4 perlfunc - Perl builtin functions
8 The functions in this section can serve as terms in an expression.
9 They fall into two major categories: list operators and named unary
10 operators. These differ in their precedence relationship with a
11 following comma. (See the precedence table in L<perlop>.) List
12 operators take more than one argument, while unary operators can never
13 take more than one argument. Thus, a comma terminates the argument of
14 a unary operator, but merely separates the arguments of a list
15 operator. A unary operator generally provides a scalar context to its
16 argument, while a list operator may provide either scalar or list
17 contexts for its arguments. If it does both, the scalar arguments will
18 be first, and the list argument will follow. (Note that there can ever
19 be only one such list argument.) For instance, splice() has three scalar
20 arguments followed by a list, whereas gethostbyname() has four scalar
23 In the syntax descriptions that follow, list operators that expect a
24 list (and provide list context for the elements of the list) are shown
25 with LIST as an argument. Such a list may consist of any combination
26 of scalar arguments or list values; the list values will be included
27 in the list as if each individual element were interpolated at that
28 point in the list, forming a longer single-dimensional list value.
29 Commas should separate elements of the LIST.
31 Any function in the list below may be used either with or without
32 parentheses around its arguments. (The syntax descriptions omit the
33 parentheses.) If you use the parentheses, the simple (but occasionally
34 surprising) rule is this: It I<looks> like a function, therefore it I<is> a
35 function, and precedence doesn't matter. Otherwise it's a list
36 operator or unary operator, and precedence does matter. And whitespace
37 between the function and left parenthesis doesn't count--so you need to
40 print 1+2+4; # Prints 7.
41 print(1+2) + 4; # Prints 3.
42 print (1+2)+4; # Also prints 3!
43 print +(1+2)+4; # Prints 7.
44 print ((1+2)+4); # Prints 7.
46 If you run Perl with the B<-w> switch it can warn you about this. For
47 example, the third line above produces:
49 print (...) interpreted as function at - line 1.
50 Useless use of integer addition in void context at - line 1.
52 A few functions take no arguments at all, and therefore work as neither
53 unary nor list operators. These include such functions as C<time>
54 and C<endpwent>. For example, C<time+86_400> always means
57 For functions that can be used in either a scalar or list context,
58 nonabortive failure is generally indicated in a scalar context by
59 returning the undefined value, and in a list context by returning the
62 Remember the following important rule: There is B<no rule> that relates
63 the behavior of an expression in list context to its behavior in scalar
64 context, or vice versa. It might do two totally different things.
65 Each operator and function decides which sort of value it would be most
66 appropriate to return in scalar context. Some operators return the
67 length of the list that would have been returned in list context. Some
68 operators return the first value in the list. Some operators return the
69 last value in the list. Some operators return a count of successful
70 operations. In general, they do what you want, unless you want
74 A named array in scalar context is quite different from what would at
75 first glance appear to be a list in scalar context. You can't get a list
76 like C<(1,2,3)> into being in scalar context, because the compiler knows
77 the context at compile time. It would generate the scalar comma operator
78 there, not the list construction version of the comma. That means it
79 was never a list to start with.
81 In general, functions in Perl that serve as wrappers for system calls
82 of the same name (like chown(2), fork(2), closedir(2), etc.) all return
83 true when they succeed and C<undef> otherwise, as is usually mentioned
84 in the descriptions below. This is different from the C interfaces,
85 which return C<-1> on failure. Exceptions to this rule are C<wait>,
86 C<waitpid>, and C<syscall>. System calls also set the special C<$!>
87 variable on failure. Other functions do not, except accidentally.
89 =head2 Perl Functions by Category
92 Here are Perl's functions (including things that look like
93 functions, like some keywords and named operators)
94 arranged by category. Some functions appear in more
99 =item Functions for SCALARs or strings
100 X<scalar> X<string> X<character>
102 C<chomp>, C<chop>, C<chr>, C<crypt>, C<hex>, C<index>, C<lc>, C<lcfirst>,
103 C<length>, C<oct>, C<ord>, C<pack>, C<q/STRING/>, C<qq/STRING/>, C<reverse>,
104 C<rindex>, C<sprintf>, C<substr>, C<tr///>, C<uc>, C<ucfirst>, C<y///>
106 =item Regular expressions and pattern matching
107 X<regular expression> X<regex> X<regexp>
109 C<m//>, C<pos>, C<quotemeta>, C<s///>, C<split>, C<study>, C<qr//>
111 =item Numeric functions
112 X<numeric> X<number> X<trigonometric> X<trigonometry>
114 C<abs>, C<atan2>, C<cos>, C<exp>, C<hex>, C<int>, C<log>, C<oct>, C<rand>,
115 C<sin>, C<sqrt>, C<srand>
117 =item Functions for real @ARRAYs
120 C<pop>, C<push>, C<shift>, C<splice>, C<unshift>
122 =item Functions for list data
125 C<grep>, C<join>, C<map>, C<qw/STRING/>, C<reverse>, C<sort>, C<unpack>
127 =item Functions for real %HASHes
130 C<delete>, C<each>, C<exists>, C<keys>, C<values>
132 =item Input and output functions
133 X<I/O> X<input> X<output> X<dbm>
135 C<binmode>, C<close>, C<closedir>, C<dbmclose>, C<dbmopen>, C<die>, C<eof>,
136 C<fileno>, C<flock>, C<format>, C<getc>, C<print>, C<printf>, C<read>,
137 C<readdir>, C<rewinddir>, C<say>, C<seek>, C<seekdir>, C<select>, C<syscall>,
138 C<sysread>, C<sysseek>, C<syswrite>, C<tell>, C<telldir>, C<truncate>,
141 =item Functions for fixed length data or records
143 C<pack>, C<read>, C<syscall>, C<sysread>, C<syswrite>, C<unpack>, C<vec>
145 =item Functions for filehandles, files, or directories
146 X<file> X<filehandle> X<directory> X<pipe> X<link> X<symlink>
148 C<-I<X>>, C<chdir>, C<chmod>, C<chown>, C<chroot>, C<fcntl>, C<glob>,
149 C<ioctl>, C<link>, C<lstat>, C<mkdir>, C<open>, C<opendir>,
150 C<readlink>, C<rename>, C<rmdir>, C<stat>, C<symlink>, C<sysopen>,
151 C<umask>, C<unlink>, C<utime>
153 =item Keywords related to the control flow of your Perl program
156 C<caller>, C<continue>, C<die>, C<do>, C<dump>, C<eval>, C<exit>,
157 C<goto>, C<last>, C<next>, C<redo>, C<return>, C<sub>, C<wantarray>
159 =item Keywords related to switch
161 C<break>, C<continue>
163 (These are only available if you enable the "switch" feature.
164 See L<feature> and L<perlsyn/"Switch statements">.)
166 =item Keywords related to scoping
168 C<caller>, C<import>, C<local>, C<my>, C<our>, C<package>, C<use>
170 =item Miscellaneous functions
172 C<defined>, C<dump>, C<eval>, C<formline>, C<local>, C<my>, C<our>, C<reset>,
173 C<scalar>, C<undef>, C<wantarray>
175 =item Functions for processes and process groups
176 X<process> X<pid> X<process id>
178 C<alarm>, C<exec>, C<fork>, C<getpgrp>, C<getppid>, C<getpriority>, C<kill>,
179 C<pipe>, C<qx/STRING/>, C<setpgrp>, C<setpriority>, C<sleep>, C<system>,
180 C<times>, C<wait>, C<waitpid>
182 =item Keywords related to perl modules
185 C<do>, C<import>, C<no>, C<package>, C<require>, C<use>
187 =item Keywords related to classes and object-orientedness
188 X<object> X<class> X<package>
190 C<bless>, C<dbmclose>, C<dbmopen>, C<package>, C<ref>, C<tie>, C<tied>,
193 =item Low-level socket functions
196 C<accept>, C<bind>, C<connect>, C<getpeername>, C<getsockname>,
197 C<getsockopt>, C<listen>, C<recv>, C<send>, C<setsockopt>, C<shutdown>,
198 C<socket>, C<socketpair>
200 =item System V interprocess communication functions
201 X<IPC> X<System V> X<semaphore> X<shared memory> X<memory> X<message>
203 C<msgctl>, C<msgget>, C<msgrcv>, C<msgsnd>, C<semctl>, C<semget>, C<semop>,
204 C<shmctl>, C<shmget>, C<shmread>, C<shmwrite>
206 =item Fetching user and group info
207 X<user> X<group> X<password> X<uid> X<gid> X<passwd> X</etc/passwd>
209 C<endgrent>, C<endhostent>, C<endnetent>, C<endpwent>, C<getgrent>,
210 C<getgrgid>, C<getgrnam>, C<getlogin>, C<getpwent>, C<getpwnam>,
211 C<getpwuid>, C<setgrent>, C<setpwent>
213 =item Fetching network info
214 X<network> X<protocol> X<host> X<hostname> X<IP> X<address> X<service>
216 C<endprotoent>, C<endservent>, C<gethostbyaddr>, C<gethostbyname>,
217 C<gethostent>, C<getnetbyaddr>, C<getnetbyname>, C<getnetent>,
218 C<getprotobyname>, C<getprotobynumber>, C<getprotoent>,
219 C<getservbyname>, C<getservbyport>, C<getservent>, C<sethostent>,
220 C<setnetent>, C<setprotoent>, C<setservent>
222 =item Time-related functions
225 C<gmtime>, C<localtime>, C<time>, C<times>
227 =item Functions new in perl5
230 C<abs>, C<bless>, C<chomp>, C<chr>, C<exists>, C<formline>, C<glob>,
231 C<import>, C<lc>, C<lcfirst>, C<map>, C<my>, C<no>, C<our>, C<prototype>,
232 C<qx>, C<qw>, C<readline>, C<readpipe>, C<ref>, C<sub*>, C<sysopen>, C<tie>,
233 C<tied>, C<uc>, C<ucfirst>, C<untie>, C<use>
235 * - C<sub> was a keyword in perl4, but in perl5 it is an
236 operator, which can be used in expressions.
238 =item Functions obsoleted in perl5
240 C<dbmclose>, C<dbmopen>
245 X<portability> X<Unix> X<portable>
247 Perl was born in Unix and can therefore access all common Unix
248 system calls. In non-Unix environments, the functionality of some
249 Unix system calls may not be available, or details of the available
250 functionality may differ slightly. The Perl functions affected
253 C<-X>, C<binmode>, C<chmod>, C<chown>, C<chroot>, C<crypt>,
254 C<dbmclose>, C<dbmopen>, C<dump>, C<endgrent>, C<endhostent>,
255 C<endnetent>, C<endprotoent>, C<endpwent>, C<endservent>, C<exec>,
256 C<fcntl>, C<flock>, C<fork>, C<getgrent>, C<getgrgid>, C<gethostbyname>,
257 C<gethostent>, C<getlogin>, C<getnetbyaddr>, C<getnetbyname>, C<getnetent>,
258 C<getppid>, C<getpgrp>, C<getpriority>, C<getprotobynumber>,
259 C<getprotoent>, C<getpwent>, C<getpwnam>, C<getpwuid>,
260 C<getservbyport>, C<getservent>, C<getsockopt>, C<glob>, C<ioctl>,
261 C<kill>, C<link>, C<lstat>, C<msgctl>, C<msgget>, C<msgrcv>,
262 C<msgsnd>, C<open>, C<pipe>, C<readlink>, C<rename>, C<select>, C<semctl>,
263 C<semget>, C<semop>, C<setgrent>, C<sethostent>, C<setnetent>,
264 C<setpgrp>, C<setpriority>, C<setprotoent>, C<setpwent>,
265 C<setservent>, C<setsockopt>, C<shmctl>, C<shmget>, C<shmread>,
266 C<shmwrite>, C<socket>, C<socketpair>,
267 C<stat>, C<symlink>, C<syscall>, C<sysopen>, C<system>,
268 C<times>, C<truncate>, C<umask>, C<unlink>,
269 C<utime>, C<wait>, C<waitpid>
271 For more information about the portability of these functions, see
272 L<perlport> and other available platform-specific documentation.
274 =head2 Alphabetical Listing of Perl Functions
279 X<-r>X<-w>X<-x>X<-o>X<-R>X<-W>X<-X>X<-O>X<-e>X<-z>X<-s>X<-f>X<-d>X<-l>X<-p>
280 X<-S>X<-b>X<-c>X<-t>X<-u>X<-g>X<-k>X<-T>X<-B>X<-M>X<-A>X<-C>
286 A file test, where X is one of the letters listed below. This unary
287 operator takes one argument, either a filename or a filehandle, and
288 tests the associated file to see if something is true about it. If the
289 argument is omitted, tests C<$_>, except for C<-t>, which tests STDIN.
290 Unless otherwise documented, it returns C<1> for true and C<''> for false, or
291 the undefined value if the file doesn't exist. Despite the funny
292 names, precedence is the same as any other named unary operator, and
293 the argument may be parenthesized like any other unary operator. The
294 operator may be any of:
296 -r File is readable by effective uid/gid.
297 -w File is writable by effective uid/gid.
298 -x File is executable by effective uid/gid.
299 -o File is owned by effective uid.
301 -R File is readable by real uid/gid.
302 -W File is writable by real uid/gid.
303 -X File is executable by real uid/gid.
304 -O File is owned by real uid.
307 -z File has zero size (is empty).
308 -s File has nonzero size (returns size in bytes).
310 -f File is a plain file.
311 -d File is a directory.
312 -l File is a symbolic link.
313 -p File is a named pipe (FIFO), or Filehandle is a pipe.
315 -b File is a block special file.
316 -c File is a character special file.
317 -t Filehandle is opened to a tty.
319 -u File has setuid bit set.
320 -g File has setgid bit set.
321 -k File has sticky bit set.
323 -T File is an ASCII text file (heuristic guess).
324 -B File is a "binary" file (opposite of -T).
326 -M Script start time minus file modification time, in days.
327 -A Same for access time.
328 -C Same for inode change time (Unix, may differ for other platforms)
334 next unless -f $_; # ignore specials
338 The interpretation of the file permission operators C<-r>, C<-R>,
339 C<-w>, C<-W>, C<-x>, and C<-X> is by default based solely on the mode
340 of the file and the uids and gids of the user. There may be other
341 reasons you can't actually read, write, or execute the file. Such
342 reasons may be for example network filesystem access controls, ACLs
343 (access control lists), read-only filesystems, and unrecognized
346 Also note that, for the superuser on the local filesystems, the C<-r>,
347 C<-R>, C<-w>, and C<-W> tests always return 1, and C<-x> and C<-X> return 1
348 if any execute bit is set in the mode. Scripts run by the superuser
349 may thus need to do a stat() to determine the actual mode of the file,
350 or temporarily set their effective uid to something else.
352 If you are using ACLs, there is a pragma called C<filetest> that may
353 produce more accurate results than the bare stat() mode bits.
354 When under the C<use filetest 'access'> the above-mentioned filetests
355 will test whether the permission can (not) be granted using the
356 access() family of system calls. Also note that the C<-x> and C<-X> may
357 under this pragma return true even if there are no execute permission
358 bits set (nor any extra execute permission ACLs). This strangeness is
359 due to the underlying system calls' definitions. Read the
360 documentation for the C<filetest> pragma for more information.
362 Note that C<-s/a/b/> does not do a negated substitution. Saying
363 C<-exp($foo)> still works as expected, however--only single letters
364 following a minus are interpreted as file tests.
366 The C<-T> and C<-B> switches work as follows. The first block or so of the
367 file is examined for odd characters such as strange control codes or
368 characters with the high bit set. If too many strange characters (>30%)
369 are found, it's a C<-B> file; otherwise it's a C<-T> file. Also, any file
370 containing null in the first block is considered a binary file. If C<-T>
371 or C<-B> is used on a filehandle, the current IO buffer is examined
372 rather than the first block. Both C<-T> and C<-B> return true on a null
373 file, or a file at EOF when testing a filehandle. Because you have to
374 read a file to do the C<-T> test, on most occasions you want to use a C<-f>
375 against the file first, as in C<next unless -f $file && -T $file>.
377 If any of the file tests (or either the C<stat> or C<lstat> operators) are given
378 the special filehandle consisting of a solitary underline, then the stat
379 structure of the previous file test (or stat operator) is used, saving
380 a system call. (This doesn't work with C<-t>, and you need to remember
381 that lstat() and C<-l> will leave values in the stat structure for the
382 symbolic link, not the real file.) (Also, if the stat buffer was filled by
383 an C<lstat> call, C<-T> and C<-B> will reset it with the results of C<stat _>).
386 print "Can do.\n" if -r $a || -w _ || -x _;
389 print "Readable\n" if -r _;
390 print "Writable\n" if -w _;
391 print "Executable\n" if -x _;
392 print "Setuid\n" if -u _;
393 print "Setgid\n" if -g _;
394 print "Sticky\n" if -k _;
395 print "Text\n" if -T _;
396 print "Binary\n" if -B _;
398 As of Perl 5.9.1, as a form of purely syntactic sugar, you can stack file
399 test operators, in a way that C<-f -w -x $file> is equivalent to
400 C<-x $file && -w _ && -f _>. (This is only syntax fancy: if you use
401 the return value of C<-f $file> as an argument to another filetest
402 operator, no special magic will happen.)
409 Returns the absolute value of its argument.
410 If VALUE is omitted, uses C<$_>.
412 =item accept NEWSOCKET,GENERICSOCKET
415 Accepts an incoming socket connect, just as the accept(2) system call
416 does. Returns the packed address if it succeeded, false otherwise.
417 See the example in L<perlipc/"Sockets: Client/Server Communication">.
419 On systems that support a close-on-exec flag on files, the flag will
420 be set for the newly opened file descriptor, as determined by the
421 value of $^F. See L<perlvar/$^F>.
430 Arranges to have a SIGALRM delivered to this process after the
431 specified number of wallclock seconds has elapsed. If SECONDS is not
432 specified, the value stored in C<$_> is used. (On some machines,
433 unfortunately, the elapsed time may be up to one second less or more
434 than you specified because of how seconds are counted, and process
435 scheduling may delay the delivery of the signal even further.)
437 Only one timer may be counting at once. Each call disables the
438 previous timer, and an argument of C<0> may be supplied to cancel the
439 previous timer without starting a new one. The returned value is the
440 amount of time remaining on the previous timer.
442 For delays of finer granularity than one second, you may use Perl's
443 four-argument version of select() leaving the first three arguments
444 undefined, or you might be able to use the C<syscall> interface to
445 access setitimer(2) if your system supports it. The Time::HiRes
446 module (from CPAN, and starting from Perl 5.8 part of the standard
447 distribution) may also prove useful.
449 It is usually a mistake to intermix C<alarm> and C<sleep> calls.
450 (C<sleep> may be internally implemented in your system with C<alarm>)
452 If you want to use C<alarm> to time out a system call you need to use an
453 C<eval>/C<die> pair. You can't rely on the alarm causing the system call to
454 fail with C<$!> set to C<EINTR> because Perl sets up signal handlers to
455 restart system calls on some systems. Using C<eval>/C<die> always works,
456 modulo the caveats given in L<perlipc/"Signals">.
459 local $SIG{ALRM} = sub { die "alarm\n" }; # NB: \n required
461 $nread = sysread SOCKET, $buffer, $size;
465 die unless $@ eq "alarm\n"; # propagate unexpected errors
472 For more information see L<perlipc>.
475 X<atan2> X<arctangent> X<tan> X<tangent>
477 Returns the arctangent of Y/X in the range -PI to PI.
479 For the tangent operation, you may use the C<Math::Trig::tan>
480 function, or use the familiar relation:
482 sub tan { sin($_[0]) / cos($_[0]) }
484 Note that atan2(0, 0) is not well-defined.
486 =item bind SOCKET,NAME
489 Binds a network address to a socket, just as the bind system call
490 does. Returns true if it succeeded, false otherwise. NAME should be a
491 packed address of the appropriate type for the socket. See the examples in
492 L<perlipc/"Sockets: Client/Server Communication">.
494 =item binmode FILEHANDLE, LAYER
495 X<binmode> X<binary> X<text> X<DOS> X<Windows>
497 =item binmode FILEHANDLE
499 Arranges for FILEHANDLE to be read or written in "binary" or "text"
500 mode on systems where the run-time libraries distinguish between
501 binary and text files. If FILEHANDLE is an expression, the value is
502 taken as the name of the filehandle. Returns true on success,
503 otherwise it returns C<undef> and sets C<$!> (errno).
505 On some systems (in general, DOS and Windows-based systems) binmode()
506 is necessary when you're not working with a text file. For the sake
507 of portability it is a good idea to always use it when appropriate,
508 and to never use it when it isn't appropriate. Also, people can
509 set their I/O to be by default UTF-8 encoded Unicode, not bytes.
511 In other words: regardless of platform, use binmode() on binary data,
512 like for example images.
514 If LAYER is present it is a single string, but may contain multiple
515 directives. The directives alter the behaviour of the file handle.
516 When LAYER is present using binmode on text file makes sense.
518 If LAYER is omitted or specified as C<:raw> the filehandle is made
519 suitable for passing binary data. This includes turning off possible CRLF
520 translation and marking it as bytes (as opposed to Unicode characters).
521 Note that, despite what may be implied in I<"Programming Perl"> (the
522 Camel) or elsewhere, C<:raw> is I<not> the simply inverse of C<:crlf>
523 -- other layers which would affect binary nature of the stream are
524 I<also> disabled. See L<PerlIO>, L<perlrun> and the discussion about the
525 PERLIO environment variable.
527 The C<:bytes>, C<:crlf>, and C<:utf8>, and any other directives of the
528 form C<:...>, are called I/O I<layers>. The C<open> pragma can be used to
529 establish default I/O layers. See L<open>.
531 I<The LAYER parameter of the binmode() function is described as "DISCIPLINE"
532 in "Programming Perl, 3rd Edition". However, since the publishing of this
533 book, by many known as "Camel III", the consensus of the naming of this
534 functionality has moved from "discipline" to "layer". All documentation
535 of this version of Perl therefore refers to "layers" rather than to
536 "disciplines". Now back to the regularly scheduled documentation...>
538 To mark FILEHANDLE as UTF-8, use C<:utf8>.
540 In general, binmode() should be called after open() but before any I/O
541 is done on the filehandle. Calling binmode() will normally flush any
542 pending buffered output data (and perhaps pending input data) on the
543 handle. An exception to this is the C<:encoding> layer that
544 changes the default character encoding of the handle, see L<open>.
545 The C<:encoding> layer sometimes needs to be called in
546 mid-stream, and it doesn't flush the stream. The C<:encoding>
547 also implicitly pushes on top of itself the C<:utf8> layer because
548 internally Perl will operate on UTF-8 encoded Unicode characters.
550 The operating system, device drivers, C libraries, and Perl run-time
551 system all work together to let the programmer treat a single
552 character (C<\n>) as the line terminator, irrespective of the external
553 representation. On many operating systems, the native text file
554 representation matches the internal representation, but on some
555 platforms the external representation of C<\n> is made up of more than
558 Mac OS, all variants of Unix, and Stream_LF files on VMS use a single
559 character to end each line in the external representation of text (even
560 though that single character is CARRIAGE RETURN on Mac OS and LINE FEED
561 on Unix and most VMS files). In other systems like OS/2, DOS and the
562 various flavors of MS-Windows your program sees a C<\n> as a simple C<\cJ>,
563 but what's stored in text files are the two characters C<\cM\cJ>. That
564 means that, if you don't use binmode() on these systems, C<\cM\cJ>
565 sequences on disk will be converted to C<\n> on input, and any C<\n> in
566 your program will be converted back to C<\cM\cJ> on output. This is what
567 you want for text files, but it can be disastrous for binary files.
569 Another consequence of using binmode() (on some systems) is that
570 special end-of-file markers will be seen as part of the data stream.
571 For systems from the Microsoft family this means that if your binary
572 data contains C<\cZ>, the I/O subsystem will regard it as the end of
573 the file, unless you use binmode().
575 binmode() is not only important for readline() and print() operations,
576 but also when using read(), seek(), sysread(), syswrite() and tell()
577 (see L<perlport> for more details). See the C<$/> and C<$\> variables
578 in L<perlvar> for how to manually set your input and output
579 line-termination sequences.
581 =item bless REF,CLASSNAME
586 This function tells the thingy referenced by REF that it is now an object
587 in the CLASSNAME package. If CLASSNAME is omitted, the current package
588 is used. Because a C<bless> is often the last thing in a constructor,
589 it returns the reference for convenience. Always use the two-argument
590 version if a derived class might inherit the function doing the blessing.
591 See L<perltoot> and L<perlobj> for more about the blessing (and blessings)
594 Consider always blessing objects in CLASSNAMEs that are mixed case.
595 Namespaces with all lowercase names are considered reserved for
596 Perl pragmata. Builtin types have all uppercase names. To prevent
597 confusion, you may wish to avoid such package names as well. Make sure
598 that CLASSNAME is a true value.
600 See L<perlmod/"Perl Modules">.
604 Break out of a C<given()> block.
606 This keyword is enabled by the "switch" feature: see L<feature>
607 for more information.
610 X<caller> X<call stack> X<stack> X<stack trace>
614 Returns the context of the current subroutine call. In scalar context,
615 returns the caller's package name if there is a caller, that is, if
616 we're in a subroutine or C<eval> or C<require>, and the undefined value
617 otherwise. In list context, returns
619 ($package, $filename, $line) = caller;
621 With EXPR, it returns some extra information that the debugger uses to
622 print a stack trace. The value of EXPR indicates how many call frames
623 to go back before the current one.
625 ($package, $filename, $line, $subroutine, $hasargs,
626 $wantarray, $evaltext, $is_require, $hints, $bitmask) = caller($i);
628 Here $subroutine may be C<(eval)> if the frame is not a subroutine
629 call, but an C<eval>. In such a case additional elements $evaltext and
630 C<$is_require> are set: C<$is_require> is true if the frame is created by a
631 C<require> or C<use> statement, $evaltext contains the text of the
632 C<eval EXPR> statement. In particular, for an C<eval BLOCK> statement,
633 $filename is C<(eval)>, but $evaltext is undefined. (Note also that
634 each C<use> statement creates a C<require> frame inside an C<eval EXPR>
635 frame.) $subroutine may also be C<(unknown)> if this particular
636 subroutine happens to have been deleted from the symbol table.
637 C<$hasargs> is true if a new instance of C<@_> was set up for the frame.
638 C<$hints> and C<$bitmask> contain pragmatic hints that the caller was
639 compiled with. The C<$hints> and C<$bitmask> values are subject to change
640 between versions of Perl, and are not meant for external use.
642 Furthermore, when called from within the DB package, caller returns more
643 detailed information: it sets the list variable C<@DB::args> to be the
644 arguments with which the subroutine was invoked.
646 Be aware that the optimizer might have optimized call frames away before
647 C<caller> had a chance to get the information. That means that C<caller(N)>
648 might not return information about the call frame you expect it do, for
649 C<< N > 1 >>. In particular, C<@DB::args> might have information from the
650 previous time C<caller> was called.
656 =item chdir FILEHANDLE
658 =item chdir DIRHANDLE
662 Changes the working directory to EXPR, if possible. If EXPR is omitted,
663 changes to the directory specified by C<$ENV{HOME}>, if set; if not,
664 changes to the directory specified by C<$ENV{LOGDIR}>. (Under VMS, the
665 variable C<$ENV{SYS$LOGIN}> is also checked, and used if it is set.) If
666 neither is set, C<chdir> does nothing. It returns true upon success,
667 false otherwise. See the example under C<die>.
669 On systems that support fchdir, you might pass a file handle or
670 directory handle as argument. On systems that don't support fchdir,
671 passing handles produces a fatal error at run time.
674 X<chmod> X<permission> X<mode>
676 Changes the permissions of a list of files. The first element of the
677 list must be the numerical mode, which should probably be an octal
678 number, and which definitely should I<not> be a string of octal digits:
679 C<0644> is okay, C<'0644'> is not. Returns the number of files
680 successfully changed. See also L</oct>, if all you have is a string.
682 $cnt = chmod 0755, 'foo', 'bar';
683 chmod 0755, @executables;
684 $mode = '0644'; chmod $mode, 'foo'; # !!! sets mode to
686 $mode = '0644'; chmod oct($mode), 'foo'; # this is better
687 $mode = 0644; chmod $mode, 'foo'; # this is best
689 On systems that support fchmod, you might pass file handles among the
690 files. On systems that don't support fchmod, passing file handles
691 produces a fatal error at run time.
693 open(my $fh, "<", "foo");
694 my $perm = (stat $fh)[2] & 07777;
695 chmod($perm | 0600, $fh);
697 You can also import the symbolic C<S_I*> constants from the Fcntl
702 chmod S_IRWXU|S_IRGRP|S_IXGRP|S_IROTH|S_IXOTH, @executables;
703 # This is identical to the chmod 0755 of the above example.
706 X<chomp> X<INPUT_RECORD_SEPARATOR> X<$/> X<newline> X<eol>
712 This safer version of L</chop> removes any trailing string
713 that corresponds to the current value of C<$/> (also known as
714 $INPUT_RECORD_SEPARATOR in the C<English> module). It returns the total
715 number of characters removed from all its arguments. It's often used to
716 remove the newline from the end of an input record when you're worried
717 that the final record may be missing its newline. When in paragraph
718 mode (C<$/ = "">), it removes all trailing newlines from the string.
719 When in slurp mode (C<$/ = undef>) or fixed-length record mode (C<$/> is
720 a reference to an integer or the like, see L<perlvar>) chomp() won't
722 If VARIABLE is omitted, it chomps C<$_>. Example:
725 chomp; # avoid \n on last field
730 If VARIABLE is a hash, it chomps the hash's values, but not its keys.
732 You can actually chomp anything that's an lvalue, including an assignment:
735 chomp($answer = <STDIN>);
737 If you chomp a list, each element is chomped, and the total number of
738 characters removed is returned.
740 If the C<encoding> pragma is in scope then the lengths returned are
741 calculated from the length of C<$/> in Unicode characters, which is not
742 always the same as the length of C<$/> in the native encoding.
744 Note that parentheses are necessary when you're chomping anything
745 that is not a simple variable. This is because C<chomp $cwd = `pwd`;>
746 is interpreted as C<(chomp $cwd) = `pwd`;>, rather than as
747 C<chomp( $cwd = `pwd` )> which you might expect. Similarly,
748 C<chomp $a, $b> is interpreted as C<chomp($a), $b> rather than
758 Chops off the last character of a string and returns the character
759 chopped. It is much more efficient than C<s/.$//s> because it neither
760 scans nor copies the string. If VARIABLE is omitted, chops C<$_>.
761 If VARIABLE is a hash, it chops the hash's values, but not its keys.
763 You can actually chop anything that's an lvalue, including an assignment.
765 If you chop a list, each element is chopped. Only the value of the
766 last C<chop> is returned.
768 Note that C<chop> returns the last character. To return all but the last
769 character, use C<substr($string, 0, -1)>.
774 X<chown> X<owner> X<user> X<group>
776 Changes the owner (and group) of a list of files. The first two
777 elements of the list must be the I<numeric> uid and gid, in that
778 order. A value of -1 in either position is interpreted by most
779 systems to leave that value unchanged. Returns the number of files
780 successfully changed.
782 $cnt = chown $uid, $gid, 'foo', 'bar';
783 chown $uid, $gid, @filenames;
785 On systems that support fchown, you might pass file handles among the
786 files. On systems that don't support fchown, passing file handles
787 produces a fatal error at run time.
789 Here's an example that looks up nonnumeric uids in the passwd file:
792 chomp($user = <STDIN>);
794 chomp($pattern = <STDIN>);
796 ($login,$pass,$uid,$gid) = getpwnam($user)
797 or die "$user not in passwd file";
799 @ary = glob($pattern); # expand filenames
800 chown $uid, $gid, @ary;
802 On most systems, you are not allowed to change the ownership of the
803 file unless you're the superuser, although you should be able to change
804 the group to any of your secondary groups. On insecure systems, these
805 restrictions may be relaxed, but this is not a portable assumption.
806 On POSIX systems, you can detect this condition this way:
808 use POSIX qw(sysconf _PC_CHOWN_RESTRICTED);
809 $can_chown_giveaway = not sysconf(_PC_CHOWN_RESTRICTED);
812 X<chr> X<character> X<ASCII> X<Unicode>
816 Returns the character represented by that NUMBER in the character set.
817 For example, C<chr(65)> is C<"A"> in either ASCII or Unicode, and
818 chr(0x263a) is a Unicode smiley face. Note that characters from 128
819 to 255 (inclusive) are by default not encoded in UTF-8 Unicode for
820 backward compatibility reasons (but see L<encoding>).
822 Negative values give the Unicode replacement character (chr(0xfffd)),
823 except under the L<bytes> pragma, where low eight bits of the value
824 (truncated to an integer) are used.
826 If NUMBER is omitted, uses C<$_>.
828 For the reverse, use L</ord>.
830 Note that under the C<bytes> pragma the NUMBER is masked to
833 See L<perlunicode> and L<encoding> for more about Unicode.
835 =item chroot FILENAME
840 This function works like the system call by the same name: it makes the
841 named directory the new root directory for all further pathnames that
842 begin with a C</> by your process and all its children. (It doesn't
843 change your current working directory, which is unaffected.) For security
844 reasons, this call is restricted to the superuser. If FILENAME is
845 omitted, does a C<chroot> to C<$_>.
847 =item close FILEHANDLE
852 Closes the file or pipe associated with the file handle, returning
853 true only if IO buffers are successfully flushed and closes the system
854 file descriptor. Closes the currently selected filehandle if the
857 You don't have to close FILEHANDLE if you are immediately going to do
858 another C<open> on it, because C<open> will close it for you. (See
859 C<open>.) However, an explicit C<close> on an input file resets the line
860 counter (C<$.>), while the implicit close done by C<open> does not.
862 If the file handle came from a piped open, C<close> will additionally
863 return false if one of the other system calls involved fails, or if the
864 program exits with non-zero status. (If the only problem was that the
865 program exited non-zero, C<$!> will be set to C<0>.) Closing a pipe
866 also waits for the process executing on the pipe to complete, in case you
867 want to look at the output of the pipe afterwards, and
868 implicitly puts the exit status value of that command into C<$?> and
869 C<${^CHILD_ERROR_NATIVE}>.
871 Prematurely closing the read end of a pipe (i.e. before the process
872 writing to it at the other end has closed it) will result in a
873 SIGPIPE being delivered to the writer. If the other end can't
874 handle that, be sure to read all the data before closing the pipe.
