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<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 scoping
161 C<caller>, C<import>, C<local>, C<my>, C<our>, C<package>, C<use>
163 =item Miscellaneous functions
165 C<defined>, C<dump>, C<eval>, C<formline>, C<local>, C<my>, C<our>, C<reset>,
166 C<scalar>, C<undef>, C<wantarray>
168 =item Functions for processes and process groups
169 X<process> X<pid> X<process id>
171 C<alarm>, C<exec>, C<fork>, C<getpgrp>, C<getppid>, C<getpriority>, C<kill>,
172 C<pipe>, C<qx/STRING/>, C<setpgrp>, C<setpriority>, C<sleep>, C<system>,
173 C<times>, C<wait>, C<waitpid>
175 =item Keywords related to perl modules
178 C<do>, C<import>, C<no>, C<package>, C<require>, C<use>
180 =item Keywords related to classes and object-orientedness
181 X<object> X<class> X<package>
183 C<bless>, C<dbmclose>, C<dbmopen>, C<package>, C<ref>, C<tie>, C<tied>,
186 =item Low-level socket functions
189 C<accept>, C<bind>, C<connect>, C<getpeername>, C<getsockname>,
190 C<getsockopt>, C<listen>, C<recv>, C<send>, C<setsockopt>, C<shutdown>,
191 C<socket>, C<socketpair>
193 =item System V interprocess communication functions
194 X<IPC> X<System V> X<semaphore> X<shared memory> X<memory> X<message>
196 C<msgctl>, C<msgget>, C<msgrcv>, C<msgsnd>, C<semctl>, C<semget>, C<semop>,
197 C<shmctl>, C<shmget>, C<shmread>, C<shmwrite>
199 =item Fetching user and group info
200 X<user> X<group> X<password> X<uid> X<gid> X<passwd> X</etc/passwd>
202 C<endgrent>, C<endhostent>, C<endnetent>, C<endpwent>, C<getgrent>,
203 C<getgrgid>, C<getgrnam>, C<getlogin>, C<getpwent>, C<getpwnam>,
204 C<getpwuid>, C<setgrent>, C<setpwent>
206 =item Fetching network info
207 X<network> X<protocol> X<host> X<hostname> X<IP> X<address> X<service>
209 C<endprotoent>, C<endservent>, C<gethostbyaddr>, C<gethostbyname>,
210 C<gethostent>, C<getnetbyaddr>, C<getnetbyname>, C<getnetent>,
211 C<getprotobyname>, C<getprotobynumber>, C<getprotoent>,
212 C<getservbyname>, C<getservbyport>, C<getservent>, C<sethostent>,
213 C<setnetent>, C<setprotoent>, C<setservent>
215 =item Time-related functions
218 C<gmtime>, C<localtime>, C<time>, C<times>
220 =item Functions new in perl5
223 C<abs>, C<bless>, C<chomp>, C<chr>, C<exists>, C<formline>, C<glob>,
224 C<import>, C<lc>, C<lcfirst>, C<map>, C<my>, C<no>, C<our>, C<prototype>,
225 C<qx>, C<qw>, C<readline>, C<readpipe>, C<ref>, C<sub*>, C<sysopen>, C<tie>,
226 C<tied>, C<uc>, C<ucfirst>, C<untie>, C<use>
228 * - C<sub> was a keyword in perl4, but in perl5 it is an
229 operator, which can be used in expressions.
231 =item Functions obsoleted in perl5
233 C<dbmclose>, C<dbmopen>
238 X<portability> X<Unix> X<portable>
240 Perl was born in Unix and can therefore access all common Unix
241 system calls. In non-Unix environments, the functionality of some
242 Unix system calls may not be available, or details of the available
243 functionality may differ slightly. The Perl functions affected
246 C<-X>, C<binmode>, C<chmod>, C<chown>, C<chroot>, C<crypt>,
247 C<dbmclose>, C<dbmopen>, C<dump>, C<endgrent>, C<endhostent>,
248 C<endnetent>, C<endprotoent>, C<endpwent>, C<endservent>, C<exec>,
249 C<fcntl>, C<flock>, C<fork>, C<getgrent>, C<getgrgid>, C<gethostbyname>,
250 C<gethostent>, C<getlogin>, C<getnetbyaddr>, C<getnetbyname>, C<getnetent>,
251 C<getppid>, C<getpgrp>, C<getpriority>, C<getprotobynumber>,
252 C<getprotoent>, C<getpwent>, C<getpwnam>, C<getpwuid>,
253 C<getservbyport>, C<getservent>, C<getsockopt>, C<glob>, C<ioctl>,
254 C<kill>, C<link>, C<lstat>, C<msgctl>, C<msgget>, C<msgrcv>,
255 C<msgsnd>, C<open>, C<pipe>, C<readlink>, C<rename>, C<select>, C<semctl>,
256 C<semget>, C<semop>, C<setgrent>, C<sethostent>, C<setnetent>,
257 C<setpgrp>, C<setpriority>, C<setprotoent>, C<setpwent>,
258 C<setservent>, C<setsockopt>, C<shmctl>, C<shmget>, C<shmread>,
259 C<shmwrite>, C<socket>, C<socketpair>,
260 C<stat>, C<symlink>, C<syscall>, C<sysopen>, C<system>,
261 C<times>, C<truncate>, C<umask>, C<unlink>,
262 C<utime>, C<wait>, C<waitpid>
264 For more information about the portability of these functions, see
265 L<perlport> and other available platform-specific documentation.
267 =head2 Alphabetical Listing of Perl Functions
272 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>
273 X<-S>X<-b>X<-c>X<-t>X<-u>X<-g>X<-k>X<-T>X<-B>X<-M>X<-A>X<-C>
279 A file test, where X is one of the letters listed below. This unary
280 operator takes one argument, either a filename or a filehandle, and
281 tests the associated file to see if something is true about it. If the
282 argument is omitted, tests C<$_>, except for C<-t>, which tests STDIN.
283 Unless otherwise documented, it returns C<1> for true and C<''> for false, or
284 the undefined value if the file doesn't exist. Despite the funny
285 names, precedence is the same as any other named unary operator, and
286 the argument may be parenthesized like any other unary operator. The
287 operator may be any of:
289 -r File is readable by effective uid/gid.
290 -w File is writable by effective uid/gid.
291 -x File is executable by effective uid/gid.
292 -o File is owned by effective uid.
294 -R File is readable by real uid/gid.
295 -W File is writable by real uid/gid.
296 -X File is executable by real uid/gid.
297 -O File is owned by real uid.
300 -z File has zero size (is empty).
301 -s File has nonzero size (returns size in bytes).
303 -f File is a plain file.
304 -d File is a directory.
305 -l File is a symbolic link.
306 -p File is a named pipe (FIFO), or Filehandle is a pipe.
308 -b File is a block special file.
309 -c File is a character special file.
310 -t Filehandle is opened to a tty.
312 -u File has setuid bit set.
313 -g File has setgid bit set.
314 -k File has sticky bit set.
316 -T File is an ASCII text file (heuristic guess).
317 -B File is a "binary" file (opposite of -T).
319 -M Script start time minus file modification time, in days.
320 -A Same for access time.
321 -C Same for inode change time (Unix, may differ for other platforms)
327 next unless -f $_; # ignore specials
331 The interpretation of the file permission operators C<-r>, C<-R>,
332 C<-w>, C<-W>, C<-x>, and C<-X> is by default based solely on the mode
333 of the file and the uids and gids of the user. There may be other
334 reasons you can't actually read, write, or execute the file. Such
335 reasons may be for example network filesystem access controls, ACLs
336 (access control lists), read-only filesystems, and unrecognized
339 Also note that, for the superuser on the local filesystems, the C<-r>,
340 C<-R>, C<-w>, and C<-W> tests always return 1, and C<-x> and C<-X> return 1
341 if any execute bit is set in the mode. Scripts run by the superuser
342 may thus need to do a stat() to determine the actual mode of the file,
343 or temporarily set their effective uid to something else.
345 If you are using ACLs, there is a pragma called C<filetest> that may
346 produce more accurate results than the bare stat() mode bits.
347 When under the C<use filetest 'access'> the above-mentioned filetests
348 will test whether the permission can (not) be granted using the
349 access() family of system calls. Also note that the C<-x> and C<-X> may
350 under this pragma return true even if there are no execute permission
351 bits set (nor any extra execute permission ACLs). This strangeness is
352 due to the underlying system calls' definitions. Read the
353 documentation for the C<filetest> pragma for more information.
355 Note that C<-s/a/b/> does not do a negated substitution. Saying
356 C<-exp($foo)> still works as expected, however--only single letters
357 following a minus are interpreted as file tests.
359 The C<-T> and C<-B> switches work as follows. The first block or so of the
360 file is examined for odd characters such as strange control codes or
361 characters with the high bit set. If too many strange characters (>30%)
362 are found, it's a C<-B> file; otherwise it's a C<-T> file. Also, any file
363 containing null in the first block is considered a binary file. If C<-T>
364 or C<-B> is used on a filehandle, the current IO buffer is examined
365 rather than the first block. Both C<-T> and C<-B> return true on a null
366 file, or a file at EOF when testing a filehandle. Because you have to
367 read a file to do the C<-T> test, on most occasions you want to use a C<-f>
368 against the file first, as in C<next unless -f $file && -T $file>.
370 If any of the file tests (or either the C<stat> or C<lstat> operators) are given
371 the special filehandle consisting of a solitary underline, then the stat
372 structure of the previous file test (or stat operator) is used, saving
373 a system call. (This doesn't work with C<-t>, and you need to remember
374 that lstat() and C<-l> will leave values in the stat structure for the
375 symbolic link, not the real file.) (Also, if the stat buffer was filled by
376 an C<lstat> call, C<-T> and C<-B> will reset it with the results of C<stat _>).
379 print "Can do.\n" if -r $a || -w _ || -x _;
382 print "Readable\n" if -r _;
383 print "Writable\n" if -w _;
384 print "Executable\n" if -x _;
385 print "Setuid\n" if -u _;
386 print "Setgid\n" if -g _;
387 print "Sticky\n" if -k _;
388 print "Text\n" if -T _;
389 print "Binary\n" if -B _;
391 As of Perl 5.9.1, as a form of purely syntactic sugar, you can stack file
392 test operators, in a way that C<-f -w -x $file> is equivalent to
393 C<-x $file && -w _ && -f _>. (This is only syntax fancy: if you use
394 the return value of C<-f $file> as an argument to another filetest
395 operator, no special magic will happen.)
402 Returns the absolute value of its argument.
403 If VALUE is omitted, uses C<$_>.
405 =item accept NEWSOCKET,GENERICSOCKET
408 Accepts an incoming socket connect, just as the accept(2) system call
409 does. Returns the packed address if it succeeded, false otherwise.
410 See the example in L<perlipc/"Sockets: Client/Server Communication">.
412 On systems that support a close-on-exec flag on files, the flag will
413 be set for the newly opened file descriptor, as determined by the
414 value of $^F. See L<perlvar/$^F>.
423 Arranges to have a SIGALRM delivered to this process after the
424 specified number of wallclock seconds has elapsed. If SECONDS is not
425 specified, the value stored in C<$_> is used. (On some machines,
426 unfortunately, the elapsed time may be up to one second less or more
427 than you specified because of how seconds are counted, and process
428 scheduling may delay the delivery of the signal even further.)
430 Only one timer may be counting at once. Each call disables the
431 previous timer, and an argument of C<0> may be supplied to cancel the
432 previous timer without starting a new one. The returned value is the
433 amount of time remaining on the previous timer.
435 For delays of finer granularity than one second, you may use Perl's
436 four-argument version of select() leaving the first three arguments
437 undefined, or you might be able to use the C<syscall> interface to
438 access setitimer(2) if your system supports it. The Time::HiRes
439 module (from CPAN, and starting from Perl 5.8 part of the standard
440 distribution) may also prove useful.
442 It is usually a mistake to intermix C<alarm> and C<sleep> calls.
443 (C<sleep> may be internally implemented in your system with C<alarm>)
445 If you want to use C<alarm> to time out a system call you need to use an
446 C<eval>/C<die> pair. You can't rely on the alarm causing the system call to
447 fail with C<$!> set to C<EINTR> because Perl sets up signal handlers to
448 restart system calls on some systems. Using C<eval>/C<die> always works,
449 modulo the caveats given in L<perlipc/"Signals">.
452 local $SIG{ALRM} = sub { die "alarm\n" }; # NB: \n required
454 $nread = sysread SOCKET, $buffer, $size;
458 die unless $@ eq "alarm\n"; # propagate unexpected errors
465 For more information see L<perlipc>.
468 X<atan2> X<arctangent> X<tan> X<tangent>
470 Returns the arctangent of Y/X in the range -PI to PI.
472 For the tangent operation, you may use the C<Math::Trig::tan>
473 function, or use the familiar relation:
475 sub tan { sin($_[0]) / cos($_[0]) }
477 Note that atan2(0, 0) is not well-defined.
479 =item bind SOCKET,NAME
482 Binds a network address to a socket, just as the bind system call
483 does. Returns true if it succeeded, false otherwise. NAME should be a
484 packed address of the appropriate type for the socket. See the examples in
485 L<perlipc/"Sockets: Client/Server Communication">.
487 =item binmode FILEHANDLE, LAYER
488 X<binmode> X<binary> X<text> X<DOS> X<Windows>
490 =item binmode FILEHANDLE
492 Arranges for FILEHANDLE to be read or written in "binary" or "text"
493 mode on systems where the run-time libraries distinguish between
494 binary and text files. If FILEHANDLE is an expression, the value is
495 taken as the name of the filehandle. Returns true on success,
496 otherwise it returns C<undef> and sets C<$!> (errno).
498 On some systems (in general, DOS and Windows-based systems) binmode()
499 is necessary when you're not working with a text file. For the sake
500 of portability it is a good idea to always use it when appropriate,
501 and to never use it when it isn't appropriate. Also, people can
502 set their I/O to be by default UTF-8 encoded Unicode, not bytes.
504 In other words: regardless of platform, use binmode() on binary data,
505 like for example images.
507 If LAYER is present it is a single string, but may contain multiple
508 directives. The directives alter the behaviour of the file handle.
509 When LAYER is present using binmode on text file makes sense.
511 If LAYER is omitted or specified as C<:raw> the filehandle is made
512 suitable for passing binary data. This includes turning off possible CRLF
513 translation and marking it as bytes (as opposed to Unicode characters).
514 Note that, despite what may be implied in I<"Programming Perl"> (the
515 Camel) or elsewhere, C<:raw> is I<not> the simply inverse of C<:crlf>
516 -- other layers which would affect binary nature of the stream are
517 I<also> disabled. See L<PerlIO>, L<perlrun> and the discussion about the
518 PERLIO environment variable.
520 The C<:bytes>, C<:crlf>, and C<:utf8>, and any other directives of the
521 form C<:...>, are called I/O I<layers>. The C<open> pragma can be used to
522 establish default I/O layers. See L<open>.
524 I<The LAYER parameter of the binmode() function is described as "DISCIPLINE"
525 in "Programming Perl, 3rd Edition". However, since the publishing of this
526 book, by many known as "Camel III", the consensus of the naming of this
527 functionality has moved from "discipline" to "layer". All documentation
528 of this version of Perl therefore refers to "layers" rather than to
529 "disciplines". Now back to the regularly scheduled documentation...>
531 To mark FILEHANDLE as UTF-8, use C<:utf8>.
533 In general, binmode() should be called after open() but before any I/O
534 is done on the filehandle. Calling binmode() will normally flush any
535 pending buffered output data (and perhaps pending input data) on the
536 handle. An exception to this is the C<:encoding> layer that
537 changes the default character encoding of the handle, see L<open>.
538 The C<:encoding> layer sometimes needs to be called in
539 mid-stream, and it doesn't flush the stream. The C<:encoding>
540 also implicitly pushes on top of itself the C<:utf8> layer because
541 internally Perl will operate on UTF-8 encoded Unicode characters.
543 The operating system, device drivers, C libraries, and Perl run-time
544 system all work together to let the programmer treat a single
545 character (C<\n>) as the line terminator, irrespective of the external
546 representation. On many operating systems, the native text file
547 representation matches the internal representation, but on some
548 platforms the external representation of C<\n> is made up of more than
551 Mac OS, all variants of Unix, and Stream_LF files on VMS use a single
552 character to end each line in the external representation of text (even
553 though that single character is CARRIAGE RETURN on Mac OS and LINE FEED
554 on Unix and most VMS files). In other systems like OS/2, DOS and the
555 various flavors of MS-Windows your program sees a C<\n> as a simple C<\cJ>,
556 but what's stored in text files are the two characters C<\cM\cJ>. That
557 means that, if you don't use binmode() on these systems, C<\cM\cJ>
558 sequences on disk will be converted to C<\n> on input, and any C<\n> in
559 your program will be converted back to C<\cM\cJ> on output. This is what
560 you want for text files, but it can be disastrous for binary files.
562 Another consequence of using binmode() (on some systems) is that
563 special end-of-file markers will be seen as part of the data stream.
564 For systems from the Microsoft family this means that if your binary
565 data contains C<\cZ>, the I/O subsystem will regard it as the end of
566 the file, unless you use binmode().
568 binmode() is not only important for readline() and print() operations,
569 but also when using read(), seek(), sysread(), syswrite() and tell()
570 (see L<perlport> for more details). See the C<$/> and C<$\> variables
571 in L<perlvar> for how to manually set your input and output
572 line-termination sequences.
574 =item bless REF,CLASSNAME
579 This function tells the thingy referenced by REF that it is now an object
580 in the CLASSNAME package. If CLASSNAME is omitted, the current package
581 is used. Because a C<bless> is often the last thing in a constructor,
582 it returns the reference for convenience. Always use the two-argument
583 version if a derived class might inherit the function doing the blessing.
584 See L<perltoot> and L<perlobj> for more about the blessing (and blessings)
587 Consider always blessing objects in CLASSNAMEs that are mixed case.
588 Namespaces with all lowercase names are considered reserved for
589 Perl pragmata. Builtin types have all uppercase names. To prevent
590 confusion, you may wish to avoid such package names as well. Make sure
591 that CLASSNAME is a true value.
593 See L<perlmod/"Perl Modules">.
596 X<caller> X<call stack> X<stack> X<stack trace>
600 Returns the context of the current subroutine call. In scalar context,
601 returns the caller's package name if there is a caller, that is, if
602 we're in a subroutine or C<eval> or C<require>, and the undefined value
603 otherwise. In list context, returns
605 ($package, $filename, $line) = caller;
607 With EXPR, it returns some extra information that the debugger uses to
608 print a stack trace. The value of EXPR indicates how many call frames
609 to go back before the current one.
611 ($package, $filename, $line, $subroutine, $hasargs,
612 $wantarray, $evaltext, $is_require, $hints, $bitmask) = caller($i);
614 Here $subroutine may be C<(eval)> if the frame is not a subroutine
615 call, but an C<eval>. In such a case additional elements $evaltext and
616 C<$is_require> are set: C<$is_require> is true if the frame is created by a
617 C<require> or C<use> statement, $evaltext contains the text of the
618 C<eval EXPR> statement. In particular, for an C<eval BLOCK> statement,
619 $filename is C<(eval)>, but $evaltext is undefined. (Note also that
620 each C<use> statement creates a C<require> frame inside an C<eval EXPR>
621 frame.) $subroutine may also be C<(unknown)> if this particular
622 subroutine happens to have been deleted from the symbol table.
623 C<$hasargs> is true if a new instance of C<@_> was set up for the frame.
624 C<$hints> and C<$bitmask> contain pragmatic hints that the caller was
625 compiled with. The C<$hints> and C<$bitmask> values are subject to change
626 between versions of Perl, and are not meant for external use.
628 Furthermore, when called from within the DB package, caller returns more
629 detailed information: it sets the list variable C<@DB::args> to be the
630 arguments with which the subroutine was invoked.
632 Be aware that the optimizer might have optimized call frames away before
633 C<caller> had a chance to get the information. That means that C<caller(N)>
634 might not return information about the call frame you expect it do, for
635 C<< N > 1 >>. In particular, C<@DB::args> might have information from the
636 previous time C<caller> was called.
642 =item chdir FILEHANDLE
644 =item chdir DIRHANDLE
648 Changes the working directory to EXPR, if possible. If EXPR is omitted,
649 changes to the directory specified by C<$ENV{HOME}>, if set; if not,
650 changes to the directory specified by C<$ENV{LOGDIR}>. (Under VMS, the
651 variable C<$ENV{SYS$LOGIN}> is also checked, and used if it is set.) If
652 neither is set, C<chdir> does nothing. It returns true upon success,
653 false otherwise. See the example under C<die>.
655 On systems that support fchdir, you might pass a file handle or
656 directory handle as argument. On systems that don't support fchdir,
657 passing handles produces a fatal error at run time.
660 X<chmod> X<permission> X<mode>
662 Changes the permissions of a list of files. The first element of the
663 list must be the numerical mode, which should probably be an octal
664 number, and which definitely should I<not> be a string of octal digits:
665 C<0644> is okay, C<'0644'> is not. Returns the number of files
666 successfully changed. See also L</oct>, if all you have is a string.
668 $cnt = chmod 0755, 'foo', 'bar';
669 chmod 0755, @executables;
670 $mode = '0644'; chmod $mode, 'foo'; # !!! sets mode to
672 $mode = '0644'; chmod oct($mode), 'foo'; # this is better
673 $mode = 0644; chmod $mode, 'foo'; # this is best
675 On systems that support fchmod, you might pass file handles among the
676 files. On systems that don't support fchmod, passing file handles
677 produces a fatal error at run time.
679 open(my $fh, "<", "foo");
680 my $perm = (stat $fh)[2] & 07777;
681 chmod($perm | 0600, $fh);
683 You can also import the symbolic C<S_I*> constants from the Fcntl
688 chmod S_IRWXU|S_IRGRP|S_IXGRP|S_IROTH|S_IXOTH, @executables;
689 # This is identical to the chmod 0755 of the above example.
692 X<chomp> X<INPUT_RECORD_SEPARATOR> X<$/> X<newline> X<eol>
698 This safer version of L</chop> removes any trailing string
699 that corresponds to the current value of C<$/> (also known as
700 $INPUT_RECORD_SEPARATOR in the C<English> module). It returns the total
701 number of characters removed from all its arguments. It's often used to
702 remove the newline from the end of an input record when you're worried
703 that the final record may be missing its newline. When in paragraph
704 mode (C<$/ = "">), it removes all trailing newlines from the string.
705 When in slurp mode (C<$/ = undef>) or fixed-length record mode (C<$/> is
706 a reference to an integer or the like, see L<perlvar>) chomp() won't
708 If VARIABLE is omitted, it chomps C<$_>. Example:
711 chomp; # avoid \n on last field
716 If VARIABLE is a hash, it chomps the hash's values, but not its keys.
718 You can actually chomp anything that's an lvalue, including an assignment:
721 chomp($answer = <STDIN>);
723 If you chomp a list, each element is chomped, and the total number of
724 characters removed is returned.
726 If the C<encoding> pragma is in scope then the lengths returned are
727 calculated from the length of C<$/> in Unicode characters, which is not
728 always the same as the length of C<$/> in the native encoding.
730 Note that parentheses are necessary when you're chomping anything
731 that is not a simple variable. This is because C<chomp $cwd = `pwd`;>
732 is interpreted as C<(chomp $cwd) = `pwd`;>, rather than as
733 C<chomp( $cwd = `pwd` )> which you might expect. Similarly,
734 C<chomp $a, $b> is interpreted as C<chomp($a), $b> rather than
744 Chops off the last character of a string and returns the character
745 chopped. It is much more efficient than C<s/.$//s> because it neither
746 scans nor copies the string. If VARIABLE is omitted, chops C<$_>.
747 If VARIABLE is a hash, it chops the hash's values, but not its keys.
749 You can actually chop anything that's an lvalue, including an assignment.
751 If you chop a list, each element is chopped. Only the value of the
752 last C<chop> is returned.
754 Note that C<chop> returns the last character. To return all but the last
755 character, use C<substr($string, 0, -1)>.
760 X<chown> X<owner> X<user> X<group>
762 Changes the owner (and group) of a list of files. The first two
763 elements of the list must be the I<numeric> uid and gid, in that
764 order. A value of -1 in either position is interpreted by most
765 systems to leave that value unchanged. Returns the number of files
766 successfully changed.
768 $cnt = chown $uid, $gid, 'foo', 'bar';
769 chown $uid, $gid, @filenames;
771 On systems that support fchown, you might pass file handles among the
772 files. On systems that don't support fchown, passing file handles
773 produces a fatal error at run time.
775 Here's an example that looks up nonnumeric uids in the passwd file:
778 chomp($user = <STDIN>);
780 chomp($pattern = <STDIN>);
782 ($login,$pass,$uid,$gid) = getpwnam($user)
783 or die "$user not in passwd file";
785 @ary = glob($pattern); # expand filenames
786 chown $uid, $gid, @ary;
788 On most systems, you are not allowed to change the ownership of the
789 file unless you're the superuser, although you should be able to change
790 the group to any of your secondary groups. On insecure systems, these
791 restrictions may be relaxed, but this is not a portable assumption.
792 On POSIX systems, you can detect this condition this way:
794 use POSIX qw(sysconf _PC_CHOWN_RESTRICTED);
795 $can_chown_giveaway = not sysconf(_PC_CHOWN_RESTRICTED);
798 X<chr> X<character> X<ASCII> X<Unicode>
802 Returns the character represented by that NUMBER in the character set.
803 For example, C<chr(65)> is C<"A"> in either ASCII or Unicode, and
804 chr(0x263a) is a Unicode smiley face. Note that characters from 128
805 to 255 (inclusive) are by default not encoded in UTF-8 Unicode for
806 backward compatibility reasons (but see L<encoding>).
808 Negative values give the Unicode replacement character (chr(0xfffd)),
809 except under the L<bytes> pragma, where low eight bits of the value
810 (truncated to an integer) are used.
812 If NUMBER is omitted, uses C<$_>.
814 For the reverse, use L</ord>.
816 Note that under the C<bytes> pragma the NUMBER is masked to
819 See L<perlunicode> and L<encoding> for more about Unicode.
821 =item chroot FILENAME
826 This function works like the system call by the same name: it makes the
827 named directory the new root directory for all further pathnames that
828 begin with a C</> by your process and all its children. (It doesn't
829 change your current working directory, which is unaffected.) For security
830 reasons, this call is restricted to the superuser. If FILENAME is
831 omitted, does a C<chroot> to C<$_>.
833 =item close FILEHANDLE
838 Closes the file or pipe associated with the file handle, returning
839 true only if IO buffers are successfully flushed and closes the system
840 file descriptor. Closes the currently selected filehandle if the
843 You don't have to close FILEHANDLE if you are immediately going to do
844 another C<open> on it, because C<open> will close it for you. (See
845 C<open>.) However, an explicit C<close> on an input file resets the line
846 counter (C<$.>), while the implicit close done by C<open> does not.
848 If the file handle came from a piped open, C<close> will additionally
849 return false if one of the other system calls involved fails, or if the
850 program exits with non-zero status. (If the only problem was that the
851 program exited non-zero, C<$!> will be set to C<0>.) Closing a pipe
852 also waits for the process executing on the pipe to complete, in case you
853 want to look at the output of the pipe afterwards, and
854 implicitly puts the exit status value of that command into C<$?> and
855 C<${^CHILD_ERROR_NATIVE}>.
857 Prematurely closing the read end of a pipe (i.e. before the process
858 writing to it at the other end has closed it) will result in a
859 SIGPIPE being delivered to the writer. If the other end can't
860 handle that, be sure to read all the data before closing the pipe.
864 open(OUTPUT, '|sort >foo') # pipe to sort
865 or die "Can't start sort: $!";
866 #... # print stuff to output
867 close OUTPUT # wait for sort to finish
868 or warn $! ? "Error closing sort pipe: $!"
869 : "Exit status $? from sort";
870 open(INPUT, 'foo') # get sort's results
871 or die "Can't open 'foo' for input: $!";
873 FILEHANDLE may be an expression whose value can be used as an indirect
874 filehandle, usually the real filehandle name.
876 =item closedir DIRHANDLE
879 Closes a directory opened by C<opendir> and returns the success of that
882 =item connect SOCKET,NAME
885 Attempts to connect to a remote socket, just as the connect system call
886 does. Returns true if it succeeded, false otherwise. NAME should be a
887 packed address of the appropriate type for the socket. See the examples in
888 L<perlipc/"Sockets: Client/Server Communication">.
893 C<continue> is actually a flow control statement rather than a function. If
894 there is a C<continue> BLOCK attached to a BLOCK (typically in a C<while> or
895 C<foreach>), it is always executed just before the conditional is about to
896 be evaluated again, just like the third part of a C<for> loop in C. Thus
897 it can be used to increment a loop variable, even when the loop has been
898 continued via the C<next> statement (which is similar to the C C<continue>
901 C<last>, C<next>, or C<redo> may appear within a C<continue>
902 block. C<last> and C<redo> will behave as if they had been executed within
903 the main block. So will C<next>, but since it will execute a C<continue>
904 block, it may be more entertaining.
907 ### redo always comes here
910 ### next always comes here
912 # then back the top to re-check EXPR
914 ### last always comes here
916 Omitting the C<continue> section is semantically equivalent to using an
917 empty one, logically enough. In that case, C<next> goes directly back
918 to check the condition at the top of the loop.
921 X<cos> X<cosine> X<acos> X<arccosine>
925 Returns the cosine of EXPR (expressed in radians). If EXPR is omitted,
926 takes cosine of C<$_>.
928 For the inverse cosine operation, you may use the C<Math::Trig::acos()>
929 function, or use this relation:
931 sub acos { atan2( sqrt(1 - $_[0] * $_[0]), $_[0] ) }
933 =item crypt PLAINTEXT,SALT
934 X<crypt> X<digest> X<hash> X<salt> X<plaintext> X<password>
935 X<decrypt> X<cryptography> X<passwd>
937 Creates a digest string exactly like the crypt(3) function in the C
938 library (assuming that you actually have a version there that has not
939 been extirpated as a potential munitions).
941 crypt() is a one-way hash function. The PLAINTEXT and SALT is turned
942 into a short string, called a digest, which is returned. The same
943 PLAINTEXT and SALT will always return the same string, but there is no
944 (known) way to get the original PLAINTEXT from the hash. Small
945 changes in the PLAINTEXT or SALT will result in large changes in the
948 There is no decrypt function. This function isn't all that useful for
949 cryptography (for that, look for F<Crypt> modules on your nearby CPAN
950 mirror) and the name "crypt" is a bit of a misnomer. Instead it is
951 primarily used to check if two pieces of text are the same without
952 having to transmit or store the text itself. An example is checking
953 if a correct password is given. The digest of the password is stored,
954 not the password itself. The user types in a password that is
955 crypt()'d with the same salt as the stored digest. If the two digests
956 match the password is correct.
958 When verifying an existing digest string you should use the digest as
959 the salt (like C<crypt($plain, $digest) eq $digest>). The SALT used
960 to create the digest is visible as part of the digest. This ensures
961 crypt() will hash the new string with the same salt as the digest.
962 This allows your code to work with the standard L<crypt|/crypt> and
963 with more exotic implementations. In other words, do not assume
964 anything about the returned string itself, or how many bytes in the
967 Traditionally the result is a string of 13 bytes: two first bytes of
968 the salt, followed by 11 bytes from the set C<[./0-9A-Za-z]>, and only
969 the first eight bytes of the digest string mattered, but alternative
970 hashing schemes (like MD5), higher level security schemes (like C2),
971 and implementations on non-UNIX platforms may produce different
974 When choosing a new salt create a random two character string whose
975 characters come from the set C<[./0-9A-Za-z]> (like C<join '', ('.',
976 '/', 0..9, 'A'..'Z', 'a'..'z')[rand 64, rand 64]>). This set of
977 characters is just a recommendation; the characters allowed in
978 the salt depend solely on your system's crypt library, and Perl can't
979 restrict what salts C<crypt()> accepts.
981 Here's an example that makes sure that whoever runs this program knows
984 $pwd = (getpwuid($<))[1];
988 chomp($word = <STDIN>);
992 if (crypt($word, $pwd) ne $pwd) {
998 Of course, typing in your own password to whoever asks you
1001 The L<crypt|/crypt> function is unsuitable for hashing large quantities
1002 of data, not least of all because you can't get the information
1003 back. Look at the L<Digest> module for more robust algorithms.
1005 If using crypt() on a Unicode string (which I<potentially> has
1006 characters with codepoints above 255), Perl tries to make sense
1007 of the situation by trying to downgrade (a copy of the string)
1008 the string back to an eight-bit byte string before calling crypt()
1009 (on that copy). If that works, good. If not, crypt() dies with
1010 C<Wide character in crypt>.
1015 [This function has been largely superseded by the C<untie> function.]
1017 Breaks the binding between a DBM file and a hash.
1019 =item dbmopen HASH,DBNAME,MASK
1020 X<dbmopen> X<dbm> X<ndbm> X<sdbm> X<gdbm>
1022 [This function has been largely superseded by the C<tie> function.]
