7 A Perl script consists of a sequence of declarations and statements.
8 The sequence of statements is executed just once, unlike in B<sed>
9 and B<awk> scripts, where the sequence of statements is executed
10 for each input line. While this means that you must explicitly
11 loop over the lines of your input file (or files), it also means
12 you have much more control over which files and which lines you look at.
13 (Actually, I'm lying--it is possible to do an implicit loop with
14 either the B<-n> or B<-p> switch. It's just not the mandatory
15 default like it is in B<sed> and B<awk>.)
17 Perl is, for the most part, a free-form language. (The only exception
18 to this is format declarations, for obvious reasons.) Text from a
19 C<"#"> character until the end of the line is a comment, and is
20 ignored. If you attempt to use C</* */> C-style comments, it will be
21 interpreted either as division or pattern matching, depending on the
22 context, and C++ C<//> comments just look like a null regular
23 expression, so don't do that.
27 The only things you need to declare in Perl are report formats
28 and subroutines--and even undefined subroutines can be handled
29 through AUTOLOAD. A variable holds the undefined value (C<undef>)
30 until it has been assigned a defined value, which is anything
31 other than C<undef>. When used as a number, C<undef> is treated
32 as C<0>; when used as a string, it is treated the empty string,
33 C<"">; and when used as a reference that isn't being assigned
34 to, it is treated as an error. If you enable warnings, you'll
35 be notified of an uninitialized value whenever you treat C<undef>
36 as a string or a number. Well, usually. Boolean ("don't-care")
37 contexts and operators such as C<++>, C<-->, C<+=>, C<-=>, and
38 C<.=> are always exempt from such warnings.
40 A declaration can be put anywhere a statement can, but has no effect on
41 the execution of the primary sequence of statements--declarations all
42 take effect at compile time. Typically all the declarations are put at
43 the beginning or the end of the script. However, if you're using
44 lexically-scoped private variables created with C<my()>, you'll
46 your format or subroutine definition is within the same block scope
47 as the my if you expect to be able to access those private variables.
49 Declaring a subroutine allows a subroutine name to be used as if it were a
50 list operator from that point forward in the program. You can declare a
51 subroutine without defining it by saying C<sub name>, thus:
54 $me = myname $0 or die "can't get myname";
56 Note that myname() functions as a list operator, not as a unary operator;
57 so be careful to use C<or> instead of C<||> in this case. However, if
58 you were to declare the subroutine as C<sub myname ($)>, then
59 C<myname> would function as a unary operator, so either C<or> or
62 Subroutines declarations can also be loaded up with the C<require> statement
63 or both loaded and imported into your namespace with a C<use> statement.
64 See L<perlmod> for details on this.
66 A statement sequence may contain declarations of lexically-scoped
67 variables, but apart from declaring a variable name, the declaration acts
68 like an ordinary statement, and is elaborated within the sequence of
69 statements as if it were an ordinary statement. That means it actually
70 has both compile-time and run-time effects.
72 =head2 Simple statements
74 The only kind of simple statement is an expression evaluated for its
75 side effects. Every simple statement must be terminated with a
76 semicolon, unless it is the final statement in a block, in which case
77 the semicolon is optional. (A semicolon is still encouraged there if the
78 block takes up more than one line, because you may eventually add another line.)
79 Note that there are some operators like C<eval {}> and C<do {}> that look
80 like compound statements, but aren't (they're just TERMs in an expression),
81 and thus need an explicit termination if used as the last item in a statement.
83 Any simple statement may optionally be followed by a I<SINGLE> modifier,
84 just before the terminating semicolon (or block ending). The possible
93 The C<if> and C<unless> modifiers have the expected semantics,
94 presuming you're a speaker of English. The C<foreach> modifier is an
95 iterator: For each value in EXPR, it aliases C<$_> to the value and
96 executes the statement. The C<while> and C<until> modifiers have the
97 usual "C<while> loop" semantics (conditional evaluated first), except
98 when applied to a C<do>-BLOCK (or to the deprecated C<do>-SUBROUTINE
99 statement), in which case the block executes once before the
100 conditional is evaluated. This is so that you can write loops like:
105 } until $line eq ".\n";
107 See L<perlfunc/do>. Note also that the loop control statements described
108 later will I<NOT> work in this construct, because modifiers don't take
109 loop labels. Sorry. You can always put another block inside of it
110 (for C<next>) or around it (for C<last>) to do that sort of thing.
