2 X<subroutine> X<function>
4 perlsub - Perl subroutines
8 To declare subroutines:
9 X<subroutine, declaration> X<sub>
11 sub NAME; # A "forward" declaration.
12 sub NAME(PROTO); # ditto, but with prototypes
13 sub NAME : ATTRS; # with attributes
14 sub NAME(PROTO) : ATTRS; # with attributes and prototypes
16 sub NAME BLOCK # A declaration and a definition.
17 sub NAME(PROTO) BLOCK # ditto, but with prototypes
18 sub NAME : ATTRS BLOCK # with attributes
19 sub NAME(PROTO) : ATTRS BLOCK # with prototypes and attributes
21 To define an anonymous subroutine at runtime:
22 X<subroutine, anonymous>
24 $subref = sub BLOCK; # no proto
25 $subref = sub (PROTO) BLOCK; # with proto
26 $subref = sub : ATTRS BLOCK; # with attributes
27 $subref = sub (PROTO) : ATTRS BLOCK; # with proto and attributes
29 To import subroutines:
32 use MODULE qw(NAME1 NAME2 NAME3);
35 X<subroutine, call> X<call>
37 NAME(LIST); # & is optional with parentheses.
38 NAME LIST; # Parentheses optional if predeclared/imported.
39 &NAME(LIST); # Circumvent prototypes.
40 &NAME; # Makes current @_ visible to called subroutine.
44 Like many languages, Perl provides for user-defined subroutines.
45 These may be located anywhere in the main program, loaded in from
46 other files via the C<do>, C<require>, or C<use> keywords, or
47 generated on the fly using C<eval> or anonymous subroutines.
48 You can even call a function indirectly using a variable containing
49 its name or a CODE reference.
51 The Perl model for function call and return values is simple: all
52 functions are passed as parameters one single flat list of scalars, and
53 all functions likewise return to their caller one single flat list of
54 scalars. Any arrays or hashes in these call and return lists will
55 collapse, losing their identities--but you may always use
56 pass-by-reference instead to avoid this. Both call and return lists may
57 contain as many or as few scalar elements as you'd like. (Often a
58 function without an explicit return statement is called a subroutine, but
59 there's really no difference from Perl's perspective.)
60 X<subroutine, parameter> X<parameter>
62 Any arguments passed in show up in the array C<@_>. Therefore, if
63 you called a function with two arguments, those would be stored in
64 C<$_[0]> and C<$_[1]>. The array C<@_> is a local array, but its
65 elements are aliases for the actual scalar parameters. In particular,
66 if an element C<$_[0]> is updated, the corresponding argument is
67 updated (or an error occurs if it is not updatable). If an argument
68 is an array or hash element which did not exist when the function
69 was called, that element is created only when (and if) it is modified
70 or a reference to it is taken. (Some earlier versions of Perl
71 created the element whether or not the element was assigned to.)
72 Assigning to the whole array C<@_> removes that aliasing, and does
73 not update any arguments.
74 X<subroutine, argument> X<argument> X<@_>
76 A C<return> statement may be used to exit a subroutine, optionally
77 specifying the returned value, which will be evaluated in the
78 appropriate context (list, scalar, or void) depending on the context of
79 the subroutine call. If you specify no return value, the subroutine
80 returns an empty list in list context, the undefined value in scalar
81 context, or nothing in void context. If you return one or more
82 aggregates (arrays and hashes), these will be flattened together into
83 one large indistinguishable list.
85 If no C<return> is found and if the last statement is an expression, its
86 value is returned. If the last statement is a loop control structure
87 like a C<foreach> or a C<while>, the returned value is unspecified. The
88 empty sub returns the empty list.
89 X<subroutine, return value> X<return value> X<return>
91 Perl does not have named formal parameters. In practice all you
92 do is assign to a C<my()> list of these. Variables that aren't
93 declared to be private are global variables. For gory details
94 on creating private variables, see L<"Private Variables via my()">
95 and L<"Temporary Values via local()">. To create protected
96 environments for a set of functions in a separate package (and
97 probably a separate file), see L<perlmod/"Packages">.
98 X<formal parameter> X<parameter, formal>
105 $max = $foo if $max < $foo;
109 $bestday = max($mon,$tue,$wed,$thu,$fri);
113 # get a line, combining continuation lines
114 # that start with whitespace
117 $thisline = $lookahead; # global variables!
118 LINE: while (defined($lookahead = <STDIN>)) {
119 if ($lookahead =~ /^[ \t]/) {
120 $thisline .= $lookahead;
129 $lookahead = <STDIN>; # get first line
130 while (defined($line = get_line())) {
134 Assigning to a list of private variables to name your arguments:
137 my($key, $value) = @_;
138 $Foo{$key} = $value unless $Foo{$key};
141 Because the assignment copies the values, this also has the effect
142 of turning call-by-reference into call-by-value. Otherwise a
143 function is free to do in-place modifications of C<@_> and change
145 X<call-by-reference> X<call-by-value>
147 upcase_in($v1, $v2); # this changes $v1 and $v2
149 for (@_) { tr/a-z/A-Z/ }
152 You aren't allowed to modify constants in this way, of course. If an
153 argument were actually literal and you tried to change it, you'd take a
154 (presumably fatal) exception. For example, this won't work:
155 X<call-by-reference> X<call-by-value>
157 upcase_in("frederick");
159 It would be much safer if the C<upcase_in()> function
160 were written to return a copy of its parameters instead
161 of changing them in place:
163 ($v3, $v4) = upcase($v1, $v2); # this doesn't change $v1 and $v2
165 return unless defined wantarray; # void context, do nothing
167 for (@parms) { tr/a-z/A-Z/ }
168 return wantarray ? @parms : $parms[0];
171 Notice how this (unprototyped) function doesn't care whether it was
172 passed real scalars or arrays. Perl sees all arguments as one big,
173 long, flat parameter list in C<@_>. This is one area where
174 Perl's simple argument-passing style shines. The C<upcase()>
175 function would work perfectly well without changing the C<upcase()>
176 definition even if we fed it things like this:
178 @newlist = upcase(@list1, @list2);
179 @newlist = upcase( split /:/, $var );
181 Do not, however, be tempted to do this:
183 (@a, @b) = upcase(@list1, @list2);
185 Like the flattened incoming parameter list, the return list is also
186 flattened on return. So all you have managed to do here is stored
187 everything in C<@a> and made C<@b> empty. See
188 L<Pass by Reference> for alternatives.
190 A subroutine may be called using an explicit C<&> prefix. The
191 C<&> is optional in modern Perl, as are parentheses if the
192 subroutine has been predeclared. The C<&> is I<not> optional
193 when just naming the subroutine, such as when it's used as
194 an argument to defined() or undef(). Nor is it optional when you
195 want to do an indirect subroutine call with a subroutine name or
196 reference using the C<&$subref()> or C<&{$subref}()> constructs,
197 although the C<< $subref->() >> notation solves that problem.