1024 This binds a dbm(3), ndbm(3), sdbm(3), gdbm(3), or Berkeley DB file to a
1025 hash. HASH is the name of the hash. (Unlike normal C<open>, the first
1026 argument is I<not> a filehandle, even though it looks like one). DBNAME
1027 is the name of the database (without the F<.dir> or F<.pag> extension if
1028 any). If the database does not exist, it is created with protection
1029 specified by MASK (as modified by the C<umask>). If your system supports
1030 only the older DBM functions, you may perform only one C<dbmopen> in your
1031 program. In older versions of Perl, if your system had neither DBM nor
1032 ndbm, calling C<dbmopen> produced a fatal error; it now falls back to
1035 If you don't have write access to the DBM file, you can only read hash
1036 variables, not set them. If you want to test whether you can write,
1037 either use file tests or try setting a dummy hash entry inside an C<eval>,
1038 which will trap the error.
1040 Note that functions such as C<keys> and C<values> may return huge lists
1041 when used on large DBM files. You may prefer to use the C<each>
1042 function to iterate over large DBM files. Example:
1044 # print out history file offsets
1045 dbmopen(%HIST,'/usr/lib/news/history',0666);
1046 while (($key,$val) = each %HIST) {
1047 print $key, ' = ', unpack('L',$val), "\n";
1051 See also L<AnyDBM_File> for a more general description of the pros and
1052 cons of the various dbm approaches, as well as L<DB_File> for a particularly
1053 rich implementation.
1055 You can control which DBM library you use by loading that library
1056 before you call dbmopen():
1059 dbmopen(%NS_Hist, "$ENV{HOME}/.netscape/history.db")
1060 or die "Can't open netscape history file: $!";
1063 X<defined> X<undef> X<undefined>
1067 Returns a Boolean value telling whether EXPR has a value other than
1068 the undefined value C<undef>. If EXPR is not present, C<$_> will be
1071 Many operations return C<undef> to indicate failure, end of file,
1072 system error, uninitialized variable, and other exceptional
1073 conditions. This function allows you to distinguish C<undef> from
1074 other values. (A simple Boolean test will not distinguish among
1075 C<undef>, zero, the empty string, and C<"0">, which are all equally
1076 false.) Note that since C<undef> is a valid scalar, its presence
1077 doesn't I<necessarily> indicate an exceptional condition: C<pop>
1078 returns C<undef> when its argument is an empty array, I<or> when the
1079 element to return happens to be C<undef>.
1081 You may also use C<defined(&func)> to check whether subroutine C<&func>
1082 has ever been defined. The return value is unaffected by any forward
1083 declarations of C<&func>. Note that a subroutine which is not defined
1084 may still be callable: its package may have an C<AUTOLOAD> method that
1085 makes it spring into existence the first time that it is called -- see
1088 Use of C<defined> on aggregates (hashes and arrays) is deprecated. It
1089 used to report whether memory for that aggregate has ever been
1090 allocated. This behavior may disappear in future versions of Perl.
1091 You should instead use a simple test for size:
1093 if (@an_array) { print "has array elements\n" }
1094 if (%a_hash) { print "has hash members\n" }
1096 When used on a hash element, it tells you whether the value is defined,
1097 not whether the key exists in the hash. Use L</exists> for the latter
1102 print if defined $switch{'D'};
1103 print "$val\n" while defined($val = pop(@ary));
1104 die "Can't readlink $sym: $!"
1105 unless defined($value = readlink $sym);
1106 sub foo { defined &$bar ? &$bar(@_) : die "No bar"; }
1107 $debugging = 0 unless defined $debugging;
1109 Note: Many folks tend to overuse C<defined>, and then are surprised to
1110 discover that the number C<0> and C<""> (the zero-length string) are, in fact,
1111 defined values. For example, if you say
1115 The pattern match succeeds, and C<$1> is defined, despite the fact that it
1116 matched "nothing". It didn't really fail to match anything. Rather, it
1117 matched something that happened to be zero characters long. This is all
1118 very above-board and honest. When a function returns an undefined value,
1119 it's an admission that it couldn't give you an honest answer. So you
1120 should use C<defined> only when you're questioning the integrity of what
1121 you're trying to do. At other times, a simple comparison to C<0> or C<""> is
1124 See also L</undef>, L</exists>, L</ref>.
1129 Given an expression that specifies a hash element, array element, hash slice,
1130 or array slice, deletes the specified element(s) from the hash or array.
1131 In the case of an array, if the array elements happen to be at the end,
1132 the size of the array will shrink to the highest element that tests
1133 true for exists() (or 0 if no such element exists).
1135 Returns a list with the same number of elements as the number of elements
1136 for which deletion was attempted. Each element of that list consists of
1137 either the value of the element deleted, or the undefined value. In scalar
1138 context, this means that you get the value of the last element deleted (or
1139 the undefined value if that element did not exist).
1141 %hash = (foo => 11, bar => 22, baz => 33);
1142 $scalar = delete $hash{foo}; # $scalar is 11
1143 $scalar = delete @hash{qw(foo bar)}; # $scalar is 22
1144 @array = delete @hash{qw(foo bar baz)}; # @array is (undef,undef,33)
1146 Deleting from C<%ENV> modifies the environment. Deleting from
1147 a hash tied to a DBM file deletes the entry from the DBM file. Deleting
1148 from a C<tie>d hash or array may not necessarily return anything.
1150 Deleting an array element effectively returns that position of the array
1151 to its initial, uninitialized state. Subsequently testing for the same
1152 element with exists() will return false. Also, deleting array elements
1153 in the middle of an array will not shift the index of the elements
1154 after them down. Use splice() for that. See L</exists>.
1156 The following (inefficiently) deletes all the values of %HASH and @ARRAY:
1158 foreach $key (keys %HASH) {
1162 foreach $index (0 .. $#ARRAY) {
1163 delete $ARRAY[$index];
1168 delete @HASH{keys %HASH};
1170 delete @ARRAY[0 .. $#ARRAY];
1172 But both of these are slower than just assigning the empty list
1173 or undefining %HASH or @ARRAY:
1175 %HASH = (); # completely empty %HASH
1176 undef %HASH; # forget %HASH ever existed
1178 @ARRAY = (); # completely empty @ARRAY
1179 undef @ARRAY; # forget @ARRAY ever existed
1181 Note that the EXPR can be arbitrarily complicated as long as the final
1182 operation is a hash element, array element, hash slice, or array slice
1185 delete $ref->[$x][$y]{$key};
1186 delete @{$ref->[$x][$y]}{$key1, $key2, @morekeys};
1188 delete $ref->[$x][$y][$index];
1189 delete @{$ref->[$x][$y]}[$index1, $index2, @moreindices];
1192 X<die> X<throw> X<exception> X<raise> X<$@> X<abort>
1194 Outside an C<eval>, prints the value of LIST to C<STDERR> and
1195 exits with the current value of C<$!> (errno). If C<$!> is C<0>,
1196 exits with the value of C<<< ($? >> 8) >>> (backtick `command`
1197 status). If C<<< ($? >> 8) >>> is C<0>, exits with C<255>. Inside
1198 an C<eval(),> the error message is stuffed into C<$@> and the
1199 C<eval> is terminated with the undefined value. This makes
1200 C<die> the way to raise an exception.
1202 Equivalent examples:
1204 die "Can't cd to spool: $!\n" unless chdir '/usr/spool/news';
1205 chdir '/usr/spool/news' or die "Can't cd to spool: $!\n"
1207 If the last element of LIST does not end in a newline, the current
1208 script line number and input line number (if any) are also printed,
1209 and a newline is supplied. Note that the "input line number" (also
1210 known as "chunk") is subject to whatever notion of "line" happens to
1211 be currently in effect, and is also available as the special variable
1212 C<$.>. See L<perlvar/"$/"> and L<perlvar/"$.">.
1214 Hint: sometimes appending C<", stopped"> to your message will cause it
1215 to make better sense when the string C<"at foo line 123"> is appended.
1216 Suppose you are running script "canasta".
1218 die "/etc/games is no good";
1219 die "/etc/games is no good, stopped";
1221 produce, respectively
1223 /etc/games is no good at canasta line 123.
1224 /etc/games is no good, stopped at canasta line 123.
1226 See also exit(), warn(), and the Carp module.
1228 If LIST is empty and C<$@> already contains a value (typically from a
1229 previous eval) that value is reused after appending C<"\t...propagated">.
1230 This is useful for propagating exceptions:
1233 die unless $@ =~ /Expected exception/;
1235 If LIST is empty and C<$@> contains an object reference that has a
1236 C<PROPAGATE> method, that method will be called with additional file
1237 and line number parameters. The return value replaces the value in
1238 C<$@>. i.e. as if C<< $@ = eval { $@->PROPAGATE(__FILE__, __LINE__) }; >>
1241 If C<$@> is empty then the string C<"Died"> is used.
1243 die() can also be called with a reference argument. If this happens to be
1244 trapped within an eval(), $@ contains the reference. This behavior permits
1245 a more elaborate exception handling implementation using objects that
1246 maintain arbitrary state about the nature of the exception. Such a scheme
1247 is sometimes preferable to matching particular string values of $@ using
1248 regular expressions. Here's an example:
1250 use Scalar::Util 'blessed';
1252 eval { ... ; die Some::Module::Exception->new( FOO => "bar" ) };
1254 if (blessed($@) && $@->isa("Some::Module::Exception")) {
1255 # handle Some::Module::Exception
1258 # handle all other possible exceptions
1262 Because perl will stringify uncaught exception messages before displaying
1263 them, you may want to overload stringification operations on such custom
1264 exception objects. See L<overload> for details about that.
1266 You can arrange for a callback to be run just before the C<die>
1267 does its deed, by setting the C<$SIG{__DIE__}> hook. The associated
1268 handler will be called with the error text and can change the error
1269 message, if it sees fit, by calling C<die> again. See
1270 L<perlvar/$SIG{expr}> for details on setting C<%SIG> entries, and
1271 L<"eval BLOCK"> for some examples. Although this feature was
1272 to be run only right before your program was to exit, this is not
1273 currently the case--the C<$SIG{__DIE__}> hook is currently called
1274 even inside eval()ed blocks/strings! If one wants the hook to do
1275 nothing in such situations, put
1279 as the first line of the handler (see L<perlvar/$^S>). Because
1280 this promotes strange action at a distance, this counterintuitive
1281 behavior may be fixed in a future release.
1286 Not really a function. Returns the value of the last command in the
1287 sequence of commands indicated by BLOCK. When modified by the C<while> or
1288 C<until> loop modifier, executes the BLOCK once before testing the loop
1289 condition. (On other statements the loop modifiers test the conditional
1292 C<do BLOCK> does I<not> count as a loop, so the loop control statements
1293 C<next>, C<last>, or C<redo> cannot be used to leave or restart the block.
1294 See L<perlsyn> for alternative strategies.
1296 =item do SUBROUTINE(LIST)
1299 This form of subroutine call is deprecated. See L<perlsub>.
1304 Uses the value of EXPR as a filename and executes the contents of the
1305 file as a Perl script.
1313 except that it's more efficient and concise, keeps track of the current
1314 filename for error messages, searches the @INC directories, and updates
1315 C<%INC> if the file is found. See L<perlvar/Predefined Names> for these
1316 variables. It also differs in that code evaluated with C<do FILENAME>
1317 cannot see lexicals in the enclosing scope; C<eval STRING> does. It's the
1318 same, however, in that it does reparse the file every time you call it,
1319 so you probably don't want to do this inside a loop.
1321 If C<do> cannot read the file, it returns undef and sets C<$!> to the
1322 error. If C<do> can read the file but cannot compile it, it
1323 returns undef and sets an error message in C<$@>. If the file is
1324 successfully compiled, C<do> returns the value of the last expression
1327 Note that inclusion of library modules is better done with the
1328 C<use> and C<require> operators, which also do automatic error checking
1329 and raise an exception if there's a problem.
1331 You might like to use C<do> to read in a program configuration
1332 file. Manual error checking can be done this way:
1334 # read in config files: system first, then user
1335 for $file ("/share/prog/defaults.rc",
1336 "$ENV{HOME}/.someprogrc")
1338 unless ($return = do $file) {
1339 warn "couldn't parse $file: $@" if $@;
1340 warn "couldn't do $file: $!" unless defined $return;
1341 warn "couldn't run $file" unless $return;
1346 X<dump> X<core> X<undump>
1350 This function causes an immediate core dump. See also the B<-u>
1351 command-line switch in L<perlrun>, which does the same thing.
1352 Primarily this is so that you can use the B<undump> program (not
1353 supplied) to turn your core dump into an executable binary after
1354 having initialized all your variables at the beginning of the
1355 program. When the new binary is executed it will begin by executing
1356 a C<goto LABEL> (with all the restrictions that C<goto> suffers).
1357 Think of it as a goto with an intervening core dump and reincarnation.
1358 If C<LABEL> is omitted, restarts the program from the top.
1360 B<WARNING>: Any files opened at the time of the dump will I<not>
1361 be open any more when the program is reincarnated, with possible
1362 resulting confusion on the part of Perl.
1364 This function is now largely obsolete, partly because it's very
1365 hard to convert a core file into an executable, and because the
1366 real compiler backends for generating portable bytecode and compilable
1367 C code have superseded it. That's why you should now invoke it as
1368 C<CORE::dump()>, if you don't want to be warned against a possible
1371 If you're looking to use L<dump> to speed up your program, consider
1372 generating bytecode or native C code as described in L<perlcc>. If
1373 you're just trying to accelerate a CGI script, consider using the
1374 C<mod_perl> extension to B<Apache>, or the CPAN module, CGI::Fast.
1375 You might also consider autoloading or selfloading, which at least
1376 make your program I<appear> to run faster.
1379 X<each> X<hash, iterator>
1381 When called in list context, returns a 2-element list consisting of the
1382 key and value for the next element of a hash, so that you can iterate over
1383 it. When called in scalar context, returns only the key for the next
1384 element in the hash.
1386 Entries are returned in an apparently random order. The actual random
1387 order is subject to change in future versions of perl, but it is
1388 guaranteed to be in the same order as either the C<keys> or C<values>
1389 function would produce on the same (unmodified) hash. Since Perl
1390 5.8.1 the ordering is different even between different runs of Perl
1391 for security reasons (see L<perlsec/"Algorithmic Complexity Attacks">).
1393 When the hash is entirely read, a null array is returned in list context
1394 (which when assigned produces a false (C<0>) value), and C<undef> in
1395 scalar context. The next call to C<each> after that will start iterating
1396 again. There is a single iterator for each hash, shared by all C<each>,
1397 C<keys>, and C<values> function calls in the program; it can be reset by
1398 reading all the elements from the hash, or by evaluating C<keys HASH> or
1399 C<values HASH>. If you add or delete elements of a hash while you're
1400 iterating over it, you may get entries skipped or duplicated, so
1401 don't. Exception: It is always safe to delete the item most recently
1402 returned by C<each()>, which means that the following code will work:
1404 while (($key, $value) = each %hash) {
1406 delete $hash{$key}; # This is safe
1409 The following prints out your environment like the printenv(1) program,
1410 only in a different order:
1412 while (($key,$value) = each %ENV) {
1413 print "$key=$value\n";
1416 See also C<keys>, C<values> and C<sort>.
1418 =item eof FILEHANDLE
1427 Returns 1 if the next read on FILEHANDLE will return end of file, or if
1428 FILEHANDLE is not open. FILEHANDLE may be an expression whose value
1429 gives the real filehandle. (Note that this function actually
1430 reads a character and then C<ungetc>s it, so isn't very useful in an
1431 interactive context.) Do not read from a terminal file (or call
1432 C<eof(FILEHANDLE)> on it) after end-of-file is reached. File types such
1433 as terminals may lose the end-of-file condition if you do.
1435 An C<eof> without an argument uses the last file read. Using C<eof()>
1436 with empty parentheses is very different. It refers to the pseudo file
1437 formed from the files listed on the command line and accessed via the
1438 C<< <> >> operator. Since C<< <> >> isn't explicitly opened,
1439 as a normal filehandle is, an C<eof()> before C<< <> >> has been
1440 used will cause C<@ARGV> to be examined to determine if input is
1441 available. Similarly, an C<eof()> after C<< <> >> has returned
1442 end-of-file will assume you are processing another C<@ARGV> list,
1443 and if you haven't set C<@ARGV>, will read input from C<STDIN>;
1444 see L<perlop/"I/O Operators">.
1446 In a C<< while (<>) >> loop, C<eof> or C<eof(ARGV)> can be used to
1447 detect the end of each file, C<eof()> will only detect the end of the
1448 last file. Examples:
1450 # reset line numbering on each input file
1452 next if /^\s*#/; # skip comments
1455 close ARGV if eof; # Not eof()!
1458 # insert dashes just before last line of last file
1460 if (eof()) { # check for end of last file
1461 print "--------------\n";
1464 last if eof(); # needed if we're reading from a terminal
1467 Practical hint: you almost never need to use C<eof> in Perl, because the
1468 input operators typically return C<undef> when they run out of data, or if
1472 X<eval> X<try> X<catch> X<evaluate> X<parse> X<execute>
1478 In the first form, the return value of EXPR is parsed and executed as if it
1479 were a little Perl program. The value of the expression (which is itself
1480 determined within scalar context) is first parsed, and if there weren't any
1481 errors, executed in the lexical context of the current Perl program, so
1482 that any variable settings or subroutine and format definitions remain
1483 afterwards. Note that the value is parsed every time the C<eval> executes.
1484 If EXPR is omitted, evaluates C<$_>. This form is typically used to
1485 delay parsing and subsequent execution of the text of EXPR until run time.
1487 In the second form, the code within the BLOCK is parsed only once--at the
1488 same time the code surrounding the C<eval> itself was parsed--and executed
1489 within the context of the current Perl program. This form is typically
1490 used to trap exceptions more efficiently than the first (see below), while
1491 also providing the benefit of checking the code within BLOCK at compile
1494 The final semicolon, if any, may be omitted from the value of EXPR or within
1497 In both forms, the value returned is the value of the last expression
1498 evaluated inside the mini-program; a return statement may be also used, just
1499 as with subroutines. The expression providing the return value is evaluated
1500 in void, scalar, or list context, depending on the context of the C<eval>
1501 itself. See L</wantarray> for more on how the evaluation context can be
1504 If there is a syntax error or runtime error, or a C<die> statement is
1505 executed, an undefined value is returned by C<eval>, and C<$@> is set to the
1506 error message. If there was no error, C<$@> is guaranteed to be a null
1507 string. Beware that using C<eval> neither silences perl from printing
1508 warnings to STDERR, nor does it stuff the text of warning messages into C<$@>.
1509 To do either of those, you have to use the C<$SIG{__WARN__}> facility, or
1510 turn off warnings inside the BLOCK or EXPR using S<C<no warnings 'all'>>.
1511 See L</warn>, L<perlvar>, L<warnings> and L<perllexwarn>.
1513 Note that, because C<eval> traps otherwise-fatal errors, it is useful for
1514 determining whether a particular feature (such as C<socket> or C<symlink>)
1515 is implemented. It is also Perl's exception trapping mechanism, where
1516 the die operator is used to raise exceptions.
1518 If the code to be executed doesn't vary, you may use the eval-BLOCK
1519 form to trap run-time errors without incurring the penalty of
1520 recompiling each time. The error, if any, is still returned in C<$@>.
1523 # make divide-by-zero nonfatal
1524 eval { $answer = $a / $b; }; warn $@ if $@;
1526 # same thing, but less efficient
1527 eval '$answer = $a / $b'; warn $@ if $@;
1529 # a compile-time error
1530 eval { $answer = }; # WRONG
1533 eval '$answer ='; # sets $@
1535 Using the C<eval{}> form as an exception trap in libraries does have some
1536 issues. Due to the current arguably broken state of C<__DIE__> hooks, you
1537 may wish not to trigger any C<__DIE__> hooks that user code may have installed.
1538 You can use the C<local $SIG{__DIE__}> construct for this purpose,
1539 as shown in this example:
1541 # a very private exception trap for divide-by-zero
1542 eval { local $SIG{'__DIE__'}; $answer = $a / $b; };
1545 This is especially significant, given that C<__DIE__> hooks can call
1546 C<die> again, which has the effect of changing their error messages:
1548 # __DIE__ hooks may modify error messages
1550 local $SIG{'__DIE__'} =
1551 sub { (my $x = $_[0]) =~ s/foo/bar/g; die $x };
1552 eval { die "foo lives here" };
1553 print $@ if $@; # prints "bar lives here"
1556 Because this promotes action at a distance, this counterintuitive behavior
1557 may be fixed in a future release.
1559 With an C<eval>, you should be especially careful to remember what's
1560 being looked at when:
1566 eval { $x }; # CASE 4
1568 eval "\$$x++"; # CASE 5
1571 Cases 1 and 2 above behave identically: they run the code contained in
1572 the variable $x. (Although case 2 has misleading double quotes making
1573 the reader wonder what else might be happening (nothing is).) Cases 3
1574 and 4 likewise behave in the same way: they run the code C<'$x'>, which
1575 does nothing but return the value of $x. (Case 4 is preferred for
1576 purely visual reasons, but it also has the advantage of compiling at
1577 compile-time instead of at run-time.) Case 5 is a place where
1578 normally you I<would> like to use double quotes, except that in this
1579 particular situation, you can just use symbolic references instead, as
1582 C<eval BLOCK> does I<not> count as a loop, so the loop control statements
1583 C<next>, C<last>, or C<redo> cannot be used to leave or restart the block.
1585 Note that as a very special case, an C<eval ''> executed within the C<DB>
1586 package doesn't see the usual surrounding lexical scope, but rather the
1587 scope of the first non-DB piece of code that called it. You don't normally
1588 need to worry about this unless you are writing a Perl debugger.
1593 =item exec PROGRAM LIST
1595 The C<exec> function executes a system command I<and never returns>--
1596 use C<system> instead of C<exec> if you want it to return. It fails and
1597 returns false only if the command does not exist I<and> it is executed
1598 directly instead of via your system's command shell (see below).
1600 Since it's a common mistake to use C<exec> instead of C<system>, Perl
1601 warns you if there is a following statement which isn't C<die>, C<warn>,
1602 or C<exit> (if C<-w> is set - but you always do that). If you
1603 I<really> want to follow an C<exec> with some other statement, you
1604 can use one of these styles to avoid the warning:
1606 exec ('foo') or print STDERR "couldn't exec foo: $!";
1607 { exec ('foo') }; print STDERR "couldn't exec foo: $!";
1609 If there is more than one argument in LIST, or if LIST is an array
1610 with more than one value, calls execvp(3) with the arguments in LIST.
1611 If there is only one scalar argument or an array with one element in it,
1612 the argument is checked for shell metacharacters, and if there are any,
1613 the entire argument is passed to the system's command shell for parsing
1614 (this is C</bin/sh -c> on Unix platforms, but varies on other platforms).
1615 If there are no shell metacharacters in the argument, it is split into
1616 words and passed directly to C<execvp>, which is more efficient.
1619 exec '/bin/echo', 'Your arguments are: ', @ARGV;
1620 exec "sort $outfile | uniq";
1622 If you don't really want to execute the first argument, but want to lie
1623 to the program you are executing about its own name, you can specify
1624 the program you actually want to run as an "indirect object" (without a
1625 comma) in front of the LIST. (This always forces interpretation of the
1626 LIST as a multivalued list, even if there is only a single scalar in
1629 $shell = '/bin/csh';
1630 exec $shell '-sh'; # pretend it's a login shell
1634 exec {'/bin/csh'} '-sh'; # pretend it's a login shell
1636 When the arguments get executed via the system shell, results will
1637 be subject to its quirks and capabilities. See L<perlop/"`STRING`">
1640 Using an indirect object with C<exec> or C<system> is also more
1641 secure. This usage (which also works fine with system()) forces
1642 interpretation of the arguments as a multivalued list, even if the
1643 list had just one argument. That way you're safe from the shell
1644 expanding wildcards or splitting up words with whitespace in them.
1646 @args = ( "echo surprise" );
1648 exec @args; # subject to shell escapes
1650 exec { $args[0] } @args; # safe even with one-arg list
1652 The first version, the one without the indirect object, ran the I<echo>
1653 program, passing it C<"surprise"> an argument. The second version
1654 didn't--it tried to run a program literally called I<"echo surprise">,
1655 didn't find it, and set C<$?> to a non-zero value indicating failure.
1657 Beginning with v5.6.0, Perl will attempt to flush all files opened for
1658 output before the exec, but this may not be supported on some platforms
1659 (see L<perlport>). To be safe, you may need to set C<$|> ($AUTOFLUSH
1660 in English) or call the C<autoflush()> method of C<IO::Handle> on any
1661 open handles in order to avoid lost output.
1663 Note that C<exec> will not call your C<END> blocks, nor will it call
1664 any C<DESTROY> methods in your objects.
1667 X<exists> X<autovivification>
1669 Given an expression that specifies a hash element or array element,
1670 returns true if the specified element in the hash or array has ever
1671 been initialized, even if the corresponding value is undefined. The
1672 element is not autovivified if it doesn't exist.
1674 print "Exists\n" if exists $hash{$key};
1675 print "Defined\n" if defined $hash{$key};
1676 print "True\n" if $hash{$key};
1678 print "Exists\n" if exists $array[$index];
1679 print "Defined\n" if defined $array[$index];
1680 print "True\n" if $array[$index];
1682 A hash or array element can be true only if it's defined, and defined if
1683 it exists, but the reverse doesn't necessarily hold true.
1685 Given an expression that specifies the name of a subroutine,
1686 returns true if the specified subroutine has ever been declared, even
1687 if it is undefined. Mentioning a subroutine name for exists or defined
1688 does not count as declaring it. Note that a subroutine which does not
1689 exist may still be callable: its package may have an C<AUTOLOAD>
1690 method that makes it spring into existence the first time that it is
1691 called -- see L<perlsub>.
1693 print "Exists\n" if exists &subroutine;
1694 print "Defined\n" if defined &subroutine;
1696 Note that the EXPR can be arbitrarily complicated as long as the final
1697 operation is a hash or array key lookup or subroutine name:
1699 if (exists $ref->{A}->{B}->{$key}) { }
1700 if (exists $hash{A}{B}{$key}) { }
1702 if (exists $ref->{A}->{B}->[$ix]) { }
1703 if (exists $hash{A}{B}[$ix]) { }
1705 if (exists &{$ref->{A}{B}{$key}}) { }
1707 Although the deepest nested array or hash will not spring into existence
1708 just because its existence was tested, any intervening ones will.
1709 Thus C<< $ref->{"A"} >> and C<< $ref->{"A"}->{"B"} >> will spring
1710 into existence due to the existence test for the $key element above.
1711 This happens anywhere the arrow operator is used, including even:
1714 if (exists $ref->{"Some key"}) { }
1715 print $ref; # prints HASH(0x80d3d5c)
1717 This surprising autovivification in what does not at first--or even
1718 second--glance appear to be an lvalue context may be fixed in a future
1721 Use of a subroutine call, rather than a subroutine name, as an argument
1722 to exists() is an error.
1725 exists &sub(); # Error
1728 X<exit> X<terminate> X<abort>
1732 Evaluates EXPR and exits immediately with that value. Example:
1735 exit 0 if $ans =~ /^[Xx]/;
1737 See also C<die>. If EXPR is omitted, exits with C<0> status. The only
1738 universally recognized values for EXPR are C<0> for success and C<1>
1739 for error; other values are subject to interpretation depending on the
1740 environment in which the Perl program is running. For example, exiting
1741 69 (EX_UNAVAILABLE) from a I<sendmail> incoming-mail filter will cause
1742 the mailer to return the item undelivered, but that's not true everywhere.
1744 Don't use C<exit> to abort a subroutine if there's any chance that
1745 someone might want to trap whatever error happened. Use C<die> instead,
1746 which can be trapped by an C<eval>.
1748 The exit() function does not always exit immediately. It calls any
1749 defined C<END> routines first, but these C<END> routines may not
1750 themselves abort the exit. Likewise any object destructors that need to
1751 be called are called before the real exit. If this is a problem, you
1752 can call C<POSIX:_exit($status)> to avoid END and destructor processing.
1753 See L<perlmod> for details.
1756 X<exp> X<exponential> X<antilog> X<antilogarithm> X<e>
1760 Returns I<e> (the natural logarithm base) to the power of EXPR.
1761 If EXPR is omitted, gives C<exp($_)>.
1763 =item fcntl FILEHANDLE,FUNCTION,SCALAR
1766 Implements the fcntl(2) function. You'll probably have to say
1770 first to get the correct constant definitions. Argument processing and
1771 value return works just like C<ioctl> below.
1775 fcntl($filehandle, F_GETFL, $packed_return_buffer)
1776 or die "can't fcntl F_GETFL: $!";
1778 You don't have to check for C<defined> on the return from C<fcntl>.
1779 Like C<ioctl>, it maps a C<0> return from the system call into
1780 C<"0 but true"> in Perl. This string is true in boolean context and C<0>
1781 in numeric context. It is also exempt from the normal B<-w> warnings
1782 on improper numeric conversions.
1784 Note that C<fcntl> will produce a fatal error if used on a machine that
1785 doesn't implement fcntl(2). See the Fcntl module or your fcntl(2)
1786 manpage to learn what functions are available on your system.
1788 Here's an example of setting a filehandle named C<REMOTE> to be
1789 non-blocking at the system level. You'll have to negotiate C<$|>
1790 on your own, though.
1792 use Fcntl qw(F_GETFL F_SETFL O_NONBLOCK);
1794 $flags = fcntl(REMOTE, F_GETFL, 0)
1795 or die "Can't get flags for the socket: $!\n";
1797 $flags = fcntl(REMOTE, F_SETFL, $flags | O_NONBLOCK)
1798 or die "Can't set flags for the socket: $!\n";
1800 =item fileno FILEHANDLE
1803 Returns the file descriptor for a filehandle, or undefined if the
1804 filehandle is not open. This is mainly useful for constructing
1805 bitmaps for C<select> and low-level POSIX tty-handling operations.
1806 If FILEHANDLE is an expression, the value is taken as an indirect
1807 filehandle, generally its name.
1809 You can use this to find out whether two handles refer to the
1810 same underlying descriptor:
1812 if (fileno(THIS) == fileno(THAT)) {
1813 print "THIS and THAT are dups\n";
1816 (Filehandles connected to memory objects via new features of C<open> may
1817 return undefined even though they are open.)
1820 =item flock FILEHANDLE,OPERATION
1821 X<flock> X<lock> X<locking>
1823 Calls flock(2), or an emulation of it, on FILEHANDLE. Returns true
1824 for success, false on failure. Produces a fatal error if used on a
1825 machine that doesn't implement flock(2), fcntl(2) locking, or lockf(3).
1826 C<flock> is Perl's portable file locking interface, although it locks
1827 only entire files, not records.
1829 Two potentially non-obvious but traditional C<flock> semantics are
1830 that it waits indefinitely until the lock is granted, and that its locks
1831 B<merely advisory>. Such discretionary locks are more flexible, but offer
1832 fewer guarantees. This means that programs that do not also use C<flock>
1833 may modify files locked with C<flock>. See L<perlport>,
1834 your port's specific documentation, or your system-specific local manpages
1835 for details. It's best to assume traditional behavior if you're writing
1836 portable programs. (But if you're not, you should as always feel perfectly
1837 free to write for your own system's idiosyncrasies (sometimes called
1838 "features"). Slavish adherence to portability concerns shouldn't get
1839 in the way of your getting your job done.)
1841 OPERATION is one of LOCK_SH, LOCK_EX, or LOCK_UN, possibly combined with
1842 LOCK_NB. These constants are traditionally valued 1, 2, 8 and 4, but
1843 you can use the symbolic names if you import them from the Fcntl module,
1844 either individually, or as a group using the ':flock' tag. LOCK_SH
1845 requests a shared lock, LOCK_EX requests an exclusive lock, and LOCK_UN
1846 releases a previously requested lock. If LOCK_NB is bitwise-or'ed with
1847 LOCK_SH or LOCK_EX then C<flock> will return immediately rather than blocking
1848 waiting for the lock (check the return status to see if you got it).
1850 To avoid the possibility of miscoordination, Perl now flushes FILEHANDLE
1851 before locking or unlocking it.
1853 Note that the emulation built with lockf(3) doesn't provide shared
1854 locks, and it requires that FILEHANDLE be open with write intent. These
1855 are the semantics that lockf(3) implements. Most if not all systems
1856 implement lockf(3) in terms of fcntl(2) locking, though, so the
1857 differing semantics shouldn't bite too many people.
1859 Note that the fcntl(2) emulation of flock(3) requires that FILEHANDLE
1860 be open with read intent to use LOCK_SH and requires that it be open
1861 with write intent to use LOCK_EX.
1863 Note also that some versions of C<flock> cannot lock things over the
1864 network; you would need to use the more system-specific C<fcntl> for
1865 that. If you like you can force Perl to ignore your system's flock(2)
1866 function, and so provide its own fcntl(2)-based emulation, by passing
1867 the switch C<-Ud_flock> to the F<Configure> program when you configure
1870 Here's a mailbox appender for BSD systems.
1872 use Fcntl ':flock'; # import LOCK_* constants
1875 flock(MBOX,LOCK_EX);
1876 # and, in case someone appended
1877 # while we were waiting...
1882 flock(MBOX,LOCK_UN);
1885 open(MBOX, ">>/usr/spool/mail/$ENV{'USER'}")
1886 or die "Can't open mailbox: $!";
1889 print MBOX $msg,"\n\n";
1892 On systems that support a real flock(), locks are inherited across fork()
1893 calls, whereas those that must resort to the more capricious fcntl()
1894 function lose the locks, making it harder to write servers.
