Mathemagical | |
220 | methods refer to methods triggered by an overloaded mathematical | |
221 | operator.) | |
222 | ||
223 | Since overloading respects inheritance via the @ISA hierarchy, the | |
224 | above declaration would also trigger overloading of C<+> and C<*=> in | |
225 | all the packages which inherit from C<Number>. | |
226 | ||
227 | =head2 Calling Conventions for Binary Operations | |
228 | ||
229 | The functions specified in the C<use overload ...> directive are called | |
230 | with three (in one particular case with four, see L<Last Resort>) | |
231 | arguments. If the corresponding operation is binary, then the first | |
232 | two arguments are the two arguments of the operation. However, due to | |
233 | general object calling conventions, the first argument should always be | |
234 | an object in the package, so in the situation of C<7+$a>, the | |
235 | order of the arguments is interchanged. It probably does not matter | |
236 | when implementing the addition method, but whether the arguments | |
237 | are reversed is vital to the subtraction method. The method can | |
238 | query this information by examining the third argument, which can take | |
239 | three different values: | |
240 | ||
241 | =over 7 | |
242 | ||
243 | =item FALSE | |
244 | ||
245 | the order of arguments is as in the current operation. | |
246 | ||
247 | =item TRUE | |
248 | ||
249 | the arguments are reversed. | |
250 | ||
251 | =item C<undef> | |
252 | ||
253 | the current operation is an assignment variant (as in | |
254 | C<$a+=7>), but the usual function is called instead. This additional | |
255 | information can be used to generate some optimizations. Compare | |
256 | L<Calling Conventions for Mutators>. | |
257 | ||
258 | =back | |
259 | ||
260 | =head2 Calling Conventions for Unary Operations | |
261 | ||
262 | Unary operation are considered binary operations with the second | |
263 | argument being C<undef>. Thus the functions that overloads C<{"++"}> | |
264 | is called with arguments C<($a,undef,'')> when $a++ is executed. | |
265 | ||
266 | =head2 Calling Conventions for Mutators | |
267 | ||
268 | Two types of mutators have different calling conventions: | |
269 | ||
270 | =over | |
271 | ||
272 | =item C<++> and C<--> | |
273 | ||
274 | The routines which implement these operators are expected to actually | |
275 | I<mutate> their arguments. So, assuming that $obj is a reference to a | |
276 | number, | |
277 | ||
278 | sub incr { my $n = $ {$_[0]}; ++$n; $_[0] = bless \$n} | |
279 | ||
280 | is an appropriate implementation of overloaded C<++>. Note that | |
281 | ||
282 | sub incr { ++$ {$_[0]} ; shift } | |
283 | ||
284 | is OK if used with preincrement and with postincrement. (In the case | |
285 | of postincrement a copying will be performed, see L<Copy Constructor>.) | |
286 | ||
287 | =item C<x=> and other assignment versions | |
288 | ||
289 | There is nothing special about these methods. They may change the | |
290 | value of their arguments, and may leave it as is. The result is going | |
291 | to be assigned to the value in the left-hand-side if different from | |
292 | this value. | |
293 | ||
294 | This allows for the same method to be used as overloaded C<+=> and | |
295 | C<+>. Note that this is I<allowed>, but not recommended, since by the | |
296 | semantic of L<"Fallback"> Perl will call the method for C<+> anyway, | |
297 | if C<+=> is not overloaded. | |
298 | ||
299 | =back | |
300 | ||
301 | B<Warning.> Due to the presence of assignment versions of operations, | |
302 | routines which may be called in assignment context may create | |
303 | self-referential structures. Currently Perl will not free self-referential | |
304 | structures until cycles are C<explicitly> broken. You may get problems | |
305 | when traversing your structures too. | |
306 | ||
307 | Say, | |
308 | ||
309 | use overload '+' => sub { bless [ \$_[0], \$_[1] ] }; | |
310 | ||
311 | is asking for trouble, since for code C<$obj += $foo> the subroutine | |
312 | is called as C<$obj = add($obj, $foo, undef)>, or C<$obj = [\$obj, | |
313 | \$foo]>. If using such a subroutine is an important optimization, one | |
314 | can overload C<+=> explicitly by a non-"optimized" version, or switch | |
315 | to non-optimized version if C<not defined $_[2]> (see | |
316 | L<Calling Conventions for Binary Operations>). | |
317 | ||
318 | Even if no I<explicit> assignment-variants of operators are present in | |
319 | the script, they may be generated by the optimizer. Say, C<",$obj,"> or | |
320 | C<',' . $obj . ','> may be both optimized to | |
321 | ||
322 | my $tmp = ',' . $obj; $tmp .= ','; | |
323 | ||
324 | =head2 Overloadable Operations | |
325 | ||
326 | The following symbols can be specified in C<use overload> directive: | |
327 | ||
328 | =over 5 | |
329 | ||
330 | =item * I<Arithmetic operations> | |
331 | ||
332 | "+", "+=", "-", "-=", "*", "*=", "/", "/=", "%", "%=", | |
333 | "**", "**=", "<<", "<<=", ">>", ">>=", "x", "x=", ".", ".=", | |
334 | ||
335 | For these operations a substituted non-assignment variant can be called if | |
336 | the assignment variant is not available. Methods for operations C<+>, | |
337 | C<->, C<+=>, and C<-=> can be called to automatically generate | |
338 | increment and decrement methods. The operation C<-> can be used to | |
339 | autogenerate missing methods for unary minus or C<abs>. | |
340 | ||
341 | See L<"MAGIC AUTOGENERATION">, L<"Calling Conventions for Mutators"> and | |
342 | L<"Calling Conventions for Binary Operations">) for details of these | |
343 | substitutions. | |
344 | ||
345 | =item * I<Comparison operations> | |
346 | ||
347 | "<", "<=", ">", ">=", "==", "!=", "<=>", | |
348 | "lt", "le", "gt", "ge", "eq", "ne", "cmp", | |
349 | ||
350 | If the corresponding "spaceship" variant is available, it can be | |
351 | used to substitute for the missing operation. During C<sort>ing | |
352 | arrays, C<cmp> is used to compare values subject to C<use overload>. | |
353 | ||
354 | =item * I<Bit operations> | |
355 | ||
356 | "&", "^", "|", "neg", "!", "~", | |
357 | ||
358 | C<neg> stands for unary minus. If the method for C<neg> is not | |
359 | specified, it can be autogenerated using the method for | |
360 | subtraction. If the method for C<!> is not specified, it can be | |