878 open(OUTPUT, '|sort >foo') # pipe to sort
879 or die "Can't start sort: $!";
880 #... # print stuff to output
881 close OUTPUT # wait for sort to finish
882 or warn $! ? "Error closing sort pipe: $!"
883 : "Exit status $? from sort";
884 open(INPUT, 'foo') # get sort's results
885 or die "Can't open 'foo' for input: $!";
887 FILEHANDLE may be an expression whose value can be used as an indirect
888 filehandle, usually the real filehandle name.
890 =item closedir DIRHANDLE
893 Closes a directory opened by C<opendir> and returns the success of that
896 =item connect SOCKET,NAME
899 Attempts to connect to a remote socket, just as the connect system call
900 does. Returns true if it succeeded, false otherwise. NAME should be a
901 packed address of the appropriate type for the socket. See the examples in
902 L<perlipc/"Sockets: Client/Server Communication">.
909 C<continue> is actually a flow control statement rather than a function. If
910 there is a C<continue> BLOCK attached to a BLOCK (typically in a C<while> or
911 C<foreach>), it is always executed just before the conditional is about to
912 be evaluated again, just like the third part of a C<for> loop in C. Thus
913 it can be used to increment a loop variable, even when the loop has been
914 continued via the C<next> statement (which is similar to the C C<continue>
917 C<last>, C<next>, or C<redo> may appear within a C<continue>
918 block. C<last> and C<redo> will behave as if they had been executed within
919 the main block. So will C<next>, but since it will execute a C<continue>
920 block, it may be more entertaining.
923 ### redo always comes here
926 ### next always comes here
928 # then back the top to re-check EXPR
930 ### last always comes here
932 Omitting the C<continue> section is semantically equivalent to using an
933 empty one, logically enough. In that case, C<next> goes directly back
934 to check the condition at the top of the loop.
936 If the "switch" feature is enabled, C<continue> is also a
937 function that will break out of the current C<when> or C<default>
938 block, and fall through to the next case. See L<feature> and
939 L<perlsyn/"Switch statements"> for more information.
943 X<cos> X<cosine> X<acos> X<arccosine>
947 Returns the cosine of EXPR (expressed in radians). If EXPR is omitted,
948 takes cosine of C<$_>.
950 For the inverse cosine operation, you may use the C<Math::Trig::acos()>
951 function, or use this relation:
953 sub acos { atan2( sqrt(1 - $_[0] * $_[0]), $_[0] ) }
955 =item crypt PLAINTEXT,SALT
956 X<crypt> X<digest> X<hash> X<salt> X<plaintext> X<password>
957 X<decrypt> X<cryptography> X<passwd>
959 Creates a digest string exactly like the crypt(3) function in the C
960 library (assuming that you actually have a version there that has not
961 been extirpated as a potential munitions).
963 crypt() is a one-way hash function. The PLAINTEXT and SALT is turned
964 into a short string, called a digest, which is returned. The same
965 PLAINTEXT and SALT will always return the same string, but there is no
966 (known) way to get the original PLAINTEXT from the hash. Small
967 changes in the PLAINTEXT or SALT will result in large changes in the
970 There is no decrypt function. This function isn't all that useful for
971 cryptography (for that, look for F<Crypt> modules on your nearby CPAN
972 mirror) and the name "crypt" is a bit of a misnomer. Instead it is
973 primarily used to check if two pieces of text are the same without
974 having to transmit or store the text itself. An example is checking
975 if a correct password is given. The digest of the password is stored,
976 not the password itself. The user types in a password that is
977 crypt()'d with the same salt as the stored digest. If the two digests
978 match the password is correct.
980 When verifying an existing digest string you should use the digest as
981 the salt (like C<crypt($plain, $digest) eq $digest>). The SALT used
982 to create the digest is visible as part of the digest. This ensures
983 crypt() will hash the new string with the same salt as the digest.
984 This allows your code to work with the standard L<crypt|/crypt> and
985 with more exotic implementations. In other words, do not assume
986 anything about the returned string itself, or how many bytes in the
989 Traditionally the result is a string of 13 bytes: two first bytes of
990 the salt, followed by 11 bytes from the set C<[./0-9A-Za-z]>, and only
991 the first eight bytes of the digest string mattered, but alternative
992 hashing schemes (like MD5), higher level security schemes (like C2),
993 and implementations on non-UNIX platforms may produce different
996 When choosing a new salt create a random two character string whose
997 characters come from the set C<[./0-9A-Za-z]> (like C<join '', ('.',
998 '/', 0..9, 'A'..'Z', 'a'..'z')[rand 64, rand 64]>). This set of
999 characters is just a recommendation; the characters allowed in
1000 the salt depend solely on your system's crypt library, and Perl can't
1001 restrict what salts C<crypt()> accepts.
1003 Here's an example that makes sure that whoever runs this program knows
1006 $pwd = (getpwuid($<))[1];
1008 system "stty -echo";
1010 chomp($word = <STDIN>);
1014 if (crypt($word, $pwd) ne $pwd) {
1020 Of course, typing in your own password to whoever asks you
1023 The L<crypt|/crypt> function is unsuitable for hashing large quantities
1024 of data, not least of all because you can't get the information
1025 back. Look at the L<Digest> module for more robust algorithms.
1027 If using crypt() on a Unicode string (which I<potentially> has
1028 characters with codepoints above 255), Perl tries to make sense
1029 of the situation by trying to downgrade (a copy of the string)
1030 the string back to an eight-bit byte string before calling crypt()
1031 (on that copy). If that works, good. If not, crypt() dies with
1032 C<Wide character in crypt>.
1037 [This function has been largely superseded by the C<untie> function.]
1039 Breaks the binding between a DBM file and a hash.
1041 =item dbmopen HASH,DBNAME,MASK
1042 X<dbmopen> X<dbm> X<ndbm> X<sdbm> X<gdbm>
1044 [This function has been largely superseded by the C<tie> function.]
1046 This binds a dbm(3), ndbm(3), sdbm(3), gdbm(3), or Berkeley DB file to a
1047 hash. HASH is the name of the hash. (Unlike normal C<open>, the first
1048 argument is I<not> a filehandle, even though it looks like one). DBNAME
1049 is the name of the database (without the F<.dir> or F<.pag> extension if
1050 any). If the database does not exist, it is created with protection
1051 specified by MASK (as modified by the C<umask>). If your system supports
1052 only the older DBM functions, you may perform only one C<dbmopen> in your
1053 program. In older versions of Perl, if your system had neither DBM nor
1054 ndbm, calling C<dbmopen> produced a fatal error; it now falls back to
1057 If you don't have write access to the DBM file, you can only read hash
1058 variables, not set them. If you want to test whether you can write,
1059 either use file tests or try setting a dummy hash entry inside an C<eval>,
1060 which will trap the error.
1062 Note that functions such as C<keys> and C<values> may return huge lists
1063 when used on large DBM files. You may prefer to use the C<each>
1064 function to iterate over large DBM files. Example:
1066 # print out history file offsets
1067 dbmopen(%HIST,'/usr/lib/news/history',0666);
1068 while (($key,$val) = each %HIST) {
1069 print $key, ' = ', unpack('L',$val), "\n";
1073 See also L<AnyDBM_File> for a more general description of the pros and
1074 cons of the various dbm approaches, as well as L<DB_File> for a particularly
1075 rich implementation.
1077 You can control which DBM library you use by loading that library
1078 before you call dbmopen():
1081 dbmopen(%NS_Hist, "$ENV{HOME}/.netscape/history.db")
1082 or die "Can't open netscape history file: $!";
1085 X<defined> X<undef> X<undefined>
1089 Returns a Boolean value telling whether EXPR has a value other than
1090 the undefined value C<undef>. If EXPR is not present, C<$_> will be
1093 Many operations return C<undef> to indicate failure, end of file,
1094 system error, uninitialized variable, and other exceptional
1095 conditions. This function allows you to distinguish C<undef> from
1096 other values. (A simple Boolean test will not distinguish among
1097 C<undef>, zero, the empty string, and C<"0">, which are all equally
1098 false.) Note that since C<undef> is a valid scalar, its presence
1099 doesn't I<necessarily> indicate an exceptional condition: C<pop>
1100 returns C<undef> when its argument is an empty array, I<or> when the
1101 element to return happens to be C<undef>.
1103 You may also use C<defined(&func)> to check whether subroutine C<&func>
1104 has ever been defined. The return value is unaffected by any forward
1105 declarations of C<&func>. Note that a subroutine which is not defined
1106 may still be callable: its package may have an C<AUTOLOAD> method that
1107 makes it spring into existence the first time that it is called -- see
1110 Use of C<defined> on aggregates (hashes and arrays) is deprecated. It
1111 used to report whether memory for that aggregate has ever been
1112 allocated. This behavior may disappear in future versions of Perl.
1113 You should instead use a simple test for size:
1115 if (@an_array) { print "has array elements\n" }
1116 if (%a_hash) { print "has hash members\n" }
1118 When used on a hash element, it tells you whether the value is defined,
1119 not whether the key exists in the hash. Use L</exists> for the latter
1124 print if defined $switch{'D'};
1125 print "$val\n" while defined($val = pop(@ary));
1126 die "Can't readlink $sym: $!"
1127 unless defined($value = readlink $sym);
1128 sub foo { defined &$bar ? &$bar(@_) : die "No bar"; }
1129 $debugging = 0 unless defined $debugging;
1131 Note: Many folks tend to overuse C<defined>, and then are surprised to
1132 discover that the number C<0> and C<""> (the zero-length string) are, in fact,
1133 defined values. For example, if you say
1137 The pattern match succeeds, and C<$1> is defined, despite the fact that it
1138 matched "nothing". It didn't really fail to match anything. Rather, it
1139 matched something that happened to be zero characters long. This is all
1140 very above-board and honest. When a function returns an undefined value,
1141 it's an admission that it couldn't give you an honest answer. So you
1142 should use C<defined> only when you're questioning the integrity of what
1143 you're trying to do. At other times, a simple comparison to C<0> or C<""> is
1146 See also L</undef>, L</exists>, L</ref>.
1151 Given an expression that specifies a hash element, array element, hash slice,
1152 or array slice, deletes the specified element(s) from the hash or array.
1153 In the case of an array, if the array elements happen to be at the end,
1154 the size of the array will shrink to the highest element that tests
1155 true for exists() (or 0 if no such element exists).
1157 Returns a list with the same number of elements as the number of elements
1158 for which deletion was attempted. Each element of that list consists of
1159 either the value of the element deleted, or the undefined value. In scalar
1160 context, this means that you get the value of the last element deleted (or
1161 the undefined value if that element did not exist).
1163 %hash = (foo => 11, bar => 22, baz => 33);
1164 $scalar = delete $hash{foo}; # $scalar is 11
1165 $scalar = delete @hash{qw(foo bar)}; # $scalar is 22
1166 @array = delete @hash{qw(foo bar baz)}; # @array is (undef,undef,33)
1168 Deleting from C<%ENV> modifies the environment. Deleting from
1169 a hash tied to a DBM file deletes the entry from the DBM file. Deleting
1170 from a C<tie>d hash or array may not necessarily return anything.
1172 Deleting an array element effectively returns that position of the array
1173 to its initial, uninitialized state. Subsequently testing for the same
1174 element with exists() will return false. Also, deleting array elements
1175 in the middle of an array will not shift the index of the elements
1176 after them down. Use splice() for that. See L</exists>.
1178 The following (inefficiently) deletes all the values of %HASH and @ARRAY:
1180 foreach $key (keys %HASH) {
1184 foreach $index (0 .. $#ARRAY) {
1185 delete $ARRAY[$index];
1190 delete @HASH{keys %HASH};
1192 delete @ARRAY[0 .. $#ARRAY];
1194 But both of these are slower than just assigning the empty list
1195 or undefining %HASH or @ARRAY:
1197 %HASH = (); # completely empty %HASH
1198 undef %HASH; # forget %HASH ever existed
1200 @ARRAY = (); # completely empty @ARRAY
1201 undef @ARRAY; # forget @ARRAY ever existed
1203 Note that the EXPR can be arbitrarily complicated as long as the final
1204 operation is a hash element, array element, hash slice, or array slice
1207 delete $ref->[$x][$y]{$key};
1208 delete @{$ref->[$x][$y]}{$key1, $key2, @morekeys};
1210 delete $ref->[$x][$y][$index];
1211 delete @{$ref->[$x][$y]}[$index1, $index2, @moreindices];
1214 X<die> X<throw> X<exception> X<raise> X<$@> X<abort>
1216 Outside an C<eval>, prints the value of LIST to C<STDERR> and
1217 exits with the current value of C<$!> (errno). If C<$!> is C<0>,
1218 exits with the value of C<<< ($? >> 8) >>> (backtick `command`
1219 status). If C<<< ($? >> 8) >>> is C<0>, exits with C<255>. Inside
1220 an C<eval(),> the error message is stuffed into C<$@> and the
1221 C<eval> is terminated with the undefined value. This makes
1222 C<die> the way to raise an exception.
1224 Equivalent examples:
1226 die "Can't cd to spool: $!\n" unless chdir '/usr/spool/news';
1227 chdir '/usr/spool/news' or die "Can't cd to spool: $!\n"
1229 If the last element of LIST does not end in a newline, the current
1230 script line number and input line number (if any) are also printed,
1231 and a newline is supplied. Note that the "input line number" (also
1232 known as "chunk") is subject to whatever notion of "line" happens to
1233 be currently in effect, and is also available as the special variable
1234 C<$.>. See L<perlvar/"$/"> and L<perlvar/"$.">.
1236 Hint: sometimes appending C<", stopped"> to your message will cause it
1237 to make better sense when the string C<"at foo line 123"> is appended.
1238 Suppose you are running script "canasta".
1240 die "/etc/games is no good";
1241 die "/etc/games is no good, stopped";
1243 produce, respectively
1245 /etc/games is no good at canasta line 123.
1246 /etc/games is no good, stopped at canasta line 123.
1248 See also exit(), warn(), and the Carp module.
1250 If LIST is empty and C<$@> already contains a value (typically from a
1251 previous eval) that value is reused after appending C<"\t...propagated">.
1252 This is useful for propagating exceptions:
1255 die unless $@ =~ /Expected exception/;
1257 If LIST is empty and C<$@> contains an object reference that has a
1258 C<PROPAGATE> method, that method will be called with additional file
1259 and line number parameters. The return value replaces the value in
1260 C<$@>. i.e. as if C<< $@ = eval { $@->PROPAGATE(__FILE__, __LINE__) }; >>
1263 If C<$@> is empty then the string C<"Died"> is used.
1265 die() can also be called with a reference argument. If this happens to be
1266 trapped within an eval(), $@ contains the reference. This behavior permits
1267 a more elaborate exception handling implementation using objects that
1268 maintain arbitrary state about the nature of the exception. Such a scheme
1269 is sometimes preferable to matching particular string values of $@ using
1270 regular expressions. Here's an example:
1272 use Scalar::Util 'blessed';
1274 eval { ... ; die Some::Module::Exception->new( FOO => "bar" ) };
1276 if (blessed($@) && $@->isa("Some::Module::Exception")) {
1277 # handle Some::Module::Exception
1280 # handle all other possible exceptions
1284 Because perl will stringify uncaught exception messages before displaying
1285 them, you may want to overload stringification operations on such custom
1286 exception objects. See L<overload> for details about that.
1288 You can arrange for a callback to be run just before the C<die>
1289 does its deed, by setting the C<$SIG{__DIE__}> hook. The associated
1290 handler will be called with the error text and can change the error
1291 message, if it sees fit, by calling C<die> again. See
1292 L<perlvar/$SIG{expr}> for details on setting C<%SIG> entries, and
1293 L<"eval BLOCK"> for some examples. Although this feature was
1294 to be run only right before your program was to exit, this is not
1295 currently the case--the C<$SIG{__DIE__}> hook is currently called
1296 even inside eval()ed blocks/strings! If one wants the hook to do
1297 nothing in such situations, put
1301 as the first line of the handler (see L<perlvar/$^S>). Because
1302 this promotes strange action at a distance, this counterintuitive
1303 behavior may be fixed in a future release.
1308 Not really a function. Returns the value of the last command in the
1309 sequence of commands indicated by BLOCK. When modified by the C<while> or
1310 C<until> loop modifier, executes the BLOCK once before testing the loop
1311 condition. (On other statements the loop modifiers test the conditional
1314 C<do BLOCK> does I<not> count as a loop, so the loop control statements
1315 C<next>, C<last>, or C<redo> cannot be used to leave or restart the block.
1316 See L<perlsyn> for alternative strategies.
1318 =item do SUBROUTINE(LIST)
1321 This form of subroutine call is deprecated. See L<perlsub>.
1326 Uses the value of EXPR as a filename and executes the contents of the
1327 file as a Perl script.
1335 except that it's more efficient and concise, keeps track of the current
1336 filename for error messages, searches the @INC directories, and updates
1337 C<%INC> if the file is found. See L<perlvar/Predefined Names> for these
1338 variables. It also differs in that code evaluated with C<do FILENAME>
1339 cannot see lexicals in the enclosing scope; C<eval STRING> does. It's the
1340 same, however, in that it does reparse the file every time you call it,
1341 so you probably don't want to do this inside a loop.
1343 If C<do> cannot read the file, it returns undef and sets C<$!> to the
1344 error. If C<do> can read the file but cannot compile it, it
1345 returns undef and sets an error message in C<$@>. If the file is
1346 successfully compiled, C<do> returns the value of the last expression
1349 Note that inclusion of library modules is better done with the
1350 C<use> and C<require> operators, which also do automatic error checking
1351 and raise an exception if there's a problem.
1353 You might like to use C<do> to read in a program configuration
1354 file. Manual error checking can be done this way:
1356 # read in config files: system first, then user
1357 for $file ("/share/prog/defaults.rc",
1358 "$ENV{HOME}/.someprogrc")
1360 unless ($return = do $file) {
1361 warn "couldn't parse $file: $@" if $@;
1362 warn "couldn't do $file: $!" unless defined $return;
1363 warn "couldn't run $file" unless $return;
1368 X<dump> X<core> X<undump>
1372 This function causes an immediate core dump. See also the B<-u>
1373 command-line switch in L<perlrun>, which does the same thing.
1374 Primarily this is so that you can use the B<undump> program (not
1375 supplied) to turn your core dump into an executable binary after
1376 having initialized all your variables at the beginning of the
1377 program. When the new binary is executed it will begin by executing
1378 a C<goto LABEL> (with all the restrictions that C<goto> suffers).
1379 Think of it as a goto with an intervening core dump and reincarnation.
1380 If C<LABEL> is omitted, restarts the program from the top.
1382 B<WARNING>: Any files opened at the time of the dump will I<not>
1383 be open any more when the program is reincarnated, with possible
1384 resulting confusion on the part of Perl.
1386 This function is now largely obsolete, partly because it's very
1387 hard to convert a core file into an executable, and because the
1388 real compiler backends for generating portable bytecode and compilable
1389 C code have superseded it. That's why you should now invoke it as
1390 C<CORE::dump()>, if you don't want to be warned against a possible
1393 If you're looking to use L<dump> to speed up your program, consider
1394 generating bytecode or native C code as described in L<perlcc>. If
1395 you're just trying to accelerate a CGI script, consider using the
1396 C<mod_perl> extension to B<Apache>, or the CPAN module, CGI::Fast.
1397 You might also consider autoloading or selfloading, which at least
1398 make your program I<appear> to run faster.
1401 X<each> X<hash, iterator>
1403 When called in list context, returns a 2-element list consisting of the
1404 key and value for the next element of a hash, so that you can iterate over
1405 it. When called in scalar context, returns only the key for the next
1406 element in the hash.
1408 Entries are returned in an apparently random order. The actual random
1409 order is subject to change in future versions of perl, but it is
1410 guaranteed to be in the same order as either the C<keys> or C<values>
1411 function would produce on the same (unmodified) hash. Since Perl
1412 5.8.1 the ordering is different even between different runs of Perl
1413 for security reasons (see L<perlsec/"Algorithmic Complexity Attacks">).
1415 When the hash is entirely read, a null array is returned in list context
1416 (which when assigned produces a false (C<0>) value), and C<undef> in
1417 scalar context. The next call to C<each> after that will start iterating
1418 again. There is a single iterator for each hash, shared by all C<each>,
1419 C<keys>, and C<values> function calls in the program; it can be reset by
1420 reading all the elements from the hash, or by evaluating C<keys HASH> or
1421 C<values HASH>. If you add or delete elements of a hash while you're
1422 iterating over it, you may get entries skipped or duplicated, so
1423 don't. Exception: It is always safe to delete the item most recently
1424 returned by C<each()>, which means that the following code will work:
1426 while (($key, $value) = each %hash) {
1428 delete $hash{$key}; # This is safe
1431 The following prints out your environment like the printenv(1) program,
1432 only in a different order:
1434 while (($key,$value) = each %ENV) {
1435 print "$key=$value\n";
1438 See also C<keys>, C<values> and C<sort>.
1440 =item eof FILEHANDLE
1449 Returns 1 if the next read on FILEHANDLE will return end of file, or if
1450 FILEHANDLE is not open. FILEHANDLE may be an expression whose value
1451 gives the real filehandle. (Note that this function actually
1452 reads a character and then C<ungetc>s it, so isn't very useful in an
1453 interactive context.) Do not read from a terminal file (or call
1454 C<eof(FILEHANDLE)> on it) after end-of-file is reached. File types such
1455 as terminals may lose the end-of-file condition if you do.
1457 An C<eof> without an argument uses the last file read. Using C<eof()>
1458 with empty parentheses is very different. It refers to the pseudo file
1459 formed from the files listed on the command line and accessed via the
1460 C<< <> >> operator. Since C<< <> >> isn't explicitly opened,
1461 as a normal filehandle is, an C<eof()> before C<< <> >> has been
1462 used will cause C<@ARGV> to be examined to determine if input is
1463 available. Similarly, an C<eof()> after C<< <> >> has returned
1464 end-of-file will assume you are processing another C<@ARGV> list,
1465 and if you haven't set C<@ARGV>, will read input from C<STDIN>;
1466 see L<perlop/"I/O Operators">.
1468 In a C<< while (<>) >> loop, C<eof> or C<eof(ARGV)> can be used to
1469 detect the end of each file, C<eof()> will only detect the end of the
1470 last file. Examples:
1472 # reset line numbering on each input file
1474 next if /^\s*#/; # skip comments
1477 close ARGV if eof; # Not eof()!
1480 # insert dashes just before last line of last file
1482 if (eof()) { # check for end of last file
1483 print "--------------\n";
1486 last if eof(); # needed if we're reading from a terminal
1489 Practical hint: you almost never need to use C<eof> in Perl, because the
1490 input operators typically return C<undef> when they run out of data, or if
1494 X<eval> X<try> X<catch> X<evaluate> X<parse> X<execute>
1500 In the first form, the return value of EXPR is parsed and executed as if it
1501 were a little Perl program. The value of the expression (which is itself
1502 determined within scalar context) is first parsed, and if there weren't any
1503 errors, executed in the lexical context of the current Perl program, so
1504 that any variable settings or subroutine and format definitions remain
1505 afterwards. Note that the value is parsed every time the C<eval> executes.
1506 If EXPR is omitted, evaluates C<$_>. This form is typically used to
1507 delay parsing and subsequent execution of the text of EXPR until run time.
1509 In the second form, the code within the BLOCK is parsed only once--at the
1510 same time the code surrounding the C<eval> itself was parsed--and executed
1511 within the context of the current Perl program. This form is typically
1512 used to trap exceptions more efficiently than the first (see below), while
1513 also providing the benefit of checking the code within BLOCK at compile
1516 The final semicolon, if any, may be omitted from the value of EXPR or within
1519 In both forms, the value returned is the value of the last expression
1520 evaluated inside the mini-program; a return statement may be also used, just
1521 as with subroutines. The expression providing the return value is evaluated
1522 in void, scalar, or list context, depending on the context of the C<eval>
1523 itself. See L</wantarray> for more on how the evaluation context can be
1526 If there is a syntax error or runtime error, or a C<die> statement is
1527 executed, an undefined value is returned by C<eval>, and C<$@> is set to the
1528 error message. If there was no error, C<$@> is guaranteed to be a null
1529 string. Beware that using C<eval> neither silences perl from printing
1530 warnings to STDERR, nor does it stuff the text of warning messages into C<$@>.
1531 To do either of those, you have to use the C<$SIG{__WARN__}> facility, or
1532 turn off warnings inside the BLOCK or EXPR using S<C<no warnings 'all'>>.
1533 See L</warn>, L<perlvar>, L<warnings> and L<perllexwarn>.
1535 Note that, because C<eval> traps otherwise-fatal errors, it is useful for
1536 determining whether a particular feature (such as C<socket> or C<symlink>)
1537 is implemented. It is also Perl's exception trapping mechanism, where
1538 the die operator is used to raise exceptions.
1540 If the code to be executed doesn't vary, you may use the eval-BLOCK
1541 form to trap run-time errors without incurring the penalty of
1542 recompiling each time. The error, if any, is still returned in C<$@>.
1545 # make divide-by-zero nonfatal
1546 eval { $answer = $a / $b; }; warn $@ if $@;
1548 # same thing, but less efficient
1549 eval '$answer = $a / $b'; warn $@ if $@;
1551 # a compile-time error
1552 eval { $answer = }; # WRONG
1555 eval '$answer ='; # sets $@
1557 Using the C<eval{}> form as an exception trap in libraries does have some
1558 issues. Due to the current arguably broken state of C<__DIE__> hooks, you
1559 may wish not to trigger any C<__DIE__> hooks that user code may have installed.
1560 You can use the C<local $SIG{__DIE__}> construct for this purpose,
1561 as shown in this example:
1563 # a very private exception trap for divide-by-zero
1564 eval { local $SIG{'__DIE__'}; $answer = $a / $b; };
1567 This is especially significant, given that C<__DIE__> hooks can call
1568 C<die> again, which has the effect of changing their error messages:
1570 # __DIE__ hooks may modify error messages
1572 local $SIG{'__DIE__'} =
1573 sub { (my $x = $_[0]) =~ s/foo/bar/g; die $x };
1574 eval { die "foo lives here" };
1575 print $@ if $@; # prints "bar lives here"
1578 Because this promotes action at a distance, this counterintuitive behavior
1579 may be fixed in a future release.
1581 With an C<eval>, you should be especially careful to remember what's
1582 being looked at when:
1588 eval { $x }; # CASE 4
1590 eval "\$$x++"; # CASE 5
1593 Cases 1 and 2 above behave identically: they run the code contained in
1594 the variable $x. (Although case 2 has misleading double quotes making
1595 the reader wonder what else might be happening (nothing is).) Cases 3
1596 and 4 likewise behave in the same way: they run the code C<'$x'>, which
1597 does nothing but return the value of $x. (Case 4 is preferred for
1598 purely visual reasons, but it also has the advantage of compiling at
1599 compile-time instead of at run-time.) Case 5 is a place where
1600 normally you I<would> like to use double quotes, except that in this
1601 particular situation, you can just use symbolic references instead, as
1604 C<eval BLOCK> does I<not> count as a loop, so the loop control statements
1605 C<next>, C<last>, or C<redo> cannot be used to leave or restart the block.
1607 Note that as a very special case, an C<eval ''> executed within the C<DB>
1608 package doesn't see the usual surrounding lexical scope, but rather the
1609 scope of the first non-DB piece of code that called it. You don't normally
1610 need to worry about this unless you are writing a Perl debugger.
1615 =item exec PROGRAM LIST
1617 The C<exec> function executes a system command I<and never returns>--
1618 use C<system> instead of C<exec> if you want it to return. It fails and
1619 returns false only if the command does not exist I<and> it is executed
1620 directly instead of via your system's command shell (see below).
1622 Since it's a common mistake to use C<exec> instead of C<system>, Perl
1623 warns you if there is a following statement which isn't C<die>, C<warn>,
1624 or C<exit> (if C<-w> is set - but you always do that). If you
1625 I<really> want to follow an C<exec> with some other statement, you
1626 can use one of these styles to avoid the warning:
1628 exec ('foo') or print STDERR "couldn't exec foo: $!";
1629 { exec ('foo') }; print STDERR "couldn't exec foo: $!";
1631 If there is more than one argument in LIST, or if LIST is an array
1632 with more than one value, calls execvp(3) with the arguments in LIST.
1633 If there is only one scalar argument or an array with one element in it,
1634 the argument is checked for shell metacharacters, and if there are any,
1635 the entire argument is passed to the system's command shell for parsing
1636 (this is C</bin/sh -c> on Unix platforms, but varies on other platforms).
1637 If there are no shell metacharacters in the argument, it is split into
1638 words and passed directly to C<execvp>, which is more efficient.
1641 exec '/bin/echo', 'Your arguments are: ', @ARGV;
1642 exec "sort $outfile | uniq";
1644 If you don't really want to execute the first argument, but want to lie
1645 to the program you are executing about its own name, you can specify
1646 the program you actually want to run as an "indirect object" (without a
1647 comma) in front of the LIST. (This always forces interpretation of the
1648 LIST as a multivalued list, even if there is only a single scalar in
1651 $shell = '/bin/csh';
1652 exec $shell '-sh'; # pretend it's a login shell
1656 exec {'/bin/csh'} '-sh'; # pretend it's a login shell
1658 When the arguments get executed via the system shell, results will
1659 be subject to its quirks and capabilities. See L<perlop/"`STRING`">
1662 Using an indirect object with C<exec> or C<system> is also more
1663 secure. This usage (which also works fine with system()) forces
1664 interpretation of the arguments as a multivalued list, even if the
1665 list had just one argument. That way you're safe from the shell
1666 expanding wildcards or splitting up words with whitespace in them.
1668 @args = ( "echo surprise" );
1670 exec @args; # subject to shell escapes
1672 exec { $args[0] } @args; # safe even with one-arg list
1674 The first version, the one without the indirect object, ran the I<echo>
1675 program, passing it C<"surprise"> an argument. The second version
1676 didn't--it tried to run a program literally called I<"echo surprise">,
1677 didn't find it, and set C<$?> to a non-zero value indicating failure.
1679 Beginning with v5.6.0, Perl will attempt to flush all files opened for
1680 output before the exec, but this may not be supported on some platforms
1681 (see L<perlport>). To be safe, you may need to set C<$|> ($AUTOFLUSH
1682 in English) or call the C<autoflush()> method of C<IO::Handle> on any
1683 open handles in order to avoid lost output.
1685 Note that C<exec> will not call your C<END> blocks, nor will it call
1686 any C<DESTROY> methods in your objects.
1689 X<exists> X<autovivification>
1691 Given an expression that specifies a hash element or array element,
1692 returns true if the specified element in the hash or array has ever
1693 been initialized, even if the corresponding value is undefined. The
1694 element is not autovivified if it doesn't exist.
1696 print "Exists\n" if exists $hash{$key};
1697 print "Defined\n" if defined $hash{$key};
1698 print "True\n" if $hash{$key};
1700 print "Exists\n" if exists $array[$index];
1701 print "Defined\n" if defined $array[$index];
1702 print "True\n" if $array[$index];
1704 A hash or array element can be true only if it's defined, and defined if
1705 it exists, but the reverse doesn't necessarily hold true.
1707 Given an expression that specifies the name of a subroutine,
1708 returns true if the specified subroutine has ever been declared, even
1709 if it is undefined. Mentioning a subroutine name for exists or defined
1710 does not count as declaring it. Note that a subroutine which does not
1711 exist may still be callable: its package may have an C<AUTOLOAD>
1712 method that makes it spring into existence the first time that it is
1713 called -- see L<perlsub>.
1715 print "Exists\n" if exists &subroutine;
1716 print "Defined\n" if defined &subroutine;
1718 Note that the EXPR can be arbitrarily complicated as long as the final
1719 operation is a hash or array key lookup or subroutine name:
1721 if (exists $ref->{A}->{B}->{$key}) { }
1722 if (exists $hash{A}{B}{$key}) { }
1724 if (exists $ref->{A}->{B}->[$ix]) { }
1725 if (exists $hash{A}{B}[$ix]) { }
1727 if (exists &{$ref->{A}{B}{$key}}) { }
1729 Although the deepest nested array or hash will not spring into existence
1730 just because its existence was tested, any intervening ones will.
1731 Thus C<< $ref->{"A"} >> and C<< $ref->{"A"}->{"B"} >> will spring
1732 into existence due to the existence test for the $key element above.
1733 This happens anywhere the arrow operator is used, including even:
1736 if (exists $ref->{"Some key"}) { }
1737 print $ref; # prints HASH(0x80d3d5c)
1739 This surprising autovivification in what does not at first--or even
1740 second--glance appear to be an lvalue context may be fixed in a future
1743 Use of a subroutine call, rather than a subroutine name, as an argument
1744 to exists() is an error.
1747 exists &sub(); # Error
1750 X<exit> X<terminate> X<abort>
1754 Evaluates EXPR and exits immediately with that value. Example:
1757 exit 0 if $ans =~ /^[Xx]/;
1759 See also C<die>. If EXPR is omitted, exits with C<0> status. The only
1760 universally recognized values for EXPR are C<0> for success and C<1>
1761 for error; other values are subject to interpretation depending on the
1762 environment in which the Perl program is running. For example, exiting
1763 69 (EX_UNAVAILABLE) from a I<sendmail> incoming-mail filter will cause
1764 the mailer to return the item undelivered, but that's not true everywhere.
1766 Don't use C<exit> to abort a subroutine if there's any chance that
1767 someone might want to trap whatever error happened. Use C<die> instead,
1768 which can be trapped by an C<eval>.
1770 The exit() function does not always exit immediately. It calls any
1771 defined C<END> routines first, but these C<END> routines may not
1772 themselves abort the exit. Likewise any object destructors that need to
1773 be called are called before the real exit. If this is a problem, you
1774 can call C<POSIX:_exit($status)> to avoid END and destructor processing.
1775 See L<perlmod> for details.
1778 X<exp> X<exponential> X<antilog> X<antilogarithm> X<e>
1782 Returns I<e> (the natural logarithm base) to the power of EXPR.
1783 If EXPR is omitted, gives C<exp($_)>.
1785 =item fcntl FILEHANDLE,FUNCTION,SCALAR
1788 Implements the fcntl(2) function. You'll probably have to say
1792 first to get the correct constant definitions. Argument processing and
1793 value return works just like C<ioctl> below.