1896 See also L<DB_File> for other flock() examples.
1899 X<fork> X<child> X<parent>
1901 Does a fork(2) system call to create a new process running the
1902 same program at the same point. It returns the child pid to the
1903 parent process, C<0> to the child process, or C<undef> if the fork is
1904 unsuccessful. File descriptors (and sometimes locks on those descriptors)
1905 are shared, while everything else is copied. On most systems supporting
1906 fork(), great care has gone into making it extremely efficient (for
1907 example, using copy-on-write technology on data pages), making it the
1908 dominant paradigm for multitasking over the last few decades.
1910 Beginning with v5.6.0, Perl will attempt to flush all files opened for
1911 output before forking the child process, but this may not be supported
1912 on some platforms (see L<perlport>). To be safe, you may need to set
1913 C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method of
1914 C<IO::Handle> on any open handles in order to avoid duplicate output.
1916 If you C<fork> without ever waiting on your children, you will
1917 accumulate zombies. On some systems, you can avoid this by setting
1918 C<$SIG{CHLD}> to C<"IGNORE">. See also L<perlipc> for more examples of
1919 forking and reaping moribund children.
1921 Note that if your forked child inherits system file descriptors like
1922 STDIN and STDOUT that are actually connected by a pipe or socket, even
1923 if you exit, then the remote server (such as, say, a CGI script or a
1924 backgrounded job launched from a remote shell) won't think you're done.
1925 You should reopen those to F</dev/null> if it's any issue.
1930 Declare a picture format for use by the C<write> function. For
1934 Test: @<<<<<<<< @||||| @>>>>>
1935 $str, $%, '$' . int($num)
1939 $num = $cost/$quantity;
1943 See L<perlform> for many details and examples.
1945 =item formline PICTURE,LIST
1948 This is an internal function used by C<format>s, though you may call it,
1949 too. It formats (see L<perlform>) a list of values according to the
1950 contents of PICTURE, placing the output into the format output
1951 accumulator, C<$^A> (or C<$ACCUMULATOR> in English).
1952 Eventually, when a C<write> is done, the contents of
1953 C<$^A> are written to some filehandle. You could also read C<$^A>
1954 and then set C<$^A> back to C<"">. Note that a format typically
1955 does one C<formline> per line of form, but the C<formline> function itself
1956 doesn't care how many newlines are embedded in the PICTURE. This means
1957 that the C<~> and C<~~> tokens will treat the entire PICTURE as a single line.
1958 You may therefore need to use multiple formlines to implement a single
1959 record format, just like the format compiler.
1961 Be careful if you put double quotes around the picture, because an C<@>
1962 character may be taken to mean the beginning of an array name.
1963 C<formline> always returns true. See L<perlform> for other examples.
1965 =item getc FILEHANDLE
1970 Returns the next character from the input file attached to FILEHANDLE,
1971 or the undefined value at end of file, or if there was an error (in
1972 the latter case C<$!> is set). If FILEHANDLE is omitted, reads from
1973 STDIN. This is not particularly efficient. However, it cannot be
1974 used by itself to fetch single characters without waiting for the user
1975 to hit enter. For that, try something more like:
1978 system "stty cbreak </dev/tty >/dev/tty 2>&1";
1981 system "stty", '-icanon', 'eol', "\001";
1987 system "stty -cbreak </dev/tty >/dev/tty 2>&1";
1990 system "stty", 'icanon', 'eol', '^@'; # ASCII null
1994 Determination of whether $BSD_STYLE should be set
1995 is left as an exercise to the reader.
1997 The C<POSIX::getattr> function can do this more portably on
1998 systems purporting POSIX compliance. See also the C<Term::ReadKey>
1999 module from your nearest CPAN site; details on CPAN can be found on
2003 X<getlogin> X<login>
2005 This implements the C library function of the same name, which on most
2006 systems returns the current login from F</etc/utmp>, if any. If null,
2009 $login = getlogin || getpwuid($<) || "Kilroy";
2011 Do not consider C<getlogin> for authentication: it is not as
2012 secure as C<getpwuid>.
2014 =item getpeername SOCKET
2015 X<getpeername> X<peer>
2017 Returns the packed sockaddr address of other end of the SOCKET connection.
2020 $hersockaddr = getpeername(SOCK);
2021 ($port, $iaddr) = sockaddr_in($hersockaddr);
2022 $herhostname = gethostbyaddr($iaddr, AF_INET);
2023 $herstraddr = inet_ntoa($iaddr);
2028 Returns the current process group for the specified PID. Use
2029 a PID of C<0> to get the current process group for the
2030 current process. Will raise an exception if used on a machine that
2031 doesn't implement getpgrp(2). If PID is omitted, returns process
2032 group of current process. Note that the POSIX version of C<getpgrp>
2033 does not accept a PID argument, so only C<PID==0> is truly portable.
2036 X<getppid> X<parent> X<pid>
2038 Returns the process id of the parent process.
2040 Note for Linux users: on Linux, the C functions C<getpid()> and
2041 C<getppid()> return different values from different threads. In order to
2042 be portable, this behavior is not reflected by the perl-level function
2043 C<getppid()>, that returns a consistent value across threads. If you want
2044 to call the underlying C<getppid()>, you may use the CPAN module
2047 =item getpriority WHICH,WHO
2048 X<getpriority> X<priority> X<nice>
2050 Returns the current priority for a process, a process group, or a user.
2051 (See L<getpriority(2)>.) Will raise a fatal exception if used on a
2052 machine that doesn't implement getpriority(2).
2055 X<getpwnam> X<getgrnam> X<gethostbyname> X<getnetbyname> X<getprotobyname>
2056 X<getpwuid> X<getgrgid> X<getservbyname> X<gethostbyaddr> X<getnetbyaddr>
2057 X<getprotobynumber> X<getservbyport> X<getpwent> X<getgrent> X<gethostent>
2058 X<getnetent> X<getprotoent> X<getservent> X<setpwent> X<setgrent> X<sethostent>
2059 X<setnetent> X<setprotoent> X<setservent> X<endpwent> X<endgrent> X<endhostent>
2060 X<endnetent> X<endprotoent> X<endservent>
2064 =item gethostbyname NAME
2066 =item getnetbyname NAME
2068 =item getprotobyname NAME
2074 =item getservbyname NAME,PROTO
2076 =item gethostbyaddr ADDR,ADDRTYPE
2078 =item getnetbyaddr ADDR,ADDRTYPE
2080 =item getprotobynumber NUMBER
2082 =item getservbyport PORT,PROTO
2100 =item sethostent STAYOPEN
2102 =item setnetent STAYOPEN
2104 =item setprotoent STAYOPEN
2106 =item setservent STAYOPEN
2120 These routines perform the same functions as their counterparts in the
2121 system library. In list context, the return values from the
2122 various get routines are as follows:
2124 ($name,$passwd,$uid,$gid,
2125 $quota,$comment,$gcos,$dir,$shell,$expire) = getpw*
2126 ($name,$passwd,$gid,$members) = getgr*
2127 ($name,$aliases,$addrtype,$length,@addrs) = gethost*
2128 ($name,$aliases,$addrtype,$net) = getnet*
2129 ($name,$aliases,$proto) = getproto*
2130 ($name,$aliases,$port,$proto) = getserv*
2132 (If the entry doesn't exist you get a null list.)
2134 The exact meaning of the $gcos field varies but it usually contains
2135 the real name of the user (as opposed to the login name) and other
2136 information pertaining to the user. Beware, however, that in many
2137 system users are able to change this information and therefore it
2138 cannot be trusted and therefore the $gcos is tainted (see
2139 L<perlsec>). The $passwd and $shell, user's encrypted password and
2140 login shell, are also tainted, because of the same reason.
2142 In scalar context, you get the name, unless the function was a
2143 lookup by name, in which case you get the other thing, whatever it is.
2144 (If the entry doesn't exist you get the undefined value.) For example:
2146 $uid = getpwnam($name);
2147 $name = getpwuid($num);
2149 $gid = getgrnam($name);
2150 $name = getgrgid($num);
2154 In I<getpw*()> the fields $quota, $comment, and $expire are special
2155 cases in the sense that in many systems they are unsupported. If the
2156 $quota is unsupported, it is an empty scalar. If it is supported, it
2157 usually encodes the disk quota. If the $comment field is unsupported,
2158 it is an empty scalar. If it is supported it usually encodes some
2159 administrative comment about the user. In some systems the $quota
2160 field may be $change or $age, fields that have to do with password
2161 aging. In some systems the $comment field may be $class. The $expire
2162 field, if present, encodes the expiration period of the account or the
2163 password. For the availability and the exact meaning of these fields
2164 in your system, please consult your getpwnam(3) documentation and your
2165 F<pwd.h> file. You can also find out from within Perl what your
2166 $quota and $comment fields mean and whether you have the $expire field
2167 by using the C<Config> module and the values C<d_pwquota>, C<d_pwage>,
2168 C<d_pwchange>, C<d_pwcomment>, and C<d_pwexpire>. Shadow password
2169 files are only supported if your vendor has implemented them in the
2170 intuitive fashion that calling the regular C library routines gets the
2171 shadow versions if you're running under privilege or if there exists
2172 the shadow(3) functions as found in System V (this includes Solaris
2173 and Linux.) Those systems that implement a proprietary shadow password
2174 facility are unlikely to be supported.
2176 The $members value returned by I<getgr*()> is a space separated list of
2177 the login names of the members of the group.
2179 For the I<gethost*()> functions, if the C<h_errno> variable is supported in
2180 C, it will be returned to you via C<$?> if the function call fails. The
2181 C<@addrs> value returned by a successful call is a list of the raw
2182 addresses returned by the corresponding system library call. In the
2183 Internet domain, each address is four bytes long and you can unpack it
2184 by saying something like:
2186 ($a,$b,$c,$d) = unpack('W4',$addr[0]);
2188 The Socket library makes this slightly easier:
2191 $iaddr = inet_aton("127.1"); # or whatever address
2192 $name = gethostbyaddr($iaddr, AF_INET);
2194 # or going the other way
2195 $straddr = inet_ntoa($iaddr);
2197 If you get tired of remembering which element of the return list
2198 contains which return value, by-name interfaces are provided
2199 in standard modules: C<File::stat>, C<Net::hostent>, C<Net::netent>,
2200 C<Net::protoent>, C<Net::servent>, C<Time::gmtime>, C<Time::localtime>,
2201 and C<User::grent>. These override the normal built-ins, supplying
2202 versions that return objects with the appropriate names
2203 for each field. For example:
2207 $is_his = (stat($filename)->uid == pwent($whoever)->uid);
2209 Even though it looks like they're the same method calls (uid),
2210 they aren't, because a C<File::stat> object is different from
2211 a C<User::pwent> object.
2213 =item getsockname SOCKET
2216 Returns the packed sockaddr address of this end of the SOCKET connection,
2217 in case you don't know the address because you have several different
2218 IPs that the connection might have come in on.
2221 $mysockaddr = getsockname(SOCK);
2222 ($port, $myaddr) = sockaddr_in($mysockaddr);
2223 printf "Connect to %s [%s]\n",
2224 scalar gethostbyaddr($myaddr, AF_INET),
2227 =item getsockopt SOCKET,LEVEL,OPTNAME
2230 Queries the option named OPTNAME associated with SOCKET at a given LEVEL.
2231 Options may exist at multiple protocol levels depending on the socket
2232 type, but at least the uppermost socket level SOL_SOCKET (defined in the
2233 C<Socket> module) will exist. To query options at another level the
2234 protocol number of the appropriate protocol controlling the option
2235 should be supplied. For example, to indicate that an option is to be
2236 interpreted by the TCP protocol, LEVEL should be set to the protocol
2237 number of TCP, which you can get using getprotobyname.
2239 The call returns a packed string representing the requested socket option,
2240 or C<undef> if there is an error (the error reason will be in $!). What
2241 exactly is in the packed string depends in the LEVEL and OPTNAME, consult
2242 your system documentation for details. A very common case however is that
2243 the option is an integer, in which case the result will be a packed
2244 integer which you can decode using unpack with the C<i> (or C<I>) format.
2246 An example testing if Nagle's algorithm is turned on on a socket:
2248 use Socket qw(:all);
2250 defined(my $tcp = getprotobyname("tcp"))
2251 or die "Could not determine the protocol number for tcp";
2252 # my $tcp = IPPROTO_TCP; # Alternative
2253 my $packed = getsockopt($socket, $tcp, TCP_NODELAY)
2254 or die "Could not query TCP_NODELAY socket option: $!";
2255 my $nodelay = unpack("I", $packed);
2256 print "Nagle's algorithm is turned ", $nodelay ? "off\n" : "on\n";
2260 X<glob> X<wildcard> X<filename, expansion> X<expand>
2264 In list context, returns a (possibly empty) list of filename expansions on
2265 the value of EXPR such as the standard Unix shell F</bin/csh> would do. In
2266 scalar context, glob iterates through such filename expansions, returning
2267 undef when the list is exhausted. This is the internal function
2268 implementing the C<< <*.c> >> operator, but you can use it directly. If
2269 EXPR is omitted, C<$_> is used. The C<< <*.c> >> operator is discussed in
2270 more detail in L<perlop/"I/O Operators">.
2272 Beginning with v5.6.0, this operator is implemented using the standard
2273 C<File::Glob> extension. See L<File::Glob> for details.
2276 X<gmtime> X<UTC> X<Greenwich>
2280 Converts a time as returned by the time function to an 9-element list
2281 with the time localized for the standard Greenwich time zone.
2282 Typically used as follows:
2285 ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) =
2288 All list elements are numeric, and come straight out of the C `struct
2289 tm'. $sec, $min, and $hour are the seconds, minutes, and hours of the
2290 specified time. $mday is the day of the month, and $mon is the month
2291 itself, in the range C<0..11> with 0 indicating January and 11
2292 indicating December. $year is the number of years since 1900. That
2293 is, $year is C<123> in year 2023. $wday is the day of the week, with
2294 0 indicating Sunday and 3 indicating Wednesday. $yday is the day of
2295 the year, in the range C<0..364> (or C<0..365> in leap years). $isdst
2298 Note that the $year element is I<not> simply the last two digits of
2299 the year. If you assume it is then you create non-Y2K-compliant
2300 programs--and you wouldn't want to do that, would you?
2302 The proper way to get a complete 4-digit year is simply:
2306 And to get the last two digits of the year (e.g., '01' in 2001) do:
2308 $year = sprintf("%02d", $year % 100);
2310 If EXPR is omitted, C<gmtime()> uses the current time (C<gmtime(time)>).
2312 In scalar context, C<gmtime()> returns the ctime(3) value:
2314 $now_string = gmtime; # e.g., "Thu Oct 13 04:54:34 1994"
2316 If you need local time instead of GMT use the L</localtime> builtin.
2317 See also the C<timegm> function provided by the C<Time::Local> module,
2318 and the strftime(3) and mktime(3) functions available via the L<POSIX> module.
2320 This scalar value is B<not> locale dependent (see L<perllocale>), but is
2321 instead a Perl builtin. To get somewhat similar but locale dependent date
2322 strings, see the example in L</localtime>.
2324 See L<perlport/gmtime> for portability concerns.
2327 X<goto> X<jump> X<jmp>
2333 The C<goto-LABEL> form finds the statement labeled with LABEL and resumes
2334 execution there. It may not be used to go into any construct that
2335 requires initialization, such as a subroutine or a C<foreach> loop. It
2336 also can't be used to go into a construct that is optimized away,
2337 or to get out of a block or subroutine given to C<sort>.
2338 It can be used to go almost anywhere else within the dynamic scope,
2339 including out of subroutines, but it's usually better to use some other
2340 construct such as C<last> or C<die>. The author of Perl has never felt the
2341 need to use this form of C<goto> (in Perl, that is--C is another matter).
2342 (The difference being that C does not offer named loops combined with
2343 loop control. Perl does, and this replaces most structured uses of C<goto>
2344 in other languages.)
2346 The C<goto-EXPR> form expects a label name, whose scope will be resolved
2347 dynamically. This allows for computed C<goto>s per FORTRAN, but isn't
2348 necessarily recommended if you're optimizing for maintainability:
2350 goto ("FOO", "BAR", "GLARCH")[$i];
2352 The C<goto-&NAME> form is quite different from the other forms of
2353 C<goto>. In fact, it isn't a goto in the normal sense at all, and
2354 doesn't have the stigma associated with other gotos. Instead, it
2355 exits the current subroutine (losing any changes set by local()) and
2356 immediately calls in its place the named subroutine using the current
2357 value of @_. This is used by C<AUTOLOAD> subroutines that wish to
2358 load another subroutine and then pretend that the other subroutine had
2359 been called in the first place (except that any modifications to C<@_>
2360 in the current subroutine are propagated to the other subroutine.)
2361 After the C<goto>, not even C<caller> will be able to tell that this
2362 routine was called first.
2364 NAME needn't be the name of a subroutine; it can be a scalar variable
2365 containing a code reference, or a block that evaluates to a code
2368 =item grep BLOCK LIST
2371 =item grep EXPR,LIST
2373 This is similar in spirit to, but not the same as, grep(1) and its
2374 relatives. In particular, it is not limited to using regular expressions.
2376 Evaluates the BLOCK or EXPR for each element of LIST (locally setting
2377 C<$_> to each element) and returns the list value consisting of those
2378 elements for which the expression evaluated to true. In scalar
2379 context, returns the number of times the expression was true.
2381 @foo = grep(!/^#/, @bar); # weed out comments
2385 @foo = grep {!/^#/} @bar; # weed out comments
2387 Note that C<$_> is an alias to the list value, so it can be used to
2388 modify the elements of the LIST. While this is useful and supported,
2389 it can cause bizarre results if the elements of LIST are not variables.
2390 Similarly, grep returns aliases into the original list, much as a for
2391 loop's index variable aliases the list elements. That is, modifying an
2392 element of a list returned by grep (for example, in a C<foreach>, C<map>
2393 or another C<grep>) actually modifies the element in the original list.
2394 This is usually something to be avoided when writing clear code.
2396 If C<$_> is lexical in the scope where the C<grep> appears (because it has
2397 been declared with C<my $_>) then, in addition to being locally aliased to
2398 the list elements, C<$_> keeps being lexical inside the block; i.e. it
2399 can't be seen from the outside, avoiding any potential side-effects.
2401 See also L</map> for a list composed of the results of the BLOCK or EXPR.
2404 X<hex> X<hexadecimal>
2408 Interprets EXPR as a hex string and returns the corresponding value.
2409 (To convert strings that might start with either C<0>, C<0x>, or C<0b>, see
2410 L</oct>.) If EXPR is omitted, uses C<$_>.
2412 print hex '0xAf'; # prints '175'
2413 print hex 'aF'; # same
2415 Hex strings may only represent integers. Strings that would cause
2416 integer overflow trigger a warning. Leading whitespace is not stripped,
2417 unlike oct(). To present something as hex, look into L</printf>,
2418 L</sprintf>, or L</unpack>.
2423 There is no builtin C<import> function. It is just an ordinary
2424 method (subroutine) defined (or inherited) by modules that wish to export
2425 names to another module. The C<use> function calls the C<import> method
2426 for the package used. See also L</use>, L<perlmod>, and L<Exporter>.
2428 =item index STR,SUBSTR,POSITION
2429 X<index> X<indexOf> X<InStr>
2431 =item index STR,SUBSTR
2433 The index function searches for one string within another, but without
2434 the wildcard-like behavior of a full regular-expression pattern match.
2435 It returns the position of the first occurrence of SUBSTR in STR at
2436 or after POSITION. If POSITION is omitted, starts searching from the
2437 beginning of the string. POSITION before the beginning of the string
2438 or after its end is treated as if it were the beginning or the end,
2439 respectively. POSITION and the return value are based at C<0> (or whatever
2440 you've set the C<$[> variable to--but don't do that). If the substring
2441 is not found, C<index> returns one less than the base, ordinarily C<-1>.
2444 X<int> X<integer> X<truncate> X<trunc>
2448 Returns the integer portion of EXPR. If EXPR is omitted, uses C<$_>.
2449 You should not use this function for rounding: one because it truncates
2450 towards C<0>, and two because machine representations of floating point
2451 numbers can sometimes produce counterintuitive results. For example,
2452 C<int(-6.725/0.025)> produces -268 rather than the correct -269; that's
2453 because it's really more like -268.99999999999994315658 instead. Usually,
2454 the C<sprintf>, C<printf>, or the C<POSIX::floor> and C<POSIX::ceil>
2455 functions will serve you better than will int().
2457 =item ioctl FILEHANDLE,FUNCTION,SCALAR
2460 Implements the ioctl(2) function. You'll probably first have to say
2462 require "sys/ioctl.ph"; # probably in $Config{archlib}/sys/ioctl.ph
2464 to get the correct function definitions. If F<sys/ioctl.ph> doesn't
2465 exist or doesn't have the correct definitions you'll have to roll your
2466 own, based on your C header files such as F<< <sys/ioctl.h> >>.
2467 (There is a Perl script called B<h2ph> that comes with the Perl kit that
2468 may help you in this, but it's nontrivial.) SCALAR will be read and/or
2469 written depending on the FUNCTION--a pointer to the string value of SCALAR
2470 will be passed as the third argument of the actual C<ioctl> call. (If SCALAR
2471 has no string value but does have a numeric value, that value will be
2472 passed rather than a pointer to the string value. To guarantee this to be
2473 true, add a C<0> to the scalar before using it.) The C<pack> and C<unpack>
2474 functions may be needed to manipulate the values of structures used by
2477 The return value of C<ioctl> (and C<fcntl>) is as follows:
2479 if OS returns: then Perl returns:
2481 0 string "0 but true"
2482 anything else that number
2484 Thus Perl returns true on success and false on failure, yet you can
2485 still easily determine the actual value returned by the operating
2488 $retval = ioctl(...) || -1;
2489 printf "System returned %d\n", $retval;
2491 The special string C<"0 but true"> is exempt from B<-w> complaints
2492 about improper numeric conversions.
2494 =item join EXPR,LIST
2497 Joins the separate strings of LIST into a single string with fields
2498 separated by the value of EXPR, and returns that new string. Example:
2500 $rec = join(':', $login,$passwd,$uid,$gid,$gcos,$home,$shell);
2502 Beware that unlike C<split>, C<join> doesn't take a pattern as its
2503 first argument. Compare L</split>.
2508 Returns a list consisting of all the keys of the named hash.
2509 (In scalar context, returns the number of keys.)
2511 The keys are returned in an apparently random order. The actual
2512 random order is subject to change in future versions of perl, but it
2513 is guaranteed to be the same order as either the C<values> or C<each>
2514 function produces (given that the hash has not been modified). Since
2515 Perl 5.8.1 the ordering is different even between different runs of
2516 Perl for security reasons (see L<perlsec/"Algorithmic Complexity
2519 As a side effect, calling keys() resets the HASH's internal iterator
2520 (see L</each>). In particular, calling keys() in void context resets
2521 the iterator with no other overhead.
2523 Here is yet another way to print your environment:
2526 @values = values %ENV;
2528 print pop(@keys), '=', pop(@values), "\n";
2531 or how about sorted by key:
2533 foreach $key (sort(keys %ENV)) {
2534 print $key, '=', $ENV{$key}, "\n";
2537 The returned values are copies of the original keys in the hash, so
2538 modifying them will not affect the original hash. Compare L</values>.
2540 To sort a hash by value, you'll need to use a C<sort> function.
2541 Here's a descending numeric sort of a hash by its values:
2543 foreach $key (sort { $hash{$b} <=> $hash{$a} } keys %hash) {
2544 printf "%4d %s\n", $hash{$key}, $key;
2547 As an lvalue C<keys> allows you to increase the number of hash buckets
2548 allocated for the given hash. This can gain you a measure of efficiency if
2549 you know the hash is going to get big. (This is similar to pre-extending
2550 an array by assigning a larger number to $#array.) If you say
2554 then C<%hash> will have at least 200 buckets allocated for it--256 of them,
2555 in fact, since it rounds up to the next power of two. These
2556 buckets will be retained even if you do C<%hash = ()>, use C<undef
2557 %hash> if you want to free the storage while C<%hash> is still in scope.
2558 You can't shrink the number of buckets allocated for the hash using
2559 C<keys> in this way (but you needn't worry about doing this by accident,
2560 as trying has no effect).
2562 See also C<each>, C<values> and C<sort>.
2564 =item kill SIGNAL, LIST
2567 Sends a signal to a list of processes. Returns the number of
2568 processes successfully signaled (which is not necessarily the
2569 same as the number actually killed).
2571 $cnt = kill 1, $child1, $child2;
2574 If SIGNAL is zero, no signal is sent to the process. This is a
2575 useful way to check that a child process is alive and hasn't changed
2576 its UID. See L<perlport> for notes on the portability of this
2579 Unlike in the shell, if SIGNAL is negative, it kills
2580 process groups instead of processes. (On System V, a negative I<PROCESS>
2581 number will also kill process groups, but that's not portable.) That
2582 means you usually want to use positive not negative signals. You may also
2583 use a signal name in quotes.
2585 See L<perlipc/"Signals"> for more details.
2592 The C<last> command is like the C<break> statement in C (as used in
2593 loops); it immediately exits the loop in question. If the LABEL is
2594 omitted, the command refers to the innermost enclosing loop. The
2595 C<continue> block, if any, is not executed:
2597 LINE: while (<STDIN>) {
2598 last LINE if /^$/; # exit when done with header
2602 C<last> cannot be used to exit a block which returns a value such as
2603 C<eval {}>, C<sub {}> or C<do {}>, and should not be used to exit
2604 a grep() or map() operation.
2606 Note that a block by itself is semantically identical to a loop
2607 that executes once. Thus C<last> can be used to effect an early
2608 exit out of such a block.
2610 See also L</continue> for an illustration of how C<last>, C<next>, and
2618 Returns a lowercased version of EXPR. This is the internal function
2619 implementing the C<\L> escape in double-quoted strings. Respects
2620 current LC_CTYPE locale if C<use locale> in force. See L<perllocale>
2621 and L<perlunicode> for more details about locale and Unicode support.
2623 If EXPR is omitted, uses C<$_>.
2626 X<lcfirst> X<lowercase>
2630 Returns the value of EXPR with the first character lowercased. This
2631 is the internal function implementing the C<\l> escape in
2632 double-quoted strings. Respects current LC_CTYPE locale if C<use
2633 locale> in force. See L<perllocale> and L<perlunicode> for more
2634 details about locale and Unicode support.
2636 If EXPR is omitted, uses C<$_>.
2643 Returns the length in I<characters> of the value of EXPR. If EXPR is
2644 omitted, returns length of C<$_>. Note that this cannot be used on
2645 an entire array or hash to find out how many elements these have.
2646 For that, use C<scalar @array> and C<scalar keys %hash> respectively.
2648 Note the I<characters>: if the EXPR is in Unicode, you will get the
2649 number of characters, not the number of bytes. To get the length
2650 in bytes, use C<do { use bytes; length(EXPR) }>, see L<bytes>.
2652 =item link OLDFILE,NEWFILE
2655 Creates a new filename linked to the old filename. Returns true for
2656 success, false otherwise.
2658 =item listen SOCKET,QUEUESIZE
2661 Does the same thing that the listen system call does. Returns true if
2662 it succeeded, false otherwise. See the example in
2663 L<perlipc/"Sockets: Client/Server Communication">.
2668 You really probably want to be using C<my> instead, because C<local> isn't
2669 what most people think of as "local". See
2670 L<perlsub/"Private Variables via my()"> for details.
2672 A local modifies the listed variables to be local to the enclosing
2673 block, file, or eval. If more than one value is listed, the list must
2674 be placed in parentheses. See L<perlsub/"Temporary Values via local()">
2675 for details, including issues with tied arrays and hashes.
2677 =item localtime EXPR
2682 Converts a time as returned by the time function to a 9-element list
2683 with the time analyzed for the local time zone. Typically used as
2687 ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) =
2690 All list elements are numeric, and come straight out of the C `struct
2691 tm'. C<$sec>, C<$min>, and C<$hour> are the seconds, minutes, and hours
2692 of the specified time.
2694 C<$mday> is the day of the month, and C<$mon> is the month itself, in
2695 the range C<0..11> with 0 indicating January and 11 indicating December.
2696 This makes it easy to get a month name from a list:
2698 my @abbr = qw( Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec );
2699 print "$abbr[$mon] $mday";
2700 # $mon=9, $mday=18 gives "Oct 18"
2702 C<$year> is the number of years since 1900, not just the last two digits
2703 of the year. That is, C<$year> is C<123> in year 2023. The proper way
2704 to get a complete 4-digit year is simply:
2708 To get the last two digits of the year (e.g., '01' in 2001) do:
2710 $year = sprintf("%02d", $year % 100);
2712 C<$wday> is the day of the week, with 0 indicating Sunday and 3 indicating
2713 Wednesday. C<$yday> is the day of the year, in the range C<0..364>
2714 (or C<0..365> in leap years.)
2716 C<$isdst> is true if the specified time occurs during Daylight Saving
2717 Time, false otherwise.
2719 If EXPR is omitted, C<localtime()> uses the current time (C<localtime(time)>).
2721 In scalar context, C<localtime()> returns the ctime(3) value:
2723 $now_string = localtime; # e.g., "Thu Oct 13 04:54:34 1994"
2725 This scalar value is B<not> locale dependent but is a Perl builtin. For GMT
2726 instead of local time use the L</gmtime> builtin. See also the
2727 C<Time::Local> module (to convert the second, minutes, hours, ... back to
2728 the integer value returned by time()), and the L<POSIX> module's strftime(3)
2729 and mktime(3) functions.
2731 To get somewhat similar but locale dependent date strings, set up your
2732 locale environment variables appropriately (please see L<perllocale>) and
2735 use POSIX qw(strftime);
2736 $now_string = strftime "%a %b %e %H:%M:%S %Y", localtime;
2737 # or for GMT formatted appropriately for your locale:
2738 $now_string = strftime "%a %b %e %H:%M:%S %Y", gmtime;
2740 Note that the C<%a> and C<%b>, the short forms of the day of the week
2741 and the month of the year, may not necessarily be three characters wide.
2743 See L<perlport/localtime> for portability concerns.
2748 This function places an advisory lock on a shared variable, or referenced
2749 object contained in I<THING> until the lock goes out of scope.
2751 lock() is a "weak keyword" : this means that if you've defined a function
2752 by this name (before any calls to it), that function will be called
2753 instead. (However, if you've said C<use threads>, lock() is always a
2754 keyword.) See L<threads>.
2757 X<log> X<logarithm> X<e> X<ln> X<base>
2761 Returns the natural logarithm (base I<e>) of EXPR. If EXPR is omitted,
2762 returns log of C<$_>. To get the log of another base, use basic algebra:
2763 The base-N log of a number is equal to the natural log of that number
2764 divided by the natural log of N. For example:
2768 return log($n)/log(10);
2771 See also L</exp> for the inverse operation.
2778 Does the same thing as the C<stat> function (including setting the
2779 special C<_> filehandle) but stats a symbolic link instead of the file
2780 the symbolic link points to. If symbolic links are unimplemented on
2781 your system, a normal C<stat> is done. For much more detailed
2782 information, please see the documentation for C<stat>.
2784 If EXPR is omitted, stats C<$_>.
2788 The match operator. See L<perlop>.
2790 =item map BLOCK LIST
2795 Evaluates the BLOCK or EXPR for each element of LIST (locally setting
2796 C<$_> to each element) and returns the list value composed of the
2797 results of each such evaluation. In scalar context, returns the
2798 total number of elements so generated. Evaluates BLOCK or EXPR in
2799 list context, so each element of LIST may produce zero, one, or
2800 more elements in the returned value.
2802 @chars = map(chr, @nums);
2804 translates a list of numbers to the corresponding characters. And
2806 %hash = map { getkey($_) => $_ } @array;
2808 is just a funny way to write
2811 foreach $_ (@array) {
2812 $hash{getkey($_)} = $_;
2815 Note that C<$_> is an alias to the list value, so it can be used to
2816 modify the elements of the LIST. While this is useful and supported,
2817 it can cause bizarre results if the elements of LIST are not variables.
2818 Using a regular C<foreach> loop for this purpose would be clearer in
2819 most cases. See also L</grep> for an array composed of those items of
2820 the original list for which the BLOCK or EXPR evaluates to true.
2822 If C<$_> is lexical in the scope where the C<map> appears (because it has
2823 been declared with C<my $_>) then, in addition to being locally aliased to
2824 the list elements, C<$_> keeps being lexical inside the block; i.e. it
2825 can't be seen from the outside, avoiding any potential side-effects.
2827 C<{> starts both hash references and blocks, so C<map { ...> could be either
2828 the start of map BLOCK LIST or map EXPR, LIST. Because perl doesn't look
2829 ahead for the closing C<}> it has to take a guess at which its dealing with
2830 based what it finds just after the C<{>. Usually it gets it right, but if it
2831 doesn't it won't realize something is wrong until it gets to the C<}> and
2832 encounters the missing (or unexpected) comma. The syntax error will be
2833 reported close to the C<}> but you'll need to change something near the C<{>
2834 such as using a unary C<+> to give perl some help:
2836 %hash = map { "\L$_", 1 } @array # perl guesses EXPR. wrong
2837 %hash = map { +"\L$_", 1 } @array # perl guesses BLOCK. right
2838 %hash = map { ("\L$_", 1) } @array # this also works
2839 %hash = map { lc($_), 1 } @array # as does this.
2840 %hash = map +( lc($_), 1 ), @array # this is EXPR and works!
2842 %hash = map ( lc($_), 1 ), @array # evaluates to (1, @array)
2844 or to force an anon hash constructor use C<+{>
2846 @hashes = map +{ lc($_), 1 }, @array # EXPR, so needs , at end
2848 and you get list of anonymous hashes each with only 1 entry.
