1PERLOBJ(1)             Perl Programmers Reference Guide             PERLOBJ(1)
2
3
4

NAME

6       perlobj - Perl objects
7

DESCRIPTION

9       First you need to understand what references are in Perl.  See perlref
10       for that.  Second, if you still find the following reference work too
11       complicated, a tutorial on object-oriented programming in Perl can be
12       found in perltoot and perltooc.
13
14       If you're still with us, then here are three very simple definitions
15       that you should find reassuring.
16
17       1.  An object is simply a reference that happens to know which class it
18           belongs to.
19
20       2.  A class is simply a package that happens to provide methods to deal
21           with object references.
22
23       3.  A method is simply a subroutine that expects an object reference
24           (or a package name, for class methods) as the first argument.
25
26       We'll cover these points now in more depth.
27
28       An Object is Simply a Reference
29
30       Unlike say C++, Perl doesn't provide any special syntax for construc‐
31       tors.  A constructor is merely a subroutine that returns a reference to
32       something "blessed" into a class, generally the class that the subrou‐
33       tine is defined in.  Here is a typical constructor:
34
35           package Critter;
36           sub new { bless {} }
37
38       That word "new" isn't special.  You could have written a construct this
39       way, too:
40
41           package Critter;
42           sub spawn { bless {} }
43
44       This might even be preferable, because the C++ programmers won't be
45       tricked into thinking that "new" works in Perl as it does in C++.  It
46       doesn't.  We recommend that you name your constructors whatever makes
47       sense in the context of the problem you're solving.  For example, con‐
48       structors in the Tk extension to Perl are named after the widgets they
49       create.
50
51       One thing that's different about Perl constructors compared with those
52       in C++ is that in Perl, they have to allocate their own memory.  (The
53       other things is that they don't automatically call overridden base-
54       class constructors.)  The "{}" allocates an anonymous hash containing
55       no key/value pairs, and returns it  The bless() takes that reference
56       and tells the object it references that it's now a Critter, and returns
57       the reference.  This is for convenience, because the referenced object
58       itself knows that it has been blessed, and the reference to it could
59       have been returned directly, like this:
60
61           sub new {
62               my $self = {};
63               bless $self;
64               return $self;
65           }
66
67       You often see such a thing in more complicated constructors that wish
68       to call methods in the class as part of the construction:
69
70           sub new {
71               my $self = {};
72               bless $self;
73               $self->initialize();
74               return $self;
75           }
76
77       If you care about inheritance (and you should; see "Modules: Creation,
78       Use, and Abuse" in perlmodlib), then you want to use the two-arg form
79       of bless so that your constructors may be inherited:
80
81           sub new {
82               my $class = shift;
83               my $self = {};
84               bless $self, $class;
85               $self->initialize();
86               return $self;
87           }
88
89       Or if you expect people to call not just "CLASS->new()" but also
90       "$obj->new()", then use something like the following.  (Note that using
91       this to call new() on an instance does not automatically perform any
92       copying.  If you want a shallow or deep copy of an object, you'll have
93       to specifically allow for that.)  The initialize() method used will be
94       of whatever $class we blessed the object into:
95
96           sub new {
97               my $this = shift;
98               my $class = ref($this) ⎪⎪ $this;
99               my $self = {};
100               bless $self, $class;
101               $self->initialize();
102               return $self;
103           }
104
105       Within the class package, the methods will typically deal with the ref‐
106       erence as an ordinary reference.  Outside the class package, the refer‐
107       ence is generally treated as an opaque value that may be accessed only
108       through the class's methods.
109
110       Although a constructor can in theory re-bless a referenced object cur‐
111       rently belonging to another class, this is almost certainly going to
112       get you into trouble.  The new class is responsible for all cleanup
113       later.  The previous blessing is forgotten, as an object may belong to
114       only one class at a time.  (Although of course it's free to inherit
115       methods from many classes.)  If you find yourself having to do this,
116       the parent class is probably misbehaving, though.
117
118       A clarification:  Perl objects are blessed.  References are not.
119       Objects know which package they belong to.  References do not.  The
120       bless() function uses the reference to find the object.  Consider the
121       following example:
122
123           $a = {};
124           $b = $a;
125           bless $a, BLAH;
126           print "\$b is a ", ref($b), "\n";
127
128       This reports $b as being a BLAH, so obviously bless() operated on the
129       object and not on the reference.