361 | autogenerated using the methods for C<bool>, or C<"">, or C<0+>. | |
362 | ||
363 | =item * I<Increment and decrement> | |
364 | ||
365 | "++", "--", | |
366 | ||
367 | If undefined, addition and subtraction methods can be | |
368 | used instead. These operations are called both in prefix and | |
369 | postfix form. | |
370 | ||
371 | =item * I<Transcendental functions> | |
372 | ||
373 | "atan2", "cos", "sin", "exp", "abs", "log", "sqrt", "int" | |
374 | ||
375 | If C<abs> is unavailable, it can be autogenerated using methods | |
376 | for "E<lt>" or "E<lt>=E<gt>" combined with either unary minus or subtraction. | |
377 | ||
378 | Note that traditionally the Perl function L<int> rounds to 0, thus for | |
379 | floating-point-like types one should follow the same semantic. If | |
380 | C<int> is unavailable, it can be autogenerated using the overloading of | |
381 | C<0+>. | |
382 | ||
383 | =item * I<Boolean, string and numeric conversion> | |
384 | ||
385 | 'bool', '""', '0+', | |
386 | ||
387 | If one or two of these operations are not overloaded, the remaining ones can | |
388 | be used instead. C<bool> is used in the flow control operators | |
389 | (like C<while>) and for the ternary C<?:> operation. These functions can | |
390 | return any arbitrary Perl value. If the corresponding operation for this value | |
391 | is overloaded too, that operation will be called again with this value. | |
392 | ||
393 | As a special case if the overload returns the object itself then it will | |
394 | be used directly. An overloaded conversion returning the object is | |
395 | probably a bug, because you're likely to get something that looks like | |
396 | C<YourPackage=HASH(0x8172b34)>. | |
397 | ||
398 | =item * I<Iteration> | |
399 | ||
400 | "<>" | |
401 | ||
402 | If not overloaded, the argument will be converted to a filehandle or | |
403 | glob (which may require a stringification). The same overloading | |
404 | happens both for the I<read-filehandle> syntax C<E<lt>$varE<gt>> and | |
405 | I<globbing> syntax C<E<lt>${var}E<gt>>. | |
406 | ||
407 | B<BUGS> Even in list context, the iterator is currently called only | |
408 | once and with scalar context. | |
409 | ||
410 | =item * I<Dereferencing> | |
411 | ||
412 | '${}', '@{}', '%{}', '&{}', '*{}'. | |
413 | ||
414 | If not overloaded, the argument will be dereferenced I<as is>, thus | |
415 | should be of correct type. These functions should return a reference | |
416 | of correct type, or another object with overloaded dereferencing. | |
417 | ||
418 | As a special case if the overload returns the object itself then it | |
419 | will be used directly (provided it is the correct type). | |
420 | ||
421 | The dereference operators must be specified explicitly they will not be passed to | |
422 | "nomethod". | |
423 | ||
424 | =item * I<Special> | |
425 | ||
426 | "nomethod", "fallback", "=", | |
427 | ||
428 | see L<SPECIAL SYMBOLS FOR C<use overload>>. | |
429 | ||
430 | =back | |
431 | ||
432 | See L<"Fallback"> for an explanation of when a missing method can be | |
433 | autogenerated. | |
434 | ||
435 | A computer-readable form of the above table is available in the hash | |
436 | %overload::ops, with values being space-separated lists of names: | |
437 | ||
438 | with_assign => '+ - * / % ** << >> x .', | |
439 | assign => '+= -= *= /= %= **= <<= >>= x= .=', | |
440 | num_comparison => '< <= > >= == !=', | |
441 | '3way_comparison'=> '<=> cmp', | |
442 | str_comparison => 'lt le gt ge eq ne', | |
443 | binary => '& | ^', | |
444 | unary => 'neg ! ~', | |
445 | mutators => '++ --', | |
446 | func => 'atan2 cos sin exp abs log sqrt', | |
447 | conversion => 'bool "" 0+', | |
448 | iterators => '<>', | |
449 | dereferencing => '${} @{} %{} &{} *{}', | |
450 | special => 'nomethod fallback =' | |
451 | ||
452 | =head2 Inheritance and overloading | |
453 | ||
454 | Inheritance interacts with overloading in two ways. | |
455 | ||
456 | =over | |
457 | ||
458 | =item Strings as values of C<use overload> directive | |
459 | ||
460 | If C<value> in | |
461 | ||
462 | use overload key => value; | |
463 | ||
464 | is a string, it is interpreted as a method name. | |
465 | ||
466 | =item Overloading of an operation is inherited by derived classes | |
467 | ||
468 | Any class derived from an overloaded class is also overloaded. The | |
469 | set of overloaded methods is the union of overloaded methods of all | |
470 | the ancestors. If some method is overloaded in several ancestor, then | |
471 | which description will be used is decided by the usual inheritance | |
472 | rules: | |
473 | ||
474 | If C<A> inherits from C<B> and C<C> (in this order), C<B> overloads | |
475 | C<+> with C<\&D::plus_sub>, and C<C> overloads C<+> by C<"plus_meth">, | |
476 | then the subroutine C<D::plus_sub> will be called to implement | |
477 | operation C<+> for an object in package C<A>. | |
478 | ||
479 | =back | |
480 | ||
481 | Note that since the value of the C<fallback> key is not a subroutine, | |
482 | its inheritance is not governed by the above rules. In the current | |
483 | implementation, the value of C<fallback> in the first overloaded | |
484 | ancestor is used, but this is accidental and subject to change. | |
485 | ||
486 | =head1 SPECIAL SYMBOLS FOR C<use overload> | |
487 | ||
488 | Three keys are recognized by Perl that are not covered by the above | |
489 | description. | |
490 | ||
491 | =head2 Last Resort | |
492 | ||
493 | C<"nomethod"> should be followed by a reference to a function of four | |
494 | parameters. If defined, it is called when the overloading mechanism | |
495 | cannot find a method for some operation. The first three arguments of | |
496 | this function coincide with the arguments for the corresponding method if | |
497 | it were found, the fourth argument is the symbol | |
498 | corresponding to the missing method. If several methods are tried, | |
499 | the last one is used. Say, C<1-$a> can be equivalent to | |
500 | ||
501 | &nomethodMethod($a,1,1,"-") | |
502 | ||
503 | if the pair C<"nomethod" =E<gt> "nomethodMethod"> was specified in the | |
504 | C<use overload> directive. | |
505 | ||
506 | The C<"nomethod"> mechanism is I<not> used for the dereference operators | |
507 | ( ${} @{} %{} &{} *{} ). | |
508 | ||
509 | ||
510 | If some operation cannot be resolved, and there is no function | |