2850 =item mkdir FILENAME,MASK
2851 X<mkdir> X<md> X<directory, create>
2853 =item mkdir FILENAME
2857 Creates the directory specified by FILENAME, with permissions
2858 specified by MASK (as modified by C<umask>). If it succeeds it
2859 returns true, otherwise it returns false and sets C<$!> (errno).
2860 If omitted, MASK defaults to 0777. If omitted, FILENAME defaults
2863 In general, it is better to create directories with permissive MASK,
2864 and let the user modify that with their C<umask>, than it is to supply
2865 a restrictive MASK and give the user no way to be more permissive.
2866 The exceptions to this rule are when the file or directory should be
2867 kept private (mail files, for instance). The perlfunc(1) entry on
2868 C<umask> discusses the choice of MASK in more detail.
2870 Note that according to the POSIX 1003.1-1996 the FILENAME may have any
2871 number of trailing slashes. Some operating and filesystems do not get
2872 this right, so Perl automatically removes all trailing slashes to keep
2875 =item msgctl ID,CMD,ARG
2878 Calls the System V IPC function msgctl(2). You'll probably have to say
2882 first to get the correct constant definitions. If CMD is C<IPC_STAT>,
2883 then ARG must be a variable that will hold the returned C<msqid_ds>
2884 structure. Returns like C<ioctl>: the undefined value for error,
2885 C<"0 but true"> for zero, or the actual return value otherwise. See also
2886 L<perlipc/"SysV IPC">, C<IPC::SysV>, and C<IPC::Semaphore> documentation.
2888 =item msgget KEY,FLAGS
2891 Calls the System V IPC function msgget(2). Returns the message queue
2892 id, or the undefined value if there is an error. See also
2893 L<perlipc/"SysV IPC"> and C<IPC::SysV> and C<IPC::Msg> documentation.
2895 =item msgrcv ID,VAR,SIZE,TYPE,FLAGS
2898 Calls the System V IPC function msgrcv to receive a message from
2899 message queue ID into variable VAR with a maximum message size of
2900 SIZE. Note that when a message is received, the message type as a
2901 native long integer will be the first thing in VAR, followed by the
2902 actual message. This packing may be opened with C<unpack("l! a*")>.
2903 Taints the variable. Returns true if successful, or false if there is
2904 an error. See also L<perlipc/"SysV IPC">, C<IPC::SysV>, and
2905 C<IPC::SysV::Msg> documentation.
2907 =item msgsnd ID,MSG,FLAGS
2910 Calls the System V IPC function msgsnd to send the message MSG to the
2911 message queue ID. MSG must begin with the native long integer message
2912 type, and be followed by the length of the actual message, and finally
2913 the message itself. This kind of packing can be achieved with
2914 C<pack("l! a*", $type, $message)>. Returns true if successful,
2915 or false if there is an error. See also C<IPC::SysV>
2916 and C<IPC::SysV::Msg> documentation.
2923 =item my EXPR : ATTRS
2925 =item my TYPE EXPR : ATTRS
2927 A C<my> declares the listed variables to be local (lexically) to the
2928 enclosing block, file, or C<eval>. If more than one value is listed,
2929 the list must be placed in parentheses.
2931 The exact semantics and interface of TYPE and ATTRS are still
2932 evolving. TYPE is currently bound to the use of C<fields> pragma,
2933 and attributes are handled using the C<attributes> pragma, or starting
2934 from Perl 5.8.0 also via the C<Attribute::Handlers> module. See
2935 L<perlsub/"Private Variables via my()"> for details, and L<fields>,
2936 L<attributes>, and L<Attribute::Handlers>.
2943 The C<next> command is like the C<continue> statement in C; it starts
2944 the next iteration of the loop:
2946 LINE: while (<STDIN>) {
2947 next LINE if /^#/; # discard comments
2951 Note that if there were a C<continue> block on the above, it would get
2952 executed even on discarded lines. If the LABEL is omitted, the command
2953 refers to the innermost enclosing loop.
2955 C<next> cannot be used to exit a block which returns a value such as
2956 C<eval {}>, C<sub {}> or C<do {}>, and should not be used to exit
2957 a grep() or map() operation.
2959 Note that a block by itself is semantically identical to a loop
2960 that executes once. Thus C<next> will exit such a block early.
2962 See also L</continue> for an illustration of how C<last>, C<next>, and
2965 =item no Module VERSION LIST
2968 =item no Module VERSION
2970 =item no Module LIST
2974 See the C<use> function, of which C<no> is the opposite.
2977 X<oct> X<octal> X<hex> X<hexadecimal> X<binary> X<bin>
2981 Interprets EXPR as an octal string and returns the corresponding
2982 value. (If EXPR happens to start off with C<0x>, interprets it as a
2983 hex string. If EXPR starts off with C<0b>, it is interpreted as a
2984 binary string. Leading whitespace is ignored in all three cases.)
2985 The following will handle decimal, binary, octal, and hex in the standard
2988 $val = oct($val) if $val =~ /^0/;
2990 If EXPR is omitted, uses C<$_>. To go the other way (produce a number
2991 in octal), use sprintf() or printf():
2993 $perms = (stat("filename"))[2] & 07777;
2994 $oct_perms = sprintf "%lo", $perms;
2996 The oct() function is commonly used when a string such as C<644> needs
2997 to be converted into a file mode, for example. (Although perl will
2998 automatically convert strings into numbers as needed, this automatic
2999 conversion assumes base 10.)
3001 =item open FILEHANDLE,EXPR
3002 X<open> X<pipe> X<file, open> X<fopen>
3004 =item open FILEHANDLE,MODE,EXPR
3006 =item open FILEHANDLE,MODE,EXPR,LIST
3008 =item open FILEHANDLE,MODE,REFERENCE
3010 =item open FILEHANDLE
3012 Opens the file whose filename is given by EXPR, and associates it with
3015 (The following is a comprehensive reference to open(): for a gentler
3016 introduction you may consider L<perlopentut>.)
3018 If FILEHANDLE is an undefined scalar variable (or array or hash element)
3019 the variable is assigned a reference to a new anonymous filehandle,
3020 otherwise if FILEHANDLE is an expression, its value is used as the name of
3021 the real filehandle wanted. (This is considered a symbolic reference, so
3022 C<use strict 'refs'> should I<not> be in effect.)
3024 If EXPR is omitted, the scalar variable of the same name as the
3025 FILEHANDLE contains the filename. (Note that lexical variables--those
3026 declared with C<my>--will not work for this purpose; so if you're
3027 using C<my>, specify EXPR in your call to open.)
3029 If three or more arguments are specified then the mode of opening and
3030 the file name are separate. If MODE is C<< '<' >> or nothing, the file
3031 is opened for input. If MODE is C<< '>' >>, the file is truncated and
3032 opened for output, being created if necessary. If MODE is C<<< '>>' >>>,
3033 the file is opened for appending, again being created if necessary.
3035 You can put a C<'+'> in front of the C<< '>' >> or C<< '<' >> to
3036 indicate that you want both read and write access to the file; thus
3037 C<< '+<' >> is almost always preferred for read/write updates--the C<<
3038 '+>' >> mode would clobber the file first. You can't usually use
3039 either read-write mode for updating textfiles, since they have
3040 variable length records. See the B<-i> switch in L<perlrun> for a
3041 better approach. The file is created with permissions of C<0666>
3042 modified by the process' C<umask> value.
3044 These various prefixes correspond to the fopen(3) modes of C<'r'>,
3045 C<'r+'>, C<'w'>, C<'w+'>, C<'a'>, and C<'a+'>.
3047 In the 2-arguments (and 1-argument) form of the call the mode and
3048 filename should be concatenated (in this order), possibly separated by
3049 spaces. It is possible to omit the mode in these forms if the mode is
3052 If the filename begins with C<'|'>, the filename is interpreted as a
3053 command to which output is to be piped, and if the filename ends with a
3054 C<'|'>, the filename is interpreted as a command which pipes output to
3055 us. See L<perlipc/"Using open() for IPC">
3056 for more examples of this. (You are not allowed to C<open> to a command
3057 that pipes both in I<and> out, but see L<IPC::Open2>, L<IPC::Open3>,
3058 and L<perlipc/"Bidirectional Communication with Another Process">
3061 For three or more arguments if MODE is C<'|-'>, the filename is
3062 interpreted as a command to which output is to be piped, and if MODE
3063 is C<'-|'>, the filename is interpreted as a command which pipes
3064 output to us. In the 2-arguments (and 1-argument) form one should
3065 replace dash (C<'-'>) with the command.
3066 See L<perlipc/"Using open() for IPC"> for more examples of this.
3067 (You are not allowed to C<open> to a command that pipes both in I<and>
3068 out, but see L<IPC::Open2>, L<IPC::Open3>, and
3069 L<perlipc/"Bidirectional Communication"> for alternatives.)
3071 In the three-or-more argument form of pipe opens, if LIST is specified
3072 (extra arguments after the command name) then LIST becomes arguments
3073 to the command invoked if the platform supports it. The meaning of
3074 C<open> with more than three arguments for non-pipe modes is not yet
3075 specified. Experimental "layers" may give extra LIST arguments
3078 In the 2-arguments (and 1-argument) form opening C<'-'> opens STDIN
3079 and opening C<< '>-' >> opens STDOUT.
3081 You may use the three-argument form of open to specify IO "layers"
3082 (sometimes also referred to as "disciplines") to be applied to the handle
3083 that affect how the input and output are processed (see L<open> and
3084 L<PerlIO> for more details). For example
3086 open(FH, "<:utf8", "file")
3088 will open the UTF-8 encoded file containing Unicode characters,
3089 see L<perluniintro>. (Note that if layers are specified in the
3090 three-arg form then default layers set by the C<open> pragma are
3093 Open returns nonzero upon success, the undefined value otherwise. If
3094 the C<open> involved a pipe, the return value happens to be the pid of
3097 If you're running Perl on a system that distinguishes between text
3098 files and binary files, then you should check out L</binmode> for tips
3099 for dealing with this. The key distinction between systems that need
3100 C<binmode> and those that don't is their text file formats. Systems
3101 like Unix, Mac OS, and Plan 9, which delimit lines with a single
3102 character, and which encode that character in C as C<"\n">, do not
3103 need C<binmode>. The rest need it.
3105 When opening a file, it's usually a bad idea to continue normal execution
3106 if the request failed, so C<open> is frequently used in connection with
3107 C<die>. Even if C<die> won't do what you want (say, in a CGI script,
3108 where you want to make a nicely formatted error message (but there are
3109 modules that can help with that problem)) you should always check
3110 the return value from opening a file. The infrequent exception is when
3111 working with an unopened filehandle is actually what you want to do.
3113 As a special case the 3-arg form with a read/write mode and the third
3114 argument being C<undef>:
3116 open(TMP, "+>", undef) or die ...
3118 opens a filehandle to an anonymous temporary file. Also using "+<"
3119 works for symmetry, but you really should consider writing something
3120 to the temporary file first. You will need to seek() to do the
3123 Since v5.8.0, perl has built using PerlIO by default. Unless you've
3124 changed this (i.e. Configure -Uuseperlio), you can open file handles to
3125 "in memory" files held in Perl scalars via:
3127 open($fh, '>', \$variable) || ..
3129 Though if you try to re-open C<STDOUT> or C<STDERR> as an "in memory"
3130 file, you have to close it first:
3133 open STDOUT, '>', \$variable or die "Can't open STDOUT: $!";
3138 open ARTICLE or die "Can't find article $ARTICLE: $!\n";
3139 while (<ARTICLE>) {...
3141 open(LOG, '>>/usr/spool/news/twitlog'); # (log is reserved)
3142 # if the open fails, output is discarded
3144 open(DBASE, '+<', 'dbase.mine') # open for update
3145 or die "Can't open 'dbase.mine' for update: $!";
3147 open(DBASE, '+<dbase.mine') # ditto
3148 or die "Can't open 'dbase.mine' for update: $!";
3150 open(ARTICLE, '-|', "caesar <$article") # decrypt article
3151 or die "Can't start caesar: $!";
3153 open(ARTICLE, "caesar <$article |") # ditto
3154 or die "Can't start caesar: $!";
3156 open(EXTRACT, "|sort >Tmp$$") # $$ is our process id
3157 or die "Can't start sort: $!";
3160 open(MEMORY,'>', \$var)
3161 or die "Can't open memory file: $!";
3162 print MEMORY "foo!\n"; # output will end up in $var
3164 # process argument list of files along with any includes
3166 foreach $file (@ARGV) {
3167 process($file, 'fh00');
3171 my($filename, $input) = @_;
3172 $input++; # this is a string increment
3173 unless (open($input, $filename)) {
3174 print STDERR "Can't open $filename: $!\n";
3179 while (<$input>) { # note use of indirection
3180 if (/^#include "(.*)"/) {
3181 process($1, $input);
3188 See L<perliol> for detailed info on PerlIO.
3190 You may also, in the Bourne shell tradition, specify an EXPR beginning
3191 with C<< '>&' >>, in which case the rest of the string is interpreted
3192 as the name of a filehandle (or file descriptor, if numeric) to be
3193 duped (as L<dup(2)>) and opened. You may use C<&> after C<< > >>,
3194 C<<< >> >>>, C<< < >>, C<< +> >>, C<<< +>> >>>, and C<< +< >>.
3195 The mode you specify should match the mode of the original filehandle.
3196 (Duping a filehandle does not take into account any existing contents
3197 of IO buffers.) If you use the 3-arg form then you can pass either a
3198 number, the name of a filehandle or the normal "reference to a glob".
3200 Here is a script that saves, redirects, and restores C<STDOUT> and
3201 C<STDERR> using various methods:
3204 open my $oldout, ">&STDOUT" or die "Can't dup STDOUT: $!";
3205 open OLDERR, ">&", \*STDERR or die "Can't dup STDERR: $!";
3207 open STDOUT, '>', "foo.out" or die "Can't redirect STDOUT: $!";
3208 open STDERR, ">&STDOUT" or die "Can't dup STDOUT: $!";
3210 select STDERR; $| = 1; # make unbuffered
3211 select STDOUT; $| = 1; # make unbuffered
3213 print STDOUT "stdout 1\n"; # this works for
3214 print STDERR "stderr 1\n"; # subprocesses too
3216 open STDOUT, ">&", $oldout or die "Can't dup \$oldout: $!";
3217 open STDERR, ">&OLDERR" or die "Can't dup OLDERR: $!";
3219 print STDOUT "stdout 2\n";
3220 print STDERR "stderr 2\n";
3222 If you specify C<< '<&=X' >>, where C<X> is a file descriptor number
3223 or a filehandle, then Perl will do an equivalent of C's C<fdopen> of
3224 that file descriptor (and not call L<dup(2)>); this is more
3225 parsimonious of file descriptors. For example:
3227 # open for input, reusing the fileno of $fd
3228 open(FILEHANDLE, "<&=$fd")
3232 open(FILEHANDLE, "<&=", $fd)
3236 # open for append, using the fileno of OLDFH
3237 open(FH, ">>&=", OLDFH)
3241 open(FH, ">>&=OLDFH")
3243 Being parsimonious on filehandles is also useful (besides being
3244 parsimonious) for example when something is dependent on file
3245 descriptors, like for example locking using flock(). If you do just
3246 C<< open(A, '>>&B') >>, the filehandle A will not have the same file
3247 descriptor as B, and therefore flock(A) will not flock(B), and vice
3248 versa. But with C<< open(A, '>>&=B') >> the filehandles will share
3249 the same file descriptor.
3251 Note that if you are using Perls older than 5.8.0, Perl will be using
3252 the standard C libraries' fdopen() to implement the "=" functionality.
3253 On many UNIX systems fdopen() fails when file descriptors exceed a
3254 certain value, typically 255. For Perls 5.8.0 and later, PerlIO is
3255 most often the default.
3257 You can see whether Perl has been compiled with PerlIO or not by
3258 running C<perl -V> and looking for C<useperlio=> line. If C<useperlio>
3259 is C<define>, you have PerlIO, otherwise you don't.
3261 If you open a pipe on the command C<'-'>, i.e., either C<'|-'> or C<'-|'>
3262 with 2-arguments (or 1-argument) form of open(), then
3263 there is an implicit fork done, and the return value of open is the pid
3264 of the child within the parent process, and C<0> within the child
3265 process. (Use C<defined($pid)> to determine whether the open was successful.)
3266 The filehandle behaves normally for the parent, but i/o to that
3267 filehandle is piped from/to the STDOUT/STDIN of the child process.
3268 In the child process the filehandle isn't opened--i/o happens from/to
3269 the new STDOUT or STDIN. Typically this is used like the normal
3270 piped open when you want to exercise more control over just how the
3271 pipe command gets executed, such as when you are running setuid, and
3272 don't want to have to scan shell commands for metacharacters.
3273 The following triples are more or less equivalent:
3275 open(FOO, "|tr '[a-z]' '[A-Z]'");
3276 open(FOO, '|-', "tr '[a-z]' '[A-Z]'");
3277 open(FOO, '|-') || exec 'tr', '[a-z]', '[A-Z]';
3278 open(FOO, '|-', "tr", '[a-z]', '[A-Z]');
3280 open(FOO, "cat -n '$file'|");
3281 open(FOO, '-|', "cat -n '$file'");
3282 open(FOO, '-|') || exec 'cat', '-n', $file;
3283 open(FOO, '-|', "cat", '-n', $file);
3285 The last example in each block shows the pipe as "list form", which is
3286 not yet supported on all platforms. A good rule of thumb is that if
3287 your platform has true C<fork()> (in other words, if your platform is
3288 UNIX) you can use the list form.
3290 See L<perlipc/"Safe Pipe Opens"> for more examples of this.
3292 Beginning with v5.6.0, Perl will attempt to flush all files opened for
3293 output before any operation that may do a fork, but this may not be
3294 supported on some platforms (see L<perlport>). To be safe, you may need
3295 to set C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method
3296 of C<IO::Handle> on any open handles.
3298 On systems that support a close-on-exec flag on files, the flag will
3299 be set for the newly opened file descriptor as determined by the value
3300 of $^F. See L<perlvar/$^F>.
3302 Closing any piped filehandle causes the parent process to wait for the
3303 child to finish, and returns the status value in C<$?> and
3304 C<${^CHILD_ERROR_NATIVE}>.
3306 The filename passed to 2-argument (or 1-argument) form of open() will
3307 have leading and trailing whitespace deleted, and the normal
3308 redirection characters honored. This property, known as "magic open",
3309 can often be used to good effect. A user could specify a filename of
3310 F<"rsh cat file |">, or you could change certain filenames as needed:
3312 $filename =~ s/(.*\.gz)\s*$/gzip -dc < $1|/;
3313 open(FH, $filename) or die "Can't open $filename: $!";
3315 Use 3-argument form to open a file with arbitrary weird characters in it,
3317 open(FOO, '<', $file);
3319 otherwise it's necessary to protect any leading and trailing whitespace:
3321 $file =~ s#^(\s)#./$1#;
3322 open(FOO, "< $file\0");
3324 (this may not work on some bizarre filesystems). One should
3325 conscientiously choose between the I<magic> and 3-arguments form
3330 will allow the user to specify an argument of the form C<"rsh cat file |">,
3331 but will not work on a filename which happens to have a trailing space, while
3333 open IN, '<', $ARGV[0];
3335 will have exactly the opposite restrictions.
3337 If you want a "real" C C<open> (see L<open(2)> on your system), then you
3338 should use the C<sysopen> function, which involves no such magic (but
3339 may use subtly different filemodes than Perl open(), which is mapped
3340 to C fopen()). This is
3341 another way to protect your filenames from interpretation. For example:
3344 sysopen(HANDLE, $path, O_RDWR|O_CREAT|O_EXCL)
3345 or die "sysopen $path: $!";
3346 $oldfh = select(HANDLE); $| = 1; select($oldfh);
3347 print HANDLE "stuff $$\n";
3349 print "File contains: ", <HANDLE>;
3351 Using the constructor from the C<IO::Handle> package (or one of its
3352 subclasses, such as C<IO::File> or C<IO::Socket>), you can generate anonymous
3353 filehandles that have the scope of whatever variables hold references to
3354 them, and automatically close whenever and however you leave that scope:
3358 sub read_myfile_munged {
3360 my $handle = new IO::File;
3361 open($handle, "myfile") or die "myfile: $!";
3363 or return (); # Automatically closed here.
3364 mung $first or die "mung failed"; # Or here.
3365 return $first, <$handle> if $ALL; # Or here.
3369 See L</seek> for some details about mixing reading and writing.
3371 =item opendir DIRHANDLE,EXPR
3374 Opens a directory named EXPR for processing by C<readdir>, C<telldir>,
3375 C<seekdir>, C<rewinddir>, and C<closedir>. Returns true if successful.
3376 DIRHANDLE may be an expression whose value can be used as an indirect
3377 dirhandle, usually the real dirhandle name. If DIRHANDLE is an undefined
3378 scalar variable (or array or hash element), the variable is assigned a
3379 reference to a new anonymous dirhandle.
3380 DIRHANDLEs have their own namespace separate from FILEHANDLEs.
3387 Returns the numeric (the native 8-bit encoding, like ASCII or EBCDIC,
3388 or Unicode) value of the first character of EXPR. If EXPR is omitted,
3391 For the reverse, see L</chr>.
3392 See L<perlunicode> and L<encoding> for more about Unicode.
3399 =item our EXPR : ATTRS
3401 =item our TYPE EXPR : ATTRS
3403 C<our> associates a simple name with a package variable in the current
3404 package for use within the current scope. When C<use strict 'vars'> is in
3405 effect, C<our> lets you use declared global variables without qualifying
3406 them with package names, within the lexical scope of the C<our> declaration.
3407 In this way C<our> differs from C<use vars>, which is package scoped.
3409 Unlike C<my>, which both allocates storage for a variable and associates
3410 a simple name with that storage for use within the current scope, C<our>
3411 associates a simple name with a package variable in the current package,
3412 for use within the current scope. In other words, C<our> has the same
3413 scoping rules as C<my>, but does not necessarily create a
3416 If more than one value is listed, the list must be placed
3422 An C<our> declaration declares a global variable that will be visible
3423 across its entire lexical scope, even across package boundaries. The
3424 package in which the variable is entered is determined at the point
3425 of the declaration, not at the point of use. This means the following
3429 our $bar; # declares $Foo::bar for rest of lexical scope
3433 print $bar; # prints 20, as it refers to $Foo::bar
3435 Multiple C<our> declarations with the same name in the same lexical
3436 scope are allowed if they are in different packages. If they happen
3437 to be in the same package, Perl will emit warnings if you have asked
3438 for them, just like multiple C<my> declarations. Unlike a second
3439 C<my> declaration, which will bind the name to a fresh variable, a
3440 second C<our> declaration in the same package, in the same scope, is
3445 our $bar; # declares $Foo::bar for rest of lexical scope
3449 our $bar = 30; # declares $Bar::bar for rest of lexical scope
3450 print $bar; # prints 30
3452 our $bar; # emits warning but has no other effect
3453 print $bar; # still prints 30
3455 An C<our> declaration may also have a list of attributes associated
3458 The exact semantics and interface of TYPE and ATTRS are still
3459 evolving. TYPE is currently bound to the use of C<fields> pragma,
3460 and attributes are handled using the C<attributes> pragma, or starting
3461 from Perl 5.8.0 also via the C<Attribute::Handlers> module. See
3462 L<perlsub/"Private Variables via my()"> for details, and L<fields>,
3463 L<attributes>, and L<Attribute::Handlers>.
3465 The only currently recognized C<our()> attribute is C<unique> which
3466 indicates that a single copy of the global is to be used by all
3467 interpreters should the program happen to be running in a
3468 multi-interpreter environment. (The default behaviour would be for
3469 each interpreter to have its own copy of the global.) Examples:
3471 our @EXPORT : unique = qw(foo);
3472 our %EXPORT_TAGS : unique = (bar => [qw(aa bb cc)]);
3473 our $VERSION : unique = "1.00";
3475 Note that this attribute also has the effect of making the global
3476 readonly when the first new interpreter is cloned (for example,
3477 when the first new thread is created).
3479 Multi-interpreter environments can come to being either through the
3480 fork() emulation on Windows platforms, or by embedding perl in a
3481 multi-threaded application. The C<unique> attribute does nothing in
3482 all other environments.
3484 Warning: the current implementation of this attribute operates on the
3485 typeglob associated with the variable; this means that C<our $x : unique>
3486 also has the effect of C<our @x : unique; our %x : unique>. This may be
3489 =item pack TEMPLATE,LIST
3492 Takes a LIST of values and converts it into a string using the rules
3493 given by the TEMPLATE. The resulting string is the concatenation of
3494 the converted values. Typically, each converted value looks
3495 like its machine-level representation. For example, on 32-bit machines
3496 an integer may be represented by a sequence of 4 bytes that will be
3497 converted to a sequence of 4 characters.
3499 The TEMPLATE is a sequence of characters that give the order and type
3500 of values, as follows:
3502 a A string with arbitrary binary data, will be null padded.
3503 A A text (ASCII) string, will be space padded.
3504 Z A null terminated (ASCIZ) string, will be null padded.
3506 b A bit string (ascending bit order inside each byte, like vec()).
3507 B A bit string (descending bit order inside each byte).
3508 h A hex string (low nybble first).
3509 H A hex string (high nybble first).
3511 c A signed char (8-bit) value.
3512 C An unsigned C char (octet) even under Unicode. Should normally not
3513 be used. See U and W instead.
3514 W An unsigned char value (can be greater than 255).
3516 s A signed short (16-bit) value.
3517 S An unsigned short value.
3519 l A signed long (32-bit) value.
3520 L An unsigned long value.
3522 q A signed quad (64-bit) value.
3523 Q An unsigned quad value.
3524 (Quads are available only if your system supports 64-bit
3525 integer values _and_ if Perl has been compiled to support those.
3526 Causes a fatal error otherwise.)
3528 i A signed integer value.
3529 I A unsigned integer value.
3530 (This 'integer' is _at_least_ 32 bits wide. Its exact
3531 size depends on what a local C compiler calls 'int'.)
3533 n An unsigned short (16-bit) in "network" (big-endian) order.
3534 N An unsigned long (32-bit) in "network" (big-endian) order.
3535 v An unsigned short (16-bit) in "VAX" (little-endian) order.
3536 V An unsigned long (32-bit) in "VAX" (little-endian) order.
3538 j A Perl internal signed integer value (IV).
3539 J A Perl internal unsigned integer value (UV).
3541 f A single-precision float in the native format.
3542 d A double-precision float in the native format.
3544 F A Perl internal floating point value (NV) in the native format
3545 D A long double-precision float in the native format.
3546 (Long doubles are available only if your system supports long
3547 double values _and_ if Perl has been compiled to support those.
3548 Causes a fatal error otherwise.)
3550 p A pointer to a null-terminated string.
3551 P A pointer to a structure (fixed-length string).
3553 u A uuencoded string.
3554 U A Unicode character number. Encodes to UTF-8 internally
3555 (or UTF-EBCDIC in EBCDIC platforms).
3557 w A BER compressed integer (not an ASN.1 BER, see perlpacktut for
3558 details). Its bytes represent an unsigned integer in base 128,
3559 most significant digit first, with as few digits as possible. Bit
3560 eight (the high bit) is set on each byte except the last.
3564 @ Null fill or truncate to absolute position, counted from the
3565 start of the innermost ()-group.
3566 . Null fill or truncate to absolute position specified by value.
3567 ( Start of a ()-group.
3569 One or more of the modifiers below may optionally follow some letters in the
3570 TEMPLATE (the second column lists the letters for which the modifier is
3573 ! sSlLiI Forces native (short, long, int) sizes instead
3574 of fixed (16-/32-bit) sizes.
3576 xX Make x and X act as alignment commands.
3578 nNvV Treat integers as signed instead of unsigned.
3580 @. Specify position as byte offset in the internal
3581 representation of the packed string. Efficient but
3584 > sSiIlLqQ Force big-endian byte-order on the type.
3585 jJfFdDpP (The "big end" touches the construct.)
3587 < sSiIlLqQ Force little-endian byte-order on the type.
3588 jJfFdDpP (The "little end" touches the construct.)
3590 The C<E<gt>> and C<E<lt>> modifiers can also be used on C<()>-groups,
3591 in which case they force a certain byte-order on all components of
3592 that group, including subgroups.
3594 The following rules apply:
3600 Each letter may optionally be followed by a number giving a repeat
3601 count. With all types except C<a>, C<A>, C<Z>, C<b>, C<B>, C<h>,
3602 C<H>, C<@>, C<.>, C<x>, C<X> and C<P> the pack function will gobble up
3603 that many values from the LIST. A C<*> for the repeat count means to
3604 use however many items are left, except for C<@>, C<x>, C<X>, where it
3605 is equivalent to C<0>, for <.> where it means relative to string start
3606 and C<u>, where it is equivalent to 1 (or 45, which is the same).
3607 A numeric repeat count may optionally be enclosed in brackets, as in
3608 C<pack 'C[80]', @arr>.
3610 One can replace the numeric repeat count by a template enclosed in brackets;
3611 then the packed length of this template in bytes is used as a count.
3612 For example, C<x[L]> skips a long (it skips the number of bytes in a long);
3613 the template C<$t X[$t] $t> unpack()s twice what $t unpacks.
3614 If the template in brackets contains alignment commands (such as C<x![d]>),
3615 its packed length is calculated as if the start of the template has the maximal
3618 When used with C<Z>, C<*> results in the addition of a trailing null
3619 byte (so the packed result will be one longer than the byte C<length>
3622 When used with C<@>, the repeat count represents an offset from the start
3623 of the innermost () group.
3625 When used with C<.>, the repeat count is used to determine the starting
3626 position from where the value offset is calculated. If the repeat count
3627 is 0, it's relative to the current position. If the repeat count is C<*>,
3628 the offset is relative to the start of the packed string. And if its an
3629 integer C<n> the offset is relative to the start of the n-th innermost
3630 () group (or the start of the string if C<n> is bigger then the group
3633 The repeat count for C<u> is interpreted as the maximal number of bytes
3634 to encode per line of output, with 0, 1 and 2 replaced by 45. The repeat
3635 count should not be more than 65.
3639 The C<a>, C<A>, and C<Z> types gobble just one value, but pack it as a
3640 string of length count, padding with nulls or spaces as necessary. When
3641 unpacking, C<A> strips trailing whitespace and nulls, C<Z> strips everything
3642 after the first null, and C<a> returns data verbatim.
3644 If the value-to-pack is too long, it is truncated. If too long and an
3645 explicit count is provided, C<Z> packs only C<$count-1> bytes, followed
3646 by a null byte. Thus C<Z> always packs a trailing null (except when the
3651 Likewise, the C<b> and C<B> fields pack a string that many bits long.
3652 Each character of the input field of pack() generates 1 bit of the result.
3653 Each result bit is based on the least-significant bit of the corresponding
3654 input character, i.e., on C<ord($char)%2>. In particular, characters C<"0">
3655 and C<"1"> generate bits 0 and 1, as do characters C<"\0"> and C<"\1">.
3657 Starting from the beginning of the input string of pack(), each 8-tuple
3658 of characters is converted to 1 character of output. With format C<b>
3659 the first character of the 8-tuple determines the least-significant bit of a
3660 character, and with format C<B> it determines the most-significant bit of
3663 If the length of the input string is not exactly divisible by 8, the
3664 remainder is packed as if the input string were padded by null characters
3665 at the end. Similarly, during unpack()ing the "extra" bits are ignored.
3667 If the input string of pack() is longer than needed, extra characters are
3668 ignored. A C<*> for the repeat count of pack() means to use all the
3669 characters of the input field. On unpack()ing the bits are converted to a
3670 string of C<"0">s and C<"1">s.
3674 The C<h> and C<H> fields pack a string that many nybbles (4-bit groups,
3675 representable as hexadecimal digits, 0-9a-f) long.
3677 Each character of the input field of pack() generates 4 bits of the result.
3678 For non-alphabetical characters the result is based on the 4 least-significant
3679 bits of the input character, i.e., on C<ord($char)%16>. In particular,
3680 characters C<"0"> and C<"1"> generate nybbles 0 and 1, as do bytes
3681 C<"\0"> and C<"\1">. For characters C<"a".."f"> and C<"A".."F"> the result
3682 is compatible with the usual hexadecimal digits, so that C<"a"> and
3683 C<"A"> both generate the nybble C<0xa==10>. The result for characters
3684 C<"g".."z"> and C<"G".."Z"> is not well-defined.
3686 Starting from the beginning of the input string of pack(), each pair
3687 of characters is converted to 1 character of output. With format C<h> the
3688 first character of the pair determines the least-significant nybble of the
3689 output character, and with format C<H> it determines the most-significant
3692 If the length of the input string is not even, it behaves as if padded
3693 by a null character at the end. Similarly, during unpack()ing the "extra"
3694 nybbles are ignored.
3696 If the input string of pack() is longer than needed, extra characters are
3698 A C<*> for the repeat count of pack() means to use all the characters of
3699 the input field. On unpack()ing the nybbles are converted to a string
3700 of hexadecimal digits.
3704 The C<p> type packs a pointer to a null-terminated string. You are
3705 responsible for ensuring the string is not a temporary value (which can
3706 potentially get deallocated before you get around to using the packed result).
