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, scalar arguments
18 come first and list argument follow, and there can only ever
19 be 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 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 literal 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 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. Whitespace
37 between the function and left parenthesis doesn't count, so sometimes
38 you need to be careful:
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 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 ("syscalls")
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 Extension modules can also hook into the Perl parser to define new
90 kinds of keyword-headed expression. These may look like functions, but
91 may also look completely different. The syntax following the keyword
92 is defined entirely by the extension. If you are an implementor, see
93 L<perlapi/PL_keyword_plugin> for the mechanism. If you are using such
94 a module, see the module's documentation for details of the syntax that
97 =head2 Perl Functions by Category
100 Here are Perl's functions (including things that look like
101 functions, like some keywords and named operators)
102 arranged by category. Some functions appear in more
107 =item Functions for SCALARs or strings
108 X<scalar> X<string> X<character>
110 C<chomp>, C<chop>, C<chr>, C<crypt>, C<hex>, C<index>, C<lc>, C<lcfirst>,
111 C<length>, C<oct>, C<ord>, C<pack>, C<q//>, C<qq//>, C<reverse>,
112 C<rindex>, C<sprintf>, C<substr>, C<tr///>, C<uc>, C<ucfirst>, C<y///>
114 =item Regular expressions and pattern matching
115 X<regular expression> X<regex> X<regexp>
117 C<m//>, C<pos>, C<quotemeta>, C<s///>, C<split>, C<study>, C<qr//>
119 =item Numeric functions
120 X<numeric> X<number> X<trigonometric> X<trigonometry>
122 C<abs>, C<atan2>, C<cos>, C<exp>, C<hex>, C<int>, C<log>, C<oct>, C<rand>,
123 C<sin>, C<sqrt>, C<srand>
125 =item Functions for real @ARRAYs
128 C<pop>, C<push>, C<shift>, C<splice>, C<unshift>
130 =item Functions for list data
133 C<grep>, C<join>, C<map>, C<qw//>, C<reverse>, C<sort>, C<unpack>
135 =item Functions for real %HASHes
138 C<delete>, C<each>, C<exists>, C<keys>, C<values>
140 =item Input and output functions
141 X<I/O> X<input> X<output> X<dbm>
143 C<binmode>, C<close>, C<closedir>, C<dbmclose>, C<dbmopen>, C<die>, C<eof>,
144 C<fileno>, C<flock>, C<format>, C<getc>, C<print>, C<printf>, C<read>,
145 C<readdir>, C<rewinddir>, C<say>, C<seek>, C<seekdir>, C<select>, C<syscall>,
146 C<sysread>, C<sysseek>, C<syswrite>, C<tell>, C<telldir>, C<truncate>,
149 =item Functions for fixed length data or records
151 C<pack>, C<read>, C<syscall>, C<sysread>, C<syswrite>, C<unpack>, C<vec>
153 =item Functions for filehandles, files, or directories
154 X<file> X<filehandle> X<directory> X<pipe> X<link> X<symlink>
156 C<-I<X>>, C<chdir>, C<chmod>, C<chown>, C<chroot>, C<fcntl>, C<glob>,
157 C<ioctl>, C<link>, C<lstat>, C<mkdir>, C<open>, C<opendir>,
158 C<readlink>, C<rename>, C<rmdir>, C<stat>, C<symlink>, C<sysopen>,
159 C<umask>, C<unlink>, C<utime>
161 =item Keywords related to the control flow of your Perl program
164 C<caller>, C<continue>, C<die>, C<do>, C<dump>, C<eval>, C<exit>,
165 C<goto>, C<last>, C<next>, C<redo>, C<return>, C<sub>, C<wantarray>
167 =item Keywords related to switch
169 C<break>, C<continue>, C<given>, C<when>, C<default>
171 (These are available only if you enable the C<"switch"> feature.
172 See L<feature> and L<perlsyn/"Switch statements">.)
174 =item Keywords related to scoping
176 C<caller>, C<import>, C<local>, C<my>, C<our>, C<state>, C<package>,
179 (C<state> is available only if the C<"state"> feature is enabled. See
182 =item Miscellaneous functions
184 C<defined>, C<dump>, C<eval>, C<formline>, C<local>, C<my>, C<our>,
185 C<reset>, C<scalar>, C<state>, C<undef>, C<wantarray>
187 =item Functions for processes and process groups
188 X<process> X<pid> X<process id>
190 C<alarm>, C<exec>, C<fork>, C<getpgrp>, C<getppid>, C<getpriority>, C<kill>,
191 C<pipe>, C<qx//>, C<setpgrp>, C<setpriority>, C<sleep>, C<system>,
192 C<times>, C<wait>, C<waitpid>
194 =item Keywords related to Perl modules
197 C<do>, C<import>, C<no>, C<package>, C<require>, C<use>
199 =item Keywords related to classes and object-orientation
200 X<object> X<class> X<package>
202 C<bless>, C<dbmclose>, C<dbmopen>, C<package>, C<ref>, C<tie>, C<tied>,
205 =item Low-level socket functions
208 C<accept>, C<bind>, C<connect>, C<getpeername>, C<getsockname>,
209 C<getsockopt>, C<listen>, C<recv>, C<send>, C<setsockopt>, C<shutdown>,
210 C<socket>, C<socketpair>
212 =item System V interprocess communication functions
213 X<IPC> X<System V> X<semaphore> X<shared memory> X<memory> X<message>
215 C<msgctl>, C<msgget>, C<msgrcv>, C<msgsnd>, C<semctl>, C<semget>, C<semop>,
216 C<shmctl>, C<shmget>, C<shmread>, C<shmwrite>
218 =item Fetching user and group info
219 X<user> X<group> X<password> X<uid> X<gid> X<passwd> X</etc/passwd>
221 C<endgrent>, C<endhostent>, C<endnetent>, C<endpwent>, C<getgrent>,
222 C<getgrgid>, C<getgrnam>, C<getlogin>, C<getpwent>, C<getpwnam>,
223 C<getpwuid>, C<setgrent>, C<setpwent>
225 =item Fetching network info
226 X<network> X<protocol> X<host> X<hostname> X<IP> X<address> X<service>
228 C<endprotoent>, C<endservent>, C<gethostbyaddr>, C<gethostbyname>,
229 C<gethostent>, C<getnetbyaddr>, C<getnetbyname>, C<getnetent>,
230 C<getprotobyname>, C<getprotobynumber>, C<getprotoent>,
231 C<getservbyname>, C<getservbyport>, C<getservent>, C<sethostent>,
232 C<setnetent>, C<setprotoent>, C<setservent>
234 =item Time-related functions
237 C<gmtime>, C<localtime>, C<time>, C<times>
239 =item Functions new in perl5
242 C<abs>, C<bless>, C<break>, C<chomp>, C<chr>, C<continue>, C<default>,
243 C<exists>, C<formline>, C<given>, C<glob>, C<import>, C<lc>, C<lcfirst>,
244 C<lock>, C<map>, C<my>, C<no>, C<our>, C<prototype>, C<qr//>, C<qw//>, C<qx//>,
245 C<readline>, C<readpipe>, C<ref>, C<sub>*, C<sysopen>, C<tie>, C<tied>, C<uc>,
246 C<ucfirst>, C<untie>, C<use>, C<when>
248 * C<sub> was a keyword in Perl 4, but in Perl 5 it is an
249 operator, which can be used in expressions.
251 =item Functions obsoleted in perl5
253 C<dbmclose>, C<dbmopen>
258 X<portability> X<Unix> X<portable>
260 Perl was born in Unix and can therefore access all common Unix
261 system calls. In non-Unix environments, the functionality of some
262 Unix system calls may not be available, or details of the available
263 functionality may differ slightly. The Perl functions affected
266 C<-X>, C<binmode>, C<chmod>, C<chown>, C<chroot>, C<crypt>,
267 C<dbmclose>, C<dbmopen>, C<dump>, C<endgrent>, C<endhostent>,
268 C<endnetent>, C<endprotoent>, C<endpwent>, C<endservent>, C<exec>,
269 C<fcntl>, C<flock>, C<fork>, C<getgrent>, C<getgrgid>, C<gethostbyname>,
270 C<gethostent>, C<getlogin>, C<getnetbyaddr>, C<getnetbyname>, C<getnetent>,
271 C<getppid>, C<getpgrp>, C<getpriority>, C<getprotobynumber>,
272 C<getprotoent>, C<getpwent>, C<getpwnam>, C<getpwuid>,
273 C<getservbyport>, C<getservent>, C<getsockopt>, C<glob>, C<ioctl>,
274 C<kill>, C<link>, C<lstat>, C<msgctl>, C<msgget>, C<msgrcv>,
275 C<msgsnd>, C<open>, C<pipe>, C<readlink>, C<rename>, C<select>, C<semctl>,
276 C<semget>, C<semop>, C<setgrent>, C<sethostent>, C<setnetent>,
277 C<setpgrp>, C<setpriority>, C<setprotoent>, C<setpwent>,
278 C<setservent>, C<setsockopt>, C<shmctl>, C<shmget>, C<shmread>,
279 C<shmwrite>, C<socket>, C<socketpair>,
280 C<stat>, C<symlink>, C<syscall>, C<sysopen>, C<system>,
281 C<times>, C<truncate>, C<umask>, C<unlink>,
282 C<utime>, C<wait>, C<waitpid>
284 For more information about the portability of these functions, see
285 L<perlport> and other available platform-specific documentation.
287 =head2 Alphabetical Listing of Perl Functions
292 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>
293 X<-S>X<-b>X<-c>X<-t>X<-u>X<-g>X<-k>X<-T>X<-B>X<-M>X<-A>X<-C>
301 A file test, where X is one of the letters listed below. This unary
302 operator takes one argument, either a filename, a filehandle, or a dirhandle,
303 and tests the associated file to see if something is true about it. If the
304 argument is omitted, tests C<$_>, except for C<-t>, which tests STDIN.
305 Unless otherwise documented, it returns C<1> for true and C<''> for false, or
306 the undefined value if the file doesn't exist. Despite the funny
307 names, precedence is the same as any other named unary operator. The
308 operator may be any of:
310 -r File is readable by effective uid/gid.
311 -w File is writable by effective uid/gid.
312 -x File is executable by effective uid/gid.
313 -o File is owned by effective uid.
315 -R File is readable by real uid/gid.
316 -W File is writable by real uid/gid.
317 -X File is executable by real uid/gid.
318 -O File is owned by real uid.
321 -z File has zero size (is empty).
322 -s File has nonzero size (returns size in bytes).
324 -f File is a plain file.
325 -d File is a directory.
326 -l File is a symbolic link.
327 -p File is a named pipe (FIFO), or Filehandle is a pipe.
329 -b File is a block special file.
330 -c File is a character special file.
331 -t Filehandle is opened to a tty.
333 -u File has setuid bit set.
334 -g File has setgid bit set.
335 -k File has sticky bit set.
337 -T File is an ASCII text file (heuristic guess).
338 -B File is a "binary" file (opposite of -T).
340 -M Script start time minus file modification time, in days.
341 -A Same for access time.
342 -C Same for inode change time (Unix, may differ for other platforms)
348 next unless -f $_; # ignore specials
352 The interpretation of the file permission operators C<-r>, C<-R>,
353 C<-w>, C<-W>, C<-x>, and C<-X> is by default based solely on the mode
354 of the file and the uids and gids of the user. There may be other
355 reasons you can't actually read, write, or execute the file: for
356 example network filesystem access controls, ACLs (access control lists),
357 read-only filesystems, and unrecognized executable formats. Note
358 that the use of these six specific operators to verify if some operation
359 is possible is usually a mistake, because it may be open to race
362 Also note that, for the superuser on the local filesystems, the C<-r>,
363 C<-R>, C<-w>, and C<-W> tests always return 1, and C<-x> and C<-X> return 1
364 if any execute bit is set in the mode. Scripts run by the superuser
365 may thus need to do a stat() to determine the actual mode of the file,
366 or temporarily set their effective uid to something else.
368 If you are using ACLs, there is a pragma called C<filetest> that may
369 produce more accurate results than the bare stat() mode bits.
370 When under the C<use filetest 'access'> the above-mentioned filetests
371 test whether the permission can (not) be granted using the
372 access(2) family of system calls. Also note that the C<-x> and C<-X> may
373 under this pragma return true even if there are no execute permission
374 bits set (nor any extra execute permission ACLs). This strangeness is
375 due to the underlying system calls' definitions. Note also that, due to
376 the implementation of C<use filetest 'access'>, the C<_> special
377 filehandle won't cache the results of the file tests when this pragma is
378 in effect. Read the documentation for the C<filetest> pragma for more
381 Note that C<-s/a/b/> does not do a negated substitution. Saying
382 C<-exp($foo)> still works as expected, however: only single letters
383 following a minus are interpreted as file tests.
385 The C<-T> and C<-B> switches work as follows. The first block or so of the
386 file is examined for odd characters such as strange control codes or
387 characters with the high bit set. If too many strange characters (>30%)
388 are found, it's a C<-B> file; otherwise it's a C<-T> file. Also, any file
389 containing a zero byte in the first block is considered a binary file. If C<-T>
390 or C<-B> is used on a filehandle, the current IO buffer is examined
391 rather than the first block. Both C<-T> and C<-B> return true on an empty
392 file, or a file at EOF when testing a filehandle. Because you have to
393 read a file to do the C<-T> test, on most occasions you want to use a C<-f>
394 against the file first, as in C<next unless -f $file && -T $file>.
396 If any of the file tests (or either the C<stat> or C<lstat> operators) are given
397 the special filehandle consisting of a solitary underline, then the stat
398 structure of the previous file test (or stat operator) is used, saving
399 a system call. (This doesn't work with C<-t>, and you need to remember
400 that lstat() and C<-l> leave values in the stat structure for the
401 symbolic link, not the real file.) (Also, if the stat buffer was filled by
402 an C<lstat> call, C<-T> and C<-B> will reset it with the results of C<stat _>).
405 print "Can do.\n" if -r $a || -w _ || -x _;
408 print "Readable\n" if -r _;
409 print "Writable\n" if -w _;
410 print "Executable\n" if -x _;
411 print "Setuid\n" if -u _;
412 print "Setgid\n" if -g _;
413 print "Sticky\n" if -k _;
414 print "Text\n" if -T _;
415 print "Binary\n" if -B _;
417 As of Perl 5.9.1, as a form of purely syntactic sugar, you can stack file
418 test operators, in a way that C<-f -w -x $file> is equivalent to
419 C<-x $file && -w _ && -f _>. (This is only fancy fancy: if you use
420 the return value of C<-f $file> as an argument to another filetest
421 operator, no special magic will happen.)
428 Returns the absolute value of its argument.
429 If VALUE is omitted, uses C<$_>.
431 =item accept NEWSOCKET,GENERICSOCKET
434 Accepts an incoming socket connect, just as accept(2)
435 does. Returns the packed address if it succeeded, false otherwise.
436 See the example in L<perlipc/"Sockets: Client/Server Communication">.
438 On systems that support a close-on-exec flag on files, the flag will
439 be set for the newly opened file descriptor, as determined by the
440 value of $^F. See L<perlvar/$^F>.
449 Arranges to have a SIGALRM delivered to this process after the
450 specified number of wallclock seconds has elapsed. If SECONDS is not
451 specified, the value stored in C<$_> is used. (On some machines,
452 unfortunately, the elapsed time may be up to one second less or more
453 than you specified because of how seconds are counted, and process
454 scheduling may delay the delivery of the signal even further.)
456 Only one timer may be counting at once. Each call disables the
457 previous timer, and an argument of C<0> may be supplied to cancel the
458 previous timer without starting a new one. The returned value is the
459 amount of time remaining on the previous timer.
461 For delays of finer granularity than one second, the Time::HiRes module
462 (from CPAN, and starting from Perl 5.8 part of the standard
463 distribution) provides ualarm(). You may also use Perl's four-argument
464 version of select() leaving the first three arguments undefined, or you
465 might be able to use the C<syscall> interface to access setitimer(2) if
466 your system supports it. See L<perlfaq8> for details.
468 It is usually a mistake to intermix C<alarm> and C<sleep> calls, because
469 C<sleep> may be internally implemented on your system with C<alarm>.
471 If you want to use C<alarm> to time out a system call you need to use an
472 C<eval>/C<die> pair. You can't rely on the alarm causing the system call to
473 fail with C<$!> set to C<EINTR> because Perl sets up signal handlers to
474 restart system calls on some systems. Using C<eval>/C<die> always works,
475 modulo the caveats given in L<perlipc/"Signals">.
478 local $SIG{ALRM} = sub { die "alarm\n" }; # NB: \n required
480 $nread = sysread SOCKET, $buffer, $size;
484 die unless $@ eq "alarm\n"; # propagate unexpected errors
491 For more information see L<perlipc>.
494 X<atan2> X<arctangent> X<tan> X<tangent>
496 Returns the arctangent of Y/X in the range -PI to PI.
498 For the tangent operation, you may use the C<Math::Trig::tan>
499 function, or use the familiar relation:
501 sub tan { sin($_[0]) / cos($_[0]) }
503 The return value for C<atan2(0,0)> is implementation-defined; consult
504 your atan2(3) manpage for more information.
506 =item bind SOCKET,NAME
509 Binds a network address to a socket, just as bind(2)
510 does. Returns true if it succeeded, false otherwise. NAME should be a
511 packed address of the appropriate type for the socket. See the examples in
512 L<perlipc/"Sockets: Client/Server Communication">.
514 =item binmode FILEHANDLE, LAYER
515 X<binmode> X<binary> X<text> X<DOS> X<Windows>
517 =item binmode FILEHANDLE
519 Arranges for FILEHANDLE to be read or written in "binary" or "text"
520 mode on systems where the run-time libraries distinguish between
521 binary and text files. If FILEHANDLE is an expression, the value is
522 taken as the name of the filehandle. Returns true on success,
523 otherwise it returns C<undef> and sets C<$!> (errno).