198 See L<perlref> for more about all that.
201 Subroutines may be called recursively. If a subroutine is called
202 using the C<&> form, the argument list is optional, and if omitted,
203 no C<@_> array is set up for the subroutine: the C<@_> array at the
204 time of the call is visible to subroutine instead. This is an
205 efficiency mechanism that new users may wish to avoid.
208 &foo(1,2,3); # pass three arguments
209 foo(1,2,3); # the same
211 foo(); # pass a null list
214 &foo; # foo() get current args, like foo(@_) !!
215 foo; # like foo() IFF sub foo predeclared, else "foo"
217 Not only does the C<&> form make the argument list optional, it also
218 disables any prototype checking on arguments you do provide. This
219 is partly for historical reasons, and partly for having a convenient way
220 to cheat if you know what you're doing. See L<Prototypes> below.
223 Subroutines whose names are in all upper case are reserved to the Perl
224 core, as are modules whose names are in all lower case. A subroutine in
225 all capitals is a loosely-held convention meaning it will be called
226 indirectly by the run-time system itself, usually due to a triggered event.
227 Subroutines that do special, pre-defined things include C<AUTOLOAD>, C<CLONE>,
228 C<DESTROY> plus all functions mentioned in L<perltie> and L<PerlIO::via>.
230 The C<BEGIN>, C<CHECK>, C<INIT> and C<END> subroutines are not so much
231 subroutines as named special code blocks, of which you can have more
232 than one in a package, and which you can B<not> call explicitly. See
233 L<perlmod/"BEGIN, CHECK, INIT and END">
235 =head2 Private Variables via my()
236 X<my> X<variable, lexical> X<lexical> X<lexical variable> X<scope, lexical>
237 X<lexical scope> X<attributes, my>
241 my $foo; # declare $foo lexically local
242 my (@wid, %get); # declare list of variables local
243 my $foo = "flurp"; # declare $foo lexical, and init it
244 my @oof = @bar; # declare @oof lexical, and init it
245 my $x : Foo = $y; # similar, with an attribute applied
247 B<WARNING>: The use of attribute lists on C<my> declarations is still
248 evolving. The current semantics and interface are subject to change.
249 See L<attributes> and L<Attribute::Handlers>.
251 The C<my> operator declares the listed variables to be lexically
252 confined to the enclosing block, conditional (C<if/unless/elsif/else>),
253 loop (C<for/foreach/while/until/continue>), subroutine, C<eval>,
254 or C<do/require/use>'d file. If more than one value is listed, the
255 list must be placed in parentheses. All listed elements must be
256 legal lvalues. Only alphanumeric identifiers may be lexically
257 scoped--magical built-ins like C<$/> must currently be C<local>ized
258 with C<local> instead.
260 Unlike dynamic variables created by the C<local> operator, lexical
261 variables declared with C<my> are totally hidden from the outside
262 world, including any called subroutines. This is true if it's the
263 same subroutine called from itself or elsewhere--every call gets
267 This doesn't mean that a C<my> variable declared in a statically
268 enclosing lexical scope would be invisible. Only dynamic scopes
269 are cut off. For example, the C<bumpx()> function below has access
270 to the lexical $x variable because both the C<my> and the C<sub>
271 occurred at the same scope, presumably file scope.
276 An C<eval()>, however, can see lexical variables of the scope it is
277 being evaluated in, so long as the names aren't hidden by declarations within
278 the C<eval()> itself. See L<perlref>.
281 The parameter list to my() may be assigned to if desired, which allows you
282 to initialize your variables. (If no initializer is given for a
283 particular variable, it is created with the undefined value.) Commonly
284 this is used to name input parameters to a subroutine. Examples:
286 $arg = "fred"; # "global" variable
288 print "$arg thinks the root is $n\n";
289 fred thinks the root is 3
292 my $arg = shift; # name doesn't matter
297 The C<my> is simply a modifier on something you might assign to. So when
298 you do assign to variables in its argument list, C<my> doesn't
299 change whether those variables are viewed as a scalar or an array. So
301 my ($foo) = <STDIN>; # WRONG?
304 both supply a list context to the right-hand side, while
308 supplies a scalar context. But the following declares only one variable:
310 my $foo, $bar = 1; # WRONG
312 That has the same effect as
317 The declared variable is not introduced (is not visible) until after
318 the current statement. Thus,
322 can be used to initialize a new $x with the value of the old $x, and
325 my $x = 123 and $x == 123
327 is false unless the old $x happened to have the value C<123>.
329 Lexical scopes of control structures are not bounded precisely by the
330 braces that delimit their controlled blocks; control expressions are
331 part of that scope, too. Thus in the loop
333 while (my $line = <>) {
339 the scope of $line extends from its declaration throughout the rest of
340 the loop construct (including the C<continue> clause), but not beyond
341 it. Similarly, in the conditional
343 if ((my $answer = <STDIN>) =~ /^yes$/i) {
345 } elsif ($answer =~ /^no$/i) {
349 die "'$answer' is neither 'yes' nor 'no'";
352 the scope of $answer extends from its declaration through the rest
353 of that conditional, including any C<elsif> and C<else> clauses,
354 but not beyond it. See L<perlsyn/"Simple statements"> for information
355 on the scope of variables in statements with modifiers.