3707 The C<P> type packs a pointer to a structure of the size indicated by the
3708 length. A NULL pointer is created if the corresponding value for C<p> or
3709 C<P> is C<undef>, similarly for unpack().
3711 If your system has a strange pointer size (i.e. a pointer is neither as
3712 big as an int nor as big as a long), it may not be possible to pack or
3713 unpack pointers in big- or little-endian byte order. Attempting to do
3714 so will result in a fatal error.
3718 The C</> template character allows packing and unpacking of a sequence of
3719 items where the packed structure contains a packed item count followed by
3720 the packed items themselves.
3721 You write I<length-item>C</>I<sequence-item>.
3723 The I<length-item> can be any C<pack> template letter, and describes
3724 how the length value is packed. The ones likely to be of most use are
3725 integer-packing ones like C<n> (for Java strings), C<w> (for ASN.1 or
3726 SNMP) and C<N> (for Sun XDR).
3728 For C<pack>, the I<sequence-item> may have a repeat count, in which case
3729 the minimum of that and the number of available items is used as argument
3730 for the I<length-item>. If it has no repeat count or uses a '*', the number
3731 of available items is used. For C<unpack> the repeat count is always obtained
3732 by decoding the packed item count, and the I<sequence-item> must not have a
3735 If the I<sequence-item> refers to a string type (C<"A">, C<"a"> or C<"Z">),
3736 the I<length-item> is a string length, not a number of strings. If there is
3737 an explicit repeat count for pack, the packed string will be adjusted to that
3740 unpack 'W/a', "\04Gurusamy"; gives ('Guru')
3741 unpack 'a3/A* A*', '007 Bond J '; gives (' Bond', 'J')
3742 pack 'n/a* w/a','hello,','world'; gives "\000\006hello,\005world"
3743 pack 'a/W2', ord('a') .. ord('z'); gives '2ab'
3745 The I<length-item> is not returned explicitly from C<unpack>.
3747 Adding a count to the I<length-item> letter is unlikely to do anything
3748 useful, unless that letter is C<A>, C<a> or C<Z>. Packing with a
3749 I<length-item> of C<a> or C<Z> may introduce C<"\000"> characters,
3750 which Perl does not regard as legal in numeric strings.
3754 The integer types C<s>, C<S>, C<l>, and C<L> may be
3755 followed by a C<!> modifier to signify native shorts or
3756 longs--as you can see from above for example a bare C<l> does mean
3757 exactly 32 bits, the native C<long> (as seen by the local C compiler)
3758 may be larger. This is an issue mainly in 64-bit platforms. You can
3759 see whether using C<!> makes any difference by
3761 print length(pack("s")), " ", length(pack("s!")), "\n";
3762 print length(pack("l")), " ", length(pack("l!")), "\n";
3764 C<i!> and C<I!> also work but only because of completeness;
3765 they are identical to C<i> and C<I>.
3767 The actual sizes (in bytes) of native shorts, ints, longs, and long
3768 longs on the platform where Perl was built are also available via
3772 print $Config{shortsize}, "\n";
3773 print $Config{intsize}, "\n";
3774 print $Config{longsize}, "\n";
3775 print $Config{longlongsize}, "\n";
3777 (The C<$Config{longlongsize}> will be undefined if your system does
3778 not support long longs.)
3782 The integer formats C<s>, C<S>, C<i>, C<I>, C<l>, C<L>, C<j>, and C<J>
3783 are inherently non-portable between processors and operating systems
3784 because they obey the native byteorder and endianness. For example a
3785 4-byte integer 0x12345678 (305419896 decimal) would be ordered natively
3786 (arranged in and handled by the CPU registers) into bytes as
3788 0x12 0x34 0x56 0x78 # big-endian
3789 0x78 0x56 0x34 0x12 # little-endian
3791 Basically, the Intel and VAX CPUs are little-endian, while everybody
3792 else, for example Motorola m68k/88k, PPC, Sparc, HP PA, Power, and
3793 Cray are big-endian. Alpha and MIPS can be either: Digital/Compaq
3794 used/uses them in little-endian mode; SGI/Cray uses them in big-endian
3797 The names `big-endian' and `little-endian' are comic references to
3798 the classic "Gulliver's Travels" (via the paper "On Holy Wars and a
3799 Plea for Peace" by Danny Cohen, USC/ISI IEN 137, April 1, 1980) and
3800 the egg-eating habits of the Lilliputians.
3802 Some systems may have even weirder byte orders such as
3807 You can see your system's preference with
3809 print join(" ", map { sprintf "%#02x", $_ }
3810 unpack("W*",pack("L",0x12345678))), "\n";
3812 The byteorder on the platform where Perl was built is also available
3816 print $Config{byteorder}, "\n";
3818 Byteorders C<'1234'> and C<'12345678'> are little-endian, C<'4321'>
3819 and C<'87654321'> are big-endian.
3821 If you want portable packed integers you can either use the formats
3822 C<n>, C<N>, C<v>, and C<V>, or you can use the C<E<gt>> and C<E<lt>>
3823 modifiers. These modifiers are only available as of perl 5.9.2.
3824 See also L<perlport>.
3828 All integer and floating point formats as well as C<p> and C<P> and
3829 C<()>-groups may be followed by the C<E<gt>> or C<E<lt>> modifiers
3830 to force big- or little- endian byte-order, respectively.
3831 This is especially useful, since C<n>, C<N>, C<v> and C<V> don't cover
3832 signed integers, 64-bit integers and floating point values. However,
3833 there are some things to keep in mind.
3835 Exchanging signed integers between different platforms only works
3836 if all platforms store them in the same format. Most platforms store
3837 signed integers in two's complement, so usually this is not an issue.
3839 The C<E<gt>> or C<E<lt>> modifiers can only be used on floating point
3840 formats on big- or little-endian machines. Otherwise, attempting to
3841 do so will result in a fatal error.
3843 Forcing big- or little-endian byte-order on floating point values for
3844 data exchange can only work if all platforms are using the same
3845 binary representation (e.g. IEEE floating point format). Even if all
3846 platforms are using IEEE, there may be subtle differences. Being able
3847 to use C<E<gt>> or C<E<lt>> on floating point values can be very useful,
3848 but also very dangerous if you don't know exactly what you're doing.
3849 It is definitely not a general way to portably store floating point
3852 When using C<E<gt>> or C<E<lt>> on an C<()>-group, this will affect
3853 all types inside the group that accept the byte-order modifiers,
3854 including all subgroups. It will silently be ignored for all other
3855 types. You are not allowed to override the byte-order within a group
3856 that already has a byte-order modifier suffix.
3860 Real numbers (floats and doubles) are in the native machine format only;
3861 due to the multiplicity of floating formats around, and the lack of a
3862 standard "network" representation, no facility for interchange has been
3863 made. This means that packed floating point data written on one machine
3864 may not be readable on another - even if both use IEEE floating point
3865 arithmetic (as the endian-ness of the memory representation is not part
3866 of the IEEE spec). See also L<perlport>.
3868 If you know exactly what you're doing, you can use the C<E<gt>> or C<E<lt>>
3869 modifiers to force big- or little-endian byte-order on floating point values.
3871 Note that Perl uses doubles (or long doubles, if configured) internally for
3872 all numeric calculation, and converting from double into float and thence back
3873 to double again will lose precision (i.e., C<unpack("f", pack("f", $foo)>)
3874 will not in general equal $foo).
3878 Pack and unpack can operate in two modes, character mode (C<C0> mode) where
3879 the packed string is processed per character and UTF-8 mode (C<U0> mode)
3880 where the packed string is processed in its UTF-8-encoded Unicode form on
3881 a byte by byte basis. Character mode is the default unless the format string
3882 starts with an C<U>. You can switch mode at any moment with an explicit
3883 C<C0> or C<U0> in the format. A mode is in effect until the next mode switch
3884 or until the end of the ()-group in which it was entered.
3888 You must yourself do any alignment or padding by inserting for example
3889 enough C<'x'>es while packing. There is no way to pack() and unpack()
3890 could know where the characters are going to or coming from. Therefore
3891 C<pack> (and C<unpack>) handle their output and input as flat
3892 sequences of characters.
3896 A ()-group is a sub-TEMPLATE enclosed in parentheses. A group may
3897 take a repeat count, both as postfix, and for unpack() also via the C</>
3898 template character. Within each repetition of a group, positioning with
3899 C<@> starts again at 0. Therefore, the result of
3901 pack( '@1A((@2A)@3A)', 'a', 'b', 'c' )
3903 is the string "\0a\0\0bc".
3907 C<x> and C<X> accept C<!> modifier. In this case they act as
3908 alignment commands: they jump forward/back to the closest position
3909 aligned at a multiple of C<count> characters. For example, to pack() or
3910 unpack() C's C<struct {char c; double d; char cc[2]}> one may need to
3911 use the template C<W x![d] d W[2]>; this assumes that doubles must be
3912 aligned on the double's size.
3914 For alignment commands C<count> of 0 is equivalent to C<count> of 1;
3915 both result in no-ops.
3919 C<n>, C<N>, C<v> and C<V> accept the C<!> modifier. In this case they
3920 will represent signed 16-/32-bit integers in big-/little-endian order.
3921 This is only portable if all platforms sharing the packed data use the
3922 same binary representation for signed integers (e.g. all platforms are
3923 using two's complement representation).
3927 A comment in a TEMPLATE starts with C<#> and goes to the end of line.
3928 White space may be used to separate pack codes from each other, but
3929 modifiers and a repeat count must follow immediately.
3933 If TEMPLATE requires more arguments to pack() than actually given, pack()
3934 assumes additional C<""> arguments. If TEMPLATE requires fewer arguments
3935 to pack() than actually given, extra arguments are ignored.
3941 $foo = pack("WWWW",65,66,67,68);
3943 $foo = pack("W4",65,66,67,68);
3945 $foo = pack("W4",0x24b6,0x24b7,0x24b8,0x24b9);
3946 # same thing with Unicode circled letters.
3947 $foo = pack("U4",0x24b6,0x24b7,0x24b8,0x24b9);
3948 # same thing with Unicode circled letters. You don't get the UTF-8
3949 # bytes because the U at the start of the format caused a switch to
3950 # U0-mode, so the UTF-8 bytes get joined into characters
3951 $foo = pack("C0U4",0x24b6,0x24b7,0x24b8,0x24b9);
3952 # foo eq "\xe2\x92\xb6\xe2\x92\xb7\xe2\x92\xb8\xe2\x92\xb9"
3953 # This is the UTF-8 encoding of the string in the previous example
3955 $foo = pack("ccxxcc",65,66,67,68);
3958 # note: the above examples featuring "W" and "c" are true
3959 # only on ASCII and ASCII-derived systems such as ISO Latin 1
3960 # and UTF-8. In EBCDIC the first example would be
3961 # $foo = pack("WWWW",193,194,195,196);
3963 $foo = pack("s2",1,2);
3964 # "\1\0\2\0" on little-endian
3965 # "\0\1\0\2" on big-endian
3967 $foo = pack("a4","abcd","x","y","z");
3970 $foo = pack("aaaa","abcd","x","y","z");
3973 $foo = pack("a14","abcdefg");
3974 # "abcdefg\0\0\0\0\0\0\0"
3976 $foo = pack("i9pl", gmtime);
3977 # a real struct tm (on my system anyway)
3979 $utmp_template = "Z8 Z8 Z16 L";
3980 $utmp = pack($utmp_template, @utmp1);
3981 # a struct utmp (BSDish)
3983 @utmp2 = unpack($utmp_template, $utmp);
3984 # "@utmp1" eq "@utmp2"
3987 unpack("N", pack("B32", substr("0" x 32 . shift, -32)));
3990 $foo = pack('sx2l', 12, 34);
3991 # short 12, two zero bytes padding, long 34
3992 $bar = pack('s@4l', 12, 34);
3993 # short 12, zero fill to position 4, long 34
3995 $baz = pack('s.l', 12, 4, 34);
3996 # short 12, zero fill to position 4, long 34
3998 $foo = pack('nN', 42, 4711);
3999 # pack big-endian 16- and 32-bit unsigned integers
4000 $foo = pack('S>L>', 42, 4711);
4002 $foo = pack('s<l<', -42, 4711);
4003 # pack little-endian 16- and 32-bit signed integers
4004 $foo = pack('(sl)<', -42, 4711);
4007 The same template may generally also be used in unpack().
4009 =item package NAMESPACE
4010 X<package> X<module> X<namespace>
4014 Declares the compilation unit as being in the given namespace. The scope
4015 of the package declaration is from the declaration itself through the end
4016 of the enclosing block, file, or eval (the same as the C<my> operator).
4017 All further unqualified dynamic identifiers will be in this namespace.
4018 A package statement affects only dynamic variables--including those
4019 you've used C<local> on--but I<not> lexical variables, which are created
4020 with C<my>. Typically it would be the first declaration in a file to
4021 be included by the C<require> or C<use> operator. You can switch into a
4022 package in more than one place; it merely influences which symbol table
4023 is used by the compiler for the rest of that block. You can refer to
4024 variables and filehandles in other packages by prefixing the identifier
4025 with the package name and a double colon: C<$Package::Variable>.
4026 If the package name is null, the C<main> package as assumed. That is,
4027 C<$::sail> is equivalent to C<$main::sail> (as well as to C<$main'sail>,
4028 still seen in older code).
4030 If NAMESPACE is omitted, then there is no current package, and all
4031 identifiers must be fully qualified or lexicals. However, you are
4032 strongly advised not to make use of this feature. Its use can cause
4033 unexpected behaviour, even crashing some versions of Perl. It is
4034 deprecated, and will be removed from a future release.
4036 See L<perlmod/"Packages"> for more information about packages, modules,
4037 and classes. See L<perlsub> for other scoping issues.
4039 =item pipe READHANDLE,WRITEHANDLE
4042 Opens a pair of connected pipes like the corresponding system call.
4043 Note that if you set up a loop of piped processes, deadlock can occur
4044 unless you are very careful. In addition, note that Perl's pipes use
4045 IO buffering, so you may need to set C<$|> to flush your WRITEHANDLE
4046 after each command, depending on the application.
4048 See L<IPC::Open2>, L<IPC::Open3>, and L<perlipc/"Bidirectional Communication">
4049 for examples of such things.
4051 On systems that support a close-on-exec flag on files, the flag will be set
4052 for the newly opened file descriptors as determined by the value of $^F.
4060 Pops and returns the last value of the array, shortening the array by
4061 one element. Has an effect similar to
4065 If there are no elements in the array, returns the undefined value
4066 (although this may happen at other times as well). If ARRAY is
4067 omitted, pops the C<@ARGV> array in the main program, and the C<@_>
4068 array in subroutines, just like C<shift>.
4071 X<pos> X<match, position>
4075 Returns the offset of where the last C<m//g> search left off for the variable
4076 in question (C<$_> is used when the variable is not specified). Note that
4077 0 is a valid match offset. C<undef> indicates that the search position
4078 is reset (usually due to match failure, but can also be because no match has
4079 yet been performed on the scalar). C<pos> directly accesses the location used
4080 by the regexp engine to store the offset, so assigning to C<pos> will change
4081 that offset, and so will also influence the C<\G> zero-width assertion in
4082 regular expressions. Because a failed C<m//gc> match doesn't reset the offset,
4083 the return from C<pos> won't change either in this case. See L<perlre> and
4086 =item print FILEHANDLE LIST
4093 Prints a string or a list of strings. Returns true if successful.
4094 FILEHANDLE may be a scalar variable name, in which case the variable
4095 contains the name of or a reference to the filehandle, thus introducing
4096 one level of indirection. (NOTE: If FILEHANDLE is a variable and
4097 the next token is a term, it may be misinterpreted as an operator
4098 unless you interpose a C<+> or put parentheses around the arguments.)
4099 If FILEHANDLE is omitted, prints by default to standard output (or
4100 to the last selected output channel--see L</select>). If LIST is
4101 also omitted, prints C<$_> to the currently selected output channel.
4102 To set the default output channel to something other than STDOUT
4103 use the select operation. The current value of C<$,> (if any) is
4104 printed between each LIST item. The current value of C<$\> (if
4105 any) is printed after the entire LIST has been printed. Because
4106 print takes a LIST, anything in the LIST is evaluated in list
4107 context, and any subroutine that you call will have one or more of
4108 its expressions evaluated in list context. Also be careful not to
4109 follow the print keyword with a left parenthesis unless you want
4110 the corresponding right parenthesis to terminate the arguments to
4111 the print--interpose a C<+> or put parentheses around all the
4114 Note that if you're storing FILEHANDLEs in an array, or if you're using
4115 any other expression more complex than a scalar variable to retrieve it,
4116 you will have to use a block returning the filehandle value instead:
4118 print { $files[$i] } "stuff\n";
4119 print { $OK ? STDOUT : STDERR } "stuff\n";
4121 =item printf FILEHANDLE FORMAT, LIST
4124 =item printf FORMAT, LIST
4126 Equivalent to C<print FILEHANDLE sprintf(FORMAT, LIST)>, except that C<$\>
4127 (the output record separator) is not appended. The first argument
4128 of the list will be interpreted as the C<printf> format. See C<sprintf>
4129 for an explanation of the format argument. If C<use locale> is in effect,
4130 the character used for the decimal point in formatted real numbers is
4131 affected by the LC_NUMERIC locale. See L<perllocale>.
4133 Don't fall into the trap of using a C<printf> when a simple
4134 C<print> would do. The C<print> is more efficient and less
4137 =item prototype FUNCTION
4140 Returns the prototype of a function as a string (or C<undef> if the
4141 function has no prototype). FUNCTION is a reference to, or the name of,
4142 the function whose prototype you want to retrieve.
4144 If FUNCTION is a string starting with C<CORE::>, the rest is taken as a
4145 name for Perl builtin. If the builtin is not I<overridable> (such as
4146 C<qw//>) or its arguments cannot be expressed by a prototype (such as
4147 C<system>) returns C<undef> because the builtin does not really behave
4148 like a Perl function. Otherwise, the string describing the equivalent
4149 prototype is returned.
4151 =item push ARRAY,LIST
4154 Treats ARRAY as a stack, and pushes the values of LIST
4155 onto the end of ARRAY. The length of ARRAY increases by the length of
4156 LIST. Has the same effect as
4159 $ARRAY[++$#ARRAY] = $value;
4162 but is more efficient. Returns the number of elements in the array following
4163 the completed C<push>.
4175 Generalized quotes. See L<perlop/"Regexp Quote-Like Operators">.
4177 =item quotemeta EXPR
4178 X<quotemeta> X<metacharacter>
4182 Returns the value of EXPR with all non-"word"
4183 characters backslashed. (That is, all characters not matching
4184 C</[A-Za-z_0-9]/> will be preceded by a backslash in the
4185 returned string, regardless of any locale settings.)
4186 This is the internal function implementing
4187 the C<\Q> escape in double-quoted strings.
4189 If EXPR is omitted, uses C<$_>.
4196 Returns a random fractional number greater than or equal to C<0> and less
4197 than the value of EXPR. (EXPR should be positive.) If EXPR is
4198 omitted, the value C<1> is used. Currently EXPR with the value C<0> is
4199 also special-cased as C<1> - this has not been documented before perl 5.8.0
4200 and is subject to change in future versions of perl. Automatically calls
4201 C<srand> unless C<srand> has already been called. See also C<srand>.
4203 Apply C<int()> to the value returned by C<rand()> if you want random
4204 integers instead of random fractional numbers. For example,
4208 returns a random integer between C<0> and C<9>, inclusive.
4210 (Note: If your rand function consistently returns numbers that are too
4211 large or too small, then your version of Perl was probably compiled
4212 with the wrong number of RANDBITS.)
4214 =item read FILEHANDLE,SCALAR,LENGTH,OFFSET
4217 =item read FILEHANDLE,SCALAR,LENGTH
4219 Attempts to read LENGTH I<characters> of data into variable SCALAR
4220 from the specified FILEHANDLE. Returns the number of characters
4221 actually read, C<0> at end of file, or undef if there was an error (in
4222 the latter case C<$!> is also set). SCALAR will be grown or shrunk
4223 so that the last character actually read is the last character of the
4224 scalar after the read.
4226 An OFFSET may be specified to place the read data at some place in the
4227 string other than the beginning. A negative OFFSET specifies
4228 placement at that many characters counting backwards from the end of
4229 the string. A positive OFFSET greater than the length of SCALAR
4230 results in the string being padded to the required size with C<"\0">
4231 bytes before the result of the read is appended.
4233 The call is actually implemented in terms of either Perl's or system's
4234 fread() call. To get a true read(2) system call, see C<sysread>.
4236 Note the I<characters>: depending on the status of the filehandle,
4237 either (8-bit) bytes or characters are read. By default all
4238 filehandles operate on bytes, but for example if the filehandle has
4239 been opened with the C<:utf8> I/O layer (see L</open>, and the C<open>
4240 pragma, L<open>), the I/O will operate on UTF-8 encoded Unicode
4241 characters, not bytes. Similarly for the C<:encoding> pragma:
4242 in that case pretty much any characters can be read.
4244 =item readdir DIRHANDLE
4247 Returns the next directory entry for a directory opened by C<opendir>.
4248 If used in list context, returns all the rest of the entries in the
4249 directory. If there are no more entries, returns an undefined value in
4250 scalar context or a null list in list context.
4252 If you're planning to filetest the return values out of a C<readdir>, you'd
4253 better prepend the directory in question. Otherwise, because we didn't
4254 C<chdir> there, it would have been testing the wrong file.
4256 opendir(DIR, $some_dir) || die "can't opendir $some_dir: $!";
4257 @dots = grep { /^\./ && -f "$some_dir/$_" } readdir(DIR);
4261 X<readline> X<gets> X<fgets>
4263 Reads from the filehandle whose typeglob is contained in EXPR. In scalar
4264 context, each call reads and returns the next line, until end-of-file is
4265 reached, whereupon the subsequent call returns undef. In list context,
4266 reads until end-of-file is reached and returns a list of lines. Note that
4267 the notion of "line" used here is however you may have defined it
4268 with C<$/> or C<$INPUT_RECORD_SEPARATOR>). See L<perlvar/"$/">.
4270 When C<$/> is set to C<undef>, when readline() is in scalar
4271 context (i.e. file slurp mode), and when an empty file is read, it
4272 returns C<''> the first time, followed by C<undef> subsequently.
4274 This is the internal function implementing the C<< <EXPR> >>
4275 operator, but you can use it directly. The C<< <EXPR> >>
4276 operator is discussed in more detail in L<perlop/"I/O Operators">.
4279 $line = readline(*STDIN); # same thing
4281 If readline encounters an operating system error, C<$!> will be set with the
4282 corresponding error message. It can be helpful to check C<$!> when you are
4283 reading from filehandles you don't trust, such as a tty or a socket. The
4284 following example uses the operator form of C<readline>, and takes the necessary
4285 steps to ensure that C<readline> was successful.
4289 unless (defined( $line = <> )) {
4301 Returns the value of a symbolic link, if symbolic links are
4302 implemented. If not, gives a fatal error. If there is some system
4303 error, returns the undefined value and sets C<$!> (errno). If EXPR is
4304 omitted, uses C<$_>.
4309 EXPR is executed as a system command.
4310 The collected standard output of the command is returned.
4311 In scalar context, it comes back as a single (potentially
4312 multi-line) string. In list context, returns a list of lines
4313 (however you've defined lines with C<$/> or C<$INPUT_RECORD_SEPARATOR>).
4314 This is the internal function implementing the C<qx/EXPR/>
4315 operator, but you can use it directly. The C<qx/EXPR/>
4316 operator is discussed in more detail in L<perlop/"I/O Operators">.
4318 =item recv SOCKET,SCALAR,LENGTH,FLAGS
4321 Receives a message on a socket. Attempts to receive LENGTH characters
4322 of data into variable SCALAR from the specified SOCKET filehandle.
4323 SCALAR will be grown or shrunk to the length actually read. Takes the
4324 same flags as the system call of the same name. Returns the address
4325 of the sender if SOCKET's protocol supports this; returns an empty
4326 string otherwise. If there's an error, returns the undefined value.
4327 This call is actually implemented in terms of recvfrom(2) system call.
4328 See L<perlipc/"UDP: Message Passing"> for examples.
4330 Note the I<characters>: depending on the status of the socket, either
4331 (8-bit) bytes or characters are received. By default all sockets
4332 operate on bytes, but for example if the socket has been changed using
4333 binmode() to operate with the C<:utf8> I/O layer (see the C<open>
4334 pragma, L<open>), the I/O will operate on UTF-8 encoded Unicode
4335 characters, not bytes. Similarly for the C<:encoding> pragma:
4336 in that case pretty much any characters can be read.
4343 The C<redo> command restarts the loop block without evaluating the
4344 conditional again. The C<continue> block, if any, is not executed. If
4345 the LABEL is omitted, the command refers to the innermost enclosing
4346 loop. Programs that want to lie to themselves about what was just input
4347 normally use this command:
4349 # a simpleminded Pascal comment stripper
4350 # (warning: assumes no { or } in strings)
4351 LINE: while (<STDIN>) {
4352 while (s|({.*}.*){.*}|$1 |) {}
4357 if (/}/) { # end of comment?
4366 C<redo> cannot be used to retry a block which returns a value such as
4367 C<eval {}>, C<sub {}> or C<do {}>, and should not be used to exit
4368 a grep() or map() operation.
4370 Note that a block by itself is semantically identical to a loop
4371 that executes once. Thus C<redo> inside such a block will effectively
4372 turn it into a looping construct.
4374 See also L</continue> for an illustration of how C<last>, C<next>, and
4382 Returns a non-empty string if EXPR is a reference, the empty
4383 string otherwise. If EXPR
4384 is not specified, C<$_> will be used. The value returned depends on the
4385 type of thing the reference is a reference to.
4386 Builtin types include:
4396 If the referenced object has been blessed into a package, then that package
4397 name is returned instead. You can think of C<ref> as a C<typeof> operator.
4399 if (ref($r) eq "HASH") {
4400 print "r is a reference to a hash.\n";
4403 print "r is not a reference at all.\n";
4406 See also L<perlref>.
4408 =item rename OLDNAME,NEWNAME
4409 X<rename> X<move> X<mv> X<ren>
4411 Changes the name of a file; an existing file NEWNAME will be
4412 clobbered. Returns true for success, false otherwise.
4414 Behavior of this function varies wildly depending on your system
4415 implementation. For example, it will usually not work across file system
4416 boundaries, even though the system I<mv> command sometimes compensates
4417 for this. Other restrictions include whether it works on directories,
4418 open files, or pre-existing files. Check L<perlport> and either the
4419 rename(2) manpage or equivalent system documentation for details.
4421 =item require VERSION
4428 Demands a version of Perl specified by VERSION, or demands some semantics
4429 specified by EXPR or by C<$_> if EXPR is not supplied.
4431 VERSION may be either a numeric argument such as 5.006, which will be
4432 compared to C<$]>, or a literal of the form v5.6.1, which will be compared
4433 to C<$^V> (aka $PERL_VERSION). A fatal error is produced at run time if
4434 VERSION is greater than the version of the current Perl interpreter.
4435 Compare with L</use>, which can do a similar check at compile time.
4437 Specifying VERSION as a literal of the form v5.6.1 should generally be
4438 avoided, because it leads to misleading error messages under earlier
4439 versions of Perl that do not support this syntax. The equivalent numeric
4440 version should be used instead.
4442 require v5.6.1; # run time version check
4443 require 5.6.1; # ditto
4444 require 5.006_001; # ditto; preferred for backwards compatibility
4446 Otherwise, C<require> demands that a library file be included if it
4447 hasn't already been included. The file is included via the do-FILE
4448 mechanism, which is essentially just a variety of C<eval>. Has
4449 semantics similar to the following subroutine:
4452 my ($filename) = @_;
4453 if (exists $INC{$filename}) {
4454 return 1 if $INC{$filename};
4455 die "Compilation failed in require";
4457 my ($realfilename,$result);
4459 foreach $prefix (@INC) {
4460 $realfilename = "$prefix/$filename";
4461 if (-f $realfilename) {
4462 $INC{$filename} = $realfilename;
4463 $result = do $realfilename;
4467 die "Can't find $filename in \@INC";
4470 $INC{$filename} = undef;
4472 } elsif (!$result) {
4473 delete $INC{$filename};
4474 die "$filename did not return true value";
4480 Note that the file will not be included twice under the same specified
4483 The file must return true as the last statement to indicate
4484 successful execution of any initialization code, so it's customary to
4485 end such a file with C<1;> unless you're sure it'll return true
4486 otherwise. But it's better just to put the C<1;>, in case you add more
4489 If EXPR is a bareword, the require assumes a "F<.pm>" extension and
4490 replaces "F<::>" with "F</>" in the filename for you,
4491 to make it easy to load standard modules. This form of loading of
4492 modules does not risk altering your namespace.
4494 In other words, if you try this:
4496 require Foo::Bar; # a splendid bareword
4498 The require function will actually look for the "F<Foo/Bar.pm>" file in the
4499 directories specified in the C<@INC> array.
4501 But if you try this:
4503 $class = 'Foo::Bar';
4504 require $class; # $class is not a bareword
4506 require "Foo::Bar"; # not a bareword because of the ""
4508 The require function will look for the "F<Foo::Bar>" file in the @INC array and
4509 will complain about not finding "F<Foo::Bar>" there. In this case you can do:
4511 eval "require $class";
4513 Now that you understand how C<require> looks for files in the case of
4514 a bareword argument, there is a little extra functionality going on
4515 behind the scenes. Before C<require> looks for a "F<.pm>" extension,
4516 it will first look for a filename with a "F<.pmc>" extension. A file
4517 with this extension is assumed to be Perl bytecode generated by
4518 L<B::Bytecode|B::Bytecode>. If this file is found, and its modification
4519 time is newer than a coinciding "F<.pm>" non-compiled file, it will be
4520 loaded in place of that non-compiled file ending in a "F<.pm>" extension.
4522 You can also insert hooks into the import facility, by putting directly
4523 Perl code into the @INC array. There are three forms of hooks: subroutine
4524 references, array references and blessed objects.
4526 Subroutine references are the simplest case. When the inclusion system
4527 walks through @INC and encounters a subroutine, this subroutine gets
4528 called with two parameters, the first being a reference to itself, and the
4529 second the name of the file to be included (e.g. "F<Foo/Bar.pm>"). The
4530 subroutine should return C<undef> or a filehandle, from which the file to
4531 include will be read. If C<undef> is returned, C<require> will look at
4532 the remaining elements of @INC.
4534 If the hook is an array reference, its first element must be a subroutine
4535 reference. This subroutine is called as above, but the first parameter is
4536 the array reference. This enables to pass indirectly some arguments to
4539 In other words, you can write:
4541 push @INC, \&my_sub;
4543 my ($coderef, $filename) = @_; # $coderef is \&my_sub
4549 push @INC, [ \&my_sub, $x, $y, ... ];
4551 my ($arrayref, $filename) = @_;
4552 # Retrieve $x, $y, ...
4553 my @parameters = @$arrayref[1..$#$arrayref];
4557 If the hook is an object, it must provide an INC method that will be
4558 called as above, the first parameter being the object itself. (Note that
4559 you must fully qualify the sub's name, as it is always forced into package
4560 C<main>.) Here is a typical code layout:
4566 my ($self, $filename) = @_;
4570 # In the main program
4571 push @INC, new Foo(...);
4573 Note that these hooks are also permitted to set the %INC entry
4574 corresponding to the files they have loaded. See L<perlvar/%INC>.
4576 For a yet-more-powerful import facility, see L</use> and L<perlmod>.
4583 Generally used in a C<continue> block at the end of a loop to clear
4584 variables and reset C<??> searches so that they work again. The
4585 expression is interpreted as a list of single characters (hyphens
4586 allowed for ranges). All variables and arrays beginning with one of
4587 those letters are reset to their pristine state. If the expression is
4588 omitted, one-match searches (C<?pattern?>) are reset to match again. Resets
4589 only variables or searches in the current package. Always returns
4592 reset 'X'; # reset all X variables
4593 reset 'a-z'; # reset lower case variables
4594 reset; # just reset ?one-time? searches
4596 Resetting C<"A-Z"> is not recommended because you'll wipe out your
4597 C<@ARGV> and C<@INC> arrays and your C<%ENV> hash. Resets only package
4598 variables--lexical variables are unaffected, but they clean themselves
4599 up on scope exit anyway, so you'll probably want to use them instead.
4607 Returns from a subroutine, C<eval>, or C<do FILE> with the value
4608 given in EXPR. Evaluation of EXPR may be in list, scalar, or void
4609 context, depending on how the return value will be used, and the context
4610 may vary from one execution to the next (see C<wantarray>). If no EXPR
4611 is given, returns an empty list in list context, the undefined value in
4612 scalar context, and (of course) nothing at all in a void context.
4614 (Note that in the absence of an explicit C<return>, a subroutine, eval,
4615 or do FILE will automatically return the value of the last expression
4619 X<reverse> X<rev> X<invert>
4621 In list context, returns a list value consisting of the elements
4622 of LIST in the opposite order. In scalar context, concatenates the
4623 elements of LIST and returns a string value with all characters
4624 in the opposite order.
4626 print reverse <>; # line tac, last line first
4628 undef $/; # for efficiency of <>
4629 print scalar reverse <>; # character tac, last line tsrif
4631 Used without arguments in scalar context, reverse() reverses C<$_>.
4633 This operator is also handy for inverting a hash, although there are some
4634 caveats. If a value is duplicated in the original hash, only one of those
4635 can be represented as a key in the inverted hash. Also, this has to
4636 unwind one hash and build a whole new one, which may take some time
4637 on a large hash, such as from a DBM file.
