This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Remove bad advice from perllocale.pod
[perl5.git] / pod / perlsyn.pod
CommitLineData
a0d0e21e
LW
1=head1 NAME
2
3perlsyn - Perl syntax
4
5=head1 DESCRIPTION
6
7A Perl script consists of a sequence of declarations and statements.
8The only things that need to be declared in Perl are report formats
9and subroutines. See the sections below for more information on those
10declarations. All uninitialized user-created objects are assumed to
11start with a null or 0 value until they are defined by some explicit
12operation such as assignment. (Though you can get warnings about the
13use of undefined values if you like.) The sequence of statements is
14executed just once, unlike in B<sed> and B<awk> scripts, where the
15sequence of statements is executed for each input line. While this means
16that you must explicitly loop over the lines of your input file (or
17files), it also means you have much more control over which files and
18which lines you look at. (Actually, I'm lying--it is possible to do an
19implicit loop with either the B<-n> or B<-p> switch. It's just not the
20mandatory default like it is in B<sed> and B<awk>.)
21
4633a7c4
LW
22=head2 Declarations
23
a0d0e21e
LW
24Perl is, for the most part, a free-form language. (The only
25exception to this is format declarations, for obvious reasons.) Comments
26are indicated by the "#" character, and extend to the end of the line. If
27you attempt to use C</* */> C-style comments, it will be interpreted
28either as division or pattern matching, depending on the context, and C++
4633a7c4 29C<//> comments just look like a null regular expression, so don't do
a0d0e21e
LW
30that.
31
32A declaration can be put anywhere a statement can, but has no effect on
33the execution of the primary sequence of statements--declarations all
34take effect at compile time. Typically all the declarations are put at
4633a7c4
LW
35the beginning or the end of the script. However, if you're using
36lexically-scoped private variables created with my(), you'll have to make sure
37your format or subroutine definition is within the same block scope
5f05dabc 38as the my if you expect to be able to access those private variables.
a0d0e21e 39
4633a7c4
LW
40Declaring a subroutine allows a subroutine name to be used as if it were a
41list operator from that point forward in the program. You can declare a
c07a80fd 42subroutine (prototyped to take one scalar parameter) without defining it by saying just:
a0d0e21e 43
c07a80fd 44 sub myname ($);
a0d0e21e
LW
45 $me = myname $0 or die "can't get myname";
46
4633a7c4 47Note that it functions as a list operator though, not as a unary
a0d0e21e
LW
48operator, so be careful to use C<or> instead of C<||> there.
49
4633a7c4
LW
50Subroutines declarations can also be loaded up with the C<require> statement
51or both loaded and imported into your namespace with a C<use> statement.
52See L<perlmod> for details on this.
a0d0e21e 53
4633a7c4
LW
54A statement sequence may contain declarations of lexically-scoped
55variables, but apart from declaring a variable name, the declaration acts
56like an ordinary statement, and is elaborated within the sequence of
57statements as if it were an ordinary statement. That means it actually
58has both compile-time and run-time effects.
a0d0e21e
LW
59
60=head2 Simple statements
61
62The only kind of simple statement is an expression evaluated for its
63side effects. Every simple statement must be terminated with a
64semicolon, unless it is the final statement in a block, in which case
65the semicolon is optional. (A semicolon is still encouraged there if the
5f05dabc 66block takes up more than one line, because you may eventually add another line.)
a0d0e21e
LW
67Note that there are some operators like C<eval {}> and C<do {}> that look
68like compound statements, but aren't (they're just TERMs in an expression),
4633a7c4 69and thus need an explicit termination if used as the last item in a statement.
a0d0e21e
LW
70
71Any simple statement may optionally be followed by a I<SINGLE> modifier,
72just before the terminating semicolon (or block ending). The possible
73modifiers are:
74
75 if EXPR
76 unless EXPR
77 while EXPR
78 until EXPR
79
80The C<if> and C<unless> modifiers have the expected semantics,
81presuming you're a speaker of English. The C<while> and C<until>
82modifiers also have the usual "while loop" semantics (conditional
83evaluated first), except when applied to a do-BLOCK (or to the
84now-deprecated do-SUBROUTINE statement), in which case the block
85executes once before the conditional is evaluated. This is so that you
86can write loops like:
87
88 do {
4633a7c4 89 $line = <STDIN>;
a0d0e21e 90 ...