357 The C<foreach> loop defaults to scoping its index variable dynamically
358 in the manner of C<local>. However, if the index variable is
359 prefixed with the keyword C<my>, or if there is already a lexical
360 by that name in scope, then a new lexical is created instead. Thus
364 for my $i (1, 2, 3) {
368 the scope of $i extends to the end of the loop, but not beyond it,
369 rendering the value of $i inaccessible within C<some_function()>.
372 Some users may wish to encourage the use of lexically scoped variables.
373 As an aid to catching implicit uses to package variables,
374 which are always global, if you say
378 then any variable mentioned from there to the end of the enclosing
379 block must either refer to a lexical variable, be predeclared via
380 C<our> or C<use vars>, or else must be fully qualified with the package name.
381 A compilation error results otherwise. An inner block may countermand
382 this with C<no strict 'vars'>.
384 A C<my> has both a compile-time and a run-time effect. At compile
385 time, the compiler takes notice of it. The principal usefulness
386 of this is to quiet C<use strict 'vars'>, but it is also essential
387 for generation of closures as detailed in L<perlref>. Actual
388 initialization is delayed until run time, though, so it gets executed
389 at the appropriate time, such as each time through a loop, for
392 Variables declared with C<my> are not part of any package and are therefore
393 never fully qualified with the package name. In particular, you're not
394 allowed to try to make a package variable (or other global) lexical:
396 my $pack::var; # ERROR! Illegal syntax
398 In fact, a dynamic variable (also known as package or global variables)
399 are still accessible using the fully qualified C<::> notation even while a
400 lexical of the same name is also visible:
405 print "$x and $::x\n";
407 That will print out C<20> and C<10>.
409 You may declare C<my> variables at the outermost scope of a file
410 to hide any such identifiers from the world outside that file. This
411 is similar in spirit to C's static variables when they are used at
412 the file level. To do this with a subroutine requires the use of
413 a closure (an anonymous function that accesses enclosing lexicals).
414 If you want to create a private subroutine that cannot be called
415 from outside that block, it can declare a lexical variable containing
416 an anonymous sub reference:
418 my $secret_version = '1.001-beta';
419 my $secret_sub = sub { print $secret_version };
422 As long as the reference is never returned by any function within the
423 module, no outside module can see the subroutine, because its name is not in
424 any package's symbol table. Remember that it's not I<REALLY> called
425 C<$some_pack::secret_version> or anything; it's just $secret_version,
426 unqualified and unqualifiable.
428 This does not work with object methods, however; all object methods
429 have to be in the symbol table of some package to be found. See
430 L<perlref/"Function Templates"> for something of a work-around to
433 =head2 Persistent Private Variables
434 X<state> X<state variable> X<static> X<variable, persistent> X<variable, static> X<closure>
436 There are two ways to build persistent private variables in Perl 5.10.
437 First, you can simply use the C<state> feature. Or, you can use closures,
438 if you want to stay compatible with releases older than 5.10.
440 =head3 Persistent variables via state()
442 Beginning with perl 5.9.4, you can declare variables with the C<state>
443 keyword in place of C<my>. For that to work, though, you must have
444 enabled that feature beforehand, either by using the C<feature> pragma, or
445 by using C<-E> on one-liners. (see L<feature>)
447 For example, the following code maintains a private counter, incremented
448 each time the gimme_another() function is called:
451 sub gimme_another { state $x; return ++$x }
453 Also, since C<$x> is lexical, it can't be reached or modified by any Perl
456 You can initialize state variables, and the assigment will be executed
459 sub starts_from_42 { state $x = 42; return ++$x }
461 You can also, as a syntactic shortcut, initialize more than one if they're
462 all declared within the same state() clause:
464 state ($a, $b, $c) = ( 'one', 'two', 'three' );
466 However, be warned that state variables declared as part of a list will
467 get assigned each time the statement will be executed, since it will be
468 considered as a regular list assigment, not one to be executed only once:
470 (state $x, my $y) = (1, 2); # $x gets reinitialized every time !
472 B<Caveat>: the code at the right side of the assignment to a state
473 variable will be executed every time; only the assignment is disabled. So,
474 avoid code that has side-effects, or that is slow to execute. This might
475 be optimized out in a future version of Perl.
477 =head3 Persistent variables with closures
479 Just because a lexical variable is lexically (also called statically)
480 scoped to its enclosing block, C<eval>, or C<do> FILE, this doesn't mean that
481 within a function it works like a C static. It normally works more
482 like a C auto, but with implicit garbage collection.
484 Unlike local variables in C or C++, Perl's lexical variables don't
485 necessarily get recycled just because their scope has exited.
486 If something more permanent is still aware of the lexical, it will
487 stick around. So long as something else references a lexical, that
488 lexical won't be freed--which is as it should be. You wouldn't want
489 memory being free until you were done using it, or kept around once you
490 were done. Automatic garbage collection takes care of this for you.
492 This means that you can pass back or save away references to lexical
493 variables, whereas to return a pointer to a C auto is a grave error.
494 It also gives us a way to simulate C's function statics. Here's a
495 mechanism for giving a function private variables with both lexical
496 scoping and a static lifetime. If you do want to create something like
497 C's static variables, just enclose the whole function in an extra block,
498 and put the static variable outside the function but in the block.
503 return ++$secret_val;
506 # $secret_val now becomes unreachable by the outside
507 # world, but retains its value between calls to gimme_another
509 If this function is being sourced in from a separate file
510 via C<require> or C<use>, then this is probably just fine. If it's
511 all in the main program, you'll need to arrange for the C<my>
512 to be executed early, either by putting the whole block above
513 your main program, or more likely, placing merely a C<BEGIN>
514 code block around it to make sure it gets executed before your program
520 return ++$secret_val;
524 See L<perlmod/"BEGIN, CHECK, INIT and END"> about the
525 special triggered code blocks, C<BEGIN>, C<CHECK>, C<INIT> and C<END>.
527 If declared at the outermost scope (the file scope), then lexicals
528 work somewhat like C's file statics. They are available to all
529 functions in that same file declared below them, but are inaccessible
530 from outside that file. This strategy is sometimes used in modules
531 to create private variables that the whole module can see.