1797 fcntl($filehandle, F_GETFL, $packed_return_buffer)
1798 or die "can't fcntl F_GETFL: $!";
1800 You don't have to check for C<defined> on the return from C<fcntl>.
1801 Like C<ioctl>, it maps a C<0> return from the system call into
1802 C<"0 but true"> in Perl. This string is true in boolean context and C<0>
1803 in numeric context. It is also exempt from the normal B<-w> warnings
1804 on improper numeric conversions.
1806 Note that C<fcntl> will produce a fatal error if used on a machine that
1807 doesn't implement fcntl(2). See the Fcntl module or your fcntl(2)
1808 manpage to learn what functions are available on your system.
1810 Here's an example of setting a filehandle named C<REMOTE> to be
1811 non-blocking at the system level. You'll have to negotiate C<$|>
1812 on your own, though.
1814 use Fcntl qw(F_GETFL F_SETFL O_NONBLOCK);
1816 $flags = fcntl(REMOTE, F_GETFL, 0)
1817 or die "Can't get flags for the socket: $!\n";
1819 $flags = fcntl(REMOTE, F_SETFL, $flags | O_NONBLOCK)
1820 or die "Can't set flags for the socket: $!\n";
1822 =item fileno FILEHANDLE
1825 Returns the file descriptor for a filehandle, or undefined if the
1826 filehandle is not open. This is mainly useful for constructing
1827 bitmaps for C<select> and low-level POSIX tty-handling operations.
1828 If FILEHANDLE is an expression, the value is taken as an indirect
1829 filehandle, generally its name.
1831 You can use this to find out whether two handles refer to the
1832 same underlying descriptor:
1834 if (fileno(THIS) == fileno(THAT)) {
1835 print "THIS and THAT are dups\n";
1838 (Filehandles connected to memory objects via new features of C<open> may
1839 return undefined even though they are open.)
1842 =item flock FILEHANDLE,OPERATION
1843 X<flock> X<lock> X<locking>
1845 Calls flock(2), or an emulation of it, on FILEHANDLE. Returns true
1846 for success, false on failure. Produces a fatal error if used on a
1847 machine that doesn't implement flock(2), fcntl(2) locking, or lockf(3).
1848 C<flock> is Perl's portable file locking interface, although it locks
1849 only entire files, not records.
1851 Two potentially non-obvious but traditional C<flock> semantics are
1852 that it waits indefinitely until the lock is granted, and that its locks
1853 B<merely advisory>. Such discretionary locks are more flexible, but offer
1854 fewer guarantees. This means that programs that do not also use C<flock>
1855 may modify files locked with C<flock>. See L<perlport>,
1856 your port's specific documentation, or your system-specific local manpages
1857 for details. It's best to assume traditional behavior if you're writing
1858 portable programs. (But if you're not, you should as always feel perfectly
1859 free to write for your own system's idiosyncrasies (sometimes called
1860 "features"). Slavish adherence to portability concerns shouldn't get
1861 in the way of your getting your job done.)
1863 OPERATION is one of LOCK_SH, LOCK_EX, or LOCK_UN, possibly combined with
1864 LOCK_NB. These constants are traditionally valued 1, 2, 8 and 4, but
1865 you can use the symbolic names if you import them from the Fcntl module,
1866 either individually, or as a group using the ':flock' tag. LOCK_SH
1867 requests a shared lock, LOCK_EX requests an exclusive lock, and LOCK_UN
1868 releases a previously requested lock. If LOCK_NB is bitwise-or'ed with
1869 LOCK_SH or LOCK_EX then C<flock> will return immediately rather than blocking
1870 waiting for the lock (check the return status to see if you got it).
1872 To avoid the possibility of miscoordination, Perl now flushes FILEHANDLE
1873 before locking or unlocking it.
1875 Note that the emulation built with lockf(3) doesn't provide shared
1876 locks, and it requires that FILEHANDLE be open with write intent. These
1877 are the semantics that lockf(3) implements. Most if not all systems
1878 implement lockf(3) in terms of fcntl(2) locking, though, so the
1879 differing semantics shouldn't bite too many people.
1881 Note that the fcntl(2) emulation of flock(3) requires that FILEHANDLE
1882 be open with read intent to use LOCK_SH and requires that it be open
1883 with write intent to use LOCK_EX.
1885 Note also that some versions of C<flock> cannot lock things over the
1886 network; you would need to use the more system-specific C<fcntl> for
1887 that. If you like you can force Perl to ignore your system's flock(2)
1888 function, and so provide its own fcntl(2)-based emulation, by passing
1889 the switch C<-Ud_flock> to the F<Configure> program when you configure
1892 Here's a mailbox appender for BSD systems.
1894 use Fcntl ':flock'; # import LOCK_* constants
1897 flock(MBOX,LOCK_EX);
1898 # and, in case someone appended
1899 # while we were waiting...
1904 flock(MBOX,LOCK_UN);
1907 open(MBOX, ">>/usr/spool/mail/$ENV{'USER'}")
1908 or die "Can't open mailbox: $!";
1911 print MBOX $msg,"\n\n";
1914 On systems that support a real flock(), locks are inherited across fork()
1915 calls, whereas those that must resort to the more capricious fcntl()
1916 function lose the locks, making it harder to write servers.
1918 See also L<DB_File> for other flock() examples.
1921 X<fork> X<child> X<parent>
1923 Does a fork(2) system call to create a new process running the
1924 same program at the same point. It returns the child pid to the
1925 parent process, C<0> to the child process, or C<undef> if the fork is
1926 unsuccessful. File descriptors (and sometimes locks on those descriptors)
1927 are shared, while everything else is copied. On most systems supporting
1928 fork(), great care has gone into making it extremely efficient (for
1929 example, using copy-on-write technology on data pages), making it the
1930 dominant paradigm for multitasking over the last few decades.
1932 Beginning with v5.6.0, Perl will attempt to flush all files opened for
1933 output before forking the child process, but this may not be supported
1934 on some platforms (see L<perlport>). To be safe, you may need to set
1935 C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method of
1936 C<IO::Handle> on any open handles in order to avoid duplicate output.
1938 If you C<fork> without ever waiting on your children, you will
1939 accumulate zombies. On some systems, you can avoid this by setting
1940 C<$SIG{CHLD}> to C<"IGNORE">. See also L<perlipc> for more examples of
1941 forking and reaping moribund children.
1943 Note that if your forked child inherits system file descriptors like
1944 STDIN and STDOUT that are actually connected by a pipe or socket, even
1945 if you exit, then the remote server (such as, say, a CGI script or a
1946 backgrounded job launched from a remote shell) won't think you're done.
1947 You should reopen those to F</dev/null> if it's any issue.
1952 Declare a picture format for use by the C<write> function. For
1956 Test: @<<<<<<<< @||||| @>>>>>
1957 $str, $%, '$' . int($num)
1961 $num = $cost/$quantity;
1965 See L<perlform> for many details and examples.
1967 =item formline PICTURE,LIST
1970 This is an internal function used by C<format>s, though you may call it,
1971 too. It formats (see L<perlform>) a list of values according to the
1972 contents of PICTURE, placing the output into the format output
1973 accumulator, C<$^A> (or C<$ACCUMULATOR> in English).
1974 Eventually, when a C<write> is done, the contents of
1975 C<$^A> are written to some filehandle. You could also read C<$^A>
1976 and then set C<$^A> back to C<"">. Note that a format typically
1977 does one C<formline> per line of form, but the C<formline> function itself
1978 doesn't care how many newlines are embedded in the PICTURE. This means
1979 that the C<~> and C<~~> tokens will treat the entire PICTURE as a single line.
1980 You may therefore need to use multiple formlines to implement a single
1981 record format, just like the format compiler.
1983 Be careful if you put double quotes around the picture, because an C<@>
1984 character may be taken to mean the beginning of an array name.
1985 C<formline> always returns true. See L<perlform> for other examples.
1987 =item getc FILEHANDLE
1992 Returns the next character from the input file attached to FILEHANDLE,
1993 or the undefined value at end of file, or if there was an error (in
1994 the latter case C<$!> is set). If FILEHANDLE is omitted, reads from
1995 STDIN. This is not particularly efficient. However, it cannot be
1996 used by itself to fetch single characters without waiting for the user
1997 to hit enter. For that, try something more like:
2000 system "stty cbreak </dev/tty >/dev/tty 2>&1";
2003 system "stty", '-icanon', 'eol', "\001";
2009 system "stty -cbreak </dev/tty >/dev/tty 2>&1";
2012 system "stty", 'icanon', 'eol', '^@'; # ASCII null
2016 Determination of whether $BSD_STYLE should be set
2017 is left as an exercise to the reader.
2019 The C<POSIX::getattr> function can do this more portably on
2020 systems purporting POSIX compliance. See also the C<Term::ReadKey>
2021 module from your nearest CPAN site; details on CPAN can be found on
2025 X<getlogin> X<login>
2027 This implements the C library function of the same name, which on most
2028 systems returns the current login from F</etc/utmp>, if any. If null,
2031 $login = getlogin || getpwuid($<) || "Kilroy";
2033 Do not consider C<getlogin> for authentication: it is not as
2034 secure as C<getpwuid>.
2036 =item getpeername SOCKET
2037 X<getpeername> X<peer>
2039 Returns the packed sockaddr address of other end of the SOCKET connection.
2042 $hersockaddr = getpeername(SOCK);
2043 ($port, $iaddr) = sockaddr_in($hersockaddr);
2044 $herhostname = gethostbyaddr($iaddr, AF_INET);
2045 $herstraddr = inet_ntoa($iaddr);
2050 Returns the current process group for the specified PID. Use
2051 a PID of C<0> to get the current process group for the
2052 current process. Will raise an exception if used on a machine that
2053 doesn't implement getpgrp(2). If PID is omitted, returns process
2054 group of current process. Note that the POSIX version of C<getpgrp>
2055 does not accept a PID argument, so only C<PID==0> is truly portable.
2058 X<getppid> X<parent> X<pid>
2060 Returns the process id of the parent process.
2062 Note for Linux users: on Linux, the C functions C<getpid()> and
2063 C<getppid()> return different values from different threads. In order to
2064 be portable, this behavior is not reflected by the perl-level function
2065 C<getppid()>, that returns a consistent value across threads. If you want
2066 to call the underlying C<getppid()>, you may use the CPAN module
2069 =item getpriority WHICH,WHO
2070 X<getpriority> X<priority> X<nice>
2072 Returns the current priority for a process, a process group, or a user.
2073 (See L<getpriority(2)>.) Will raise a fatal exception if used on a
2074 machine that doesn't implement getpriority(2).
2077 X<getpwnam> X<getgrnam> X<gethostbyname> X<getnetbyname> X<getprotobyname>
2078 X<getpwuid> X<getgrgid> X<getservbyname> X<gethostbyaddr> X<getnetbyaddr>
2079 X<getprotobynumber> X<getservbyport> X<getpwent> X<getgrent> X<gethostent>
2080 X<getnetent> X<getprotoent> X<getservent> X<setpwent> X<setgrent> X<sethostent>
2081 X<setnetent> X<setprotoent> X<setservent> X<endpwent> X<endgrent> X<endhostent>
2082 X<endnetent> X<endprotoent> X<endservent>
2086 =item gethostbyname NAME
2088 =item getnetbyname NAME
2090 =item getprotobyname NAME
2096 =item getservbyname NAME,PROTO
2098 =item gethostbyaddr ADDR,ADDRTYPE
2100 =item getnetbyaddr ADDR,ADDRTYPE
2102 =item getprotobynumber NUMBER
2104 =item getservbyport PORT,PROTO
2122 =item sethostent STAYOPEN
2124 =item setnetent STAYOPEN
2126 =item setprotoent STAYOPEN
2128 =item setservent STAYOPEN
2142 These routines perform the same functions as their counterparts in the
2143 system library. In list context, the return values from the
2144 various get routines are as follows:
2146 ($name,$passwd,$uid,$gid,
2147 $quota,$comment,$gcos,$dir,$shell,$expire) = getpw*
2148 ($name,$passwd,$gid,$members) = getgr*
2149 ($name,$aliases,$addrtype,$length,@addrs) = gethost*
2150 ($name,$aliases,$addrtype,$net) = getnet*
2151 ($name,$aliases,$proto) = getproto*
2152 ($name,$aliases,$port,$proto) = getserv*
2154 (If the entry doesn't exist you get a null list.)
2156 The exact meaning of the $gcos field varies but it usually contains
2157 the real name of the user (as opposed to the login name) and other
2158 information pertaining to the user. Beware, however, that in many
2159 system users are able to change this information and therefore it
2160 cannot be trusted and therefore the $gcos is tainted (see
2161 L<perlsec>). The $passwd and $shell, user's encrypted password and
2162 login shell, are also tainted, because of the same reason.
2164 In scalar context, you get the name, unless the function was a
2165 lookup by name, in which case you get the other thing, whatever it is.
2166 (If the entry doesn't exist you get the undefined value.) For example:
2168 $uid = getpwnam($name);
2169 $name = getpwuid($num);
2171 $gid = getgrnam($name);
2172 $name = getgrgid($num);
2176 In I<getpw*()> the fields $quota, $comment, and $expire are special
2177 cases in the sense that in many systems they are unsupported. If the
2178 $quota is unsupported, it is an empty scalar. If it is supported, it
2179 usually encodes the disk quota. If the $comment field is unsupported,
2180 it is an empty scalar. If it is supported it usually encodes some
2181 administrative comment about the user. In some systems the $quota
2182 field may be $change or $age, fields that have to do with password
2183 aging. In some systems the $comment field may be $class. The $expire
2184 field, if present, encodes the expiration period of the account or the
2185 password. For the availability and the exact meaning of these fields
2186 in your system, please consult your getpwnam(3) documentation and your
2187 F<pwd.h> file. You can also find out from within Perl what your
2188 $quota and $comment fields mean and whether you have the $expire field
2189 by using the C<Config> module and the values C<d_pwquota>, C<d_pwage>,
2190 C<d_pwchange>, C<d_pwcomment>, and C<d_pwexpire>. Shadow password
2191 files are only supported if your vendor has implemented them in the
2192 intuitive fashion that calling the regular C library routines gets the
2193 shadow versions if you're running under privilege or if there exists
2194 the shadow(3) functions as found in System V (this includes Solaris
2195 and Linux.) Those systems that implement a proprietary shadow password
2196 facility are unlikely to be supported.
2198 The $members value returned by I<getgr*()> is a space separated list of
2199 the login names of the members of the group.
2201 For the I<gethost*()> functions, if the C<h_errno> variable is supported in
2202 C, it will be returned to you via C<$?> if the function call fails. The
2203 C<@addrs> value returned by a successful call is a list of the raw
2204 addresses returned by the corresponding system library call. In the
2205 Internet domain, each address is four bytes long and you can unpack it
2206 by saying something like:
2208 ($a,$b,$c,$d) = unpack('W4',$addr[0]);
2210 The Socket library makes this slightly easier:
2213 $iaddr = inet_aton("127.1"); # or whatever address
2214 $name = gethostbyaddr($iaddr, AF_INET);
2216 # or going the other way
2217 $straddr = inet_ntoa($iaddr);
2219 If you get tired of remembering which element of the return list
2220 contains which return value, by-name interfaces are provided
2221 in standard modules: C<File::stat>, C<Net::hostent>, C<Net::netent>,
2222 C<Net::protoent>, C<Net::servent>, C<Time::gmtime>, C<Time::localtime>,
2223 and C<User::grent>. These override the normal built-ins, supplying
2224 versions that return objects with the appropriate names
2225 for each field. For example:
2229 $is_his = (stat($filename)->uid == pwent($whoever)->uid);
2231 Even though it looks like they're the same method calls (uid),
2232 they aren't, because a C<File::stat> object is different from
2233 a C<User::pwent> object.
2235 =item getsockname SOCKET
2238 Returns the packed sockaddr address of this end of the SOCKET connection,
2239 in case you don't know the address because you have several different
2240 IPs that the connection might have come in on.
2243 $mysockaddr = getsockname(SOCK);
2244 ($port, $myaddr) = sockaddr_in($mysockaddr);
2245 printf "Connect to %s [%s]\n",
2246 scalar gethostbyaddr($myaddr, AF_INET),
2249 =item getsockopt SOCKET,LEVEL,OPTNAME
2252 Queries the option named OPTNAME associated with SOCKET at a given LEVEL.
2253 Options may exist at multiple protocol levels depending on the socket
2254 type, but at least the uppermost socket level SOL_SOCKET (defined in the
2255 C<Socket> module) will exist. To query options at another level the
2256 protocol number of the appropriate protocol controlling the option
2257 should be supplied. For example, to indicate that an option is to be
2258 interpreted by the TCP protocol, LEVEL should be set to the protocol
2259 number of TCP, which you can get using getprotobyname.
2261 The call returns a packed string representing the requested socket option,
2262 or C<undef> if there is an error (the error reason will be in $!). What
2263 exactly is in the packed string depends in the LEVEL and OPTNAME, consult
2264 your system documentation for details. A very common case however is that
2265 the option is an integer, in which case the result will be a packed
2266 integer which you can decode using unpack with the C<i> (or C<I>) format.
2268 An example testing if Nagle's algorithm is turned on on a socket:
2270 use Socket qw(:all);
2272 defined(my $tcp = getprotobyname("tcp"))
2273 or die "Could not determine the protocol number for tcp";
2274 # my $tcp = IPPROTO_TCP; # Alternative
2275 my $packed = getsockopt($socket, $tcp, TCP_NODELAY)
2276 or die "Could not query TCP_NODELAY socket option: $!";
2277 my $nodelay = unpack("I", $packed);
2278 print "Nagle's algorithm is turned ", $nodelay ? "off\n" : "on\n";
2282 X<glob> X<wildcard> X<filename, expansion> X<expand>
2286 In list context, returns a (possibly empty) list of filename expansions on
2287 the value of EXPR such as the standard Unix shell F</bin/csh> would do. In
2288 scalar context, glob iterates through such filename expansions, returning
2289 undef when the list is exhausted. This is the internal function
2290 implementing the C<< <*.c> >> operator, but you can use it directly. If
2291 EXPR is omitted, C<$_> is used. The C<< <*.c> >> operator is discussed in
2292 more detail in L<perlop/"I/O Operators">.
2294 Beginning with v5.6.0, this operator is implemented using the standard
2295 C<File::Glob> extension. See L<File::Glob> for details.
2298 X<gmtime> X<UTC> X<Greenwich>
2302 Converts a time as returned by the time function to an 9-element list
2303 with the time localized for the standard Greenwich time zone.
2304 Typically used as follows:
2307 ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) =
2310 All list elements are numeric, and come straight out of the C `struct
2311 tm'. $sec, $min, and $hour are the seconds, minutes, and hours of the
2312 specified time. $mday is the day of the month, and $mon is the month
2313 itself, in the range C<0..11> with 0 indicating January and 11
2314 indicating December. $year is the number of years since 1900. That
2315 is, $year is C<123> in year 2023. $wday is the day of the week, with
2316 0 indicating Sunday and 3 indicating Wednesday. $yday is the day of
2317 the year, in the range C<0..364> (or C<0..365> in leap years). $isdst
2320 Note that the $year element is I<not> simply the last two digits of
2321 the year. If you assume it is then you create non-Y2K-compliant
2322 programs--and you wouldn't want to do that, would you?
2324 The proper way to get a complete 4-digit year is simply:
2328 And to get the last two digits of the year (e.g., '01' in 2001) do:
2330 $year = sprintf("%02d", $year % 100);
2332 If EXPR is omitted, C<gmtime()> uses the current time (C<gmtime(time)>).
2334 In scalar context, C<gmtime()> returns the ctime(3) value:
2336 $now_string = gmtime; # e.g., "Thu Oct 13 04:54:34 1994"
2338 If you need local time instead of GMT use the L</localtime> builtin.
2339 See also the C<timegm> function provided by the C<Time::Local> module,
2340 and the strftime(3) and mktime(3) functions available via the L<POSIX> module.
2342 This scalar value is B<not> locale dependent (see L<perllocale>), but is
2343 instead a Perl builtin. To get somewhat similar but locale dependent date
2344 strings, see the example in L</localtime>.
2346 See L<perlport/gmtime> for portability concerns.
2349 X<goto> X<jump> X<jmp>
2355 The C<goto-LABEL> form finds the statement labeled with LABEL and resumes
2356 execution there. It may not be used to go into any construct that
2357 requires initialization, such as a subroutine or a C<foreach> loop. It
2358 also can't be used to go into a construct that is optimized away,
2359 or to get out of a block or subroutine given to C<sort>.
2360 It can be used to go almost anywhere else within the dynamic scope,
2361 including out of subroutines, but it's usually better to use some other
2362 construct such as C<last> or C<die>. The author of Perl has never felt the
2363 need to use this form of C<goto> (in Perl, that is--C is another matter).
2364 (The difference being that C does not offer named loops combined with
2365 loop control. Perl does, and this replaces most structured uses of C<goto>
2366 in other languages.)
2368 The C<goto-EXPR> form expects a label name, whose scope will be resolved
2369 dynamically. This allows for computed C<goto>s per FORTRAN, but isn't
2370 necessarily recommended if you're optimizing for maintainability:
2372 goto ("FOO", "BAR", "GLARCH")[$i];
2374 The C<goto-&NAME> form is quite different from the other forms of
2375 C<goto>. In fact, it isn't a goto in the normal sense at all, and
2376 doesn't have the stigma associated with other gotos. Instead, it
2377 exits the current subroutine (losing any changes set by local()) and
2378 immediately calls in its place the named subroutine using the current
2379 value of @_. This is used by C<AUTOLOAD> subroutines that wish to
2380 load another subroutine and then pretend that the other subroutine had
2381 been called in the first place (except that any modifications to C<@_>
2382 in the current subroutine are propagated to the other subroutine.)
2383 After the C<goto>, not even C<caller> will be able to tell that this
2384 routine was called first.
2386 NAME needn't be the name of a subroutine; it can be a scalar variable
2387 containing a code reference, or a block that evaluates to a code
2390 =item grep BLOCK LIST
2393 =item grep EXPR,LIST
2395 This is similar in spirit to, but not the same as, grep(1) and its
2396 relatives. In particular, it is not limited to using regular expressions.
2398 Evaluates the BLOCK or EXPR for each element of LIST (locally setting
2399 C<$_> to each element) and returns the list value consisting of those
2400 elements for which the expression evaluated to true. In scalar
2401 context, returns the number of times the expression was true.
2403 @foo = grep(!/^#/, @bar); # weed out comments
2407 @foo = grep {!/^#/} @bar; # weed out comments
2409 Note that C<$_> is an alias to the list value, so it can be used to
2410 modify the elements of the LIST. While this is useful and supported,
2411 it can cause bizarre results if the elements of LIST are not variables.
2412 Similarly, grep returns aliases into the original list, much as a for
2413 loop's index variable aliases the list elements. That is, modifying an
2414 element of a list returned by grep (for example, in a C<foreach>, C<map>
2415 or another C<grep>) actually modifies the element in the original list.
2416 This is usually something to be avoided when writing clear code.
2418 If C<$_> is lexical in the scope where the C<grep> appears (because it has
2419 been declared with C<my $_>) then, in addition to being locally aliased to
2420 the list elements, C<$_> keeps being lexical inside the block; i.e. it
2421 can't be seen from the outside, avoiding any potential side-effects.
2423 See also L</map> for a list composed of the results of the BLOCK or EXPR.
2426 X<hex> X<hexadecimal>
2430 Interprets EXPR as a hex string and returns the corresponding value.
2431 (To convert strings that might start with either C<0>, C<0x>, or C<0b>, see
2432 L</oct>.) If EXPR is omitted, uses C<$_>.
2434 print hex '0xAf'; # prints '175'
2435 print hex 'aF'; # same
2437 Hex strings may only represent integers. Strings that would cause
2438 integer overflow trigger a warning. Leading whitespace is not stripped,
2439 unlike oct(). To present something as hex, look into L</printf>,
2440 L</sprintf>, or L</unpack>.
2445 There is no builtin C<import> function. It is just an ordinary
2446 method (subroutine) defined (or inherited) by modules that wish to export
2447 names to another module. The C<use> function calls the C<import> method
2448 for the package used. See also L</use>, L<perlmod>, and L<Exporter>.
2450 =item index STR,SUBSTR,POSITION
2451 X<index> X<indexOf> X<InStr>
2453 =item index STR,SUBSTR
2455 The index function searches for one string within another, but without
2456 the wildcard-like behavior of a full regular-expression pattern match.
2457 It returns the position of the first occurrence of SUBSTR in STR at
2458 or after POSITION. If POSITION is omitted, starts searching from the
2459 beginning of the string. POSITION before the beginning of the string
2460 or after its end is treated as if it were the beginning or the end,
2461 respectively. POSITION and the return value are based at C<0> (or whatever
2462 you've set the C<$[> variable to--but don't do that). If the substring
2463 is not found, C<index> returns one less than the base, ordinarily C<-1>.
2466 X<int> X<integer> X<truncate> X<trunc>
2470 Returns the integer portion of EXPR. If EXPR is omitted, uses C<$_>.
2471 You should not use this function for rounding: one because it truncates
2472 towards C<0>, and two because machine representations of floating point
2473 numbers can sometimes produce counterintuitive results. For example,
2474 C<int(-6.725/0.025)> produces -268 rather than the correct -269; that's
2475 because it's really more like -268.99999999999994315658 instead. Usually,
2476 the C<sprintf>, C<printf>, or the C<POSIX::floor> and C<POSIX::ceil>
2477 functions will serve you better than will int().
2479 =item ioctl FILEHANDLE,FUNCTION,SCALAR
2482 Implements the ioctl(2) function. You'll probably first have to say
2484 require "sys/ioctl.ph"; # probably in $Config{archlib}/sys/ioctl.ph
2486 to get the correct function definitions. If F<sys/ioctl.ph> doesn't
2487 exist or doesn't have the correct definitions you'll have to roll your
2488 own, based on your C header files such as F<< <sys/ioctl.h> >>.
2489 (There is a Perl script called B<h2ph> that comes with the Perl kit that
2490 may help you in this, but it's nontrivial.) SCALAR will be read and/or
2491 written depending on the FUNCTION--a pointer to the string value of SCALAR
2492 will be passed as the third argument of the actual C<ioctl> call. (If SCALAR
2493 has no string value but does have a numeric value, that value will be
2494 passed rather than a pointer to the string value. To guarantee this to be
2495 true, add a C<0> to the scalar before using it.) The C<pack> and C<unpack>
2496 functions may be needed to manipulate the values of structures used by
2499 The return value of C<ioctl> (and C<fcntl>) is as follows:
2501 if OS returns: then Perl returns:
2503 0 string "0 but true"
2504 anything else that number
2506 Thus Perl returns true on success and false on failure, yet you can
2507 still easily determine the actual value returned by the operating
2510 $retval = ioctl(...) || -1;
2511 printf "System returned %d\n", $retval;
2513 The special string C<"0 but true"> is exempt from B<-w> complaints
2514 about improper numeric conversions.
2516 =item join EXPR,LIST
2519 Joins the separate strings of LIST into a single string with fields
2520 separated by the value of EXPR, and returns that new string. Example:
2522 $rec = join(':', $login,$passwd,$uid,$gid,$gcos,$home,$shell);
2524 Beware that unlike C<split>, C<join> doesn't take a pattern as its
2525 first argument. Compare L</split>.
2530 Returns a list consisting of all the keys of the named hash.
2531 (In scalar context, returns the number of keys.)
2533 The keys are returned in an apparently random order. The actual
2534 random order is subject to change in future versions of perl, but it
2535 is guaranteed to be the same order as either the C<values> or C<each>
2536 function produces (given that the hash has not been modified). Since
2537 Perl 5.8.1 the ordering is different even between different runs of
2538 Perl for security reasons (see L<perlsec/"Algorithmic Complexity
2541 As a side effect, calling keys() resets the HASH's internal iterator
2542 (see L</each>). In particular, calling keys() in void context resets
2543 the iterator with no other overhead.
2545 Here is yet another way to print your environment:
2548 @values = values %ENV;
2550 print pop(@keys), '=', pop(@values), "\n";
2553 or how about sorted by key:
2555 foreach $key (sort(keys %ENV)) {
2556 print $key, '=', $ENV{$key}, "\n";
2559 The returned values are copies of the original keys in the hash, so
2560 modifying them will not affect the original hash. Compare L</values>.
2562 To sort a hash by value, you'll need to use a C<sort> function.
2563 Here's a descending numeric sort of a hash by its values:
2565 foreach $key (sort { $hash{$b} <=> $hash{$a} } keys %hash) {
2566 printf "%4d %s\n", $hash{$key}, $key;
2569 As an lvalue C<keys> allows you to increase the number of hash buckets
2570 allocated for the given hash. This can gain you a measure of efficiency if
2571 you know the hash is going to get big. (This is similar to pre-extending
2572 an array by assigning a larger number to $#array.) If you say
2576 then C<%hash> will have at least 200 buckets allocated for it--256 of them,
2577 in fact, since it rounds up to the next power of two. These
2578 buckets will be retained even if you do C<%hash = ()>, use C<undef
2579 %hash> if you want to free the storage while C<%hash> is still in scope.
2580 You can't shrink the number of buckets allocated for the hash using
2581 C<keys> in this way (but you needn't worry about doing this by accident,
2582 as trying has no effect).
2584 See also C<each>, C<values> and C<sort>.
2586 =item kill SIGNAL, LIST
2589 Sends a signal to a list of processes. Returns the number of
2590 processes successfully signaled (which is not necessarily the
2591 same as the number actually killed).
2593 $cnt = kill 1, $child1, $child2;
2596 If SIGNAL is zero, no signal is sent to the process. This is a
2597 useful way to check that a child process is alive and hasn't changed
2598 its UID. See L<perlport> for notes on the portability of this
2601 Unlike in the shell, if SIGNAL is negative, it kills
2602 process groups instead of processes. (On System V, a negative I<PROCESS>
2603 number will also kill process groups, but that's not portable.) That
2604 means you usually want to use positive not negative signals. You may also
2605 use a signal name in quotes.
2607 See L<perlipc/"Signals"> for more details.
2614 The C<last> command is like the C<break> statement in C (as used in
2615 loops); it immediately exits the loop in question. If the LABEL is
2616 omitted, the command refers to the innermost enclosing loop. The
2617 C<continue> block, if any, is not executed:
2619 LINE: while (<STDIN>) {
2620 last LINE if /^$/; # exit when done with header
2624 C<last> cannot be used to exit a block which returns a value such as
2625 C<eval {}>, C<sub {}> or C<do {}>, and should not be used to exit
2626 a grep() or map() operation.
2628 Note that a block by itself is semantically identical to a loop
2629 that executes once. Thus C<last> can be used to effect an early
2630 exit out of such a block.
2632 See also L</continue> for an illustration of how C<last>, C<next>, and
2640 Returns a lowercased version of EXPR. This is the internal function
2641 implementing the C<\L> escape in double-quoted strings. Respects
2642 current LC_CTYPE locale if C<use locale> in force. See L<perllocale>
2643 and L<perlunicode> for more details about locale and Unicode support.
2645 If EXPR is omitted, uses C<$_>.
2648 X<lcfirst> X<lowercase>
2652 Returns the value of EXPR with the first character lowercased. This
2653 is the internal function implementing the C<\l> escape in
2654 double-quoted strings. Respects current LC_CTYPE locale if C<use
2655 locale> in force. See L<perllocale> and L<perlunicode> for more
2656 details about locale and Unicode support.
2658 If EXPR is omitted, uses C<$_>.
2665 Returns the length in I<characters> of the value of EXPR. If EXPR is
2666 omitted, returns length of C<$_>. Note that this cannot be used on
2667 an entire array or hash to find out how many elements these have.
2668 For that, use C<scalar @array> and C<scalar keys %hash> respectively.
2670 Note the I<characters>: if the EXPR is in Unicode, you will get the
2671 number of characters, not the number of bytes. To get the length
2672 in bytes, use C<do { use bytes; length(EXPR) }>, see L<bytes>.
2674 =item link OLDFILE,NEWFILE
2677 Creates a new filename linked to the old filename. Returns true for
2678 success, false otherwise.
2680 =item listen SOCKET,QUEUESIZE
2683 Does the same thing that the listen system call does. Returns true if
2684 it succeeded, false otherwise. See the example in
2685 L<perlipc/"Sockets: Client/Server Communication">.
2690 You really probably want to be using C<my> instead, because C<local> isn't
2691 what most people think of as "local". See
2692 L<perlsub/"Private Variables via my()"> for details.
2694 A local modifies the listed variables to be local to the enclosing
2695 block, file, or eval. If more than one value is listed, the list must
2696 be placed in parentheses. See L<perlsub/"Temporary Values via local()">
2697 for details, including issues with tied arrays and hashes.
2699 =item localtime EXPR
2704 Converts a time as returned by the time function to a 9-element list
2705 with the time analyzed for the local time zone. Typically used as
2709 ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) =
2712 All list elements are numeric, and come straight out of the C `struct
2713 tm'. C<$sec>, C<$min>, and C<$hour> are the seconds, minutes, and hours
2714 of the specified time.
2716 C<$mday> is the day of the month, and C<$mon> is the month itself, in
2717 the range C<0..11> with 0 indicating January and 11 indicating December.
2718 This makes it easy to get a month name from a list:
2720 my @abbr = qw( Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec );
2721 print "$abbr[$mon] $mday";
2722 # $mon=9, $mday=18 gives "Oct 18"
2724 C<$year> is the number of years since 1900, not just the last two digits
2725 of the year. That is, C<$year> is C<123> in year 2023. The proper way
2726 to get a complete 4-digit year is simply:
2730 To get the last two digits of the year (e.g., '01' in 2001) do:
2732 $year = sprintf("%02d", $year % 100);
2734 C<$wday> is the day of the week, with 0 indicating Sunday and 3 indicating
2735 Wednesday. C<$yday> is the day of the year, in the range C<0..364>
2736 (or C<0..365> in leap years.)
2738 C<$isdst> is true if the specified time occurs during Daylight Saving
2739 Time, false otherwise.
2741 If EXPR is omitted, C<localtime()> uses the current time (C<localtime(time)>).
2743 In scalar context, C<localtime()> returns the ctime(3) value:
2745 $now_string = localtime; # e.g., "Thu Oct 13 04:54:34 1994"
2747 This scalar value is B<not> locale dependent but is a Perl builtin. For GMT
2748 instead of local time use the L</gmtime> builtin. See also the
2749 C<Time::Local> module (to convert the second, minutes, hours, ... back to
2750 the integer value returned by time()), and the L<POSIX> module's strftime(3)
2751 and mktime(3) functions.