4633a7c4 91 } until $line eq ".\n";
a0d0e21e
LW
92
93See L<perlfunc/do>. Note also that the loop control
5f05dabc 94statements described later will I<NOT> work in this construct, because
a0d0e21e 95modifiers don't take loop labels. Sorry. You can always wrap
4633a7c4 96another block around it to do that sort of thing.
a0d0e21e
LW
97
98=head2 Compound statements
99
100In Perl, a sequence of statements that defines a scope is called a block.
101Sometimes a block is delimited by the file containing it (in the case
102of a required file, or the program as a whole), and sometimes a block
103is delimited by the extent of a string (in the case of an eval).
104
105But generally, a block is delimited by curly brackets, also known as braces.
106We will call this syntactic construct a BLOCK.
107
108The following compound statements may be used to control flow:
109
110 if (EXPR) BLOCK
111 if (EXPR) BLOCK else BLOCK
112 if (EXPR) BLOCK elsif (EXPR) BLOCK ... else BLOCK
113 LABEL while (EXPR) BLOCK
114 LABEL while (EXPR) BLOCK continue BLOCK
115 LABEL for (EXPR; EXPR; EXPR) BLOCK
748a9306 116 LABEL foreach VAR (LIST) BLOCK
a0d0e21e
LW
117 LABEL BLOCK continue BLOCK
118
119Note that, unlike C and Pascal, these are defined in terms of BLOCKs,
120not statements. This means that the curly brackets are I<required>--no
121dangling statements allowed. If you want to write conditionals without
122curly brackets there are several other ways to do it. The following
123all do the same thing:
124
125 if (!open(FOO)) { die "Can't open $FOO: $!"; }
126 die "Can't open $FOO: $!" unless open(FOO);
127 open(FOO) or die "Can't open $FOO: $!"; # FOO or bust!
128 open(FOO) ? 'hi mom' : die "Can't open $FOO: $!";
129 # a bit exotic, that last one
130
5f05dabc 131The C<if> statement is straightforward. Because BLOCKs are always
a0d0e21e
LW
132bounded by curly brackets, there is never any ambiguity about which
133C<if> an C<else> goes with. If you use C<unless> in place of C<if>,
134the sense of the test is reversed.
135
136The C<while> statement executes the block as long as the expression is
137true (does not evaluate to the null string or 0 or "0"). The LABEL is
4633a7c4
LW
138optional, and if present, consists of an identifier followed by a colon.
139The LABEL identifies the loop for the loop control statements C<next>,
140C<last>, and C<redo>. If the LABEL is omitted, the loop control statement
141refers to the innermost enclosing loop. This may include dynamically
142looking back your call-stack at run time to find the LABEL. Such
143desperate behavior triggers a warning if you use the B<-w> flag.
144
145If there is a C<continue> BLOCK, it is always executed just before the
146conditional is about to be evaluated again, just like the third part of a
147C<for> loop in C. Thus it can be used to increment a loop variable, even
148when the loop has been continued via the C<next> statement (which is
149similar to the C C<continue> statement).
150
151=head2 Loop Control
152
153The C<next> command is like the C<continue> statement in C; it starts
154the next iteration of the loop:
155
156 LINE: while (<STDIN>) {
157 next LINE if /^#/; # discard comments
158 ...
159 }
160
161The C<last> command is like the C<break> statement in C (as used in
162loops); it immediately exits the loop in question. The
163C<continue> block, if any, is not executed:
164
165 LINE: while (<STDIN>) {
166 last LINE if /^$/; # exit when done with header
167 ...
168 }
169
170The C<redo> command restarts the loop block without evaluating the
171conditional again. The C<continue> block, if any, is I<not> executed.
172This command is normally used by programs that want to lie to themselves
173about what was just input.
174
175For example, when processing a file like F</etc/termcap>.
176If your input lines might end in backslashes to indicate continuation, you
177want to skip ahead and get the next record.
