7 A Perl script consists of a sequence of declarations and statements.
8 The only things that need to be declared in Perl are report formats
9 and subroutines. See the sections below for more information on those
10 declarations. All uninitialized user-created objects are assumed to
11 start with a C<null> or C<0> value until they are defined by some explicit
12 operation such as assignment. (Though you can get warnings about the
13 use of undefined values if you like.) The sequence of statements is
14 executed just once, unlike in B<sed> and B<awk> scripts, where the
15 sequence of statements is executed for each input line. While this means
16 that you must explicitly loop over the lines of your input file (or
17 files), it also means you have much more control over which files and
18 which lines you look at. (Actually, I'm lying--it is possible to do an
19 implicit loop with either the B<-n> or B<-p> switch. It's just not the
20 mandatory default like it is in B<sed> and B<awk>.)
24 Perl is, for the most part, a free-form language. (The only exception
25 to this is format declarations, for obvious reasons.) Text from a
26 C<"#"> character until the end of the line is a comment, and is
27 ignored. If you attempt to use C</* */> C-style comments, it will be
28 interpreted either as division or pattern matching, depending on the
29 context, and C++ C<//> comments just look like a null regular
30 expression, so don't do that.
32 A declaration can be put anywhere a statement can, but has no effect on
33 the execution of the primary sequence of statements--declarations all
34 take effect at compile time. Typically all the declarations are put at
35 the beginning or the end of the script. However, if you're using
36 lexically-scoped private variables created with C<my()>, you'll have to make sure
37 your format or subroutine definition is within the same block scope
38 as the my if you expect to be able to access those private variables.
40 Declaring a subroutine allows a subroutine name to be used as if it were a
41 list operator from that point forward in the program. You can declare a
42 subroutine without defining it by saying C<sub name>, thus:
45 $me = myname $0 or die "can't get myname";
47 Note that my() functions as a list operator, not as a unary operator; so
48 be careful to use C<or> instead of C<||> in this case. However, if
49 you were to declare the subroutine as C<sub myname ($)>, then
50 C<myname> would function as a unary operator, so either C<or> or
53 Subroutines declarations can also be loaded up with the C<require> statement
54 or both loaded and imported into your namespace with a C<use> statement.
55 See L<perlmod> for details on this.
57 A statement sequence may contain declarations of lexically-scoped
58 variables, but apart from declaring a variable name, the declaration acts
59 like an ordinary statement, and is elaborated within the sequence of
60 statements as if it were an ordinary statement. That means it actually
61 has both compile-time and run-time effects.
63 =head2 Simple statements
65 The only kind of simple statement is an expression evaluated for its
66 side effects. Every simple statement must be terminated with a
67 semicolon, unless it is the final statement in a block, in which case
68 the semicolon is optional. (A semicolon is still encouraged there if the
69 block takes up more than one line, because you may eventually add another line.)
70 Note that there are some operators like C<eval {}> and C<do {}> that look
71 like compound statements, but aren't (they're just TERMs in an expression),
72 and thus need an explicit termination if used as the last item in a statement.
74 Any simple statement may optionally be followed by a I<SINGLE> modifier,
75 just before the terminating semicolon (or block ending). The possible
84 The C<if> and C<unless> modifiers have the expected semantics,
85 presuming you're a speaker of English. The C<foreach> modifier is an
86 iterator: For each value in EXPR, it aliases C<$_> to the value and
87 executes the statement. The C<while> and C<until> modifiers have the
88 usual "C<while> loop" semantics (conditional evaluated first), except
89 when applied to a C<do>-BLOCK (or to the deprecated C<do>-SUBROUTINE
90 statement), in which case the block executes once before the
91 conditional is evaluated. This is so that you can write loops like:
96 } until $line eq ".\n";
98 See L<perlfunc/do>. Note also that the loop control statements described
99 later will I<NOT> work in this construct, because modifiers don't take
100 loop labels. Sorry. You can always put another block inside of it
101 (for C<next>) or around it (for C<last>) to do that sort of thing.