111 For C<next>, just double the braces:
118 For C<last>, you have to be more elaborate:
127 =head2 Compound statements
129 In Perl, a sequence of statements that defines a scope is called a block.
130 Sometimes a block is delimited by the file containing it (in the case
131 of a required file, or the program as a whole), and sometimes a block
132 is delimited by the extent of a string (in the case of an eval).
134 But generally, a block is delimited by curly brackets, also known as braces.
135 We will call this syntactic construct a BLOCK.
137 The following compound statements may be used to control flow:
140 if (EXPR) BLOCK else BLOCK
141 if (EXPR) BLOCK elsif (EXPR) BLOCK ... else BLOCK
142 LABEL while (EXPR) BLOCK
143 LABEL while (EXPR) BLOCK continue BLOCK
144 LABEL for (EXPR; EXPR; EXPR) BLOCK
145 LABEL foreach VAR (LIST) BLOCK
146 LABEL foreach VAR (LIST) BLOCK continue BLOCK
147 LABEL BLOCK continue BLOCK
149 Note that, unlike C and Pascal, these are defined in terms of BLOCKs,
150 not statements. This means that the curly brackets are I<required>--no
151 dangling statements allowed. If you want to write conditionals without
152 curly brackets there are several other ways to do it. The following
153 all do the same thing:
155 if (!open(FOO)) { die "Can't open $FOO: $!"; }
156 die "Can't open $FOO: $!" unless open(FOO);
157 open(FOO) or die "Can't open $FOO: $!"; # FOO or bust!
158 open(FOO) ? 'hi mom' : die "Can't open $FOO: $!";
159 # a bit exotic, that last one
161 The C<if> statement is straightforward. Because BLOCKs are always
162 bounded by curly brackets, there is never any ambiguity about which
163 C<if> an C<else> goes with. If you use C<unless> in place of C<if>,
164 the sense of the test is reversed.
166 The C<while> statement executes the block as long as the expression is
167 true (does not evaluate to the null string C<""> or C<0> or C<"0">).
168 The LABEL is optional, and if present, consists of an identifier followed
169 by a colon. The LABEL identifies the loop for the loop control
170 statements C<next>, C<last>, and C<redo>.
171 If the LABEL is omitted, the loop control statement
172 refers to the innermost enclosing loop. This may include dynamically
173 looking back your call-stack at run time to find the LABEL. Such
174 desperate behavior triggers a warning if you use the C<use warnings>
175 pragma or the B<-w> flag.
176 Unlike a C<foreach> statement, a C<while> statement never implicitly
177 localises any variables.
179 If there is a C<continue> BLOCK, it is always executed just before the
180 conditional is about to be evaluated again, just like the third part of a
181 C<for> loop in C. Thus it can be used to increment a loop variable, even
182 when the loop has been continued via the C<next> statement (which is
183 similar to the C C<continue> statement).
187 The C<next> command is like the C<continue> statement in C; it starts
188 the next iteration of the loop:
190 LINE: while (<STDIN>) {
191 next LINE if /^#/; # discard comments
195 The C<last> command is like the C<break> statement in C (as used in
196 loops); it immediately exits the loop in question. The
197 C<continue> block, if any, is not executed:
199 LINE: while (<STDIN>) {
200 last LINE if /^$/; # exit when done with header
204 The C<redo> command restarts the loop block without evaluating the
205 conditional again. The C<continue> block, if any, is I<not> executed.
206 This command is normally used by programs that want to lie to themselves
207 about what was just input.
209 For example, when processing a file like F</etc/termcap>.
210 If your input lines might end in backslashes to indicate continuation, you
211 want to skip ahead and get the next record.
222 which is Perl short-hand for the more explicitly written version:
224 LINE: while (defined($line = <ARGV>)) {
226 if ($line =~ s/\\$//) {
228 redo LINE unless eof(); # not eof(ARGV)!