511 | assigned to C<"nomethod">, then an exception will be raised via die()-- | |
512 | unless C<"fallback"> was specified as a key in C<use overload> directive. | |
513 | ||
514 | ||
515 | =head2 Fallback | |
516 | ||
517 | The key C<"fallback"> governs what to do if a method for a particular | |
518 | operation is not found. Three different cases are possible depending on | |
519 | the value of C<"fallback">: | |
520 | ||
521 | =over 16 | |
522 | ||
523 | =item * C<undef> | |
524 | ||
525 | Perl tries to use a | |
526 | substituted method (see L<MAGIC AUTOGENERATION>). If this fails, it | |
527 | then tries to calls C<"nomethod"> value; if missing, an exception | |
528 | will be raised. | |
529 | ||
530 | =item * TRUE | |
531 | ||
532 | The same as for the C<undef> value, but no exception is raised. Instead, | |
533 | it silently reverts to what it would have done were there no C<use overload> | |
534 | present. | |
535 | ||
536 | =item * defined, but FALSE | |
537 | ||
538 | No autogeneration is tried. Perl tries to call | |
539 | C<"nomethod"> value, and if this is missing, raises an exception. | |
540 | ||
541 | =back | |
542 | ||
543 | B<Note.> C<"fallback"> inheritance via @ISA is not carved in stone | |
544 | yet, see L<"Inheritance and overloading">. | |
545 | ||
546 | =head2 Copy Constructor | |
547 | ||
548 | The value for C<"="> is a reference to a function with three | |
549 | arguments, i.e., it looks like the other values in C<use | |
550 | overload>. However, it does not overload the Perl assignment | |
551 | operator. This would go against Camel hair. | |
552 | ||
553 | This operation is called in the situations when a mutator is applied | |
554 | to a reference that shares its object with some other reference, such | |
555 | as | |
556 | ||
557 | $a=$b; | |
558 | ++$a; | |
559 | ||
560 | To make this change $a and not change $b, a copy of C<$$a> is made, | |
561 | and $a is assigned a reference to this new object. This operation is | |
562 | done during execution of the C<++$a>, and not during the assignment, | |
563 | (so before the increment C<$$a> coincides with C<$$b>). This is only | |
564 | done if C<++> is expressed via a method for C<'++'> or C<'+='> (or | |
565 | C<nomethod>). Note that if this operation is expressed via C<'+'> | |
566 | a nonmutator, i.e., as in | |
567 | ||
568 | $a=$b; | |
569 | $a=$a+1; | |
570 | ||
571 | then C<$a> does not reference a new copy of C<$$a>, since $$a does not | |
572 | appear as lvalue when the above code is executed. | |
573 | ||
574 | If the copy constructor is required during the execution of some mutator, | |
575 | but a method for C<'='> was not specified, it can be autogenerated as a | |
576 | string copy if the object is a plain scalar. | |
577 | ||
578 | =over 5 | |
579 | ||
580 | =item B<Example> | |
581 | ||
582 | The actually executed code for | |
583 | ||
584 | $a=$b; | |
585 | Something else which does not modify $a or $b.... | |
586 | ++$a; | |
587 | ||
588 | may be | |
589 | ||
590 | $a=$b; | |
591 | Something else which does not modify $a or $b.... | |
592 | $a = $a->clone(undef,""); | |
593 | $a->incr(undef,""); | |
594 | ||
595 | if $b was mathemagical, and C<'++'> was overloaded with C<\&incr>, | |
596 | C<'='> was overloaded with C<\&clone>. | |
597 | ||
598 | =back | |
599 | ||
600 | Same behaviour is triggered by C<$b = $a++>, which is consider a synonym for | |
601 | C<$b = $a; ++$a>. | |
602 | ||
603 | =head1 MAGIC AUTOGENERATION | |
604 | ||
605 | If a method for an operation is not found, and the value for C<"fallback"> is | |
606 | TRUE or undefined, Perl tries to autogenerate a substitute method for | |
607 | the missing operation based on the defined operations. Autogenerated method | |
608 | substitutions are possible for the following operations: | |
609 | ||
610 | =over 16 | |
611 | ||
612 | =item I<Assignment forms of arithmetic operations> | |
613 | ||
614 | C<$a+=$b> can use the method for C<"+"> if the method for C<"+="> | |
615 | is not defined. | |
616 | ||
617 | =item I<Conversion operations> | |
618 | ||
619 | String, numeric, and boolean conversion are calculated in terms of one | |
620 | another if not all of them are defined. | |
621 | ||
622 | =item I<Increment and decrement> | |
623 | ||
624 | The C<++$a> operation can be expressed in terms of C<$a+=1> or C<$a+1>, | |
625 | and C<$a--> in terms of C<$a-=1> and C<$a-1>. | |
626 | ||
627 | =item C<abs($a)> | |
628 | ||
629 | can be expressed in terms of C<$aE<lt>0> and C<-$a> (or C<0-$a>). | |
630 | ||
631 | =item I<Unary minus> | |
632 | ||
633 | can be expressed in terms of subtraction. | |
634 | ||
635 | =item I<Negation> | |
636 | ||
637 | C<!> and C<not> can be expressed in terms of boolean conversion, or | |
638 | string or numerical conversion. | |
639 | ||
640 | =item I<Concatenation> | |
641 | ||
642 | can be expressed in terms of string conversion. | |
643 | ||
644 | =item I<Comparison operations> | |
645 | ||
646 | can be expressed in terms of its "spaceship" counterpart: either | |
647 | C<E<lt>=E<gt>> or C<cmp>: | |
648 | ||
649 | <, >, <=, >=, ==, != in terms of <=> | |
650 | lt, gt, le, ge, eq, ne in terms of cmp | |
651 | ||
652 | =item I<Iterator> | |
653 | ||
654 | <> in terms of builtin operations | |
655 | ||
656 | =item I<Dereferencing> | |
657 | ||
658 | ${} @{} %{} &{} *{} in terms of builtin operations | |
659 | ||
660 | =item I<Copy operator> | |
661 | ||
662 | can be expressed in terms of an assignment to the dereferenced value, if this | |
663 | value is a scalar and not a reference. | |
664 | ||
665 | =back | |
666 | ||
667 | =head1 Losing overloading | |
668 | ||
669 | The restriction for the comparison operation is that even if, for example, | |
670 | `C<cmp>' should return a blessed reference, the autogenerated `C<lt>' | |
671 | function will produce only a standard logical value based on the | |
672 | numerical value of the result of `C<cmp>'. In particular, a working | |
673 | numeric conversion is needed in this case (possibly expressed in terms of | |
674 | other conversions). | |
675 | ||
676 | Similarly, C<.=> and C<x=> operators lose their mathemagical properties | |
677 | if the string conversion substitution is applied. | |
678 | ||
679 | When you chop() a mathemagical object it is promoted to a string and its | |
680 | mathemagical properties are lost. The same can happen with other | |