2753 To get somewhat similar but locale dependent date strings, set up your
2754 locale environment variables appropriately (please see L<perllocale>) and
2757 use POSIX qw(strftime);
2758 $now_string = strftime "%a %b %e %H:%M:%S %Y", localtime;
2759 # or for GMT formatted appropriately for your locale:
2760 $now_string = strftime "%a %b %e %H:%M:%S %Y", gmtime;
2762 Note that the C<%a> and C<%b>, the short forms of the day of the week
2763 and the month of the year, may not necessarily be three characters wide.
2765 See L<perlport/localtime> for portability concerns.
2770 This function places an advisory lock on a shared variable, or referenced
2771 object contained in I<THING> until the lock goes out of scope.
2773 lock() is a "weak keyword" : this means that if you've defined a function
2774 by this name (before any calls to it), that function will be called
2775 instead. (However, if you've said C<use threads>, lock() is always a
2776 keyword.) See L<threads>.
2779 X<log> X<logarithm> X<e> X<ln> X<base>
2783 Returns the natural logarithm (base I<e>) of EXPR. If EXPR is omitted,
2784 returns log of C<$_>. To get the log of another base, use basic algebra:
2785 The base-N log of a number is equal to the natural log of that number
2786 divided by the natural log of N. For example:
2790 return log($n)/log(10);
2793 See also L</exp> for the inverse operation.
2800 Does the same thing as the C<stat> function (including setting the
2801 special C<_> filehandle) but stats a symbolic link instead of the file
2802 the symbolic link points to. If symbolic links are unimplemented on
2803 your system, a normal C<stat> is done. For much more detailed
2804 information, please see the documentation for C<stat>.
2806 If EXPR is omitted, stats C<$_>.
2810 The match operator. See L<perlop>.
2812 =item map BLOCK LIST
2817 Evaluates the BLOCK or EXPR for each element of LIST (locally setting
2818 C<$_> to each element) and returns the list value composed of the
2819 results of each such evaluation. In scalar context, returns the
2820 total number of elements so generated. Evaluates BLOCK or EXPR in
2821 list context, so each element of LIST may produce zero, one, or
2822 more elements in the returned value.
2824 @chars = map(chr, @nums);
2826 translates a list of numbers to the corresponding characters. And
2828 %hash = map { getkey($_) => $_ } @array;
2830 is just a funny way to write
2833 foreach $_ (@array) {
2834 $hash{getkey($_)} = $_;
2837 Note that C<$_> is an alias to the list value, so it can be used to
2838 modify the elements of the LIST. While this is useful and supported,
2839 it can cause bizarre results if the elements of LIST are not variables.
2840 Using a regular C<foreach> loop for this purpose would be clearer in
2841 most cases. See also L</grep> for an array composed of those items of
2842 the original list for which the BLOCK or EXPR evaluates to true.
2844 If C<$_> is lexical in the scope where the C<map> appears (because it has
2845 been declared with C<my $_>) then, in addition to being locally aliased to
2846 the list elements, C<$_> keeps being lexical inside the block; i.e. it
2847 can't be seen from the outside, avoiding any potential side-effects.
2849 C<{> starts both hash references and blocks, so C<map { ...> could be either
2850 the start of map BLOCK LIST or map EXPR, LIST. Because perl doesn't look
2851 ahead for the closing C<}> it has to take a guess at which its dealing with
2852 based what it finds just after the C<{>. Usually it gets it right, but if it
2853 doesn't it won't realize something is wrong until it gets to the C<}> and
2854 encounters the missing (or unexpected) comma. The syntax error will be
2855 reported close to the C<}> but you'll need to change something near the C<{>
2856 such as using a unary C<+> to give perl some help:
2858 %hash = map { "\L$_", 1 } @array # perl guesses EXPR. wrong
2859 %hash = map { +"\L$_", 1 } @array # perl guesses BLOCK. right
2860 %hash = map { ("\L$_", 1) } @array # this also works
2861 %hash = map { lc($_), 1 } @array # as does this.
2862 %hash = map +( lc($_), 1 ), @array # this is EXPR and works!
2864 %hash = map ( lc($_), 1 ), @array # evaluates to (1, @array)
2866 or to force an anon hash constructor use C<+{>
2868 @hashes = map +{ lc($_), 1 }, @array # EXPR, so needs , at end
2870 and you get list of anonymous hashes each with only 1 entry.
2872 =item mkdir FILENAME,MASK
2873 X<mkdir> X<md> X<directory, create>
2875 =item mkdir FILENAME
2879 Creates the directory specified by FILENAME, with permissions
2880 specified by MASK (as modified by C<umask>). If it succeeds it
2881 returns true, otherwise it returns false and sets C<$!> (errno).
2882 If omitted, MASK defaults to 0777. If omitted, FILENAME defaults
2885 In general, it is better to create directories with permissive MASK,
2886 and let the user modify that with their C<umask>, than it is to supply
2887 a restrictive MASK and give the user no way to be more permissive.
2888 The exceptions to this rule are when the file or directory should be
2889 kept private (mail files, for instance). The perlfunc(1) entry on
2890 C<umask> discusses the choice of MASK in more detail.
2892 Note that according to the POSIX 1003.1-1996 the FILENAME may have any
2893 number of trailing slashes. Some operating and filesystems do not get
2894 this right, so Perl automatically removes all trailing slashes to keep
2897 =item msgctl ID,CMD,ARG
2900 Calls the System V IPC function msgctl(2). You'll probably have to say
2904 first to get the correct constant definitions. If CMD is C<IPC_STAT>,
2905 then ARG must be a variable that will hold the returned C<msqid_ds>
2906 structure. Returns like C<ioctl>: the undefined value for error,
2907 C<"0 but true"> for zero, or the actual return value otherwise. See also
2908 L<perlipc/"SysV IPC">, C<IPC::SysV>, and C<IPC::Semaphore> documentation.
2910 =item msgget KEY,FLAGS
2913 Calls the System V IPC function msgget(2). Returns the message queue
2914 id, or the undefined value if there is an error. See also
2915 L<perlipc/"SysV IPC"> and C<IPC::SysV> and C<IPC::Msg> documentation.
2917 =item msgrcv ID,VAR,SIZE,TYPE,FLAGS
2920 Calls the System V IPC function msgrcv to receive a message from
2921 message queue ID into variable VAR with a maximum message size of
2922 SIZE. Note that when a message is received, the message type as a
2923 native long integer will be the first thing in VAR, followed by the
2924 actual message. This packing may be opened with C<unpack("l! a*")>.
2925 Taints the variable. Returns true if successful, or false if there is
2926 an error. See also L<perlipc/"SysV IPC">, C<IPC::SysV>, and
2927 C<IPC::SysV::Msg> documentation.
2929 =item msgsnd ID,MSG,FLAGS
2932 Calls the System V IPC function msgsnd to send the message MSG to the
2933 message queue ID. MSG must begin with the native long integer message
2934 type, and be followed by the length of the actual message, and finally
2935 the message itself. This kind of packing can be achieved with
2936 C<pack("l! a*", $type, $message)>. Returns true if successful,
2937 or false if there is an error. See also C<IPC::SysV>
2938 and C<IPC::SysV::Msg> documentation.
2945 =item my EXPR : ATTRS
2947 =item my TYPE EXPR : ATTRS
2949 A C<my> declares the listed variables to be local (lexically) to the
2950 enclosing block, file, or C<eval>. If more than one value is listed,
2951 the list must be placed in parentheses.
2953 The exact semantics and interface of TYPE and ATTRS are still
2954 evolving. TYPE is currently bound to the use of C<fields> pragma,
2955 and attributes are handled using the C<attributes> pragma, or starting
2956 from Perl 5.8.0 also via the C<Attribute::Handlers> module. See
2957 L<perlsub/"Private Variables via my()"> for details, and L<fields>,
2958 L<attributes>, and L<Attribute::Handlers>.
2965 The C<next> command is like the C<continue> statement in C; it starts
2966 the next iteration of the loop:
2968 LINE: while (<STDIN>) {
2969 next LINE if /^#/; # discard comments
2973 Note that if there were a C<continue> block on the above, it would get
2974 executed even on discarded lines. If the LABEL is omitted, the command
2975 refers to the innermost enclosing loop.
2977 C<next> cannot be used to exit a block which returns a value such as
2978 C<eval {}>, C<sub {}> or C<do {}>, and should not be used to exit
2979 a grep() or map() operation.
2981 Note that a block by itself is semantically identical to a loop
2982 that executes once. Thus C<next> will exit such a block early.
2984 See also L</continue> for an illustration of how C<last>, C<next>, and
2987 =item no Module VERSION LIST
2990 =item no Module VERSION
2992 =item no Module LIST
2996 See the C<use> function, of which C<no> is the opposite.
2999 X<oct> X<octal> X<hex> X<hexadecimal> X<binary> X<bin>
3003 Interprets EXPR as an octal string and returns the corresponding
3004 value. (If EXPR happens to start off with C<0x>, interprets it as a
3005 hex string. If EXPR starts off with C<0b>, it is interpreted as a
3006 binary string. Leading whitespace is ignored in all three cases.)
3007 The following will handle decimal, binary, octal, and hex in the standard
3010 $val = oct($val) if $val =~ /^0/;
3012 If EXPR is omitted, uses C<$_>. To go the other way (produce a number
3013 in octal), use sprintf() or printf():
3015 $perms = (stat("filename"))[2] & 07777;
3016 $oct_perms = sprintf "%lo", $perms;
3018 The oct() function is commonly used when a string such as C<644> needs
3019 to be converted into a file mode, for example. (Although perl will
3020 automatically convert strings into numbers as needed, this automatic
3021 conversion assumes base 10.)
3023 =item open FILEHANDLE,EXPR
3024 X<open> X<pipe> X<file, open> X<fopen>
3026 =item open FILEHANDLE,MODE,EXPR
3028 =item open FILEHANDLE,MODE,EXPR,LIST
3030 =item open FILEHANDLE,MODE,REFERENCE
3032 =item open FILEHANDLE
3034 Opens the file whose filename is given by EXPR, and associates it with
3037 (The following is a comprehensive reference to open(): for a gentler
3038 introduction you may consider L<perlopentut>.)
3040 If FILEHANDLE is an undefined scalar variable (or array or hash element)
3041 the variable is assigned a reference to a new anonymous filehandle,
3042 otherwise if FILEHANDLE is an expression, its value is used as the name of
3043 the real filehandle wanted. (This is considered a symbolic reference, so
3044 C<use strict 'refs'> should I<not> be in effect.)
3046 If EXPR is omitted, the scalar variable of the same name as the
3047 FILEHANDLE contains the filename. (Note that lexical variables--those
3048 declared with C<my>--will not work for this purpose; so if you're
3049 using C<my>, specify EXPR in your call to open.)
3051 If three or more arguments are specified then the mode of opening and
3052 the file name are separate. If MODE is C<< '<' >> or nothing, the file
3053 is opened for input. If MODE is C<< '>' >>, the file is truncated and
3054 opened for output, being created if necessary. If MODE is C<<< '>>' >>>,
3055 the file is opened for appending, again being created if necessary.
3057 You can put a C<'+'> in front of the C<< '>' >> or C<< '<' >> to
3058 indicate that you want both read and write access to the file; thus
3059 C<< '+<' >> is almost always preferred for read/write updates--the C<<
3060 '+>' >> mode would clobber the file first. You can't usually use
3061 either read-write mode for updating textfiles, since they have
3062 variable length records. See the B<-i> switch in L<perlrun> for a
3063 better approach. The file is created with permissions of C<0666>
3064 modified by the process' C<umask> value.
3066 These various prefixes correspond to the fopen(3) modes of C<'r'>,
3067 C<'r+'>, C<'w'>, C<'w+'>, C<'a'>, and C<'a+'>.
3069 In the 2-arguments (and 1-argument) form of the call the mode and
3070 filename should be concatenated (in this order), possibly separated by
3071 spaces. It is possible to omit the mode in these forms if the mode is
3074 If the filename begins with C<'|'>, the filename is interpreted as a
3075 command to which output is to be piped, and if the filename ends with a
3076 C<'|'>, the filename is interpreted as a command which pipes output to
3077 us. See L<perlipc/"Using open() for IPC">
3078 for more examples of this. (You are not allowed to C<open> to a command
3079 that pipes both in I<and> out, but see L<IPC::Open2>, L<IPC::Open3>,
3080 and L<perlipc/"Bidirectional Communication with Another Process">
3083 For three or more arguments if MODE is C<'|-'>, the filename is
3084 interpreted as a command to which output is to be piped, and if MODE
3085 is C<'-|'>, the filename is interpreted as a command which pipes
3086 output to us. In the 2-arguments (and 1-argument) form one should
3087 replace dash (C<'-'>) with the command.
3088 See L<perlipc/"Using open() for IPC"> for more examples of this.
3089 (You are not allowed to C<open> to a command that pipes both in I<and>
3090 out, but see L<IPC::Open2>, L<IPC::Open3>, and
3091 L<perlipc/"Bidirectional Communication"> for alternatives.)
3093 In the three-or-more argument form of pipe opens, if LIST is specified
3094 (extra arguments after the command name) then LIST becomes arguments
3095 to the command invoked if the platform supports it. The meaning of
3096 C<open> with more than three arguments for non-pipe modes is not yet
3097 specified. Experimental "layers" may give extra LIST arguments
3100 In the 2-arguments (and 1-argument) form opening C<'-'> opens STDIN
3101 and opening C<< '>-' >> opens STDOUT.
3103 You may use the three-argument form of open to specify IO "layers"
3104 (sometimes also referred to as "disciplines") to be applied to the handle
3105 that affect how the input and output are processed (see L<open> and
3106 L<PerlIO> for more details). For example
3108 open(FH, "<:utf8", "file")
3110 will open the UTF-8 encoded file containing Unicode characters,
3111 see L<perluniintro>. Note that if layers are specified in the
3112 three-arg form then default layers stored in ${^OPEN} (see L<perlvar>;
3113 usually set by the B<open> pragma or the switch B<-CioD>) are ignored.
3115 Open returns nonzero upon success, the undefined value otherwise. If
3116 the C<open> involved a pipe, the return value happens to be the pid of
3119 If you're running Perl on a system that distinguishes between text
3120 files and binary files, then you should check out L</binmode> for tips
3121 for dealing with this. The key distinction between systems that need
3122 C<binmode> and those that don't is their text file formats. Systems
3123 like Unix, Mac OS, and Plan 9, which delimit lines with a single
3124 character, and which encode that character in C as C<"\n">, do not
3125 need C<binmode>. The rest need it.
3127 When opening a file, it's usually a bad idea to continue normal execution
3128 if the request failed, so C<open> is frequently used in connection with
3129 C<die>. Even if C<die> won't do what you want (say, in a CGI script,
3130 where you want to make a nicely formatted error message (but there are
3131 modules that can help with that problem)) you should always check
3132 the return value from opening a file. The infrequent exception is when
3133 working with an unopened filehandle is actually what you want to do.
3135 As a special case the 3-arg form with a read/write mode and the third
3136 argument being C<undef>:
3138 open(TMP, "+>", undef) or die ...
3140 opens a filehandle to an anonymous temporary file. Also using "+<"
3141 works for symmetry, but you really should consider writing something
3142 to the temporary file first. You will need to seek() to do the
3145 Since v5.8.0, perl has built using PerlIO by default. Unless you've
3146 changed this (i.e. Configure -Uuseperlio), you can open file handles to
3147 "in memory" files held in Perl scalars via:
3149 open($fh, '>', \$variable) || ..
3151 Though if you try to re-open C<STDOUT> or C<STDERR> as an "in memory"
3152 file, you have to close it first:
3155 open STDOUT, '>', \$variable or die "Can't open STDOUT: $!";
3160 open ARTICLE or die "Can't find article $ARTICLE: $!\n";
3161 while (<ARTICLE>) {...
3163 open(LOG, '>>/usr/spool/news/twitlog'); # (log is reserved)
3164 # if the open fails, output is discarded
3166 open(DBASE, '+<', 'dbase.mine') # open for update
3167 or die "Can't open 'dbase.mine' for update: $!";
3169 open(DBASE, '+<dbase.mine') # ditto
3170 or die "Can't open 'dbase.mine' for update: $!";
3172 open(ARTICLE, '-|', "caesar <$article") # decrypt article
3173 or die "Can't start caesar: $!";
3175 open(ARTICLE, "caesar <$article |") # ditto
3176 or die "Can't start caesar: $!";
3178 open(EXTRACT, "|sort >Tmp$$") # $$ is our process id
3179 or die "Can't start sort: $!";
3182 open(MEMORY,'>', \$var)
3183 or die "Can't open memory file: $!";
3184 print MEMORY "foo!\n"; # output will end up in $var
3186 # process argument list of files along with any includes
3188 foreach $file (@ARGV) {
3189 process($file, 'fh00');
3193 my($filename, $input) = @_;
3194 $input++; # this is a string increment
3195 unless (open($input, $filename)) {
3196 print STDERR "Can't open $filename: $!\n";
3201 while (<$input>) { # note use of indirection
3202 if (/^#include "(.*)"/) {
3203 process($1, $input);
3210 See L<perliol> for detailed info on PerlIO.
3212 You may also, in the Bourne shell tradition, specify an EXPR beginning
3213 with C<< '>&' >>, in which case the rest of the string is interpreted
3214 as the name of a filehandle (or file descriptor, if numeric) to be
3215 duped (as L<dup(2)>) and opened. You may use C<&> after C<< > >>,
3216 C<<< >> >>>, C<< < >>, C<< +> >>, C<<< +>> >>>, and C<< +< >>.
3217 The mode you specify should match the mode of the original filehandle.
3218 (Duping a filehandle does not take into account any existing contents
3219 of IO buffers.) If you use the 3-arg form then you can pass either a
3220 number, the name of a filehandle or the normal "reference to a glob".
3222 Here is a script that saves, redirects, and restores C<STDOUT> and
3223 C<STDERR> using various methods:
3226 open my $oldout, ">&STDOUT" or die "Can't dup STDOUT: $!";
3227 open OLDERR, ">&", \*STDERR or die "Can't dup STDERR: $!";
3229 open STDOUT, '>', "foo.out" or die "Can't redirect STDOUT: $!";
3230 open STDERR, ">&STDOUT" or die "Can't dup STDOUT: $!";
3232 select STDERR; $| = 1; # make unbuffered
3233 select STDOUT; $| = 1; # make unbuffered
3235 print STDOUT "stdout 1\n"; # this works for
3236 print STDERR "stderr 1\n"; # subprocesses too
3238 open STDOUT, ">&", $oldout or die "Can't dup \$oldout: $!";
3239 open STDERR, ">&OLDERR" or die "Can't dup OLDERR: $!";
3241 print STDOUT "stdout 2\n";
3242 print STDERR "stderr 2\n";
3244 If you specify C<< '<&=X' >>, where C<X> is a file descriptor number
3245 or a filehandle, then Perl will do an equivalent of C's C<fdopen> of
3246 that file descriptor (and not call L<dup(2)>); this is more
3247 parsimonious of file descriptors. For example:
3249 # open for input, reusing the fileno of $fd
3250 open(FILEHANDLE, "<&=$fd")
3254 open(FILEHANDLE, "<&=", $fd)
3258 # open for append, using the fileno of OLDFH
3259 open(FH, ">>&=", OLDFH)
3263 open(FH, ">>&=OLDFH")
3265 Being parsimonious on filehandles is also useful (besides being
3266 parsimonious) for example when something is dependent on file
3267 descriptors, like for example locking using flock(). If you do just
3268 C<< open(A, '>>&B') >>, the filehandle A will not have the same file
3269 descriptor as B, and therefore flock(A) will not flock(B), and vice
3270 versa. But with C<< open(A, '>>&=B') >> the filehandles will share
3271 the same file descriptor.
3273 Note that if you are using Perls older than 5.8.0, Perl will be using
3274 the standard C libraries' fdopen() to implement the "=" functionality.
3275 On many UNIX systems fdopen() fails when file descriptors exceed a
3276 certain value, typically 255. For Perls 5.8.0 and later, PerlIO is
3277 most often the default.
3279 You can see whether Perl has been compiled with PerlIO or not by
3280 running C<perl -V> and looking for C<useperlio=> line. If C<useperlio>
3281 is C<define>, you have PerlIO, otherwise you don't.
3283 If you open a pipe on the command C<'-'>, i.e., either C<'|-'> or C<'-|'>
3284 with 2-arguments (or 1-argument) form of open(), then
3285 there is an implicit fork done, and the return value of open is the pid
3286 of the child within the parent process, and C<0> within the child
3287 process. (Use C<defined($pid)> to determine whether the open was successful.)
3288 The filehandle behaves normally for the parent, but i/o to that
3289 filehandle is piped from/to the STDOUT/STDIN of the child process.
3290 In the child process the filehandle isn't opened--i/o happens from/to
3291 the new STDOUT or STDIN. Typically this is used like the normal
3292 piped open when you want to exercise more control over just how the
3293 pipe command gets executed, such as when you are running setuid, and
3294 don't want to have to scan shell commands for metacharacters.
3295 The following triples are more or less equivalent:
3297 open(FOO, "|tr '[a-z]' '[A-Z]'");
3298 open(FOO, '|-', "tr '[a-z]' '[A-Z]'");
3299 open(FOO, '|-') || exec 'tr', '[a-z]', '[A-Z]';
3300 open(FOO, '|-', "tr", '[a-z]', '[A-Z]');
3302 open(FOO, "cat -n '$file'|");
3303 open(FOO, '-|', "cat -n '$file'");
3304 open(FOO, '-|') || exec 'cat', '-n', $file;
3305 open(FOO, '-|', "cat", '-n', $file);
3307 The last example in each block shows the pipe as "list form", which is
3308 not yet supported on all platforms. A good rule of thumb is that if
3309 your platform has true C<fork()> (in other words, if your platform is
3310 UNIX) you can use the list form.
3312 See L<perlipc/"Safe Pipe Opens"> for more examples of this.
3314 Beginning with v5.6.0, Perl will attempt to flush all files opened for
3315 output before any operation that may do a fork, but this may not be
3316 supported on some platforms (see L<perlport>). To be safe, you may need
3317 to set C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method
3318 of C<IO::Handle> on any open handles.
3320 On systems that support a close-on-exec flag on files, the flag will
3321 be set for the newly opened file descriptor as determined by the value
3322 of $^F. See L<perlvar/$^F>.
3324 Closing any piped filehandle causes the parent process to wait for the
3325 child to finish, and returns the status value in C<$?> and
3326 C<${^CHILD_ERROR_NATIVE}>.
3328 The filename passed to 2-argument (or 1-argument) form of open() will
3329 have leading and trailing whitespace deleted, and the normal
3330 redirection characters honored. This property, known as "magic open",
3331 can often be used to good effect. A user could specify a filename of
3332 F<"rsh cat file |">, or you could change certain filenames as needed:
3334 $filename =~ s/(.*\.gz)\s*$/gzip -dc < $1|/;
3335 open(FH, $filename) or die "Can't open $filename: $!";
3337 Use 3-argument form to open a file with arbitrary weird characters in it,
3339 open(FOO, '<', $file);
3341 otherwise it's necessary to protect any leading and trailing whitespace:
3343 $file =~ s#^(\s)#./$1#;
3344 open(FOO, "< $file\0");
3346 (this may not work on some bizarre filesystems). One should
3347 conscientiously choose between the I<magic> and 3-arguments form
3352 will allow the user to specify an argument of the form C<"rsh cat file |">,
3353 but will not work on a filename which happens to have a trailing space, while
3355 open IN, '<', $ARGV[0];
3357 will have exactly the opposite restrictions.
3359 If you want a "real" C C<open> (see L<open(2)> on your system), then you
3360 should use the C<sysopen> function, which involves no such magic (but
3361 may use subtly different filemodes than Perl open(), which is mapped
3362 to C fopen()). This is
3363 another way to protect your filenames from interpretation. For example:
3366 sysopen(HANDLE, $path, O_RDWR|O_CREAT|O_EXCL)
3367 or die "sysopen $path: $!";
3368 $oldfh = select(HANDLE); $| = 1; select($oldfh);
3369 print HANDLE "stuff $$\n";
3371 print "File contains: ", <HANDLE>;
3373 Using the constructor from the C<IO::Handle> package (or one of its
3374 subclasses, such as C<IO::File> or C<IO::Socket>), you can generate anonymous
3375 filehandles that have the scope of whatever variables hold references to
3376 them, and automatically close whenever and however you leave that scope:
3380 sub read_myfile_munged {
3382 my $handle = new IO::File;
3383 open($handle, "myfile") or die "myfile: $!";
3385 or return (); # Automatically closed here.
3386 mung $first or die "mung failed"; # Or here.
3387 return $first, <$handle> if $ALL; # Or here.
3391 See L</seek> for some details about mixing reading and writing.
3393 =item opendir DIRHANDLE,EXPR
3396 Opens a directory named EXPR for processing by C<readdir>, C<telldir>,
3397 C<seekdir>, C<rewinddir>, and C<closedir>. Returns true if successful.
3398 DIRHANDLE may be an expression whose value can be used as an indirect
3399 dirhandle, usually the real dirhandle name. If DIRHANDLE is an undefined
3400 scalar variable (or array or hash element), the variable is assigned a
3401 reference to a new anonymous dirhandle.
3402 DIRHANDLEs have their own namespace separate from FILEHANDLEs.
3409 Returns the numeric (the native 8-bit encoding, like ASCII or EBCDIC,
3410 or Unicode) value of the first character of EXPR. If EXPR is omitted,
3413 For the reverse, see L</chr>.
3414 See L<perlunicode> and L<encoding> for more about Unicode.
3421 =item our EXPR : ATTRS
3423 =item our TYPE EXPR : ATTRS
3425 C<our> associates a simple name with a package variable in the current
3426 package for use within the current scope. When C<use strict 'vars'> is in
3427 effect, C<our> lets you use declared global variables without qualifying
3428 them with package names, within the lexical scope of the C<our> declaration.
3429 In this way C<our> differs from C<use vars>, which is package scoped.
3431 Unlike C<my>, which both allocates storage for a variable and associates
3432 a simple name with that storage for use within the current scope, C<our>
3433 associates a simple name with a package variable in the current package,
3434 for use within the current scope. In other words, C<our> has the same
3435 scoping rules as C<my>, but does not necessarily create a
3438 If more than one value is listed, the list must be placed
3444 An C<our> declaration declares a global variable that will be visible
3445 across its entire lexical scope, even across package boundaries. The
3446 package in which the variable is entered is determined at the point
3447 of the declaration, not at the point of use. This means the following
3451 our $bar; # declares $Foo::bar for rest of lexical scope
3455 print $bar; # prints 20, as it refers to $Foo::bar
3457 Multiple C<our> declarations with the same name in the same lexical
3458 scope are allowed if they are in different packages. If they happen
3459 to be in the same package, Perl will emit warnings if you have asked
3460 for them, just like multiple C<my> declarations. Unlike a second
3461 C<my> declaration, which will bind the name to a fresh variable, a
3462 second C<our> declaration in the same package, in the same scope, is
3467 our $bar; # declares $Foo::bar for rest of lexical scope
3471 our $bar = 30; # declares $Bar::bar for rest of lexical scope
3472 print $bar; # prints 30
3474 our $bar; # emits warning but has no other effect
3475 print $bar; # still prints 30
3477 An C<our> declaration may also have a list of attributes associated
3480 The exact semantics and interface of TYPE and ATTRS are still
3481 evolving. TYPE is currently bound to the use of C<fields> pragma,
3482 and attributes are handled using the C<attributes> pragma, or starting
3483 from Perl 5.8.0 also via the C<Attribute::Handlers> module. See
3484 L<perlsub/"Private Variables via my()"> for details, and L<fields>,
3485 L<attributes>, and L<Attribute::Handlers>.
3487 The only currently recognized C<our()> attribute is C<unique> which
3488 indicates that a single copy of the global is to be used by all
3489 interpreters should the program happen to be running in a
3490 multi-interpreter environment. (The default behaviour would be for
3491 each interpreter to have its own copy of the global.) Examples:
3493 our @EXPORT : unique = qw(foo);
3494 our %EXPORT_TAGS : unique = (bar => [qw(aa bb cc)]);
3495 our $VERSION : unique = "1.00";
3497 Note that this attribute also has the effect of making the global
3498 readonly when the first new interpreter is cloned (for example,
3499 when the first new thread is created).
3501 Multi-interpreter environments can come to being either through the
3502 fork() emulation on Windows platforms, or by embedding perl in a
3503 multi-threaded application. The C<unique> attribute does nothing in
3504 all other environments.
3506 Warning: the current implementation of this attribute operates on the
3507 typeglob associated with the variable; this means that C<our $x : unique>
3508 also has the effect of C<our @x : unique; our %x : unique>. This may be
3511 =item pack TEMPLATE,LIST
3514 Takes a LIST of values and converts it into a string using the rules
3515 given by the TEMPLATE. The resulting string is the concatenation of
3516 the converted values. Typically, each converted value looks
3517 like its machine-level representation. For example, on 32-bit machines
3518 an integer may be represented by a sequence of 4 bytes that will be
3519 converted to a sequence of 4 characters.
3521 The TEMPLATE is a sequence of characters that give the order and type
3522 of values, as follows:
3524 a A string with arbitrary binary data, will be null padded.
3525 A A text (ASCII) string, will be space padded.
3526 Z A null terminated (ASCIZ) string, will be null padded.
3528 b A bit string (ascending bit order inside each byte, like vec()).
3529 B A bit string (descending bit order inside each byte).
3530 h A hex string (low nybble first).
3531 H A hex string (high nybble first).
3533 c A signed char (8-bit) value.
3534 C An unsigned C char (octet) even under Unicode. Should normally not
3535 be used. See U and W instead.
3536 W An unsigned char value (can be greater than 255).
3538 s A signed short (16-bit) value.
3539 S An unsigned short value.
3541 l A signed long (32-bit) value.
3542 L An unsigned long value.
3544 q A signed quad (64-bit) value.
3545 Q An unsigned quad value.
3546 (Quads are available only if your system supports 64-bit
3547 integer values _and_ if Perl has been compiled to support those.
3548 Causes a fatal error otherwise.)
3550 i A signed integer value.
3551 I A unsigned integer value.
3552 (This 'integer' is _at_least_ 32 bits wide. Its exact
3553 size depends on what a local C compiler calls 'int'.)
3555 n An unsigned short (16-bit) in "network" (big-endian) order.
3556 N An unsigned long (32-bit) in "network" (big-endian) order.
3557 v An unsigned short (16-bit) in "VAX" (little-endian) order.
3558 V An unsigned long (32-bit) in "VAX" (little-endian) order.
3560 j A Perl internal signed integer value (IV).
3561 J A Perl internal unsigned integer value (UV).
3563 f A single-precision float in the native format.
3564 d A double-precision float in the native format.
3566 F A Perl internal floating point value (NV) in the native format
3567 D A long double-precision float in the native format.
3568 (Long doubles are available only if your system supports long
3569 double values _and_ if Perl has been compiled to support those.
3570 Causes a fatal error otherwise.)
3572 p A pointer to a null-terminated string.
3573 P A pointer to a structure (fixed-length string).
3575 u A uuencoded string.
3576 U A Unicode character number. Encodes to UTF-8 internally
3577 (or UTF-EBCDIC in EBCDIC platforms).
3579 w A BER compressed integer (not an ASN.1 BER, see perlpacktut for
3580 details). Its bytes represent an unsigned integer in base 128,
3581 most significant digit first, with as few digits as possible. Bit
3582 eight (the high bit) is set on each byte except the last.
3586 @ Null fill or truncate to absolute position, counted from the
3587 start of the innermost ()-group.
3588 . Null fill or truncate to absolute position specified by value.
3589 ( Start of a ()-group.
3591 One or more of the modifiers below may optionally follow some letters in the
3592 TEMPLATE (the second column lists the letters for which the modifier is
3595 ! sSlLiI Forces native (short, long, int) sizes instead
3596 of fixed (16-/32-bit) sizes.
3598 xX Make x and X act as alignment commands.
3600 nNvV Treat integers as signed instead of unsigned.
3602 @. Specify position as byte offset in the internal
3603 representation of the packed string. Efficient but
3606 > sSiIlLqQ Force big-endian byte-order on the type.
3607 jJfFdDpP (The "big end" touches the construct.)
3609 < sSiIlLqQ Force little-endian byte-order on the type.
3610 jJfFdDpP (The "little end" touches the construct.)
3612 The C<E<gt>> and C<E<lt>> modifiers can also be used on C<()>-groups,
3613 in which case they force a certain byte-order on all components of
3614 that group, including subgroups.
3616 The following rules apply:
3622 Each letter may optionally be followed by a number giving a repeat
3623 count. With all types except C<a>, C<A>, C<Z>, C<b>, C<B>, C<h>,
3624 C<H>, C<@>, C<.>, C<x>, C<X> and C<P> the pack function will gobble up
3625 that many values from the LIST. A C<*> for the repeat count means to
3626 use however many items are left, except for C<@>, C<x>, C<X>, where it
3627 is equivalent to C<0>, for <.> where it means relative to string start
3628 and C<u>, where it is equivalent to 1 (or 45, which is the same).
3629 A numeric repeat count may optionally be enclosed in brackets, as in
3630 C<pack 'C[80]', @arr>.
3632 One can replace the numeric repeat count by a template enclosed in brackets;
3633 then the packed length of this template in bytes is used as a count.
3634 For example, C<x[L]> skips a long (it skips the number of bytes in a long);
3635 the template C<$t X[$t] $t> unpack()s twice what $t unpacks.
3636 If the template in brackets contains alignment commands (such as C<x![d]>),
3637 its packed length is calculated as if the start of the template has the maximal
3640 When used with C<Z>, C<*> results in the addition of a trailing null
3641 byte (so the packed result will be one longer than the byte C<length>
3644 When used with C<@>, the repeat count represents an offset from the start
3645 of the innermost () group.
3647 When used with C<.>, the repeat count is used to determine the starting
3648 position from where the value offset is calculated. If the repeat count
3649 is 0, it's relative to the current position. If the repeat count is C<*>,
3650 the offset is relative to the start of the packed string. And if its an
3651 integer C<n> the offset is relative to the start of the n-th innermost
3652 () group (or the start of the string if C<n> is bigger then the group
3655 The repeat count for C<u> is interpreted as the maximal number of bytes
3656 to encode per line of output, with 0, 1 and 2 replaced by 45. The repeat
3657 count should not be more than 65.