130
131       A Class is Simply a Package
132
133       Unlike say C++, Perl doesn't provide any special syntax for class defi‐
134       nitions.  You use a package as a class by putting method definitions
135       into the class.
136
137       There is a special array within each package called @ISA, which says
138       where else to look for a method if you can't find it in the current
139       package.  This is how Perl implements inheritance.  Each element of the
140       @ISA array is just the name of another package that happens to be a
141       class package.  The classes are searched (depth first) for missing
142       methods in the order that they occur in @ISA.  The classes accessible
143       through @ISA are known as base classes of the current class.
144
145       All classes implicitly inherit from class "UNIVERSAL" as their last
146       base class.  Several commonly used methods are automatically supplied
147       in the UNIVERSAL class; see "Default UNIVERSAL methods" for more
148       details.
149
150       If a missing method is found in a base class, it is cached in the cur‐
151       rent class for efficiency.  Changing @ISA or defining new subroutines
152       invalidates the cache and causes Perl to do the lookup again.
153
154       If neither the current class, its named base classes, nor the UNIVERSAL
155       class contains the requested method, these three places are searched
156       all over again, this time looking for a method named AUTOLOAD().  If an
157       AUTOLOAD is found, this method is called on behalf of the missing
158       method, setting the package global $AUTOLOAD to be the fully qualified
159       name of the method that was intended to be called.
160
161       If none of that works, Perl finally gives up and complains.
162
163       If you want to stop the AUTOLOAD inheritance say simply
164
165               sub AUTOLOAD;
166
167       and the call will die using the name of the sub being called.
168
169       Perl classes do method inheritance only.  Data inheritance is left up
170       to the class itself.  By and large, this is not a problem in Perl,
171       because most classes model the attributes of their object using an
172       anonymous hash, which serves as its own little namespace to be carved
173       up by the various classes that might want to do something with the
174       object.  The only problem with this is that you can't sure that you
175       aren't using a piece of the hash that isn't already used.  A reasonable
176       workaround is to prepend your fieldname in the hash with the package
177       name.
178
179           sub bump {
180               my $self = shift;
181               $self->{ __PACKAGE__ . ".count"}++;
182           }
183
184       A Method is Simply a Subroutine
185
186       Unlike say C++, Perl doesn't provide any special syntax for method def‐
187       inition.  (It does provide a little syntax for method invocation
188       though.  More on that later.)  A method expects its first argument to
189       be the object (reference) or package (string) it is being invoked on.
190       There are two ways of calling methods, which we'll call class methods
191       and instance methods.
192
193       A class method expects a class name as the first argument.  It provides
194       functionality for the class as a whole, not for any individual object
195       belonging to the class.  Constructors are often class methods, but see
196       perltoot and perltooc for alternatives.  Many class methods simply
197       ignore their first argument, because they already know what package
198       they're in and don't care what package they were invoked via.  (These
199       aren't necessarily the same, because class methods follow the inheri‐
200       tance tree just like ordinary instance methods.)  Another typical use
201       for class methods is to look up an object by name:
202
203           sub find {
204               my ($class, $name) = @_;
205               $objtable{$name};
206           }
207
208       An instance method expects an object reference as its first argument.
209       Typically it shifts the first argument into a "self" or "this" vari‐
210       able, and then uses that as an ordinary reference.
211
212           sub display {
213               my $self = shift;
214               my @keys = @_ ? @_ : sort keys %$self;
215               foreach $key (@keys) {
216                   print "\t$key => $self->{$key}\n";
217               }
218           }
219
220       Method Invocation
221
222       For various historical and other reasons, Perl offers two equivalent
223       ways to write a method call.  The simpler and more common way is to use
224       the arrow notation:
225
226           my $fred = Critter->find("Fred");
227           $fred->display("Height", "Weight");
228
229       You should already be familiar with the use of the "->" operator with
230       references.  In fact, since $fred above is a reference to an object,
231       you could think of the method call as just another form of dereferenc‐
232       ing.
233
234       Whatever is on the left side of the arrow, whether a reference or a
235       class name, is passed to the method subroutine as its first argument.