178
179 while (<>) {
180 chomp;
181 if (s/\\$//) {
182 $_ .= <>;
183 redo unless eof();
184 }
185 # now process $_
186 }
187
188which is Perl short-hand for the more explicitly written version:
189
190 LINE: while ($line = <ARGV>) {
191 chomp($line);
192 if ($line =~ s/\\$//) {
193 $line .= <ARGV>;
194 redo LINE unless eof(); # not eof(ARGV)!
195 }
196 # now process $line
197 }
198
184e9718 199Or here's a simpleminded Pascal comment stripper (warning: assumes no { or } in strings).
4633a7c4
LW
200
201 LINE: while (<STDIN>) {
202 while (s|({.*}.*){.*}|$1 |) {}
203 s|{.*}| |;
204 if (s|{.*| |) {
205 $front = $_;
206 while (<STDIN>) {
207 if (/}/) { # end of comment?
208 s|^|$front{|;
209 redo LINE;
210 }
211 }
212 }
213 print;
214 }
215
216Note that if there were a C<continue> block on the above code, it would get
217executed even on discarded lines.
a0d0e21e
LW
218
219If the word C<while> is replaced by the word C<until>, the sense of the
220test is reversed, but the conditional is still tested before the first
221iteration.
222
5b23ba8b
G
223The form C<while/if BLOCK BLOCK>, available in Perl 4, is no longer
224available. Replace any occurrence of C<if BLOCK> by C<if (do BLOCK)>.
4633a7c4 225
cb1a09d0 226=head2 For Loops
a0d0e21e 227
cb1a09d0
AD
228Perl's C-style C<for> loop works exactly like the corresponding C<while> loop;
229that means that this:
a0d0e21e
LW
230
231 for ($i = 1; $i < 10; $i++) {
232 ...
233 }
234
cb1a09d0 235is the same as this:
a0d0e21e
LW
236
237 $i = 1;
238 while ($i < 10) {
239 ...
240 } continue {
241 $i++;
242 }
243
55497cff
PP
244(There is one minor difference: The first form implies a lexical scope
245for variables declared with C<my> in the initialization expression.)
246
cb1a09d0
AD
247Besides the normal array index looping, C<for> can lend itself
248to many other interesting applications. Here's one that avoids the
249problem you get into if you explicitly test for end-of-file on
250an interactive file descriptor causing your program to appear to
251hang.
252
253 $on_a_tty = -t STDIN && -t STDOUT;
254 sub prompt { print "yes? " if $on_a_tty }
255 for ( prompt(); <STDIN>; prompt() ) {
256 # do something
257 }
258
259=head2 Foreach Loops
260
4633a7c4 261The C<foreach> loop iterates over a normal list value and sets the
55497cff
PP
262variable VAR to be each element of the list in turn. If the variable
263is preceded with the keyword C<my>, then it is lexically scoped, and
264is therefore visible only within the loop. Otherwise, the variable is
265implicitly local to the loop and regains its former value upon exiting
266the loop. If the variable was previously declared with C<my>, it uses
267that variable instead of the global one, but it's still localized to
268the loop. (Note that a lexically scoped variable can cause problems
269with you have subroutine or format declarations.)
4633a7c4
LW
270
271The C<foreach> keyword is actually a synonym for the C<for> keyword, so
272you can use C<foreach> for readability or C<for> for brevity. If VAR is
273omitted, $_ is set to each value. If LIST is an actual array (as opposed
274to an expression returning a list value), you can modify each element of
275the array by modifying VAR inside the loop. That's because the C<foreach>
276loop index variable is an implicit alias for each item in the list that
277you're looping over.