525 On some systems (in general, DOS and Windows-based systems) binmode()
526 is necessary when you're not working with a text file. For the sake
527 of portability it is a good idea to always use it when appropriate,
528 and to never use it when it isn't appropriate. Also, people can
529 set their I/O to be by default UTF-8 encoded Unicode, not bytes.
531 In other words: regardless of platform, use binmode() on binary data,
532 like for example images.
534 If LAYER is present it is a single string, but may contain multiple
535 directives. The directives alter the behaviour of the filehandle.
536 When LAYER is present using binmode on a text file makes sense.
538 If LAYER is omitted or specified as C<:raw> the filehandle is made
539 suitable for passing binary data. This includes turning off possible CRLF
540 translation and marking it as bytes (as opposed to Unicode characters).
541 Note that, despite what may be implied in I<"Programming Perl"> (the
542 Camel, 3rd edition) or elsewhere, C<:raw> is I<not> simply the inverse of C<:crlf>.
543 Other layers that would affect the binary nature of the stream are
544 I<also> disabled. See L<PerlIO>, L<perlrun>, and the discussion about the
545 PERLIO environment variable.
547 The C<:bytes>, C<:crlf>, C<:utf8>, and any other directives of the
548 form C<:...>, are called I/O I<layers>. The C<open> pragma can be used to
549 establish default I/O layers. See L<open>.
551 I<The LAYER parameter of the binmode() function is described as "DISCIPLINE"
552 in "Programming Perl, 3rd Edition". However, since the publishing of this
553 book, by many known as "Camel III", the consensus of the naming of this
554 functionality has moved from "discipline" to "layer". All documentation
555 of this version of Perl therefore refers to "layers" rather than to
556 "disciplines". Now back to the regularly scheduled documentation...>
558 To mark FILEHANDLE as UTF-8, use C<:utf8> or C<:encoding(utf8)>.
559 C<:utf8> just marks the data as UTF-8 without further checking,
560 while C<:encoding(utf8)> checks the data for actually being valid
561 UTF-8. More details can be found in L<PerlIO::encoding>.
563 In general, binmode() should be called after open() but before any I/O
564 is done on the filehandle. Calling binmode() normally flushes any
565 pending buffered output data (and perhaps pending input data) on the
566 handle. An exception to this is the C<:encoding> layer that
567 changes the default character encoding of the handle, see L<open>.
568 The C<:encoding> layer sometimes needs to be called in
569 mid-stream, and it doesn't flush the stream. The C<:encoding>
570 also implicitly pushes on top of itself the C<:utf8> layer because
571 internally Perl operates on UTF8-encoded Unicode characters.
573 The operating system, device drivers, C libraries, and Perl run-time
574 system all work together to let the programmer treat a single
575 character (C<\n>) as the line terminator, irrespective of the external
576 representation. On many operating systems, the native text file
577 representation matches the internal representation, but on some
578 platforms the external representation of C<\n> is made up of more than
581 Mac OS, all variants of Unix, and Stream_LF files on VMS use a single
582 character to end each line in the external representation of text (even
583 though that single character is CARRIAGE RETURN on Mac OS and LINE FEED
584 on Unix and most VMS files). In other systems like OS/2, DOS and the
585 various flavors of MS-Windows your program sees a C<\n> as a simple C<\cJ>,
586 but what's stored in text files are the two characters C<\cM\cJ>. That
587 means that, if you don't use binmode() on these systems, C<\cM\cJ>
588 sequences on disk will be converted to C<\n> on input, and any C<\n> in
589 your program will be converted back to C<\cM\cJ> on output. This is what
590 you want for text files, but it can be disastrous for binary files.
592 Another consequence of using binmode() (on some systems) is that
593 special end-of-file markers will be seen as part of the data stream.
594 For systems from the Microsoft family this means that if your binary
595 data contains C<\cZ>, the I/O subsystem will regard it as the end of
596 the file, unless you use binmode().
598 binmode() is important not only for readline() and print() operations,
599 but also when using read(), seek(), sysread(), syswrite() and tell()
600 (see L<perlport> for more details). See the C<$/> and C<$\> variables
601 in L<perlvar> for how to manually set your input and output
602 line-termination sequences.
604 =item bless REF,CLASSNAME
609 This function tells the thingy referenced by REF that it is now an object
610 in the CLASSNAME package. If CLASSNAME is omitted, the current package
611 is used. Because a C<bless> is often the last thing in a constructor,
612 it returns the reference for convenience. Always use the two-argument
613 version if a derived class might inherit the function doing the blessing.
614 See L<perltoot> and L<perlobj> for more about the blessing (and blessings)
617 Consider always blessing objects in CLASSNAMEs that are mixed case.
618 Namespaces with all lowercase names are considered reserved for
619 Perl pragmata. Builtin types have all uppercase names. To prevent
620 confusion, you may wish to avoid such package names as well. Make sure
621 that CLASSNAME is a true value.
623 See L<perlmod/"Perl Modules">.
627 Break out of a C<given()> block.
629 This keyword is enabled by the C<"switch"> feature: see L<feature>
630 for more information.
633 X<caller> X<call stack> X<stack> X<stack trace>
637 Returns the context of the current subroutine call. In scalar context,
638 returns the caller's package name if there I<is> a caller (that is, if
639 we're in a subroutine or C<eval> or C<require>) and the undefined value
640 otherwise. In list context, returns
643 ($package, $filename, $line) = caller;
645 With EXPR, it returns some extra information that the debugger uses to
646 print a stack trace. The value of EXPR indicates how many call frames
647 to go back before the current one.
650 ($package, $filename, $line, $subroutine, $hasargs,
653 $wantarray, $evaltext, $is_require, $hints, $bitmask, $hinthash)
656 Here $subroutine may be C<(eval)> if the frame is not a subroutine
657 call, but an C<eval>. In such a case additional elements $evaltext and
658 C<$is_require> are set: C<$is_require> is true if the frame is created by a
659 C<require> or C<use> statement, $evaltext contains the text of the
660 C<eval EXPR> statement. In particular, for an C<eval BLOCK> statement,
661 $subroutine is C<(eval)>, but $evaltext is undefined. (Note also that
662 each C<use> statement creates a C<require> frame inside an C<eval EXPR>
663 frame.) $subroutine may also be C<(unknown)> if this particular
664 subroutine happens to have been deleted from the symbol table.
665 C<$hasargs> is true if a new instance of C<@_> was set up for the frame.
666 C<$hints> and C<$bitmask> contain pragmatic hints that the caller was
667 compiled with. The C<$hints> and C<$bitmask> values are subject to change
668 between versions of Perl, and are not meant for external use.
670 C<$hinthash> is a reference to a hash containing the value of C<%^H> when the
671 caller was compiled, or C<undef> if C<%^H> was empty. Do not modify the values
672 of this hash, as they are the actual values stored in the optree.
674 Furthermore, when called from within the DB package, caller returns more
675 detailed information: it sets the list variable C<@DB::args> to be the
676 arguments with which the subroutine was invoked.
678 Be aware that the optimizer might have optimized call frames away before
679 C<caller> had a chance to get the information. That means that C<caller(N)>
680 might not return information about the call frame you expect it to, for
681 C<< N > 1 >>. In particular, C<@DB::args> might have information from the
682 previous time C<caller> was called.
684 Also be aware that setting C<@DB::args> is I<best effort>, intended for
685 debugging or generating backtraces, and should not be relied upon. In
686 particular, as C<@_> contains aliases to the caller's arguments, Perl does
687 not take a copy of C<@_>, so C<@DB::args> will contain modifications the
688 subroutine makes to C<@_> or its contents, not the original values at call
689 time. C<@DB::args>, like C<@_>, does not hold explicit references to its
690 elements, so under certain cases its elements may have become freed and
691 reallocated for other variables or temporary values. Finally, a side effect
692 of the current implementation means that the effects of C<shift @_> can
693 I<normally> be undone (but not C<pop @_> or other splicing, and not if a
694 reference to C<@_> has been taken, and subject to the caveat about reallocated
695 elements), so C<@DB::args> is actually a hybrid of the current state and
696 initial state of C<@_>. Buyer beware.
703 =item chdir FILEHANDLE
705 =item chdir DIRHANDLE
709 Changes the working directory to EXPR, if possible. If EXPR is omitted,
710 changes to the directory specified by C<$ENV{HOME}>, if set; if not,
711 changes to the directory specified by C<$ENV{LOGDIR}>. (Under VMS, the
712 variable C<$ENV{SYS$LOGIN}> is also checked, and used if it is set.) If
713 neither is set, C<chdir> does nothing. It returns true on success,
714 false otherwise. See the example under C<die>.
716 On systems that support fchdir(2), you may pass a filehandle or
717 directory handle as argument. On systems that don't support fchdir(2),
718 passing handles raises an exception.
721 X<chmod> X<permission> X<mode>
723 Changes the permissions of a list of files. The first element of the
724 list must be the numerical mode, which should probably be an octal
725 number, and which definitely should I<not> be a string of octal digits:
726 C<0644> is okay, but C<"0644"> is not. Returns the number of files
727 successfully changed. See also L</oct>, if all you have is a string.
729 $cnt = chmod 0755, "foo", "bar";
730 chmod 0755, @executables;
731 $mode = "0644"; chmod $mode, "foo"; # !!! sets mode to
733 $mode = "0644"; chmod oct($mode), "foo"; # this is better
734 $mode = 0644; chmod $mode, "foo"; # this is best
736 On systems that support fchmod(2), you may pass filehandles among the
737 files. On systems that don't support fchmod(2), passing filehandles raises
738 an exception. Filehandles must be passed as globs or glob references to be
739 recognized; barewords are considered filenames.
741 open(my $fh, "<", "foo");
742 my $perm = (stat $fh)[2] & 07777;
743 chmod($perm | 0600, $fh);
745 You can also import the symbolic C<S_I*> constants from the C<Fcntl>
748 use Fcntl qw( :mode );
749 chmod S_IRWXU|S_IRGRP|S_IXGRP|S_IROTH|S_IXOTH, @executables;
750 # Identical to the chmod 0755 of the example above.
753 X<chomp> X<INPUT_RECORD_SEPARATOR> X<$/> X<newline> X<eol>
759 This safer version of L</chop> removes any trailing string
760 that corresponds to the current value of C<$/> (also known as
761 $INPUT_RECORD_SEPARATOR in the C<English> module). It returns the total
762 number of characters removed from all its arguments. It's often used to
763 remove the newline from the end of an input record when you're worried
764 that the final record may be missing its newline. When in paragraph
765 mode (C<$/ = "">), it removes all trailing newlines from the string.
766 When in slurp mode (C<$/ = undef>) or fixed-length record mode (C<$/> is
767 a reference to an integer or the like, see L<perlvar>) chomp() won't
769 If VARIABLE is omitted, it chomps C<$_>. Example:
772 chomp; # avoid \n on last field
777 If VARIABLE is a hash, it chomps the hash's values, but not its keys.
779 You can actually chomp anything that's an lvalue, including an assignment:
782 chomp($answer = <STDIN>);
784 If you chomp a list, each element is chomped, and the total number of
785 characters removed is returned.
787 Note that parentheses are necessary when you're chomping anything
788 that is not a simple variable. This is because C<chomp $cwd = `pwd`;>
789 is interpreted as C<(chomp $cwd) = `pwd`;>, rather than as
790 C<chomp( $cwd = `pwd` )> which you might expect. Similarly,
791 C<chomp $a, $b> is interpreted as C<chomp($a), $b> rather than
801 Chops off the last character of a string and returns the character
802 chopped. It is much more efficient than C<s/.$//s> because it neither
803 scans nor copies the string. If VARIABLE is omitted, chops C<$_>.
804 If VARIABLE is a hash, it chops the hash's values, but not its keys.
806 You can actually chop anything that's an lvalue, including an assignment.
808 If you chop a list, each element is chopped. Only the value of the
809 last C<chop> is returned.
811 Note that C<chop> returns the last character. To return all but the last
812 character, use C<substr($string, 0, -1)>.
817 X<chown> X<owner> X<user> X<group>
819 Changes the owner (and group) of a list of files. The first two
820 elements of the list must be the I<numeric> uid and gid, in that
821 order. A value of -1 in either position is interpreted by most
822 systems to leave that value unchanged. Returns the number of files
823 successfully changed.
825 $cnt = chown $uid, $gid, 'foo', 'bar';
826 chown $uid, $gid, @filenames;
828 On systems that support fchown(2), you may pass filehandles among the
829 files. On systems that don't support fchown(2), passing filehandles raises
830 an exception. Filehandles must be passed as globs or glob references to be
831 recognized; barewords are considered filenames.
833 Here's an example that looks up nonnumeric uids in the passwd file:
836 chomp($user = <STDIN>);
838 chomp($pattern = <STDIN>);
840 ($login,$pass,$uid,$gid) = getpwnam($user)
841 or die "$user not in passwd file";
843 @ary = glob($pattern); # expand filenames
844 chown $uid, $gid, @ary;
846 On most systems, you are not allowed to change the ownership of the
847 file unless you're the superuser, although you should be able to change
848 the group to any of your secondary groups. On insecure systems, these
849 restrictions may be relaxed, but this is not a portable assumption.
850 On POSIX systems, you can detect this condition this way:
852 use POSIX qw(sysconf _PC_CHOWN_RESTRICTED);
853 $can_chown_giveaway = not sysconf(_PC_CHOWN_RESTRICTED);
856 X<chr> X<character> X<ASCII> X<Unicode>
860 Returns the character represented by that NUMBER in the character set.
861 For example, C<chr(65)> is C<"A"> in either ASCII or Unicode, and
862 chr(0x263a) is a Unicode smiley face.
864 Negative values give the Unicode replacement character (chr(0xfffd)),
865 except under the L<bytes> pragma, where the low eight bits of the value
866 (truncated to an integer) are used.
868 If NUMBER is omitted, uses C<$_>.
870 For the reverse, use L</ord>.
872 Note that characters from 128 to 255 (inclusive) are by default
873 internally not encoded as UTF-8 for backward compatibility reasons.
875 See L<perlunicode> for more about Unicode.
877 =item chroot FILENAME
882 This function works like the system call by the same name: it makes the
883 named directory the new root directory for all further pathnames that
884 begin with a C</> by your process and all its children. (It doesn't
885 change your current working directory, which is unaffected.) For security
886 reasons, this call is restricted to the superuser. If FILENAME is
887 omitted, does a C<chroot> to C<$_>.
889 =item close FILEHANDLE
894 Closes the file or pipe associated with the filehandle, flushes the IO
895 buffers, and closes the system file descriptor. Returns true if those
896 operations have succeeded and if no error was reported by any PerlIO
897 layer. Closes the currently selected filehandle if the argument is
900 You don't have to close FILEHANDLE if you are immediately going to do
901 another C<open> on it, because C<open> closes it for you. (See
902 C<open>.) However, an explicit C<close> on an input file resets the line
903 counter (C<$.>), while the implicit close done by C<open> does not.
905 If the filehandle came from a piped open, C<close> returns false if one of
906 the other syscalls involved fails or if its program exits with non-zero
907 status. If the only problem was that the program exited non-zero, C<$!>
908 will be set to C<0>. Closing a pipe also waits for the process executing
909 on the pipe to exit--in case you wish to look at the output of the pipe
910 afterwards--and implicitly puts the exit status value of that command into
911 C<$?> and C<${^CHILD_ERROR_NATIVE}>.
913 Closing the read end of a pipe before the process writing to it at the
914 other end is done writing results in the writer receiving a SIGPIPE. If
915 the other end can't handle that, be sure to read all the data before
920 open(OUTPUT, '|sort >foo') # pipe to sort
921 or die "Can't start sort: $!";
922 #... # print stuff to output
923 close OUTPUT # wait for sort to finish
924 or warn $! ? "Error closing sort pipe: $!"
925 : "Exit status $? from sort";
926 open(INPUT, 'foo') # get sort's results
927 or die "Can't open 'foo' for input: $!";
929 FILEHANDLE may be an expression whose value can be used as an indirect
930 filehandle, usually the real filehandle name.
932 =item closedir DIRHANDLE
935 Closes a directory opened by C<opendir> and returns the success of that
938 =item connect SOCKET,NAME
941 Attempts to connect to a remote socket, just like connect(2).
942 Returns true if it succeeded, false otherwise. NAME should be a
943 packed address of the appropriate type for the socket. See the examples in
944 L<perlipc/"Sockets: Client/Server Communication">.
951 C<continue> is actually a flow control statement rather than a function. If
952 there is a C<continue> BLOCK attached to a BLOCK (typically in a C<while> or
953 C<foreach>), it is always executed just before the conditional is about to
954 be evaluated again, just like the third part of a C<for> loop in C. Thus
955 it can be used to increment a loop variable, even when the loop has been
956 continued via the C<next> statement (which is similar to the C C<continue>
959 C<last>, C<next>, or C<redo> may appear within a C<continue>
960 block; C<last> and C<redo> behave as if they had been executed within
961 the main block. So will C<next>, but since it will execute a C<continue>
962 block, it may be more entertaining.
965 ### redo always comes here
968 ### next always comes here
970 # then back the top to re-check EXPR
972 ### last always comes here
974 Omitting the C<continue> section is equivalent to using an
975 empty one, logically enough, so C<next> goes directly back
976 to check the condition at the top of the loop.
978 If the C<"switch"> feature is enabled, C<continue> is also a
979 function that exits the current C<when> (or C<default>) block and
980 falls through to the next one. See L<feature> and
981 L<perlsyn/"Switch statements"> for more information.
985 X<cos> X<cosine> X<acos> X<arccosine>
989 Returns the cosine of EXPR (expressed in radians). If EXPR is omitted,
990 takes cosine of C<$_>.