236       So the above code is mostly equivalent to:
237
238           my $fred = Critter::find("Critter", "Fred");
239           Critter::display($fred, "Height", "Weight");
240
241       How does Perl know which package the subroutine is in?  By looking at
242       the left side of the arrow, which must be either a package name or a
243       reference to an object, i.e. something that has been blessed to a pack‐
244       age.  Either way, that's the package where Perl starts looking.  If
245       that package has no subroutine with that name, Perl starts looking for
246       it in any base classes of that package, and so on.
247
248       If you need to, you can force Perl to start looking in some other pack‐
249       age:
250
251           my $barney = MyCritter->Critter::find("Barney");
252           $barney->Critter::display("Height", "Weight");
253
254       Here "MyCritter" is presumably a subclass of "Critter" that defines its
255       own versions of find() and display().  We haven't specified what those
256       methods do, but that doesn't matter above since we've forced Perl to
257       start looking for the subroutines in "Critter".
258
259       As a special case of the above, you may use the "SUPER" pseudo-class to
260       tell Perl to start looking for the method in the packages named in the
261       current class's @ISA list.
262
263           package MyCritter;
264           use base 'Critter';    # sets @MyCritter::ISA = ('Critter');
265
266           sub display {
267               my ($self, @args) = @_;
268               $self->SUPER::display("Name", @args);
269           }
270
271       It is important to note that "SUPER" refers to the superclass(es) of
272       the current package and not to the superclass(es) of the object. Also,
273       the "SUPER" pseudo-class can only currently be used as a modifier to a
274       method name, but not in any of the other ways that class names are nor‐
275       mally used, eg:
276
277           something->SUPER::method(...);      # OK
278           SUPER::method(...);                 # WRONG
279           SUPER->method(...);                 # WRONG
280
281       Instead of a class name or an object reference, you can also use any
282       expression that returns either of those on the left side of the arrow.
283       So the following statement is valid:
284
285           Critter->find("Fred")->display("Height", "Weight");
286
287       and so is the following:
288
289           my $fred = (reverse "rettirC")->find(reverse "derF");
290
291       The right side of the arrow typically is the method name, but a simple
292       scalar variable containing either the method name or a subroutine ref‐
293       erence can also be used.
294
295       Indirect Object Syntax
296
297       The other way to invoke a method is by using the so-called "indirect
298       object" notation.  This syntax was available in Perl 4 long before
299       objects were introduced, and is still used with filehandles like this:
300
301          print STDERR "help!!!\n";
302
303       The same syntax can be used to call either object or class methods.
304
305          my $fred = find Critter "Fred";
306          display $fred "Height", "Weight";
307
308       Notice that there is no comma between the object or class name and the
309       parameters.  This is how Perl can tell you want an indirect method call
310       instead of an ordinary subroutine call.
311
312       But what if there are no arguments?  In that case, Perl must guess what
313       you want.  Even worse, it must make that guess at compile time.  Usu‐
314       ally Perl gets it right, but when it doesn't you get a function call
315       compiled as a method, or vice versa.  This can introduce subtle bugs
316       that are hard to detect.
317
318       For example, a call to a method "new" in indirect notation -- as C++
319       programmers are wont to make -- can be miscompiled into a subroutine
320       call if there's already a "new" function in scope.  You'd end up call‐
321       ing the current package's "new" as a subroutine, rather than the
322       desired class's method.  The compiler tries to cheat by remembering
323       bareword "require"s, but the grief when it messes up just isn't worth
324       the years of debugging it will take you to track down such subtle bugs.
325
326       There is another problem with this syntax: the indirect object is lim‐
327       ited to a name, a scalar variable, or a block, because it would have to
328       do too much lookahead otherwise, just like any other postfix derefer‐
329       ence in the language.  (These are the same quirky rules as are used for
330       the filehandle slot in functions like "print" and "printf".)  This can
331       lead to horribly confusing precedence problems, as in these next two
332       lines:
333
334           move $obj->{FIELD};                 # probably wrong!
335           move $ary[$i];                      # probably wrong!
336
337       Those actually parse as the very surprising:
338
339           $obj->move->{FIELD};                # Well, lookee here
340           $ary->move([$i]);                   # Didn't expect this one, eh?