533 =head2 Temporary Values via local()
534 X<local> X<scope, dynamic> X<dynamic scope> X<variable, local>
535 X<variable, temporary>
537 B<WARNING>: In general, you should be using C<my> instead of C<local>, because
538 it's faster and safer. Exceptions to this include the global punctuation
539 variables, global filehandles and formats, and direct manipulation of the
540 Perl symbol table itself. C<local> is mostly used when the current value
541 of a variable must be visible to called subroutines.
545 # localization of values
547 local $foo; # make $foo dynamically local
548 local (@wid, %get); # make list of variables local
549 local $foo = "flurp"; # make $foo dynamic, and init it
550 local @oof = @bar; # make @oof dynamic, and init it
552 local $hash{key} = "val"; # sets a local value for this hash entry
553 local ($cond ? $v1 : $v2); # several types of lvalues support
556 # localization of symbols
558 local *FH; # localize $FH, @FH, %FH, &FH ...
559 local *merlyn = *randal; # now $merlyn is really $randal, plus
560 # @merlyn is really @randal, etc
561 local *merlyn = 'randal'; # SAME THING: promote 'randal' to *randal
562 local *merlyn = \$randal; # just alias $merlyn, not @merlyn etc
564 A C<local> modifies its listed variables to be "local" to the
565 enclosing block, C<eval>, or C<do FILE>--and to I<any subroutine
566 called from within that block>. A C<local> just gives temporary
567 values to global (meaning package) variables. It does I<not> create
568 a local variable. This is known as dynamic scoping. Lexical scoping
569 is done with C<my>, which works more like C's auto declarations.
571 Some types of lvalues can be localized as well : hash and array elements
572 and slices, conditionals (provided that their result is always
573 localizable), and symbolic references. As for simple variables, this
574 creates new, dynamically scoped values.
576 If more than one variable or expression is given to C<local>, they must be
577 placed in parentheses. This operator works
578 by saving the current values of those variables in its argument list on a
579 hidden stack and restoring them upon exiting the block, subroutine, or
580 eval. This means that called subroutines can also reference the local
581 variable, but not the global one. The argument list may be assigned to if
582 desired, which allows you to initialize your local variables. (If no
583 initializer is given for a particular variable, it is created with an
586 Because C<local> is a run-time operator, it gets executed each time
587 through a loop. Consequently, it's more efficient to localize your
588 variables outside the loop.
590 =head3 Grammatical note on local()
593 A C<local> is simply a modifier on an lvalue expression. When you assign to
594 a C<local>ized variable, the C<local> doesn't change whether its list is viewed
595 as a scalar or an array. So
597 local($foo) = <STDIN>;
598 local @FOO = <STDIN>;
600 both supply a list context to the right-hand side, while
602 local $foo = <STDIN>;
604 supplies a scalar context.
606 =head3 Localization of special variables
607 X<local, special variable>
609 If you localize a special variable, you'll be giving a new value to it,
610 but its magic won't go away. That means that all side-effects related
611 to this magic still work with the localized value.
613 This feature allows code like this to work :
615 # Read the whole contents of FILE in $slurp
616 { local $/ = undef; $slurp = <FILE>; }
618 Note, however, that this restricts localization of some values ; for
619 example, the following statement dies, as of perl 5.9.0, with an error
620 I<Modification of a read-only value attempted>, because the $1 variable is
621 magical and read-only :
625 Similarly, but in a way more difficult to spot, the following snippet will
628 sub f { local $_ = "foo"; print }
630 # now $_ is aliased to $1, thus is magic and readonly
634 See next section for an alternative to this situation.
636 B<WARNING>: Localization of tied arrays and hashes does not currently
638 This will be fixed in a future release of Perl; in the meantime, avoid
639 code that relies on any particular behaviour of localising tied arrays
640 or hashes (localising individual elements is still okay).
641 See L<perl58delta/"Localising Tied Arrays and Hashes Is Broken"> for more
645 =head3 Localization of globs
646 X<local, glob> X<glob>
652 creates a whole new symbol table entry for the glob C<name> in the
653 current package. That means that all variables in its glob slot ($name,
654 @name, %name, &name, and the C<name> filehandle) are dynamically reset.
656 This implies, among other things, that any magic eventually carried by
657 those variables is locally lost. In other words, saying C<local */>
658 will not have any effect on the internal value of the input record
661 Notably, if you want to work with a brand new value of the default scalar
662 $_, and avoid the potential problem listed above about $_ previously
663 carrying a magic value, you should use C<local *_> instead of C<local $_>.
664 As of perl 5.9.1, you can also use the lexical form of C<$_> (declaring it
665 with C<my $_>), which avoids completely this problem.
667 =head3 Localization of elements of composite types
668 X<local, composite type element> X<local, array element> X<local, hash element>
670 It's also worth taking a moment to explain what happens when you
671 C<local>ize a member of a composite type (i.e. an array or hash element).
672 In this case, the element is C<local>ized I<by name>. This means that
673 when the scope of the C<local()> ends, the saved value will be
674 restored to the hash element whose key was named in the C<local()>, or
675 the array element whose index was named in the C<local()>. If that
676 element was deleted while the C<local()> was in effect (e.g. by a
677 C<delete()> from a hash or a C<shift()> of an array), it will spring
678 back into existence, possibly extending an array and filling in the
679 skipped elements with C<undef>. For instance, if you say
681 %hash = ( 'This' => 'is', 'a' => 'test' );
685 local($hash{'a'}) = 'drill';
686 while (my $e = pop(@ary)) {
691 $hash{'only a'} = 'test';
695 print join(' ', map { "$_ $hash{$_}" } sort keys %hash),".\n";
696 print "The array has ",scalar(@ary)," elements: ",
697 join(', ', map { defined $_ ? $_ : 'undef' } @ary),"\n";
704 This is a test only a test.
705 The array has 6 elements: 0, 1, 2, undef, undef, 5
707 The behavior of local() on non-existent members of composite
708 types is subject to change in future.
710 =head2 Lvalue subroutines
711 X<lvalue> X<subroutine, lvalue>
713 B<WARNING>: Lvalue subroutines are still experimental and the
714 implementation may change in future versions of Perl.