233 Note that if there were a C<continue> block on the above code, it would get
234 executed even on discarded lines. This is often used to reset line counters
235 or C<?pat?> one-time matches.
237 # inspired by :1,$g/fred/s//WILMA/
239 ?(fred)? && s//WILMA $1 WILMA/;
240 ?(barney)? && s//BETTY $1 BETTY/;
241 ?(homer)? && s//MARGE $1 MARGE/;
243 print "$ARGV $.: $_";
244 close ARGV if eof(); # reset $.
245 reset if eof(); # reset ?pat?
248 If the word C<while> is replaced by the word C<until>, the sense of the
249 test is reversed, but the conditional is still tested before the first
252 The loop control statements don't work in an C<if> or C<unless>, since
253 they aren't loops. You can double the braces to make them such, though.
261 The form C<while/if BLOCK BLOCK>, available in Perl 4, is no longer
262 available. Replace any occurrence of C<if BLOCK> by C<if (do BLOCK)>.
266 Perl's C-style C<for> loop works exactly like the corresponding C<while> loop;
267 that means that this:
269 for ($i = 1; $i < 10; $i++) {
282 (There is one minor difference: The first form implies a lexical scope
283 for variables declared with C<my> in the initialization expression.)
285 Besides the normal array index looping, C<for> can lend itself
286 to many other interesting applications. Here's one that avoids the
287 problem you get into if you explicitly test for end-of-file on
288 an interactive file descriptor causing your program to appear to
291 $on_a_tty = -t STDIN && -t STDOUT;
292 sub prompt { print "yes? " if $on_a_tty }
293 for ( prompt(); <STDIN>; prompt() ) {
299 The C<foreach> loop iterates over a normal list value and sets the
300 variable VAR to be each element of the list in turn. If the variable
301 is preceded with the keyword C<my>, then it is lexically scoped, and
302 is therefore visible only within the loop. Otherwise, the variable is
303 implicitly local to the loop and regains its former value upon exiting
304 the loop. If the variable was previously declared with C<my>, it uses
305 that variable instead of the global one, but it's still localized to
308 The C<foreach> keyword is actually a synonym for the C<for> keyword, so
309 you can use C<foreach> for readability or C<for> for brevity. (Or because
310 the Bourne shell is more familiar to you than I<csh>, so writing C<for>
311 comes more naturally.) If VAR is omitted, C<$_> is set to each value.
313 If any element of LIST is an lvalue, you can modify it by modifying
314 VAR inside the loop. Conversely, if any element of LIST is NOT an
315 lvalue, any attempt to modify that element will fail. In other words,
316 the C<foreach> loop index variable is an implicit alias for each item
317 in the list that you're looping over.
319 If any part of LIST is an array, C<foreach> will get very confused if
320 you add or remove elements within the loop body, for example with
321 C<splice>. So don't do that.
323 C<foreach> probably won't do what you expect if VAR is a tied or other
324 special variable. Don't do that either.
328 for (@ary) { s/foo/bar/ }
330 for my $elem (@elements) {
334 for $count (10,9,8,7,6,5,4,3,2,1,'BOOM') {
335 print $count, "\n"; sleep(1);
338 for (1..15) { print "Merry Christmas\n"; }
340 foreach $item (split(/:[\\\n:]*/, $ENV{TERMCAP})) {
341 print "Item: $item\n";
344 Here's how a C programmer might code up a particular algorithm in Perl:
346 for (my $i = 0; $i < @ary1; $i++) {
347 for (my $j = 0; $j < @ary2; $j++) {
348 if ($ary1[$i] > $ary2[$j]) {
349 last; # can't go to outer :-(
351 $ary1[$i] += $ary2[$j];
353 # this is where that last takes me
356 Whereas here's how a Perl programmer more comfortable with the idiom might
359 OUTER: for my $wid (@ary1) {
360 INNER: for my $jet (@ary2) {
361 next OUTER if $wid > $jet;
366 See how much easier this is? It's cleaner, safer, and faster. It's
367 cleaner because it's less noisy. It's safer because if code gets added
368 between the inner and outer loops later on, the new code won't be
369 accidentally executed. The C<next> explicitly iterates the other loop
370 rather than merely terminating the inner one. And it's faster because
371 Perl executes a C<foreach> statement more rapidly than it would the
372 equivalent C<for> loop.