102 For C<next>, just double the braces:
109 For C<last>, you have to be more elaborate:
118 =head2 Compound statements
120 In Perl, a sequence of statements that defines a scope is called a block.
121 Sometimes a block is delimited by the file containing it (in the case
122 of a required file, or the program as a whole), and sometimes a block
123 is delimited by the extent of a string (in the case of an eval).
125 But generally, a block is delimited by curly brackets, also known as braces.
126 We will call this syntactic construct a BLOCK.
128 The following compound statements may be used to control flow:
131 if (EXPR) BLOCK else BLOCK
132 if (EXPR) BLOCK elsif (EXPR) BLOCK ... else BLOCK
133 LABEL while (EXPR) BLOCK
134 LABEL while (EXPR) BLOCK continue BLOCK
135 LABEL for (EXPR; EXPR; EXPR) BLOCK
136 LABEL foreach VAR (LIST) BLOCK
137 LABEL foreach VAR (LIST) BLOCK continue BLOCK
138 LABEL BLOCK continue BLOCK
140 Note that, unlike C and Pascal, these are defined in terms of BLOCKs,
141 not statements. This means that the curly brackets are I<required>--no
142 dangling statements allowed. If you want to write conditionals without
143 curly brackets there are several other ways to do it. The following
144 all do the same thing:
146 if (!open(FOO)) { die "Can't open $FOO: $!"; }
147 die "Can't open $FOO: $!" unless open(FOO);
148 open(FOO) or die "Can't open $FOO: $!"; # FOO or bust!
149 open(FOO) ? 'hi mom' : die "Can't open $FOO: $!";
150 # a bit exotic, that last one
152 The C<if> statement is straightforward. Because BLOCKs are always
153 bounded by curly brackets, there is never any ambiguity about which
154 C<if> an C<else> goes with. If you use C<unless> in place of C<if>,
155 the sense of the test is reversed.
157 The C<while> statement executes the block as long as the expression is
158 true (does not evaluate to the null string (C<"">) or C<0> or C<"0")>. The LABEL is
159 optional, and if present, consists of an identifier followed by a colon.
160 The LABEL identifies the loop for the loop control statements C<next>,
161 C<last>, and C<redo>. If the LABEL is omitted, the loop control statement
162 refers to the innermost enclosing loop. This may include dynamically
163 looking back your call-stack at run time to find the LABEL. Such
164 desperate behavior triggers a warning if you use the B<-w> flag.
166 If there is a C<continue> BLOCK, it is always executed just before the
167 conditional is about to be evaluated again, just like the third part of a
168 C<for> loop in C. Thus it can be used to increment a loop variable, even
169 when the loop has been continued via the C<next> statement (which is
170 similar to the C C<continue> statement).
174 The C<next> command is like the C<continue> statement in C; it starts
175 the next iteration of the loop:
177 LINE: while (<STDIN>) {
178 next LINE if /^#/; # discard comments
182 The C<last> command is like the C<break> statement in C (as used in
183 loops); it immediately exits the loop in question. The
184 C<continue> block, if any, is not executed:
186 LINE: while (<STDIN>) {
187 last LINE if /^$/; # exit when done with header
191 The C<redo> command restarts the loop block without evaluating the
192 conditional again. The C<continue> block, if any, is I<not> executed.
193 This command is normally used by programs that want to lie to themselves
194 about what was just input.
196 For example, when processing a file like F</etc/termcap>.
197 If your input lines might end in backslashes to indicate continuation, you
198 want to skip ahead and get the next record.
209 which is Perl short-hand for the more explicitly written version:
211 LINE: while (defined($line = <ARGV>)) {
213 if ($line =~ s/\\$//) {
215 redo LINE unless eof(); # not eof(ARGV)!
220 Note that if there were a C<continue> block on the above code, it would get
221 executed even on discarded lines. This is often used to reset line counters
222 or C<?pat?> one-time matches.