716 It is possible to return a modifiable value from a subroutine.
717 To do this, you have to declare the subroutine to return an lvalue.
720 sub canmod : lvalue {
721 # return $val; this doesn't work, don't say "return"
728 canmod() = 5; # assigns to $val
731 The scalar/list context for the subroutine and for the right-hand
732 side of assignment is determined as if the subroutine call is replaced
733 by a scalar. For example, consider:
735 data(2,3) = get_data(3,4);
737 Both subroutines here are called in a scalar context, while in:
739 (data(2,3)) = get_data(3,4);
743 (data(2),data(3)) = get_data(3,4);
745 all the subroutines are called in a list context.
749 =item Lvalue subroutines are EXPERIMENTAL
751 They appear to be convenient, but there are several reasons to be
754 You can't use the return keyword, you must pass out the value before
755 falling out of subroutine scope. (see comment in example above). This
756 is usually not a problem, but it disallows an explicit return out of a
757 deeply nested loop, which is sometimes a nice way out.
759 They violate encapsulation. A normal mutator can check the supplied
760 argument before setting the attribute it is protecting, an lvalue
761 subroutine never gets that chance. Consider;
763 my $some_array_ref = []; # protected by mutators ??
765 sub set_arr { # normal mutator
767 die("expected array, you supplied ", ref $val)
768 unless ref $val eq 'ARRAY';
769 $some_array_ref = $val;
771 sub set_arr_lv : lvalue { # lvalue mutator
775 # set_arr_lv cannot stop this !
776 set_arr_lv() = { a => 1 };
780 =head2 Passing Symbol Table Entries (typeglobs)
783 B<WARNING>: The mechanism described in this section was originally
784 the only way to simulate pass-by-reference in older versions of
785 Perl. While it still works fine in modern versions, the new reference
786 mechanism is generally easier to work with. See below.
788 Sometimes you don't want to pass the value of an array to a subroutine
789 but rather the name of it, so that the subroutine can modify the global
790 copy of it rather than working with a local copy. In perl you can
791 refer to all objects of a particular name by prefixing the name
792 with a star: C<*foo>. This is often known as a "typeglob", because the
793 star on the front can be thought of as a wildcard match for all the
794 funny prefix characters on variables and subroutines and such.
796 When evaluated, the typeglob produces a scalar value that represents
797 all the objects of that name, including any filehandle, format, or
798 subroutine. When assigned to, it causes the name mentioned to refer to
799 whatever C<*> value was assigned to it. Example:
802 local(*someary) = @_;
803 foreach $elem (@someary) {
810 Scalars are already passed by reference, so you can modify
811 scalar arguments without using this mechanism by referring explicitly
812 to C<$_[0]> etc. You can modify all the elements of an array by passing
813 all the elements as scalars, but you have to use the C<*> mechanism (or
814 the equivalent reference mechanism) to C<push>, C<pop>, or change the size of
815 an array. It will certainly be faster to pass the typeglob (or reference).
817 Even if you don't want to modify an array, this mechanism is useful for
818 passing multiple arrays in a single LIST, because normally the LIST
819 mechanism will merge all the array values so that you can't extract out
820 the individual arrays. For more on typeglobs, see
821 L<perldata/"Typeglobs and Filehandles">.
823 =head2 When to Still Use local()
824 X<local> X<variable, local>
826 Despite the existence of C<my>, there are still three places where the
827 C<local> operator still shines. In fact, in these three places, you
828 I<must> use C<local> instead of C<my>.
834 You need to give a global variable a temporary value, especially $_.
836 The global variables, like C<@ARGV> or the punctuation variables, must be
837 C<local>ized with C<local()>. This block reads in F</etc/motd>, and splits
838 it up into chunks separated by lines of equal signs, which are placed
842 local @ARGV = ("/etc/motd");
845 @Fields = split /^\s*=+\s*$/;
848 It particular, it's important to C<local>ize $_ in any routine that assigns
849 to it. Look out for implicit assignments in C<while> conditionals.
853 You need to create a local file or directory handle or a local function.
855 A function that needs a filehandle of its own must use
856 C<local()> on a complete typeglob. This can be used to create new symbol
860 local (*READER, *WRITER); # not my!
861 pipe (READER, WRITER) or die "pipe: $!";
862 return (*READER, *WRITER);
864 ($head, $tail) = ioqueue();
866 See the Symbol module for a way to create anonymous symbol table
869 Because assignment of a reference to a typeglob creates an alias, this
870 can be used to create what is effectively a local function, or at least,
874 local *grow = \&shrink; # only until this block exists
875 grow(); # really calls shrink()
876 move(); # if move() grow()s, it shrink()s too
878 grow(); # get the real grow() again
880 See L<perlref/"Function Templates"> for more about manipulating
881 functions by name in this way.
885 You want to temporarily change just one element of an array or hash.
887 You can C<local>ize just one element of an aggregate. Usually this
891 local $SIG{INT} = 'IGNORE';
892 funct(); # uninterruptible
894 # interruptibility automatically restored here
896 But it also works on lexically declared aggregates. Prior to 5.005,
897 this operation could on occasion misbehave.
901 =head2 Pass by Reference
902 X<pass by reference> X<pass-by-reference> X<reference>
904 If you want to pass more than one array or hash into a function--or
905 return them from it--and have them maintain their integrity, then
906 you're going to have to use an explicit pass-by-reference. Before you
907 do that, you need to understand references as detailed in L<perlref>.
908 This section may not make much sense to you otherwise.
910 Here are a few simple examples. First, let's pass in several arrays
911 to a function and have it C<pop> all of then, returning a new list
912 of all their former last elements:
914 @tailings = popmany ( \@a, \@b, \@c, \@d );
919 foreach $aref ( @_ ) {
920 push @retlist, pop @$aref;
925 Here's how you might write a function that returns a
926 list of keys occurring in all the hashes passed to it:
928 @common = inter( \%foo, \%bar, \%joe );
930 my ($k, $href, %seen); # locals
932 while ( $k = each %$href ) {
936 return grep { $seen{$_} == @_ } keys %seen;