374 =head2 Basic BLOCKs and Switch Statements
376 A BLOCK by itself (labeled or not) is semantically equivalent to a
377 loop that executes once. Thus you can use any of the loop control
378 statements in it to leave or restart the block. (Note that this is
379 I<NOT> true in C<eval{}>, C<sub{}>, or contrary to popular belief
380 C<do{}> blocks, which do I<NOT> count as loops.) The C<continue>
383 The BLOCK construct is particularly nice for doing case
387 if (/^abc/) { $abc = 1; last SWITCH; }
388 if (/^def/) { $def = 1; last SWITCH; }
389 if (/^xyz/) { $xyz = 1; last SWITCH; }
393 There is no official C<switch> statement in Perl, because there are
394 already several ways to write the equivalent. In addition to the
395 above, you could write
398 $abc = 1, last SWITCH if /^abc/;
399 $def = 1, last SWITCH if /^def/;
400 $xyz = 1, last SWITCH if /^xyz/;
404 (That's actually not as strange as it looks once you realize that you can
405 use loop control "operators" within an expression, That's just the normal
411 /^abc/ && do { $abc = 1; last SWITCH; };
412 /^def/ && do { $def = 1; last SWITCH; };
413 /^xyz/ && do { $xyz = 1; last SWITCH; };
417 or formatted so it stands out more as a "proper" C<switch> statement:
440 /^abc/ and $abc = 1, last SWITCH;
441 /^def/ and $def = 1, last SWITCH;
442 /^xyz/ and $xyz = 1, last SWITCH;
457 A common idiom for a C<switch> statement is to use C<foreach>'s aliasing to make
458 a temporary assignment to C<$_> for convenient matching:
460 SWITCH: for ($where) {
461 /In Card Names/ && do { push @flags, '-e'; last; };
462 /Anywhere/ && do { push @flags, '-h'; last; };
463 /In Rulings/ && do { last; };
464 die "unknown value for form variable where: `$where'";
467 Another interesting approach to a switch statement is arrange
468 for a C<do> block to return the proper value:
471 if ($flag & O_RDONLY) { "r" } # XXX: isn't this 0?
472 elsif ($flag & O_WRONLY) { ($flag & O_APPEND) ? "a" : "w" }
473 elsif ($flag & O_RDWR) {
474 if ($flag & O_CREAT) { "w+" }
475 else { ($flag & O_APPEND) ? "a+" : "r+" }
482 ($flags & O_WRONLY) ? "write-only" :
483 ($flags & O_RDWR) ? "read-write" :
487 Or if you are certainly that all the C<&&> clauses are true, you can use
488 something like this, which "switches" on the value of the
489 C<HTTP_USER_AGENT> environment variable.
492 # pick out jargon file page based on browser
493 $dir = 'http://www.wins.uva.nl/~mes/jargon';
494 for ($ENV{HTTP_USER_AGENT}) {
495 $page = /Mac/ && 'm/Macintrash.html'
496 || /Win(dows )?NT/ && 'e/evilandrude.html'
497 || /Win|MSIE|WebTV/ && 'm/MicroslothWindows.html'
498 || /Linux/ && 'l/Linux.html'
499 || /HP-UX/ && 'h/HP-SUX.html'
500 || /SunOS/ && 's/ScumOS.html'
501 || 'a/AppendixB.html';
503 print "Location: $dir/$page\015\012\015\012";
505 That kind of switch statement only works when you know the C<&&> clauses
506 will be true. If you don't, the previous C<?:> example should be used.
508 You might also consider writing a hash of subroutine references
509 instead of synthesizing a C<switch> statement.
513 Although not for the faint of heart, Perl does support a C<goto>
514 statement. There are three forms: C<goto>-LABEL, C<goto>-EXPR, and
515 C<goto>-&NAME. A loop's LABEL is not actually a valid target for
516 a C<goto>; it's just the name of the loop.