224 # inspired by :1,$g/fred/s//WILMA/
226 ?(fred)? && s//WILMA $1 WILMA/;
227 ?(barney)? && s//BETTY $1 BETTY/;
228 ?(homer)? && s//MARGE $1 MARGE/;
230 print "$ARGV $.: $_";
231 close ARGV if eof(); # reset $.
232 reset if eof(); # reset ?pat?
235 If the word C<while> is replaced by the word C<until>, the sense of the
236 test is reversed, but the conditional is still tested before the first
239 The loop control statements don't work in an C<if> or C<unless>, since
240 they aren't loops. You can double the braces to make them such, though.
248 The form C<while/if BLOCK BLOCK>, available in Perl 4, is no longer
249 available. Replace any occurrence of C<if BLOCK> by C<if (do BLOCK)>.
253 Perl's C-style C<for> loop works exactly like the corresponding C<while> loop;
254 that means that this:
256 for ($i = 1; $i < 10; $i++) {
269 (There is one minor difference: The first form implies a lexical scope
270 for variables declared with C<my> in the initialization expression.)
272 Besides the normal array index looping, C<for> can lend itself
273 to many other interesting applications. Here's one that avoids the
274 problem you get into if you explicitly test for end-of-file on
275 an interactive file descriptor causing your program to appear to
278 $on_a_tty = -t STDIN && -t STDOUT;
279 sub prompt { print "yes? " if $on_a_tty }
280 for ( prompt(); <STDIN>; prompt() ) {
286 The C<foreach> loop iterates over a normal list value and sets the
287 variable VAR to be each element of the list in turn. If the variable
288 is preceded with the keyword C<my>, then it is lexically scoped, and
289 is therefore visible only within the loop. Otherwise, the variable is
290 implicitly local to the loop and regains its former value upon exiting
291 the loop. If the variable was previously declared with C<my>, it uses
292 that variable instead of the global one, but it's still localized to
295 The C<foreach> keyword is actually a synonym for the C<for> keyword, so
296 you can use C<foreach> for readability or C<for> for brevity. (Or because
297 the Bourne shell is more familiar to you than I<csh>, so writing C<for>
298 comes more naturally.) If VAR is omitted, C<$_> is set to each value.
299 If any element of LIST is an lvalue, you can modify it by modifying VAR
300 inside the loop. That's because the C<foreach> loop index variable is
301 an implicit alias for each item in the list that you're looping over.
303 If any part of LIST is an array, C<foreach> will get very confused if
304 you add or remove elements within the loop body, for example with
305 C<splice>. So don't do that.
307 C<foreach> probably won't do what you expect if VAR is a tied or other
308 special variable. Don't do that either.
312 for (@ary) { s/foo/bar/ }
314 foreach my $elem (@elements) {
318 for $count (10,9,8,7,6,5,4,3,2,1,'BOOM') {
319 print $count, "\n"; sleep(1);
322 for (1..15) { print "Merry Christmas\n"; }
324 foreach $item (split(/:[\\\n:]*/, $ENV{TERMCAP})) {
325 print "Item: $item\n";
328 Here's how a C programmer might code up a particular algorithm in Perl:
330 for (my $i = 0; $i < @ary1; $i++) {
331 for (my $j = 0; $j < @ary2; $j++) {
332 if ($ary1[$i] > $ary2[$j]) {
333 last; # can't go to outer :-(
335 $ary1[$i] += $ary2[$j];
337 # this is where that last takes me
340 Whereas here's how a Perl programmer more comfortable with the idiom might
343 OUTER: foreach my $wid (@ary1) {
344 INNER: foreach my $jet (@ary2) {
345 next OUTER if $wid > $jet;
350 See how much easier this is? It's cleaner, safer, and faster. It's
351 cleaner because it's less noisy. It's safer because if code gets added
352 between the inner and outer loops later on, the new code won't be
353 accidentally executed. The C<next> explicitly iterates the other loop
354 rather than merely terminating the inner one. And it's faster because
355 Perl executes a C<foreach> statement more rapidly than it would the
356 equivalent C<for> loop.