939 So far, we're using just the normal list return mechanism.
940 What happens if you want to pass or return a hash? Well,
941 if you're using only one of them, or you don't mind them
942 concatenating, then the normal calling convention is ok, although
945 Where people get into trouble is here:
947 (@a, @b) = func(@c, @d);
949 (%a, %b) = func(%c, %d);
951 That syntax simply won't work. It sets just C<@a> or C<%a> and
952 clears the C<@b> or C<%b>. Plus the function didn't get passed
953 into two separate arrays or hashes: it got one long list in C<@_>,
956 If you can arrange for everyone to deal with this through references, it's
957 cleaner code, although not so nice to look at. Here's a function that
958 takes two array references as arguments, returning the two array elements
959 in order of how many elements they have in them:
961 ($aref, $bref) = func(\@c, \@d);
962 print "@$aref has more than @$bref\n";
964 my ($cref, $dref) = @_;
965 if (@$cref > @$dref) {
966 return ($cref, $dref);
968 return ($dref, $cref);
972 It turns out that you can actually do this also:
974 (*a, *b) = func(\@c, \@d);
975 print "@a has more than @b\n";
985 Here we're using the typeglobs to do symbol table aliasing. It's
986 a tad subtle, though, and also won't work if you're using C<my>
987 variables, because only globals (even in disguise as C<local>s)
988 are in the symbol table.
990 If you're passing around filehandles, you could usually just use the bare
991 typeglob, like C<*STDOUT>, but typeglobs references work, too.
997 print $fh "her um well a hmmm\n";
1000 $rec = get_rec(\*STDIN);
1003 return scalar <$fh>;
1006 If you're planning on generating new filehandles, you could do this.
1007 Notice to pass back just the bare *FH, not its reference.
1012 return open (FH, $path) ? *FH : undef;
1016 X<prototype> X<subroutine, prototype>
1018 Perl supports a very limited kind of compile-time argument checking
1019 using function prototyping. If you declare
1023 then C<mypush()> takes arguments exactly like C<push()> does. The
1024 function declaration must be visible at compile time. The prototype
1025 affects only interpretation of new-style calls to the function,
1026 where new-style is defined as not using the C<&> character. In
1027 other words, if you call it like a built-in function, then it behaves
1028 like a built-in function. If you call it like an old-fashioned
1029 subroutine, then it behaves like an old-fashioned subroutine. It
1030 naturally falls out from this rule that prototypes have no influence
1031 on subroutine references like C<\&foo> or on indirect subroutine
1032 calls like C<&{$subref}> or C<< $subref->() >>.
1034 Method calls are not influenced by prototypes either, because the
1035 function to be called is indeterminate at compile time, since
1036 the exact code called depends on inheritance.
1038 Because the intent of this feature is primarily to let you define
1039 subroutines that work like built-in functions, here are prototypes
1040 for some other functions that parse almost exactly like the
1041 corresponding built-in.
1043 Declared as Called as
1045 sub mylink ($$) mylink $old, $new
1046 sub myvec ($$$) myvec $var, $offset, 1
1047 sub myindex ($$;$) myindex &getstring, "substr"
1048 sub mysyswrite ($$$;$) mysyswrite $buf, 0, length($buf) - $off, $off
1049 sub myreverse (@) myreverse $a, $b, $c
1050 sub myjoin ($@) myjoin ":", $a, $b, $c
1051 sub mypop (\@) mypop @array
1052 sub mysplice (\@$$@) mysplice @array, @array, 0, @pushme
1053 sub mykeys (\%) mykeys %{$hashref}
1054 sub myopen (*;$) myopen HANDLE, $name
1055 sub mypipe (**) mypipe READHANDLE, WRITEHANDLE
1056 sub mygrep (&@) mygrep { /foo/ } $a, $b, $c
1057 sub myrand (;$) myrand 42
1058 sub mytime () mytime
1060 Any backslashed prototype character represents an actual argument
1061 that absolutely must start with that character. The value passed
1062 as part of C<@_> will be a reference to the actual argument given
1063 in the subroutine call, obtained by applying C<\> to that argument.
1065 You can also backslash several argument types simultaneously by using
1066 the C<\[]> notation:
1068 sub myref (\[$@%&*])
1070 will allow calling myref() as
1078 and the first argument of myref() will be a reference to
1079 a scalar, an array, a hash, a code, or a glob.
1081 Unbackslashed prototype characters have special meanings. Any
1082 unbackslashed C<@> or C<%> eats all remaining arguments, and forces
1083 list context. An argument represented by C<$> forces scalar context. An
1084 C<&> requires an anonymous subroutine, which, if passed as the first
1085 argument, does not require the C<sub> keyword or a subsequent comma.
1087 A C<*> allows the subroutine to accept a bareword, constant, scalar expression,
1088 typeglob, or a reference to a typeglob in that slot. The value will be
1089 available to the subroutine either as a simple scalar, or (in the latter
1090 two cases) as a reference to the typeglob. If you wish to always convert
1091 such arguments to a typeglob reference, use Symbol::qualify_to_ref() as
1094 use Symbol 'qualify_to_ref';
1097 my $fh = qualify_to_ref(shift, caller);
1101 A semicolon (C<;>) separates mandatory arguments from optional arguments.
1102 It is redundant before C<@> or C<%>, which gobble up everything else.
1104 As the last character of a prototype, you can use C<_> in place of C<$>:
1105 if this argument is not provided, C<$_> will be used instead.
1107 Note how the last three examples in the table above are treated
1108 specially by the parser. C<mygrep()> is parsed as a true list
1109 operator, C<myrand()> is parsed as a true unary operator with unary
1110 precedence the same as C<rand()>, and C<mytime()> is truly without
1111 arguments, just like C<time()>. That is, if you say
1115 you'll get C<mytime() + 2>, not C<mytime(2)>, which is how it would be parsed
1116 without a prototype.
1118 The interesting thing about C<&> is that you can generate new syntax with it,
1119 provided it's in the initial position:
1123 my($try,$catch) = @_;
1130 sub catch (&) { $_[0] }
1135 /phooey/ and print "unphooey\n";
1138 That prints C<"unphooey">. (Yes, there are still unresolved
1139 issues having to do with visibility of C<@_>. I'm ignoring that
1140 question for the moment. (But note that if we make C<@_> lexically
1141 scoped, those anonymous subroutines can act like closures... (Gee,
1142 is this sounding a little Lispish? (Never mind.))))