3661 The C<a>, C<A>, and C<Z> types gobble just one value, but pack it as a
3662 string of length count, padding with nulls or spaces as necessary. When
3663 unpacking, C<A> strips trailing whitespace and nulls, C<Z> strips everything
3664 after the first null, and C<a> returns data verbatim.
3666 If the value-to-pack is too long, it is truncated. If too long and an
3667 explicit count is provided, C<Z> packs only C<$count-1> bytes, followed
3668 by a null byte. Thus C<Z> always packs a trailing null (except when the
3673 Likewise, the C<b> and C<B> fields pack a string that many bits long.
3674 Each character of the input field of pack() generates 1 bit of the result.
3675 Each result bit is based on the least-significant bit of the corresponding
3676 input character, i.e., on C<ord($char)%2>. In particular, characters C<"0">
3677 and C<"1"> generate bits 0 and 1, as do characters C<"\0"> and C<"\1">.
3679 Starting from the beginning of the input string of pack(), each 8-tuple
3680 of characters is converted to 1 character of output. With format C<b>
3681 the first character of the 8-tuple determines the least-significant bit of a
3682 character, and with format C<B> it determines the most-significant bit of
3685 If the length of the input string is not exactly divisible by 8, the
3686 remainder is packed as if the input string were padded by null characters
3687 at the end. Similarly, during unpack()ing the "extra" bits are ignored.
3689 If the input string of pack() is longer than needed, extra characters are
3690 ignored. A C<*> for the repeat count of pack() means to use all the
3691 characters of the input field. On unpack()ing the bits are converted to a
3692 string of C<"0">s and C<"1">s.
3696 The C<h> and C<H> fields pack a string that many nybbles (4-bit groups,
3697 representable as hexadecimal digits, 0-9a-f) long.
3699 Each character of the input field of pack() generates 4 bits of the result.
3700 For non-alphabetical characters the result is based on the 4 least-significant
3701 bits of the input character, i.e., on C<ord($char)%16>. In particular,
3702 characters C<"0"> and C<"1"> generate nybbles 0 and 1, as do bytes
3703 C<"\0"> and C<"\1">. For characters C<"a".."f"> and C<"A".."F"> the result
3704 is compatible with the usual hexadecimal digits, so that C<"a"> and
3705 C<"A"> both generate the nybble C<0xa==10>. The result for characters
3706 C<"g".."z"> and C<"G".."Z"> is not well-defined.
3708 Starting from the beginning of the input string of pack(), each pair
3709 of characters is converted to 1 character of output. With format C<h> the
3710 first character of the pair determines the least-significant nybble of the
3711 output character, and with format C<H> it determines the most-significant
3714 If the length of the input string is not even, it behaves as if padded
3715 by a null character at the end. Similarly, during unpack()ing the "extra"
3716 nybbles are ignored.
3718 If the input string of pack() is longer than needed, extra characters are
3720 A C<*> for the repeat count of pack() means to use all the characters of
3721 the input field. On unpack()ing the nybbles are converted to a string
3722 of hexadecimal digits.
3726 The C<p> type packs a pointer to a null-terminated string. You are
3727 responsible for ensuring the string is not a temporary value (which can
3728 potentially get deallocated before you get around to using the packed result).
3729 The C<P> type packs a pointer to a structure of the size indicated by the
3730 length. A NULL pointer is created if the corresponding value for C<p> or
3731 C<P> is C<undef>, similarly for unpack().
3733 If your system has a strange pointer size (i.e. a pointer is neither as
3734 big as an int nor as big as a long), it may not be possible to pack or
3735 unpack pointers in big- or little-endian byte order. Attempting to do
3736 so will result in a fatal error.
3740 The C</> template character allows packing and unpacking of a sequence of
3741 items where the packed structure contains a packed item count followed by
3742 the packed items themselves.
3743 You write I<length-item>C</>I<sequence-item>.
3745 The I<length-item> can be any C<pack> template letter, and describes
3746 how the length value is packed. The ones likely to be of most use are
3747 integer-packing ones like C<n> (for Java strings), C<w> (for ASN.1 or
3748 SNMP) and C<N> (for Sun XDR).
3750 For C<pack>, the I<sequence-item> may have a repeat count, in which case
3751 the minimum of that and the number of available items is used as argument
3752 for the I<length-item>. If it has no repeat count or uses a '*', the number
3753 of available items is used. For C<unpack> the repeat count is always obtained
3754 by decoding the packed item count, and the I<sequence-item> must not have a
3757 If the I<sequence-item> refers to a string type (C<"A">, C<"a"> or C<"Z">),
3758 the I<length-item> is a string length, not a number of strings. If there is
3759 an explicit repeat count for pack, the packed string will be adjusted to that
3762 unpack 'W/a', "\04Gurusamy"; gives ('Guru')
3763 unpack 'a3/A* A*', '007 Bond J '; gives (' Bond', 'J')
3764 pack 'n/a* w/a','hello,','world'; gives "\000\006hello,\005world"
3765 pack 'a/W2', ord('a') .. ord('z'); gives '2ab'
3767 The I<length-item> is not returned explicitly from C<unpack>.
3769 Adding a count to the I<length-item> letter is unlikely to do anything
3770 useful, unless that letter is C<A>, C<a> or C<Z>. Packing with a
3771 I<length-item> of C<a> or C<Z> may introduce C<"\000"> characters,
3772 which Perl does not regard as legal in numeric strings.
3776 The integer types C<s>, C<S>, C<l>, and C<L> may be
3777 followed by a C<!> modifier to signify native shorts or
3778 longs--as you can see from above for example a bare C<l> does mean
3779 exactly 32 bits, the native C<long> (as seen by the local C compiler)
3780 may be larger. This is an issue mainly in 64-bit platforms. You can
3781 see whether using C<!> makes any difference by
3783 print length(pack("s")), " ", length(pack("s!")), "\n";
3784 print length(pack("l")), " ", length(pack("l!")), "\n";
3786 C<i!> and C<I!> also work but only because of completeness;
3787 they are identical to C<i> and C<I>.
3789 The actual sizes (in bytes) of native shorts, ints, longs, and long
3790 longs on the platform where Perl was built are also available via
3794 print $Config{shortsize}, "\n";
3795 print $Config{intsize}, "\n";
3796 print $Config{longsize}, "\n";
3797 print $Config{longlongsize}, "\n";
3799 (The C<$Config{longlongsize}> will be undefined if your system does
3800 not support long longs.)
3804 The integer formats C<s>, C<S>, C<i>, C<I>, C<l>, C<L>, C<j>, and C<J>
3805 are inherently non-portable between processors and operating systems
3806 because they obey the native byteorder and endianness. For example a
3807 4-byte integer 0x12345678 (305419896 decimal) would be ordered natively
3808 (arranged in and handled by the CPU registers) into bytes as
3810 0x12 0x34 0x56 0x78 # big-endian
3811 0x78 0x56 0x34 0x12 # little-endian
3813 Basically, the Intel and VAX CPUs are little-endian, while everybody
3814 else, for example Motorola m68k/88k, PPC, Sparc, HP PA, Power, and
3815 Cray are big-endian. Alpha and MIPS can be either: Digital/Compaq
3816 used/uses them in little-endian mode; SGI/Cray uses them in big-endian
3819 The names `big-endian' and `little-endian' are comic references to
3820 the classic "Gulliver's Travels" (via the paper "On Holy Wars and a
3821 Plea for Peace" by Danny Cohen, USC/ISI IEN 137, April 1, 1980) and
3822 the egg-eating habits of the Lilliputians.
3824 Some systems may have even weirder byte orders such as
3829 You can see your system's preference with
3831 print join(" ", map { sprintf "%#02x", $_ }
3832 unpack("W*",pack("L",0x12345678))), "\n";
3834 The byteorder on the platform where Perl was built is also available
3838 print $Config{byteorder}, "\n";
3840 Byteorders C<'1234'> and C<'12345678'> are little-endian, C<'4321'>
3841 and C<'87654321'> are big-endian.
3843 If you want portable packed integers you can either use the formats
3844 C<n>, C<N>, C<v>, and C<V>, or you can use the C<E<gt>> and C<E<lt>>
3845 modifiers. These modifiers are only available as of perl 5.9.2.
3846 See also L<perlport>.
3850 All integer and floating point formats as well as C<p> and C<P> and
3851 C<()>-groups may be followed by the C<E<gt>> or C<E<lt>> modifiers
3852 to force big- or little- endian byte-order, respectively.
3853 This is especially useful, since C<n>, C<N>, C<v> and C<V> don't cover
3854 signed integers, 64-bit integers and floating point values. However,
3855 there are some things to keep in mind.
3857 Exchanging signed integers between different platforms only works
3858 if all platforms store them in the same format. Most platforms store
3859 signed integers in two's complement, so usually this is not an issue.
3861 The C<E<gt>> or C<E<lt>> modifiers can only be used on floating point
3862 formats on big- or little-endian machines. Otherwise, attempting to
3863 do so will result in a fatal error.
3865 Forcing big- or little-endian byte-order on floating point values for
3866 data exchange can only work if all platforms are using the same
3867 binary representation (e.g. IEEE floating point format). Even if all
3868 platforms are using IEEE, there may be subtle differences. Being able
3869 to use C<E<gt>> or C<E<lt>> on floating point values can be very useful,
3870 but also very dangerous if you don't know exactly what you're doing.
3871 It is definitely not a general way to portably store floating point
3874 When using C<E<gt>> or C<E<lt>> on an C<()>-group, this will affect
3875 all types inside the group that accept the byte-order modifiers,
3876 including all subgroups. It will silently be ignored for all other
3877 types. You are not allowed to override the byte-order within a group
3878 that already has a byte-order modifier suffix.
3882 Real numbers (floats and doubles) are in the native machine format only;
3883 due to the multiplicity of floating formats around, and the lack of a
3884 standard "network" representation, no facility for interchange has been
3885 made. This means that packed floating point data written on one machine
3886 may not be readable on another - even if both use IEEE floating point
3887 arithmetic (as the endian-ness of the memory representation is not part
3888 of the IEEE spec). See also L<perlport>.
3890 If you know exactly what you're doing, you can use the C<E<gt>> or C<E<lt>>
3891 modifiers to force big- or little-endian byte-order on floating point values.
3893 Note that Perl uses doubles (or long doubles, if configured) internally for
3894 all numeric calculation, and converting from double into float and thence back
3895 to double again will lose precision (i.e., C<unpack("f", pack("f", $foo)>)
3896 will not in general equal $foo).
3900 Pack and unpack can operate in two modes, character mode (C<C0> mode) where
3901 the packed string is processed per character and UTF-8 mode (C<U0> mode)
3902 where the packed string is processed in its UTF-8-encoded Unicode form on
3903 a byte by byte basis. Character mode is the default unless the format string
3904 starts with an C<U>. You can switch mode at any moment with an explicit
3905 C<C0> or C<U0> in the format. A mode is in effect until the next mode switch
3906 or until the end of the ()-group in which it was entered.
3910 You must yourself do any alignment or padding by inserting for example
3911 enough C<'x'>es while packing. There is no way to pack() and unpack()
3912 could know where the characters are going to or coming from. Therefore
3913 C<pack> (and C<unpack>) handle their output and input as flat
3914 sequences of characters.
3918 A ()-group is a sub-TEMPLATE enclosed in parentheses. A group may
3919 take a repeat count, both as postfix, and for unpack() also via the C</>
3920 template character. Within each repetition of a group, positioning with
3921 C<@> starts again at 0. Therefore, the result of
3923 pack( '@1A((@2A)@3A)', 'a', 'b', 'c' )
3925 is the string "\0a\0\0bc".
3929 C<x> and C<X> accept C<!> modifier. In this case they act as
3930 alignment commands: they jump forward/back to the closest position
3931 aligned at a multiple of C<count> characters. For example, to pack() or
3932 unpack() C's C<struct {char c; double d; char cc[2]}> one may need to
3933 use the template C<W x![d] d W[2]>; this assumes that doubles must be
3934 aligned on the double's size.
3936 For alignment commands C<count> of 0 is equivalent to C<count> of 1;
3937 both result in no-ops.
3941 C<n>, C<N>, C<v> and C<V> accept the C<!> modifier. In this case they
3942 will represent signed 16-/32-bit integers in big-/little-endian order.
3943 This is only portable if all platforms sharing the packed data use the
3944 same binary representation for signed integers (e.g. all platforms are
3945 using two's complement representation).
3949 A comment in a TEMPLATE starts with C<#> and goes to the end of line.
3950 White space may be used to separate pack codes from each other, but
3951 modifiers and a repeat count must follow immediately.
3955 If TEMPLATE requires more arguments to pack() than actually given, pack()
3956 assumes additional C<""> arguments. If TEMPLATE requires fewer arguments
3957 to pack() than actually given, extra arguments are ignored.
3963 $foo = pack("WWWW",65,66,67,68);
3965 $foo = pack("W4",65,66,67,68);
3967 $foo = pack("W4",0x24b6,0x24b7,0x24b8,0x24b9);
3968 # same thing with Unicode circled letters.
3969 $foo = pack("U4",0x24b6,0x24b7,0x24b8,0x24b9);
3970 # same thing with Unicode circled letters. You don't get the UTF-8
3971 # bytes because the U at the start of the format caused a switch to
3972 # U0-mode, so the UTF-8 bytes get joined into characters
3973 $foo = pack("C0U4",0x24b6,0x24b7,0x24b8,0x24b9);
3974 # foo eq "\xe2\x92\xb6\xe2\x92\xb7\xe2\x92\xb8\xe2\x92\xb9"
3975 # This is the UTF-8 encoding of the string in the previous example
3977 $foo = pack("ccxxcc",65,66,67,68);
3980 # note: the above examples featuring "W" and "c" are true
3981 # only on ASCII and ASCII-derived systems such as ISO Latin 1
3982 # and UTF-8. In EBCDIC the first example would be
3983 # $foo = pack("WWWW",193,194,195,196);
3985 $foo = pack("s2",1,2);
3986 # "\1\0\2\0" on little-endian
3987 # "\0\1\0\2" on big-endian
3989 $foo = pack("a4","abcd","x","y","z");
3992 $foo = pack("aaaa","abcd","x","y","z");
3995 $foo = pack("a14","abcdefg");
3996 # "abcdefg\0\0\0\0\0\0\0"
3998 $foo = pack("i9pl", gmtime);
3999 # a real struct tm (on my system anyway)
4001 $utmp_template = "Z8 Z8 Z16 L";
4002 $utmp = pack($utmp_template, @utmp1);
4003 # a struct utmp (BSDish)
4005 @utmp2 = unpack($utmp_template, $utmp);
4006 # "@utmp1" eq "@utmp2"
4009 unpack("N", pack("B32", substr("0" x 32 . shift, -32)));
4012 $foo = pack('sx2l', 12, 34);
4013 # short 12, two zero bytes padding, long 34
4014 $bar = pack('s@4l', 12, 34);
4015 # short 12, zero fill to position 4, long 34
4017 $baz = pack('s.l', 12, 4, 34);
4018 # short 12, zero fill to position 4, long 34
4020 $foo = pack('nN', 42, 4711);
4021 # pack big-endian 16- and 32-bit unsigned integers
4022 $foo = pack('S>L>', 42, 4711);
4024 $foo = pack('s<l<', -42, 4711);
4025 # pack little-endian 16- and 32-bit signed integers
4026 $foo = pack('(sl)<', -42, 4711);
4029 The same template may generally also be used in unpack().
4031 =item package NAMESPACE
4032 X<package> X<module> X<namespace>
4036 Declares the compilation unit as being in the given namespace. The scope
4037 of the package declaration is from the declaration itself through the end
4038 of the enclosing block, file, or eval (the same as the C<my> operator).
4039 All further unqualified dynamic identifiers will be in this namespace.
4040 A package statement affects only dynamic variables--including those
4041 you've used C<local> on--but I<not> lexical variables, which are created
4042 with C<my>. Typically it would be the first declaration in a file to
4043 be included by the C<require> or C<use> operator. You can switch into a
4044 package in more than one place; it merely influences which symbol table
4045 is used by the compiler for the rest of that block. You can refer to
4046 variables and filehandles in other packages by prefixing the identifier
4047 with the package name and a double colon: C<$Package::Variable>.
4048 If the package name is null, the C<main> package as assumed. That is,
4049 C<$::sail> is equivalent to C<$main::sail> (as well as to C<$main'sail>,
4050 still seen in older code).
4052 If NAMESPACE is omitted, then there is no current package, and all
4053 identifiers must be fully qualified or lexicals. However, you are
4054 strongly advised not to make use of this feature. Its use can cause
4055 unexpected behaviour, even crashing some versions of Perl. It is
4056 deprecated, and will be removed from a future release.
4058 See L<perlmod/"Packages"> for more information about packages, modules,
4059 and classes. See L<perlsub> for other scoping issues.
4061 =item pipe READHANDLE,WRITEHANDLE
4064 Opens a pair of connected pipes like the corresponding system call.
4065 Note that if you set up a loop of piped processes, deadlock can occur
4066 unless you are very careful. In addition, note that Perl's pipes use
4067 IO buffering, so you may need to set C<$|> to flush your WRITEHANDLE
4068 after each command, depending on the application.
4070 See L<IPC::Open2>, L<IPC::Open3>, and L<perlipc/"Bidirectional Communication">
4071 for examples of such things.
4073 On systems that support a close-on-exec flag on files, the flag will be set
4074 for the newly opened file descriptors as determined by the value of $^F.
4082 Pops and returns the last value of the array, shortening the array by
4083 one element. Has an effect similar to
4087 If there are no elements in the array, returns the undefined value
4088 (although this may happen at other times as well). If ARRAY is
4089 omitted, pops the C<@ARGV> array in the main program, and the C<@_>
4090 array in subroutines, just like C<shift>.
4093 X<pos> X<match, position>
4097 Returns the offset of where the last C<m//g> search left off for the variable
4098 in question (C<$_> is used when the variable is not specified). Note that
4099 0 is a valid match offset. C<undef> indicates that the search position
4100 is reset (usually due to match failure, but can also be because no match has
4101 yet been performed on the scalar). C<pos> directly accesses the location used
4102 by the regexp engine to store the offset, so assigning to C<pos> will change
4103 that offset, and so will also influence the C<\G> zero-width assertion in
4104 regular expressions. Because a failed C<m//gc> match doesn't reset the offset,
4105 the return from C<pos> won't change either in this case. See L<perlre> and
4108 =item print FILEHANDLE LIST
4115 Prints a string or a list of strings. Returns true if successful.
4116 FILEHANDLE may be a scalar variable name, in which case the variable
4117 contains the name of or a reference to the filehandle, thus introducing
4118 one level of indirection. (NOTE: If FILEHANDLE is a variable and
4119 the next token is a term, it may be misinterpreted as an operator
4120 unless you interpose a C<+> or put parentheses around the arguments.)
4121 If FILEHANDLE is omitted, prints by default to standard output (or
4122 to the last selected output channel--see L</select>). If LIST is
4123 also omitted, prints C<$_> to the currently selected output channel.
4124 To set the default output channel to something other than STDOUT
4125 use the select operation. The current value of C<$,> (if any) is
4126 printed between each LIST item. The current value of C<$\> (if
4127 any) is printed after the entire LIST has been printed. Because
4128 print takes a LIST, anything in the LIST is evaluated in list
4129 context, and any subroutine that you call will have one or more of
4130 its expressions evaluated in list context. Also be careful not to
4131 follow the print keyword with a left parenthesis unless you want
4132 the corresponding right parenthesis to terminate the arguments to
4133 the print--interpose a C<+> or put parentheses around all the
4136 Note that if you're storing FILEHANDLEs in an array, or if you're using
4137 any other expression more complex than a scalar variable to retrieve it,
4138 you will have to use a block returning the filehandle value instead:
4140 print { $files[$i] } "stuff\n";
4141 print { $OK ? STDOUT : STDERR } "stuff\n";
4143 =item printf FILEHANDLE FORMAT, LIST
4146 =item printf FORMAT, LIST
4148 Equivalent to C<print FILEHANDLE sprintf(FORMAT, LIST)>, except that C<$\>
4149 (the output record separator) is not appended. The first argument
4150 of the list will be interpreted as the C<printf> format. See C<sprintf>
4151 for an explanation of the format argument. If C<use locale> is in effect,
4152 the character used for the decimal point in formatted real numbers is
4153 affected by the LC_NUMERIC locale. See L<perllocale>.
4155 Don't fall into the trap of using a C<printf> when a simple
4156 C<print> would do. The C<print> is more efficient and less
4159 =item prototype FUNCTION
4162 Returns the prototype of a function as a string (or C<undef> if the
4163 function has no prototype). FUNCTION is a reference to, or the name of,
4164 the function whose prototype you want to retrieve.
4166 If FUNCTION is a string starting with C<CORE::>, the rest is taken as a
4167 name for Perl builtin. If the builtin is not I<overridable> (such as
4168 C<qw//>) or its arguments cannot be expressed by a prototype (such as
4169 C<system>) returns C<undef> because the builtin does not really behave
4170 like a Perl function. Otherwise, the string describing the equivalent
4171 prototype is returned.
4173 =item push ARRAY,LIST
4176 Treats ARRAY as a stack, and pushes the values of LIST
4177 onto the end of ARRAY. The length of ARRAY increases by the length of
4178 LIST. Has the same effect as
4181 $ARRAY[++$#ARRAY] = $value;
4184 but is more efficient. Returns the number of elements in the array following
4185 the completed C<push>.
4197 Generalized quotes. See L<perlop/"Regexp Quote-Like Operators">.
4199 =item quotemeta EXPR
4200 X<quotemeta> X<metacharacter>
4204 Returns the value of EXPR with all non-"word"
4205 characters backslashed. (That is, all characters not matching
4206 C</[A-Za-z_0-9]/> will be preceded by a backslash in the
4207 returned string, regardless of any locale settings.)
4208 This is the internal function implementing
4209 the C<\Q> escape in double-quoted strings.
4211 If EXPR is omitted, uses C<$_>.
4218 Returns a random fractional number greater than or equal to C<0> and less
4219 than the value of EXPR. (EXPR should be positive.) If EXPR is
4220 omitted, the value C<1> is used. Currently EXPR with the value C<0> is
4221 also special-cased as C<1> - this has not been documented before perl 5.8.0
4222 and is subject to change in future versions of perl. Automatically calls
4223 C<srand> unless C<srand> has already been called. See also C<srand>.
4225 Apply C<int()> to the value returned by C<rand()> if you want random
4226 integers instead of random fractional numbers. For example,
4230 returns a random integer between C<0> and C<9>, inclusive.
4232 (Note: If your rand function consistently returns numbers that are too
4233 large or too small, then your version of Perl was probably compiled
4234 with the wrong number of RANDBITS.)
4236 =item read FILEHANDLE,SCALAR,LENGTH,OFFSET
4239 =item read FILEHANDLE,SCALAR,LENGTH
4241 Attempts to read LENGTH I<characters> of data into variable SCALAR
4242 from the specified FILEHANDLE. Returns the number of characters
4243 actually read, C<0> at end of file, or undef if there was an error (in
4244 the latter case C<$!> is also set). SCALAR will be grown or shrunk
4245 so that the last character actually read is the last character of the
4246 scalar after the read.
4248 An OFFSET may be specified to place the read data at some place in the
4249 string other than the beginning. A negative OFFSET specifies
4250 placement at that many characters counting backwards from the end of
4251 the string. A positive OFFSET greater than the length of SCALAR
4252 results in the string being padded to the required size with C<"\0">
4253 bytes before the result of the read is appended.
4255 The call is actually implemented in terms of either Perl's or system's
4256 fread() call. To get a true read(2) system call, see C<sysread>.
4258 Note the I<characters>: depending on the status of the filehandle,
4259 either (8-bit) bytes or characters are read. By default all
4260 filehandles operate on bytes, but for example if the filehandle has
4261 been opened with the C<:utf8> I/O layer (see L</open>, and the C<open>
4262 pragma, L<open>), the I/O will operate on UTF-8 encoded Unicode
4263 characters, not bytes. Similarly for the C<:encoding> pragma:
4264 in that case pretty much any characters can be read.
4266 =item readdir DIRHANDLE
4269 Returns the next directory entry for a directory opened by C<opendir>.
4270 If used in list context, returns all the rest of the entries in the
4271 directory. If there are no more entries, returns an undefined value in
4272 scalar context or a null list in list context.
4274 If you're planning to filetest the return values out of a C<readdir>, you'd
4275 better prepend the directory in question. Otherwise, because we didn't
4276 C<chdir> there, it would have been testing the wrong file.
4278 opendir(DIR, $some_dir) || die "can't opendir $some_dir: $!";
4279 @dots = grep { /^\./ && -f "$some_dir/$_" } readdir(DIR);
4283 X<readline> X<gets> X<fgets>
4285 Reads from the filehandle whose typeglob is contained in EXPR. In scalar
4286 context, each call reads and returns the next line, until end-of-file is
4287 reached, whereupon the subsequent call returns undef. In list context,
4288 reads until end-of-file is reached and returns a list of lines. Note that
4289 the notion of "line" used here is however you may have defined it
4290 with C<$/> or C<$INPUT_RECORD_SEPARATOR>). See L<perlvar/"$/">.
4292 When C<$/> is set to C<undef>, when readline() is in scalar
4293 context (i.e. file slurp mode), and when an empty file is read, it
4294 returns C<''> the first time, followed by C<undef> subsequently.
4296 This is the internal function implementing the C<< <EXPR> >>
4297 operator, but you can use it directly. The C<< <EXPR> >>
4298 operator is discussed in more detail in L<perlop/"I/O Operators">.
4301 $line = readline(*STDIN); # same thing
4303 If readline encounters an operating system error, C<$!> will be set with the
4304 corresponding error message. It can be helpful to check C<$!> when you are
4305 reading from filehandles you don't trust, such as a tty or a socket. The
4306 following example uses the operator form of C<readline>, and takes the necessary
4307 steps to ensure that C<readline> was successful.
4311 unless (defined( $line = <> )) {
4323 Returns the value of a symbolic link, if symbolic links are
4324 implemented. If not, gives a fatal error. If there is some system
4325 error, returns the undefined value and sets C<$!> (errno). If EXPR is
4326 omitted, uses C<$_>.
4331 EXPR is executed as a system command.
4332 The collected standard output of the command is returned.
4333 In scalar context, it comes back as a single (potentially
4334 multi-line) string. In list context, returns a list of lines
4335 (however you've defined lines with C<$/> or C<$INPUT_RECORD_SEPARATOR>).
4336 This is the internal function implementing the C<qx/EXPR/>
4337 operator, but you can use it directly. The C<qx/EXPR/>
4338 operator is discussed in more detail in L<perlop/"I/O Operators">.
4340 =item recv SOCKET,SCALAR,LENGTH,FLAGS
4343 Receives a message on a socket. Attempts to receive LENGTH characters
4344 of data into variable SCALAR from the specified SOCKET filehandle.
4345 SCALAR will be grown or shrunk to the length actually read. Takes the
4346 same flags as the system call of the same name. Returns the address
4347 of the sender if SOCKET's protocol supports this; returns an empty
4348 string otherwise. If there's an error, returns the undefined value.
4349 This call is actually implemented in terms of recvfrom(2) system call.
4350 See L<perlipc/"UDP: Message Passing"> for examples.
4352 Note the I<characters>: depending on the status of the socket, either
4353 (8-bit) bytes or characters are received. By default all sockets
4354 operate on bytes, but for example if the socket has been changed using
4355 binmode() to operate with the C<:utf8> I/O layer (see the C<open>
4356 pragma, L<open>), the I/O will operate on UTF-8 encoded Unicode
4357 characters, not bytes. Similarly for the C<:encoding> pragma:
4358 in that case pretty much any characters can be read.
4365 The C<redo> command restarts the loop block without evaluating the
4366 conditional again. The C<continue> block, if any, is not executed. If
4367 the LABEL is omitted, the command refers to the innermost enclosing
4368 loop. Programs that want to lie to themselves about what was just input
4369 normally use this command:
4371 # a simpleminded Pascal comment stripper
4372 # (warning: assumes no { or } in strings)
4373 LINE: while (<STDIN>) {
4374 while (s|({.*}.*){.*}|$1 |) {}
4379 if (/}/) { # end of comment?
4388 C<redo> cannot be used to retry a block which returns a value such as
4389 C<eval {}>, C<sub {}> or C<do {}>, and should not be used to exit
4390 a grep() or map() operation.
4392 Note that a block by itself is semantically identical to a loop
4393 that executes once. Thus C<redo> inside such a block will effectively
4394 turn it into a looping construct.
4396 See also L</continue> for an illustration of how C<last>, C<next>, and
4404 Returns a non-empty string if EXPR is a reference, the empty
4405 string otherwise. If EXPR
4406 is not specified, C<$_> will be used. The value returned depends on the
4407 type of thing the reference is a reference to.
4408 Builtin types include:
4418 If the referenced object has been blessed into a package, then that package
4419 name is returned instead. You can think of C<ref> as a C<typeof> operator.
4421 if (ref($r) eq "HASH") {
4422 print "r is a reference to a hash.\n";
4425 print "r is not a reference at all.\n";
4428 See also L<perlref>.
4430 =item rename OLDNAME,NEWNAME
4431 X<rename> X<move> X<mv> X<ren>
4433 Changes the name of a file; an existing file NEWNAME will be
4434 clobbered. Returns true for success, false otherwise.
4436 Behavior of this function varies wildly depending on your system
4437 implementation. For example, it will usually not work across file system
4438 boundaries, even though the system I<mv> command sometimes compensates
4439 for this. Other restrictions include whether it works on directories,
4440 open files, or pre-existing files. Check L<perlport> and either the
4441 rename(2) manpage or equivalent system documentation for details.
4443 =item require VERSION
4450 Demands a version of Perl specified by VERSION, or demands some semantics
4451 specified by EXPR or by C<$_> if EXPR is not supplied.
4453 VERSION may be either a numeric argument such as 5.006, which will be
4454 compared to C<$]>, or a literal of the form v5.6.1, which will be compared
4455 to C<$^V> (aka $PERL_VERSION). A fatal error is produced at run time if
4456 VERSION is greater than the version of the current Perl interpreter.
4457 Compare with L</use>, which can do a similar check at compile time.
4459 Specifying VERSION as a literal of the form v5.6.1 should generally be
4460 avoided, because it leads to misleading error messages under earlier
4461 versions of Perl that do not support this syntax. The equivalent numeric
4462 version should be used instead.
4464 require v5.6.1; # run time version check
4465 require 5.6.1; # ditto
4466 require 5.006_001; # ditto; preferred for backwards compatibility
4468 Otherwise, C<require> demands that a library file be included if it
4469 hasn't already been included. The file is included via the do-FILE
4470 mechanism, which is essentially just a variety of C<eval>. Has
4471 semantics similar to the following subroutine:
4474 my ($filename) = @_;
4475 if (exists $INC{$filename}) {
4476 return 1 if $INC{$filename};
4477 die "Compilation failed in require";
4479 my ($realfilename,$result);
4481 foreach $prefix (@INC) {
4482 $realfilename = "$prefix/$filename";
4483 if (-f $realfilename) {
4484 $INC{$filename} = $realfilename;
4485 $result = do $realfilename;
4489 die "Can't find $filename in \@INC";
4492 $INC{$filename} = undef;
4494 } elsif (!$result) {
4495 delete $INC{$filename};
4496 die "$filename did not return true value";
4502 Note that the file will not be included twice under the same specified
4505 The file must return true as the last statement to indicate
4506 successful execution of any initialization code, so it's customary to
4507 end such a file with C<1;> unless you're sure it'll return true
4508 otherwise. But it's better just to put the C<1;>, in case you add more
4511 If EXPR is a bareword, the require assumes a "F<.pm>" extension and
4512 replaces "F<::>" with "F</>" in the filename for you,
4513 to make it easy to load standard modules. This form of loading of
4514 modules does not risk altering your namespace.
4516 In other words, if you try this:
4518 require Foo::Bar; # a splendid bareword
4520 The require function will actually look for the "F<Foo/Bar.pm>" file in the
4521 directories specified in the C<@INC> array.
4523 But if you try this:
4525 $class = 'Foo::Bar';
4526 require $class; # $class is not a bareword
4528 require "Foo::Bar"; # not a bareword because of the ""
4530 The require function will look for the "F<Foo::Bar>" file in the @INC array and
4531 will complain about not finding "F<Foo::Bar>" there. In this case you can do:
4533 eval "require $class";
4535 Now that you understand how C<require> looks for files in the case of
4536 a bareword argument, there is a little extra functionality going on
4537 behind the scenes. Before C<require> looks for a "F<.pm>" extension,
4538 it will first look for a filename with a "F<.pmc>" extension. A file
4539 with this extension is assumed to be Perl bytecode generated by
4540 L<B::Bytecode|B::Bytecode>. If this file is found, and its modification
4541 time is newer than a coinciding "F<.pm>" non-compiled file, it will be
4542 loaded in place of that non-compiled file ending in a "F<.pm>" extension.
4544 You can also insert hooks into the import facility, by putting directly
4545 Perl code into the @INC array. There are three forms of hooks: subroutine
4546 references, array references and blessed objects.
4548 Subroutine references are the simplest case. When the inclusion system
4549 walks through @INC and encounters a subroutine, this subroutine gets
4550 called with two parameters, the first being a reference to itself, and the
4551 second the name of the file to be included (e.g. "F<Foo/Bar.pm>"). The
4552 subroutine should return C<undef> or a filehandle, from which the file to
4553 include will be read. If C<undef> is returned, C<require> will look at
4554 the remaining elements of @INC.
4556 If the hook is an array reference, its first element must be a subroutine
4557 reference. This subroutine is called as above, but the first parameter is
4558 the array reference. This enables to pass indirectly some arguments to
4561 In other words, you can write:
4563 push @INC, \&my_sub;
4565 my ($coderef, $filename) = @_; # $coderef is \&my_sub
4571 push @INC, [ \&my_sub, $x, $y, ... ];
4573 my ($arrayref, $filename) = @_;
4574 # Retrieve $x, $y, ...
4575 my @parameters = @$arrayref[1..$#$arrayref];
4579 If the hook is an object, it must provide an INC method that will be
4580 called as above, the first parameter being the object itself. (Note that
4581 you must fully qualify the sub's name, as it is always forced into package
4582 C<main>.) Here is a typical code layout:
4588 my ($self, $filename) = @_;
4592 # In the main program
4593 push @INC, new Foo(...);
4595 Note that these hooks are also permitted to set the %INC entry
4596 corresponding to the files they have loaded. See L<perlvar/%INC>.