358 =head2 Basic BLOCKs and Switch Statements
360 A BLOCK by itself (labeled or not) is semantically equivalent to a
361 loop that executes once. Thus you can use any of the loop control
362 statements in it to leave or restart the block. (Note that this is
363 I<NOT> true in C<eval{}>, C<sub{}>, or contrary to popular belief
364 C<do{}> blocks, which do I<NOT> count as loops.) The C<continue>
367 The BLOCK construct is particularly nice for doing case
371 if (/^abc/) { $abc = 1; last SWITCH; }
372 if (/^def/) { $def = 1; last SWITCH; }
373 if (/^xyz/) { $xyz = 1; last SWITCH; }
377 There is no official C<switch> statement in Perl, because there are
378 already several ways to write the equivalent. In addition to the
379 above, you could write
382 $abc = 1, last SWITCH if /^abc/;
383 $def = 1, last SWITCH if /^def/;
384 $xyz = 1, last SWITCH if /^xyz/;
388 (That's actually not as strange as it looks once you realize that you can
389 use loop control "operators" within an expression, That's just the normal
395 /^abc/ && do { $abc = 1; last SWITCH; };
396 /^def/ && do { $def = 1; last SWITCH; };
397 /^xyz/ && do { $xyz = 1; last SWITCH; };
401 or formatted so it stands out more as a "proper" C<switch> statement:
424 /^abc/ and $abc = 1, last SWITCH;
425 /^def/ and $def = 1, last SWITCH;
426 /^xyz/ and $xyz = 1, last SWITCH;
441 A common idiom for a C<switch> statement is to use C<foreach>'s aliasing to make
442 a temporary assignment to C<$_> for convenient matching:
444 SWITCH: for ($where) {
445 /In Card Names/ && do { push @flags, '-e'; last; };
446 /Anywhere/ && do { push @flags, '-h'; last; };
447 /In Rulings/ && do { last; };
448 die "unknown value for form variable where: `$where'";
451 Another interesting approach to a switch statement is arrange
452 for a C<do> block to return the proper value:
455 if ($flag & O_RDONLY) { "r" } # XXX: isn't this 0?
456 elsif ($flag & O_WRONLY) { ($flag & O_APPEND) ? "a" : "w" }
457 elsif ($flag & O_RDWR) {
458 if ($flag & O_CREAT) { "w+" }
459 else { ($flag & O_APPEND) ? "a+" : "r+" }
466 ($flags & O_WRONLY) ? "write-only" :
467 ($flags & O_RDWR) ? "read-write" :
471 Or if you are certainly that all the C<&&> clauses are true, you can use
472 something like this, which "switches" on the value of the
473 C<HTTP_USER_AGENT> envariable.
476 # pick out jargon file page based on browser
477 $dir = 'http://www.wins.uva.nl/~mes/jargon';
478 for ($ENV{HTTP_USER_AGENT}) {
479 $page = /Mac/ && 'm/Macintrash.html'
480 || /Win(dows )?NT/ && 'e/evilandrude.html'
481 || /Win|MSIE|WebTV/ && 'm/MicroslothWindows.html'
482 || /Linux/ && 'l/Linux.html'
483 || /HP-UX/ && 'h/HP-SUX.html'
484 || /SunOS/ && 's/ScumOS.html'
485 || 'a/AppendixB.html';
487 print "Location: $dir/$page\015\012\015\012";
489 That kind of switch statement only works when you know the C<&&> clauses
490 will be true. If you don't, the previous C<?:> example should be used.
492 You might also consider writing a hash of subroutine references
493 instead of synthesizing a C<switch> statement.