1144 And here's a reimplementation of the Perl C<grep> operator:
1151 push(@result, $_) if &$code;
1156 Some folks would prefer full alphanumeric prototypes. Alphanumerics have
1157 been intentionally left out of prototypes for the express purpose of
1158 someday in the future adding named, formal parameters. The current
1159 mechanism's main goal is to let module writers provide better diagnostics
1160 for module users. Larry feels the notation quite understandable to Perl
1161 programmers, and that it will not intrude greatly upon the meat of the
1162 module, nor make it harder to read. The line noise is visually
1163 encapsulated into a small pill that's easy to swallow.
1165 If you try to use an alphanumeric sequence in a prototype you will
1166 generate an optional warning - "Illegal character in prototype...".
1167 Unfortunately earlier versions of Perl allowed the prototype to be
1168 used as long as its prefix was a valid prototype. The warning may be
1169 upgraded to a fatal error in a future version of Perl once the
1170 majority of offending code is fixed.
1172 It's probably best to prototype new functions, not retrofit prototyping
1173 into older ones. That's because you must be especially careful about
1174 silent impositions of differing list versus scalar contexts. For example,
1175 if you decide that a function should take just one parameter, like this:
1179 print "you gave me $n\n";
1182 and someone has been calling it with an array or expression
1188 Then you've just supplied an automatic C<scalar> in front of their
1189 argument, which can be more than a bit surprising. The old C<@foo>
1190 which used to hold one thing doesn't get passed in. Instead,
1191 C<func()> now gets passed in a C<1>; that is, the number of elements
1192 in C<@foo>. And the C<split> gets called in scalar context so it
1193 starts scribbling on your C<@_> parameter list. Ouch!
1195 This is all very powerful, of course, and should be used only in moderation
1196 to make the world a better place.
1198 =head2 Constant Functions
1201 Functions with a prototype of C<()> are potential candidates for
1202 inlining. If the result after optimization and constant folding
1203 is either a constant or a lexically-scoped scalar which has no other
1204 references, then it will be used in place of function calls made
1205 without C<&>. Calls made using C<&> are never inlined. (See
1206 F<constant.pm> for an easy way to declare most constants.)
1208 The following functions would all be inlined:
1210 sub pi () { 3.14159 } # Not exact, but close.
1211 sub PI () { 4 * atan2 1, 1 } # As good as it gets,
1212 # and it's inlined, too!
1216 sub FLAG_FOO () { 1 << 8 }
1217 sub FLAG_BAR () { 1 << 9 }
1218 sub FLAG_MASK () { FLAG_FOO | FLAG_BAR }
1220 sub OPT_BAZ () { not (0x1B58 & FLAG_MASK) }
1222 sub N () { int(OPT_BAZ) / 3 }
1224 sub FOO_SET () { 1 if FLAG_MASK & FLAG_FOO }
1226 Be aware that these will not be inlined; as they contain inner scopes,
1227 the constant folding doesn't reduce them to a single constant:
1229 sub foo_set () { if (FLAG_MASK & FLAG_FOO) { 1 } }
1240 If you redefine a subroutine that was eligible for inlining, you'll get
1241 a mandatory warning. (You can use this warning to tell whether or not a
1242 particular subroutine is considered constant.) The warning is
1243 considered severe enough not to be optional because previously compiled
1244 invocations of the function will still be using the old value of the
1245 function. If you need to be able to redefine the subroutine, you need to
1246 ensure that it isn't inlined, either by dropping the C<()> prototype
1247 (which changes calling semantics, so beware) or by thwarting the
1248 inlining mechanism in some other way, such as
1250 sub not_inlined () {
1254 =head2 Overriding Built-in Functions
1255 X<built-in> X<override> X<CORE> X<CORE::GLOBAL>
1257 Many built-in functions may be overridden, though this should be tried
1258 only occasionally and for good reason. Typically this might be
1259 done by a package attempting to emulate missing built-in functionality
1260 on a non-Unix system.
1262 Overriding may be done only by importing the name from a module at
1263 compile time--ordinary predeclaration isn't good enough. However, the
1264 C<use subs> pragma lets you, in effect, predeclare subs
1265 via the import syntax, and these names may then override built-in ones:
1267 use subs 'chdir', 'chroot', 'chmod', 'chown';
1271 To unambiguously refer to the built-in form, precede the
1272 built-in name with the special package qualifier C<CORE::>. For example,
1273 saying C<CORE::open()> always refers to the built-in C<open()>, even
1274 if the current package has imported some other subroutine called
1275 C<&open()> from elsewhere. Even though it looks like a regular
1276 function call, it isn't: you can't take a reference to it, such as
1277 the incorrect C<\&CORE::open> might appear to produce.
1279 Library modules should not in general export built-in names like C<open>
1280 or C<chdir> as part of their default C<@EXPORT> list, because these may
1281 sneak into someone else's namespace and change the semantics unexpectedly.
1282 Instead, if the module adds that name to C<@EXPORT_OK>, then it's
1283 possible for a user to import the name explicitly, but not implicitly.
1284 That is, they could say
1288 and it would import the C<open> override. But if they said
1292 they would get the default imports without overrides.
1294 The foregoing mechanism for overriding built-in is restricted, quite
1295 deliberately, to the package that requests the import. There is a second
1296 method that is sometimes applicable when you wish to override a built-in
1297 everywhere, without regard to namespace boundaries. This is achieved by
1298 importing a sub into the special namespace C<CORE::GLOBAL::>. Here is an
1299 example that quite brazenly replaces the C<glob> operator with something
1300 that understands regular expressions.
1305 @EXPORT_OK = 'glob';
1311 my $where = ($sym =~ s/^GLOBAL_// ? 'CORE::GLOBAL' : caller(0));
1312 $pkg->export($where, $sym, @_);
1319 if (opendir D, '.') {
1320 @got = grep /$pat/, readdir D;
1327 And here's how it could be (ab)used:
1329 #use REGlob 'GLOBAL_glob'; # override glob() in ALL namespaces
1331 use REGlob 'glob'; # override glob() in Foo:: only
1332 print for <^[a-z_]+\.pm\$>; # show all pragmatic modules
1334 The initial comment shows a contrived, even dangerous example.
1335 By overriding C<glob> globally, you would be forcing the new (and
1336 subversive) behavior for the C<glob> operator for I<every> namespace,
1337 without the complete cognizance or cooperation of the modules that own
1338 those namespaces. Naturally, this should be done with extreme caution--if
1339 it must be done at all.