278
748a9306 279Examples:
a0d0e21e 280
4633a7c4 281 for (@ary) { s/foo/bar/ }
a0d0e21e 282
55497cff 283 foreach my $elem (@elements) {
a0d0e21e
LW
284 $elem *= 2;
285 }
286
4633a7c4
LW
287 for $count (10,9,8,7,6,5,4,3,2,1,'BOOM') {
288 print $count, "\n"; sleep(1);
a0d0e21e
LW
289 }
290
291 for (1..15) { print "Merry Christmas\n"; }
292
4633a7c4 293 foreach $item (split(/:[\\\n:]*/, $ENV{TERMCAP})) {
a0d0e21e
LW
294 print "Item: $item\n";
295 }
296
4633a7c4
LW
297Here's how a C programmer might code up a particular algorithm in Perl:
298
55497cff
PP
299 for (my $i = 0; $i < @ary1; $i++) {
300 for (my $j = 0; $j < @ary2; $j++) {
4633a7c4
LW
301 if ($ary1[$i] > $ary2[$j]) {
302 last; # can't go to outer :-(
303 }
304 $ary1[$i] += $ary2[$j];
305 }
cb1a09d0 306 # this is where that last takes me
4633a7c4
LW
307 }
308
184e9718 309Whereas here's how a Perl programmer more comfortable with the idiom might
cb1a09d0 310do it:
4633a7c4 311
55497cff
PP
312 OUTER: foreach my $wid (@ary1) {
313 INNER: foreach my $jet (@ary2) {
cb1a09d0
AD
314 next OUTER if $wid > $jet;
315 $wid += $jet;
4633a7c4
LW
316 }
317 }
318
cb1a09d0
AD
319See how much easier this is? It's cleaner, safer, and faster. It's
320cleaner because it's less noisy. It's safer because if code gets added
c07a80fd 321between the inner and outer loops later on, the new code won't be
5f05dabc 322accidentally executed. The C<next> explicitly iterates the other loop
c07a80fd
PP
323rather than merely terminating the inner one. And it's faster because
324Perl executes a C<foreach> statement more rapidly than it would the
325equivalent C<for> loop.
4633a7c4
LW
326
327=head2 Basic BLOCKs and Switch Statements
328
55497cff
PP
329A BLOCK by itself (labeled or not) is semantically equivalent to a
330loop that executes once. Thus you can use any of the loop control
331statements in it to leave or restart the block. (Note that this is
332I<NOT> true in C<eval{}>, C<sub{}>, or contrary to popular belief
333C<do{}> blocks, which do I<NOT> count as loops.) The C<continue>
334block is optional.
4633a7c4
LW
335
336The BLOCK construct is particularly nice for doing case
a0d0e21e
LW
337structures.
338
339 SWITCH: {
340 if (/^abc/) { $abc = 1; last SWITCH; }
341 if (/^def/) { $def = 1; last SWITCH; }
342 if (/^xyz/) { $xyz = 1; last SWITCH; }
343 $nothing = 1;
344 }
345
346There is no official switch statement in Perl, because there are
347already several ways to write the equivalent. In addition to the
348above, you could write
349
350 SWITCH: {
351 $abc = 1, last SWITCH if /^abc/;
352 $def = 1, last SWITCH if /^def/;
353 $xyz = 1, last SWITCH if /^xyz/;
354 $nothing = 1;
355 }
356
cb1a09d0 357(That's actually not as strange as it looks once you realize that you can
a0d0e21e
LW
358use loop control "operators" within an expression, That's just the normal
359C comma operator.)
360
361or
362
363 SWITCH: {
364 /^abc/ && do { $abc = 1; last SWITCH; };
365 /^def/ && do { $def = 1; last SWITCH; };
366 /^xyz/ && do { $xyz = 1; last SWITCH; };
367 $nothing = 1;
368 }
369
370or formatted so it stands out more as a "proper" switch statement:
371
372 SWITCH: {
373 /^abc/ && do {
374 $abc = 1;
375 last SWITCH;
376 };
377
378 /^def/ && do {
379 $def = 1;
380 last SWITCH;
381 };
382
383 /^xyz/ && do {
384 $xyz = 1;
385 last SWITCH;
386 };
387 $nothing = 1;
388 }
389
390or
391
392 SWITCH: {
393 /^abc/ and $abc = 1, last SWITCH;
394 /^def/ and $def = 1, last SWITCH;
395 /^xyz/ and $xyz = 1, last SWITCH;
396 $nothing = 1;
397 }
398
399or even, horrors,
400
401 if (/^abc/)
402 { $abc = 1 }
403 elsif (/^def/)
404 { $def = 1 }
405 elsif (/^xyz/)
406 { $xyz = 1 }
407 else
408 { $nothing = 1 }
409
4633a7c4
LW
410
411A common idiom for a switch statement is to use C<foreach>'s aliasing to make
412a temporary assignment to $_ for convenient matching:
413
414 SWITCH: for ($where) {
415 /In Card Names/ && do { push @flags, '-e'; last; };
416 /Anywhere/ && do { push @flags, '-h'; last; };
417 /In Rulings/ && do { last; };
418 die "unknown value for form variable where: `$where'";
419 }
420
cb1a09d0
AD
421Another interesting approach to a switch statement is arrange
422for a C<do> block to return the proper value:
423
424 $amode = do {
425 if ($flag & O_RDONLY) { "r" }
c07a80fd 426 elsif ($flag & O_WRONLY) { ($flag & O_APPEND) ? "a" : "w" }
cb1a09d0
AD
427 elsif ($flag & O_RDWR) {
428 if ($flag & O_CREAT) { "w+" }
c07a80fd 429 else { ($flag & O_APPEND) ? "a+" : "r+" }
cb1a09d0
AD
430 }
431 };
432
4633a7c4
LW
433=head2 Goto
434
435Although not for the faint of heart, Perl does support a C<goto> statement.