497 Although not for the faint of heart, Perl does support a C<goto>
498 statement. There are three forms: C<goto>-LABEL, C<goto>-EXPR, and
499 C<goto>-&NAME. A loop's LABEL is not actually a valid target for
500 a C<goto>; it's just the name of the loop.
502 The C<goto>-LABEL form finds the statement labeled with LABEL and resumes
503 execution there. It may not be used to go into any construct that
504 requires initialization, such as a subroutine or a C<foreach> loop. It
505 also can't be used to go into a construct that is optimized away. It
506 can be used to go almost anywhere else within the dynamic scope,
507 including out of subroutines, but it's usually better to use some other
508 construct such as C<last> or C<die>. The author of Perl has never felt the
509 need to use this form of C<goto> (in Perl, that is--C is another matter).
511 The C<goto>-EXPR form expects a label name, whose scope will be resolved
512 dynamically. This allows for computed C<goto>s per FORTRAN, but isn't
513 necessarily recommended if you're optimizing for maintainability:
515 goto ("FOO", "BAR", "GLARCH")[$i];
517 The C<goto>-&NAME form is highly magical, and substitutes a call to the
518 named subroutine for the currently running subroutine. This is used by
519 C<AUTOLOAD()> subroutines that wish to load another subroutine and then
520 pretend that the other subroutine had been called in the first place
521 (except that any modifications to C<@_> in the current subroutine are
522 propagated to the other subroutine.) After the C<goto>, not even C<caller()>
523 will be able to tell that this routine was called first.
525 In almost all cases like this, it's usually a far, far better idea to use the
526 structured control flow mechanisms of C<next>, C<last>, or C<redo> instead of
527 resorting to a C<goto>. For certain applications, the catch and throw pair of
528 C<eval{}> and die() for exception processing can also be a prudent approach.
530 =head2 PODs: Embedded Documentation
532 Perl has a mechanism for intermixing documentation with source code.
533 While it's expecting the beginning of a new statement, if the compiler
534 encounters a line that begins with an equal sign and a word, like this
536 =head1 Here There Be Pods!
538 Then that text and all remaining text up through and including a line
539 beginning with C<=cut> will be ignored. The format of the intervening
540 text is described in L<perlpod>.
542 This allows you to intermix your source code
543 and your documentation text freely, as in
547 The snazzle() function will behave in the most spectacular
548 form that you can possibly imagine, not even excepting
549 cybernetic pyrotechnics.
551 =cut back to the compiler, nuff of this pod stuff!
558 Note that pod translators should look at only paragraphs beginning
559 with a pod directive (it makes parsing easier), whereas the compiler
560 actually knows to look for pod escapes even in the middle of a
561 paragraph. This means that the following secret stuff will be
562 ignored by both the compiler and the translators.
566 warn "Neither POD nor CODE!?"
570 You probably shouldn't rely upon the C<warn()> being podded out forever.
571 Not all pod translators are well-behaved in this regard, and perhaps
572 the compiler will become pickier.
574 One may also use pod directives to quickly comment out a section
577 =head2 Plain Old Comments (Not!)
579 Much like the C preprocessor, Perl can process line directives. Using
580 this, one can control Perl's idea of filenames and line numbers in
581 error or warning messages (especially for strings that are processed
582 with C<eval()>). The syntax for this mechanism is the same as for most
583 C preprocessors: it matches the regular expression
584 C</^#\s*line\s+(\d+)\s*(?:\s"([^"]*)")?/> with C<$1> being the line
585 number for the next line, and C<$2> being the optional filename
586 (specified within quotes).
588 Here are some examples that you should be able to type into your command
593 # the `#' on the previous line must be the first char on line
596 foo at bzzzt line 201.
600 eval qq[\n#line 2001 ""\ndie 'foo']; print $@;
605 eval qq[\n#line 200 "foo bar"\ndie 'foo']; print $@;
607 foo at foo bar line 200.
611 eval "\n#line " . __LINE__ . ' "' . __FILE__ ."\"\ndie 'foo'";
614 foo at goop line 345.