992 For the inverse cosine operation, you may use the C<Math::Trig::acos()>
993 function, or use this relation:
995 sub acos { atan2( sqrt(1 - $_[0] * $_[0]), $_[0] ) }
997 =item crypt PLAINTEXT,SALT
998 X<crypt> X<digest> X<hash> X<salt> X<plaintext> X<password>
999 X<decrypt> X<cryptography> X<passwd> X<encrypt>
1001 Creates a digest string exactly like the crypt(3) function in the C
1002 library (assuming that you actually have a version there that has not
1003 been extirpated as a potential munition).
1005 crypt() is a one-way hash function. The PLAINTEXT and SALT is turned
1006 into a short string, called a digest, which is returned. The same
1007 PLAINTEXT and SALT will always return the same string, but there is no
1008 (known) way to get the original PLAINTEXT from the hash. Small
1009 changes in the PLAINTEXT or SALT will result in large changes in the
1012 There is no decrypt function. This function isn't all that useful for
1013 cryptography (for that, look for F<Crypt> modules on your nearby CPAN
1014 mirror) and the name "crypt" is a bit of a misnomer. Instead it is
1015 primarily used to check if two pieces of text are the same without
1016 having to transmit or store the text itself. An example is checking
1017 if a correct password is given. The digest of the password is stored,
1018 not the password itself. The user types in a password that is
1019 crypt()'d with the same salt as the stored digest. If the two digests
1020 match the password is correct.
1022 When verifying an existing digest string you should use the digest as
1023 the salt (like C<crypt($plain, $digest) eq $digest>). The SALT used
1024 to create the digest is visible as part of the digest. This ensures
1025 crypt() will hash the new string with the same salt as the digest.
1026 This allows your code to work with the standard L<crypt|/crypt> and
1027 with more exotic implementations. In other words, do not assume
1028 anything about the returned string itself, or how many bytes in the
1031 Traditionally the result is a string of 13 bytes: two first bytes of
1032 the salt, followed by 11 bytes from the set C<[./0-9A-Za-z]>, and only
1033 the first eight bytes of PLAINTEXT mattered. But alternative
1034 hashing schemes (like MD5), higher level security schemes (like C2),
1035 and implementations on non-Unix platforms may produce different
1038 When choosing a new salt create a random two character string whose
1039 characters come from the set C<[./0-9A-Za-z]> (like C<join '', ('.',
1040 '/', 0..9, 'A'..'Z', 'a'..'z')[rand 64, rand 64]>). This set of
1041 characters is just a recommendation; the characters allowed in
1042 the salt depend solely on your system's crypt library, and Perl can't
1043 restrict what salts C<crypt()> accepts.
1045 Here's an example that makes sure that whoever runs this program knows
1048 $pwd = (getpwuid($<))[1];
1050 system "stty -echo";
1052 chomp($word = <STDIN>);
1056 if (crypt($word, $pwd) ne $pwd) {
1062 Of course, typing in your own password to whoever asks you
1065 The L<crypt|/crypt> function is unsuitable for hashing large quantities
1066 of data, not least of all because you can't get the information
1067 back. Look at the L<Digest> module for more robust algorithms.
1069 If using crypt() on a Unicode string (which I<potentially> has
1070 characters with codepoints above 255), Perl tries to make sense
1071 of the situation by trying to downgrade (a copy of the string)
1072 the string back to an eight-bit byte string before calling crypt()
1073 (on that copy). If that works, good. If not, crypt() dies with
1074 C<Wide character in crypt>.
1079 [This function has been largely superseded by the C<untie> function.]
1081 Breaks the binding between a DBM file and a hash.
1083 =item dbmopen HASH,DBNAME,MASK
1084 X<dbmopen> X<dbm> X<ndbm> X<sdbm> X<gdbm>
1086 [This function has been largely superseded by the C<tie> function.]
1088 This binds a dbm(3), ndbm(3), sdbm(3), gdbm(3), or Berkeley DB file to a
1089 hash. HASH is the name of the hash. (Unlike normal C<open>, the first
1090 argument is I<not> a filehandle, even though it looks like one). DBNAME
1091 is the name of the database (without the F<.dir> or F<.pag> extension if
1092 any). If the database does not exist, it is created with protection
1093 specified by MASK (as modified by the C<umask>). If your system supports
1094 only the older DBM functions, you may make only one C<dbmopen> call in your
1095 program. In older versions of Perl, if your system had neither DBM nor
1096 ndbm, calling C<dbmopen> produced a fatal error; it now falls back to
1099 If you don't have write access to the DBM file, you can only read hash
1100 variables, not set them. If you want to test whether you can write,
1101 either use file tests or try setting a dummy hash entry inside an C<eval>
1104 Note that functions such as C<keys> and C<values> may return huge lists
1105 when used on large DBM files. You may prefer to use the C<each>
1106 function to iterate over large DBM files. Example:
1108 # print out history file offsets
1109 dbmopen(%HIST,'/usr/lib/news/history',0666);
1110 while (($key,$val) = each %HIST) {
1111 print $key, ' = ', unpack('L',$val), "\n";
1115 See also L<AnyDBM_File> for a more general description of the pros and
1116 cons of the various dbm approaches, as well as L<DB_File> for a particularly
1117 rich implementation.
1119 You can control which DBM library you use by loading that library
1120 before you call dbmopen():
1123 dbmopen(%NS_Hist, "$ENV{HOME}/.netscape/history.db")
1124 or die "Can't open netscape history file: $!";
1127 X<defined> X<undef> X<undefined>
1131 Returns a Boolean value telling whether EXPR has a value other than
1132 the undefined value C<undef>. If EXPR is not present, C<$_> is
1135 Many operations return C<undef> to indicate failure, end of file,
1136 system error, uninitialized variable, and other exceptional
1137 conditions. This function allows you to distinguish C<undef> from
1138 other values. (A simple Boolean test will not distinguish among
1139 C<undef>, zero, the empty string, and C<"0">, which are all equally
1140 false.) Note that since C<undef> is a valid scalar, its presence
1141 doesn't I<necessarily> indicate an exceptional condition: C<pop>
1142 returns C<undef> when its argument is an empty array, I<or> when the
1143 element to return happens to be C<undef>.
1145 You may also use C<defined(&func)> to check whether subroutine C<&func>
1146 has ever been defined. The return value is unaffected by any forward
1147 declarations of C<&func>. A subroutine that is not defined
1148 may still be callable: its package may have an C<AUTOLOAD> method that
1149 makes it spring into existence the first time that it is called; see
1152 Use of C<defined> on aggregates (hashes and arrays) is deprecated. It
1153 used to report whether memory for that aggregate has ever been
1154 allocated. This behavior may disappear in future versions of Perl.
1155 You should instead use a simple test for size:
1157 if (@an_array) { print "has array elements\n" }
1158 if (%a_hash) { print "has hash members\n" }
1160 When used on a hash element, it tells you whether the value is defined,
1161 not whether the key exists in the hash. Use L</exists> for the latter
1166 print if defined $switch{'D'};
1167 print "$val\n" while defined($val = pop(@ary));
1168 die "Can't readlink $sym: $!"
1169 unless defined($value = readlink $sym);
1170 sub foo { defined &$bar ? &$bar(@_) : die "No bar"; }
1171 $debugging = 0 unless defined $debugging;
1173 Note: Many folks tend to overuse C<defined>, and then are surprised to
1174 discover that the number C<0> and C<""> (the zero-length string) are, in fact,
1175 defined values. For example, if you say
1179 The pattern match succeeds and C<$1> is defined, although it
1180 matched "nothing". It didn't really fail to match anything. Rather, it
1181 matched something that happened to be zero characters long. This is all
1182 very above-board and honest. When a function returns an undefined value,
1183 it's an admission that it couldn't give you an honest answer. So you
1184 should use C<defined> only when questioning the integrity of what
1185 you're trying to do. At other times, a simple comparison to C<0> or C<""> is
1188 See also L</undef>, L</exists>, L</ref>.
1193 Given an expression that specifies an element or slice of a hash, C<delete>
1194 deletes the specified elements from that hash so that exists() on that element
1195 no longer returns true. Setting a hash element to the undefined value does
1196 not remove its key, but deleting it does; see L</exists>.
1198 It returns the value or values deleted in list context, or the last such
1199 element in scalar context. The return list's length always matches that of
1200 the argument list: deleting non-existent elements returns the undefined value
1201 in their corresponding positions.
1203 delete() may also be used on arrays and array slices, but its behavior is less
1204 straightforward. Although exists() will return false for deleted entries,
1205 deleting array elements never changes indices of existing values; use shift()
1206 or splice() for that. However, if all deleted elements fall at the end of an
1207 array, the array's size shrinks to the position of the highest element that
1208 still tests true for exists(), or to 0 if none do.
1210 B<Be aware> that calling delete on array values is deprecated and likely to
1211 be removed in a future version of Perl.
1213 Deleting from C<%ENV> modifies the environment. Deleting from a hash tied to
1214 a DBM file deletes the entry from the DBM file. Deleting from a C<tied> hash
1215 or array may not necessarily return anything; it depends on the implementation
1216 of the C<tied> package's DELETE method, which may do whatever it pleases.
1218 The C<delete local EXPR> construct localizes the deletion to the current
1219 block at run time. Until the block exits, elements locally deleted
1220 temporarily no longer exist. See L<perlsub/"Localized deletion of elements
1221 of composite types">.
1223 %hash = (foo => 11, bar => 22, baz => 33);
1224 $scalar = delete $hash{foo}; # $scalar is 11
1225 $scalar = delete @hash{qw(foo bar)}; # $scalar is 22
1226 @array = delete @hash{qw(foo bar baz)}; # @array is (undef,undef,33)
1228 The following (inefficiently) deletes all the values of %HASH and @ARRAY:
1230 foreach $key (keys %HASH) {
1234 foreach $index (0 .. $#ARRAY) {
1235 delete $ARRAY[$index];
1240 delete @HASH{keys %HASH};
1242 delete @ARRAY[0 .. $#ARRAY];
1244 But both are slower than assigning the empty list
1245 or undefining %HASH or @ARRAY, which is the customary
1246 way to empty out an aggregate:
1248 %HASH = (); # completely empty %HASH
1249 undef %HASH; # forget %HASH ever existed
1251 @ARRAY = (); # completely empty @ARRAY
1252 undef @ARRAY; # forget @ARRAY ever existed
1254 The EXPR can be arbitrarily complicated provided its
1255 final operation is an element or slice of an aggregate:
1257 delete $ref->[$x][$y]{$key};
1258 delete @{$ref->[$x][$y]}{$key1, $key2, @morekeys};
1260 delete $ref->[$x][$y][$index];
1261 delete @{$ref->[$x][$y]}[$index1, $index2, @moreindices];
1264 X<die> X<throw> X<exception> X<raise> X<$@> X<abort>
1266 Outside an C<eval>, prints the value of LIST to C<STDERR> and
1267 exits with the current value of C<$!> (errno). If C<$!> is C<0>,
1268 exits with the value of C<<< ($? >> 8) >>> (backtick `command`
1269 status). If C<<< ($? >> 8) >>> is C<0>, exits with C<255>. Inside
1270 an C<eval(),> the error message is stuffed into C<$@> and the
1271 C<eval> is terminated with the undefined value. This makes
1272 C<die> the way to raise an exception.
1274 Equivalent examples:
1276 die "Can't cd to spool: $!\n" unless chdir '/usr/spool/news';
1277 chdir '/usr/spool/news' or die "Can't cd to spool: $!\n"
1279 If the last element of LIST does not end in a newline, the current
1280 script line number and input line number (if any) are also printed,
1281 and a newline is supplied. Note that the "input line number" (also
1282 known as "chunk") is subject to whatever notion of "line" happens to
1283 be currently in effect, and is also available as the special variable
1284 C<$.>. See L<perlvar/"$/"> and L<perlvar/"$.">.
1286 Hint: sometimes appending C<", stopped"> to your message will cause it
1287 to make better sense when the string C<"at foo line 123"> is appended.
1288 Suppose you are running script "canasta".
1290 die "/etc/games is no good";
1291 die "/etc/games is no good, stopped";
1293 produce, respectively
1295 /etc/games is no good at canasta line 123.
1296 /etc/games is no good, stopped at canasta line 123.
1298 See also exit(), warn(), and the Carp module.
1300 If the output is empty and C<$@> already contains a value (typically from a
1301 previous eval) that value is reused after appending C<"\t...propagated">.
1302 This is useful for propagating exceptions:
1305 die unless $@ =~ /Expected exception/;
1307 If the output is empty and C<$@> contains an object reference that has a
1308 C<PROPAGATE> method, that method will be called with additional file
1309 and line number parameters. The return value replaces the value in
1310 C<$@>. i.e., as if C<< $@ = eval { $@->PROPAGATE(__FILE__, __LINE__) }; >>
1313 If C<$@> is empty then the string C<"Died"> is used.
1315 You can also call C<die> with a reference argument, and if this is trapped
1316 within an C<eval>, C<$@> contains that reference. This permits more
1317 elaborate exception handling using objects that maintain arbitrary state
1318 about the exception. Such a scheme is sometimes preferable to matching
1319 particular string values of C<$@> with regular expressions. Because C<$@>
1320 is a global variable and C<eval> may be used within object implementations,
1321 be careful that analyzing the error object doesn't replace the reference in
1322 the global variable. It's easiest to make a local copy of the reference
1323 before any manipulations. Here's an example:
1325 use Scalar::Util "blessed";
1327 eval { ... ; die Some::Module::Exception->new( FOO => "bar" ) };
1328 if (my $ev_err = $@) {
1329 if (blessed($ev_err) && $ev_err->isa("Some::Module::Exception")) {
1330 # handle Some::Module::Exception
1333 # handle all other possible exceptions
1337 Because Perl stringifies uncaught exception messages before display,
1338 you'll probably want to overload stringification operations on
1339 exception objects. See L<overload> for details about that.
1341 You can arrange for a callback to be run just before the C<die>
1342 does its deed, by setting the C<$SIG{__DIE__}> hook. The associated
1343 handler is called with the error text and can change the error
1344 message, if it sees fit, by calling C<die> again. See
1345 L<perlvar/$SIG{expr}> for details on setting C<%SIG> entries, and
1346 L<"eval BLOCK"> for some examples. Although this feature was
1347 to be run only right before your program was to exit, this is not
1348 currently so: the C<$SIG{__DIE__}> hook is currently called
1349 even inside eval()ed blocks/strings! If one wants the hook to do
1350 nothing in such situations, put
1354 as the first line of the handler (see L<perlvar/$^S>). Because
1355 this promotes strange action at a distance, this counterintuitive
1356 behavior may be fixed in a future release.
1361 Not really a function. Returns the value of the last command in the
1362 sequence of commands indicated by BLOCK. When modified by the C<while> or
1363 C<until> loop modifier, executes the BLOCK once before testing the loop
1364 condition. (On other statements the loop modifiers test the conditional
1367 C<do BLOCK> does I<not> count as a loop, so the loop control statements
1368 C<next>, C<last>, or C<redo> cannot be used to leave or restart the block.
1369 See L<perlsyn> for alternative strategies.
1371 =item do SUBROUTINE(LIST)
1374 This form of subroutine call is deprecated. See L<perlsub>.
1379 Uses the value of EXPR as a filename and executes the contents of the
1380 file as a Perl script.
1388 except that it's more efficient and concise, keeps track of the current
1389 filename for error messages, searches the @INC directories, and updates
1390 C<%INC> if the file is found. See L<perlvar/Predefined Names> for these
1391 variables. It also differs in that code evaluated with C<do FILENAME>
1392 cannot see lexicals in the enclosing scope; C<eval STRING> does. It's the
1393 same, however, in that it does reparse the file every time you call it,
1394 so you probably don't want to do this inside a loop.
1396 If C<do> cannot read the file, it returns undef and sets C<$!> to the
1397 error. If C<do> can read the file but cannot compile it, it
1398 returns undef and sets an error message in C<$@>. If the file is
1399 successfully compiled, C<do> returns the value of the last expression
1402 Inclusion of library modules is better done with the
1403 C<use> and C<require> operators, which also do automatic error checking
1404 and raise an exception if there's a problem.
1406 You might like to use C<do> to read in a program configuration
1407 file. Manual error checking can be done this way:
1409 # read in config files: system first, then user
1410 for $file ("/share/prog/defaults.rc",
1411 "$ENV{HOME}/.someprogrc")
1413 unless ($return = do $file) {
1414 warn "couldn't parse $file: $@" if $@;
1415 warn "couldn't do $file: $!" unless defined $return;
1416 warn "couldn't run $file" unless $return;
1421 X<dump> X<core> X<undump>
1425 This function causes an immediate core dump. See also the B<-u>
1426 command-line switch in L<perlrun>, which does the same thing.
1427 Primarily this is so that you can use the B<undump> program (not
1428 supplied) to turn your core dump into an executable binary after
1429 having initialized all your variables at the beginning of the
1430 program. When the new binary is executed it will begin by executing
1431 a C<goto LABEL> (with all the restrictions that C<goto> suffers).
1432 Think of it as a goto with an intervening core dump and reincarnation.
1433 If C<LABEL> is omitted, restarts the program from the top.
1435 B<WARNING>: Any files opened at the time of the dump will I<not>
1436 be open any more when the program is reincarnated, with possible
1437 resulting confusion by Perl.
1439 This function is now largely obsolete, mostly because it's very hard to
1440 convert a core file into an executable. That's why you should now invoke
1441 it as C<CORE::dump()>, if you don't want to be warned against a possible
1445 X<each> X<hash, iterator>
1450 When called in list context, returns a 2-element list consisting of the key
1451 and value for the next element of a hash, or the index and value for the
1452 next element of an array, so that you can iterate over it. When called in
1453 scalar context, returns only the key (not the value) in a hash, or the index
1456 Hash entries are returned in an apparently random order. The actual random
1457 order is subject to change in future versions of Perl, but it is
1458 guaranteed to be in the same order as either the C<keys> or C<values>
1459 function would produce on the same (unmodified) hash. Since Perl
1460 5.8.2 the ordering can be different even between different runs of Perl
1461 for security reasons (see L<perlsec/"Algorithmic Complexity Attacks">).