681 | operations as well. | |
682 | ||
683 | =head1 Run-time Overloading | |
684 | ||
685 | Since all C<use> directives are executed at compile-time, the only way to | |
686 | change overloading during run-time is to | |
687 | ||
688 | eval 'use overload "+" => \&addmethod'; | |
689 | ||
690 | You can also use | |
691 | ||
692 | eval 'no overload "+", "--", "<="'; | |
693 | ||
694 | though the use of these constructs during run-time is questionable. | |
695 | ||
696 | =head1 Public functions | |
697 | ||
698 | Package C<overload.pm> provides the following public functions: | |
699 | ||
700 | =over 5 | |
701 | ||
702 | =item overload::StrVal(arg) | |
703 | ||
704 | Gives string value of C<arg> as in absence of stringify overloading. | |
705 | ||
706 | =item overload::Overloaded(arg) | |
707 | ||
708 | Returns true if C<arg> is subject to overloading of some operations. | |
709 | ||
710 | =item overload::Method(obj,op) | |
711 | ||
712 | Returns C<undef> or a reference to the method that implements C<op>. | |
713 | ||
714 | =back | |
715 | ||
716 | =head1 Overloading constants | |
717 | ||
718 | For some application Perl parser mangles constants too much. It is possible | |
719 | to hook into this process via overload::constant() and overload::remove_constant() | |
720 | functions. | |
721 | ||
722 | These functions take a hash as an argument. The recognized keys of this hash | |
723 | are | |
724 | ||
725 | =over 8 | |
726 | ||
727 | =item integer | |
728 | ||
729 | to overload integer constants, | |
730 | ||
731 | =item float | |
732 | ||
733 | to overload floating point constants, | |
734 | ||
735 | =item binary | |
736 | ||
737 | to overload octal and hexadecimal constants, | |
738 | ||
739 | =item q | |
740 | ||
741 | to overload C<q>-quoted strings, constant pieces of C<qq>- and C<qx>-quoted | |
742 | strings and here-documents, | |
743 | ||
744 | =item qr | |
745 | ||
746 | to overload constant pieces of regular expressions. | |
747 | ||
748 | =back | |
749 | ||
750 | The corresponding values are references to functions which take three arguments: | |
751 | the first one is the I<initial> string form of the constant, the second one | |
752 | is how Perl interprets this constant, the third one is how the constant is used. | |
753 | Note that the initial string form does not | |
754 | contain string delimiters, and has backslashes in backslash-delimiter | |
755 | combinations stripped (thus the value of delimiter is not relevant for | |
756 | processing of this string). The return value of this function is how this | |
757 | constant is going to be interpreted by Perl. The third argument is undefined | |
758 | unless for overloaded C<q>- and C<qr>- constants, it is C<q> in single-quote | |
759 | context (comes from strings, regular expressions, and single-quote HERE | |
760 | documents), it is C<tr> for arguments of C<tr>/C<y> operators, | |
761 | it is C<s> for right-hand side of C<s>-operator, and it is C<qq> otherwise. | |
762 | ||
763 | Since an expression C<"ab$cd,,"> is just a shortcut for C<'ab' . $cd . ',,'>, | |
764 | it is expected that overloaded constant strings are equipped with reasonable | |
765 | overloaded catenation operator, otherwise absurd results will result. | |
766 | Similarly, negative numbers are considered as negations of positive constants. | |
767 | ||
768 | Note that it is probably meaningless to call the functions overload::constant() | |
769 | and overload::remove_constant() from anywhere but import() and unimport() methods. | |
770 | From these methods they may be called as | |
771 | ||
772 | sub import { | |
773 | shift; | |
774 | return unless @_; | |
775 | die "unknown import: @_" unless @_ == 1 and $_[0] eq ':constant'; | |
776 | overload::constant integer => sub {Math::BigInt->new(shift)}; | |
777 | } | |
778 | ||
779 | B<BUGS> Currently overloaded-ness of constants does not propagate | |
780 | into C<eval '...'>. | |
781 | ||
782 | =head1 IMPLEMENTATION | |
783 | ||
784 | What follows is subject to change RSN. | |
785 | ||
786 | The table of methods for all operations is cached in magic for the | |
787 | symbol table hash for the package. The cache is invalidated during | |
788 | processing of C<use overload>, C<no overload>, new function | |
789 | definitions, and changes in @ISA. However, this invalidation remains | |
790 | unprocessed until the next C<bless>ing into the package. Hence if you | |
791 | want to change overloading structure dynamically, you'll need an | |
792 | additional (fake) C<bless>ing to update the table. | |
793 | ||
794 | (Every SVish thing has a magic queue, and magic is an entry in that | |
795 | queue. This is how a single variable may participate in multiple | |
796 | forms of magic simultaneously. For instance, environment variables | |
797 | regularly have two forms at once: their %ENV magic and their taint | |
798 | magic. However, the magic which implements overloading is applied to | |
799 | the stashes, which are rarely used directly, thus should not slow down | |
800 | Perl.) | |
801 | ||
802 | If an object belongs to a package using overload, it carries a special | |
803 | flag. Thus the only speed penalty during arithmetic operations without | |
804 | overloading is the checking of this flag. | |
805 | ||
806 | In fact, if C<use overload> is not present, there is almost no overhead | |
807 | for overloadable operations, so most programs should not suffer | |
808 | measurable performance penalties. A considerable effort was made to | |
809 | minimize the overhead when overload is used in some package, but the | |
810 | arguments in question do not belong to packages using overload. When | |
811 | in doubt, test your speed with C<use overload> and without it. So far | |
812 | there have been no reports of substantial speed degradation if Perl is | |
813 | compiled with optimization turned on. | |
814 | ||
815 | There is no size penalty for data if overload is not used. The only | |
816 | size penalty if overload is used in some package is that I<all> the | |
817 | packages acquire a magic during the next C<bless>ing into the | |
818 | package. This magic is three-words-long for packages without | |
819 | overloading, and carries the cache table if the package is overloaded. | |
820 | ||
821 | Copying (C<$a=$b>) is shallow; however, a one-level-deep copying is | |