4598 For a yet-more-powerful import facility, see L</use> and L<perlmod>.
4605 Generally used in a C<continue> block at the end of a loop to clear
4606 variables and reset C<??> searches so that they work again. The
4607 expression is interpreted as a list of single characters (hyphens
4608 allowed for ranges). All variables and arrays beginning with one of
4609 those letters are reset to their pristine state. If the expression is
4610 omitted, one-match searches (C<?pattern?>) are reset to match again. Resets
4611 only variables or searches in the current package. Always returns
4614 reset 'X'; # reset all X variables
4615 reset 'a-z'; # reset lower case variables
4616 reset; # just reset ?one-time? searches
4618 Resetting C<"A-Z"> is not recommended because you'll wipe out your
4619 C<@ARGV> and C<@INC> arrays and your C<%ENV> hash. Resets only package
4620 variables--lexical variables are unaffected, but they clean themselves
4621 up on scope exit anyway, so you'll probably want to use them instead.
4629 Returns from a subroutine, C<eval>, or C<do FILE> with the value
4630 given in EXPR. Evaluation of EXPR may be in list, scalar, or void
4631 context, depending on how the return value will be used, and the context
4632 may vary from one execution to the next (see C<wantarray>). If no EXPR
4633 is given, returns an empty list in list context, the undefined value in
4634 scalar context, and (of course) nothing at all in a void context.
4636 (Note that in the absence of an explicit C<return>, a subroutine, eval,
4637 or do FILE will automatically return the value of the last expression
4641 X<reverse> X<rev> X<invert>
4643 In list context, returns a list value consisting of the elements
4644 of LIST in the opposite order. In scalar context, concatenates the
4645 elements of LIST and returns a string value with all characters
4646 in the opposite order.
4648 print reverse <>; # line tac, last line first
4650 undef $/; # for efficiency of <>
4651 print scalar reverse <>; # character tac, last line tsrif
4653 Used without arguments in scalar context, reverse() reverses C<$_>.
4655 This operator is also handy for inverting a hash, although there are some
4656 caveats. If a value is duplicated in the original hash, only one of those
4657 can be represented as a key in the inverted hash. Also, this has to
4658 unwind one hash and build a whole new one, which may take some time
4659 on a large hash, such as from a DBM file.
4661 %by_name = reverse %by_address; # Invert the hash
4663 =item rewinddir DIRHANDLE
4666 Sets the current position to the beginning of the directory for the
4667 C<readdir> routine on DIRHANDLE.
4669 =item rindex STR,SUBSTR,POSITION
4672 =item rindex STR,SUBSTR
4674 Works just like index() except that it returns the position of the I<last>
4675 occurrence of SUBSTR in STR. If POSITION is specified, returns the
4676 last occurrence beginning at or before that position.
4678 =item rmdir FILENAME
4679 X<rmdir> X<rd> X<directory, remove>
4683 Deletes the directory specified by FILENAME if that directory is
4684 empty. If it succeeds it returns true, otherwise it returns false and
4685 sets C<$!> (errno). If FILENAME is omitted, uses C<$_>.
4689 The substitution operator. See L<perlop>.
4691 =item say FILEHANDLE LIST
4698 Just like C<print>, but implicitly appends a newline.
4699 C<say LIST> is simply an abbreviation for C<print LIST, "\n">,
4700 and C<say()> works just like C<print($_, "\n")>.
4702 That means that a call to say() appends any output record separator
4703 I<after> the added newline.
4705 This keyword is only available when the "say" feature is
4706 enabled: see L<feature>.
4709 X<scalar> X<context>
4711 Forces EXPR to be interpreted in scalar context and returns the value
4714 @counts = ( scalar @a, scalar @b, scalar @c );
4716 There is no equivalent operator to force an expression to
4717 be interpolated in list context because in practice, this is never
4718 needed. If you really wanted to do so, however, you could use
4719 the construction C<@{[ (some expression) ]}>, but usually a simple
4720 C<(some expression)> suffices.
4722 Because C<scalar> is unary operator, if you accidentally use for EXPR a
4723 parenthesized list, this behaves as a scalar comma expression, evaluating
4724 all but the last element in void context and returning the final element
4725 evaluated in scalar context. This is seldom what you want.
4727 The following single statement:
4729 print uc(scalar(&foo,$bar)),$baz;
4731 is the moral equivalent of these two:
4734 print(uc($bar),$baz);
4736 See L<perlop> for more details on unary operators and the comma operator.
4738 =item seek FILEHANDLE,POSITION,WHENCE
4739 X<seek> X<fseek> X<filehandle, position>
4741 Sets FILEHANDLE's position, just like the C<fseek> call of C<stdio>.
4742 FILEHANDLE may be an expression whose value gives the name of the
4743 filehandle. The values for WHENCE are C<0> to set the new position
4744 I<in bytes> to POSITION, C<1> to set it to the current position plus
4745 POSITION, and C<2> to set it to EOF plus POSITION (typically
4746 negative). For WHENCE you may use the constants C<SEEK_SET>,
4747 C<SEEK_CUR>, and C<SEEK_END> (start of the file, current position, end
4748 of the file) from the Fcntl module. Returns C<1> upon success, C<0>
4751 Note the I<in bytes>: even if the filehandle has been set to
4752 operate on characters (for example by using the C<:utf8> open
4753 layer), tell() will return byte offsets, not character offsets
4754 (because implementing that would render seek() and tell() rather slow).
4756 If you want to position file for C<sysread> or C<syswrite>, don't use
4757 C<seek>--buffering makes its effect on the file's system position
4758 unpredictable and non-portable. Use C<sysseek> instead.
4760 Due to the rules and rigors of ANSI C, on some systems you have to do a
4761 seek whenever you switch between reading and writing. Amongst other
4762 things, this may have the effect of calling stdio's clearerr(3).
4763 A WHENCE of C<1> (C<SEEK_CUR>) is useful for not moving the file position:
4767 This is also useful for applications emulating C<tail -f>. Once you hit
4768 EOF on your read, and then sleep for a while, you might have to stick in a
4769 seek() to reset things. The C<seek> doesn't change the current position,
4770 but it I<does> clear the end-of-file condition on the handle, so that the
4771 next C<< <FILE> >> makes Perl try again to read something. We hope.
4773 If that doesn't work (some IO implementations are particularly
4774 cantankerous), then you may need something more like this:
4777 for ($curpos = tell(FILE); $_ = <FILE>;
4778 $curpos = tell(FILE)) {
4779 # search for some stuff and put it into files
4781 sleep($for_a_while);
4782 seek(FILE, $curpos, 0);
4785 =item seekdir DIRHANDLE,POS
4788 Sets the current position for the C<readdir> routine on DIRHANDLE. POS
4789 must be a value returned by C<telldir>. C<seekdir> also has the same caveats
4790 about possible directory compaction as the corresponding system library
4793 =item select FILEHANDLE
4794 X<select> X<filehandle, default>
4798 Returns the currently selected filehandle. Sets the current default
4799 filehandle for output, if FILEHANDLE is supplied. This has two
4800 effects: first, a C<write> or a C<print> without a filehandle will
4801 default to this FILEHANDLE. Second, references to variables related to
4802 output will refer to this output channel. For example, if you have to
4803 set the top of form format for more than one output channel, you might
4811 FILEHANDLE may be an expression whose value gives the name of the
4812 actual filehandle. Thus:
4814 $oldfh = select(STDERR); $| = 1; select($oldfh);
4816 Some programmers may prefer to think of filehandles as objects with
4817 methods, preferring to write the last example as:
4820 STDERR->autoflush(1);
4822 =item select RBITS,WBITS,EBITS,TIMEOUT
4825 This calls the select(2) system call with the bit masks specified, which
4826 can be constructed using C<fileno> and C<vec>, along these lines:
4828 $rin = $win = $ein = '';
4829 vec($rin,fileno(STDIN),1) = 1;
4830 vec($win,fileno(STDOUT),1) = 1;
4833 If you want to select on many filehandles you might wish to write a
4837 my(@fhlist) = split(' ',$_[0]);
4840 vec($bits,fileno($_),1) = 1;
4844 $rin = fhbits('STDIN TTY SOCK');
4848 ($nfound,$timeleft) =
4849 select($rout=$rin, $wout=$win, $eout=$ein, $timeout);
4851 or to block until something becomes ready just do this
4853 $nfound = select($rout=$rin, $wout=$win, $eout=$ein, undef);
4855 Most systems do not bother to return anything useful in $timeleft, so
4856 calling select() in scalar context just returns $nfound.
4858 Any of the bit masks can also be undef. The timeout, if specified, is
4859 in seconds, which may be fractional. Note: not all implementations are
4860 capable of returning the $timeleft. If not, they always return
4861 $timeleft equal to the supplied $timeout.
4863 You can effect a sleep of 250 milliseconds this way:
4865 select(undef, undef, undef, 0.25);
4867 Note that whether C<select> gets restarted after signals (say, SIGALRM)
4868 is implementation-dependent. See also L<perlport> for notes on the
4869 portability of C<select>.
4871 On error, C<select> behaves like the select(2) system call : it returns
4874 Note: on some Unixes, the select(2) system call may report a socket file
4875 descriptor as "ready for reading", when actually no data is available,
4876 thus a subsequent read blocks. It can be avoided using always the
4877 O_NONBLOCK flag on the socket. See select(2) and fcntl(2) for further
4880 B<WARNING>: One should not attempt to mix buffered I/O (like C<read>
4881 or <FH>) with C<select>, except as permitted by POSIX, and even
4882 then only on POSIX systems. You have to use C<sysread> instead.
4884 =item semctl ID,SEMNUM,CMD,ARG
4887 Calls the System V IPC function C<semctl>. You'll probably have to say
4891 first to get the correct constant definitions. If CMD is IPC_STAT or
4892 GETALL, then ARG must be a variable that will hold the returned
4893 semid_ds structure or semaphore value array. Returns like C<ioctl>:
4894 the undefined value for error, "C<0 but true>" for zero, or the actual
4895 return value otherwise. The ARG must consist of a vector of native
4896 short integers, which may be created with C<pack("s!",(0)x$nsem)>.
4897 See also L<perlipc/"SysV IPC">, C<IPC::SysV>, C<IPC::Semaphore>
4900 =item semget KEY,NSEMS,FLAGS
4903 Calls the System V IPC function semget. Returns the semaphore id, or
4904 the undefined value if there is an error. See also
4905 L<perlipc/"SysV IPC">, C<IPC::SysV>, C<IPC::SysV::Semaphore>
4908 =item semop KEY,OPSTRING
4911 Calls the System V IPC function semop to perform semaphore operations
4912 such as signalling and waiting. OPSTRING must be a packed array of
4913 semop structures. Each semop structure can be generated with
4914 C<pack("s!3", $semnum, $semop, $semflag)>. The length of OPSTRING
4915 implies the number of semaphore operations. Returns true if
4916 successful, or false if there is an error. As an example, the
4917 following code waits on semaphore $semnum of semaphore id $semid:
4919 $semop = pack("s!3", $semnum, -1, 0);
4920 die "Semaphore trouble: $!\n" unless semop($semid, $semop);
4922 To signal the semaphore, replace C<-1> with C<1>. See also
4923 L<perlipc/"SysV IPC">, C<IPC::SysV>, and C<IPC::SysV::Semaphore>
4926 =item send SOCKET,MSG,FLAGS,TO
4929 =item send SOCKET,MSG,FLAGS
4931 Sends a message on a socket. Attempts to send the scalar MSG to the
4932 SOCKET filehandle. Takes the same flags as the system call of the
4933 same name. On unconnected sockets you must specify a destination to
4934 send TO, in which case it does a C C<sendto>. Returns the number of
4935 characters sent, or the undefined value if there is an error. The C
4936 system call sendmsg(2) is currently unimplemented. See
4937 L<perlipc/"UDP: Message Passing"> for examples.
4939 Note the I<characters>: depending on the status of the socket, either
4940 (8-bit) bytes or characters are sent. By default all sockets operate
4941 on bytes, but for example if the socket has been changed using
4942 binmode() to operate with the C<:utf8> I/O layer (see L</open>, or the
4943 C<open> pragma, L<open>), the I/O will operate on UTF-8 encoded
4944 Unicode characters, not bytes. Similarly for the C<:encoding> pragma:
4945 in that case pretty much any characters can be sent.
4947 =item setpgrp PID,PGRP
4950 Sets the current process group for the specified PID, C<0> for the current
4951 process. Will produce a fatal error if used on a machine that doesn't
4952 implement POSIX setpgid(2) or BSD setpgrp(2). If the arguments are omitted,
4953 it defaults to C<0,0>. Note that the BSD 4.2 version of C<setpgrp> does not
4954 accept any arguments, so only C<setpgrp(0,0)> is portable. See also
4957 =item setpriority WHICH,WHO,PRIORITY
4958 X<setpriority> X<priority> X<nice> X<renice>
4960 Sets the current priority for a process, a process group, or a user.
4961 (See setpriority(2).) Will produce a fatal error if used on a machine
4962 that doesn't implement setpriority(2).
4964 =item setsockopt SOCKET,LEVEL,OPTNAME,OPTVAL
4967 Sets the socket option requested. Returns undefined if there is an
4968 error. Use integer constants provided by the C<Socket> module for
4969 LEVEL and OPNAME. Values for LEVEL can also be obtained from
4970 getprotobyname. OPTVAL might either be a packed string or an integer.
4971 An integer OPTVAL is shorthand for pack("i", OPTVAL).
4973 An example disabling the Nagle's algorithm for a socket:
4975 use Socket qw(IPPROTO_TCP TCP_NODELAY);
4976 setsockopt($socket, IPPROTO_TCP, TCP_NODELAY, 1);
4983 Shifts the first value of the array off and returns it, shortening the
4984 array by 1 and moving everything down. If there are no elements in the
4985 array, returns the undefined value. If ARRAY is omitted, shifts the
4986 C<@_> array within the lexical scope of subroutines and formats, and the
4987 C<@ARGV> array outside of a subroutine and also within the lexical scopes
4988 established by the C<eval STRING>, C<BEGIN {}>, C<INIT {}>, C<CHECK {}>
4989 and C<END {}> constructs.
4991 See also C<unshift>, C<push>, and C<pop>. C<shift> and C<unshift> do the
4992 same thing to the left end of an array that C<pop> and C<push> do to the
4995 =item shmctl ID,CMD,ARG
4998 Calls the System V IPC function shmctl. You'll probably have to say
5002 first to get the correct constant definitions. If CMD is C<IPC_STAT>,
5003 then ARG must be a variable that will hold the returned C<shmid_ds>
5004 structure. Returns like ioctl: the undefined value for error, "C<0> but
5005 true" for zero, or the actual return value otherwise.
5006 See also L<perlipc/"SysV IPC"> and C<IPC::SysV> documentation.
5008 =item shmget KEY,SIZE,FLAGS
5011 Calls the System V IPC function shmget. Returns the shared memory
5012 segment id, or the undefined value if there is an error.
5013 See also L<perlipc/"SysV IPC"> and C<IPC::SysV> documentation.
5015 =item shmread ID,VAR,POS,SIZE
5019 =item shmwrite ID,STRING,POS,SIZE
5021 Reads or writes the System V shared memory segment ID starting at
5022 position POS for size SIZE by attaching to it, copying in/out, and
5023 detaching from it. When reading, VAR must be a variable that will
5024 hold the data read. When writing, if STRING is too long, only SIZE
5025 bytes are used; if STRING is too short, nulls are written to fill out
5026 SIZE bytes. Return true if successful, or false if there is an error.
5027 shmread() taints the variable. See also L<perlipc/"SysV IPC">,
5028 C<IPC::SysV> documentation, and the C<IPC::Shareable> module from CPAN.
5030 =item shutdown SOCKET,HOW
5033 Shuts down a socket connection in the manner indicated by HOW, which
5034 has the same interpretation as in the system call of the same name.
5036 shutdown(SOCKET, 0); # I/we have stopped reading data
5037 shutdown(SOCKET, 1); # I/we have stopped writing data
5038 shutdown(SOCKET, 2); # I/we have stopped using this socket
5040 This is useful with sockets when you want to tell the other
5041 side you're done writing but not done reading, or vice versa.
5042 It's also a more insistent form of close because it also
5043 disables the file descriptor in any forked copies in other
5047 X<sin> X<sine> X<asin> X<arcsine>
5051 Returns the sine of EXPR (expressed in radians). If EXPR is omitted,
5052 returns sine of C<$_>.
5054 For the inverse sine operation, you may use the C<Math::Trig::asin>
5055 function, or use this relation:
5057 sub asin { atan2($_[0], sqrt(1 - $_[0] * $_[0])) }
5064 Causes the script to sleep for EXPR seconds, or forever if no EXPR.
5065 May be interrupted if the process receives a signal such as C<SIGALRM>.
5066 Returns the number of seconds actually slept. You probably cannot
5067 mix C<alarm> and C<sleep> calls, because C<sleep> is often implemented
5070 On some older systems, it may sleep up to a full second less than what
5071 you requested, depending on how it counts seconds. Most modern systems
5072 always sleep the full amount. They may appear to sleep longer than that,
5073 however, because your process might not be scheduled right away in a
5074 busy multitasking system.
5076 For delays of finer granularity than one second, you may use Perl's
5077 C<syscall> interface to access setitimer(2) if your system supports
5078 it, or else see L</select> above. The Time::HiRes module (from CPAN,
5079 and starting from Perl 5.8 part of the standard distribution) may also
5082 See also the POSIX module's C<pause> function.
5084 =item socket SOCKET,DOMAIN,TYPE,PROTOCOL
5087 Opens a socket of the specified kind and attaches it to filehandle
5088 SOCKET. DOMAIN, TYPE, and PROTOCOL are specified the same as for
5089 the system call of the same name. You should C<use Socket> first
5090 to get the proper definitions imported. See the examples in
5091 L<perlipc/"Sockets: Client/Server Communication">.
5093 On systems that support a close-on-exec flag on files, the flag will
5094 be set for the newly opened file descriptor, as determined by the
5095 value of $^F. See L<perlvar/$^F>.
5097 =item socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL
5100 Creates an unnamed pair of sockets in the specified domain, of the
5101 specified type. DOMAIN, TYPE, and PROTOCOL are specified the same as
5102 for the system call of the same name. If unimplemented, yields a fatal
5103 error. Returns true if successful.
5105 On systems that support a close-on-exec flag on files, the flag will
5106 be set for the newly opened file descriptors, as determined by the value
5107 of $^F. See L<perlvar/$^F>.
5109 Some systems defined C<pipe> in terms of C<socketpair>, in which a call
5110 to C<pipe(Rdr, Wtr)> is essentially:
5113 socketpair(Rdr, Wtr, AF_UNIX, SOCK_STREAM, PF_UNSPEC);
5114 shutdown(Rdr, 1); # no more writing for reader
5115 shutdown(Wtr, 0); # no more reading for writer
5117 See L<perlipc> for an example of socketpair use. Perl 5.8 and later will
5118 emulate socketpair using IP sockets to localhost if your system implements
5119 sockets but not socketpair.
5121 =item sort SUBNAME LIST
5122 X<sort> X<qsort> X<quicksort> X<mergesort>
5124 =item sort BLOCK LIST
5128 In list context, this sorts the LIST and returns the sorted list value.
5129 In scalar context, the behaviour of C<sort()> is undefined.
5131 If SUBNAME or BLOCK is omitted, C<sort>s in standard string comparison
5132 order. If SUBNAME is specified, it gives the name of a subroutine
5133 that returns an integer less than, equal to, or greater than C<0>,
5134 depending on how the elements of the list are to be ordered. (The C<<
5135 <=> >> and C<cmp> operators are extremely useful in such routines.)
5136 SUBNAME may be a scalar variable name (unsubscripted), in which case
5137 the value provides the name of (or a reference to) the actual
5138 subroutine to use. In place of a SUBNAME, you can provide a BLOCK as
5139 an anonymous, in-line sort subroutine.
5141 If the subroutine's prototype is C<($$)>, the elements to be compared
5142 are passed by reference in C<@_>, as for a normal subroutine. This is
5143 slower than unprototyped subroutines, where the elements to be
5144 compared are passed into the subroutine
5145 as the package global variables $a and $b (see example below). Note that
5146 in the latter case, it is usually counter-productive to declare $a and
5149 The values to be compared are always passed by reference and should not
5152 You also cannot exit out of the sort block or subroutine using any of the
5153 loop control operators described in L<perlsyn> or with C<goto>.
5155 When C<use locale> is in effect, C<sort LIST> sorts LIST according to the
5156 current collation locale. See L<perllocale>.
5158 sort() returns aliases into the original list, much as a for loop's index
5159 variable aliases the list elements. That is, modifying an element of a
5160 list returned by sort() (for example, in a C<foreach>, C<map> or C<grep>)
5161 actually modifies the element in the original list. This is usually
5162 something to be avoided when writing clear code.
5164 Perl 5.6 and earlier used a quicksort algorithm to implement sort.
5165 That algorithm was not stable, and I<could> go quadratic. (A I<stable> sort
5166 preserves the input order of elements that compare equal. Although
5167 quicksort's run time is O(NlogN) when averaged over all arrays of
5168 length N, the time can be O(N**2), I<quadratic> behavior, for some
5169 inputs.) In 5.7, the quicksort implementation was replaced with
5170 a stable mergesort algorithm whose worst-case behavior is O(NlogN).
5171 But benchmarks indicated that for some inputs, on some platforms,
5172 the original quicksort was faster. 5.8 has a sort pragma for
5173 limited control of the sort. Its rather blunt control of the
5174 underlying algorithm may not persist into future Perls, but the
5175 ability to characterize the input or output in implementation
5176 independent ways quite probably will. See L<sort>.
5181 @articles = sort @files;
5183 # same thing, but with explicit sort routine
5184 @articles = sort {$a cmp $b} @files;
5186 # now case-insensitively
5187 @articles = sort {uc($a) cmp uc($b)} @files;
5189 # same thing in reversed order
5190 @articles = sort {$b cmp $a} @files;
5192 # sort numerically ascending
5193 @articles = sort {$a <=> $b} @files;
5195 # sort numerically descending
5196 @articles = sort {$b <=> $a} @files;
5198 # this sorts the %age hash by value instead of key
5199 # using an in-line function
5200 @eldest = sort { $age{$b} <=> $age{$a} } keys %age;
5202 # sort using explicit subroutine name
5204 $age{$a} <=> $age{$b}; # presuming numeric
5206 @sortedclass = sort byage @class;
5208 sub backwards { $b cmp $a }
5209 @harry = qw(dog cat x Cain Abel);
5210 @george = qw(gone chased yz Punished Axed);
5212 # prints AbelCaincatdogx
5213 print sort backwards @harry;
5214 # prints xdogcatCainAbel
5215 print sort @george, 'to', @harry;
5216 # prints AbelAxedCainPunishedcatchaseddoggonetoxyz
5218 # inefficiently sort by descending numeric compare using
5219 # the first integer after the first = sign, or the
5220 # whole record case-insensitively otherwise
5223 ($b =~ /=(\d+)/)[0] <=> ($a =~ /=(\d+)/)[0]
5228 # same thing, but much more efficiently;
5229 # we'll build auxiliary indices instead
5233 push @nums, /=(\d+)/;
5238 $nums[$b] <=> $nums[$a]
5240 $caps[$a] cmp $caps[$b]
5244 # same thing, but without any temps
5245 @new = map { $_->[0] }
5246 sort { $b->[1] <=> $a->[1]
5249 } map { [$_, /=(\d+)/, uc($_)] } @old;
5251 # using a prototype allows you to use any comparison subroutine
5252 # as a sort subroutine (including other package's subroutines)
5254 sub backwards ($$) { $_[1] cmp $_[0]; } # $a and $b are not set here
5257 @new = sort other::backwards @old;
5259 # guarantee stability, regardless of algorithm
5261 @new = sort { substr($a, 3, 5) cmp substr($b, 3, 5) } @old;
5263 # force use of mergesort (not portable outside Perl 5.8)
5264 use sort '_mergesort'; # note discouraging _
5265 @new = sort { substr($a, 3, 5) cmp substr($b, 3, 5) } @old;
5267 If you're using strict, you I<must not> declare $a
5268 and $b as lexicals. They are package globals. That means
5269 if you're in the C<main> package and type
5271 @articles = sort {$b <=> $a} @files;
5273 then C<$a> and C<$b> are C<$main::a> and C<$main::b> (or C<$::a> and C<$::b>),
5274 but if you're in the C<FooPack> package, it's the same as typing
5276 @articles = sort {$FooPack::b <=> $FooPack::a} @files;
5278 The comparison function is required to behave. If it returns
5279 inconsistent results (sometimes saying C<$x[1]> is less than C<$x[2]> and
5280 sometimes saying the opposite, for example) the results are not
5283 Because C<< <=> >> returns C<undef> when either operand is C<NaN>
5284 (not-a-number), and because C<sort> will trigger a fatal error unless the
5285 result of a comparison is defined, when sorting with a comparison function
5286 like C<< $a <=> $b >>, be careful about lists that might contain a C<NaN>.
5287 The following example takes advantage of the fact that C<NaN != NaN> to
5288 eliminate any C<NaN>s from the input.
5290 @result = sort { $a <=> $b } grep { $_ == $_ } @input;
5292 =item splice ARRAY,OFFSET,LENGTH,LIST
5295 =item splice ARRAY,OFFSET,LENGTH
5297 =item splice ARRAY,OFFSET
5301 Removes the elements designated by OFFSET and LENGTH from an array, and
5302 replaces them with the elements of LIST, if any. In list context,
5303 returns the elements removed from the array. In scalar context,
5304 returns the last element removed, or C<undef> if no elements are
5305 removed. The array grows or shrinks as necessary.
5306 If OFFSET is negative then it starts that far from the end of the array.
5307 If LENGTH is omitted, removes everything from OFFSET onward.
5308 If LENGTH is negative, removes the elements from OFFSET onward
5309 except for -LENGTH elements at the end of the array.
5310 If both OFFSET and LENGTH are omitted, removes everything. If OFFSET is
5311 past the end of the array, perl issues a warning, and splices at the
5314 The following equivalences hold (assuming C<< $[ == 0 and $#a >= $i >> )
5316 push(@a,$x,$y) splice(@a,@a,0,$x,$y)
5317 pop(@a) splice(@a,-1)
5318 shift(@a) splice(@a,0,1)
5319 unshift(@a,$x,$y) splice(@a,0,0,$x,$y)
5320 $a[$i] = $y splice(@a,$i,1,$y)
5322 Example, assuming array lengths are passed before arrays:
5324 sub aeq { # compare two list values
5325 my(@a) = splice(@_,0,shift);
5326 my(@b) = splice(@_,0,shift);
5327 return 0 unless @a == @b; # same len?
5329 return 0 if pop(@a) ne pop(@b);
5333 if (&aeq($len,@foo[1..$len],0+@bar,@bar)) { ... }
5335 =item split /PATTERN/,EXPR,LIMIT
5338 =item split /PATTERN/,EXPR
5340 =item split /PATTERN/
5344 Splits the string EXPR into a list of strings and returns that list. By
5345 default, empty leading fields are preserved, and empty trailing ones are
5346 deleted. (If all fields are empty, they are considered to be trailing.)
5348 In scalar context, returns the number of fields found and splits into
5349 the C<@_> array. Use of split in scalar context is deprecated, however,
5350 because it clobbers your subroutine arguments.
5352 If EXPR is omitted, splits the C<$_> string. If PATTERN is also omitted,
5353 splits on whitespace (after skipping any leading whitespace). Anything
5354 matching PATTERN is taken to be a delimiter separating the fields. (Note
5355 that the delimiter may be longer than one character.)
5357 If LIMIT is specified and positive, it represents the maximum number
5358 of fields the EXPR will be split into, though the actual number of
5359 fields returned depends on the number of times PATTERN matches within
5360 EXPR. If LIMIT is unspecified or zero, trailing null fields are
5361 stripped (which potential users of C<pop> would do well to remember).
5362 If LIMIT is negative, it is treated as if an arbitrarily large LIMIT
5363 had been specified. Note that splitting an EXPR that evaluates to the
5364 empty string always returns the empty list, regardless of the LIMIT
5367 A pattern matching the null string (not to be confused with
5368 a null pattern C<//>, which is just one member of the set of patterns
5369 matching a null string) will split the value of EXPR into separate
5370 characters at each point it matches that way. For example:
5372 print join(':', split(/ */, 'hi there'));
5374 produces the output 'h:i:t:h:e:r:e'.
5376 As a special case for C<split>, using the empty pattern C<//> specifically
5377 matches only the null string, and is not be confused with the regular use
5378 of C<//> to mean "the last successful pattern match". So, for C<split>,
5381 print join(':', split(//, 'hi there'));
5383 produces the output 'h:i: :t:h:e:r:e'.
5385 Empty leading (or trailing) fields are produced when there are positive
5386 width matches at the beginning (or end) of the string; a zero-width match
5387 at the beginning (or end) of the string does not produce an empty field.
5390 print join(':', split(/(?=\w)/, 'hi there!'));
5392 produces the output 'h:i :t:h:e:r:e!'.
5394 The LIMIT parameter can be used to split a line partially
5396 ($login, $passwd, $remainder) = split(/:/, $_, 3);
5398 When assigning to a list, if LIMIT is omitted, or zero, Perl supplies
5399 a LIMIT one larger than the number of variables in the list, to avoid
5400 unnecessary work. For the list above LIMIT would have been 4 by
5401 default. In time critical applications it behooves you not to split
5402 into more fields than you really need.
5404 If the PATTERN contains parentheses, additional list elements are
5405 created from each matching substring in the delimiter.
5407 split(/([,-])/, "1-10,20", 3);
5409 produces the list value
5411 (1, '-', 10, ',', 20)
5413 If you had the entire header of a normal Unix email message in $header,
5414 you could split it up into fields and their values this way:
5416 $header =~ s/\n\s+/ /g; # fix continuation lines
5417 %hdrs = (UNIX_FROM => split /^(\S*?):\s*/m, $header);
5419 The pattern C</PATTERN/> may be replaced with an expression to specify
5420 patterns that vary at runtime. (To do runtime compilation only once,
5421 use C</$variable/o>.)
5423 As a special case, specifying a PATTERN of space (S<C<' '>>) will split on
5424 white space just as C<split> with no arguments does. Thus, S<C<split(' ')>> can
5425 be used to emulate B<awk>'s default behavior, whereas S<C<split(/ /)>>
5426 will give you as many null initial fields as there are leading spaces.
5427 A C<split> on C</\s+/> is like a S<C<split(' ')>> except that any leading
5428 whitespace produces a null first field. A C<split> with no arguments
5429 really does a S<C<split(' ', $_)>> internally.
5431 A PATTERN of C</^/> is treated as if it were C</^/m>, since it isn't
5436 open(PASSWD, '/etc/passwd');
5439 ($login, $passwd, $uid, $gid,
5440 $gcos, $home, $shell) = split(/:/);
5444 As with regular pattern matching, any capturing parentheses that are not
5445 matched in a C<split()> will be set to C<undef> when returned:
5447 @fields = split /(A)|B/, "1A2B3";
5448 # @fields is (1, 'A', 2, undef, 3)
5450 =item sprintf FORMAT, LIST
5453 Returns a string formatted by the usual C<printf> conventions of the C
5454 library function C<sprintf>. See below for more details
5455 and see L<sprintf(3)> or L<printf(3)> on your system for an explanation of
5456 the general principles.
5460 # Format number with up to 8 leading zeroes
5461 $result = sprintf("%08d", $number);
5463 # Round number to 3 digits after decimal point
5464 $rounded = sprintf("%.3f", $number);
5466 Perl does its own C<sprintf> formatting--it emulates the C
5467 function C<sprintf>, but it doesn't use it (except for floating-point
5468 numbers, and even then only the standard modifiers are allowed). As a
5469 result, any non-standard extensions in your local C<sprintf> are not
5470 available from Perl.
5472 Unlike C<printf>, C<sprintf> does not do what you probably mean when you
5473 pass it an array as your first argument. The array is given scalar context,
5474 and instead of using the 0th element of the array as the format, Perl will
5475 use the count of elements in the array as the format, which is almost never
5478 Perl's C<sprintf> permits the following universally-known conversions:
5481 %c a character with the given number
5483 %d a signed integer, in decimal
5484 %u an unsigned integer, in decimal
5485 %o an unsigned integer, in octal
5486 %x an unsigned integer, in hexadecimal
5487 %e a floating-point number, in scientific notation
5488 %f a floating-point number, in fixed decimal notation
5489 %g a floating-point number, in %e or %f notation
5491 In addition, Perl permits the following widely-supported conversions:
5493 %X like %x, but using upper-case letters
5494 %E like %e, but using an upper-case "E"
5495 %G like %g, but with an upper-case "E" (if applicable)
5496 %b an unsigned integer, in binary
5497 %p a pointer (outputs the Perl value's address in hexadecimal)
5498 %n special: *stores* the number of characters output so far
5499 into the next variable in the parameter list
5501 Finally, for backward (and we do mean "backward") compatibility, Perl
5502 permits these unnecessary but widely-supported conversions:
5505 %D a synonym for %ld
5506 %U a synonym for %lu
5507 %O a synonym for %lo
5510 Note that the number of exponent digits in the scientific notation produced
5511 by C<%e>, C<%E>, C<%g> and C<%G> for numbers with the modulus of the
5512 exponent less than 100 is system-dependent: it may be three or less
5513 (zero-padded as necessary). In other words, 1.23 times ten to the
5514 99th may be either "1.23e99" or "1.23e099".
5516 Between the C<%> and the format letter, you may specify a number of
5517 additional attributes controlling the interpretation of the format.