1463 After C<each> has returned all entries from the hash or array, the next
1464 call to C<each> returns the empty list in list context and C<undef> in
1465 scalar context. The next call following that one restarts iteration. Each
1466 hash or array has its own internal iterator, accessed by C<each>, C<keys>,
1467 and C<values>. The iterator is implicitly reset when C<each> has reached
1468 the end as just described; it can be explicitly reset by calling C<keys> or
1469 C<values> on the hash or array. If you add or delete a hash's elements
1470 while iterating over it, entries may be skipped or duplicated--so don't do
1471 that. Exception: It is always safe to delete the item most recently
1472 returned by C<each()>, so the following code works properly:
1474 while (($key, $value) = each %hash) {
1476 delete $hash{$key}; # This is safe
1479 This prints out your environment like the printenv(1) program,
1480 but in a different order:
1482 while (($key,$value) = each %ENV) {
1483 print "$key=$value\n";
1486 See also C<keys>, C<values> and C<sort>.
1488 =item eof FILEHANDLE
1497 Returns 1 if the next read on FILEHANDLE will return end of file, or if
1498 FILEHANDLE is not open. FILEHANDLE may be an expression whose value
1499 gives the real filehandle. (Note that this function actually
1500 reads a character and then C<ungetc>s it, so isn't useful in an
1501 interactive context.) Do not read from a terminal file (or call
1502 C<eof(FILEHANDLE)> on it) after end-of-file is reached. File types such
1503 as terminals may lose the end-of-file condition if you do.
1505 An C<eof> without an argument uses the last file read. Using C<eof()>
1506 with empty parentheses is different. It refers to the pseudo file
1507 formed from the files listed on the command line and accessed via the
1508 C<< <> >> operator. Since C<< <> >> isn't explicitly opened,
1509 as a normal filehandle is, an C<eof()> before C<< <> >> has been
1510 used will cause C<@ARGV> to be examined to determine if input is
1511 available. Similarly, an C<eof()> after C<< <> >> has returned
1512 end-of-file will assume you are processing another C<@ARGV> list,
1513 and if you haven't set C<@ARGV>, will read input from C<STDIN>;
1514 see L<perlop/"I/O Operators">.
1516 In a C<< while (<>) >> loop, C<eof> or C<eof(ARGV)> can be used to
1517 detect the end of each file, C<eof()> will detect the end of only the
1518 last file. Examples:
1520 # reset line numbering on each input file
1522 next if /^\s*#/; # skip comments
1525 close ARGV if eof; # Not eof()!
1528 # insert dashes just before last line of last file
1530 if (eof()) { # check for end of last file
1531 print "--------------\n";
1534 last if eof(); # needed if we're reading from a terminal
1537 Practical hint: you almost never need to use C<eof> in Perl, because the
1538 input operators typically return C<undef> when they run out of data, or if
1542 X<eval> X<try> X<catch> X<evaluate> X<parse> X<execute>
1543 X<error, handling> X<exception, handling>
1549 In the first form, the return value of EXPR is parsed and executed as if it
1550 were a little Perl program. The value of the expression (which is itself
1551 determined within scalar context) is first parsed, and if there weren't any
1552 errors, executed in the lexical context of the current Perl program, so
1553 that any variable settings or subroutine and format definitions remain
1554 afterwards. Note that the value is parsed every time the C<eval> executes.
1555 If EXPR is omitted, evaluates C<$_>. This form is typically used to
1556 delay parsing and subsequent execution of the text of EXPR until run time.
1558 In the second form, the code within the BLOCK is parsed only once--at the
1559 same time the code surrounding the C<eval> itself was parsed--and executed
1560 within the context of the current Perl program. This form is typically
1561 used to trap exceptions more efficiently than the first (see below), while
1562 also providing the benefit of checking the code within BLOCK at compile
1565 The final semicolon, if any, may be omitted from the value of EXPR or within
1568 In both forms, the value returned is the value of the last expression
1569 evaluated inside the mini-program; a return statement may be also used, just
1570 as with subroutines. The expression providing the return value is evaluated
1571 in void, scalar, or list context, depending on the context of the C<eval>
1572 itself. See L</wantarray> for more on how the evaluation context can be
1575 If there is a syntax error or runtime error, or a C<die> statement is
1576 executed, C<eval> returns an undefined value in scalar context
1577 or an empty list in list context, and C<$@> is set to the
1578 error message. If there was no error, C<$@> is guaranteed to be the empty
1579 string. Beware that using C<eval> neither silences Perl from printing
1580 warnings to STDERR, nor does it stuff the text of warning messages into C<$@>.
1581 To do either of those, you have to use the C<$SIG{__WARN__}> facility, or
1582 turn off warnings inside the BLOCK or EXPR using S<C<no warnings 'all'>>.
1583 See L</warn>, L<perlvar>, L<warnings> and L<perllexwarn>.
1585 Note that, because C<eval> traps otherwise-fatal errors, it is useful for
1586 determining whether a particular feature (such as C<socket> or C<symlink>)
1587 is implemented. It is also Perl's exception trapping mechanism, where
1588 the die operator is used to raise exceptions.
1590 If you want to trap errors when loading an XS module, some problems with
1591 the binary interface (such as Perl version skew) may be fatal even with
1592 C<eval> unless C<$ENV{PERL_DL_NONLAZY}> is set. See L<perlrun>.
1594 If the code to be executed doesn't vary, you may use the eval-BLOCK
1595 form to trap run-time errors without incurring the penalty of
1596 recompiling each time. The error, if any, is still returned in C<$@>.
1599 # make divide-by-zero nonfatal
1600 eval { $answer = $a / $b; }; warn $@ if $@;
1602 # same thing, but less efficient
1603 eval '$answer = $a / $b'; warn $@ if $@;
1605 # a compile-time error
1606 eval { $answer = }; # WRONG
1609 eval '$answer ='; # sets $@
1611 Using the C<eval{}> form as an exception trap in libraries does have some
1612 issues. Due to the current arguably broken state of C<__DIE__> hooks, you
1613 may wish not to trigger any C<__DIE__> hooks that user code may have installed.
1614 You can use the C<local $SIG{__DIE__}> construct for this purpose,
1615 as this example shows:
1617 # a private exception trap for divide-by-zero
1618 eval { local $SIG{'__DIE__'}; $answer = $a / $b; };
1621 This is especially significant, given that C<__DIE__> hooks can call
1622 C<die> again, which has the effect of changing their error messages:
1624 # __DIE__ hooks may modify error messages
1626 local $SIG{'__DIE__'} =
1627 sub { (my $x = $_[0]) =~ s/foo/bar/g; die $x };
1628 eval { die "foo lives here" };
1629 print $@ if $@; # prints "bar lives here"
1632 Because this promotes action at a distance, this counterintuitive behavior
1633 may be fixed in a future release.
1635 With an C<eval>, you should be especially careful to remember what's
1636 being looked at when:
1642 eval { $x }; # CASE 4
1644 eval "\$$x++"; # CASE 5
1647 Cases 1 and 2 above behave identically: they run the code contained in
1648 the variable $x. (Although case 2 has misleading double quotes making
1649 the reader wonder what else might be happening (nothing is).) Cases 3
1650 and 4 likewise behave in the same way: they run the code C<'$x'>, which
1651 does nothing but return the value of $x. (Case 4 is preferred for
1652 purely visual reasons, but it also has the advantage of compiling at
1653 compile-time instead of at run-time.) Case 5 is a place where
1654 normally you I<would> like to use double quotes, except that in this
1655 particular situation, you can just use symbolic references instead, as
1658 The assignment to C<$@> occurs before restoration of localised variables,
1659 which means a temporary is required if you want to mask some but not all
1662 # alter $@ on nefarious repugnancy only
1666 local $@; # protect existing $@
1667 eval { test_repugnancy() };
1668 # $@ =~ /nefarious/ and die $@; # DOES NOT WORK
1669 $@ =~ /nefarious/ and $e = $@;
1671 die $e if defined $e
1674 C<eval BLOCK> does I<not> count as a loop, so the loop control statements
1675 C<next>, C<last>, or C<redo> cannot be used to leave or restart the block.
1677 An C<eval ''> executed within the C<DB> package doesn't see the usual
1678 surrounding lexical scope, but rather the scope of the first non-DB piece
1679 of code that called it. You don't normally need to worry about this unless
1680 you are writing a Perl debugger.
1685 =item exec PROGRAM LIST
1687 The C<exec> function executes a system command I<and never returns>;
1688 use C<system> instead of C<exec> if you want it to return. It fails and
1689 returns false only if the command does not exist I<and> it is executed
1690 directly instead of via your system's command shell (see below).
1692 Since it's a common mistake to use C<exec> instead of C<system>, Perl
1693 warns you if there is a following statement that isn't C<die>, C<warn>,
1694 or C<exit> (if C<-w> is set--but you always do that, right?). If you
1695 I<really> want to follow an C<exec> with some other statement, you
1696 can use one of these styles to avoid the warning:
1698 exec ('foo') or print STDERR "couldn't exec foo: $!";
1699 { exec ('foo') }; print STDERR "couldn't exec foo: $!";
1701 If there is more than one argument in LIST, or if LIST is an array
1702 with more than one value, calls execvp(3) with the arguments in LIST.
1703 If there is only one scalar argument or an array with one element in it,
1704 the argument is checked for shell metacharacters, and if there are any,
1705 the entire argument is passed to the system's command shell for parsing
1706 (this is C</bin/sh -c> on Unix platforms, but varies on other platforms).
1707 If there are no shell metacharacters in the argument, it is split into
1708 words and passed directly to C<execvp>, which is more efficient.
1711 exec '/bin/echo', 'Your arguments are: ', @ARGV;
1712 exec "sort $outfile | uniq";
1714 If you don't really want to execute the first argument, but want to lie
1715 to the program you are executing about its own name, you can specify
1716 the program you actually want to run as an "indirect object" (without a
1717 comma) in front of the LIST. (This always forces interpretation of the
1718 LIST as a multivalued list, even if there is only a single scalar in
1721 $shell = '/bin/csh';
1722 exec $shell '-sh'; # pretend it's a login shell
1726 exec {'/bin/csh'} '-sh'; # pretend it's a login shell
1728 When the arguments get executed via the system shell, results are
1729 subject to its quirks and capabilities. See L<perlop/"`STRING`">
1732 Using an indirect object with C<exec> or C<system> is also more
1733 secure. This usage (which also works fine with system()) forces
1734 interpretation of the arguments as a multivalued list, even if the
1735 list had just one argument. That way you're safe from the shell
1736 expanding wildcards or splitting up words with whitespace in them.
1738 @args = ( "echo surprise" );
1740 exec @args; # subject to shell escapes
1742 exec { $args[0] } @args; # safe even with one-arg list
1744 The first version, the one without the indirect object, ran the I<echo>
1745 program, passing it C<"surprise"> an argument. The second version didn't;
1746 it tried to run a program named I<"echo surprise">, didn't find it, and set
1747 C<$?> to a non-zero value indicating failure.
1749 Beginning with v5.6.0, Perl attempts to flush all files opened for
1750 output before the exec, but this may not be supported on some platforms
1751 (see L<perlport>). To be safe, you may need to set C<$|> ($AUTOFLUSH
1752 in English) or call the C<autoflush()> method of C<IO::Handle> on any
1753 open handles to avoid lost output.
1755 Note that C<exec> will not call your C<END> blocks, nor will it invoke
1756 C<DESTROY> methods on your objects.
1759 X<exists> X<autovivification>
1761 Given an expression that specifies an element of a hash, returns true if the
1762 specified element in the hash has ever been initialized, even if the
1763 corresponding value is undefined.
1765 print "Exists\n" if exists $hash{$key};
1766 print "Defined\n" if defined $hash{$key};
1767 print "True\n" if $hash{$key};
1769 exists may also be called on array elements, but its behavior is much less
1770 obvious, and is strongly tied to the use of L</delete> on arrays. B<Be aware>
1771 that calling exists on array values is deprecated and likely to be removed in
1772 a future version of Perl.
1774 print "Exists\n" if exists $array[$index];
1775 print "Defined\n" if defined $array[$index];
1776 print "True\n" if $array[$index];
1778 A hash or array element can be true only if it's defined, and defined if
1779 it exists, but the reverse doesn't necessarily hold true.
1781 Given an expression that specifies the name of a subroutine,
1782 returns true if the specified subroutine has ever been declared, even
1783 if it is undefined. Mentioning a subroutine name for exists or defined
1784 does not count as declaring it. Note that a subroutine that does not
1785 exist may still be callable: its package may have an C<AUTOLOAD>
1786 method that makes it spring into existence the first time that it is
1787 called; see L<perlsub>.
1789 print "Exists\n" if exists &subroutine;
1790 print "Defined\n" if defined &subroutine;
1792 Note that the EXPR can be arbitrarily complicated as long as the final
1793 operation is a hash or array key lookup or subroutine name:
1795 if (exists $ref->{A}->{B}->{$key}) { }
1796 if (exists $hash{A}{B}{$key}) { }
1798 if (exists $ref->{A}->{B}->[$ix]) { }
1799 if (exists $hash{A}{B}[$ix]) { }
1801 if (exists &{$ref->{A}{B}{$key}}) { }
1803 Although the mostly deeply nested array or hash will not spring into
1804 existence just because its existence was tested, any intervening ones will.
1805 Thus C<< $ref->{"A"} >> and C<< $ref->{"A"}->{"B"} >> will spring
1806 into existence due to the existence test for the $key element above.
1807 This happens anywhere the arrow operator is used, including even here:
1810 if (exists $ref->{"Some key"}) { }
1811 print $ref; # prints HASH(0x80d3d5c)
1813 This surprising autovivification in what does not at first--or even
1814 second--glance appear to be an lvalue context may be fixed in a future
1817 Use of a subroutine call, rather than a subroutine name, as an argument
1818 to exists() is an error.
1821 exists &sub(); # Error
1824 X<exit> X<terminate> X<abort>
1828 Evaluates EXPR and exits immediately with that value. Example:
1831 exit 0 if $ans =~ /^[Xx]/;
1833 See also C<die>. If EXPR is omitted, exits with C<0> status. The only
1834 universally recognized values for EXPR are C<0> for success and C<1>
1835 for error; other values are subject to interpretation depending on the
1836 environment in which the Perl program is running. For example, exiting
1837 69 (EX_UNAVAILABLE) from a I<sendmail> incoming-mail filter will cause
1838 the mailer to return the item undelivered, but that's not true everywhere.
1840 Don't use C<exit> to abort a subroutine if there's any chance that
1841 someone might want to trap whatever error happened. Use C<die> instead,
1842 which can be trapped by an C<eval>.
1844 The exit() function does not always exit immediately. It calls any
1845 defined C<END> routines first, but these C<END> routines may not
1846 themselves abort the exit. Likewise any object destructors that need to
1847 be called are called before the real exit. If this is a problem, you
1848 can call C<POSIX:_exit($status)> to avoid END and destructor processing.
1849 See L<perlmod> for details.
1852 X<exp> X<exponential> X<antilog> X<antilogarithm> X<e>
1856 Returns I<e> (the natural logarithm base) to the power of EXPR.
1857 If EXPR is omitted, gives C<exp($_)>.
1859 =item fcntl FILEHANDLE,FUNCTION,SCALAR
1862 Implements the fcntl(2) function. You'll probably have to say
1866 first to get the correct constant definitions. Argument processing and
1867 value returned work just like C<ioctl> below.
1871 fcntl($filehandle, F_GETFL, $packed_return_buffer)
1872 or die "can't fcntl F_GETFL: $!";
1874 You don't have to check for C<defined> on the return from C<fcntl>.
1875 Like C<ioctl>, it maps a C<0> return from the system call into
1876 C<"0 but true"> in Perl. This string is true in boolean context and C<0>
1877 in numeric context. It is also exempt from the normal B<-w> warnings
1878 on improper numeric conversions.
1880 Note that C<fcntl> raises an exception if used on a machine that
1881 doesn't implement fcntl(2). See the Fcntl module or your fcntl(2)
1882 manpage to learn what functions are available on your system.
1884 Here's an example of setting a filehandle named C<REMOTE> to be
1885 non-blocking at the system level. You'll have to negotiate C<$|>
1886 on your own, though.
1888 use Fcntl qw(F_GETFL F_SETFL O_NONBLOCK);
1890 $flags = fcntl(REMOTE, F_GETFL, 0)
1891 or die "Can't get flags for the socket: $!\n";
1893 $flags = fcntl(REMOTE, F_SETFL, $flags | O_NONBLOCK)
1894 or die "Can't set flags for the socket: $!\n";
1896 =item fileno FILEHANDLE
1899 Returns the file descriptor for a filehandle, or undefined if the
1900 filehandle is not open. This is mainly useful for constructing
1901 bitmaps for C<select> and low-level POSIX tty-handling operations.
1902 If FILEHANDLE is an expression, the value is taken as an indirect
1903 filehandle, generally its name.
1905 You can use this to find out whether two handles refer to the
1906 same underlying descriptor:
1908 if (fileno(THIS) == fileno(THAT)) {
1909 print "THIS and THAT are dups\n";
1912 (Filehandles connected to memory objects via new features of C<open> may
1913 return undefined even though they are open.)
1916 =item flock FILEHANDLE,OPERATION
1917 X<flock> X<lock> X<locking>
1919 Calls flock(2), or an emulation of it, on FILEHANDLE. Returns true
1920 for success, false on failure. Produces a fatal error if used on a
1921 machine that doesn't implement flock(2), fcntl(2) locking, or lockf(3).