822 | carried out before any operation that can imply an assignment to the | |
823 | object $a (or $b) refers to, like C<$a++>. You can override this | |
824 | behavior by defining your own copy constructor (see L<"Copy Constructor">). | |
825 | ||
826 | It is expected that arguments to methods that are not explicitly supposed | |
827 | to be changed are constant (but this is not enforced). | |
828 | ||
829 | =head1 Metaphor clash | |
830 | ||
831 | One may wonder why the semantic of overloaded C<=> is so counter intuitive. | |
832 | If it I<looks> counter intuitive to you, you are subject to a metaphor | |
833 | clash. | |
834 | ||
835 | Here is a Perl object metaphor: | |
836 | ||
837 | I< object is a reference to blessed data> | |
838 | ||
839 | and an arithmetic metaphor: | |
840 | ||
841 | I< object is a thing by itself>. | |
842 | ||
843 | The I<main> problem of overloading C<=> is the fact that these metaphors | |
844 | imply different actions on the assignment C<$a = $b> if $a and $b are | |
845 | objects. Perl-think implies that $a becomes a reference to whatever | |
846 | $b was referencing. Arithmetic-think implies that the value of "object" | |
847 | $a is changed to become the value of the object $b, preserving the fact | |
848 | that $a and $b are separate entities. | |
849 | ||
850 | The difference is not relevant in the absence of mutators. After | |
851 | a Perl-way assignment an operation which mutates the data referenced by $a | |
852 | would change the data referenced by $b too. Effectively, after | |
853 | C<$a = $b> values of $a and $b become I<indistinguishable>. | |
854 | ||
855 | On the other hand, anyone who has used algebraic notation knows the | |
856 | expressive power of the arithmetic metaphor. Overloading works hard | |
857 | to enable this metaphor while preserving the Perlian way as far as | |
858 | possible. Since it is not possible to freely mix two contradicting | |
859 | metaphors, overloading allows the arithmetic way to write things I<as | |
860 | far as all the mutators are called via overloaded access only>. The | |
861 | way it is done is described in L<Copy Constructor>. | |
862 | ||
863 | If some mutator methods are directly applied to the overloaded values, | |
864 | one may need to I<explicitly unlink> other values which references the | |
865 | same value: | |
866 | ||
867 | $a = new Data 23; | |
868 | ... | |
869 | $b = $a; # $b is "linked" to $a | |
870 | ... | |
871 | $a = $a->clone; # Unlink $b from $a | |
872 | $a->increment_by(4); | |
873 | ||
874 | Note that overloaded access makes this transparent: | |
875 | ||
876 | $a = new Data 23; | |
877 | $b = $a; # $b is "linked" to $a | |
878 | $a += 4; # would unlink $b automagically | |
879 | ||
880 | However, it would not make | |
881 | ||
882 | $a = new Data 23; | |
883 | $a = 4; # Now $a is a plain 4, not 'Data' | |
884 | ||
885 | preserve "objectness" of $a. But Perl I<has> a way to make assignments | |
886 | to an object do whatever you want. It is just not the overload, but | |
887 | tie()ing interface (see L<perlfunc/tie>). Adding a FETCH() method | |
888 | which returns the object itself, and STORE() method which changes the | |
889 | value of the object, one can reproduce the arithmetic metaphor in its | |
890 | completeness, at least for variables which were tie()d from the start. | |
891 | ||
892 | (Note that a workaround for a bug may be needed, see L<"BUGS">.) | |
893 | ||
894 | =head1 Cookbook | |
895 | ||
896 | Please add examples to what follows! | |
897 | ||
898 | =head2 Two-face scalars | |
899 | ||
900 | Put this in F<two_face.pm> in your Perl library directory: | |
901 | ||
902 | package two_face; # Scalars with separate string and | |
903 | # numeric values. | |
904 | sub new { my $p = shift; bless [@_], $p } | |
905 | use overload '""' => \&str, '0+' => \&num, fallback => 1; | |
906 | sub num {shift->[1]} | |
907 | sub str {shift->[0]} | |
908 | ||
909 | Use it as follows: | |
910 | ||
911 | require two_face; | |
912 | my $seven = new two_face ("vii", 7); | |
913 | printf "seven=$seven, seven=%d, eight=%d\n", $seven, $seven+1; | |
914 | print "seven contains `i'\n" if $seven =~ /i/; | |
915 | ||
916 | (The second line creates a scalar which has both a string value, and a | |
917 | numeric value.) This prints: | |
918 | ||
919 | seven=vii, seven=7, eight=8 | |
920 | seven contains `i' | |
921 | ||
922 | =head2 Two-face references | |
923 | ||
924 | Suppose you want to create an object which is accessible as both an | |
925 | array reference and a hash reference, similar to the | |
926 | L<pseudo-hash|perlref/"Pseudo-hashes: Using an array as a hash"> | |
927 | builtin Perl type. Let's make it better than a pseudo-hash by | |
928 | allowing index 0 to be treated as a normal element. | |
929 | ||
930 | package two_refs; | |
931 | use overload '%{}' => \&gethash, '@{}' => sub { $ {shift()} }; | |
932 | sub new { | |
933 | my $p = shift; | |
934 | bless \ [@_], $p; | |
935 | } | |
936 | sub gethash { | |
937 | my %h; | |
938 | my $self = shift; | |
939 | tie %h, ref $self, $self; | |
940 | \%h; | |
941 | } | |
942 | ||
943 | sub TIEHASH { my $p = shift; bless \ shift, $p } | |
944 | my %fields; | |
945 | my $i = 0; | |
946 | $fields{$_} = $i++ foreach qw{zero one two three}; | |
947 | sub STORE { | |
948 | my $self = ${shift()}; | |
949 | my $key = $fields{shift()}; | |
950 | defined $key or die "Out of band access"; | |
951 | $$self->[$key] = shift; | |
952 | } | |
953 | sub FETCH { | |
954 | my $self = ${shift()}; | |
955 | my $key = $fields{shift()}; | |
956 | defined $key or die "Out of band access"; | |
957 | $$self->[$key]; | |
958 | } | |
959 | ||
960 | Now one can access an object using both the array and hash syntax: | |
961 | ||
962 | my $bar = new two_refs 3,4,5,6; | |
963 | $bar->[2] = 11; | |
964 | $bar->{two} == 11 or die 'bad hash fetch'; | |
965 | ||
966 | Note several important features of this example. First of all, the | |
967 | I<actual> type of $bar is a scalar reference, and we do not overload | |
968 | the scalar dereference. Thus we can get the I<actual> non-overloaded | |
969 | contents of $bar by just using C<$$bar> (what we do in functions which | |
970 | overload dereference). Similarly, the object returned by the | |
971 | TIEHASH() method is a scalar reference. | |
972 | ||
973 | Second, we create a new tied hash each time the hash syntax is used. | |