5518 In order, these are:
5522 =item format parameter index
5524 An explicit format parameter index, such as C<2$>. By default sprintf
5525 will format the next unused argument in the list, but this allows you
5526 to take the arguments out of order, e.g.:
5528 printf '%2$d %1$d', 12, 34; # prints "34 12"
5529 printf '%3$d %d %1$d', 1, 2, 3; # prints "3 1 1"
5534 space prefix positive number with a space
5535 + prefix positive number with a plus sign
5536 - left-justify within the field
5537 0 use zeros, not spaces, to right-justify
5538 # prefix non-zero octal with "0", non-zero hex with "0x",
5539 non-zero binary with "0b"
5543 printf '<% d>', 12; # prints "< 12>"
5544 printf '<%+d>', 12; # prints "<+12>"
5545 printf '<%6s>', 12; # prints "< 12>"
5546 printf '<%-6s>', 12; # prints "<12 >"
5547 printf '<%06s>', 12; # prints "<000012>"
5548 printf '<%#x>', 12; # prints "<0xc>"
5552 This flag tells perl to interpret the supplied string as a vector of
5553 integers, one for each character in the string. Perl applies the format to
5554 each integer in turn, then joins the resulting strings with a separator (a
5555 dot C<.> by default). This can be useful for displaying ordinal values of
5556 characters in arbitrary strings:
5558 printf "%vd", "AB\x{100}"; # prints "65.66.256"
5559 printf "version is v%vd\n", $^V; # Perl's version
5561 Put an asterisk C<*> before the C<v> to override the string to
5562 use to separate the numbers:
5564 printf "address is %*vX\n", ":", $addr; # IPv6 address
5565 printf "bits are %0*v8b\n", " ", $bits; # random bitstring
5567 You can also explicitly specify the argument number to use for
5568 the join string using e.g. C<*2$v>:
5570 printf '%*4$vX %*4$vX %*4$vX', @addr[1..3], ":"; # 3 IPv6 addresses
5572 =item (minimum) width
5574 Arguments are usually formatted to be only as wide as required to
5575 display the given value. You can override the width by putting
5576 a number here, or get the width from the next argument (with C<*>)
5577 or from a specified argument (with e.g. C<*2$>):
5579 printf '<%s>', "a"; # prints "<a>"
5580 printf '<%6s>', "a"; # prints "< a>"
5581 printf '<%*s>', 6, "a"; # prints "< a>"
5582 printf '<%*2$s>', "a", 6; # prints "< a>"
5583 printf '<%2s>', "long"; # prints "<long>" (does not truncate)
5585 If a field width obtained through C<*> is negative, it has the same
5586 effect as the C<-> flag: left-justification.
5588 =item precision, or maximum width
5591 You can specify a precision (for numeric conversions) or a maximum
5592 width (for string conversions) by specifying a C<.> followed by a number.
5593 For floating point formats, with the exception of 'g' and 'G', this specifies
5594 the number of decimal places to show (the default being 6), e.g.:
5596 # these examples are subject to system-specific variation
5597 printf '<%f>', 1; # prints "<1.000000>"
5598 printf '<%.1f>', 1; # prints "<1.0>"
5599 printf '<%.0f>', 1; # prints "<1>"
5600 printf '<%e>', 10; # prints "<1.000000e+01>"
5601 printf '<%.1e>', 10; # prints "<1.0e+01>"
5603 For 'g' and 'G', this specifies the maximum number of digits to show,
5604 including prior to the decimal point as well as after it, e.g.:
5606 # these examples are subject to system-specific variation
5607 printf '<%g>', 1; # prints "<1>"
5608 printf '<%.10g>', 1; # prints "<1>"
5609 printf '<%g>', 100; # prints "<100>"
5610 printf '<%.1g>', 100; # prints "<1e+02>"
5611 printf '<%.2g>', 100.01; # prints "<1e+02>"
5612 printf '<%.5g>', 100.01; # prints "<100.01>"
5613 printf '<%.4g>', 100.01; # prints "<100>"
5615 For integer conversions, specifying a precision implies that the
5616 output of the number itself should be zero-padded to this width:
5618 printf '<%.6x>', 1; # prints "<000001>"
5619 printf '<%#.6x>', 1; # prints "<0x000001>"
5620 printf '<%-10.6x>', 1; # prints "<000001 >"
5622 For string conversions, specifying a precision truncates the string
5623 to fit in the specified width:
5625 printf '<%.5s>', "truncated"; # prints "<trunc>"
5626 printf '<%10.5s>', "truncated"; # prints "< trunc>"
5628 You can also get the precision from the next argument using C<.*>:
5630 printf '<%.6x>', 1; # prints "<000001>"
5631 printf '<%.*x>', 6, 1; # prints "<000001>"
5633 You cannot currently get the precision from a specified number,
5634 but it is intended that this will be possible in the future using
5637 printf '<%.*2$x>', 1, 6; # INVALID, but in future will print "<000001>"
5641 For numeric conversions, you can specify the size to interpret the
5642 number as using C<l>, C<h>, C<V>, C<q>, C<L>, or C<ll>. For integer
5643 conversions (C<d u o x X b i D U O>), numbers are usually assumed to be
5644 whatever the default integer size is on your platform (usually 32 or 64
5645 bits), but you can override this to use instead one of the standard C types,
5646 as supported by the compiler used to build Perl:
5648 l interpret integer as C type "long" or "unsigned long"
5649 h interpret integer as C type "short" or "unsigned short"
5650 q, L or ll interpret integer as C type "long long", "unsigned long long".
5651 or "quads" (typically 64-bit integers)
5653 The last will produce errors if Perl does not understand "quads" in your
5654 installation. (This requires that either the platform natively supports quads
5655 or Perl was specifically compiled to support quads.) You can find out
5656 whether your Perl supports quads via L<Config>:
5659 ($Config{use64bitint} eq 'define' || $Config{longsize} >= 8) &&
5662 For floating point conversions (C<e f g E F G>), numbers are usually assumed
5663 to be the default floating point size on your platform (double or long double),
5664 but you can force 'long double' with C<q>, C<L>, or C<ll> if your
5665 platform supports them. You can find out whether your Perl supports long
5666 doubles via L<Config>:
5669 $Config{d_longdbl} eq 'define' && print "long doubles\n";
5671 You can find out whether Perl considers 'long double' to be the default
5672 floating point size to use on your platform via L<Config>:
5675 ($Config{uselongdouble} eq 'define') &&
5676 print "long doubles by default\n";
5678 It can also be the case that long doubles and doubles are the same thing:
5681 ($Config{doublesize} == $Config{longdblsize}) &&
5682 print "doubles are long doubles\n";
5684 The size specifier C<V> has no effect for Perl code, but it is supported
5685 for compatibility with XS code; it means 'use the standard size for
5686 a Perl integer (or floating-point number)', which is already the
5687 default for Perl code.
5689 =item order of arguments
5691 Normally, sprintf takes the next unused argument as the value to
5692 format for each format specification. If the format specification
5693 uses C<*> to require additional arguments, these are consumed from
5694 the argument list in the order in which they appear in the format
5695 specification I<before> the value to format. Where an argument is
5696 specified using an explicit index, this does not affect the normal
5697 order for the arguments (even when the explicitly specified index
5698 would have been the next argument in any case).
5702 printf '<%*.*s>', $a, $b, $c;
5704 would use C<$a> for the width, C<$b> for the precision and C<$c>
5705 as the value to format, while:
5707 print '<%*1$.*s>', $a, $b;
5709 would use C<$a> for the width and the precision, and C<$b> as the
5712 Here are some more examples - beware that when using an explicit
5713 index, the C<$> may need to be escaped:
5715 printf "%2\$d %d\n", 12, 34; # will print "34 12\n"
5716 printf "%2\$d %d %d\n", 12, 34; # will print "34 12 34\n"
5717 printf "%3\$d %d %d\n", 12, 34, 56; # will print "56 12 34\n"
5718 printf "%2\$*3\$d %d\n", 12, 34, 3; # will print " 34 12\n"
5722 If C<use locale> is in effect, the character used for the decimal
5723 point in formatted real numbers is affected by the LC_NUMERIC locale.
5727 X<sqrt> X<root> X<square root>
5731 Return the square root of EXPR. If EXPR is omitted, returns square
5732 root of C<$_>. Only works on non-negative operands, unless you've
5733 loaded the standard Math::Complex module.
5736 print sqrt(-2); # prints 1.4142135623731i
5739 X<srand> X<seed> X<randseed>
5743 Sets the random number seed for the C<rand> operator.
5745 The point of the function is to "seed" the C<rand> function so that
5746 C<rand> can produce a different sequence each time you run your
5749 If srand() is not called explicitly, it is called implicitly at the
5750 first use of the C<rand> operator. However, this was not the case in
5751 versions of Perl before 5.004, so if your script will run under older
5752 Perl versions, it should call C<srand>.
5754 Most programs won't even call srand() at all, except those that
5755 need a cryptographically-strong starting point rather than the
5756 generally acceptable default, which is based on time of day,
5757 process ID, and memory allocation, or the F</dev/urandom> device,
5760 You can call srand($seed) with the same $seed to reproduce the
5761 I<same> sequence from rand(), but this is usually reserved for
5762 generating predictable results for testing or debugging.
5763 Otherwise, don't call srand() more than once in your program.
5765 Do B<not> call srand() (i.e. without an argument) more than once in
5766 a script. The internal state of the random number generator should
5767 contain more entropy than can be provided by any seed, so calling
5768 srand() again actually I<loses> randomness.
5770 Most implementations of C<srand> take an integer and will silently
5771 truncate decimal numbers. This means C<srand(42)> will usually
5772 produce the same results as C<srand(42.1)>. To be safe, always pass
5773 C<srand> an integer.
5775 In versions of Perl prior to 5.004 the default seed was just the
5776 current C<time>. This isn't a particularly good seed, so many old
5777 programs supply their own seed value (often C<time ^ $$> or C<time ^
5778 ($$ + ($$ << 15))>), but that isn't necessary any more.
5780 For cryptographic purposes, however, you need something much more random
5781 than the default seed. Checksumming the compressed output of one or more
5782 rapidly changing operating system status programs is the usual method. For
5785 srand (time ^ $$ ^ unpack "%L*", `ps axww | gzip`);
5787 If you're particularly concerned with this, see the C<Math::TrulyRandom>
5790 Frequently called programs (like CGI scripts) that simply use
5794 for a seed can fall prey to the mathematical property that
5798 one-third of the time. So don't do that.
5800 =item stat FILEHANDLE
5801 X<stat> X<file, status>
5807 Returns a 13-element list giving the status info for a file, either
5808 the file opened via FILEHANDLE, or named by EXPR. If EXPR is omitted,
5809 it stats C<$_>. Returns a null list if the stat fails. Typically used
5812 ($dev,$ino,$mode,$nlink,$uid,$gid,$rdev,$size,
5813 $atime,$mtime,$ctime,$blksize,$blocks)
5816 Not all fields are supported on all filesystem types. Here are the
5817 meanings of the fields:
5819 0 dev device number of filesystem
5821 2 mode file mode (type and permissions)
5822 3 nlink number of (hard) links to the file
5823 4 uid numeric user ID of file's owner
5824 5 gid numeric group ID of file's owner
5825 6 rdev the device identifier (special files only)
5826 7 size total size of file, in bytes
5827 8 atime last access time in seconds since the epoch
5828 9 mtime last modify time in seconds since the epoch
5829 10 ctime inode change time in seconds since the epoch (*)
5830 11 blksize preferred block size for file system I/O
5831 12 blocks actual number of blocks allocated
5833 (The epoch was at 00:00 January 1, 1970 GMT.)
5835 (*) Not all fields are supported on all filesystem types. Notably, the
5836 ctime field is non-portable. In particular, you cannot expect it to be a
5837 "creation time", see L<perlport/"Files and Filesystems"> for details.
5839 If C<stat> is passed the special filehandle consisting of an underline, no
5840 stat is done, but the current contents of the stat structure from the
5841 last C<stat>, C<lstat>, or filetest are returned. Example:
5843 if (-x $file && (($d) = stat(_)) && $d < 0) {
5844 print "$file is executable NFS file\n";
5847 (This works on machines only for which the device number is negative
5850 Because the mode contains both the file type and its permissions, you
5851 should mask off the file type portion and (s)printf using a C<"%o">
5852 if you want to see the real permissions.
5854 $mode = (stat($filename))[2];
5855 printf "Permissions are %04o\n", $mode & 07777;
5857 In scalar context, C<stat> returns a boolean value indicating success
5858 or failure, and, if successful, sets the information associated with
5859 the special filehandle C<_>.
5861 The File::stat module provides a convenient, by-name access mechanism:
5864 $sb = stat($filename);
5865 printf "File is %s, size is %s, perm %04o, mtime %s\n",
5866 $filename, $sb->size, $sb->mode & 07777,
5867 scalar localtime $sb->mtime;
5869 You can import symbolic mode constants (C<S_IF*>) and functions
5870 (C<S_IS*>) from the Fcntl module:
5874 $mode = (stat($filename))[2];
5876 $user_rwx = ($mode & S_IRWXU) >> 6;
5877 $group_read = ($mode & S_IRGRP) >> 3;
5878 $other_execute = $mode & S_IXOTH;
5880 printf "Permissions are %04o\n", S_IMODE($mode), "\n";
5882 $is_setuid = $mode & S_ISUID;
5883 $is_setgid = S_ISDIR($mode);
5885 You could write the last two using the C<-u> and C<-d> operators.
5886 The commonly available C<S_IF*> constants are
5888 # Permissions: read, write, execute, for user, group, others.
5890 S_IRWXU S_IRUSR S_IWUSR S_IXUSR
5891 S_IRWXG S_IRGRP S_IWGRP S_IXGRP
5892 S_IRWXO S_IROTH S_IWOTH S_IXOTH
5894 # Setuid/Setgid/Stickiness/SaveText.
5895 # Note that the exact meaning of these is system dependent.
5897 S_ISUID S_ISGID S_ISVTX S_ISTXT
5899 # File types. Not necessarily all are available on your system.
5901 S_IFREG S_IFDIR S_IFLNK S_IFBLK S_ISCHR S_IFIFO S_IFSOCK S_IFWHT S_ENFMT
5903 # The following are compatibility aliases for S_IRUSR, S_IWUSR, S_IXUSR.
5905 S_IREAD S_IWRITE S_IEXEC
5907 and the C<S_IF*> functions are
5909 S_IMODE($mode) the part of $mode containing the permission bits
5910 and the setuid/setgid/sticky bits
5912 S_IFMT($mode) the part of $mode containing the file type
5913 which can be bit-anded with e.g. S_IFREG
5914 or with the following functions
5916 # The operators -f, -d, -l, -b, -c, -p, and -S.
5918 S_ISREG($mode) S_ISDIR($mode) S_ISLNK($mode)
5919 S_ISBLK($mode) S_ISCHR($mode) S_ISFIFO($mode) S_ISSOCK($mode)
5921 # No direct -X operator counterpart, but for the first one
5922 # the -g operator is often equivalent. The ENFMT stands for
5923 # record flocking enforcement, a platform-dependent feature.
5925 S_ISENFMT($mode) S_ISWHT($mode)
5927 See your native chmod(2) and stat(2) documentation for more details
5928 about the C<S_*> constants. To get status info for a symbolic link
5929 instead of the target file behind the link, use the C<lstat> function.
5936 Takes extra time to study SCALAR (C<$_> if unspecified) in anticipation of
5937 doing many pattern matches on the string before it is next modified.
5938 This may or may not save time, depending on the nature and number of
5939 patterns you are searching on, and on the distribution of character
5940 frequencies in the string to be searched--you probably want to compare
5941 run times with and without it to see which runs faster. Those loops
5942 that scan for many short constant strings (including the constant
5943 parts of more complex patterns) will benefit most. You may have only
5944 one C<study> active at a time--if you study a different scalar the first
5945 is "unstudied". (The way C<study> works is this: a linked list of every
5946 character in the string to be searched is made, so we know, for
5947 example, where all the C<'k'> characters are. From each search string,
5948 the rarest character is selected, based on some static frequency tables
5949 constructed from some C programs and English text. Only those places
5950 that contain this "rarest" character are examined.)
5952 For example, here is a loop that inserts index producing entries
5953 before any line containing a certain pattern:
5957 print ".IX foo\n" if /\bfoo\b/;
5958 print ".IX bar\n" if /\bbar\b/;
5959 print ".IX blurfl\n" if /\bblurfl\b/;
5964 In searching for C</\bfoo\b/>, only those locations in C<$_> that contain C<f>
5965 will be looked at, because C<f> is rarer than C<o>. In general, this is
5966 a big win except in pathological cases. The only question is whether
5967 it saves you more time than it took to build the linked list in the
5970 Note that if you have to look for strings that you don't know till
5971 runtime, you can build an entire loop as a string and C<eval> that to
5972 avoid recompiling all your patterns all the time. Together with
5973 undefining C<$/> to input entire files as one record, this can be very
5974 fast, often faster than specialized programs like fgrep(1). The following
5975 scans a list of files (C<@files>) for a list of words (C<@words>), and prints
5976 out the names of those files that contain a match:
5978 $search = 'while (<>) { study;';
5979 foreach $word (@words) {
5980 $search .= "++\$seen{\$ARGV} if /\\b$word\\b/;\n";
5985 eval $search; # this screams
5986 $/ = "\n"; # put back to normal input delimiter
5987 foreach $file (sort keys(%seen)) {
5991 =item sub NAME BLOCK
5994 =item sub NAME (PROTO) BLOCK
5996 =item sub NAME : ATTRS BLOCK
5998 =item sub NAME (PROTO) : ATTRS BLOCK
6000 This is subroutine definition, not a real function I<per se>.
6001 Without a BLOCK it's just a forward declaration. Without a NAME,
6002 it's an anonymous function declaration, and does actually return
6003 a value: the CODE ref of the closure you just created.
6005 See L<perlsub> and L<perlref> for details about subroutines and
6006 references, and L<attributes> and L<Attribute::Handlers> for more
6007 information about attributes.
6009 =item substr EXPR,OFFSET,LENGTH,REPLACEMENT
6010 X<substr> X<substring> X<mid> X<left> X<right>
6012 =item substr EXPR,OFFSET,LENGTH
6014 =item substr EXPR,OFFSET
6016 Extracts a substring out of EXPR and returns it. First character is at
6017 offset C<0>, or whatever you've set C<$[> to (but don't do that).
6018 If OFFSET is negative (or more precisely, less than C<$[>), starts
6019 that far from the end of the string. If LENGTH is omitted, returns
6020 everything to the end of the string. If LENGTH is negative, leaves that
6021 many characters off the end of the string.
6023 You can use the substr() function as an lvalue, in which case EXPR
6024 must itself be an lvalue. If you assign something shorter than LENGTH,
6025 the string will shrink, and if you assign something longer than LENGTH,
6026 the string will grow to accommodate it. To keep the string the same
6027 length you may need to pad or chop your value using C<sprintf>.
6029 If OFFSET and LENGTH specify a substring that is partly outside the
6030 string, only the part within the string is returned. If the substring
6031 is beyond either end of the string, substr() returns the undefined
6032 value and produces a warning. When used as an lvalue, specifying a
6033 substring that is entirely outside the string is a fatal error.
6034 Here's an example showing the behavior for boundary cases:
6037 substr($name, 4) = 'dy'; # $name is now 'freddy'
6038 my $null = substr $name, 6, 2; # returns '' (no warning)
6039 my $oops = substr $name, 7; # returns undef, with warning
6040 substr($name, 7) = 'gap'; # fatal error
6042 An alternative to using substr() as an lvalue is to specify the
6043 replacement string as the 4th argument. This allows you to replace
6044 parts of the EXPR and return what was there before in one operation,
6045 just as you can with splice().
6047 Note that the lvalue returned by the 3-arg version of substr() acts as
6048 a 'magic bullet'; each time it is assigned to, it remembers which part
6049 of the original string is being modified; for example:
6052 for (substr($x,1,2)) {
6053 $_ = 'a'; print $x,"\n"; # prints 1a4
6054 $_ = 'xyz'; print $x,"\n"; # prints 1xyz4
6056 $_ = 'pq'; print $x,"\n"; # prints 5pq9
6060 Prior to Perl version 5.9.1, the result of using an lvalue multiple times was
6063 =item symlink OLDFILE,NEWFILE
6064 X<symlink> X<link> X<symbolic link> X<link, symbolic>
6066 Creates a new filename symbolically linked to the old filename.
6067 Returns C<1> for success, C<0> otherwise. On systems that don't support
6068 symbolic links, produces a fatal error at run time. To check for that,
6071 $symlink_exists = eval { symlink("",""); 1 };
6073 =item syscall NUMBER, LIST
6074 X<syscall> X<system call>
6076 Calls the system call specified as the first element of the list,
6077 passing the remaining elements as arguments to the system call. If
6078 unimplemented, produces a fatal error. The arguments are interpreted
6079 as follows: if a given argument is numeric, the argument is passed as
6080 an int. If not, the pointer to the string value is passed. You are
6081 responsible to make sure a string is pre-extended long enough to
6082 receive any result that might be written into a string. You can't use a
6083 string literal (or other read-only string) as an argument to C<syscall>
6084 because Perl has to assume that any string pointer might be written
6086 integer arguments are not literals and have never been interpreted in a
6087 numeric context, you may need to add C<0> to them to force them to look
6088 like numbers. This emulates the C<syswrite> function (or vice versa):
6090 require 'syscall.ph'; # may need to run h2ph
6092 syscall(&SYS_write, fileno(STDOUT), $s, length $s);
6094 Note that Perl supports passing of up to only 14 arguments to your system call,
6095 which in practice should usually suffice.
6097 Syscall returns whatever value returned by the system call it calls.
6098 If the system call fails, C<syscall> returns C<-1> and sets C<$!> (errno).
6099 Note that some system calls can legitimately return C<-1>. The proper
6100 way to handle such calls is to assign C<$!=0;> before the call and
6101 check the value of C<$!> if syscall returns C<-1>.
6103 There's a problem with C<syscall(&SYS_pipe)>: it returns the file
6104 number of the read end of the pipe it creates. There is no way
6105 to retrieve the file number of the other end. You can avoid this
6106 problem by using C<pipe> instead.
6108 =item sysopen FILEHANDLE,FILENAME,MODE
6111 =item sysopen FILEHANDLE,FILENAME,MODE,PERMS
6113 Opens the file whose filename is given by FILENAME, and associates it
6114 with FILEHANDLE. If FILEHANDLE is an expression, its value is used as
6115 the name of the real filehandle wanted. This function calls the
6116 underlying operating system's C<open> function with the parameters
6117 FILENAME, MODE, PERMS.
6119 The possible values and flag bits of the MODE parameter are
6120 system-dependent; they are available via the standard module C<Fcntl>.
6121 See the documentation of your operating system's C<open> to see which
6122 values and flag bits are available. You may combine several flags
6123 using the C<|>-operator.
6125 Some of the most common values are C<O_RDONLY> for opening the file in
6126 read-only mode, C<O_WRONLY> for opening the file in write-only mode,
6127 and C<O_RDWR> for opening the file in read-write mode.
6128 X<O_RDONLY> X<O_RDWR> X<O_WRONLY>
6130 For historical reasons, some values work on almost every system
6131 supported by perl: zero means read-only, one means write-only, and two
6132 means read/write. We know that these values do I<not> work under
6133 OS/390 & VM/ESA Unix and on the Macintosh; you probably don't want to
6134 use them in new code.
6136 If the file named by FILENAME does not exist and the C<open> call creates
6137 it (typically because MODE includes the C<O_CREAT> flag), then the value of
6138 PERMS specifies the permissions of the newly created file. If you omit
6139 the PERMS argument to C<sysopen>, Perl uses the octal value C<0666>.
6140 These permission values need to be in octal, and are modified by your
6141 process's current C<umask>.
6144 In many systems the C<O_EXCL> flag is available for opening files in
6145 exclusive mode. This is B<not> locking: exclusiveness means here that
6146 if the file already exists, sysopen() fails. C<O_EXCL> may not work
6147 on network filesystems, and has no effect unless the C<O_CREAT> flag
6148 is set as well. Setting C<O_CREAT|O_EXCL> prevents the file from
6149 being opened if it is a symbolic link. It does not protect against
6150 symbolic links in the file's path.
6153 Sometimes you may want to truncate an already-existing file. This
6154 can be done using the C<O_TRUNC> flag. The behavior of
6155 C<O_TRUNC> with C<O_RDONLY> is undefined.
6158 You should seldom if ever use C<0644> as argument to C<sysopen>, because
6159 that takes away the user's option to have a more permissive umask.
6160 Better to omit it. See the perlfunc(1) entry on C<umask> for more
6163 Note that C<sysopen> depends on the fdopen() C library function.
6164 On many UNIX systems, fdopen() is known to fail when file descriptors
6165 exceed a certain value, typically 255. If you need more file
6166 descriptors than that, consider rebuilding Perl to use the C<sfio>
6167 library, or perhaps using the POSIX::open() function.
6169 See L<perlopentut> for a kinder, gentler explanation of opening files.
6171 =item sysread FILEHANDLE,SCALAR,LENGTH,OFFSET
6174 =item sysread FILEHANDLE,SCALAR,LENGTH
6176 Attempts to read LENGTH bytes of data into variable SCALAR from the
6177 specified FILEHANDLE, using the system call read(2). It bypasses
6178 buffered IO, so mixing this with other kinds of reads, C<print>,
6179 C<write>, C<seek>, C<tell>, or C<eof> can cause confusion because the
6180 perlio or stdio layers usually buffers data. Returns the number of
6181 bytes actually read, C<0> at end of file, or undef if there was an
6182 error (in the latter case C<$!> is also set). SCALAR will be grown or
6183 shrunk so that the last byte actually read is the last byte of the
6184 scalar after the read.
6186 An OFFSET may be specified to place the read data at some place in the
6187 string other than the beginning. A negative OFFSET specifies
6188 placement at that many characters counting backwards from the end of
6189 the string. A positive OFFSET greater than the length of SCALAR
6190 results in the string being padded to the required size with C<"\0">
6191 bytes before the result of the read is appended.
6193 There is no syseof() function, which is ok, since eof() doesn't work
6194 very well on device files (like ttys) anyway. Use sysread() and check
6195 for a return value for 0 to decide whether you're done.
6197 Note that if the filehandle has been marked as C<:utf8> Unicode
6198 characters are read instead of bytes (the LENGTH, OFFSET, and the
6199 return value of sysread() are in Unicode characters).
6200 The C<:encoding(...)> layer implicitly introduces the C<:utf8> layer.
6201 See L</binmode>, L</open>, and the C<open> pragma, L<open>.
6203 =item sysseek FILEHANDLE,POSITION,WHENCE
6206 Sets FILEHANDLE's system position in bytes using the system call
6207 lseek(2). FILEHANDLE may be an expression whose value gives the name
6208 of the filehandle. The values for WHENCE are C<0> to set the new
6209 position to POSITION, C<1> to set the it to the current position plus
6210 POSITION, and C<2> to set it to EOF plus POSITION (typically
6213 Note the I<in bytes>: even if the filehandle has been set to operate
6214 on characters (for example by using the C<:utf8> I/O layer), tell()
6215 will return byte offsets, not character offsets (because implementing
6216 that would render sysseek() very slow).
6218 sysseek() bypasses normal buffered IO, so mixing this with reads (other
6219 than C<sysread>, for example C<< <> >> or read()) C<print>, C<write>,
6220 C<seek>, C<tell>, or C<eof> may cause confusion.
6222 For WHENCE, you may also use the constants C<SEEK_SET>, C<SEEK_CUR>,
6223 and C<SEEK_END> (start of the file, current position, end of the file)
6224 from the Fcntl module. Use of the constants is also more portable
6225 than relying on 0, 1, and 2. For example to define a "systell" function:
6227 use Fcntl 'SEEK_CUR';
6228 sub systell { sysseek($_[0], 0, SEEK_CUR) }
6230 Returns the new position, or the undefined value on failure. A position
6231 of zero is returned as the string C<"0 but true">; thus C<sysseek> returns
6232 true on success and false on failure, yet you can still easily determine
6238 =item system PROGRAM LIST
6240 Does exactly the same thing as C<exec LIST>, except that a fork is
6241 done first, and the parent process waits for the child process to
6242 complete. Note that argument processing varies depending on the
6243 number of arguments. If there is more than one argument in LIST,
6244 or if LIST is an array with more than one value, starts the program
6245 given by the first element of the list with arguments given by the
6246 rest of the list. If there is only one scalar argument, the argument
6247 is checked for shell metacharacters, and if there are any, the
6248 entire argument is passed to the system's command shell for parsing
6249 (this is C</bin/sh -c> on Unix platforms, but varies on other
6250 platforms). If there are no shell metacharacters in the argument,
6251 it is split into words and passed directly to C<execvp>, which is
6254 Beginning with v5.6.0, Perl will attempt to flush all files opened for
6255 output before any operation that may do a fork, but this may not be
6256 supported on some platforms (see L<perlport>). To be safe, you may need
6257 to set C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method
6258 of C<IO::Handle> on any open handles.
6260 The return value is the exit status of the program as returned by the
6261 C<wait> call. To get the actual exit value, shift right by eight (see
6262 below). See also L</exec>. This is I<not> what you want to use to capture
6263 the output from a command, for that you should use merely backticks or
6264 C<qx//>, as described in L<perlop/"`STRING`">. Return value of -1
6265 indicates a failure to start the program or an error of the wait(2) system
6266 call (inspect $! for the reason).
6268 Like C<exec>, C<system> allows you to lie to a program about its name if
6269 you use the C<system PROGRAM LIST> syntax. Again, see L</exec>.
6271 Since C<SIGINT> and C<SIGQUIT> are ignored during the execution of
6272 C<system>, if you expect your program to terminate on receipt of these
6273 signals you will need to arrange to do so yourself based on the return
6276 @args = ("command", "arg1", "arg2");
6278 or die "system @args failed: $?"
6280 You can check all the failure possibilities by inspecting
6284 print "failed to execute: $!\n";
6287 printf "child died with signal %d, %s coredump\n",
6288 ($? & 127), ($? & 128) ? 'with' : 'without';
6291 printf "child exited with value %d\n", $? >> 8;
6294 Alternatively you might inspect the value of C<${^CHILD_ERROR_NATIVE}>
6295 with the W*() calls of the POSIX extension.
6297 When the arguments get executed via the system shell, results
6298 and return codes will be subject to its quirks and capabilities.
6299 See L<perlop/"`STRING`"> and L</exec> for details.
6301 =item syswrite FILEHANDLE,SCALAR,LENGTH,OFFSET
6304 =item syswrite FILEHANDLE,SCALAR,LENGTH
6306 =item syswrite FILEHANDLE,SCALAR
6308 Attempts to write LENGTH bytes of data from variable SCALAR to the
6309 specified FILEHANDLE, using the system call write(2). If LENGTH is
6310 not specified, writes whole SCALAR. It bypasses buffered IO, so
6311 mixing this with reads (other than C<sysread())>, C<print>, C<write>,
6312 C<seek>, C<tell>, or C<eof> may cause confusion because the perlio and
6313 stdio layers usually buffers data. Returns the number of bytes
6314 actually written, or C<undef> if there was an error (in this case the
6315 errno variable C<$!> is also set). If the LENGTH is greater than the
6316 available data in the SCALAR after the OFFSET, only as much data as is
6317 available will be written.
6319 An OFFSET may be specified to write the data from some part of the
6320 string other than the beginning. A negative OFFSET specifies writing
6321 that many characters counting backwards from the end of the string.
6322 In the case the SCALAR is empty you can use OFFSET but only zero offset.
6324 Note that if the filehandle has been marked as C<:utf8>, Unicode
6325 characters are written instead of bytes (the LENGTH, OFFSET, and the
6326 return value of syswrite() are in UTF-8 encoded Unicode characters).
6327 The C<:encoding(...)> layer implicitly introduces the C<:utf8> layer.
6328 See L</binmode>, L</open>, and the C<open> pragma, L<open>.
6330 =item tell FILEHANDLE
6335 Returns the current position I<in bytes> for FILEHANDLE, or -1 on
6336 error. FILEHANDLE may be an expression whose value gives the name of
6337 the actual filehandle. If FILEHANDLE is omitted, assumes the file
6340 Note the I<in bytes>: even if the filehandle has been set to
6341 operate on characters (for example by using the C<:utf8> open
6342 layer), tell() will return byte offsets, not character offsets
6343 (because that would render seek() and tell() rather slow).
6345 The return value of tell() for the standard streams like the STDIN
6346 depends on the operating system: it may return -1 or something else.
6347 tell() on pipes, fifos, and sockets usually returns -1.
6349 There is no C<systell> function. Use C<sysseek(FH, 0, 1)> for that.
6351 Do not use tell() (or other buffered I/O operations) on a file handle
6352 that has been manipulated by sysread(), syswrite() or sysseek().
6353 Those functions ignore the buffering, while tell() does not.
6355 =item telldir DIRHANDLE
6358 Returns the current position of the C<readdir> routines on DIRHANDLE.
6359 Value may be given to C<seekdir> to access a particular location in a
6360 directory. C<telldir> has the same caveats about possible directory
6361 compaction as the corresponding system library routine.
6363 =item tie VARIABLE,CLASSNAME,LIST
6366 This function binds a variable to a package class that will provide the
6367 implementation for the variable. VARIABLE is the name of the variable
6368 to be enchanted. CLASSNAME is the name of a class implementing objects
6369 of correct type. Any additional arguments are passed to the C<new>
6370 method of the class (meaning C<TIESCALAR>, C<TIEHANDLE>, C<TIEARRAY>,
6371 or C<TIEHASH>). Typically these are arguments such as might be passed
6372 to the C<dbm_open()> function of C. The object returned by the C<new>
6373 method is also returned by the C<tie> function, which would be useful
6374 if you want to access other methods in CLASSNAME.
6376 Note that functions such as C<keys> and C<values> may return huge lists
6377 when used on large objects, like DBM files. You may prefer to use the
6378 C<each> function to iterate over such. Example:
6380 # print out history file offsets
6382 tie(%HIST, 'NDBM_File', '/usr/lib/news/history', 1, 0);
6383 while (($key,$val) = each %HIST) {
6384 print $key, ' = ', unpack('L',$val), "\n";
6388 A class implementing a hash should have the following methods:
6390 TIEHASH classname, LIST
6392 STORE this, key, value
6397 NEXTKEY this, lastkey
6402 A class implementing an ordinary array should have the following methods:
6404 TIEARRAY classname, LIST
6406 STORE this, key, value
6408 STORESIZE this, count
6414 SPLICE this, offset, length, LIST
6419 A class implementing a file handle should have the following methods:
6421 TIEHANDLE classname, LIST
6422 READ this, scalar, length, offset
6425 WRITE this, scalar, length, offset
6427 PRINTF this, format, LIST
6431 SEEK this, position, whence
6433 OPEN this, mode, LIST
6438 A class implementing a scalar should have the following methods:
6440 TIESCALAR classname, LIST
6446 Not all methods indicated above need be implemented. See L<perltie>,
6447 L<Tie::Hash>, L<Tie::Array>, L<Tie::Scalar>, and L<Tie::Handle>.
6449 Unlike C<dbmopen>, the C<tie> function will not use or require a module
6450 for you--you need to do that explicitly yourself. See L<DB_File>
6451 or the F<Config> module for interesting C<tie> implementations.