4639 %by_name = reverse %by_address; # Invert the hash
4641 =item rewinddir DIRHANDLE
4644 Sets the current position to the beginning of the directory for the
4645 C<readdir> routine on DIRHANDLE.
4647 =item rindex STR,SUBSTR,POSITION
4650 =item rindex STR,SUBSTR
4652 Works just like index() except that it returns the position of the LAST
4653 occurrence of SUBSTR in STR. If POSITION is specified, returns the
4654 last occurrence at or before that position.
4656 =item rmdir FILENAME
4657 X<rmdir> X<rd> X<directory, remove>
4661 Deletes the directory specified by FILENAME if that directory is
4662 empty. If it succeeds it returns true, otherwise it returns false and
4663 sets C<$!> (errno). If FILENAME is omitted, uses C<$_>.
4667 The substitution operator. See L<perlop>.
4670 X<scalar> X<context>
4672 Forces EXPR to be interpreted in scalar context and returns the value
4675 @counts = ( scalar @a, scalar @b, scalar @c );
4677 There is no equivalent operator to force an expression to
4678 be interpolated in list context because in practice, this is never
4679 needed. If you really wanted to do so, however, you could use
4680 the construction C<@{[ (some expression) ]}>, but usually a simple
4681 C<(some expression)> suffices.
4683 Because C<scalar> is unary operator, if you accidentally use for EXPR a
4684 parenthesized list, this behaves as a scalar comma expression, evaluating
4685 all but the last element in void context and returning the final element
4686 evaluated in scalar context. This is seldom what you want.
4688 The following single statement:
4690 print uc(scalar(&foo,$bar)),$baz;
4692 is the moral equivalent of these two:
4695 print(uc($bar),$baz);
4697 See L<perlop> for more details on unary operators and the comma operator.
4699 =item seek FILEHANDLE,POSITION,WHENCE
4700 X<seek> X<fseek> X<filehandle, position>
4702 Sets FILEHANDLE's position, just like the C<fseek> call of C<stdio>.
4703 FILEHANDLE may be an expression whose value gives the name of the
4704 filehandle. The values for WHENCE are C<0> to set the new position
4705 I<in bytes> to POSITION, C<1> to set it to the current position plus
4706 POSITION, and C<2> to set it to EOF plus POSITION (typically
4707 negative). For WHENCE you may use the constants C<SEEK_SET>,
4708 C<SEEK_CUR>, and C<SEEK_END> (start of the file, current position, end
4709 of the file) from the Fcntl module. Returns C<1> upon success, C<0>
4712 Note the I<in bytes>: even if the filehandle has been set to
4713 operate on characters (for example by using the C<:utf8> open
4714 layer), tell() will return byte offsets, not character offsets
4715 (because implementing that would render seek() and tell() rather slow).
4717 If you want to position file for C<sysread> or C<syswrite>, don't use
4718 C<seek>--buffering makes its effect on the file's system position
4719 unpredictable and non-portable. Use C<sysseek> instead.
4721 Due to the rules and rigors of ANSI C, on some systems you have to do a
4722 seek whenever you switch between reading and writing. Amongst other
4723 things, this may have the effect of calling stdio's clearerr(3).
4724 A WHENCE of C<1> (C<SEEK_CUR>) is useful for not moving the file position:
4728 This is also useful for applications emulating C<tail -f>. Once you hit
4729 EOF on your read, and then sleep for a while, you might have to stick in a
4730 seek() to reset things. The C<seek> doesn't change the current position,
4731 but it I<does> clear the end-of-file condition on the handle, so that the
4732 next C<< <FILE> >> makes Perl try again to read something. We hope.
4734 If that doesn't work (some IO implementations are particularly
4735 cantankerous), then you may need something more like this:
4738 for ($curpos = tell(FILE); $_ = <FILE>;
4739 $curpos = tell(FILE)) {
4740 # search for some stuff and put it into files
4742 sleep($for_a_while);
4743 seek(FILE, $curpos, 0);
4746 =item seekdir DIRHANDLE,POS
4749 Sets the current position for the C<readdir> routine on DIRHANDLE. POS
4750 must be a value returned by C<telldir>. C<seekdir> also has the same caveats
4751 about possible directory compaction as the corresponding system library
4754 =item select FILEHANDLE
4755 X<select> X<filehandle, default>
4759 Returns the currently selected filehandle. Sets the current default
4760 filehandle for output, if FILEHANDLE is supplied. This has two
4761 effects: first, a C<write> or a C<print> without a filehandle will
4762 default to this FILEHANDLE. Second, references to variables related to
4763 output will refer to this output channel. For example, if you have to
4764 set the top of form format for more than one output channel, you might
4772 FILEHANDLE may be an expression whose value gives the name of the
4773 actual filehandle. Thus:
4775 $oldfh = select(STDERR); $| = 1; select($oldfh);
4777 Some programmers may prefer to think of filehandles as objects with
4778 methods, preferring to write the last example as:
4781 STDERR->autoflush(1);
4783 =item select RBITS,WBITS,EBITS,TIMEOUT
4786 This calls the select(2) system call with the bit masks specified, which
4787 can be constructed using C<fileno> and C<vec>, along these lines:
4789 $rin = $win = $ein = '';
4790 vec($rin,fileno(STDIN),1) = 1;
4791 vec($win,fileno(STDOUT),1) = 1;
4794 If you want to select on many filehandles you might wish to write a
4798 my(@fhlist) = split(' ',$_[0]);
4801 vec($bits,fileno($_),1) = 1;
4805 $rin = fhbits('STDIN TTY SOCK');
4809 ($nfound,$timeleft) =
4810 select($rout=$rin, $wout=$win, $eout=$ein, $timeout);
4812 or to block until something becomes ready just do this
4814 $nfound = select($rout=$rin, $wout=$win, $eout=$ein, undef);
4816 Most systems do not bother to return anything useful in $timeleft, so
4817 calling select() in scalar context just returns $nfound.
4819 Any of the bit masks can also be undef. The timeout, if specified, is
4820 in seconds, which may be fractional. Note: not all implementations are
4821 capable of returning the $timeleft. If not, they always return
4822 $timeleft equal to the supplied $timeout.
4824 You can effect a sleep of 250 milliseconds this way:
4826 select(undef, undef, undef, 0.25);
4828 Note that whether C<select> gets restarted after signals (say, SIGALRM)
4829 is implementation-dependent. See also L<perlport> for notes on the
4830 portability of C<select>.
4832 On error, C<select> behaves like the select(2) system call : it returns
4835 Note: on some Unixes, the select(2) system call may report a socket file
4836 descriptor as "ready for reading", when actually no data is available,
4837 thus a subsequent read blocks. It can be avoided using always the
4838 O_NONBLOCK flag on the socket. See select(2) and fcntl(2) for further
4841 B<WARNING>: One should not attempt to mix buffered I/O (like C<read>
4842 or <FH>) with C<select>, except as permitted by POSIX, and even
4843 then only on POSIX systems. You have to use C<sysread> instead.
4845 =item semctl ID,SEMNUM,CMD,ARG
4848 Calls the System V IPC function C<semctl>. You'll probably have to say
4852 first to get the correct constant definitions. If CMD is IPC_STAT or
4853 GETALL, then ARG must be a variable that will hold the returned
4854 semid_ds structure or semaphore value array. Returns like C<ioctl>:
4855 the undefined value for error, "C<0 but true>" for zero, or the actual
4856 return value otherwise. The ARG must consist of a vector of native
4857 short integers, which may be created with C<pack("s!",(0)x$nsem)>.
4858 See also L<perlipc/"SysV IPC">, C<IPC::SysV>, C<IPC::Semaphore>
4861 =item semget KEY,NSEMS,FLAGS
4864 Calls the System V IPC function semget. Returns the semaphore id, or
4865 the undefined value if there is an error. See also
4866 L<perlipc/"SysV IPC">, C<IPC::SysV>, C<IPC::SysV::Semaphore>
4869 =item semop KEY,OPSTRING
4872 Calls the System V IPC function semop to perform semaphore operations
4873 such as signalling and waiting. OPSTRING must be a packed array of
4874 semop structures. Each semop structure can be generated with
4875 C<pack("s!3", $semnum, $semop, $semflag)>. The length of OPSTRING
4876 implies the number of semaphore operations. Returns true if
4877 successful, or false if there is an error. As an example, the
4878 following code waits on semaphore $semnum of semaphore id $semid:
4880 $semop = pack("s!3", $semnum, -1, 0);
4881 die "Semaphore trouble: $!\n" unless semop($semid, $semop);
4883 To signal the semaphore, replace C<-1> with C<1>. See also
4884 L<perlipc/"SysV IPC">, C<IPC::SysV>, and C<IPC::SysV::Semaphore>
4887 =item send SOCKET,MSG,FLAGS,TO
4890 =item send SOCKET,MSG,FLAGS
4892 Sends a message on a socket. Attempts to send the scalar MSG to the
4893 SOCKET filehandle. Takes the same flags as the system call of the
4894 same name. On unconnected sockets you must specify a destination to
4895 send TO, in which case it does a C C<sendto>. Returns the number of
4896 characters sent, or the undefined value if there is an error. The C
4897 system call sendmsg(2) is currently unimplemented. See
4898 L<perlipc/"UDP: Message Passing"> for examples.
4900 Note the I<characters>: depending on the status of the socket, either
4901 (8-bit) bytes or characters are sent. By default all sockets operate
4902 on bytes, but for example if the socket has been changed using
4903 binmode() to operate with the C<:utf8> I/O layer (see L</open>, or the
4904 C<open> pragma, L<open>), the I/O will operate on UTF-8 encoded
4905 Unicode characters, not bytes. Similarly for the C<:encoding> pragma:
4906 in that case pretty much any characters can be sent.
4908 =item setpgrp PID,PGRP
4911 Sets the current process group for the specified PID, C<0> for the current
4912 process. Will produce a fatal error if used on a machine that doesn't
4913 implement POSIX setpgid(2) or BSD setpgrp(2). If the arguments are omitted,
4914 it defaults to C<0,0>. Note that the BSD 4.2 version of C<setpgrp> does not
4915 accept any arguments, so only C<setpgrp(0,0)> is portable. See also
4918 =item setpriority WHICH,WHO,PRIORITY
4919 X<setpriority> X<priority> X<nice> X<renice>
4921 Sets the current priority for a process, a process group, or a user.
4922 (See setpriority(2).) Will produce a fatal error if used on a machine
4923 that doesn't implement setpriority(2).
4925 =item setsockopt SOCKET,LEVEL,OPTNAME,OPTVAL
4928 Sets the socket option requested. Returns undefined if there is an
4929 error. OPTVAL may be specified as C<undef> if you don't want to pass an
4937 Shifts the first value of the array off and returns it, shortening the
4938 array by 1 and moving everything down. If there are no elements in the
4939 array, returns the undefined value. If ARRAY is omitted, shifts the
4940 C<@_> array within the lexical scope of subroutines and formats, and the
4941 C<@ARGV> array outside of a subroutine and also within the lexical scopes
4942 established by the C<eval STRING>, C<BEGIN {}>, C<INIT {}>, C<CHECK {}>
4943 and C<END {}> constructs.
4945 See also C<unshift>, C<push>, and C<pop>. C<shift> and C<unshift> do the
4946 same thing to the left end of an array that C<pop> and C<push> do to the
4949 =item shmctl ID,CMD,ARG
4952 Calls the System V IPC function shmctl. You'll probably have to say
4956 first to get the correct constant definitions. If CMD is C<IPC_STAT>,
4957 then ARG must be a variable that will hold the returned C<shmid_ds>
4958 structure. Returns like ioctl: the undefined value for error, "C<0> but
4959 true" for zero, or the actual return value otherwise.
4960 See also L<perlipc/"SysV IPC"> and C<IPC::SysV> documentation.
4962 =item shmget KEY,SIZE,FLAGS
4965 Calls the System V IPC function shmget. Returns the shared memory
4966 segment id, or the undefined value if there is an error.
4967 See also L<perlipc/"SysV IPC"> and C<IPC::SysV> documentation.
4969 =item shmread ID,VAR,POS,SIZE
4973 =item shmwrite ID,STRING,POS,SIZE
4975 Reads or writes the System V shared memory segment ID starting at
4976 position POS for size SIZE by attaching to it, copying in/out, and
4977 detaching from it. When reading, VAR must be a variable that will
4978 hold the data read. When writing, if STRING is too long, only SIZE
4979 bytes are used; if STRING is too short, nulls are written to fill out
4980 SIZE bytes. Return true if successful, or false if there is an error.
4981 shmread() taints the variable. See also L<perlipc/"SysV IPC">,
4982 C<IPC::SysV> documentation, and the C<IPC::Shareable> module from CPAN.
4984 =item shutdown SOCKET,HOW
4987 Shuts down a socket connection in the manner indicated by HOW, which
4988 has the same interpretation as in the system call of the same name.
4990 shutdown(SOCKET, 0); # I/we have stopped reading data
4991 shutdown(SOCKET, 1); # I/we have stopped writing data
4992 shutdown(SOCKET, 2); # I/we have stopped using this socket
4994 This is useful with sockets when you want to tell the other
4995 side you're done writing but not done reading, or vice versa.
4996 It's also a more insistent form of close because it also
4997 disables the file descriptor in any forked copies in other
5001 X<sin> X<sine> X<asin> X<arcsine>
5005 Returns the sine of EXPR (expressed in radians). If EXPR is omitted,
5006 returns sine of C<$_>.
5008 For the inverse sine operation, you may use the C<Math::Trig::asin>
5009 function, or use this relation:
5011 sub asin { atan2($_[0], sqrt(1 - $_[0] * $_[0])) }
5018 Causes the script to sleep for EXPR seconds, or forever if no EXPR.
5019 May be interrupted if the process receives a signal such as C<SIGALRM>.
5020 Returns the number of seconds actually slept. You probably cannot
5021 mix C<alarm> and C<sleep> calls, because C<sleep> is often implemented
5024 On some older systems, it may sleep up to a full second less than what
5025 you requested, depending on how it counts seconds. Most modern systems
5026 always sleep the full amount. They may appear to sleep longer than that,
5027 however, because your process might not be scheduled right away in a
5028 busy multitasking system.
5030 For delays of finer granularity than one second, you may use Perl's
5031 C<syscall> interface to access setitimer(2) if your system supports
5032 it, or else see L</select> above. The Time::HiRes module (from CPAN,
5033 and starting from Perl 5.8 part of the standard distribution) may also
5036 See also the POSIX module's C<pause> function.
5038 =item socket SOCKET,DOMAIN,TYPE,PROTOCOL
5041 Opens a socket of the specified kind and attaches it to filehandle
5042 SOCKET. DOMAIN, TYPE, and PROTOCOL are specified the same as for
5043 the system call of the same name. You should C<use Socket> first
5044 to get the proper definitions imported. See the examples in
5045 L<perlipc/"Sockets: Client/Server Communication">.
5047 On systems that support a close-on-exec flag on files, the flag will
5048 be set for the newly opened file descriptor, as determined by the
5049 value of $^F. See L<perlvar/$^F>.
5051 =item socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL
5054 Creates an unnamed pair of sockets in the specified domain, of the
5055 specified type. DOMAIN, TYPE, and PROTOCOL are specified the same as
5056 for the system call of the same name. If unimplemented, yields a fatal
5057 error. Returns true if successful.
5059 On systems that support a close-on-exec flag on files, the flag will
5060 be set for the newly opened file descriptors, as determined by the value
5061 of $^F. See L<perlvar/$^F>.
5063 Some systems defined C<pipe> in terms of C<socketpair>, in which a call
5064 to C<pipe(Rdr, Wtr)> is essentially:
5067 socketpair(Rdr, Wtr, AF_UNIX, SOCK_STREAM, PF_UNSPEC);
5068 shutdown(Rdr, 1); # no more writing for reader
5069 shutdown(Wtr, 0); # no more reading for writer
5071 See L<perlipc> for an example of socketpair use. Perl 5.8 and later will
5072 emulate socketpair using IP sockets to localhost if your system implements
5073 sockets but not socketpair.
5075 =item sort SUBNAME LIST
5076 X<sort> X<qsort> X<quicksort> X<mergesort>
5078 =item sort BLOCK LIST
5082 In list context, this sorts the LIST and returns the sorted list value.
5083 In scalar context, the behaviour of C<sort()> is undefined.
5085 If SUBNAME or BLOCK is omitted, C<sort>s in standard string comparison
5086 order. If SUBNAME is specified, it gives the name of a subroutine
5087 that returns an integer less than, equal to, or greater than C<0>,
5088 depending on how the elements of the list are to be ordered. (The C<<
5089 <=> >> and C<cmp> operators are extremely useful in such routines.)
5090 SUBNAME may be a scalar variable name (unsubscripted), in which case
5091 the value provides the name of (or a reference to) the actual
5092 subroutine to use. In place of a SUBNAME, you can provide a BLOCK as
5093 an anonymous, in-line sort subroutine.
5095 If the subroutine's prototype is C<($$)>, the elements to be compared
5096 are passed by reference in C<@_>, as for a normal subroutine. This is
5097 slower than unprototyped subroutines, where the elements to be
5098 compared are passed into the subroutine
5099 as the package global variables $a and $b (see example below). Note that
5100 in the latter case, it is usually counter-productive to declare $a and
5103 The values to be compared are always passed by reference and should not
5106 You also cannot exit out of the sort block or subroutine using any of the
5107 loop control operators described in L<perlsyn> or with C<goto>.
5109 When C<use locale> is in effect, C<sort LIST> sorts LIST according to the
5110 current collation locale. See L<perllocale>.
5112 sort() returns aliases into the original list, much as a for loop's index
5113 variable aliases the list elements. That is, modifying an element of a
5114 list returned by sort() (for example, in a C<foreach>, C<map> or C<grep>)
5115 actually modifies the element in the original list. This is usually
5116 something to be avoided when writing clear code.
5118 Perl 5.6 and earlier used a quicksort algorithm to implement sort.
5119 That algorithm was not stable, and I<could> go quadratic. (A I<stable> sort
5120 preserves the input order of elements that compare equal. Although
5121 quicksort's run time is O(NlogN) when averaged over all arrays of
5122 length N, the time can be O(N**2), I<quadratic> behavior, for some
5123 inputs.) In 5.7, the quicksort implementation was replaced with
5124 a stable mergesort algorithm whose worst-case behavior is O(NlogN).
5125 But benchmarks indicated that for some inputs, on some platforms,
5126 the original quicksort was faster. 5.8 has a sort pragma for
5127 limited control of the sort. Its rather blunt control of the
5128 underlying algorithm may not persist into future Perls, but the
5129 ability to characterize the input or output in implementation
5130 independent ways quite probably will. See L<sort>.
5135 @articles = sort @files;
5137 # same thing, but with explicit sort routine
5138 @articles = sort {$a cmp $b} @files;
5140 # now case-insensitively
5141 @articles = sort {uc($a) cmp uc($b)} @files;
5143 # same thing in reversed order
5144 @articles = sort {$b cmp $a} @files;
5146 # sort numerically ascending
5147 @articles = sort {$a <=> $b} @files;
5149 # sort numerically descending
5150 @articles = sort {$b <=> $a} @files;
5152 # this sorts the %age hash by value instead of key
5153 # using an in-line function
5154 @eldest = sort { $age{$b} <=> $age{$a} } keys %age;
5156 # sort using explicit subroutine name
5158 $age{$a} <=> $age{$b}; # presuming numeric
5160 @sortedclass = sort byage @class;
5162 sub backwards { $b cmp $a }
5163 @harry = qw(dog cat x Cain Abel);
5164 @george = qw(gone chased yz Punished Axed);
5166 # prints AbelCaincatdogx
5167 print sort backwards @harry;
5168 # prints xdogcatCainAbel
5169 print sort @george, 'to', @harry;
5170 # prints AbelAxedCainPunishedcatchaseddoggonetoxyz
5172 # inefficiently sort by descending numeric compare using
5173 # the first integer after the first = sign, or the
5174 # whole record case-insensitively otherwise
5177 ($b =~ /=(\d+)/)[0] <=> ($a =~ /=(\d+)/)[0]
5182 # same thing, but much more efficiently;
5183 # we'll build auxiliary indices instead
5187 push @nums, /=(\d+)/;
5192 $nums[$b] <=> $nums[$a]
5194 $caps[$a] cmp $caps[$b]
5198 # same thing, but without any temps
5199 @new = map { $_->[0] }
5200 sort { $b->[1] <=> $a->[1]
5203 } map { [$_, /=(\d+)/, uc($_)] } @old;
5205 # using a prototype allows you to use any comparison subroutine
5206 # as a sort subroutine (including other package's subroutines)
5208 sub backwards ($$) { $_[1] cmp $_[0]; } # $a and $b are not set here
5211 @new = sort other::backwards @old;
5213 # guarantee stability, regardless of algorithm
5215 @new = sort { substr($a, 3, 5) cmp substr($b, 3, 5) } @old;
5217 # force use of mergesort (not portable outside Perl 5.8)
5218 use sort '_mergesort'; # note discouraging _
5219 @new = sort { substr($a, 3, 5) cmp substr($b, 3, 5) } @old;
5221 If you're using strict, you I<must not> declare $a
5222 and $b as lexicals. They are package globals. That means
5223 if you're in the C<main> package and type
5225 @articles = sort {$b <=> $a} @files;
5227 then C<$a> and C<$b> are C<$main::a> and C<$main::b> (or C<$::a> and C<$::b>),
5228 but if you're in the C<FooPack> package, it's the same as typing
5230 @articles = sort {$FooPack::b <=> $FooPack::a} @files;
5232 The comparison function is required to behave. If it returns
5233 inconsistent results (sometimes saying C<$x[1]> is less than C<$x[2]> and
5234 sometimes saying the opposite, for example) the results are not
5237 Because C<< <=> >> returns C<undef> when either operand is C<NaN>
5238 (not-a-number), and because C<sort> will trigger a fatal error unless the
5239 result of a comparison is defined, when sorting with a comparison function
5240 like C<< $a <=> $b >>, be careful about lists that might contain a C<NaN>.
5241 The following example takes advantage of the fact that C<NaN != NaN> to
5242 eliminate any C<NaN>s from the input.
5244 @result = sort { $a <=> $b } grep { $_ == $_ } @input;
5246 =item splice ARRAY,OFFSET,LENGTH,LIST
5249 =item splice ARRAY,OFFSET,LENGTH
5251 =item splice ARRAY,OFFSET
5255 Removes the elements designated by OFFSET and LENGTH from an array, and
5256 replaces them with the elements of LIST, if any. In list context,
5257 returns the elements removed from the array. In scalar context,
5258 returns the last element removed, or C<undef> if no elements are
5259 removed. The array grows or shrinks as necessary.
5260 If OFFSET is negative then it starts that far from the end of the array.
5261 If LENGTH is omitted, removes everything from OFFSET onward.
5262 If LENGTH is negative, removes the elements from OFFSET onward
5263 except for -LENGTH elements at the end of the array.
5264 If both OFFSET and LENGTH are omitted, removes everything. If OFFSET is
5265 past the end of the array, perl issues a warning, and splices at the
5268 The following equivalences hold (assuming C<< $[ == 0 and $#a >= $i >> )
5270 push(@a,$x,$y) splice(@a,@a,0,$x,$y)
5271 pop(@a) splice(@a,-1)
5272 shift(@a) splice(@a,0,1)
5273 unshift(@a,$x,$y) splice(@a,0,0,$x,$y)
5274 $a[$i] = $y splice(@a,$i,1,$y)
5276 Example, assuming array lengths are passed before arrays:
5278 sub aeq { # compare two list values
5279 my(@a) = splice(@_,0,shift);
5280 my(@b) = splice(@_,0,shift);
5281 return 0 unless @a == @b; # same len?
5283 return 0 if pop(@a) ne pop(@b);
5287 if (&aeq($len,@foo[1..$len],0+@bar,@bar)) { ... }
5289 =item split /PATTERN/,EXPR,LIMIT
5292 =item split /PATTERN/,EXPR
5294 =item split /PATTERN/
5298 Splits the string EXPR into a list of strings and returns that list. By
5299 default, empty leading fields are preserved, and empty trailing ones are
5300 deleted. (If all fields are empty, they are considered to be trailing.)
5302 In scalar context, returns the number of fields found and splits into
5303 the C<@_> array. Use of split in scalar context is deprecated, however,
5304 because it clobbers your subroutine arguments.
5306 If EXPR is omitted, splits the C<$_> string. If PATTERN is also omitted,
5307 splits on whitespace (after skipping any leading whitespace). Anything
5308 matching PATTERN is taken to be a delimiter separating the fields. (Note
5309 that the delimiter may be longer than one character.)
5311 If LIMIT is specified and positive, it represents the maximum number
5312 of fields the EXPR will be split into, though the actual number of
5313 fields returned depends on the number of times PATTERN matches within
5314 EXPR. If LIMIT is unspecified or zero, trailing null fields are
5315 stripped (which potential users of C<pop> would do well to remember).
5316 If LIMIT is negative, it is treated as if an arbitrarily large LIMIT
5317 had been specified. Note that splitting an EXPR that evaluates to the
5318 empty string always returns the empty list, regardless of the LIMIT
5321 A pattern matching the null string (not to be confused with
5322 a null pattern C<//>, which is just one member of the set of patterns
5323 matching a null string) will split the value of EXPR into separate
5324 characters at each point it matches that way. For example:
5326 print join(':', split(/ */, 'hi there'));
5328 produces the output 'h:i:t:h:e:r:e'.
5330 As a special case for C<split>, using the empty pattern C<//> specifically
5331 matches only the null string, and is not be confused with the regular use
5332 of C<//> to mean "the last successful pattern match". So, for C<split>,
5335 print join(':', split(//, 'hi there'));
5337 produces the output 'h:i: :t:h:e:r:e'.
5339 Empty leading (or trailing) fields are produced when there are positive
5340 width matches at the beginning (or end) of the string; a zero-width match
5341 at the beginning (or end) of the string does not produce an empty field.
5344 print join(':', split(/(?=\w)/, 'hi there!'));
5346 produces the output 'h:i :t:h:e:r:e!'.
5348 The LIMIT parameter can be used to split a line partially
5350 ($login, $passwd, $remainder) = split(/:/, $_, 3);
5352 When assigning to a list, if LIMIT is omitted, or zero, Perl supplies
5353 a LIMIT one larger than the number of variables in the list, to avoid
5354 unnecessary work. For the list above LIMIT would have been 4 by
5355 default. In time critical applications it behooves you not to split
5356 into more fields than you really need.
5358 If the PATTERN contains parentheses, additional list elements are
5359 created from each matching substring in the delimiter.
5361 split(/([,-])/, "1-10,20", 3);
5363 produces the list value
5365 (1, '-', 10, ',', 20)
5367 If you had the entire header of a normal Unix email message in $header,
5368 you could split it up into fields and their values this way:
5370 $header =~ s/\n\s+/ /g; # fix continuation lines
5371 %hdrs = (UNIX_FROM => split /^(\S*?):\s*/m, $header);
5373 The pattern C</PATTERN/> may be replaced with an expression to specify
5374 patterns that vary at runtime. (To do runtime compilation only once,
5375 use C</$variable/o>.)
5377 As a special case, specifying a PATTERN of space (S<C<' '>>) will split on
5378 white space just as C<split> with no arguments does. Thus, S<C<split(' ')>> can
5379 be used to emulate B<awk>'s default behavior, whereas S<C<split(/ /)>>
5380 will give you as many null initial fields as there are leading spaces.
5381 A C<split> on C</\s+/> is like a S<C<split(' ')>> except that any leading
5382 whitespace produces a null first field. A C<split> with no arguments
5383 really does a S<C<split(' ', $_)>> internally.
5385 A PATTERN of C</^/> is treated as if it were C</^/m>, since it isn't
5390 open(PASSWD, '/etc/passwd');
5393 ($login, $passwd, $uid, $gid,
5394 $gcos, $home, $shell) = split(/:/);
5398 As with regular pattern matching, any capturing parentheses that are not
5399 matched in a C<split()> will be set to C<undef> when returned:
5401 @fields = split /(A)|B/, "1A2B3";
5402 # @fields is (1, 'A', 2, undef, 3)
5404 =item sprintf FORMAT, LIST
5407 Returns a string formatted by the usual C<printf> conventions of the C
5408 library function C<sprintf>. See below for more details
5409 and see L<sprintf(3)> or L<printf(3)> on your system for an explanation of
5410 the general principles.
5414 # Format number with up to 8 leading zeroes
5415 $result = sprintf("%08d", $number);
5417 # Round number to 3 digits after decimal point
5418 $rounded = sprintf("%.3f", $number);
5420 Perl does its own C<sprintf> formatting--it emulates the C
5421 function C<sprintf>, but it doesn't use it (except for floating-point
5422 numbers, and even then only the standard modifiers are allowed). As a
5423 result, any non-standard extensions in your local C<sprintf> are not
5424 available from Perl.
5426 Unlike C<printf>, C<sprintf> does not do what you probably mean when you
5427 pass it an array as your first argument. The array is given scalar context,
5428 and instead of using the 0th element of the array as the format, Perl will
5429 use the count of elements in the array as the format, which is almost never
5432 Perl's C<sprintf> permits the following universally-known conversions:
5435 %c a character with the given number
5437 %d a signed integer, in decimal
5438 %u an unsigned integer, in decimal
5439 %o an unsigned integer, in octal
5440 %x an unsigned integer, in hexadecimal
5441 %e a floating-point number, in scientific notation
5442 %f a floating-point number, in fixed decimal notation
5443 %g a floating-point number, in %e or %f notation
5445 In addition, Perl permits the following widely-supported conversions:
5447 %X like %x, but using upper-case letters
5448 %E like %e, but using an upper-case "E"
5449 %G like %g, but with an upper-case "E" (if applicable)
5450 %b an unsigned integer, in binary
5451 %p a pointer (outputs the Perl value's address in hexadecimal)
5452 %n special: *stores* the number of characters output so far
5453 into the next variable in the parameter list
5455 Finally, for backward (and we do mean "backward") compatibility, Perl
5456 permits these unnecessary but widely-supported conversions:
5459 %D a synonym for %ld
5460 %U a synonym for %lu
5461 %O a synonym for %lo
5464 Note that the number of exponent digits in the scientific notation produced
5465 by C<%e>, C<%E>, C<%g> and C<%G> for numbers with the modulus of the
5466 exponent less than 100 is system-dependent: it may be three or less
5467 (zero-padded as necessary). In other words, 1.23 times ten to the
5468 99th may be either "1.23e99" or "1.23e099".
5470 Between the C<%> and the format letter, you may specify a number of
5471 additional attributes controlling the interpretation of the format.
5472 In order, these are:
5476 =item format parameter index
5478 An explicit format parameter index, such as C<2$>. By default sprintf
5479 will format the next unused argument in the list, but this allows you
5480 to take the arguments out of order, e.g.:
5482 printf '%2$d %1$d', 12, 34; # prints "34 12"
5483 printf '%3$d %d %1$d', 1, 2, 3; # prints "3 1 1"
5488 space prefix positive number with a space
5489 + prefix positive number with a plus sign
5490 - left-justify within the field
5491 0 use zeros, not spaces, to right-justify
5492 # prefix non-zero octal with "0", non-zero hex with "0x",
5493 non-zero binary with "0b"
5497 printf '<% d>', 12; # prints "< 12>"
5498 printf '<%+d>', 12; # prints "<+12>"
5499 printf '<%6s>', 12; # prints "< 12>"
5500 printf '<%-6s>', 12; # prints "<12 >"
5501 printf '<%06s>', 12; # prints "<000012>"
5502 printf '<%#x>', 12; # prints "<0xc>"
5506 The vector flag C<v>, optionally specifying the join string to use.
5507 This flag tells perl to interpret the supplied string as a vector
5508 of integers, one for each character in the string, separated by
5509 a given string (a dot C<.> by default). This can be useful for
5510 displaying ordinal values of characters in arbitrary strings:
5512 printf "version is v%vd\n", $^V; # Perl's version
5514 Put an asterisk C<*> before the C<v> to override the string to
5515 use to separate the numbers:
5517 printf "address is %*vX\n", ":", $addr; # IPv6 address
5518 printf "bits are %0*v8b\n", " ", $bits; # random bitstring
5520 You can also explicitly specify the argument number to use for
5521 the join string using e.g. C<*2$v>:
5523 printf '%*4$vX %*4$vX %*4$vX', @addr[1..3], ":"; # 3 IPv6 addresses
5525 =item (minimum) width
5527 Arguments are usually formatted to be only as wide as required to
5528 display the given value. You can override the width by putting
5529 a number here, or get the width from the next argument (with C<*>)
5530 or from a specified argument (with e.g. C<*2$>):
5532 printf '<%s>', "a"; # prints "<a>"
5533 printf '<%6s>', "a"; # prints "< a>"
5534 printf '<%*s>', 6, "a"; # prints "< a>"
5535 printf '<%*2$s>', "a", 6; # prints "< a>"
5536 printf '<%2s>', "long"; # prints "<long>" (does not truncate)
5538 If a field width obtained through C<*> is negative, it has the same
5539 effect as the C<-> flag: left-justification.
5541 =item precision, or maximum width
5544 You can specify a precision (for numeric conversions) or a maximum
5545 width (for string conversions) by specifying a C<.> followed by a number.