974 | This allows us not to worry about a possibility of a reference loop, | |
975 | which would lead to a memory leak. | |
976 | ||
977 | Both these problems can be cured. Say, if we want to overload hash | |
978 | dereference on a reference to an object which is I<implemented> as a | |
979 | hash itself, the only problem one has to circumvent is how to access | |
980 | this I<actual> hash (as opposed to the I<virtual> hash exhibited by the | |
981 | overloaded dereference operator). Here is one possible fetching routine: | |
982 | ||
983 | sub access_hash { | |
984 | my ($self, $key) = (shift, shift); | |
985 | my $class = ref $self; | |
986 | bless $self, 'overload::dummy'; # Disable overloading of %{} | |
987 | my $out = $self->{$key}; | |
988 | bless $self, $class; # Restore overloading | |
989 | $out; | |
990 | } | |
991 | ||
992 | To remove creation of the tied hash on each access, one may an extra | |
993 | level of indirection which allows a non-circular structure of references: | |
994 | ||
995 | package two_refs1; | |
996 | use overload '%{}' => sub { ${shift()}->[1] }, | |
997 | '@{}' => sub { ${shift()}->[0] }; | |
998 | sub new { | |
999 | my $p = shift; | |
1000 | my $a = [@_]; | |
1001 | my %h; | |
1002 | tie %h, $p, $a; | |
1003 | bless \ [$a, \%h], $p; | |
1004 | } | |
1005 | sub gethash { | |
1006 | my %h; | |
1007 | my $self = shift; | |
1008 | tie %h, ref $self, $self; | |
1009 | \%h; | |
1010 | } | |
1011 | ||
1012 | sub TIEHASH { my $p = shift; bless \ shift, $p } | |
1013 | my %fields; | |
1014 | my $i = 0; | |
1015 | $fields{$_} = $i++ foreach qw{zero one two three}; | |
1016 | sub STORE { | |
1017 | my $a = ${shift()}; | |
1018 | my $key = $fields{shift()}; | |
1019 | defined $key or die "Out of band access"; | |
1020 | $a->[$key] = shift; | |
1021 | } | |
1022 | sub FETCH { | |
1023 | my $a = ${shift()}; | |
1024 | my $key = $fields{shift()}; | |
1025 | defined $key or die "Out of band access"; | |
1026 | $a->[$key]; | |
1027 | } | |
1028 | ||
1029 | Now if $baz is overloaded like this, then C<$baz> is a reference to a | |
1030 | reference to the intermediate array, which keeps a reference to an | |
1031 | actual array, and the access hash. The tie()ing object for the access | |
1032 | hash is a reference to a reference to the actual array, so | |
1033 | ||
1034 | =over | |
1035 | ||
1036 | =item * | |
1037 | ||
1038 | There are no loops of references. | |
1039 | ||
1040 | =item * | |
1041 | ||
1042 | Both "objects" which are blessed into the class C<two_refs1> are | |
1043 | references to a reference to an array, thus references to a I<scalar>. | |
1044 | Thus the accessor expression C<$$foo-E<gt>[$ind]> involves no | |
1045 | overloaded operations. | |
1046 | ||
1047 | =back | |
1048 | ||
1049 | =head2 Symbolic calculator | |
1050 | ||
1051 | Put this in F<symbolic.pm> in your Perl library directory: | |
1052 | ||
1053 | package symbolic; # Primitive symbolic calculator | |
1054 | use overload nomethod => \&wrap; | |
1055 | ||
1056 | sub new { shift; bless ['n', @_] } | |
1057 | sub wrap { | |
1058 | my ($obj, $other, $inv, $meth) = @_; | |
1059 | ($obj, $other) = ($other, $obj) if $inv; | |
1060 | bless [$meth, $obj, $other]; | |
1061 | } | |
1062 | ||
1063 | This module is very unusual as overloaded modules go: it does not | |
1064 | provide any usual overloaded operators, instead it provides the L<Last | |
1065 | Resort> operator C<nomethod>. In this example the corresponding | |
1066 | subroutine returns an object which encapsulates operations done over | |
1067 | the objects: C<new symbolic 3> contains C<['n', 3]>, C<2 + new | |
1068 | symbolic 3> contains C<['+', 2, ['n', 3]]>. | |
1069 | ||
1070 | Here is an example of the script which "calculates" the side of | |
1071 | circumscribed octagon using the above package: | |
1072 | ||
1073 | require symbolic; | |
1074 | my $iter = 1; # 2**($iter+2) = 8 | |
1075 | my $side = new symbolic 1; | |
1076 | my $cnt = $iter; | |
1077 | ||
1078 | while ($cnt--) { | |
1079 | $side = (sqrt(1 + $side**2) - 1)/$side; | |
1080 | } | |
1081 | print "OK\n"; | |
1082 | ||
1083 | The value of $side is | |
1084 | ||
1085 | ['/', ['-', ['sqrt', ['+', 1, ['**', ['n', 1], 2]], | |
1086 | undef], 1], ['n', 1]] | |
1087 | ||
1088 | Note that while we obtained this value using a nice little script, | |
1089 | there is no simple way to I<use> this value. In fact this value may | |
1090 | be inspected in debugger (see L<perldebug>), but ony if | |
1091 | C<bareStringify> B<O>ption is set, and not via C<p> command. | |
1092 | ||
1093 | If one attempts to print this value, then the overloaded operator | |
1094 | C<""> will be called, which will call C<nomethod> operator. The | |
1095 | result of this operator will be stringified again, but this result is | |
1096 | again of type C<symbolic>, which will lead to an infinite loop. | |
1097 | ||
1098 | Add a pretty-printer method to the module F<symbolic.pm>: | |
1099 | ||
1100 | sub pretty { | |
1101 | my ($meth, $a, $b) = @{+shift}; | |
1102 | $a = 'u' unless defined $a; | |
1103 | $b = 'u' unless defined $b; | |
1104 | $a = $a->pretty if ref $a; | |
1105 | $b = $b->pretty if ref $b; | |
1106 | "[$meth $a $b]"; | |
1107 | } | |
1108 | ||
1109 | Now one can finish the script by | |
1110 | ||
1111 | print "side = ", $side->pretty, "\n"; | |
1112 | ||
1113 | The method C<pretty> is doing object-to-string conversion, so it | |
1114 | is natural to overload the operator C<""> using this method. However, | |
1115 | inside such a method it is not necessary to pretty-print the | |
1116 | I<components> $a and $b of an object. In the above subroutine | |
1117 | C<"[$meth $a $b]"> is a catenation of some strings and components $a | |
1118 | and $b. If these components use overloading, the catenation operator | |
1119 | will look for an overloaded operator C<.>; if not present, it will | |
1120 | look for an overloaded operator C<"">. Thus it is enough to use | |
1121 | ||
1122 | use overload nomethod => \&wrap, '""' => \&str; | |
1123 | sub str { | |
1124 | my ($meth, $a, $b) = @{+shift}; | |
1125 | $a = 'u' unless defined $a; | |
1126 | $b = 'u' unless defined $b; | |
1127 | "[$meth $a $b]"; | |
1128 | } | |
1129 | ||
1130 | Now one can change the last line of the script to | |
1131 | ||
1132 | print "side = $side\n"; | |
1133 | ||