341
342       Rather than what you might have expected:
343
344           $obj->{FIELD}->move();              # You should be so lucky.
345           $ary[$i]->move;                     # Yeah, sure.
346
347       To get the correct behavior with indirect object syntax, you would have
348       to use a block around the indirect object:
349
350           move {$obj->{FIELD}};
351           move {$ary[$i]};
352
353       Even then, you still have the same potential problem if there happens
354       to be a function named "move" in the current package.  The "->" nota‐
355       tion suffers from neither of these disturbing ambiguities, so we recom‐
356       mend you use it exclusively.  However, you may still end up having to
357       read code using the indirect object notation, so it's important to be
358       familiar with it.
359
360       Default UNIVERSAL methods
361
362       The "UNIVERSAL" package automatically contains the following methods
363       that are inherited by all other classes:
364
365       isa(CLASS)
366           "isa" returns true if its object is blessed into a subclass of
367           "CLASS"
368
369           You can also call "UNIVERSAL::isa" as a subroutine with two argu‐
370           ments.  Of course, this will do the wrong thing if someone has
371           overridden "isa" in a class, so don't do it.
372
373           If you need to determine whether you've received a valid invocant,
374           use the "blessed" function from Scalar::Util:
375
376               if (blessed($ref) && $ref->isa( 'Some::Class')) {
377                   # ...
378               }
379
380           "blessed" returns the name of the package the argument has been
381           blessed into, or "undef".
382
383       can(METHOD)
384           "can" checks to see if its object has a method called "METHOD", if
385           it does then a reference to the sub is returned, if it does not
386           then undef is returned.
387
388           "UNIVERSAL::can" can also be called as a subroutine with two argu‐
389           ments.  It'll always return undef if its first argument isn't an
390           object or a class name.  The same caveats for calling "UNIVER‐
391           SAL::isa" directly apply here, too.
392
393       VERSION( [NEED] )
394           "VERSION" returns the version number of the class (package).  If
395           the NEED argument is given then it will check that the current ver‐
396           sion (as defined by the $VERSION variable in the given package) not
397           less than NEED; it will die if this is not the case.  This method
398           is normally called as a class method.  This method is called auto‐
399           matically by the "VERSION" form of "use".
400
401               use A 1.2 qw(some imported subs);
402               # implies:
403               A->VERSION(1.2);
404
405       NOTE: "can" directly uses Perl's internal code for method lookup, and
406       "isa" uses a very similar method and cache-ing strategy. This may cause
407       strange effects if the Perl code dynamically changes @ISA in any pack‐
408       age.
409
410       You may add other methods to the UNIVERSAL class via Perl or XS code.
411       You do not need to "use UNIVERSAL" to make these methods available to
412       your program (and you should not do so).
413
414       Destructors
415
416       When the last reference to an object goes away, the object is automati‐
417       cally destroyed.  (This may even be after you exit, if you've stored
418       references in global variables.)  If you want to capture control just
419       before the object is freed, you may define a DESTROY method in your
420       class.  It will automatically be called at the appropriate moment, and
421       you can do any extra cleanup you need to do.  Perl passes a reference
422       to the object under destruction as the first (and only) argument.
423       Beware that the reference is a read-only value, and cannot be modified
424       by manipulating $_[0] within the destructor.  The object itself (i.e.
425       the thingy the reference points to, namely "${$_[0]}", "@{$_[0]}",
426       "%{$_[0]}" etc.) is not similarly constrained.
427
428       Since DESTROY methods can be called at unpredictable times, it is
429       important that you localise any global variables that the method may
430       update.  In particular, localise $@ if you use "eval {}" and localise
431       $? if you use "system" or backticks.
432
433       If you arrange to re-bless the reference before the destructor returns,
434       perl will again call the DESTROY method for the re-blessed object after
435       the current one returns.  This can be used for clean delegation of
436       object destruction, or for ensuring that destructors in the base
437       classes of your choosing get called.  Explicitly calling DESTROY is
438       also possible, but is usually never needed.
439
440       Do not confuse the previous discussion with how objects CONTAINED in
441       the current one are destroyed.  Such objects will be freed and
442       destroyed automatically when the current object is freed, provided no
443       other references to them exist elsewhere.