1922 C<flock> is Perl's portable file locking interface, although it locks
1923 entire files only, not records.
1925 Two potentially non-obvious but traditional C<flock> semantics are
1926 that it waits indefinitely until the lock is granted, and that its locks
1927 B<merely advisory>. Such discretionary locks are more flexible, but offer
1928 fewer guarantees. This means that programs that do not also use C<flock>
1929 may modify files locked with C<flock>. See L<perlport>,
1930 your port's specific documentation, or your system-specific local manpages
1931 for details. It's best to assume traditional behavior if you're writing
1932 portable programs. (But if you're not, you should as always feel perfectly
1933 free to write for your own system's idiosyncrasies (sometimes called
1934 "features"). Slavish adherence to portability concerns shouldn't get
1935 in the way of your getting your job done.)
1937 OPERATION is one of LOCK_SH, LOCK_EX, or LOCK_UN, possibly combined with
1938 LOCK_NB. These constants are traditionally valued 1, 2, 8 and 4, but
1939 you can use the symbolic names if you import them from the Fcntl module,
1940 either individually, or as a group using the ':flock' tag. LOCK_SH
1941 requests a shared lock, LOCK_EX requests an exclusive lock, and LOCK_UN
1942 releases a previously requested lock. If LOCK_NB is bitwise-or'ed with
1943 LOCK_SH or LOCK_EX then C<flock> returns immediately rather than blocking
1944 waiting for the lock; check the return status to see if you got it.
1946 To avoid the possibility of miscoordination, Perl now flushes FILEHANDLE
1947 before locking or unlocking it.
1949 Note that the emulation built with lockf(3) doesn't provide shared
1950 locks, and it requires that FILEHANDLE be open with write intent. These
1951 are the semantics that lockf(3) implements. Most if not all systems
1952 implement lockf(3) in terms of fcntl(2) locking, though, so the
1953 differing semantics shouldn't bite too many people.
1955 Note that the fcntl(2) emulation of flock(3) requires that FILEHANDLE
1956 be open with read intent to use LOCK_SH and requires that it be open
1957 with write intent to use LOCK_EX.
1959 Note also that some versions of C<flock> cannot lock things over the
1960 network; you would need to use the more system-specific C<fcntl> for
1961 that. If you like you can force Perl to ignore your system's flock(2)
1962 function, and so provide its own fcntl(2)-based emulation, by passing
1963 the switch C<-Ud_flock> to the F<Configure> program when you configure
1966 Here's a mailbox appender for BSD systems.
1968 use Fcntl qw(:flock SEEK_END); # import LOCK_* and SEEK_END constants
1972 flock($fh, LOCK_EX) or die "Cannot lock mailbox - $!\n";
1974 # and, in case someone appended while we were waiting...
1975 seek($fh, 0, SEEK_END) or die "Cannot seek - $!\n";
1980 flock($fh, LOCK_UN) or die "Cannot unlock mailbox - $!\n";
1983 open(my $mbox, ">>", "/usr/spool/mail/$ENV{'USER'}")
1984 or die "Can't open mailbox: $!";
1987 print $mbox $msg,"\n\n";
1990 On systems that support a real flock(2), locks are inherited across fork()
1991 calls, whereas those that must resort to the more capricious fcntl(2)
1992 function lose their locks, making it seriously harder to write servers.
1994 See also L<DB_File> for other flock() examples.
1997 X<fork> X<child> X<parent>
1999 Does a fork(2) system call to create a new process running the
2000 same program at the same point. It returns the child pid to the
2001 parent process, C<0> to the child process, or C<undef> if the fork is
2002 unsuccessful. File descriptors (and sometimes locks on those descriptors)
2003 are shared, while everything else is copied. On most systems supporting
2004 fork(), great care has gone into making it extremely efficient (for
2005 example, using copy-on-write technology on data pages), making it the
2006 dominant paradigm for multitasking over the last few decades.
2008 Beginning with v5.6.0, Perl attempts to flush all files opened for
2009 output before forking the child process, but this may not be supported
2010 on some platforms (see L<perlport>). To be safe, you may need to set
2011 C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method of
2012 C<IO::Handle> on any open handles to avoid duplicate output.
2014 If you C<fork> without ever waiting on your children, you will
2015 accumulate zombies. On some systems, you can avoid this by setting
2016 C<$SIG{CHLD}> to C<"IGNORE">. See also L<perlipc> for more examples of
2017 forking and reaping moribund children.
2019 Note that if your forked child inherits system file descriptors like
2020 STDIN and STDOUT that are actually connected by a pipe or socket, even
2021 if you exit, then the remote server (such as, say, a CGI script or a
2022 backgrounded job launched from a remote shell) won't think you're done.
2023 You should reopen those to F</dev/null> if it's any issue.
2028 Declare a picture format for use by the C<write> function. For
2032 Test: @<<<<<<<< @||||| @>>>>>
2033 $str, $%, '$' . int($num)
2037 $num = $cost/$quantity;
2041 See L<perlform> for many details and examples.
2043 =item formline PICTURE,LIST
2046 This is an internal function used by C<format>s, though you may call it,
2047 too. It formats (see L<perlform>) a list of values according to the
2048 contents of PICTURE, placing the output into the format output
2049 accumulator, C<$^A> (or C<$ACCUMULATOR> in English).
2050 Eventually, when a C<write> is done, the contents of
2051 C<$^A> are written to some filehandle. You could also read C<$^A>
2052 and then set C<$^A> back to C<"">. Note that a format typically
2053 does one C<formline> per line of form, but the C<formline> function itself
2054 doesn't care how many newlines are embedded in the PICTURE. This means
2055 that the C<~> and C<~~> tokens treat the entire PICTURE as a single line.
2056 You may therefore need to use multiple formlines to implement a single
2057 record format, just like the C<format> compiler.
2059 Be careful if you put double quotes around the picture, because an C<@>
2060 character may be taken to mean the beginning of an array name.
2061 C<formline> always returns true. See L<perlform> for other examples.
2063 =item getc FILEHANDLE
2064 X<getc> X<getchar> X<character> X<file, read>
2068 Returns the next character from the input file attached to FILEHANDLE,
2069 or the undefined value at end of file or if there was an error (in
2070 the latter case C<$!> is set). If FILEHANDLE is omitted, reads from
2071 STDIN. This is not particularly efficient. However, it cannot be
2072 used by itself to fetch single characters without waiting for the user
2073 to hit enter. For that, try something more like:
2076 system "stty cbreak </dev/tty >/dev/tty 2>&1";
2079 system "stty", '-icanon', 'eol', "\001";
2085 system "stty -cbreak </dev/tty >/dev/tty 2>&1";
2088 system 'stty', 'icanon', 'eol', '^@'; # ASCII NUL
2092 Determination of whether $BSD_STYLE should be set
2093 is left as an exercise to the reader.
2095 The C<POSIX::getattr> function can do this more portably on
2096 systems purporting POSIX compliance. See also the C<Term::ReadKey>
2097 module from your nearest CPAN site; details on CPAN can be found on
2101 X<getlogin> X<login>
2103 This implements the C library function of the same name, which on most
2104 systems returns the current login from F</etc/utmp>, if any. If it
2105 returns the empty string, use C<getpwuid>.
2107 $login = getlogin || getpwuid($<) || "Kilroy";
2109 Do not consider C<getlogin> for authentication: it is not as
2110 secure as C<getpwuid>.
2112 =item getpeername SOCKET
2113 X<getpeername> X<peer>
2115 Returns the packed sockaddr address of other end of the SOCKET connection.
2118 $hersockaddr = getpeername(SOCK);
2119 ($port, $iaddr) = sockaddr_in($hersockaddr);
2120 $herhostname = gethostbyaddr($iaddr, AF_INET);
2121 $herstraddr = inet_ntoa($iaddr);
2126 Returns the current process group for the specified PID. Use
2127 a PID of C<0> to get the current process group for the
2128 current process. Will raise an exception if used on a machine that
2129 doesn't implement getpgrp(2). If PID is omitted, returns process
2130 group of current process. Note that the POSIX version of C<getpgrp>
2131 does not accept a PID argument, so only C<PID==0> is truly portable.
2134 X<getppid> X<parent> X<pid>
2136 Returns the process id of the parent process.
2138 Note for Linux users: on Linux, the C functions C<getpid()> and
2139 C<getppid()> return different values from different threads. In order to
2140 be portable, this behavior is not reflected by the Perl-level function
2141 C<getppid()>, that returns a consistent value across threads. If you want
2142 to call the underlying C<getppid()>, you may use the CPAN module
2145 =item getpriority WHICH,WHO
2146 X<getpriority> X<priority> X<nice>
2148 Returns the current priority for a process, a process group, or a user.
2149 (See C<getpriority(2)>.) Will raise a fatal exception if used on a
2150 machine that doesn't implement getpriority(2).
2153 X<getpwnam> X<getgrnam> X<gethostbyname> X<getnetbyname> X<getprotobyname>
2154 X<getpwuid> X<getgrgid> X<getservbyname> X<gethostbyaddr> X<getnetbyaddr>
2155 X<getprotobynumber> X<getservbyport> X<getpwent> X<getgrent> X<gethostent>
2156 X<getnetent> X<getprotoent> X<getservent> X<setpwent> X<setgrent> X<sethostent>
2157 X<setnetent> X<setprotoent> X<setservent> X<endpwent> X<endgrent> X<endhostent>
2158 X<endnetent> X<endprotoent> X<endservent>
2162 =item gethostbyname NAME
2164 =item getnetbyname NAME
2166 =item getprotobyname NAME
2172 =item getservbyname NAME,PROTO
2174 =item gethostbyaddr ADDR,ADDRTYPE
2176 =item getnetbyaddr ADDR,ADDRTYPE
2178 =item getprotobynumber NUMBER
2180 =item getservbyport PORT,PROTO
2198 =item sethostent STAYOPEN
2200 =item setnetent STAYOPEN
2202 =item setprotoent STAYOPEN
2204 =item setservent STAYOPEN
2218 These routines are the same as their counterparts in the
2219 system C library. In list context, the return values from the
2220 various get routines are as follows:
2222 ($name,$passwd,$uid,$gid,
2223 $quota,$comment,$gcos,$dir,$shell,$expire) = getpw*
2224 ($name,$passwd,$gid,$members) = getgr*
2225 ($name,$aliases,$addrtype,$length,@addrs) = gethost*
2226 ($name,$aliases,$addrtype,$net) = getnet*
2227 ($name,$aliases,$proto) = getproto*
2228 ($name,$aliases,$port,$proto) = getserv*
2230 (If the entry doesn't exist you get an empty list.)
2232 The exact meaning of the $gcos field varies but it usually contains
2233 the real name of the user (as opposed to the login name) and other
2234 information pertaining to the user. Beware, however, that in many
2235 system users are able to change this information and therefore it
2236 cannot be trusted and therefore the $gcos is tainted (see
2237 L<perlsec>). The $passwd and $shell, user's encrypted password and
2238 login shell, are also tainted, because of the same reason.
2240 In scalar context, you get the name, unless the function was a
2241 lookup by name, in which case you get the other thing, whatever it is.
2242 (If the entry doesn't exist you get the undefined value.) For example:
2244 $uid = getpwnam($name);
2245 $name = getpwuid($num);
2247 $gid = getgrnam($name);
2248 $name = getgrgid($num);
2252 In I<getpw*()> the fields $quota, $comment, and $expire are special
2253 in that they are unsupported on many systems. If the
2254 $quota is unsupported, it is an empty scalar. If it is supported, it
2255 usually encodes the disk quota. If the $comment field is unsupported,
2256 it is an empty scalar. If it is supported it usually encodes some
2257 administrative comment about the user. In some systems the $quota
2258 field may be $change or $age, fields that have to do with password
2259 aging. In some systems the $comment field may be $class. The $expire
2260 field, if present, encodes the expiration period of the account or the
2261 password. For the availability and the exact meaning of these fields
2262 in your system, please consult your getpwnam(3) documentation and your
2263 F<pwd.h> file. You can also find out from within Perl what your
2264 $quota and $comment fields mean and whether you have the $expire field
2265 by using the C<Config> module and the values C<d_pwquota>, C<d_pwage>,
2266 C<d_pwchange>, C<d_pwcomment>, and C<d_pwexpire>. Shadow password
2267 files are supported only if your vendor has implemented them in the
2268 intuitive fashion that calling the regular C library routines gets the
2269 shadow versions if you're running under privilege or if there exists
2270 the shadow(3) functions as found in System V (this includes Solaris
2271 and Linux.) Those systems that implement a proprietary shadow password
2272 facility are unlikely to be supported.
2274 The $members value returned by I<getgr*()> is a space separated list of
2275 the login names of the members of the group.
2277 For the I<gethost*()> functions, if the C<h_errno> variable is supported in
2278 C, it will be returned to you via C<$?> if the function call fails. The
2279 C<@addrs> value returned by a successful call is a list of raw
2280 addresses returned by the corresponding library call. In the
2281 Internet domain, each address is four bytes long; you can unpack it
2282 by saying something like:
2284 ($a,$b,$c,$d) = unpack('W4',$addr[0]);
2286 The Socket library makes this slightly easier:
2289 $iaddr = inet_aton("127.1"); # or whatever address
2290 $name = gethostbyaddr($iaddr, AF_INET);
2292 # or going the other way
2293 $straddr = inet_ntoa($iaddr);
2295 In the opposite way, to resolve a hostname to the IP address
2299 $packed_ip = gethostbyname("www.perl.org");
2300 if (defined $packed_ip) {
2301 $ip_address = inet_ntoa($packed_ip);
2304 Make sure <gethostbyname()> is called in SCALAR context and that
2305 its return value is checked for definedness.
2307 If you get tired of remembering which element of the return list
2308 contains which return value, by-name interfaces are provided
2309 in standard modules: C<File::stat>, C<Net::hostent>, C<Net::netent>,
2310 C<Net::protoent>, C<Net::servent>, C<Time::gmtime>, C<Time::localtime>,
2311 and C<User::grent>. These override the normal built-ins, supplying
2312 versions that return objects with the appropriate names
2313 for each field. For example:
2317 $is_his = (stat($filename)->uid == pwent($whoever)->uid);
2319 Even though it looks like they're the same method calls (uid),
2320 they aren't, because a C<File::stat> object is different from
2321 a C<User::pwent> object.
2323 =item getsockname SOCKET
2326 Returns the packed sockaddr address of this end of the SOCKET connection,
2327 in case you don't know the address because you have several different
2328 IPs that the connection might have come in on.
2331 $mysockaddr = getsockname(SOCK);
2332 ($port, $myaddr) = sockaddr_in($mysockaddr);
2333 printf "Connect to %s [%s]\n",
2334 scalar gethostbyaddr($myaddr, AF_INET),
2337 =item getsockopt SOCKET,LEVEL,OPTNAME
2340 Queries the option named OPTNAME associated with SOCKET at a given LEVEL.
2341 Options may exist at multiple protocol levels depending on the socket
2342 type, but at least the uppermost socket level SOL_SOCKET (defined in the
2343 C<Socket> module) will exist. To query options at another level the
2344 protocol number of the appropriate protocol controlling the option
2345 should be supplied. For example, to indicate that an option is to be
2346 interpreted by the TCP protocol, LEVEL should be set to the protocol
2347 number of TCP, which you can get using C<getprotobyname>.
2349 The function returns a packed string representing the requested socket
2350 option, or C<undef> on error, with the reason for the error placed in
2351 C<$!>). Just what is in the packed string depends on LEVEL and OPTNAME;
2352 consult getsockopt(2) for details. A common case is that the option is an
2353 integer, in which case the result is a packed integer, which you can decode
2354 using C<unpack> with the C<i> (or C<I>) format.
2356 An example to test whether Nagle's algorithm is turned on on a socket:
2358 use Socket qw(:all);
2360 defined(my $tcp = getprotobyname("tcp"))
2361 or die "Could not determine the protocol number for tcp";
2362 # my $tcp = IPPROTO_TCP; # Alternative
2363 my $packed = getsockopt($socket, $tcp, TCP_NODELAY)
2364 or die "getsockopt TCP_NODELAY: $!";
2365 my $nodelay = unpack("I", $packed);
2366 print "Nagle's algorithm is turned ", $nodelay ? "off\n" : "on\n";
2370 X<glob> X<wildcard> X<filename, expansion> X<expand>
2374 In list context, returns a (possibly empty) list of filename expansions on
2375 the value of EXPR such as the standard Unix shell F</bin/csh> would do. In
2376 scalar context, glob iterates through such filename expansions, returning
2377 undef when the list is exhausted. This is the internal function
2378 implementing the C<< <*.c> >> operator, but you can use it directly. If
2379 EXPR is omitted, C<$_> is used. The C<< <*.c> >> operator is discussed in
2380 more detail in L<perlop/"I/O Operators">.
2382 Note that C<glob> splits its arguments on whitespace and treats
2383 each segment as separate pattern. As such, C<glob("*.c *.h")>
2384 matches all files with a F<.c> or F<.h> extension. The expression
2385 C<glob(".* *")> matchs all files in the current working directory.
2387 If non-empty braces are the only wildcard characters used in the
2388 C<glob>, no filenames are matched, but potentially many strings
2389 are returned. For example, this produces nine strings, one for
2390 each pairing of fruits and colors:
2392 @many = glob "{apple,tomato,cherry}={green,yellow,red}";
2394 Beginning with v5.6.0, this operator is implemented using the standard
2395 C<File::Glob> extension. See L<File::Glob> for details, including
2396 C<bsd_glob> which does not treat whitespace as a pattern separator.
2399 X<gmtime> X<UTC> X<Greenwich>
2403 Works just like L<localtime> but the returned values are
2404 localized for the standard Greenwich time zone.