1134 | which outputs | |
1135 | ||
1136 | side = [/ [- [sqrt [+ 1 [** [n 1 u] 2]] u] 1] [n 1 u]] | |
1137 | ||
1138 | and one can inspect the value in debugger using all the possible | |
1139 | methods. | |
1140 | ||
1141 | Something is still amiss: consider the loop variable $cnt of the | |
1142 | script. It was a number, not an object. We cannot make this value of | |
1143 | type C<symbolic>, since then the loop will not terminate. | |
1144 | ||
1145 | Indeed, to terminate the cycle, the $cnt should become false. | |
1146 | However, the operator C<bool> for checking falsity is overloaded (this | |
1147 | time via overloaded C<"">), and returns a long string, thus any object | |
1148 | of type C<symbolic> is true. To overcome this, we need a way to | |
1149 | compare an object to 0. In fact, it is easier to write a numeric | |
1150 | conversion routine. | |
1151 | ||
1152 | Here is the text of F<symbolic.pm> with such a routine added (and | |
1153 | slightly modified str()): | |
1154 | ||
1155 | package symbolic; # Primitive symbolic calculator | |
1156 | use overload | |
1157 | nomethod => \&wrap, '""' => \&str, '0+' => \# | |
1158 | ||
1159 | sub new { shift; bless ['n', @_] } | |
1160 | sub wrap { | |
1161 | my ($obj, $other, $inv, $meth) = @_; | |
1162 | ($obj, $other) = ($other, $obj) if $inv; | |
1163 | bless [$meth, $obj, $other]; | |
1164 | } | |
1165 | sub str { | |
1166 | my ($meth, $a, $b) = @{+shift}; | |
1167 | $a = 'u' unless defined $a; | |
1168 | if (defined $b) { | |
1169 | "[$meth $a $b]"; | |
1170 | } else { | |
1171 | "[$meth $a]"; | |
1172 | } | |
1173 | } | |
1174 | my %subr = ( n => sub {$_[0]}, | |
1175 | sqrt => sub {sqrt $_[0]}, | |
1176 | '-' => sub {shift() - shift()}, | |
1177 | '+' => sub {shift() + shift()}, | |
1178 | '/' => sub {shift() / shift()}, | |
1179 | '*' => sub {shift() * shift()}, | |
1180 | '**' => sub {shift() ** shift()}, | |
1181 | ); | |
1182 | sub num { | |
1183 | my ($meth, $a, $b) = @{+shift}; | |
1184 | my $subr = $subr{$meth} | |
1185 | or die "Do not know how to ($meth) in symbolic"; | |
1186 | $a = $a->num if ref $a eq __PACKAGE__; | |
1187 | $b = $b->num if ref $b eq __PACKAGE__; | |
1188 | $subr->($a,$b); | |
1189 | } | |
1190 | ||
1191 | All the work of numeric conversion is done in %subr and num(). Of | |
1192 | course, %subr is not complete, it contains only operators used in the | |
1193 | example below. Here is the extra-credit question: why do we need an | |
1194 | explicit recursion in num()? (Answer is at the end of this section.) | |
1195 | ||
1196 | Use this module like this: | |
1197 | ||
1198 | require symbolic; | |
1199 | my $iter = new symbolic 2; # 16-gon | |
1200 | my $side = new symbolic 1; | |
1201 | my $cnt = $iter; | |
1202 | ||
1203 | while ($cnt) { | |
1204 | $cnt = $cnt - 1; # Mutator `--' not implemented | |
1205 | $side = (sqrt(1 + $side**2) - 1)/$side; | |
1206 | } | |
1207 | printf "%s=%f\n", $side, $side; | |
1208 | printf "pi=%f\n", $side*(2**($iter+2)); | |
1209 | ||
1210 | It prints (without so many line breaks) | |
1211 | ||
1212 | [/ [- [sqrt [+ 1 [** [/ [- [sqrt [+ 1 [** [n 1] 2]]] 1] | |
1213 | [n 1]] 2]]] 1] | |
1214 | [/ [- [sqrt [+ 1 [** [n 1] 2]]] 1] [n 1]]]=0.198912 | |
1215 | pi=3.182598 | |
1216 | ||
1217 | The above module is very primitive. It does not implement | |
1218 | mutator methods (C<++>, C<-=> and so on), does not do deep copying | |
1219 | (not required without mutators!), and implements only those arithmetic | |
1220 | operations which are used in the example. | |
1221 | ||
1222 | To implement most arithmetic operations is easy; one should just use | |
1223 | the tables of operations, and change the code which fills %subr to | |
1224 | ||
1225 | my %subr = ( 'n' => sub {$_[0]} ); | |
1226 | foreach my $op (split " ", $overload::ops{with_assign}) { | |
1227 | $subr{$op} = $subr{"$op="} = eval "sub {shift() $op shift()}"; | |
1228 | } | |
1229 | my @bins = qw(binary 3way_comparison num_comparison str_comparison); | |
1230 | foreach my $op (split " ", "@overload::ops{ @bins }") { | |
1231 | $subr{$op} = eval "sub {shift() $op shift()}"; | |
1232 | } | |
1233 | foreach my $op (split " ", "@overload::ops{qw(unary func)}") { | |
1234 | print "defining `$op'\n"; | |
1235 | $subr{$op} = eval "sub {$op shift()}"; | |
1236 | } | |
1237 | ||
1238 | Due to L<Calling Conventions for Mutators>, we do not need anything | |
1239 | special to make C<+=> and friends work, except filling C<+=> entry of | |
1240 | %subr, and defining a copy constructor (needed since Perl has no | |
1241 | way to know that the implementation of C<'+='> does not mutate | |
1242 | the argument, compare L<Copy Constructor>). | |
1243 | ||
1244 | To implement a copy constructor, add C<< '=' => \&cpy >> to C<use overload> | |
1245 | line, and code (this code assumes that mutators change things one level | |
1246 | deep only, so recursive copying is not needed): | |
1247 | ||
1248 | sub cpy { | |
1249 | my $self = shift; | |
1250 | bless [@$self], ref $self; | |
1251 | } | |
1252 | ||
1253 | To make C<++> and C<--> work, we need to implement actual mutators, | |
1254 | either directly, or in C<nomethod>. We continue to do things inside | |
1255 | C<nomethod>, thus add | |
1256 | ||
1257 | if ($meth eq '++' or $meth eq '--') { | |
1258 | @$obj = ($meth, (bless [@$obj]), 1); # Avoid circular reference | |
1259 | return $obj; | |
1260 | } | |
1261 | ||
1262 | after the first line of wrap(). This is not a most effective | |
1263 | implementation, one may consider | |
1264 | ||
1265 | sub inc { $_[0] = bless ['++', shift, 1]; } | |
1266 | ||
1267 | instead. | |
1268 | ||
1269 | As a final remark, note that one can fill %subr by | |
1270 | ||
1271 | my %subr = ( 'n' => sub {$_[0]} ); | |
1272 | foreach my $op (split " ", $overload::ops{with_assign}) { | |
1273 | $subr{$op} = $subr{"$op="} = eval "sub {shift() $op shift()}"; | |
1274 | } | |
1275 | my @bins = qw(binary 3way_comparison num_comparison str_comparison); | |
1276 | foreach my $op (split " ", "@overload::ops{ @bins }") { | |
1277 | $subr{$op} = eval "sub {shift() $op shift()}"; | |
1278 | } | |
1279 | foreach my $op (split " ", "@overload::ops{qw(unary func)}") { | |
1280 | $subr{$op} = eval "sub {$op shift()}"; | |
1281 | } | |
1282 | $subr{'++'} = $subr{'+'}; | |
1283 | $subr{'--'} = $subr{'-'}; | |
1284 | ||