1341 The C<REGlob> example above does not implement all the support needed to
1342 cleanly override perl's C<glob> operator. The built-in C<glob> has
1343 different behaviors depending on whether it appears in a scalar or list
1344 context, but our C<REGlob> doesn't. Indeed, many perl built-in have such
1345 context sensitive behaviors, and these must be adequately supported by
1346 a properly written override. For a fully functional example of overriding
1347 C<glob>, study the implementation of C<File::DosGlob> in the standard
1350 When you override a built-in, your replacement should be consistent (if
1351 possible) with the built-in native syntax. You can achieve this by using
1352 a suitable prototype. To get the prototype of an overridable built-in,
1353 use the C<prototype> function with an argument of C<"CORE::builtin_name">
1354 (see L<perlfunc/prototype>).
1356 Note however that some built-ins can't have their syntax expressed by a
1357 prototype (such as C<system> or C<chomp>). If you override them you won't
1358 be able to fully mimic their original syntax.
1360 The built-ins C<do>, C<require> and C<glob> can also be overridden, but due
1361 to special magic, their original syntax is preserved, and you don't have
1362 to define a prototype for their replacements. (You can't override the
1363 C<do BLOCK> syntax, though).
1365 C<require> has special additional dark magic: if you invoke your
1366 C<require> replacement as C<require Foo::Bar>, it will actually receive
1367 the argument C<"Foo/Bar.pm"> in @_. See L<perlfunc/require>.
1369 And, as you'll have noticed from the previous example, if you override
1370 C<glob>, the C<< <*> >> glob operator is overridden as well.
1372 In a similar fashion, overriding the C<readline> function also overrides
1373 the equivalent I/O operator C<< <FILEHANDLE> >>.
1375 Finally, some built-ins (e.g. C<exists> or C<grep>) can't be overridden.
1378 X<autoloading> X<AUTOLOAD>
1380 If you call a subroutine that is undefined, you would ordinarily
1381 get an immediate, fatal error complaining that the subroutine doesn't
1382 exist. (Likewise for subroutines being used as methods, when the
1383 method doesn't exist in any base class of the class's package.)
1384 However, if an C<AUTOLOAD> subroutine is defined in the package or
1385 packages used to locate the original subroutine, then that
1386 C<AUTOLOAD> subroutine is called with the arguments that would have
1387 been passed to the original subroutine. The fully qualified name
1388 of the original subroutine magically appears in the global $AUTOLOAD
1389 variable of the same package as the C<AUTOLOAD> routine. The name
1390 is not passed as an ordinary argument because, er, well, just
1391 because, that's why. (As an exception, a method call to a nonexistent
1392 C<import> or C<unimport> method is just skipped instead.)
1394 Many C<AUTOLOAD> routines load in a definition for the requested
1395 subroutine using eval(), then execute that subroutine using a special
1396 form of goto() that erases the stack frame of the C<AUTOLOAD> routine
1397 without a trace. (See the source to the standard module documented
1398 in L<AutoLoader>, for example.) But an C<AUTOLOAD> routine can
1399 also just emulate the routine and never define it. For example,
1400 let's pretend that a function that wasn't defined should just invoke
1401 C<system> with those arguments. All you'd do is:
1404 my $program = $AUTOLOAD;
1405 $program =~ s/.*:://;
1406 system($program, @_);
1412 In fact, if you predeclare functions you want to call that way, you don't
1413 even need parentheses:
1415 use subs qw(date who ls);
1420 A more complete example of this is the standard Shell module, which
1421 can treat undefined subroutine calls as calls to external programs.
1423 Mechanisms are available to help modules writers split their modules
1424 into autoloadable files. See the standard AutoLoader module
1425 described in L<AutoLoader> and in L<AutoSplit>, the standard
1426 SelfLoader modules in L<SelfLoader>, and the document on adding C
1427 functions to Perl code in L<perlxs>.
1429 =head2 Subroutine Attributes
1430 X<attribute> X<subroutine, attribute> X<attrs>
1432 A subroutine declaration or definition may have a list of attributes
1433 associated with it. If such an attribute list is present, it is
1434 broken up at space or colon boundaries and treated as though a
1435 C<use attributes> had been seen. See L<attributes> for details
1436 about what attributes are currently supported.
1437 Unlike the limitation with the obsolescent C<use attrs>, the
1438 C<sub : ATTRLIST> syntax works to associate the attributes with
1439 a pre-declaration, and not just with a subroutine definition.
1441 The attributes must be valid as simple identifier names (without any
1442 punctuation other than the '_' character). They may have a parameter
1443 list appended, which is only checked for whether its parentheses ('(',')')
1446 Examples of valid syntax (even though the attributes are unknown):
1448 sub fnord (&\%) : switch(10,foo(7,3)) : expensive;
1449 sub plugh () : Ugly('\(") :Bad;
1450 sub xyzzy : _5x5 { ... }
1452 Examples of invalid syntax:
1454 sub fnord : switch(10,foo(); # ()-string not balanced
1455 sub snoid : Ugly('('); # ()-string not balanced
1456 sub xyzzy : 5x5; # "5x5" not a valid identifier
1457 sub plugh : Y2::north; # "Y2::north" not a simple identifier
1458 sub snurt : foo + bar; # "+" not a colon or space
1460 The attribute list is passed as a list of constant strings to the code
1461 which associates them with the subroutine. In particular, the second example
1462 of valid syntax above currently looks like this in terms of how it's
1465 use attributes __PACKAGE__, \&plugh, q[Ugly('\(")], 'Bad';
1467 For further details on attribute lists and their manipulation,
1468 see L<attributes> and L<Attribute::Handlers>.
1472 See L<perlref/"Function Templates"> for more about references and closures.
1473 See L<perlxs> if you'd like to learn about calling C subroutines from Perl.
1474 See L<perlembed> if you'd like to learn about calling Perl subroutines from C.
1475 See L<perlmod> to learn about bundling up your functions in separate files.
1476 See L<perlmodlib> to learn what library modules come standard on your system.
1477 See L<perltoot> to learn how to make object method calls.