5546 For floating point formats, with the exception of 'g' and 'G', this specifies
5547 the number of decimal places to show (the default being 6), e.g.:
5549 # these examples are subject to system-specific variation
5550 printf '<%f>', 1; # prints "<1.000000>"
5551 printf '<%.1f>', 1; # prints "<1.0>"
5552 printf '<%.0f>', 1; # prints "<1>"
5553 printf '<%e>', 10; # prints "<1.000000e+01>"
5554 printf '<%.1e>', 10; # prints "<1.0e+01>"
5556 For 'g' and 'G', this specifies the maximum number of digits to show,
5557 including prior to the decimal point as well as after it, e.g.:
5559 # these examples are subject to system-specific variation
5560 printf '<%g>', 1; # prints "<1>"
5561 printf '<%.10g>', 1; # prints "<1>"
5562 printf '<%g>', 100; # prints "<100>"
5563 printf '<%.1g>', 100; # prints "<1e+02>"
5564 printf '<%.2g>', 100.01; # prints "<1e+02>"
5565 printf '<%.5g>', 100.01; # prints "<100.01>"
5566 printf '<%.4g>', 100.01; # prints "<100>"
5568 For integer conversions, specifying a precision implies that the
5569 output of the number itself should be zero-padded to this width:
5571 printf '<%.6x>', 1; # prints "<000001>"
5572 printf '<%#.6x>', 1; # prints "<0x000001>"
5573 printf '<%-10.6x>', 1; # prints "<000001 >"
5575 For string conversions, specifying a precision truncates the string
5576 to fit in the specified width:
5578 printf '<%.5s>', "truncated"; # prints "<trunc>"
5579 printf '<%10.5s>', "truncated"; # prints "< trunc>"
5581 You can also get the precision from the next argument using C<.*>:
5583 printf '<%.6x>', 1; # prints "<000001>"
5584 printf '<%.*x>', 6, 1; # prints "<000001>"
5586 You cannot currently get the precision from a specified number,
5587 but it is intended that this will be possible in the future using
5590 printf '<%.*2$x>', 1, 6; # INVALID, but in future will print "<000001>"
5594 For numeric conversions, you can specify the size to interpret the
5595 number as using C<l>, C<h>, C<V>, C<q>, C<L>, or C<ll>. For integer
5596 conversions (C<d u o x X b i D U O>), numbers are usually assumed to be
5597 whatever the default integer size is on your platform (usually 32 or 64
5598 bits), but you can override this to use instead one of the standard C types,
5599 as supported by the compiler used to build Perl:
5601 l interpret integer as C type "long" or "unsigned long"
5602 h interpret integer as C type "short" or "unsigned short"
5603 q, L or ll interpret integer as C type "long long", "unsigned long long".
5604 or "quads" (typically 64-bit integers)
5606 The last will produce errors if Perl does not understand "quads" in your
5607 installation. (This requires that either the platform natively supports quads
5608 or Perl was specifically compiled to support quads.) You can find out
5609 whether your Perl supports quads via L<Config>:
5612 ($Config{use64bitint} eq 'define' || $Config{longsize} >= 8) &&
5615 For floating point conversions (C<e f g E F G>), numbers are usually assumed
5616 to be the default floating point size on your platform (double or long double),
5617 but you can force 'long double' with C<q>, C<L>, or C<ll> if your
5618 platform supports them. You can find out whether your Perl supports long
5619 doubles via L<Config>:
5622 $Config{d_longdbl} eq 'define' && print "long doubles\n";
5624 You can find out whether Perl considers 'long double' to be the default
5625 floating point size to use on your platform via L<Config>:
5628 ($Config{uselongdouble} eq 'define') &&
5629 print "long doubles by default\n";
5631 It can also be the case that long doubles and doubles are the same thing:
5634 ($Config{doublesize} == $Config{longdblsize}) &&
5635 print "doubles are long doubles\n";
5637 The size specifier C<V> has no effect for Perl code, but it is supported
5638 for compatibility with XS code; it means 'use the standard size for
5639 a Perl integer (or floating-point number)', which is already the
5640 default for Perl code.
5642 =item order of arguments
5644 Normally, sprintf takes the next unused argument as the value to
5645 format for each format specification. If the format specification
5646 uses C<*> to require additional arguments, these are consumed from
5647 the argument list in the order in which they appear in the format
5648 specification I<before> the value to format. Where an argument is
5649 specified using an explicit index, this does not affect the normal
5650 order for the arguments (even when the explicitly specified index
5651 would have been the next argument in any case).
5655 printf '<%*.*s>', $a, $b, $c;
5657 would use C<$a> for the width, C<$b> for the precision and C<$c>
5658 as the value to format, while:
5660 print '<%*1$.*s>', $a, $b;
5662 would use C<$a> for the width and the precision, and C<$b> as the
5665 Here are some more examples - beware that when using an explicit
5666 index, the C<$> may need to be escaped:
5668 printf "%2\$d %d\n", 12, 34; # will print "34 12\n"
5669 printf "%2\$d %d %d\n", 12, 34; # will print "34 12 34\n"
5670 printf "%3\$d %d %d\n", 12, 34, 56; # will print "56 12 34\n"
5671 printf "%2\$*3\$d %d\n", 12, 34, 3; # will print " 34 12\n"
5675 If C<use locale> is in effect, the character used for the decimal
5676 point in formatted real numbers is affected by the LC_NUMERIC locale.
5680 X<sqrt> X<root> X<square root>
5684 Return the square root of EXPR. If EXPR is omitted, returns square
5685 root of C<$_>. Only works on non-negative operands, unless you've
5686 loaded the standard Math::Complex module.
5689 print sqrt(-2); # prints 1.4142135623731i
5692 X<srand> X<seed> X<randseed>
5696 Sets the random number seed for the C<rand> operator.
5698 The point of the function is to "seed" the C<rand> function so that
5699 C<rand> can produce a different sequence each time you run your
5702 If srand() is not called explicitly, it is called implicitly at the
5703 first use of the C<rand> operator. However, this was not the case in
5704 versions of Perl before 5.004, so if your script will run under older
5705 Perl versions, it should call C<srand>.
5707 Most programs won't even call srand() at all, except those that
5708 need a cryptographically-strong starting point rather than the
5709 generally acceptable default, which is based on time of day,
5710 process ID, and memory allocation, or the F</dev/urandom> device,
5713 You can call srand($seed) with the same $seed to reproduce the
5714 I<same> sequence from rand(), but this is usually reserved for
5715 generating predictable results for testing or debugging.
5716 Otherwise, don't call srand() more than once in your program.
5718 Do B<not> call srand() (i.e. without an argument) more than once in
5719 a script. The internal state of the random number generator should
5720 contain more entropy than can be provided by any seed, so calling
5721 srand() again actually I<loses> randomness.
5723 Most implementations of C<srand> take an integer and will silently
5724 truncate decimal numbers. This means C<srand(42)> will usually
5725 produce the same results as C<srand(42.1)>. To be safe, always pass
5726 C<srand> an integer.
5728 In versions of Perl prior to 5.004 the default seed was just the
5729 current C<time>. This isn't a particularly good seed, so many old
5730 programs supply their own seed value (often C<time ^ $$> or C<time ^
5731 ($$ + ($$ << 15))>), but that isn't necessary any more.
5733 For cryptographic purposes, however, you need something much more random
5734 than the default seed. Checksumming the compressed output of one or more
5735 rapidly changing operating system status programs is the usual method. For
5738 srand (time ^ $$ ^ unpack "%L*", `ps axww | gzip`);
5740 If you're particularly concerned with this, see the C<Math::TrulyRandom>
5743 Frequently called programs (like CGI scripts) that simply use
5747 for a seed can fall prey to the mathematical property that
5751 one-third of the time. So don't do that.
5753 =item stat FILEHANDLE
5754 X<stat> X<file, status>
5760 Returns a 13-element list giving the status info for a file, either
5761 the file opened via FILEHANDLE, or named by EXPR. If EXPR is omitted,
5762 it stats C<$_>. Returns a null list if the stat fails. Typically used
5765 ($dev,$ino,$mode,$nlink,$uid,$gid,$rdev,$size,
5766 $atime,$mtime,$ctime,$blksize,$blocks)
5769 Not all fields are supported on all filesystem types. Here are the
5770 meanings of the fields:
5772 0 dev device number of filesystem
5774 2 mode file mode (type and permissions)
5775 3 nlink number of (hard) links to the file
5776 4 uid numeric user ID of file's owner
5777 5 gid numeric group ID of file's owner
5778 6 rdev the device identifier (special files only)
5779 7 size total size of file, in bytes
5780 8 atime last access time in seconds since the epoch
5781 9 mtime last modify time in seconds since the epoch
5782 10 ctime inode change time in seconds since the epoch (*)
5783 11 blksize preferred block size for file system I/O
5784 12 blocks actual number of blocks allocated
5786 (The epoch was at 00:00 January 1, 1970 GMT.)
5788 (*) Not all fields are supported on all filesystem types. Notably, the
5789 ctime field is non-portable. In particular, you cannot expect it to be a
5790 "creation time", see L<perlport/"Files and Filesystems"> for details.
5792 If C<stat> is passed the special filehandle consisting of an underline, no
5793 stat is done, but the current contents of the stat structure from the
5794 last C<stat>, C<lstat>, or filetest are returned. Example:
5796 if (-x $file && (($d) = stat(_)) && $d < 0) {
5797 print "$file is executable NFS file\n";
5800 (This works on machines only for which the device number is negative
5803 Because the mode contains both the file type and its permissions, you
5804 should mask off the file type portion and (s)printf using a C<"%o">
5805 if you want to see the real permissions.
5807 $mode = (stat($filename))[2];
5808 printf "Permissions are %04o\n", $mode & 07777;
5810 In scalar context, C<stat> returns a boolean value indicating success
5811 or failure, and, if successful, sets the information associated with
5812 the special filehandle C<_>.
5814 The File::stat module provides a convenient, by-name access mechanism:
5817 $sb = stat($filename);
5818 printf "File is %s, size is %s, perm %04o, mtime %s\n",
5819 $filename, $sb->size, $sb->mode & 07777,
5820 scalar localtime $sb->mtime;
5822 You can import symbolic mode constants (C<S_IF*>) and functions
5823 (C<S_IS*>) from the Fcntl module:
5827 $mode = (stat($filename))[2];
5829 $user_rwx = ($mode & S_IRWXU) >> 6;
5830 $group_read = ($mode & S_IRGRP) >> 3;
5831 $other_execute = $mode & S_IXOTH;
5833 printf "Permissions are %04o\n", S_IMODE($mode), "\n";
5835 $is_setuid = $mode & S_ISUID;
5836 $is_setgid = S_ISDIR($mode);
5838 You could write the last two using the C<-u> and C<-d> operators.
5839 The commonly available C<S_IF*> constants are
5841 # Permissions: read, write, execute, for user, group, others.
5843 S_IRWXU S_IRUSR S_IWUSR S_IXUSR
5844 S_IRWXG S_IRGRP S_IWGRP S_IXGRP
5845 S_IRWXO S_IROTH S_IWOTH S_IXOTH
5847 # Setuid/Setgid/Stickiness/SaveText.
5848 # Note that the exact meaning of these is system dependent.
5850 S_ISUID S_ISGID S_ISVTX S_ISTXT
5852 # File types. Not necessarily all are available on your system.
5854 S_IFREG S_IFDIR S_IFLNK S_IFBLK S_ISCHR S_IFIFO S_IFSOCK S_IFWHT S_ENFMT
5856 # The following are compatibility aliases for S_IRUSR, S_IWUSR, S_IXUSR.
5858 S_IREAD S_IWRITE S_IEXEC
5860 and the C<S_IF*> functions are
5862 S_IMODE($mode) the part of $mode containing the permission bits
5863 and the setuid/setgid/sticky bits
5865 S_IFMT($mode) the part of $mode containing the file type
5866 which can be bit-anded with e.g. S_IFREG
5867 or with the following functions
5869 # The operators -f, -d, -l, -b, -c, -p, and -S.
5871 S_ISREG($mode) S_ISDIR($mode) S_ISLNK($mode)
5872 S_ISBLK($mode) S_ISCHR($mode) S_ISFIFO($mode) S_ISSOCK($mode)
5874 # No direct -X operator counterpart, but for the first one
5875 # the -g operator is often equivalent. The ENFMT stands for
5876 # record flocking enforcement, a platform-dependent feature.
5878 S_ISENFMT($mode) S_ISWHT($mode)
5880 See your native chmod(2) and stat(2) documentation for more details
5881 about the C<S_*> constants. To get status info for a symbolic link
5882 instead of the target file behind the link, use the C<lstat> function.
5889 Takes extra time to study SCALAR (C<$_> if unspecified) in anticipation of
5890 doing many pattern matches on the string before it is next modified.
5891 This may or may not save time, depending on the nature and number of
5892 patterns you are searching on, and on the distribution of character
5893 frequencies in the string to be searched--you probably want to compare
5894 run times with and without it to see which runs faster. Those loops
5895 that scan for many short constant strings (including the constant
5896 parts of more complex patterns) will benefit most. You may have only
5897 one C<study> active at a time--if you study a different scalar the first
5898 is "unstudied". (The way C<study> works is this: a linked list of every
5899 character in the string to be searched is made, so we know, for
5900 example, where all the C<'k'> characters are. From each search string,
5901 the rarest character is selected, based on some static frequency tables
5902 constructed from some C programs and English text. Only those places
5903 that contain this "rarest" character are examined.)
5905 For example, here is a loop that inserts index producing entries
5906 before any line containing a certain pattern:
5910 print ".IX foo\n" if /\bfoo\b/;
5911 print ".IX bar\n" if /\bbar\b/;
5912 print ".IX blurfl\n" if /\bblurfl\b/;
5917 In searching for C</\bfoo\b/>, only those locations in C<$_> that contain C<f>
5918 will be looked at, because C<f> is rarer than C<o>. In general, this is
5919 a big win except in pathological cases. The only question is whether
5920 it saves you more time than it took to build the linked list in the
5923 Note that if you have to look for strings that you don't know till
5924 runtime, you can build an entire loop as a string and C<eval> that to
5925 avoid recompiling all your patterns all the time. Together with
5926 undefining C<$/> to input entire files as one record, this can be very
5927 fast, often faster than specialized programs like fgrep(1). The following
5928 scans a list of files (C<@files>) for a list of words (C<@words>), and prints
5929 out the names of those files that contain a match:
5931 $search = 'while (<>) { study;';
5932 foreach $word (@words) {
5933 $search .= "++\$seen{\$ARGV} if /\\b$word\\b/;\n";
5938 eval $search; # this screams
5939 $/ = "\n"; # put back to normal input delimiter
5940 foreach $file (sort keys(%seen)) {
5944 =item sub NAME BLOCK
5947 =item sub NAME (PROTO) BLOCK
5949 =item sub NAME : ATTRS BLOCK
5951 =item sub NAME (PROTO) : ATTRS BLOCK
5953 This is subroutine definition, not a real function I<per se>.
5954 Without a BLOCK it's just a forward declaration. Without a NAME,
5955 it's an anonymous function declaration, and does actually return
5956 a value: the CODE ref of the closure you just created.
5958 See L<perlsub> and L<perlref> for details about subroutines and
5959 references, and L<attributes> and L<Attribute::Handlers> for more
5960 information about attributes.
5962 =item substr EXPR,OFFSET,LENGTH,REPLACEMENT
5963 X<substr> X<substring> X<mid> X<left> X<right>
5965 =item substr EXPR,OFFSET,LENGTH
5967 =item substr EXPR,OFFSET
5969 Extracts a substring out of EXPR and returns it. First character is at
5970 offset C<0>, or whatever you've set C<$[> to (but don't do that).
5971 If OFFSET is negative (or more precisely, less than C<$[>), starts
5972 that far from the end of the string. If LENGTH is omitted, returns
5973 everything to the end of the string. If LENGTH is negative, leaves that
5974 many characters off the end of the string.
5976 You can use the substr() function as an lvalue, in which case EXPR
5977 must itself be an lvalue. If you assign something shorter than LENGTH,
5978 the string will shrink, and if you assign something longer than LENGTH,
5979 the string will grow to accommodate it. To keep the string the same
5980 length you may need to pad or chop your value using C<sprintf>.
5982 If OFFSET and LENGTH specify a substring that is partly outside the
5983 string, only the part within the string is returned. If the substring
5984 is beyond either end of the string, substr() returns the undefined
5985 value and produces a warning. When used as an lvalue, specifying a
5986 substring that is entirely outside the string is a fatal error.
5987 Here's an example showing the behavior for boundary cases:
5990 substr($name, 4) = 'dy'; # $name is now 'freddy'
5991 my $null = substr $name, 6, 2; # returns '' (no warning)
5992 my $oops = substr $name, 7; # returns undef, with warning
5993 substr($name, 7) = 'gap'; # fatal error
5995 An alternative to using substr() as an lvalue is to specify the
5996 replacement string as the 4th argument. This allows you to replace
5997 parts of the EXPR and return what was there before in one operation,
5998 just as you can with splice().
6000 Note that the lvalue returned by the 3-arg version of substr() acts as
6001 a 'magic bullet'; each time it is assigned to, it remembers which part
6002 of the original string is being modified; for example:
6005 for (substr($x,1,2)) {
6006 $_ = 'a'; print $x,"\n"; # prints 1a4
6007 $_ = 'xyz'; print $x,"\n"; # prints 1xyz4
6009 $_ = 'pq'; print $x,"\n"; # prints 5pq9
6013 Prior to Perl version 5.9.1, the result of using an lvalue multiple times was
6016 =item symlink OLDFILE,NEWFILE
6017 X<symlink> X<link> X<symbolic link> X<link, symbolic>
6019 Creates a new filename symbolically linked to the old filename.
6020 Returns C<1> for success, C<0> otherwise. On systems that don't support
6021 symbolic links, produces a fatal error at run time. To check for that,
6024 $symlink_exists = eval { symlink("",""); 1 };
6026 =item syscall NUMBER, LIST
6027 X<syscall> X<system call>
6029 Calls the system call specified as the first element of the list,
6030 passing the remaining elements as arguments to the system call. If
6031 unimplemented, produces a fatal error. The arguments are interpreted
6032 as follows: if a given argument is numeric, the argument is passed as
6033 an int. If not, the pointer to the string value is passed. You are
6034 responsible to make sure a string is pre-extended long enough to
6035 receive any result that might be written into a string. You can't use a
6036 string literal (or other read-only string) as an argument to C<syscall>
6037 because Perl has to assume that any string pointer might be written
6039 integer arguments are not literals and have never been interpreted in a
6040 numeric context, you may need to add C<0> to them to force them to look
6041 like numbers. This emulates the C<syswrite> function (or vice versa):
6043 require 'syscall.ph'; # may need to run h2ph
6045 syscall(&SYS_write, fileno(STDOUT), $s, length $s);
6047 Note that Perl supports passing of up to only 14 arguments to your system call,
6048 which in practice should usually suffice.
6050 Syscall returns whatever value returned by the system call it calls.
6051 If the system call fails, C<syscall> returns C<-1> and sets C<$!> (errno).
6052 Note that some system calls can legitimately return C<-1>. The proper
6053 way to handle such calls is to assign C<$!=0;> before the call and
6054 check the value of C<$!> if syscall returns C<-1>.
6056 There's a problem with C<syscall(&SYS_pipe)>: it returns the file
6057 number of the read end of the pipe it creates. There is no way
6058 to retrieve the file number of the other end. You can avoid this
6059 problem by using C<pipe> instead.
6061 =item sysopen FILEHANDLE,FILENAME,MODE
6064 =item sysopen FILEHANDLE,FILENAME,MODE,PERMS
6066 Opens the file whose filename is given by FILENAME, and associates it
6067 with FILEHANDLE. If FILEHANDLE is an expression, its value is used as
6068 the name of the real filehandle wanted. This function calls the
6069 underlying operating system's C<open> function with the parameters
6070 FILENAME, MODE, PERMS.
6072 The possible values and flag bits of the MODE parameter are
6073 system-dependent; they are available via the standard module C<Fcntl>.
6074 See the documentation of your operating system's C<open> to see which
6075 values and flag bits are available. You may combine several flags
6076 using the C<|>-operator.
6078 Some of the most common values are C<O_RDONLY> for opening the file in
6079 read-only mode, C<O_WRONLY> for opening the file in write-only mode,
6080 and C<O_RDWR> for opening the file in read-write mode.
6081 X<O_RDONLY> X<O_RDWR> X<O_WRONLY>
6083 For historical reasons, some values work on almost every system
6084 supported by perl: zero means read-only, one means write-only, and two
6085 means read/write. We know that these values do I<not> work under
6086 OS/390 & VM/ESA Unix and on the Macintosh; you probably don't want to
6087 use them in new code.
6089 If the file named by FILENAME does not exist and the C<open> call creates
6090 it (typically because MODE includes the C<O_CREAT> flag), then the value of
6091 PERMS specifies the permissions of the newly created file. If you omit
6092 the PERMS argument to C<sysopen>, Perl uses the octal value C<0666>.
6093 These permission values need to be in octal, and are modified by your
6094 process's current C<umask>.
6097 In many systems the C<O_EXCL> flag is available for opening files in
6098 exclusive mode. This is B<not> locking: exclusiveness means here that
6099 if the file already exists, sysopen() fails. C<O_EXCL> may not work
6100 on network filesystems, and has no effect unless the C<O_CREAT> flag
6101 is set as well. Setting C<O_CREAT|O_EXCL> prevents the file from
6102 being opened if it is a symbolic link. It does not protect against
6103 symbolic links in the file's path.
6106 Sometimes you may want to truncate an already-existing file. This
6107 can be done using the C<O_TRUNC> flag. The behavior of
6108 C<O_TRUNC> with C<O_RDONLY> is undefined.
6111 You should seldom if ever use C<0644> as argument to C<sysopen>, because
6112 that takes away the user's option to have a more permissive umask.
6113 Better to omit it. See the perlfunc(1) entry on C<umask> for more
6116 Note that C<sysopen> depends on the fdopen() C library function.
6117 On many UNIX systems, fdopen() is known to fail when file descriptors
6118 exceed a certain value, typically 255. If you need more file
6119 descriptors than that, consider rebuilding Perl to use the C<sfio>
6120 library, or perhaps using the POSIX::open() function.
6122 See L<perlopentut> for a kinder, gentler explanation of opening files.
6124 =item sysread FILEHANDLE,SCALAR,LENGTH,OFFSET
6127 =item sysread FILEHANDLE,SCALAR,LENGTH
6129 Attempts to read LENGTH bytes of data into variable SCALAR from the
6130 specified FILEHANDLE, using the system call read(2). It bypasses
6131 buffered IO, so mixing this with other kinds of reads, C<print>,
6132 C<write>, C<seek>, C<tell>, or C<eof> can cause confusion because the
6133 perlio or stdio layers usually buffers data. Returns the number of
6134 bytes actually read, C<0> at end of file, or undef if there was an
6135 error (in the latter case C<$!> is also set). SCALAR will be grown or
6136 shrunk so that the last byte actually read is the last byte of the
6137 scalar after the read.
6139 An OFFSET may be specified to place the read data at some place in the
6140 string other than the beginning. A negative OFFSET specifies
6141 placement at that many characters counting backwards from the end of
6142 the string. A positive OFFSET greater than the length of SCALAR
6143 results in the string being padded to the required size with C<"\0">
6144 bytes before the result of the read is appended.
6146 There is no syseof() function, which is ok, since eof() doesn't work
6147 very well on device files (like ttys) anyway. Use sysread() and check
6148 for a return value for 0 to decide whether you're done.
6150 Note that if the filehandle has been marked as C<:utf8> Unicode
6151 characters are read instead of bytes (the LENGTH, OFFSET, and the
6152 return value of sysread() are in Unicode characters).
6153 The C<:encoding(...)> layer implicitly introduces the C<:utf8> layer.
6154 See L</binmode>, L</open>, and the C<open> pragma, L<open>.
6156 =item sysseek FILEHANDLE,POSITION,WHENCE
6159 Sets FILEHANDLE's system position in bytes using the system call
6160 lseek(2). FILEHANDLE may be an expression whose value gives the name
6161 of the filehandle. The values for WHENCE are C<0> to set the new
6162 position to POSITION, C<1> to set the it to the current position plus
6163 POSITION, and C<2> to set it to EOF plus POSITION (typically
6166 Note the I<in bytes>: even if the filehandle has been set to operate
6167 on characters (for example by using the C<:utf8> I/O layer), tell()
6168 will return byte offsets, not character offsets (because implementing
6169 that would render sysseek() very slow).
6171 sysseek() bypasses normal buffered IO, so mixing this with reads (other
6172 than C<sysread>, for example C<< <> >> or read()) C<print>, C<write>,
6173 C<seek>, C<tell>, or C<eof> may cause confusion.
6175 For WHENCE, you may also use the constants C<SEEK_SET>, C<SEEK_CUR>,
6176 and C<SEEK_END> (start of the file, current position, end of the file)
6177 from the Fcntl module. Use of the constants is also more portable
6178 than relying on 0, 1, and 2. For example to define a "systell" function:
6180 use Fcntl 'SEEK_CUR';
6181 sub systell { sysseek($_[0], 0, SEEK_CUR) }
6183 Returns the new position, or the undefined value on failure. A position
6184 of zero is returned as the string C<"0 but true">; thus C<sysseek> returns
6185 true on success and false on failure, yet you can still easily determine
6191 =item system PROGRAM LIST
6193 Does exactly the same thing as C<exec LIST>, except that a fork is
6194 done first, and the parent process waits for the child process to
6195 complete. Note that argument processing varies depending on the
6196 number of arguments. If there is more than one argument in LIST,
6197 or if LIST is an array with more than one value, starts the program
6198 given by the first element of the list with arguments given by the
6199 rest of the list. If there is only one scalar argument, the argument
6200 is checked for shell metacharacters, and if there are any, the
6201 entire argument is passed to the system's command shell for parsing
6202 (this is C</bin/sh -c> on Unix platforms, but varies on other
6203 platforms). If there are no shell metacharacters in the argument,
6204 it is split into words and passed directly to C<execvp>, which is
6207 Beginning with v5.6.0, Perl will attempt to flush all files opened for
6208 output before any operation that may do a fork, but this may not be
6209 supported on some platforms (see L<perlport>). To be safe, you may need
6210 to set C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method
6211 of C<IO::Handle> on any open handles.
6213 The return value is the exit status of the program as returned by the
6214 C<wait> call. To get the actual exit value, shift right by eight (see
6215 below). See also L</exec>. This is I<not> what you want to use to capture
6216 the output from a command, for that you should use merely backticks or
6217 C<qx//>, as described in L<perlop/"`STRING`">. Return value of -1
6218 indicates a failure to start the program or an error of the wait(2) system
6219 call (inspect $! for the reason).
6221 Like C<exec>, C<system> allows you to lie to a program about its name if
6222 you use the C<system PROGRAM LIST> syntax. Again, see L</exec>.
6224 Since C<SIGINT> and C<SIGQUIT> are ignored during the execution of
6225 C<system>, if you expect your program to terminate on receipt of these
6226 signals you will need to arrange to do so yourself based on the return
6229 @args = ("command", "arg1", "arg2");
6231 or die "system @args failed: $?"
6233 You can check all the failure possibilities by inspecting
6237 print "failed to execute: $!\n";
6240 printf "child died with signal %d, %s coredump\n",
6241 ($? & 127), ($? & 128) ? 'with' : 'without';
6244 printf "child exited with value %d\n", $? >> 8;
6247 Alternatively you might inspect the value of C<${^CHILD_ERROR_NATIVE}>
6248 with the W*() calls of the POSIX extension.
6250 When the arguments get executed via the system shell, results
6251 and return codes will be subject to its quirks and capabilities.
6252 See L<perlop/"`STRING`"> and L</exec> for details.
6254 =item syswrite FILEHANDLE,SCALAR,LENGTH,OFFSET
6257 =item syswrite FILEHANDLE,SCALAR,LENGTH
6259 =item syswrite FILEHANDLE,SCALAR
6261 Attempts to write LENGTH bytes of data from variable SCALAR to the
6262 specified FILEHANDLE, using the system call write(2). If LENGTH is
6263 not specified, writes whole SCALAR. It bypasses buffered IO, so
6264 mixing this with reads (other than C<sysread())>, C<print>, C<write>,
6265 C<seek>, C<tell>, or C<eof> may cause confusion because the perlio and
6266 stdio layers usually buffers data. Returns the number of bytes
6267 actually written, or C<undef> if there was an error (in this case the
6268 errno variable C<$!> is also set). If the LENGTH is greater than the
6269 available data in the SCALAR after the OFFSET, only as much data as is
6270 available will be written.
6272 An OFFSET may be specified to write the data from some part of the
6273 string other than the beginning. A negative OFFSET specifies writing
6274 that many characters counting backwards from the end of the string.
6275 In the case the SCALAR is empty you can use OFFSET but only zero offset.
6277 Note that if the filehandle has been marked as C<:utf8>, Unicode
6278 characters are written instead of bytes (the LENGTH, OFFSET, and the
6279 return value of syswrite() are in UTF-8 encoded Unicode characters).
6280 The C<:encoding(...)> layer implicitly introduces the C<:utf8> layer.
6281 See L</binmode>, L</open>, and the C<open> pragma, L<open>.
6283 =item tell FILEHANDLE
6288 Returns the current position I<in bytes> for FILEHANDLE, or -1 on
6289 error. FILEHANDLE may be an expression whose value gives the name of
6290 the actual filehandle. If FILEHANDLE is omitted, assumes the file
6293 Note the I<in bytes>: even if the filehandle has been set to
6294 operate on characters (for example by using the C<:utf8> open
6295 layer), tell() will return byte offsets, not character offsets
6296 (because that would render seek() and tell() rather slow).
6298 The return value of tell() for the standard streams like the STDIN
6299 depends on the operating system: it may return -1 or something else.
6300 tell() on pipes, fifos, and sockets usually returns -1.
6302 There is no C<systell> function. Use C<sysseek(FH, 0, 1)> for that.
6304 Do not use tell() (or other buffered I/O operations) on a file handle
6305 that has been manipulated by sysread(), syswrite() or sysseek().
6306 Those functions ignore the buffering, while tell() does not.
6308 =item telldir DIRHANDLE
6311 Returns the current position of the C<readdir> routines on DIRHANDLE.
6312 Value may be given to C<seekdir> to access a particular location in a
6313 directory. C<telldir> has the same caveats about possible directory
6314 compaction as the corresponding system library routine.
6316 =item tie VARIABLE,CLASSNAME,LIST
6319 This function binds a variable to a package class that will provide the
6320 implementation for the variable. VARIABLE is the name of the variable
6321 to be enchanted. CLASSNAME is the name of a class implementing objects
6322 of correct type. Any additional arguments are passed to the C<new>
6323 method of the class (meaning C<TIESCALAR>, C<TIEHANDLE>, C<TIEARRAY>,
6324 or C<TIEHASH>). Typically these are arguments such as might be passed
6325 to the C<dbm_open()> function of C. The object returned by the C<new>
6326 method is also returned by the C<tie> function, which would be useful
6327 if you want to access other methods in CLASSNAME.
6329 Note that functions such as C<keys> and C<values> may return huge lists
6330 when used on large objects, like DBM files. You may prefer to use the
6331 C<each> function to iterate over such. Example:
6333 # print out history file offsets
6335 tie(%HIST, 'NDBM_File', '/usr/lib/news/history', 1, 0);
6336 while (($key,$val) = each %HIST) {
6337 print $key, ' = ', unpack('L',$val), "\n";
6341 A class implementing a hash should have the following methods:
6343 TIEHASH classname, LIST
6345 STORE this, key, value
6350 NEXTKEY this, lastkey
6355 A class implementing an ordinary array should have the following methods:
6357 TIEARRAY classname, LIST
6359 STORE this, key, value
6361 STORESIZE this, count
6367 SPLICE this, offset, length, LIST
6372 A class implementing a file handle should have the following methods:
6374 TIEHANDLE classname, LIST
6375 READ this, scalar, length, offset
6378 WRITE this, scalar, length, offset
6380 PRINTF this, format, LIST
6384 SEEK this, position, whence
6386 OPEN this, mode, LIST
6391 A class implementing a scalar should have the following methods:
6393 TIESCALAR classname, LIST
6399 Not all methods indicated above need be implemented. See L<perltie>,
6400 L<Tie::Hash>, L<Tie::Array>, L<Tie::Scalar>, and L<Tie::Handle>.
6402 Unlike C<dbmopen>, the C<tie> function will not use or require a module
6403 for you--you need to do that explicitly yourself. See L<DB_File>
6404 or the F<Config> module for interesting C<tie> implementations.
6406 For further details see L<perltie>, L<"tied VARIABLE">.
6411 Returns a reference to the object underlying VARIABLE (the same value
6412 that was originally returned by the C<tie> call that bound the variable
6413 to a package.) Returns the undefined value if VARIABLE isn't tied to a
6419 Returns the number of non-leap seconds since whatever time the system
6420 considers to be the epoch, suitable for feeding to C<gmtime> and
6421 C<localtime>. On most systems the epoch is 00:00:00 UTC, January 1, 1970;
6422 a prominent exception being Mac OS Classic which uses 00:00:00, January 1,
6423 1904 in the current local time zone for its epoch.
6425 For measuring time in better granularity than one second,
6426 you may use either the Time::HiRes module (from CPAN, and starting from
6427 Perl 5.8 part of the standard distribution), or if you have
6428 gettimeofday(2), you may be able to use the C<syscall> interface of Perl.