1285 | This finishes implementation of a primitive symbolic calculator in | |
1286 | 50 lines of Perl code. Since the numeric values of subexpressions | |
1287 | are not cached, the calculator is very slow. | |
1288 | ||
1289 | Here is the answer for the exercise: In the case of str(), we need no | |
1290 | explicit recursion since the overloaded C<.>-operator will fall back | |
1291 | to an existing overloaded operator C<"">. Overloaded arithmetic | |
1292 | operators I<do not> fall back to numeric conversion if C<fallback> is | |
1293 | not explicitly requested. Thus without an explicit recursion num() | |
1294 | would convert C<['+', $a, $b]> to C<$a + $b>, which would just rebuild | |
1295 | the argument of num(). | |
1296 | ||
1297 | If you wonder why defaults for conversion are different for str() and | |
1298 | num(), note how easy it was to write the symbolic calculator. This | |
1299 | simplicity is due to an appropriate choice of defaults. One extra | |
1300 | note: due to the explicit recursion num() is more fragile than sym(): | |
1301 | we need to explicitly check for the type of $a and $b. If components | |
1302 | $a and $b happen to be of some related type, this may lead to problems. | |
1303 | ||
1304 | =head2 I<Really> symbolic calculator | |
1305 | ||
1306 | One may wonder why we call the above calculator symbolic. The reason | |
1307 | is that the actual calculation of the value of expression is postponed | |
1308 | until the value is I<used>. | |
1309 | ||
1310 | To see it in action, add a method | |
1311 | ||
1312 | sub STORE { | |
1313 | my $obj = shift; | |
1314 | $#$obj = 1; | |
1315 | @$obj->[0,1] = ('=', shift); | |
1316 | } | |
1317 | ||
1318 | to the package C<symbolic>. After this change one can do | |
1319 | ||
1320 | my $a = new symbolic 3; | |
1321 | my $b = new symbolic 4; | |
1322 | my $c = sqrt($a**2 + $b**2); | |
1323 | ||
1324 | and the numeric value of $c becomes 5. However, after calling | |
1325 | ||
1326 | $a->STORE(12); $b->STORE(5); | |
1327 | ||
1328 | the numeric value of $c becomes 13. There is no doubt now that the module | |
1329 | symbolic provides a I<symbolic> calculator indeed. | |
1330 | ||
1331 | To hide the rough edges under the hood, provide a tie()d interface to the | |
1332 | package C<symbolic> (compare with L<Metaphor clash>). Add methods | |
1333 | ||
1334 | sub TIESCALAR { my $pack = shift; $pack->new(@_) } | |
1335 | sub FETCH { shift } | |
1336 | sub nop { } # Around a bug | |
1337 | ||
1338 | (the bug is described in L<"BUGS">). One can use this new interface as | |
1339 | ||
1340 | tie $a, 'symbolic', 3; | |
1341 | tie $b, 'symbolic', 4; | |
1342 | $a->nop; $b->nop; # Around a bug | |
1343 | ||
1344 | my $c = sqrt($a**2 + $b**2); | |
1345 | ||
1346 | Now numeric value of $c is 5. After C<$a = 12; $b = 5> the numeric value | |
1347 | of $c becomes 13. To insulate the user of the module add a method | |
1348 | ||
1349 | sub vars { my $p = shift; tie($_, $p), $_->nop foreach @_; } | |
1350 | ||
1351 | Now | |
1352 | ||
1353 | my ($a, $b); | |
1354 | symbolic->vars($a, $b); | |
1355 | my $c = sqrt($a**2 + $b**2); | |
1356 | ||
1357 | $a = 3; $b = 4; | |
1358 | printf "c5 %s=%f\n", $c, $c; | |
1359 | ||
1360 | $a = 12; $b = 5; | |
1361 | printf "c13 %s=%f\n", $c, $c; | |
1362 | ||
1363 | shows that the numeric value of $c follows changes to the values of $a | |
1364 | and $b. | |
1365 | ||
1366 | =head1 AUTHOR | |
1367 | ||
1368 | Ilya Zakharevich E<lt>F<ilya@math.mps.ohio-state.edu>E<gt>. | |
1369 | ||
1370 | =head1 DIAGNOSTICS | |
1371 | ||
1372 | When Perl is run with the B<-Do> switch or its equivalent, overloading | |
1373 | induces diagnostic messages. | |
1374 | ||
1375 | Using the C<m> command of Perl debugger (see L<perldebug>) one can | |
1376 | deduce which operations are overloaded (and which ancestor triggers | |
1377 | this overloading). Say, if C<eq> is overloaded, then the method C<(eq> | |
1378 | is shown by debugger. The method C<()> corresponds to the C<fallback> | |
1379 | key (in fact a presence of this method shows that this package has | |
1380 | overloading enabled, and it is what is used by the C<Overloaded> | |
1381 | function of module C<overload>). | |
1382 | ||
1383 | The module might issue the following warnings: | |
1384 | ||
1385 | =over 4 | |
1386 | ||
1387 | =item Odd number of arguments for overload::constant | |
1388 | ||
1389 | (W) The call to overload::constant contained an odd number of arguments. | |
1390 | The arguments should come in pairs. | |
1391 | ||
1392 | =item `%s' is not an overloadable type | |
1393 | ||
1394 | (W) You tried to overload a constant type the overload package is unaware of. | |
1395 | ||
1396 | =item `%s' is not a code reference | |
1397 | ||
1398 | (W) The second (fourth, sixth, ...) argument of overload::constant needs | |
1399 | to be a code reference. Either an anonymous subroutine, or a reference | |
1400 | to a subroutine. | |
1401 | ||
1402 | =back | |
1403 | ||
1404 | =head1 BUGS | |
1405 | ||
1406 | Because it is used for overloading, the per-package hash %OVERLOAD now | |
1407 | has a special meaning in Perl. The symbol table is filled with names | |
1408 | looking like line-noise. | |
1409 | ||
1410 | For the purpose of inheritance every overloaded package behaves as if | |
1411 | C<fallback> is present (possibly undefined). This may create | |
1412 | interesting effects if some package is not overloaded, but inherits | |
1413 | from two overloaded packages. | |
1414 | ||
1415 | Relation between overloading and tie()ing is broken. Overloading is | |
1416 | triggered or not basing on the I<previous> class of tie()d value. | |
1417 | ||
1418 | This happens because the presence of overloading is checked too early, | |
1419 | before any tie()d access is attempted. If the FETCH()ed class of the | |
1420 | tie()d value does not change, a simple workaround is to access the value | |
1421 | immediately after tie()ing, so that after this call the I<previous> class | |
1422 | coincides with the current one. | |
1423 | ||
1424 | B<Needed:> a way to fix this without a speed penalty. | |
1425 | ||
1426 | Barewords are not covered by overloaded string constants. | |
1427 | ||
1428 | This document is confusing. There are grammos and misleading language | |
1429 | used in places. It would seem a total rewrite is needed. | |
1430 | ||
1431 | =cut | |
1432 |