2406 Note: when called in list context, $isdst, the last value
2407 returned by gmtime is always C<0>. There is no
2408 Daylight Saving Time in GMT.
2410 See L<perlport/gmtime> for portability concerns.
2413 X<goto> X<jump> X<jmp>
2419 The C<goto-LABEL> form finds the statement labeled with LABEL and
2420 resumes execution there. It can't be used to get out of a block or
2421 subroutine given to C<sort>. It can be used to go almost anywhere
2422 else within the dynamic scope, including out of subroutines, but it's
2423 usually better to use some other construct such as C<last> or C<die>.
2424 The author of Perl has never felt the need to use this form of C<goto>
2425 (in Perl, that is; C is another matter). (The difference is that C
2426 does not offer named loops combined with loop control. Perl does, and
2427 this replaces most structured uses of C<goto> in other languages.)
2429 The C<goto-EXPR> form expects a label name, whose scope will be resolved
2430 dynamically. This allows for computed C<goto>s per FORTRAN, but isn't
2431 necessarily recommended if you're optimizing for maintainability:
2433 goto ("FOO", "BAR", "GLARCH")[$i];
2435 Use of C<goto-LABEL> or C<goto-EXPR> to jump into a construct is
2436 deprecated and will issue a warning. Even then, it may not be used to
2437 go into any construct that requires initialization, such as a
2438 subroutine or a C<foreach> loop. It also can't be used to go into a
2439 construct that is optimized away.
2441 The C<goto-&NAME> form is quite different from the other forms of
2442 C<goto>. In fact, it isn't a goto in the normal sense at all, and
2443 doesn't have the stigma associated with other gotos. Instead, it
2444 exits the current subroutine (losing any changes set by local()) and
2445 immediately calls in its place the named subroutine using the current
2446 value of @_. This is used by C<AUTOLOAD> subroutines that wish to
2447 load another subroutine and then pretend that the other subroutine had
2448 been called in the first place (except that any modifications to C<@_>
2449 in the current subroutine are propagated to the other subroutine.)
2450 After the C<goto>, not even C<caller> will be able to tell that this
2451 routine was called first.
2453 NAME needn't be the name of a subroutine; it can be a scalar variable
2454 containing a code reference, or a block that evaluates to a code
2457 =item grep BLOCK LIST
2460 =item grep EXPR,LIST
2462 This is similar in spirit to, but not the same as, grep(1) and its
2463 relatives. In particular, it is not limited to using regular expressions.
2465 Evaluates the BLOCK or EXPR for each element of LIST (locally setting
2466 C<$_> to each element) and returns the list value consisting of those
2467 elements for which the expression evaluated to true. In scalar
2468 context, returns the number of times the expression was true.
2470 @foo = grep(!/^#/, @bar); # weed out comments
2474 @foo = grep {!/^#/} @bar; # weed out comments
2476 Note that C<$_> is an alias to the list value, so it can be used to
2477 modify the elements of the LIST. While this is useful and supported,
2478 it can cause bizarre results if the elements of LIST are not variables.
2479 Similarly, grep returns aliases into the original list, much as a for
2480 loop's index variable aliases the list elements. That is, modifying an
2481 element of a list returned by grep (for example, in a C<foreach>, C<map>
2482 or another C<grep>) actually modifies the element in the original list.
2483 This is usually something to be avoided when writing clear code.
2485 If C<$_> is lexical in the scope where the C<grep> appears (because it has
2486 been declared with C<my $_>) then, in addition to being locally aliased to
2487 the list elements, C<$_> keeps being lexical inside the block; i.e., it
2488 can't be seen from the outside, avoiding any potential side-effects.
2490 See also L</map> for a list composed of the results of the BLOCK or EXPR.
2493 X<hex> X<hexadecimal>
2497 Interprets EXPR as a hex string and returns the corresponding value.
2498 (To convert strings that might start with either C<0>, C<0x>, or C<0b>, see
2499 L</oct>.) If EXPR is omitted, uses C<$_>.
2501 print hex '0xAf'; # prints '175'
2502 print hex 'aF'; # same
2504 Hex strings may only represent integers. Strings that would cause
2505 integer overflow trigger a warning. Leading whitespace is not stripped,
2506 unlike oct(). To present something as hex, look into L</printf>,
2507 L</sprintf>, or L</unpack>.
2512 There is no builtin C<import> function. It is just an ordinary
2513 method (subroutine) defined (or inherited) by modules that wish to export
2514 names to another module. The C<use> function calls the C<import> method
2515 for the package used. See also L</use>, L<perlmod>, and L<Exporter>.
2517 =item index STR,SUBSTR,POSITION
2518 X<index> X<indexOf> X<InStr>
2520 =item index STR,SUBSTR
2522 The index function searches for one string within another, but without
2523 the wildcard-like behavior of a full regular-expression pattern match.
2524 It returns the position of the first occurrence of SUBSTR in STR at
2525 or after POSITION. If POSITION is omitted, starts searching from the
2526 beginning of the string. POSITION before the beginning of the string
2527 or after its end is treated as if it were the beginning or the end,
2528 respectively. POSITION and the return value are based at C<0> (or whatever
2529 you've set the C<$[> variable to--but don't do that). If the substring
2530 is not found, C<index> returns one less than the base, ordinarily C<-1>.
2533 X<int> X<integer> X<truncate> X<trunc> X<floor>
2537 Returns the integer portion of EXPR. If EXPR is omitted, uses C<$_>.
2538 You should not use this function for rounding: one because it truncates
2539 towards C<0>, and two because machine representations of floating-point
2540 numbers can sometimes produce counterintuitive results. For example,
2541 C<int(-6.725/0.025)> produces -268 rather than the correct -269; that's
2542 because it's really more like -268.99999999999994315658 instead. Usually,
2543 the C<sprintf>, C<printf>, or the C<POSIX::floor> and C<POSIX::ceil>
2544 functions will serve you better than will int().
2546 =item ioctl FILEHANDLE,FUNCTION,SCALAR
2549 Implements the ioctl(2) function. You'll probably first have to say
2551 require "sys/ioctl.ph"; # probably in $Config{archlib}/sys/ioctl.ph
2553 to get the correct function definitions. If F<sys/ioctl.ph> doesn't
2554 exist or doesn't have the correct definitions you'll have to roll your
2555 own, based on your C header files such as F<< <sys/ioctl.h> >>.
2556 (There is a Perl script called B<h2ph> that comes with the Perl kit that
2557 may help you in this, but it's nontrivial.) SCALAR will be read and/or
2558 written depending on the FUNCTION; a C pointer to the string value of SCALAR
2559 will be passed as the third argument of the actual C<ioctl> call. (If SCALAR
2560 has no string value but does have a numeric value, that value will be
2561 passed rather than a pointer to the string value. To guarantee this to be
2562 true, add a C<0> to the scalar before using it.) The C<pack> and C<unpack>
2563 functions may be needed to manipulate the values of structures used by
2566 The return value of C<ioctl> (and C<fcntl>) is as follows:
2568 if OS returns: then Perl returns:
2570 0 string "0 but true"
2571 anything else that number
2573 Thus Perl returns true on success and false on failure, yet you can
2574 still easily determine the actual value returned by the operating
2577 $retval = ioctl(...) || -1;
2578 printf "System returned %d\n", $retval;
2580 The special string C<"0 but true"> is exempt from B<-w> complaints
2581 about improper numeric conversions.
2583 =item join EXPR,LIST
2586 Joins the separate strings of LIST into a single string with fields
2587 separated by the value of EXPR, and returns that new string. Example:
2589 $rec = join(':', $login,$passwd,$uid,$gid,$gcos,$home,$shell);
2591 Beware that unlike C<split>, C<join> doesn't take a pattern as its
2592 first argument. Compare L</split>.
2599 Returns a list consisting of all the keys of the named hash, or the indices
2600 of an array. (In scalar context, returns the number of keys or indices.)
2602 The keys of a hash are returned in an apparently random order. The actual
2603 random order is subject to change in future versions of Perl, but it
2604 is guaranteed to be the same order as either the C<values> or C<each>
2605 function produces (given that the hash has not been modified). Since
2606 Perl 5.8.1 the ordering is different even between different runs of
2607 Perl for security reasons (see L<perlsec/"Algorithmic Complexity
2610 As a side effect, calling keys() resets the HASH or ARRAY's internal iterator
2611 (see L</each>). In particular, calling keys() in void context resets
2612 the iterator with no other overhead.
2614 Here is yet another way to print your environment:
2617 @values = values %ENV;
2619 print pop(@keys), '=', pop(@values), "\n";
2622 or how about sorted by key:
2624 foreach $key (sort(keys %ENV)) {
2625 print $key, '=', $ENV{$key}, "\n";
2628 The returned values are copies of the original keys in the hash, so
2629 modifying them will not affect the original hash. Compare L</values>.
2631 To sort a hash by value, you'll need to use a C<sort> function.
2632 Here's a descending numeric sort of a hash by its values:
2634 foreach $key (sort { $hash{$b} <=> $hash{$a} } keys %hash) {
2635 printf "%4d %s\n", $hash{$key}, $key;
2638 Used as an lvalue, C<keys> allows you to increase the number of hash buckets
2639 allocated for the given hash. This can gain you a measure of efficiency if
2640 you know the hash is going to get big. (This is similar to pre-extending
2641 an array by assigning a larger number to $#array.) If you say
2645 then C<%hash> will have at least 200 buckets allocated for it--256 of them,
2646 in fact, since it rounds up to the next power of two. These
2647 buckets will be retained even if you do C<%hash = ()>, use C<undef
2648 %hash> if you want to free the storage while C<%hash> is still in scope.
2649 You can't shrink the number of buckets allocated for the hash using
2650 C<keys> in this way (but you needn't worry about doing this by accident,
2651 as trying has no effect). C<keys @array> in an lvalue context is a syntax
2654 See also C<each>, C<values> and C<sort>.
2656 =item kill SIGNAL, LIST
2659 Sends a signal to a list of processes. Returns the number of
2660 processes successfully signaled (which is not necessarily the
2661 same as the number actually killed).
2663 $cnt = kill 1, $child1, $child2;
2666 If SIGNAL is zero, no signal is sent to the process, but C<kill>
2667 checks whether it's I<possible> to send a signal to it (that
2668 means, to be brief, that the process is owned by the same user, or we are
2669 the super-user). This is useful to check that a child process is still
2670 alive (even if only as a zombie) and hasn't changed its UID. See
2671 L<perlport> for notes on the portability of this construct.
2673 Unlike in the shell, if SIGNAL is negative, it kills process groups instead
2674 of processes. That means you usually want to use positive not negative signals.
2675 You may also use a signal name in quotes.
2677 The behavior of kill when a I<PROCESS> number is zero or negative depends on
2678 the operating system. For example, on POSIX-conforming systems, zero will
2679 signal the current process group and -1 will signal all processes.
2681 See L<perlipc/"Signals"> for more details.
2688 The C<last> command is like the C<break> statement in C (as used in
2689 loops); it immediately exits the loop in question. If the LABEL is
2690 omitted, the command refers to the innermost enclosing loop. The
2691 C<continue> block, if any, is not executed:
2693 LINE: while (<STDIN>) {
2694 last LINE if /^$/; # exit when done with header
2698 C<last> cannot be used to exit a block that returns a value such as
2699 C<eval {}>, C<sub {}> or C<do {}>, and should not be used to exit
2700 a grep() or map() operation.
2702 Note that a block by itself is semantically identical to a loop
2703 that executes once. Thus C<last> can be used to effect an early
2704 exit out of such a block.
2706 See also L</continue> for an illustration of how C<last>, C<next>, and
2714 Returns a lowercased version of EXPR. This is the internal function
2715 implementing the C<\L> escape in double-quoted strings.
2717 If EXPR is omitted, uses C<$_>.
2719 What gets returned depends on several factors:
2723 =item If C<use bytes> is in effect:
2727 =item On EBCDIC platforms
2729 The results are what the C language system call C<tolower()> returns.
2731 =item On ASCII platforms
2733 The results follow ASCII semantics. Only characters C<A-Z> change, to C<a-z>
2738 =item Otherwise, If EXPR has the UTF8 flag set
2740 If the current package has a subroutine named C<ToLower>, it will be used to
2741 change the case (See L<perlunicode/User-Defined Case Mappings>.)
2742 Otherwise Unicode semantics are used for the case change.
2744 =item Otherwise, if C<use locale> is in effect
2746 Respects current LC_CTYPE locale. See L<perllocale>.
2748 =item Otherwise, if C<use feature 'unicode_strings'> is in effect:
2750 Unicode semantics are used for the case change. Any subroutine named
2751 C<ToLower> will not be used.
2757 =item On EBCDIC platforms
2759 The results are what the C language system call C<tolower()> returns.
2761 =item On ASCII platforms
2763 ASCII semantics are used for the case change. The lowercase of any character
2764 outside the ASCII range is the character itself.
2771 X<lcfirst> X<lowercase>
2775 Returns the value of EXPR with the first character lowercased. This
2776 is the internal function implementing the C<\l> escape in
2777 double-quoted strings.
2779 If EXPR is omitted, uses C<$_>.
2781 This function behaves the same way under various pragma, such as in a locale,
2789 Returns the length in I<characters> of the value of EXPR. If EXPR is
2790 omitted, returns length of C<$_>. If EXPR is undefined, returns C<undef>.
2792 This function cannot be used on an entire array or hash to find out how
2793 many elements these have. For that, use C<scalar @array> and C<scalar keys
2794 %hash>, respectively.
2796 Like all Perl character operations, length() normally deals in logical
2797 characters, not physical bytes. For how many bytes a string encoded as
2798 UTF-8 would take up, use C<length(Encode::encode_utf8(EXPR))> (you'll have
2799 to C<use Encode> first). See L<Encode> and L<perlunicode>.
2801 =item link OLDFILE,NEWFILE
2804 Creates a new filename linked to the old filename. Returns true for
2805 success, false otherwise.
2807 =item listen SOCKET,QUEUESIZE
2810 Does the same thing that the listen(2) system call does. Returns true if
2811 it succeeded, false otherwise. See the example in
2812 L<perlipc/"Sockets: Client/Server Communication">.
2817 You really probably want to be using C<my> instead, because C<local> isn't
2818 what most people think of as "local". See
2819 L<perlsub/"Private Variables via my()"> for details.
2821 A local modifies the listed variables to be local to the enclosing
2822 block, file, or eval. If more than one value is listed, the list must
2823 be placed in parentheses. See L<perlsub/"Temporary Values via local()">
2824 for details, including issues with tied arrays and hashes.
2826 The C<delete local EXPR> construct can also be used to localize the deletion
2827 of array/hash elements to the current block.
2828 See L<perlsub/"Localized deletion of elements of composite types">.
2830 =item localtime EXPR
2831 X<localtime> X<ctime>
2835 Converts a time as returned by the time function to a 9-element list
2836 with the time analyzed for the local time zone. Typically used as
2840 ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) =
2843 All list elements are numeric, and come straight out of the C `struct
2844 tm'. C<$sec>, C<$min>, and C<$hour> are the seconds, minutes, and hours
2845 of the specified time.
2847 C<$mday> is the day of the month, and C<$mon> is the month itself, in
2848 the range C<0..11> with 0 indicating January and 11 indicating December.
2849 This makes it easy to get a month name from a list:
2851 my @abbr = qw( Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec );
2852 print "$abbr[$mon] $mday";
2853 # $mon=9, $mday=18 gives "Oct 18"
2855 C<$year> is the number of years since 1900, not just the last two digits
2856 of the year. That is, C<$year> is C<123> in year 2023. The proper way
2857 to get a 4-digit year is simply:
2861 Otherwise you create non-Y2K-compliant programs--and you wouldn't want
2862 to do that, would you?
2864 To get the last two digits of the year (e.g., '01' in 2001) do:
2866 $year = sprintf("%02d", $year % 100);
2868 C<$wday> is the day of the week, with 0 indicating Sunday and 3 indicating
2869 Wednesday. C<$yday> is the day of the year, in the range C<0..364>
2870 (or C<0..365> in leap years.)
2872 C<$isdst> is true if the specified time occurs during Daylight Saving
2873 Time, false otherwise.
2875 If EXPR is omitted, C<localtime()> uses the current time (as returned
2878 In scalar context, C<localtime()> returns the ctime(3) value:
2880 $now_string = localtime; # e.g., "Thu Oct 13 04:54:34 1994"
2882 This scalar value is B<not> locale dependent but is a Perl builtin. For GMT
2883 instead of local time use the L</gmtime> builtin. See also the
2884 C<Time::Local> module (to convert the second, minutes, hours, ... back to
2885 the integer value returned by time()), and the L<POSIX> module's strftime(3)
2886 and mktime(3) functions.
2888 To get somewhat similar but locale dependent date strings, set up your
2889 locale environment variables appropriately (please see L<perllocale>) and
2892 use POSIX qw(strftime);
2893 $now_string = strftime "%a %b %e %H:%M:%S %Y", localtime;
2894 # or for GMT formatted appropriately for your locale:
2895 $now_string = strftime "%a %b %e %H:%M:%S %Y", gmtime;
2897 Note that the C<%a> and C<%b>, the short forms of the day of the week
2898 and the month of the year, may not necessarily be three characters wide.
2900 See L<perlport/localtime> for portability concerns.
2902 The L<Time::gmtime> and L<Time::localtime> modules provides a convenient,
2903 by-name access mechanism to the gmtime() and localtime() functions,
2906 For a comprehensive date and time representation look at the
2907 L<DateTime> module on CPAN.
2912 This function places an advisory lock on a shared variable, or referenced
2913 object contained in I<THING> until the lock goes out of scope.
2915 lock() is a "weak keyword" : this means that if you've defined a function
2916 by this name (before any calls to it), that function will be called
2917 instead. If you are not under C<use threads::shared> this does nothing.