6429 See L<perlfaq8> for details.
6434 Returns a four-element list giving the user and system times, in
6435 seconds, for this process and the children of this process.
6437 ($user,$system,$cuser,$csystem) = times;
6439 In scalar context, C<times> returns C<$user>.
6443 The transliteration operator. Same as C<y///>. See L<perlop>.
6445 =item truncate FILEHANDLE,LENGTH
6448 =item truncate EXPR,LENGTH
6450 Truncates the file opened on FILEHANDLE, or named by EXPR, to the
6451 specified length. Produces a fatal error if truncate isn't implemented
6452 on your system. Returns true if successful, the undefined value
6455 The behavior is undefined if LENGTH is greater than the length of the
6459 X<uc> X<uppercase> X<toupper>
6463 Returns an uppercased version of EXPR. This is the internal function
6464 implementing the C<\U> escape in double-quoted strings. Respects
6465 current LC_CTYPE locale if C<use locale> in force. See L<perllocale>
6466 and L<perlunicode> for more details about locale and Unicode support.
6467 It does not attempt to do titlecase mapping on initial letters. See
6468 C<ucfirst> for that.
6470 If EXPR is omitted, uses C<$_>.
6473 X<ucfirst> X<uppercase>
6477 Returns the value of EXPR with the first character in uppercase
6478 (titlecase in Unicode). This is the internal function implementing
6479 the C<\u> escape in double-quoted strings. Respects current LC_CTYPE
6480 locale if C<use locale> in force. See L<perllocale> and L<perlunicode>
6481 for more details about locale and Unicode support.
6483 If EXPR is omitted, uses C<$_>.
6490 Sets the umask for the process to EXPR and returns the previous value.
6491 If EXPR is omitted, merely returns the current umask.
6493 The Unix permission C<rwxr-x---> is represented as three sets of three
6494 bits, or three octal digits: C<0750> (the leading 0 indicates octal
6495 and isn't one of the digits). The C<umask> value is such a number
6496 representing disabled permissions bits. The permission (or "mode")
6497 values you pass C<mkdir> or C<sysopen> are modified by your umask, so
6498 even if you tell C<sysopen> to create a file with permissions C<0777>,
6499 if your umask is C<0022> then the file will actually be created with
6500 permissions C<0755>. If your C<umask> were C<0027> (group can't
6501 write; others can't read, write, or execute), then passing
6502 C<sysopen> C<0666> would create a file with mode C<0640> (C<0666 &~
6505 Here's some advice: supply a creation mode of C<0666> for regular
6506 files (in C<sysopen>) and one of C<0777> for directories (in
6507 C<mkdir>) and executable files. This gives users the freedom of
6508 choice: if they want protected files, they might choose process umasks
6509 of C<022>, C<027>, or even the particularly antisocial mask of C<077>.
6510 Programs should rarely if ever make policy decisions better left to
6511 the user. The exception to this is when writing files that should be
6512 kept private: mail files, web browser cookies, I<.rhosts> files, and
6515 If umask(2) is not implemented on your system and you are trying to
6516 restrict access for I<yourself> (i.e., (EXPR & 0700) > 0), produces a
6517 fatal error at run time. If umask(2) is not implemented and you are
6518 not trying to restrict access for yourself, returns C<undef>.
6520 Remember that a umask is a number, usually given in octal; it is I<not> a
6521 string of octal digits. See also L</oct>, if all you have is a string.
6524 X<undef> X<undefine>
6528 Undefines the value of EXPR, which must be an lvalue. Use only on a
6529 scalar value, an array (using C<@>), a hash (using C<%>), a subroutine
6530 (using C<&>), or a typeglob (using C<*>). (Saying C<undef $hash{$key}>
6531 will probably not do what you expect on most predefined variables or
6532 DBM list values, so don't do that; see L<delete>.) Always returns the
6533 undefined value. You can omit the EXPR, in which case nothing is
6534 undefined, but you still get an undefined value that you could, for
6535 instance, return from a subroutine, assign to a variable or pass as a
6536 parameter. Examples:
6539 undef $bar{'blurfl'}; # Compare to: delete $bar{'blurfl'};
6543 undef *xyz; # destroys $xyz, @xyz, %xyz, &xyz, etc.
6544 return (wantarray ? (undef, $errmsg) : undef) if $they_blew_it;
6545 select undef, undef, undef, 0.25;
6546 ($a, $b, undef, $c) = &foo; # Ignore third value returned
6548 Note that this is a unary operator, not a list operator.
6551 X<unlink> X<delete> X<remove> X<rm>
6555 Deletes a list of files. Returns the number of files successfully
6558 $cnt = unlink 'a', 'b', 'c';
6562 Note: C<unlink> will not delete directories unless you are superuser and
6563 the B<-U> flag is supplied to Perl. Even if these conditions are
6564 met, be warned that unlinking a directory can inflict damage on your
6565 filesystem. Use C<rmdir> instead.
6567 If LIST is omitted, uses C<$_>.
6569 =item unpack TEMPLATE,EXPR
6572 =item unpack TEMPLATE
6574 C<unpack> does the reverse of C<pack>: it takes a string
6575 and expands it out into a list of values.
6576 (In scalar context, it returns merely the first value produced.)
6578 If EXPR is omitted, unpacks the C<$_> string.
6580 The string is broken into chunks described by the TEMPLATE. Each chunk
6581 is converted separately to a value. Typically, either the string is a result
6582 of C<pack>, or the characters of the string represent a C structure of some
6585 The TEMPLATE has the same format as in the C<pack> function.
6586 Here's a subroutine that does substring:
6589 my($what,$where,$howmuch) = @_;
6590 unpack("x$where a$howmuch", $what);
6595 sub ordinal { unpack("W",$_[0]); } # same as ord()
6597 In addition to fields allowed in pack(), you may prefix a field with
6598 a %<number> to indicate that
6599 you want a <number>-bit checksum of the items instead of the items
6600 themselves. Default is a 16-bit checksum. Checksum is calculated by
6601 summing numeric values of expanded values (for string fields the sum of
6602 C<ord($char)> is taken, for bit fields the sum of zeroes and ones).
6604 For example, the following
6605 computes the same number as the System V sum program:
6609 unpack("%32W*",<>) % 65535;
6612 The following efficiently counts the number of set bits in a bit vector:
6614 $setbits = unpack("%32b*", $selectmask);
6616 The C<p> and C<P> formats should be used with care. Since Perl
6617 has no way of checking whether the value passed to C<unpack()>
6618 corresponds to a valid memory location, passing a pointer value that's
6619 not known to be valid is likely to have disastrous consequences.
6621 If there are more pack codes or if the repeat count of a field or a group
6622 is larger than what the remainder of the input string allows, the result
6623 is not well defined: in some cases, the repeat count is decreased, or
6624 C<unpack()> will produce null strings or zeroes, or terminate with an
6625 error. If the input string is longer than one described by the TEMPLATE,
6626 the rest is ignored.
6628 See L</pack> for more examples and notes.
6630 =item untie VARIABLE
6633 Breaks the binding between a variable and a package. (See C<tie>.)
6634 Has no effect if the variable is not tied.
6636 =item unshift ARRAY,LIST
6639 Does the opposite of a C<shift>. Or the opposite of a C<push>,
6640 depending on how you look at it. Prepends list to the front of the
6641 array, and returns the new number of elements in the array.
6643 unshift(@ARGV, '-e') unless $ARGV[0] =~ /^-/;
6645 Note the LIST is prepended whole, not one element at a time, so the
6646 prepended elements stay in the same order. Use C<reverse> to do the
6649 =item use Module VERSION LIST
6650 X<use> X<module> X<import>
6652 =item use Module VERSION
6654 =item use Module LIST
6660 Imports some semantics into the current package from the named module,
6661 generally by aliasing certain subroutine or variable names into your
6662 package. It is exactly equivalent to
6664 BEGIN { require Module; import Module LIST; }
6666 except that Module I<must> be a bareword.
6668 VERSION may be either a numeric argument such as 5.006, which will be
6669 compared to C<$]>, or a literal of the form v5.6.1, which will be compared
6670 to C<$^V> (aka $PERL_VERSION. A fatal error is produced if VERSION is
6671 greater than the version of the current Perl interpreter; Perl will not
6672 attempt to parse the rest of the file. Compare with L</require>, which can
6673 do a similar check at run time.
6675 Specifying VERSION as a literal of the form v5.6.1 should generally be
6676 avoided, because it leads to misleading error messages under earlier
6677 versions of Perl that do not support this syntax. The equivalent numeric
6678 version should be used instead.
6680 use v5.6.1; # compile time version check
6682 use 5.006_001; # ditto; preferred for backwards compatibility
6684 This is often useful if you need to check the current Perl version before
6685 C<use>ing library modules that have changed in incompatible ways from
6686 older versions of Perl. (We try not to do this more than we have to.)
6688 The C<BEGIN> forces the C<require> and C<import> to happen at compile time. The
6689 C<require> makes sure the module is loaded into memory if it hasn't been
6690 yet. The C<import> is not a builtin--it's just an ordinary static method
6691 call into the C<Module> package to tell the module to import the list of
6692 features back into the current package. The module can implement its
6693 C<import> method any way it likes, though most modules just choose to
6694 derive their C<import> method via inheritance from the C<Exporter> class that
6695 is defined in the C<Exporter> module. See L<Exporter>. If no C<import>
6696 method can be found then the call is skipped, even if there is an AUTOLOAD
6699 If you do not want to call the package's C<import> method (for instance,
6700 to stop your namespace from being altered), explicitly supply the empty list:
6704 That is exactly equivalent to
6706 BEGIN { require Module }
6708 If the VERSION argument is present between Module and LIST, then the
6709 C<use> will call the VERSION method in class Module with the given
6710 version as an argument. The default VERSION method, inherited from
6711 the UNIVERSAL class, croaks if the given version is larger than the
6712 value of the variable C<$Module::VERSION>.
6714 Again, there is a distinction between omitting LIST (C<import> called
6715 with no arguments) and an explicit empty LIST C<()> (C<import> not
6716 called). Note that there is no comma after VERSION!
6718 Because this is a wide-open interface, pragmas (compiler directives)
6719 are also implemented this way. Currently implemented pragmas are:
6724 use sigtrap qw(SEGV BUS);
6725 use strict qw(subs vars refs);
6726 use subs qw(afunc blurfl);
6727 use warnings qw(all);
6728 use sort qw(stable _quicksort _mergesort);
6730 Some of these pseudo-modules import semantics into the current
6731 block scope (like C<strict> or C<integer>, unlike ordinary modules,
6732 which import symbols into the current package (which are effective
6733 through the end of the file).
6735 There's a corresponding C<no> command that unimports meanings imported
6736 by C<use>, i.e., it calls C<unimport Module LIST> instead of C<import>.
6737 It behaves exactly as C<import> does with respect to VERSION, an
6738 omitted LIST, empty LIST, or no unimport method being found.
6744 See L<perlmodlib> for a list of standard modules and pragmas. See L<perlrun>
6745 for the C<-M> and C<-m> command-line options to perl that give C<use>
6746 functionality from the command-line.
6751 Changes the access and modification times on each file of a list of
6752 files. The first two elements of the list must be the NUMERICAL access
6753 and modification times, in that order. Returns the number of files
6754 successfully changed. The inode change time of each file is set
6755 to the current time. For example, this code has the same effect as the
6756 Unix touch(1) command when the files I<already exist> and belong to
6757 the user running the program:
6760 $atime = $mtime = time;
6761 utime $atime, $mtime, @ARGV;
6763 Since perl 5.7.2, if the first two elements of the list are C<undef>, then
6764 the utime(2) function in the C library will be called with a null second
6765 argument. On most systems, this will set the file's access and
6766 modification times to the current time (i.e. equivalent to the example
6767 above) and will even work on other users' files where you have write
6770 utime undef, undef, @ARGV;
6772 Under NFS this will use the time of the NFS server, not the time of
6773 the local machine. If there is a time synchronization problem, the
6774 NFS server and local machine will have different times. The Unix
6775 touch(1) command will in fact normally use this form instead of the
6776 one shown in the first example.
6778 Note that only passing one of the first two elements as C<undef> will
6779 be equivalent of passing it as 0 and will not have the same effect as
6780 described when they are both C<undef>. This case will also trigger an
6781 uninitialized warning.
6783 On systems that support futimes, you might pass file handles among the
6784 files. On systems that don't support futimes, passing file handles
6785 produces a fatal error at run time.
6790 Returns a list consisting of all the values of the named hash.
6791 (In a scalar context, returns the number of values.)
6793 The values are returned in an apparently random order. The actual
6794 random order is subject to change in future versions of perl, but it
6795 is guaranteed to be the same order as either the C<keys> or C<each>
6796 function would produce on the same (unmodified) hash. Since Perl
6797 5.8.1 the ordering is different even between different runs of Perl
6798 for security reasons (see L<perlsec/"Algorithmic Complexity Attacks">).
6800 As a side effect, calling values() resets the HASH's internal iterator,
6801 see L</each>. (In particular, calling values() in void context resets
6802 the iterator with no other overhead.)
6804 Note that the values are not copied, which means modifying them will
6805 modify the contents of the hash:
6807 for (values %hash) { s/foo/bar/g } # modifies %hash values
6808 for (@hash{keys %hash}) { s/foo/bar/g } # same
6810 See also C<keys>, C<each>, and C<sort>.
6812 =item vec EXPR,OFFSET,BITS
6813 X<vec> X<bit> X<bit vector>
6815 Treats the string in EXPR as a bit vector made up of elements of
6816 width BITS, and returns the value of the element specified by OFFSET
6817 as an unsigned integer. BITS therefore specifies the number of bits
6818 that are reserved for each element in the bit vector. This must
6819 be a power of two from 1 to 32 (or 64, if your platform supports
6822 If BITS is 8, "elements" coincide with bytes of the input string.
6824 If BITS is 16 or more, bytes of the input string are grouped into chunks
6825 of size BITS/8, and each group is converted to a number as with
6826 pack()/unpack() with big-endian formats C<n>/C<N> (and analogously
6827 for BITS==64). See L<"pack"> for details.
6829 If bits is 4 or less, the string is broken into bytes, then the bits
6830 of each byte are broken into 8/BITS groups. Bits of a byte are
6831 numbered in a little-endian-ish way, as in C<0x01>, C<0x02>,
6832 C<0x04>, C<0x08>, C<0x10>, C<0x20>, C<0x40>, C<0x80>. For example,
6833 breaking the single input byte C<chr(0x36)> into two groups gives a list
6834 C<(0x6, 0x3)>; breaking it into 4 groups gives C<(0x2, 0x1, 0x3, 0x0)>.
6836 C<vec> may also be assigned to, in which case parentheses are needed
6837 to give the expression the correct precedence as in
6839 vec($image, $max_x * $x + $y, 8) = 3;
6841 If the selected element is outside the string, the value 0 is returned.
6842 If an element off the end of the string is written to, Perl will first
6843 extend the string with sufficiently many zero bytes. It is an error
6844 to try to write off the beginning of the string (i.e. negative OFFSET).
6846 The string should not contain any character with the value > 255 (which
6847 can only happen if you're using UTF-8 encoding). If it does, it will be
6848 treated as something that is not UTF-8 encoded. When the C<vec> was
6849 assigned to, other parts of your program will also no longer consider the
6850 string to be UTF-8 encoded. In other words, if you do have such characters
6851 in your string, vec() will operate on the actual byte string, and not the
6852 conceptual character string.
6854 Strings created with C<vec> can also be manipulated with the logical
6855 operators C<|>, C<&>, C<^>, and C<~>. These operators will assume a bit
6856 vector operation is desired when both operands are strings.
6857 See L<perlop/"Bitwise String Operators">.
6859 The following code will build up an ASCII string saying C<'PerlPerlPerl'>.
6860 The comments show the string after each step. Note that this code works
6861 in the same way on big-endian or little-endian machines.
6864 vec($foo, 0, 32) = 0x5065726C; # 'Perl'
6866 # $foo eq "Perl" eq "\x50\x65\x72\x6C", 32 bits
6867 print vec($foo, 0, 8); # prints 80 == 0x50 == ord('P')
6869 vec($foo, 2, 16) = 0x5065; # 'PerlPe'
6870 vec($foo, 3, 16) = 0x726C; # 'PerlPerl'
6871 vec($foo, 8, 8) = 0x50; # 'PerlPerlP'
6872 vec($foo, 9, 8) = 0x65; # 'PerlPerlPe'
6873 vec($foo, 20, 4) = 2; # 'PerlPerlPe' . "\x02"
6874 vec($foo, 21, 4) = 7; # 'PerlPerlPer'
6876 vec($foo, 45, 2) = 3; # 'PerlPerlPer' . "\x0c"
6877 vec($foo, 93, 1) = 1; # 'PerlPerlPer' . "\x2c"
6878 vec($foo, 94, 1) = 1; # 'PerlPerlPerl'
6881 To transform a bit vector into a string or list of 0's and 1's, use these:
6883 $bits = unpack("b*", $vector);
6884 @bits = split(//, unpack("b*", $vector));
6886 If you know the exact length in bits, it can be used in place of the C<*>.
6888 Here is an example to illustrate how the bits actually fall in place:
6894 unpack("V",$_) 01234567890123456789012345678901
6895 ------------------------------------------------------------------
6900 for ($shift=0; $shift < $width; ++$shift) {
6901 for ($off=0; $off < 32/$width; ++$off) {
6902 $str = pack("B*", "0"x32);
6903 $bits = (1<<$shift);
6904 vec($str, $off, $width) = $bits;
6905 $res = unpack("b*",$str);
6906 $val = unpack("V", $str);
6913 vec($_,@#,@#) = @<< == @######### @>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
6914 $off, $width, $bits, $val, $res
6918 Regardless of the machine architecture on which it is run, the above
6919 example should print the following table:
6922 unpack("V",$_) 01234567890123456789012345678901
6923 ------------------------------------------------------------------
6924 vec($_, 0, 1) = 1 == 1 10000000000000000000000000000000
6925 vec($_, 1, 1) = 1 == 2 01000000000000000000000000000000
6926 vec($_, 2, 1) = 1 == 4 00100000000000000000000000000000
6927 vec($_, 3, 1) = 1 == 8 00010000000000000000000000000000
6928 vec($_, 4, 1) = 1 == 16 00001000000000000000000000000000
6929 vec($_, 5, 1) = 1 == 32 00000100000000000000000000000000
6930 vec($_, 6, 1) = 1 == 64 00000010000000000000000000000000
6931 vec($_, 7, 1) = 1 == 128 00000001000000000000000000000000
6932 vec($_, 8, 1) = 1 == 256 00000000100000000000000000000000
6933 vec($_, 9, 1) = 1 == 512 00000000010000000000000000000000
6934 vec($_,10, 1) = 1 == 1024 00000000001000000000000000000000
6935 vec($_,11, 1) = 1 == 2048 00000000000100000000000000000000
6936 vec($_,12, 1) = 1 == 4096 00000000000010000000000000000000
6937 vec($_,13, 1) = 1 == 8192 00000000000001000000000000000000
6938 vec($_,14, 1) = 1 == 16384 00000000000000100000000000000000
6939 vec($_,15, 1) = 1 == 32768 00000000000000010000000000000000
6940 vec($_,16, 1) = 1 == 65536 00000000000000001000000000000000
6941 vec($_,17, 1) = 1 == 131072 00000000000000000100000000000000
6942 vec($_,18, 1) = 1 == 262144 00000000000000000010000000000000
6943 vec($_,19, 1) = 1 == 524288 00000000000000000001000000000000
6944 vec($_,20, 1) = 1 == 1048576 00000000000000000000100000000000
6945 vec($_,21, 1) = 1 == 2097152 00000000000000000000010000000000
6946 vec($_,22, 1) = 1 == 4194304 00000000000000000000001000000000
6947 vec($_,23, 1) = 1 == 8388608 00000000000000000000000100000000
6948 vec($_,24, 1) = 1 == 16777216 00000000000000000000000010000000
6949 vec($_,25, 1) = 1 == 33554432 00000000000000000000000001000000
6950 vec($_,26, 1) = 1 == 67108864 00000000000000000000000000100000
6951 vec($_,27, 1) = 1 == 134217728 00000000000000000000000000010000
6952 vec($_,28, 1) = 1 == 268435456 00000000000000000000000000001000
6953 vec($_,29, 1) = 1 == 536870912 00000000000000000000000000000100
6954 vec($_,30, 1) = 1 == 1073741824 00000000000000000000000000000010
6955 vec($_,31, 1) = 1 == 2147483648 00000000000000000000000000000001
6956 vec($_, 0, 2) = 1 == 1 10000000000000000000000000000000
6957 vec($_, 1, 2) = 1 == 4 00100000000000000000000000000000
6958 vec($_, 2, 2) = 1 == 16 00001000000000000000000000000000
6959 vec($_, 3, 2) = 1 == 64 00000010000000000000000000000000
6960 vec($_, 4, 2) = 1 == 256 00000000100000000000000000000000
6961 vec($_, 5, 2) = 1 == 1024 00000000001000000000000000000000
6962 vec($_, 6, 2) = 1 == 4096 00000000000010000000000000000000
6963 vec($_, 7, 2) = 1 == 16384 00000000000000100000000000000000
6964 vec($_, 8, 2) = 1 == 65536 00000000000000001000000000000000
6965 vec($_, 9, 2) = 1 == 262144 00000000000000000010000000000000
6966 vec($_,10, 2) = 1 == 1048576 00000000000000000000100000000000
6967 vec($_,11, 2) = 1 == 4194304 00000000000000000000001000000000
6968 vec($_,12, 2) = 1 == 16777216 00000000000000000000000010000000
6969 vec($_,13, 2) = 1 == 67108864 00000000000000000000000000100000
6970 vec($_,14, 2) = 1 == 268435456 00000000000000000000000000001000
6971 vec($_,15, 2) = 1 == 1073741824 00000000000000000000000000000010
6972 vec($_, 0, 2) = 2 == 2 01000000000000000000000000000000
6973 vec($_, 1, 2) = 2 == 8 00010000000000000000000000000000
6974 vec($_, 2, 2) = 2 == 32 00000100000000000000000000000000
6975 vec($_, 3, 2) = 2 == 128 00000001000000000000000000000000
6976 vec($_, 4, 2) = 2 == 512 00000000010000000000000000000000
6977 vec($_, 5, 2) = 2 == 2048 00000000000100000000000000000000
6978 vec($_, 6, 2) = 2 == 8192 00000000000001000000000000000000
6979 vec($_, 7, 2) = 2 == 32768 00000000000000010000000000000000
6980 vec($_, 8, 2) = 2 == 131072 00000000000000000100000000000000
6981 vec($_, 9, 2) = 2 == 524288 00000000000000000001000000000000
6982 vec($_,10, 2) = 2 == 2097152 00000000000000000000010000000000
6983 vec($_,11, 2) = 2 == 8388608 00000000000000000000000100000000
6984 vec($_,12, 2) = 2 == 33554432 00000000000000000000000001000000
6985 vec($_,13, 2) = 2 == 134217728 00000000000000000000000000010000
6986 vec($_,14, 2) = 2 == 536870912 00000000000000000000000000000100
6987 vec($_,15, 2) = 2 == 2147483648 00000000000000000000000000000001
6988 vec($_, 0, 4) = 1 == 1 10000000000000000000000000000000
6989 vec($_, 1, 4) = 1 == 16 00001000000000000000000000000000
6990 vec($_, 2, 4) = 1 == 256 00000000100000000000000000000000
6991 vec($_, 3, 4) = 1 == 4096 00000000000010000000000000000000
6992 vec($_, 4, 4) = 1 == 65536 00000000000000001000000000000000
6993 vec($_, 5, 4) = 1 == 1048576 00000000000000000000100000000000
6994 vec($_, 6, 4) = 1 == 16777216 00000000000000000000000010000000
6995 vec($_, 7, 4) = 1 == 268435456 00000000000000000000000000001000
6996 vec($_, 0, 4) = 2 == 2 01000000000000000000000000000000
6997 vec($_, 1, 4) = 2 == 32 00000100000000000000000000000000
6998 vec($_, 2, 4) = 2 == 512 00000000010000000000000000000000
6999 vec($_, 3, 4) = 2 == 8192 00000000000001000000000000000000
7000 vec($_, 4, 4) = 2 == 131072 00000000000000000100000000000000
7001 vec($_, 5, 4) = 2 == 2097152 00000000000000000000010000000000
7002 vec($_, 6, 4) = 2 == 33554432 00000000000000000000000001000000
7003 vec($_, 7, 4) = 2 == 536870912 00000000000000000000000000000100
7004 vec($_, 0, 4) = 4 == 4 00100000000000000000000000000000
7005 vec($_, 1, 4) = 4 == 64 00000010000000000000000000000000
7006 vec($_, 2, 4) = 4 == 1024 00000000001000000000000000000000
7007 vec($_, 3, 4) = 4 == 16384 00000000000000100000000000000000
7008 vec($_, 4, 4) = 4 == 262144 00000000000000000010000000000000
7009 vec($_, 5, 4) = 4 == 4194304 00000000000000000000001000000000
7010 vec($_, 6, 4) = 4 == 67108864 00000000000000000000000000100000
7011 vec($_, 7, 4) = 4 == 1073741824 00000000000000000000000000000010
7012 vec($_, 0, 4) = 8 == 8 00010000000000000000000000000000
7013 vec($_, 1, 4) = 8 == 128 00000001000000000000000000000000
7014 vec($_, 2, 4) = 8 == 2048 00000000000100000000000000000000
7015 vec($_, 3, 4) = 8 == 32768 00000000000000010000000000000000
7016 vec($_, 4, 4) = 8 == 524288 00000000000000000001000000000000
7017 vec($_, 5, 4) = 8 == 8388608 00000000000000000000000100000000
7018 vec($_, 6, 4) = 8 == 134217728 00000000000000000000000000010000
7019 vec($_, 7, 4) = 8 == 2147483648 00000000000000000000000000000001
7020 vec($_, 0, 8) = 1 == 1 10000000000000000000000000000000
7021 vec($_, 1, 8) = 1 == 256 00000000100000000000000000000000
7022 vec($_, 2, 8) = 1 == 65536 00000000000000001000000000000000
7023 vec($_, 3, 8) = 1 == 16777216 00000000000000000000000010000000
7024 vec($_, 0, 8) = 2 == 2 01000000000000000000000000000000
7025 vec($_, 1, 8) = 2 == 512 00000000010000000000000000000000
7026 vec($_, 2, 8) = 2 == 131072 00000000000000000100000000000000
7027 vec($_, 3, 8) = 2 == 33554432 00000000000000000000000001000000
7028 vec($_, 0, 8) = 4 == 4 00100000000000000000000000000000
7029 vec($_, 1, 8) = 4 == 1024 00000000001000000000000000000000
7030 vec($_, 2, 8) = 4 == 262144 00000000000000000010000000000000
7031 vec($_, 3, 8) = 4 == 67108864 00000000000000000000000000100000
7032 vec($_, 0, 8) = 8 == 8 00010000000000000000000000000000
7033 vec($_, 1, 8) = 8 == 2048 00000000000100000000000000000000
7034 vec($_, 2, 8) = 8 == 524288 00000000000000000001000000000000
7035 vec($_, 3, 8) = 8 == 134217728 00000000000000000000000000010000
7036 vec($_, 0, 8) = 16 == 16 00001000000000000000000000000000
7037 vec($_, 1, 8) = 16 == 4096 00000000000010000000000000000000
7038 vec($_, 2, 8) = 16 == 1048576 00000000000000000000100000000000
7039 vec($_, 3, 8) = 16 == 268435456 00000000000000000000000000001000
7040 vec($_, 0, 8) = 32 == 32 00000100000000000000000000000000
7041 vec($_, 1, 8) = 32 == 8192 00000000000001000000000000000000
7042 vec($_, 2, 8) = 32 == 2097152 00000000000000000000010000000000
7043 vec($_, 3, 8) = 32 == 536870912 00000000000000000000000000000100
7044 vec($_, 0, 8) = 64 == 64 00000010000000000000000000000000
7045 vec($_, 1, 8) = 64 == 16384 00000000000000100000000000000000
7046 vec($_, 2, 8) = 64 == 4194304 00000000000000000000001000000000
7047 vec($_, 3, 8) = 64 == 1073741824 00000000000000000000000000000010
7048 vec($_, 0, 8) = 128 == 128 00000001000000000000000000000000
7049 vec($_, 1, 8) = 128 == 32768 00000000000000010000000000000000
7050 vec($_, 2, 8) = 128 == 8388608 00000000000000000000000100000000
7051 vec($_, 3, 8) = 128 == 2147483648 00000000000000000000000000000001
7056 Behaves like the wait(2) system call on your system: it waits for a child
7057 process to terminate and returns the pid of the deceased process, or
7058 C<-1> if there are no child processes. The status is returned in C<$?>
7059 and C<{^CHILD_ERROR_NATIVE}>.
7060 Note that a return value of C<-1> could mean that child processes are
7061 being automatically reaped, as described in L<perlipc>.
7063 =item waitpid PID,FLAGS
7066 Waits for a particular child process to terminate and returns the pid of
7067 the deceased process, or C<-1> if there is no such child process. On some
7068 systems, a value of 0 indicates that there are processes still running.
7069 The status is returned in C<$?> and C<{^CHILD_ERROR_NATIVE}>. If you say
7071 use POSIX ":sys_wait_h";
7074 $kid = waitpid(-1, WNOHANG);
7077 then you can do a non-blocking wait for all pending zombie processes.
7078 Non-blocking wait is available on machines supporting either the
7079 waitpid(2) or wait4(2) system calls. However, waiting for a particular
7080 pid with FLAGS of C<0> is implemented everywhere. (Perl emulates the
7081 system call by remembering the status values of processes that have
7082 exited but have not been harvested by the Perl script yet.)
7084 Note that on some systems, a return value of C<-1> could mean that child
7085 processes are being automatically reaped. See L<perlipc> for details,
7086 and for other examples.
7089 X<wantarray> X<context>
7091 Returns true if the context of the currently executing subroutine or
7092 C<eval> is looking for a list value. Returns false if the context is
7093 looking for a scalar. Returns the undefined value if the context is
7094 looking for no value (void context).
7096 return unless defined wantarray; # don't bother doing more
7097 my @a = complex_calculation();
7098 return wantarray ? @a : "@a";
7100 C<wantarray()>'s result is unspecified in the top level of a file,
7101 in a C<BEGIN>, C<CHECK>, C<INIT> or C<END> block, or in a C<DESTROY>
7104 This function should have been named wantlist() instead.
7107 X<warn> X<warning> X<STDERR>
7109 Produces a message on STDERR just like C<die>, but doesn't exit or throw
7112 If LIST is empty and C<$@> already contains a value (typically from a
7113 previous eval) that value is used after appending C<"\t...caught">
7114 to C<$@>. This is useful for staying almost, but not entirely similar to
7117 If C<$@> is empty then the string C<"Warning: Something's wrong"> is used.
7119 No message is printed if there is a C<$SIG{__WARN__}> handler
7120 installed. It is the handler's responsibility to deal with the message
7121 as it sees fit (like, for instance, converting it into a C<die>). Most
7122 handlers must therefore make arrangements to actually display the
7123 warnings that they are not prepared to deal with, by calling C<warn>
7124 again in the handler. Note that this is quite safe and will not
7125 produce an endless loop, since C<__WARN__> hooks are not called from
7128 You will find this behavior is slightly different from that of
7129 C<$SIG{__DIE__}> handlers (which don't suppress the error text, but can
7130 instead call C<die> again to change it).
7132 Using a C<__WARN__> handler provides a powerful way to silence all
7133 warnings (even the so-called mandatory ones). An example:
7135 # wipe out *all* compile-time warnings
7136 BEGIN { $SIG{'__WARN__'} = sub { warn $_[0] if $DOWARN } }
7138 my $foo = 20; # no warning about duplicate my $foo,
7139 # but hey, you asked for it!
7140 # no compile-time or run-time warnings before here
7143 # run-time warnings enabled after here
7144 warn "\$foo is alive and $foo!"; # does show up
7146 See L<perlvar> for details on setting C<%SIG> entries, and for more
7147 examples. See the Carp module for other kinds of warnings using its
7148 carp() and cluck() functions.
7150 =item write FILEHANDLE
7157 Writes a formatted record (possibly multi-line) to the specified FILEHANDLE,
7158 using the format associated with that file. By default the format for
7159 a file is the one having the same name as the filehandle, but the
7160 format for the current output channel (see the C<select> function) may be set
7161 explicitly by assigning the name of the format to the C<$~> variable.
7163 Top of form processing is handled automatically: if there is
7164 insufficient room on the current page for the formatted record, the
7165 page is advanced by writing a form feed, a special top-of-page format
7166 is used to format the new page header, and then the record is written.
7167 By default the top-of-page format is the name of the filehandle with
7168 "_TOP" appended, but it may be dynamically set to the format of your
7169 choice by assigning the name to the C<$^> variable while the filehandle is
7170 selected. The number of lines remaining on the current page is in
7171 variable C<$->, which can be set to C<0> to force a new page.
7173 If FILEHANDLE is unspecified, output goes to the current default output
7174 channel, which starts out as STDOUT but may be changed by the
7175 C<select> operator. If the FILEHANDLE is an EXPR, then the expression
7176 is evaluated and the resulting string is used to look up the name of
7177 the FILEHANDLE at run time. For more on formats, see L<perlform>.
7179 Note that write is I<not> the opposite of C<read>. Unfortunately.
7183 The transliteration operator. Same as C<tr///>. See L<perlop>.