518 The C<goto>-LABEL form finds the statement labeled with LABEL and resumes
519 execution there. It may not be used to go into any construct that
520 requires initialization, such as a subroutine or a C<foreach> loop. It
521 also can't be used to go into a construct that is optimized away. It
522 can be used to go almost anywhere else within the dynamic scope,
523 including out of subroutines, but it's usually better to use some other
524 construct such as C<last> or C<die>. The author of Perl has never felt the
525 need to use this form of C<goto> (in Perl, that is--C is another matter).
527 The C<goto>-EXPR form expects a label name, whose scope will be resolved
528 dynamically. This allows for computed C<goto>s per FORTRAN, but isn't
529 necessarily recommended if you're optimizing for maintainability:
531 goto(("FOO", "BAR", "GLARCH")[$i]);
533 The C<goto>-&NAME form is highly magical, and substitutes a call to the
534 named subroutine for the currently running subroutine. This is used by
535 C<AUTOLOAD()> subroutines that wish to load another subroutine and then
536 pretend that the other subroutine had been called in the first place
537 (except that any modifications to C<@_> in the current subroutine are
538 propagated to the other subroutine.) After the C<goto>, not even C<caller()>
539 will be able to tell that this routine was called first.
541 In almost all cases like this, it's usually a far, far better idea to use the
542 structured control flow mechanisms of C<next>, C<last>, or C<redo> instead of
543 resorting to a C<goto>. For certain applications, the catch and throw pair of
544 C<eval{}> and die() for exception processing can also be a prudent approach.
546 =head2 PODs: Embedded Documentation
548 Perl has a mechanism for intermixing documentation with source code.
549 While it's expecting the beginning of a new statement, if the compiler
550 encounters a line that begins with an equal sign and a word, like this
552 =head1 Here There Be Pods!
554 Then that text and all remaining text up through and including a line
555 beginning with C<=cut> will be ignored. The format of the intervening
556 text is described in L<perlpod>.
558 This allows you to intermix your source code
559 and your documentation text freely, as in
563 The snazzle() function will behave in the most spectacular
564 form that you can possibly imagine, not even excepting
565 cybernetic pyrotechnics.
567 =cut back to the compiler, nuff of this pod stuff!
574 Note that pod translators should look at only paragraphs beginning
575 with a pod directive (it makes parsing easier), whereas the compiler
576 actually knows to look for pod escapes even in the middle of a
577 paragraph. This means that the following secret stuff will be
578 ignored by both the compiler and the translators.
582 warn "Neither POD nor CODE!?"
586 You probably shouldn't rely upon the C<warn()> being podded out forever.
587 Not all pod translators are well-behaved in this regard, and perhaps
588 the compiler will become pickier.
590 One may also use pod directives to quickly comment out a section
593 =head2 Plain Old Comments (Not!)
595 Much like the C preprocessor, Perl can process line directives. Using
596 this, one can control Perl's idea of filenames and line numbers in
597 error or warning messages (especially for strings that are processed
598 with C<eval()>). The syntax for this mechanism is the same as for most
599 C preprocessors: it matches the regular expression
600 C</^#\s*line\s+(\d+)\s*(?:\s"([^"]+)")?\s*$/> with C<$1> being the line
601 number for the next line, and C<$2> being the optional filename
602 (specified within quotes).
604 There is a fairly obvious gotcha included with the line directive:
605 Debuggers and profilers will only show the last source line to appear
606 at a particular line number in a given file. Care should be taken not
607 to cause line number collisions in code you'd like to debug later.
609 Here are some examples that you should be able to type into your command
614 # the `#' on the previous line must be the first char on line
617 foo at bzzzt line 201.
621 eval qq[\n#line 2001 ""\ndie 'foo']; print $@;
626 eval qq[\n#line 200 "foo bar"\ndie 'foo']; print $@;
628 foo at foo bar line 200.
632 eval "\n#line " . __LINE__ . ' "' . __FILE__ ."\"\ndie 'foo'";
635 foo at goop line 345.