2918 See L<threads::shared>.
2921 X<log> X<logarithm> X<e> X<ln> X<base>
2925 Returns the natural logarithm (base I<e>) of EXPR. If EXPR is omitted,
2926 returns log of C<$_>. To get the log of another base, use basic algebra:
2927 The base-N log of a number is equal to the natural log of that number
2928 divided by the natural log of N. For example:
2932 return log($n)/log(10);
2935 See also L</exp> for the inverse operation.
2942 Does the same thing as the C<stat> function (including setting the
2943 special C<_> filehandle) but stats a symbolic link instead of the file
2944 the symbolic link points to. If symbolic links are unimplemented on
2945 your system, a normal C<stat> is done. For much more detailed
2946 information, please see the documentation for C<stat>.
2948 If EXPR is omitted, stats C<$_>.
2952 The match operator. See L<perlop/"Regexp Quote-Like Operators">.
2954 =item map BLOCK LIST
2959 Evaluates the BLOCK or EXPR for each element of LIST (locally setting
2960 C<$_> to each element) and returns the list value composed of the
2961 results of each such evaluation. In scalar context, returns the
2962 total number of elements so generated. Evaluates BLOCK or EXPR in
2963 list context, so each element of LIST may produce zero, one, or
2964 more elements in the returned value.
2966 @chars = map(chr, @nums);
2968 translates a list of numbers to the corresponding characters. And
2970 %hash = map { get_a_key_for($_) => $_ } @array;
2972 is just a funny way to write
2976 $hash{get_a_key_for($_)} = $_;
2979 Note that C<$_> is an alias to the list value, so it can be used to
2980 modify the elements of the LIST. While this is useful and supported,
2981 it can cause bizarre results if the elements of LIST are not variables.
2982 Using a regular C<foreach> loop for this purpose would be clearer in
2983 most cases. See also L</grep> for an array composed of those items of
2984 the original list for which the BLOCK or EXPR evaluates to true.
2986 If C<$_> is lexical in the scope where the C<map> appears (because it has
2987 been declared with C<my $_>), then, in addition to being locally aliased to
2988 the list elements, C<$_> keeps being lexical inside the block; that is, it
2989 can't be seen from the outside, avoiding any potential side-effects.
2991 C<{> starts both hash references and blocks, so C<map { ...> could be either
2992 the start of map BLOCK LIST or map EXPR, LIST. Because Perl doesn't look
2993 ahead for the closing C<}> it has to take a guess at which it's dealing with
2994 based on what it finds just after the C<{>. Usually it gets it right, but if it
2995 doesn't it won't realize something is wrong until it gets to the C<}> and
2996 encounters the missing (or unexpected) comma. The syntax error will be
2997 reported close to the C<}>, but you'll need to change something near the C<{>
2998 such as using a unary C<+> to give Perl some help:
3000 %hash = map { "\L$_" => 1 } @array # perl guesses EXPR. wrong
3001 %hash = map { +"\L$_" => 1 } @array # perl guesses BLOCK. right
3002 %hash = map { ("\L$_" => 1) } @array # this also works
3003 %hash = map { lc($_) => 1 } @array # as does this.
3004 %hash = map +( lc($_) => 1 ), @array # this is EXPR and works!
3006 %hash = map ( lc($_), 1 ), @array # evaluates to (1, @array)
3008 or to force an anon hash constructor use C<+{>:
3010 @hashes = map +{ lc($_) => 1 }, @array # EXPR, so needs comma at end
3012 to get a list of anonymous hashes each with only one entry apiece.
3014 =item mkdir FILENAME,MASK
3015 X<mkdir> X<md> X<directory, create>
3017 =item mkdir FILENAME
3021 Creates the directory specified by FILENAME, with permissions
3022 specified by MASK (as modified by C<umask>). If it succeeds it
3023 returns true, otherwise it returns false and sets C<$!> (errno).
3024 If omitted, MASK defaults to 0777. If omitted, FILENAME defaults
3027 In general, it is better to create directories with permissive MASK,
3028 and let the user modify that with their C<umask>, than it is to supply
3029 a restrictive MASK and give the user no way to be more permissive.
3030 The exceptions to this rule are when the file or directory should be
3031 kept private (mail files, for instance). The perlfunc(1) entry on
3032 C<umask> discusses the choice of MASK in more detail.
3034 Note that according to the POSIX 1003.1-1996 the FILENAME may have any
3035 number of trailing slashes. Some operating and filesystems do not get
3036 this right, so Perl automatically removes all trailing slashes to keep
3039 To recursively create a directory structure, look at
3040 the C<mkpath> function of the L<File::Path> module.
3042 =item msgctl ID,CMD,ARG
3045 Calls the System V IPC function msgctl(2). You'll probably have to say
3049 first to get the correct constant definitions. If CMD is C<IPC_STAT>,
3050 then ARG must be a variable that will hold the returned C<msqid_ds>
3051 structure. Returns like C<ioctl>: the undefined value for error,
3052 C<"0 but true"> for zero, or the actual return value otherwise. See also
3053 L<perlipc/"SysV IPC">, C<IPC::SysV>, and C<IPC::Semaphore> documentation.
3055 =item msgget KEY,FLAGS
3058 Calls the System V IPC function msgget(2). Returns the message queue
3059 id, or the undefined value if there is an error. See also
3060 L<perlipc/"SysV IPC"> and C<IPC::SysV> and C<IPC::Msg> documentation.
3062 =item msgrcv ID,VAR,SIZE,TYPE,FLAGS
3065 Calls the System V IPC function msgrcv to receive a message from
3066 message queue ID into variable VAR with a maximum message size of
3067 SIZE. Note that when a message is received, the message type as a
3068 native long integer will be the first thing in VAR, followed by the
3069 actual message. This packing may be opened with C<unpack("l! a*")>.
3070 Taints the variable. Returns true if successful, or false if there is
3071 an error. See also L<perlipc/"SysV IPC">, C<IPC::SysV>, and
3072 C<IPC::SysV::Msg> documentation.
3074 =item msgsnd ID,MSG,FLAGS
3077 Calls the System V IPC function msgsnd to send the message MSG to the
3078 message queue ID. MSG must begin with the native long integer message
3079 type, and be followed by the length of the actual message, and finally
3080 the message itself. This kind of packing can be achieved with
3081 C<pack("l! a*", $type, $message)>. Returns true if successful,
3082 or false if there is an error. See also C<IPC::SysV>
3083 and C<IPC::SysV::Msg> documentation.
3090 =item my EXPR : ATTRS
3092 =item my TYPE EXPR : ATTRS
3094 A C<my> declares the listed variables to be local (lexically) to the
3095 enclosing block, file, or C<eval>. If more than one value is listed,
3096 the list must be placed in parentheses.
3098 The exact semantics and interface of TYPE and ATTRS are still
3099 evolving. TYPE is currently bound to the use of C<fields> pragma,
3100 and attributes are handled using the C<attributes> pragma, or starting
3101 from Perl 5.8.0 also via the C<Attribute::Handlers> module. See
3102 L<perlsub/"Private Variables via my()"> for details, and L<fields>,
3103 L<attributes>, and L<Attribute::Handlers>.
3110 The C<next> command is like the C<continue> statement in C; it starts
3111 the next iteration of the loop:
3113 LINE: while (<STDIN>) {
3114 next LINE if /^#/; # discard comments
3118 Note that if there were a C<continue> block on the above, it would get
3119 executed even on discarded lines. If LABEL is omitted, the command
3120 refers to the innermost enclosing loop.
3122 C<next> cannot be used to exit a block which returns a value such as
3123 C<eval {}>, C<sub {}> or C<do {}>, and should not be used to exit
3124 a grep() or map() operation.
3126 Note that a block by itself is semantically identical to a loop
3127 that executes once. Thus C<next> will exit such a block early.
3129 See also L</continue> for an illustration of how C<last>, C<next>, and
3132 =item no MODULE VERSION LIST
3136 =item no MODULE VERSION
3138 =item no MODULE LIST
3144 See the C<use> function, of which C<no> is the opposite.
3147 X<oct> X<octal> X<hex> X<hexadecimal> X<binary> X<bin>
3151 Interprets EXPR as an octal string and returns the corresponding
3152 value. (If EXPR happens to start off with C<0x>, interprets it as a
3153 hex string. If EXPR starts off with C<0b>, it is interpreted as a
3154 binary string. Leading whitespace is ignored in all three cases.)
3155 The following will handle decimal, binary, octal, and hex in standard
3158 $val = oct($val) if $val =~ /^0/;
3160 If EXPR is omitted, uses C<$_>. To go the other way (produce a number
3161 in octal), use sprintf() or printf():
3163 $dec_perms = (stat("filename"))[2] & 07777;
3164 $oct_perm_str = sprintf "%o", $perms;
3166 The oct() function is commonly used when a string such as C<644> needs
3167 to be converted into a file mode, for example. Although Perl
3168 automatically converts strings into numbers as needed, this automatic
3169 conversion assumes base 10.
3171 Leading white space is ignored without warning, as too are any trailing
3172 non-digits, such as a decimal point (C<oct> only handles non-negative
3173 integers, not negative integers or floating point).
3175 =item open FILEHANDLE,EXPR
3176 X<open> X<pipe> X<file, open> X<fopen>
3178 =item open FILEHANDLE,MODE,EXPR
3180 =item open FILEHANDLE,MODE,EXPR,LIST
3182 =item open FILEHANDLE,MODE,REFERENCE
3184 =item open FILEHANDLE
3186 Opens the file whose filename is given by EXPR, and associates it with
3189 Simple examples to open a file for reading:
3191 open(my $fh, '<', "input.txt") or die $!;
3195 open(my $fh, '>', "output.txt") or die $!;
3197 (The following is a comprehensive reference to open(): for a gentler
3198 introduction you may consider L<perlopentut>.)
3200 If FILEHANDLE is an undefined scalar variable (or array or hash element)
3201 the variable is assigned a reference to a new anonymous filehandle,
3202 otherwise if FILEHANDLE is an expression, its value is used as the name of
3203 the real filehandle wanted. (This is considered a symbolic reference, so
3204 C<use strict 'refs'> should I<not> be in effect.)
3206 If EXPR is omitted, the scalar variable of the same name as the
3207 FILEHANDLE contains the filename. (Note that lexical variables--those
3208 declared with C<my>--will not work for this purpose; so if you're
3209 using C<my>, specify EXPR in your call to open.)
3211 If three or more arguments are specified then the mode of opening and
3212 the filename are separate. If MODE is C<< '<' >> or nothing, the file
3213 is opened for input. If MODE is C<< '>' >>, the file is truncated and
3214 opened for output, being created if necessary. If MODE is C<<< '>>' >>>,
3215 the file is opened for appending, again being created if necessary.
3217 You can put a C<'+'> in front of the C<< '>' >> or C<< '<' >> to
3218 indicate that you want both read and write access to the file; thus
3219 C<< '+<' >> is almost always preferred for read/write updates--the
3220 C<< '+>' >> mode would clobber the file first. You can't usually use
3221 either read-write mode for updating textfiles, since they have
3222 variable length records. See the B<-i> switch in L<perlrun> for a
3223 better approach. The file is created with permissions of C<0666>
3224 modified by the process's C<umask> value.
3226 These various prefixes correspond to the fopen(3) modes of C<'r'>,
3227 C<'r+'>, C<'w'>, C<'w+'>, C<'a'>, and C<'a+'>.
3229 In the two-argument (and one-argument) form of the call, the mode and
3230 filename should be concatenated (in that order), possibly separated by
3231 spaces. You may omit the mode in these forms when that mode is
3234 If the filename begins with C<'|'>, the filename is interpreted as a
3235 command to which output is to be piped, and if the filename ends with a
3236 C<'|'>, the filename is interpreted as a command that pipes output to
3237 us. See L<perlipc/"Using open() for IPC">
3238 for more examples of this. (You are not allowed to C<open> to a command
3239 that pipes both in I<and> out, but see L<IPC::Open2>, L<IPC::Open3>,
3240 and L<perlipc/"Bidirectional Communication with Another Process">
3243 For three or more arguments if MODE is C<'|-'>, the filename is
3244 interpreted as a command to which output is to be piped, and if MODE
3245 is C<'-|'>, the filename is interpreted as a command that pipes
3246 output to us. In the two-argument (and one-argument) form, one should
3247 replace dash (C<'-'>) with the command.
3248 See L<perlipc/"Using open() for IPC"> for more examples of this.
3249 (You are not allowed to C<open> to a command that pipes both in I<and>
3250 out, but see L<IPC::Open2>, L<IPC::Open3>, and
3251 L<perlipc/"Bidirectional Communication"> for alternatives.)
3253 In the form of pipe opens taking three or more arguments, if LIST is specified
3254 (extra arguments after the command name) then LIST becomes arguments
3255 to the command invoked if the platform supports it. The meaning of
3256 C<open> with more than three arguments for non-pipe modes is not yet
3257 defined, but experimental "layers" may give extra LIST arguments
3260 In the two-argument (and one-argument) form, opening C<< '<-' >>
3261 or C<'-'> opens STDIN and opening C<< '>-' >> opens STDOUT.
3263 You may use the three-argument form of open to specify I/O layers
3264 (sometimes referred to as "disciplines") to apply to the handle
3265 that affect how the input and output are processed (see L<open> and
3266 L<PerlIO> for more details). For example:
3268 open(my $fh, "<:encoding(UTF-8)", "filename")
3269 || die "can't open UTF-8 encoded filename: $!";
3271 opens the UTF-8 encoded file containing Unicode characters;
3272 see L<perluniintro>. Note that if layers are specified in the
3273 three-argument form, then default layers stored in ${^OPEN} (see L<perlvar>;
3274 usually set by the B<open> pragma or the switch B<-CioD>) are ignored.
3276 Open returns nonzero on success, the undefined value otherwise. If
3277 the C<open> involved a pipe, the return value happens to be the pid of
3280 If you're running Perl on a system that distinguishes between text
3281 files and binary files, then you should check out L</binmode> for tips
3282 for dealing with this. The key distinction between systems that need
3283 C<binmode> and those that don't is their text file formats. Systems
3284 like Unix, Mac OS, and Plan 9, that end lines with a single
3285 character and encode that character in C as C<"\n"> do not
3286 need C<binmode>. The rest need it.
3288 When opening a file, it's seldom a good idea to continue
3289 if the request failed, so C<open> is frequently used with
3290 C<die>. Even if C<die> won't do what you want (say, in a CGI script,
3291 where you want to format a suitable error message (but there are
3292 modules that can help with that problem)) always check
3293 the return value from opening a file.
3295 As a special case the 3-arg form with a read/write mode and the third
3296 argument being C<undef>:
3298 open(my $tmp, "+>", undef) or die ...
3300 opens a filehandle to an anonymous temporary file. Also using "+<"
3301 works for symmetry, but you really should consider writing something
3302 to the temporary file first. You will need to seek() to do the
3305 Since v5.8.0, Perl has built using PerlIO by default. Unless you've
3306 changed this (i.e., Configure -Uuseperlio), you can open filehandles
3307 directly to Perl scalars via:
3309 open($fh, '>', \$variable) || ..
3311 To (re)open C<STDOUT> or C<STDERR> as an in-memory file, close it first:
3314 open STDOUT, '>', \$variable or die "Can't open STDOUT: $!";
3319 open ARTICLE or die "Can't find article $ARTICLE: $!\n";
3320 while (<ARTICLE>) {...
3322 open(LOG, '>>/usr/spool/news/twitlog'); # (log is reserved)
3323 # if the open fails, output is discarded
3325 open(my $dbase, '+<', 'dbase.mine') # open for update
3326 or die "Can't open 'dbase.mine' for update: $!";
3328 open(my $dbase, '+<dbase.mine') # ditto
3329 or die "Can't open 'dbase.mine' for update: $!";
3331 open(ARTICLE, '-|', "caesar <$article") # decrypt article
3332 or die "Can't start caesar: $!";
3334 open(ARTICLE, "caesar <$article |") # ditto
3335 or die "Can't start caesar: $!";
3337 open(EXTRACT, "|sort >Tmp$$") # $$ is our process id
3338 or die "Can't start sort: $!";
3341 open(MEMORY,'>', \$var)
3342 or die "Can't open memory file: $!";
3343 print MEMORY "foo!\n"; # output will appear in $var
3345 # process argument list of files along with any includes
3347 foreach $file (@ARGV) {
3348 process($file, 'fh00');
3352 my($filename, $input) = @_;
3353 $input++; # this is a string increment
3354 unless (open($input, $filename)) {
3355 print STDERR "Can't open $filename: $!\n";
3360 while (<$input>) { # note use of indirection
3361 if (/^#include "(.*)"/) {
3362 process($1, $input);
3369 See L<perliol> for detailed info on PerlIO.
3371 You may also, in the Bourne shell tradition, specify an EXPR beginning
3372 with C<< '>&' >>, in which case the rest of the string is interpreted
3373 as the name of a filehandle (or file descriptor, if numeric) to be
3374 duped (as C<dup(2)>) and opened. You may use C<&> after C<< > >>,
3375 C<<< >> >>>, C<< < >>, C<< +> >>, C<<< +>> >>>, and C<< +< >>.
3376 The mode you specify should match the mode of the original filehandle.
3377 (Duping a filehandle does not take into account any existing contents
3378 of IO buffers.) If you use the 3-arg form then you can pass either a
3379 number, the name of a filehandle or the normal "reference to a glob".