444
445       Summary
446
447       That's about all there is to it.  Now you need just to go off and buy a
448       book about object-oriented design methodology, and bang your forehead
449       with it for the next six months or so.
450
451       Two-Phased Garbage Collection
452
453       For most purposes, Perl uses a fast and simple, reference-based garbage
454       collection system.  That means there's an extra dereference going on at
455       some level, so if you haven't built your Perl executable using your C
456       compiler's "-O" flag, performance will suffer.  If you have built Perl
457       with "cc -O", then this probably won't matter.
458
459       A more serious concern is that unreachable memory with a non-zero ref‐
460       erence count will not normally get freed.  Therefore, this is a bad
461       idea:
462
463           {
464               my $a;
465               $a = \$a;
466           }
467
468       Even thought $a should go away, it can't.  When building recursive data
469       structures, you'll have to break the self-reference yourself explicitly
470       if you don't care to leak.  For example, here's a self-referential node
471       such as one might use in a sophisticated tree structure:
472
473           sub new_node {
474               my $class = shift;
475               my $node  = {};
476               $node->{LEFT} = $node->{RIGHT} = $node;
477               $node->{DATA} = [ @_ ];
478               return bless $node => $class;
479           }
480
481       If you create nodes like that, they (currently) won't go away unless
482       you break their self reference yourself.  (In other words, this is not
483       to be construed as a feature, and you shouldn't depend on it.)
484
485       Almost.
486
487       When an interpreter thread finally shuts down (usually when your pro‐
488       gram exits), then a rather costly but complete mark-and-sweep style of
489       garbage collection is performed, and everything allocated by that
490       thread gets destroyed.  This is essential to support Perl as an embed‐
491       ded or a multithreadable language.  For example, this program demon‐
492       strates Perl's two-phased garbage collection:
493
494           #!/usr/bin/perl
495           package Subtle;
496
497           sub new {
498               my $test;
499               $test = \$test;
500               warn "CREATING " . \$test;
501               return bless \$test;
502           }
503
504           sub DESTROY {
505               my $self = shift;
506               warn "DESTROYING $self";
507           }
508
509           package main;
510
511           warn "starting program";
512           {
513               my $a = Subtle->new;
514               my $b = Subtle->new;
515               $$a = 0;  # break selfref
516               warn "leaving block";
517           }
518
519           warn "just exited block";
520           warn "time to die...";
521           exit;
522
523       When run as /foo/test, the following output is produced:
524
525           starting program at /foo/test line 18.
526           CREATING SCALAR(0x8e5b8) at /foo/test line 7.
527           CREATING SCALAR(0x8e57c) at /foo/test line 7.
528           leaving block at /foo/test line 23.
529           DESTROYING Subtle=SCALAR(0x8e5b8) at /foo/test line 13.
530           just exited block at /foo/test line 26.
531           time to die... at /foo/test line 27.
532           DESTROYING Subtle=SCALAR(0x8e57c) during global destruction.
533
534       Notice that "global destruction" bit there?  That's the thread garbage
535       collector reaching the unreachable.
536
537       Objects are always destructed, even when regular refs aren't.  Objects
538       are destructed in a separate pass before ordinary refs just to prevent
539       object destructors from using refs that have been themselves destruc‐
540       ted.  Plain refs are only garbage-collected if the destruct level is
541       greater than 0.  You can test the higher levels of global destruction
542       by setting the PERL_DESTRUCT_LEVEL environment variable, presuming
543       "-DDEBUGGING" was enabled during perl build time.  See
544       "PERL_DESTRUCT_LEVEL" in perlhack for more information.
545
546       A more complete garbage collection strategy will be implemented at a
547       future date.
548
549       In the meantime, the best solution is to create a non-recursive con‐
550       tainer class that holds a pointer to the self-referential data struc‐
551       ture.  Define a DESTROY method for the containing object's class that
552       manually breaks the circularities in the self-referential structure.
553

SEE ALSO

555       A kinder, gentler tutorial on object-oriented programming in Perl can
556       be found in perltoot, perlboot and perltooc.  You should also check out
557       perlbot for other object tricks, traps, and tips, as well as perlmodlib
558       for some style guides on constructing both modules and classes.
559
560
561
562perl v5.8.8                       2006-01-07                        PERLOBJ(1)
Impressum