6453 For further details see L<perltie>, L<"tied VARIABLE">.
6458 Returns a reference to the object underlying VARIABLE (the same value
6459 that was originally returned by the C<tie> call that bound the variable
6460 to a package.) Returns the undefined value if VARIABLE isn't tied to a
6466 Returns the number of non-leap seconds since whatever time the system
6467 considers to be the epoch, suitable for feeding to C<gmtime> and
6468 C<localtime>. On most systems the epoch is 00:00:00 UTC, January 1, 1970;
6469 a prominent exception being Mac OS Classic which uses 00:00:00, January 1,
6470 1904 in the current local time zone for its epoch.
6472 For measuring time in better granularity than one second,
6473 you may use either the Time::HiRes module (from CPAN, and starting from
6474 Perl 5.8 part of the standard distribution), or if you have
6475 gettimeofday(2), you may be able to use the C<syscall> interface of Perl.
6476 See L<perlfaq8> for details.
6481 Returns a four-element list giving the user and system times, in
6482 seconds, for this process and the children of this process.
6484 ($user,$system,$cuser,$csystem) = times;
6486 In scalar context, C<times> returns C<$user>.
6490 The transliteration operator. Same as C<y///>. See L<perlop>.
6492 =item truncate FILEHANDLE,LENGTH
6495 =item truncate EXPR,LENGTH
6497 Truncates the file opened on FILEHANDLE, or named by EXPR, to the
6498 specified length. Produces a fatal error if truncate isn't implemented
6499 on your system. Returns true if successful, the undefined value
6502 The behavior is undefined if LENGTH is greater than the length of the
6506 X<uc> X<uppercase> X<toupper>
6510 Returns an uppercased version of EXPR. This is the internal function
6511 implementing the C<\U> escape in double-quoted strings. Respects
6512 current LC_CTYPE locale if C<use locale> in force. See L<perllocale>
6513 and L<perlunicode> for more details about locale and Unicode support.
6514 It does not attempt to do titlecase mapping on initial letters. See
6515 C<ucfirst> for that.
6517 If EXPR is omitted, uses C<$_>.
6520 X<ucfirst> X<uppercase>
6524 Returns the value of EXPR with the first character in uppercase
6525 (titlecase in Unicode). This is the internal function implementing
6526 the C<\u> escape in double-quoted strings. Respects current LC_CTYPE
6527 locale if C<use locale> in force. See L<perllocale> and L<perlunicode>
6528 for more details about locale and Unicode support.
6530 If EXPR is omitted, uses C<$_>.
6537 Sets the umask for the process to EXPR and returns the previous value.
6538 If EXPR is omitted, merely returns the current umask.
6540 The Unix permission C<rwxr-x---> is represented as three sets of three
6541 bits, or three octal digits: C<0750> (the leading 0 indicates octal
6542 and isn't one of the digits). The C<umask> value is such a number
6543 representing disabled permissions bits. The permission (or "mode")
6544 values you pass C<mkdir> or C<sysopen> are modified by your umask, so
6545 even if you tell C<sysopen> to create a file with permissions C<0777>,
6546 if your umask is C<0022> then the file will actually be created with
6547 permissions C<0755>. If your C<umask> were C<0027> (group can't
6548 write; others can't read, write, or execute), then passing
6549 C<sysopen> C<0666> would create a file with mode C<0640> (C<0666 &~
6552 Here's some advice: supply a creation mode of C<0666> for regular
6553 files (in C<sysopen>) and one of C<0777> for directories (in
6554 C<mkdir>) and executable files. This gives users the freedom of
6555 choice: if they want protected files, they might choose process umasks
6556 of C<022>, C<027>, or even the particularly antisocial mask of C<077>.
6557 Programs should rarely if ever make policy decisions better left to
6558 the user. The exception to this is when writing files that should be
6559 kept private: mail files, web browser cookies, I<.rhosts> files, and
6562 If umask(2) is not implemented on your system and you are trying to
6563 restrict access for I<yourself> (i.e., (EXPR & 0700) > 0), produces a
6564 fatal error at run time. If umask(2) is not implemented and you are
6565 not trying to restrict access for yourself, returns C<undef>.
6567 Remember that a umask is a number, usually given in octal; it is I<not> a
6568 string of octal digits. See also L</oct>, if all you have is a string.
6571 X<undef> X<undefine>
6575 Undefines the value of EXPR, which must be an lvalue. Use only on a
6576 scalar value, an array (using C<@>), a hash (using C<%>), a subroutine
6577 (using C<&>), or a typeglob (using C<*>). (Saying C<undef $hash{$key}>
6578 will probably not do what you expect on most predefined variables or
6579 DBM list values, so don't do that; see L<delete>.) Always returns the
6580 undefined value. You can omit the EXPR, in which case nothing is
6581 undefined, but you still get an undefined value that you could, for
6582 instance, return from a subroutine, assign to a variable or pass as a
6583 parameter. Examples:
6586 undef $bar{'blurfl'}; # Compare to: delete $bar{'blurfl'};
6590 undef *xyz; # destroys $xyz, @xyz, %xyz, &xyz, etc.
6591 return (wantarray ? (undef, $errmsg) : undef) if $they_blew_it;
6592 select undef, undef, undef, 0.25;
6593 ($a, $b, undef, $c) = &foo; # Ignore third value returned
6595 Note that this is a unary operator, not a list operator.
6598 X<unlink> X<delete> X<remove> X<rm>
6602 Deletes a list of files. Returns the number of files successfully
6605 $cnt = unlink 'a', 'b', 'c';
6609 Note: C<unlink> will not attempt to delete directories unless you are superuser
6610 and the B<-U> flag is supplied to Perl. Even if these conditions are
6611 met, be warned that unlinking a directory can inflict damage on your
6612 filesystem. Finally, using C<unlink> on directories is not supported on
6613 many operating systems. Use C<rmdir> instead.
6615 If LIST is omitted, uses C<$_>.
6617 =item unpack TEMPLATE,EXPR
6620 =item unpack TEMPLATE
6622 C<unpack> does the reverse of C<pack>: it takes a string
6623 and expands it out into a list of values.
6624 (In scalar context, it returns merely the first value produced.)
6626 If EXPR is omitted, unpacks the C<$_> string.
6628 The string is broken into chunks described by the TEMPLATE. Each chunk
6629 is converted separately to a value. Typically, either the string is a result
6630 of C<pack>, or the characters of the string represent a C structure of some
6633 The TEMPLATE has the same format as in the C<pack> function.
6634 Here's a subroutine that does substring:
6637 my($what,$where,$howmuch) = @_;
6638 unpack("x$where a$howmuch", $what);
6643 sub ordinal { unpack("W",$_[0]); } # same as ord()
6645 In addition to fields allowed in pack(), you may prefix a field with
6646 a %<number> to indicate that
6647 you want a <number>-bit checksum of the items instead of the items
6648 themselves. Default is a 16-bit checksum. Checksum is calculated by
6649 summing numeric values of expanded values (for string fields the sum of
6650 C<ord($char)> is taken, for bit fields the sum of zeroes and ones).
6652 For example, the following
6653 computes the same number as the System V sum program:
6657 unpack("%32W*",<>) % 65535;
6660 The following efficiently counts the number of set bits in a bit vector:
6662 $setbits = unpack("%32b*", $selectmask);
6664 The C<p> and C<P> formats should be used with care. Since Perl
6665 has no way of checking whether the value passed to C<unpack()>
6666 corresponds to a valid memory location, passing a pointer value that's
6667 not known to be valid is likely to have disastrous consequences.
6669 If there are more pack codes or if the repeat count of a field or a group
6670 is larger than what the remainder of the input string allows, the result
6671 is not well defined: in some cases, the repeat count is decreased, or
6672 C<unpack()> will produce null strings or zeroes, or terminate with an
6673 error. If the input string is longer than one described by the TEMPLATE,
6674 the rest is ignored.
6676 See L</pack> for more examples and notes.
6678 =item untie VARIABLE
6681 Breaks the binding between a variable and a package. (See C<tie>.)
6682 Has no effect if the variable is not tied.
6684 =item unshift ARRAY,LIST
6687 Does the opposite of a C<shift>. Or the opposite of a C<push>,
6688 depending on how you look at it. Prepends list to the front of the
6689 array, and returns the new number of elements in the array.
6691 unshift(@ARGV, '-e') unless $ARGV[0] =~ /^-/;
6693 Note the LIST is prepended whole, not one element at a time, so the
6694 prepended elements stay in the same order. Use C<reverse> to do the
6697 =item use Module VERSION LIST
6698 X<use> X<module> X<import>
6700 =item use Module VERSION
6702 =item use Module LIST
6708 Imports some semantics into the current package from the named module,
6709 generally by aliasing certain subroutine or variable names into your
6710 package. It is exactly equivalent to
6712 BEGIN { require Module; import Module LIST; }
6714 except that Module I<must> be a bareword.
6716 VERSION may be either a numeric argument such as 5.006, which will be
6717 compared to C<$]>, or a literal of the form v5.6.1, which will be compared
6718 to C<$^V> (aka $PERL_VERSION. A fatal error is produced if VERSION is
6719 greater than the version of the current Perl interpreter; Perl will not
6720 attempt to parse the rest of the file. Compare with L</require>, which can
6721 do a similar check at run time.
6723 Specifying VERSION as a literal of the form v5.6.1 should generally be
6724 avoided, because it leads to misleading error messages under earlier
6725 versions of Perl that do not support this syntax. The equivalent numeric
6726 version should be used instead.
6728 use v5.6.1; # compile time version check
6730 use 5.006_001; # ditto; preferred for backwards compatibility
6732 This is often useful if you need to check the current Perl version before
6733 C<use>ing library modules that have changed in incompatible ways from
6734 older versions of Perl. (We try not to do this more than we have to.)
6736 The C<BEGIN> forces the C<require> and C<import> to happen at compile time. The
6737 C<require> makes sure the module is loaded into memory if it hasn't been
6738 yet. The C<import> is not a builtin--it's just an ordinary static method
6739 call into the C<Module> package to tell the module to import the list of
6740 features back into the current package. The module can implement its
6741 C<import> method any way it likes, though most modules just choose to
6742 derive their C<import> method via inheritance from the C<Exporter> class that
6743 is defined in the C<Exporter> module. See L<Exporter>. If no C<import>
6744 method can be found then the call is skipped, even if there is an AUTOLOAD
6747 If you do not want to call the package's C<import> method (for instance,
6748 to stop your namespace from being altered), explicitly supply the empty list:
6752 That is exactly equivalent to
6754 BEGIN { require Module }
6756 If the VERSION argument is present between Module and LIST, then the
6757 C<use> will call the VERSION method in class Module with the given
6758 version as an argument. The default VERSION method, inherited from
6759 the UNIVERSAL class, croaks if the given version is larger than the
6760 value of the variable C<$Module::VERSION>.
6762 Again, there is a distinction between omitting LIST (C<import> called
6763 with no arguments) and an explicit empty LIST C<()> (C<import> not
6764 called). Note that there is no comma after VERSION!
6766 Because this is a wide-open interface, pragmas (compiler directives)
6767 are also implemented this way. Currently implemented pragmas are:
6772 use sigtrap qw(SEGV BUS);
6773 use strict qw(subs vars refs);
6774 use subs qw(afunc blurfl);
6775 use warnings qw(all);
6776 use sort qw(stable _quicksort _mergesort);
6778 Some of these pseudo-modules import semantics into the current
6779 block scope (like C<strict> or C<integer>, unlike ordinary modules,
6780 which import symbols into the current package (which are effective
6781 through the end of the file).
6783 There's a corresponding C<no> command that unimports meanings imported
6784 by C<use>, i.e., it calls C<unimport Module LIST> instead of C<import>.
6785 It behaves exactly as C<import> does with respect to VERSION, an
6786 omitted LIST, empty LIST, or no unimport method being found.
6792 See L<perlmodlib> for a list of standard modules and pragmas. See L<perlrun>
6793 for the C<-M> and C<-m> command-line options to perl that give C<use>
6794 functionality from the command-line.
6799 Changes the access and modification times on each file of a list of
6800 files. The first two elements of the list must be the NUMERICAL access
6801 and modification times, in that order. Returns the number of files
6802 successfully changed. The inode change time of each file is set
6803 to the current time. For example, this code has the same effect as the
6804 Unix touch(1) command when the files I<already exist> and belong to
6805 the user running the program:
6808 $atime = $mtime = time;
6809 utime $atime, $mtime, @ARGV;
6811 Since perl 5.7.2, if the first two elements of the list are C<undef>, then
6812 the utime(2) function in the C library will be called with a null second
6813 argument. On most systems, this will set the file's access and
6814 modification times to the current time (i.e. equivalent to the example
6815 above) and will even work on other users' files where you have write
6818 utime undef, undef, @ARGV;
6820 Under NFS this will use the time of the NFS server, not the time of
6821 the local machine. If there is a time synchronization problem, the
6822 NFS server and local machine will have different times. The Unix
6823 touch(1) command will in fact normally use this form instead of the
6824 one shown in the first example.
6826 Note that only passing one of the first two elements as C<undef> will
6827 be equivalent of passing it as 0 and will not have the same effect as
6828 described when they are both C<undef>. This case will also trigger an
6829 uninitialized warning.
6831 On systems that support futimes, you might pass file handles among the
6832 files. On systems that don't support futimes, passing file handles
6833 produces a fatal error at run time.
6838 Returns a list consisting of all the values of the named hash.
6839 (In a scalar context, returns the number of values.)
6841 The values are returned in an apparently random order. The actual
6842 random order is subject to change in future versions of perl, but it
6843 is guaranteed to be the same order as either the C<keys> or C<each>
6844 function would produce on the same (unmodified) hash. Since Perl
6845 5.8.1 the ordering is different even between different runs of Perl
6846 for security reasons (see L<perlsec/"Algorithmic Complexity Attacks">).
6848 As a side effect, calling values() resets the HASH's internal iterator,
6849 see L</each>. (In particular, calling values() in void context resets
6850 the iterator with no other overhead.)
6852 Note that the values are not copied, which means modifying them will
6853 modify the contents of the hash:
6855 for (values %hash) { s/foo/bar/g } # modifies %hash values
6856 for (@hash{keys %hash}) { s/foo/bar/g } # same
6858 See also C<keys>, C<each>, and C<sort>.
6860 =item vec EXPR,OFFSET,BITS
6861 X<vec> X<bit> X<bit vector>
6863 Treats the string in EXPR as a bit vector made up of elements of
6864 width BITS, and returns the value of the element specified by OFFSET
6865 as an unsigned integer. BITS therefore specifies the number of bits
6866 that are reserved for each element in the bit vector. This must
6867 be a power of two from 1 to 32 (or 64, if your platform supports
6870 If BITS is 8, "elements" coincide with bytes of the input string.
6872 If BITS is 16 or more, bytes of the input string are grouped into chunks
6873 of size BITS/8, and each group is converted to a number as with
6874 pack()/unpack() with big-endian formats C<n>/C<N> (and analogously
6875 for BITS==64). See L<"pack"> for details.
6877 If bits is 4 or less, the string is broken into bytes, then the bits
6878 of each byte are broken into 8/BITS groups. Bits of a byte are
6879 numbered in a little-endian-ish way, as in C<0x01>, C<0x02>,
6880 C<0x04>, C<0x08>, C<0x10>, C<0x20>, C<0x40>, C<0x80>. For example,
6881 breaking the single input byte C<chr(0x36)> into two groups gives a list
6882 C<(0x6, 0x3)>; breaking it into 4 groups gives C<(0x2, 0x1, 0x3, 0x0)>.
6884 C<vec> may also be assigned to, in which case parentheses are needed
6885 to give the expression the correct precedence as in
6887 vec($image, $max_x * $x + $y, 8) = 3;
6889 If the selected element is outside the string, the value 0 is returned.
6890 If an element off the end of the string is written to, Perl will first
6891 extend the string with sufficiently many zero bytes. It is an error
6892 to try to write off the beginning of the string (i.e. negative OFFSET).
6894 The string should not contain any character with the value > 255 (which
6895 can only happen if you're using UTF-8 encoding). If it does, it will be
6896 treated as something that is not UTF-8 encoded. When the C<vec> was
6897 assigned to, other parts of your program will also no longer consider the
6898 string to be UTF-8 encoded. In other words, if you do have such characters
6899 in your string, vec() will operate on the actual byte string, and not the
6900 conceptual character string.
6902 Strings created with C<vec> can also be manipulated with the logical
6903 operators C<|>, C<&>, C<^>, and C<~>. These operators will assume a bit
6904 vector operation is desired when both operands are strings.
6905 See L<perlop/"Bitwise String Operators">.
6907 The following code will build up an ASCII string saying C<'PerlPerlPerl'>.
6908 The comments show the string after each step. Note that this code works
6909 in the same way on big-endian or little-endian machines.
6912 vec($foo, 0, 32) = 0x5065726C; # 'Perl'
6914 # $foo eq "Perl" eq "\x50\x65\x72\x6C", 32 bits
6915 print vec($foo, 0, 8); # prints 80 == 0x50 == ord('P')
6917 vec($foo, 2, 16) = 0x5065; # 'PerlPe'
6918 vec($foo, 3, 16) = 0x726C; # 'PerlPerl'
6919 vec($foo, 8, 8) = 0x50; # 'PerlPerlP'
6920 vec($foo, 9, 8) = 0x65; # 'PerlPerlPe'
6921 vec($foo, 20, 4) = 2; # 'PerlPerlPe' . "\x02"
6922 vec($foo, 21, 4) = 7; # 'PerlPerlPer'
6924 vec($foo, 45, 2) = 3; # 'PerlPerlPer' . "\x0c"
6925 vec($foo, 93, 1) = 1; # 'PerlPerlPer' . "\x2c"
6926 vec($foo, 94, 1) = 1; # 'PerlPerlPerl'
6929 To transform a bit vector into a string or list of 0's and 1's, use these:
6931 $bits = unpack("b*", $vector);
6932 @bits = split(//, unpack("b*", $vector));
6934 If you know the exact length in bits, it can be used in place of the C<*>.
6936 Here is an example to illustrate how the bits actually fall in place:
6942 unpack("V",$_) 01234567890123456789012345678901
6943 ------------------------------------------------------------------
6948 for ($shift=0; $shift < $width; ++$shift) {
6949 for ($off=0; $off < 32/$width; ++$off) {
6950 $str = pack("B*", "0"x32);
6951 $bits = (1<<$shift);
6952 vec($str, $off, $width) = $bits;
6953 $res = unpack("b*",$str);
6954 $val = unpack("V", $str);
6961 vec($_,@#,@#) = @<< == @######### @>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
6962 $off, $width, $bits, $val, $res
6966 Regardless of the machine architecture on which it is run, the above
6967 example should print the following table:
6970 unpack("V",$_) 01234567890123456789012345678901
6971 ------------------------------------------------------------------
6972 vec($_, 0, 1) = 1 == 1 10000000000000000000000000000000
6973 vec($_, 1, 1) = 1 == 2 01000000000000000000000000000000
6974 vec($_, 2, 1) = 1 == 4 00100000000000000000000000000000
6975 vec($_, 3, 1) = 1 == 8 00010000000000000000000000000000
6976 vec($_, 4, 1) = 1 == 16 00001000000000000000000000000000
6977 vec($_, 5, 1) = 1 == 32 00000100000000000000000000000000
6978 vec($_, 6, 1) = 1 == 64 00000010000000000000000000000000
6979 vec($_, 7, 1) = 1 == 128 00000001000000000000000000000000
6980 vec($_, 8, 1) = 1 == 256 00000000100000000000000000000000
6981 vec($_, 9, 1) = 1 == 512 00000000010000000000000000000000
6982 vec($_,10, 1) = 1 == 1024 00000000001000000000000000000000
6983 vec($_,11, 1) = 1 == 2048 00000000000100000000000000000000
6984 vec($_,12, 1) = 1 == 4096 00000000000010000000000000000000
6985 vec($_,13, 1) = 1 == 8192 00000000000001000000000000000000
6986 vec($_,14, 1) = 1 == 16384 00000000000000100000000000000000
6987 vec($_,15, 1) = 1 == 32768 00000000000000010000000000000000
6988 vec($_,16, 1) = 1 == 65536 00000000000000001000000000000000
6989 vec($_,17, 1) = 1 == 131072 00000000000000000100000000000000
6990 vec($_,18, 1) = 1 == 262144 00000000000000000010000000000000
6991 vec($_,19, 1) = 1 == 524288 00000000000000000001000000000000
6992 vec($_,20, 1) = 1 == 1048576 00000000000000000000100000000000
6993 vec($_,21, 1) = 1 == 2097152 00000000000000000000010000000000
6994 vec($_,22, 1) = 1 == 4194304 00000000000000000000001000000000
6995 vec($_,23, 1) = 1 == 8388608 00000000000000000000000100000000
6996 vec($_,24, 1) = 1 == 16777216 00000000000000000000000010000000
6997 vec($_,25, 1) = 1 == 33554432 00000000000000000000000001000000
6998 vec($_,26, 1) = 1 == 67108864 00000000000000000000000000100000
6999 vec($_,27, 1) = 1 == 134217728 00000000000000000000000000010000
7000 vec($_,28, 1) = 1 == 268435456 00000000000000000000000000001000
7001 vec($_,29, 1) = 1 == 536870912 00000000000000000000000000000100
7002 vec($_,30, 1) = 1 == 1073741824 00000000000000000000000000000010
7003 vec($_,31, 1) = 1 == 2147483648 00000000000000000000000000000001
7004 vec($_, 0, 2) = 1 == 1 10000000000000000000000000000000
7005 vec($_, 1, 2) = 1 == 4 00100000000000000000000000000000
7006 vec($_, 2, 2) = 1 == 16 00001000000000000000000000000000
7007 vec($_, 3, 2) = 1 == 64 00000010000000000000000000000000
7008 vec($_, 4, 2) = 1 == 256 00000000100000000000000000000000
7009 vec($_, 5, 2) = 1 == 1024 00000000001000000000000000000000
7010 vec($_, 6, 2) = 1 == 4096 00000000000010000000000000000000
7011 vec($_, 7, 2) = 1 == 16384 00000000000000100000000000000000
7012 vec($_, 8, 2) = 1 == 65536 00000000000000001000000000000000
7013 vec($_, 9, 2) = 1 == 262144 00000000000000000010000000000000
7014 vec($_,10, 2) = 1 == 1048576 00000000000000000000100000000000
7015 vec($_,11, 2) = 1 == 4194304 00000000000000000000001000000000
7016 vec($_,12, 2) = 1 == 16777216 00000000000000000000000010000000
7017 vec($_,13, 2) = 1 == 67108864 00000000000000000000000000100000
7018 vec($_,14, 2) = 1 == 268435456 00000000000000000000000000001000
7019 vec($_,15, 2) = 1 == 1073741824 00000000000000000000000000000010
7020 vec($_, 0, 2) = 2 == 2 01000000000000000000000000000000
7021 vec($_, 1, 2) = 2 == 8 00010000000000000000000000000000
7022 vec($_, 2, 2) = 2 == 32 00000100000000000000000000000000
7023 vec($_, 3, 2) = 2 == 128 00000001000000000000000000000000
7024 vec($_, 4, 2) = 2 == 512 00000000010000000000000000000000
7025 vec($_, 5, 2) = 2 == 2048 00000000000100000000000000000000
7026 vec($_, 6, 2) = 2 == 8192 00000000000001000000000000000000
7027 vec($_, 7, 2) = 2 == 32768 00000000000000010000000000000000
7028 vec($_, 8, 2) = 2 == 131072 00000000000000000100000000000000
7029 vec($_, 9, 2) = 2 == 524288 00000000000000000001000000000000
7030 vec($_,10, 2) = 2 == 2097152 00000000000000000000010000000000
7031 vec($_,11, 2) = 2 == 8388608 00000000000000000000000100000000
7032 vec($_,12, 2) = 2 == 33554432 00000000000000000000000001000000
7033 vec($_,13, 2) = 2 == 134217728 00000000000000000000000000010000
7034 vec($_,14, 2) = 2 == 536870912 00000000000000000000000000000100
7035 vec($_,15, 2) = 2 == 2147483648 00000000000000000000000000000001
7036 vec($_, 0, 4) = 1 == 1 10000000000000000000000000000000
7037 vec($_, 1, 4) = 1 == 16 00001000000000000000000000000000
7038 vec($_, 2, 4) = 1 == 256 00000000100000000000000000000000
7039 vec($_, 3, 4) = 1 == 4096 00000000000010000000000000000000
7040 vec($_, 4, 4) = 1 == 65536 00000000000000001000000000000000
7041 vec($_, 5, 4) = 1 == 1048576 00000000000000000000100000000000
7042 vec($_, 6, 4) = 1 == 16777216 00000000000000000000000010000000
7043 vec($_, 7, 4) = 1 == 268435456 00000000000000000000000000001000
7044 vec($_, 0, 4) = 2 == 2 01000000000000000000000000000000
7045 vec($_, 1, 4) = 2 == 32 00000100000000000000000000000000
7046 vec($_, 2, 4) = 2 == 512 00000000010000000000000000000000
7047 vec($_, 3, 4) = 2 == 8192 00000000000001000000000000000000
7048 vec($_, 4, 4) = 2 == 131072 00000000000000000100000000000000
7049 vec($_, 5, 4) = 2 == 2097152 00000000000000000000010000000000
7050 vec($_, 6, 4) = 2 == 33554432 00000000000000000000000001000000
7051 vec($_, 7, 4) = 2 == 536870912 00000000000000000000000000000100
7052 vec($_, 0, 4) = 4 == 4 00100000000000000000000000000000
7053 vec($_, 1, 4) = 4 == 64 00000010000000000000000000000000
7054 vec($_, 2, 4) = 4 == 1024 00000000001000000000000000000000
7055 vec($_, 3, 4) = 4 == 16384 00000000000000100000000000000000
7056 vec($_, 4, 4) = 4 == 262144 00000000000000000010000000000000
7057 vec($_, 5, 4) = 4 == 4194304 00000000000000000000001000000000
7058 vec($_, 6, 4) = 4 == 67108864 00000000000000000000000000100000
7059 vec($_, 7, 4) = 4 == 1073741824 00000000000000000000000000000010
7060 vec($_, 0, 4) = 8 == 8 00010000000000000000000000000000
7061 vec($_, 1, 4) = 8 == 128 00000001000000000000000000000000
7062 vec($_, 2, 4) = 8 == 2048 00000000000100000000000000000000
7063 vec($_, 3, 4) = 8 == 32768 00000000000000010000000000000000
7064 vec($_, 4, 4) = 8 == 524288 00000000000000000001000000000000
7065 vec($_, 5, 4) = 8 == 8388608 00000000000000000000000100000000
7066 vec($_, 6, 4) = 8 == 134217728 00000000000000000000000000010000
7067 vec($_, 7, 4) = 8 == 2147483648 00000000000000000000000000000001
7068 vec($_, 0, 8) = 1 == 1 10000000000000000000000000000000
7069 vec($_, 1, 8) = 1 == 256 00000000100000000000000000000000
7070 vec($_, 2, 8) = 1 == 65536 00000000000000001000000000000000
7071 vec($_, 3, 8) = 1 == 16777216 00000000000000000000000010000000
7072 vec($_, 0, 8) = 2 == 2 01000000000000000000000000000000
7073 vec($_, 1, 8) = 2 == 512 00000000010000000000000000000000
7074 vec($_, 2, 8) = 2 == 131072 00000000000000000100000000000000
7075 vec($_, 3, 8) = 2 == 33554432 00000000000000000000000001000000
7076 vec($_, 0, 8) = 4 == 4 00100000000000000000000000000000
7077 vec($_, 1, 8) = 4 == 1024 00000000001000000000000000000000
7078 vec($_, 2, 8) = 4 == 262144 00000000000000000010000000000000
7079 vec($_, 3, 8) = 4 == 67108864 00000000000000000000000000100000
7080 vec($_, 0, 8) = 8 == 8 00010000000000000000000000000000
7081 vec($_, 1, 8) = 8 == 2048 00000000000100000000000000000000
7082 vec($_, 2, 8) = 8 == 524288 00000000000000000001000000000000
7083 vec($_, 3, 8) = 8 == 134217728 00000000000000000000000000010000
7084 vec($_, 0, 8) = 16 == 16 00001000000000000000000000000000
7085 vec($_, 1, 8) = 16 == 4096 00000000000010000000000000000000
7086 vec($_, 2, 8) = 16 == 1048576 00000000000000000000100000000000
7087 vec($_, 3, 8) = 16 == 268435456 00000000000000000000000000001000
7088 vec($_, 0, 8) = 32 == 32 00000100000000000000000000000000
7089 vec($_, 1, 8) = 32 == 8192 00000000000001000000000000000000
7090 vec($_, 2, 8) = 32 == 2097152 00000000000000000000010000000000
7091 vec($_, 3, 8) = 32 == 536870912 00000000000000000000000000000100
7092 vec($_, 0, 8) = 64 == 64 00000010000000000000000000000000
7093 vec($_, 1, 8) = 64 == 16384 00000000000000100000000000000000
7094 vec($_, 2, 8) = 64 == 4194304 00000000000000000000001000000000
7095 vec($_, 3, 8) = 64 == 1073741824 00000000000000000000000000000010
7096 vec($_, 0, 8) = 128 == 128 00000001000000000000000000000000
7097 vec($_, 1, 8) = 128 == 32768 00000000000000010000000000000000
7098 vec($_, 2, 8) = 128 == 8388608 00000000000000000000000100000000
7099 vec($_, 3, 8) = 128 == 2147483648 00000000000000000000000000000001
7104 Behaves like the wait(2) system call on your system: it waits for a child
7105 process to terminate and returns the pid of the deceased process, or
7106 C<-1> if there are no child processes. The status is returned in C<$?>
7107 and C<{^CHILD_ERROR_NATIVE}>.
7108 Note that a return value of C<-1> could mean that child processes are
7109 being automatically reaped, as described in L<perlipc>.
7111 =item waitpid PID,FLAGS
7114 Waits for a particular child process to terminate and returns the pid of
7115 the deceased process, or C<-1> if there is no such child process. On some
7116 systems, a value of 0 indicates that there are processes still running.
7117 The status is returned in C<$?> and C<{^CHILD_ERROR_NATIVE}>. If you say
7119 use POSIX ":sys_wait_h";
7122 $kid = waitpid(-1, WNOHANG);
7125 then you can do a non-blocking wait for all pending zombie processes.
7126 Non-blocking wait is available on machines supporting either the
7127 waitpid(2) or wait4(2) system calls. However, waiting for a particular
7128 pid with FLAGS of C<0> is implemented everywhere. (Perl emulates the
7129 system call by remembering the status values of processes that have
7130 exited but have not been harvested by the Perl script yet.)
7132 Note that on some systems, a return value of C<-1> could mean that child
7133 processes are being automatically reaped. See L<perlipc> for details,
7134 and for other examples.
7137 X<wantarray> X<context>
7139 Returns true if the context of the currently executing subroutine or
7140 C<eval> is looking for a list value. Returns false if the context is
7141 looking for a scalar. Returns the undefined value if the context is
7142 looking for no value (void context).
7144 return unless defined wantarray; # don't bother doing more
7145 my @a = complex_calculation();
7146 return wantarray ? @a : "@a";
7148 C<wantarray()>'s result is unspecified in the top level of a file,
7149 in a C<BEGIN>, C<CHECK>, C<INIT> or C<END> block, or in a C<DESTROY>
7152 This function should have been named wantlist() instead.
7155 X<warn> X<warning> X<STDERR>
7157 Produces a message on STDERR just like C<die>, but doesn't exit or throw
7160 If LIST is empty and C<$@> already contains a value (typically from a
7161 previous eval) that value is used after appending C<"\t...caught">
7162 to C<$@>. This is useful for staying almost, but not entirely similar to
7165 If C<$@> is empty then the string C<"Warning: Something's wrong"> is used.
7167 No message is printed if there is a C<$SIG{__WARN__}> handler
7168 installed. It is the handler's responsibility to deal with the message
7169 as it sees fit (like, for instance, converting it into a C<die>). Most
7170 handlers must therefore make arrangements to actually display the
7171 warnings that they are not prepared to deal with, by calling C<warn>
7172 again in the handler. Note that this is quite safe and will not
7173 produce an endless loop, since C<__WARN__> hooks are not called from
7176 You will find this behavior is slightly different from that of
7177 C<$SIG{__DIE__}> handlers (which don't suppress the error text, but can
7178 instead call C<die> again to change it).
7180 Using a C<__WARN__> handler provides a powerful way to silence all
7181 warnings (even the so-called mandatory ones). An example:
7183 # wipe out *all* compile-time warnings
7184 BEGIN { $SIG{'__WARN__'} = sub { warn $_[0] if $DOWARN } }
7186 my $foo = 20; # no warning about duplicate my $foo,
7187 # but hey, you asked for it!
7188 # no compile-time or run-time warnings before here
7191 # run-time warnings enabled after here
7192 warn "\$foo is alive and $foo!"; # does show up
7194 See L<perlvar> for details on setting C<%SIG> entries, and for more
7195 examples. See the Carp module for other kinds of warnings using its
7196 carp() and cluck() functions.
7198 =item write FILEHANDLE
7205 Writes a formatted record (possibly multi-line) to the specified FILEHANDLE,
7206 using the format associated with that file. By default the format for
7207 a file is the one having the same name as the filehandle, but the
7208 format for the current output channel (see the C<select> function) may be set
7209 explicitly by assigning the name of the format to the C<$~> variable.
7211 Top of form processing is handled automatically: if there is
7212 insufficient room on the current page for the formatted record, the
7213 page is advanced by writing a form feed, a special top-of-page format
7214 is used to format the new page header, and then the record is written.
7215 By default the top-of-page format is the name of the filehandle with
7216 "_TOP" appended, but it may be dynamically set to the format of your
7217 choice by assigning the name to the C<$^> variable while the filehandle is
7218 selected. The number of lines remaining on the current page is in
7219 variable C<$->, which can be set to C<0> to force a new page.
7221 If FILEHANDLE is unspecified, output goes to the current default output
7222 channel, which starts out as STDOUT but may be changed by the
7223 C<select> operator. If the FILEHANDLE is an EXPR, then the expression
7224 is evaluated and the resulting string is used to look up the name of
7225 the FILEHANDLE at run time. For more on formats, see L<perlform>.
7227 Note that write is I<not> the opposite of C<read>. Unfortunately.
7231 The transliteration operator. Same as C<tr///>. See L<perlop>.