436A loop's LABEL is not actually a valid target for a C<goto>;
437it's just the name of the loop. There are three forms: goto-LABEL,
438goto-EXPR, and goto-&NAME.
439
440The goto-LABEL form finds the statement labeled with LABEL and resumes
441execution there. It may not be used to go into any construct that
442requires initialization, such as a subroutine or a foreach loop. It
443also can't be used to go into a construct that is optimized away. It
444can be used to go almost anywhere else within the dynamic scope,
445including out of subroutines, but it's usually better to use some other
446construct such as last or die. The author of Perl has never felt the
447need to use this form of goto (in Perl, that is--C is another matter).
448
449The goto-EXPR form expects a label name, whose scope will be resolved
450dynamically. This allows for computed gotos per FORTRAN, but isn't
451necessarily recommended if you're optimizing for maintainability:
452
453 goto ("FOO", "BAR", "GLARCH")[$i];
454
455The goto-&NAME form is highly magical, and substitutes a call to the
456named subroutine for the currently running subroutine. This is used by
457AUTOLOAD() subroutines that wish to load another subroutine and then
458pretend that the other subroutine had been called in the first place
459(except that any modifications to @_ in the current subroutine are
460propagated to the other subroutine.) After the C<goto>, not even caller()
461will be able to tell that this routine was called first.
462
c07a80fd
PP
463In almost all cases like this, it's usually a far, far better idea to use the
464structured control flow mechanisms of C<next>, C<last>, or C<redo> instead of
4633a7c4
LW
465resorting to a C<goto>. For certain applications, the catch and throw pair of
466C<eval{}> and die() for exception processing can also be a prudent approach.
cb1a09d0
AD
467
468=head2 PODs: Embedded Documentation
469
470Perl has a mechanism for intermixing documentation with source code.
c07a80fd 471While it's expecting the beginning of a new statement, if the compiler
cb1a09d0
AD
472encounters a line that begins with an equal sign and a word, like this
473
474 =head1 Here There Be Pods!
475
476Then that text and all remaining text up through and including a line
477beginning with C<=cut> will be ignored. The format of the intervening
478text is described in L<perlpod>.
479
480This allows you to intermix your source code
481and your documentation text freely, as in
482
483 =item snazzle($)
484
485 The snazzle() function will behave in the most spectacular
486 form that you can possibly imagine, not even excepting
487 cybernetic pyrotechnics.
488
489 =cut back to the compiler, nuff of this pod stuff!
490
491 sub snazzle($) {
492 my $thingie = shift;
493 .........
494 }
495
5f05dabc 496Note that pod translators should look at only paragraphs beginning
184e9718 497with a pod directive (it makes parsing easier), whereas the compiler
cb1a09d0
AD
498actually knows to look for pod escapes even in the middle of a
499paragraph. This means that the following secret stuff will be
500ignored by both the compiler and the translators.
501
502 $a=3;
503 =secret stuff
504 warn "Neither POD nor CODE!?"
505 =cut back
506 print "got $a\n";
507
508You probably shouldn't rely upon the warn() being podded out forever.
509Not all pod translators are well-behaved in this regard, and perhaps
510the compiler will become pickier.