3381 Here is a script that saves, redirects, and restores C<STDOUT> and
3382 C<STDERR> using various methods:
3385 open my $oldout, ">&STDOUT" or die "Can't dup STDOUT: $!";
3386 open OLDERR, ">&", \*STDERR or die "Can't dup STDERR: $!";
3388 open STDOUT, '>', "foo.out" or die "Can't redirect STDOUT: $!";
3389 open STDERR, ">&STDOUT" or die "Can't dup STDOUT: $!";
3391 select STDERR; $| = 1; # make unbuffered
3392 select STDOUT; $| = 1; # make unbuffered
3394 print STDOUT "stdout 1\n"; # this works for
3395 print STDERR "stderr 1\n"; # subprocesses too
3397 open STDOUT, ">&", $oldout or die "Can't dup \$oldout: $!";
3398 open STDERR, ">&OLDERR" or die "Can't dup OLDERR: $!";
3400 print STDOUT "stdout 2\n";
3401 print STDERR "stderr 2\n";
3403 If you specify C<< '<&=X' >>, where C<X> is a file descriptor number
3404 or a filehandle, then Perl will do an equivalent of C's C<fdopen> of
3405 that file descriptor (and not call C<dup(2)>); this is more
3406 parsimonious of file descriptors. For example:
3408 # open for input, reusing the fileno of $fd
3409 open(FILEHANDLE, "<&=$fd")
3413 open(FILEHANDLE, "<&=", $fd)
3417 # open for append, using the fileno of OLDFH
3418 open(FH, ">>&=", OLDFH)
3422 open(FH, ">>&=OLDFH")
3424 Being parsimonious on filehandles is also useful (besides being
3425 parsimonious) for example when something is dependent on file
3426 descriptors, like for example locking using flock(). If you do just
3427 C<< open(A, '>>&B') >>, the filehandle A will not have the same file
3428 descriptor as B, and therefore flock(A) will not flock(B), and vice
3429 versa. But with C<< open(A, '>>&=B') >> the filehandles will share
3430 the same file descriptor.
3432 Note that if you are using Perls older than 5.8.0, Perl will be using
3433 the standard C libraries' fdopen() to implement the "=" functionality.
3434 On many Unix systems fdopen() fails when file descriptors exceed a
3435 certain value, typically 255. For Perls 5.8.0 and later, PerlIO is
3436 most often the default.
3438 You can see whether Perl has been compiled with PerlIO or not by
3439 running C<perl -V> and looking for C<useperlio=> line. If C<useperlio>
3440 is C<define>, you have PerlIO, otherwise you don't.
3442 If you open a pipe on the command C<'-'>, i.e., either C<'|-'> or C<'-|'>
3443 with 2-arguments (or 1-argument) form of open(), then
3444 there is an implicit fork done, and the return value of open is the pid
3445 of the child within the parent process, and C<0> within the child
3446 process. (Use C<defined($pid)> to determine whether the open was successful.)
3447 The filehandle behaves normally for the parent, but I/O to that
3448 filehandle is piped from/to the STDOUT/STDIN of the child process.
3449 In the child process, the filehandle isn't opened--I/O happens from/to
3450 the new STDOUT/STDIN. Typically this is used like the normal
3451 piped open when you want to exercise more control over just how the
3452 pipe command gets executed, such as when running setuid and
3453 you don't want to have to scan shell commands for metacharacters.
3455 The following triples are more or less equivalent:
3457 open(FOO, "|tr '[a-z]' '[A-Z]'");
3458 open(FOO, '|-', "tr '[a-z]' '[A-Z]'");
3459 open(FOO, '|-') || exec 'tr', '[a-z]', '[A-Z]';
3460 open(FOO, '|-', "tr", '[a-z]', '[A-Z]');
3462 open(FOO, "cat -n '$file'|");
3463 open(FOO, '-|', "cat -n '$file'");
3464 open(FOO, '-|') || exec 'cat', '-n', $file;
3465 open(FOO, '-|', "cat", '-n', $file);
3467 The last example in each block shows the pipe as "list form", which is
3468 not yet supported on all platforms. A good rule of thumb is that if
3469 your platform has true C<fork()> (in other words, if your platform is
3470 Unix) you can use the list form.
3472 See L<perlipc/"Safe Pipe Opens"> for more examples of this.
3474 Beginning with v5.6.0, Perl will attempt to flush all files opened for
3475 output before any operation that may do a fork, but this may not be
3476 supported on some platforms (see L<perlport>). To be safe, you may need
3477 to set C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method
3478 of C<IO::Handle> on any open handles.
3480 On systems that support a close-on-exec flag on files, the flag will
3481 be set for the newly opened file descriptor as determined by the value
3482 of $^F. See L<perlvar/$^F>.
3484 Closing any piped filehandle causes the parent process to wait for the
3485 child to finish, and returns the status value in C<$?> and
3486 C<${^CHILD_ERROR_NATIVE}>.
3488 The filename passed to 2-argument (or 1-argument) form of open() will
3489 have leading and trailing whitespace deleted, and the normal
3490 redirection characters honored. This property, known as "magic open",
3491 can often be used to good effect. A user could specify a filename of
3492 F<"rsh cat file |">, or you could change certain filenames as needed:
3494 $filename =~ s/(.*\.gz)\s*$/gzip -dc < $1|/;
3495 open(FH, $filename) or die "Can't open $filename: $!";
3497 Use 3-argument form to open a file with arbitrary weird characters in it,
3499 open(FOO, '<', $file);
3501 otherwise it's necessary to protect any leading and trailing whitespace:
3503 $file =~ s#^(\s)#./$1#;
3504 open(FOO, "< $file\0");
3506 (this may not work on some bizarre filesystems). One should
3507 conscientiously choose between the I<magic> and 3-arguments form
3512 will allow the user to specify an argument of the form C<"rsh cat file |">,
3513 but will not work on a filename that happens to have a trailing space, while
3515 open IN, '<', $ARGV[0];
3517 will have exactly the opposite restrictions.
3519 If you want a "real" C C<open> (see C<open(2)> on your system), then you
3520 should use the C<sysopen> function, which involves no such magic (but
3521 may use subtly different filemodes than Perl open(), which is mapped
3522 to C fopen()). This is
3523 another way to protect your filenames from interpretation. For example:
3526 sysopen(HANDLE, $path, O_RDWR|O_CREAT|O_EXCL)
3527 or die "sysopen $path: $!";
3528 $oldfh = select(HANDLE); $| = 1; select($oldfh);
3529 print HANDLE "stuff $$\n";
3531 print "File contains: ", <HANDLE>;
3533 Using the constructor from the C<IO::Handle> package (or one of its
3534 subclasses, such as C<IO::File> or C<IO::Socket>), you can generate anonymous
3535 filehandles that have the scope of whatever variables hold references to
3536 them, and automatically close whenever and however you leave that scope:
3540 sub read_myfile_munged {
3542 my $handle = IO::File->new;
3543 open($handle, "myfile") or die "myfile: $!";
3545 or return (); # Automatically closed here.
3546 mung $first or die "mung failed"; # Or here.
3547 return $first, <$handle> if $ALL; # Or here.
3551 See L</seek> for some details about mixing reading and writing.
3553 =item opendir DIRHANDLE,EXPR
3556 Opens a directory named EXPR for processing by C<readdir>, C<telldir>,
3557 C<seekdir>, C<rewinddir>, and C<closedir>. Returns true if successful.
3558 DIRHANDLE may be an expression whose value can be used as an indirect
3559 dirhandle, usually the real dirhandle name. If DIRHANDLE is an undefined
3560 scalar variable (or array or hash element), the variable is assigned a
3561 reference to a new anonymous dirhandle.
3562 DIRHANDLEs have their own namespace separate from FILEHANDLEs.
3564 See example at C<readdir>.
3571 Returns the numeric (the native 8-bit encoding, like ASCII or EBCDIC,
3572 or Unicode) value of the first character of EXPR. If EXPR is omitted,
3575 For the reverse, see L</chr>.
3576 See L<perlunicode> for more about Unicode.
3583 =item our EXPR : ATTRS
3585 =item our TYPE EXPR : ATTRS
3587 C<our> associates a simple name with a package variable in the current
3588 package for use within the current scope. When C<use strict 'vars'> is in
3589 effect, C<our> lets you use declared global variables without qualifying
3590 them with package names, within the lexical scope of the C<our> declaration.
3591 In this way C<our> differs from C<use vars>, which is package scoped.
3593 Unlike C<my>, which both allocates storage for a variable and associates
3594 a simple name with that storage for use within the current scope, C<our>
3595 associates a simple name with a package variable in the current package,
3596 for use within the current scope. In other words, C<our> has the same
3597 scoping rules as C<my>, but does not necessarily create a
3600 If more than one value is listed, the list must be placed
3606 An C<our> declaration declares a global variable that will be visible
3607 across its entire lexical scope, even across package boundaries. The
3608 package in which the variable is entered is determined at the point
3609 of the declaration, not at the point of use. This means the following
3613 our $bar; # declares $Foo::bar for rest of lexical scope
3617 print $bar; # prints 20, as it refers to $Foo::bar
3619 Multiple C<our> declarations with the same name in the same lexical
3620 scope are allowed if they are in different packages. If they happen
3621 to be in the same package, Perl will emit warnings if you have asked
3622 for them, just like multiple C<my> declarations. Unlike a second
3623 C<my> declaration, which will bind the name to a fresh variable, a
3624 second C<our> declaration in the same package, in the same scope, is
3629 our $bar; # declares $Foo::bar for rest of lexical scope
3633 our $bar = 30; # declares $Bar::bar for rest of lexical scope
3634 print $bar; # prints 30
3636 our $bar; # emits warning but has no other effect
3637 print $bar; # still prints 30
3639 An C<our> declaration may also have a list of attributes associated
3642 The exact semantics and interface of TYPE and ATTRS are still
3643 evolving. TYPE is currently bound to the use of C<fields> pragma,
3644 and attributes are handled using the C<attributes> pragma, or starting
3645 from Perl 5.8.0 also via the C<Attribute::Handlers> module. See
3646 L<perlsub/"Private Variables via my()"> for details, and L<fields>,
3647 L<attributes>, and L<Attribute::Handlers>.
3649 =item pack TEMPLATE,LIST
3652 Takes a LIST of values and converts it into a string using the rules
3653 given by the TEMPLATE. The resulting string is the concatenation of
3654 the converted values. Typically, each converted value looks
3655 like its machine-level representation. For example, on 32-bit machines
3656 an integer may be represented by a sequence of 4 bytes, which will in
3657 Perl be presented as a string that's 4 characters long.
3659 See L<perlpacktut> for an introduction to this function.
3661 The TEMPLATE is a sequence of characters that give the order and type
3662 of values, as follows:
3664 a A string with arbitrary binary data, will be null padded.
3665 A A text (ASCII) string, will be space padded.
3666 Z A null-terminated (ASCIZ) string, will be null padded.
3668 b A bit string (ascending bit order inside each byte, like vec()).
3669 B A bit string (descending bit order inside each byte).
3670 h A hex string (low nybble first).
3671 H A hex string (high nybble first).
3673 c A signed char (8-bit) value.
3674 C An unsigned char (octet) value.
3675 W An unsigned char value (can be greater than 255).
3677 s A signed short (16-bit) value.
3678 S An unsigned short value.
3680 l A signed long (32-bit) value.
3681 L An unsigned long value.
3683 q A signed quad (64-bit) value.
3684 Q An unsigned quad value.
3685 (Quads are available only if your system supports 64-bit
3686 integer values _and_ if Perl has been compiled to support those.
3687 Raises an exception otherwise.)
3689 i A signed integer value.
3690 I A unsigned integer value.
3691 (This 'integer' is _at_least_ 32 bits wide. Its exact
3692 size depends on what a local C compiler calls 'int'.)
3694 n An unsigned short (16-bit) in "network" (big-endian) order.
3695 N An unsigned long (32-bit) in "network" (big-endian) order.
3696 v An unsigned short (16-bit) in "VAX" (little-endian) order.
3697 V An unsigned long (32-bit) in "VAX" (little-endian) order.
3699 j A Perl internal signed integer value (IV).
3700 J A Perl internal unsigned integer value (UV).
3702 f A single-precision float in native format.
3703 d A double-precision float in native format.
3705 F A Perl internal floating-point value (NV) in native format
3706 D A float of long-double precision in native format.
3707 (Long doubles are available only if your system supports long
3708 double values _and_ if Perl has been compiled to support those.
3709 Raises an exception otherwise.)
3711 p A pointer to a null-terminated string.
3712 P A pointer to a structure (fixed-length string).
3714 u A uuencoded string.
3715 U A Unicode character number. Encodes to a character in character mode
3716 and UTF-8 (or UTF-EBCDIC in EBCDIC platforms) in byte mode.
3718 w A BER compressed integer (not an ASN.1 BER, see perlpacktut for
3719 details). Its bytes represent an unsigned integer in base 128,
3720 most significant digit first, with as few digits as possible. Bit
3721 eight (the high bit) is set on each byte except the last.
3723 x A null byte (a.k.a ASCII NUL, "\000", chr(0))
3725 @ Null-fill or truncate to absolute position, counted from the
3726 start of the innermost ()-group.
3727 . Null-fill or truncate to absolute position specified by the value.
3728 ( Start of a ()-group.
3730 One or more modifiers below may optionally follow certain letters in the
3731 TEMPLATE (the second column lists letters for which the modifier is valid):
3733 ! sSlLiI Forces native (short, long, int) sizes instead
3734 of fixed (16-/32-bit) sizes.
3736 xX Make x and X act as alignment commands.
3738 nNvV Treat integers as signed instead of unsigned.
3740 @. Specify position as byte offset in the internal
3741 representation of the packed string. Efficient but
3744 > sSiIlLqQ Force big-endian byte-order on the type.
3745 jJfFdDpP (The "big end" touches the construct.)
3747 < sSiIlLqQ Force little-endian byte-order on the type.
3748 jJfFdDpP (The "little end" touches the construct.)
3750 The C<< > >> and C<< < >> modifiers can also be used on C<()> groups
3751 to force a particular byte-order on all components in that group,
3752 including all its subgroups.
3754 The following rules apply:
3760 Each letter may optionally be followed by a number indicating the repeat
3761 count. A numeric repeat count may optionally be enclosed in brackets, as
3762 in C<pack("C[80]", @arr)>. The repeat count gobbles that many values from
3763 the LIST when used with all format types other than C<a>, C<A>, C<Z>, C<b>,
3764 C<B>, C<h>, C<H>, C<@>, C<.>, C<x>, C<X>, and C<P>, where it means
3765 something else, dscribed below. Supplying a C<*> for the repeat count
3766 instead of a number means to use however many items are left, except for:
3772 C<@>, C<x>, and C<X>, where it is equivalent to C<0>.
3776 <.>, where it means relative to the start of the string.
3780 C<u>, where it is equivalent to 1 (or 45, which here is equivalent).
3784 One can replace a numeric repeat count with a template letter enclosed in
3785 brackets to use the packed byte length of the bracketed template for the
3788 For example, the template C<x[L]> skips as many bytes as in a packed long,
3789 and the template C<"$t X[$t] $t"> unpacks twice whatever $t (when
3790 variable-expanded) unpacks. If the template in brackets contains alignment
3791 commands (such as C<x![d]>), its packed length is calculated as if the
3792 start of the template had the maximal possible alignment.
3794 When used with C<Z>, a C<*> as the repeat count is guaranteed to add a
3795 trailing null byte, so the resulting string is always one byte longer than
3796 the byte length of the item itself.
3798 When used with C<@>, the repeat count represents an offset from the start
3799 of the innermost C<()> group.
3801 When used with C<.>, the repeat count determines the starting position to
3802 calculate the value offset as follows:
3808 If the repeat count is C<0>, it's relative to the current position.
3812 If the repeat count is C<*>, the offset is relative to the start of the
3817 And if it's an integer I<n>, the offset is relative to the start of the
3818 I<n>th innermost C<()> group, or to the start of the string if I<n> is
3819 bigger then the group level.
3823 The repeat count for C<u> is interpreted as the maximal number of bytes
3824 to encode per line of output, with 0, 1 and 2 replaced by 45. The repeat
3825 count should not be more than 65.
3829 The C<a>, C<A>, and C<Z> types gobble just one value, but pack it as a
3830 string of length count, padding with nulls or spaces as needed. When
3831 unpacking, C<A> strips trailing whitespace and nulls, C<Z> strips everything
3832 after the first null, and C<a> returns data without any sort of trimming.
3834 If the value to pack is too long, the result is truncated. If it's too
3835 long and an explicit count is provided, C<Z> packs only C<$count-1> bytes,
3836 followed by a null byte. Thus C<Z> always packs a trailing null, except
3837 for when the count is 0.
3841 Likewise, the C<b> and C<B> formats pack a string that's that many bits long.
3842 Each such format generates 1 bit of the result.
3844 Each result bit is based on the least-significant bit of the corresponding
3845 input character, i.e., on C<ord($char)%2>. In particular, characters C<"0">
3846 and C<"1"> generate bits 0 and 1, as do characters C<"\000"> and C<"\001">.
3848 Starting from the beginning of the input string, each 8-tuple
3849 of characters is converted to 1 character of output. With format C<b>,
3850 the first character of the 8-tuple determines the least-significant bit of a
3851 character; with format C<B>, it determines the most-significant bit of
3854 If the length of the input string is not evenly divisible by 8, the
3855 remainder is packed as if the input string were padded by null characters
3856 at the end. Similarly during unpacking, "extra" bits are ignored.
3858 If the input string is longer than needed, remaining characters are ignored.
3860 A C<*> for the repeat count uses all characters of the input field.
3861 On unpacking, bits are converted to a string of C<"0">s and C<"1">s.
3865 The C<h> and C<H> formats pack a string that many nybbles (4-bit groups,
3866 representable as hexadecimal digits, C<"0".."9"> C<"a".."f">) long.
3868 For each such format, pack() generates 4 bits of the result.
3869 With non-alphabetical characters, the result is based on the 4 least-significant
3870 bits of the input character, i.e., on C<ord($char)%16>. In particular,
3871 characters C<"0"> and C<"1"> generate nybbles 0 and 1, as do bytes
3872 C<"\0"> and C<"\1">. For characters C<"a".."f"> and C<"A".."F">, the result
3873 is compatible with the usual hexadecimal digits, so that C<"a"> and
3874 C<"A"> both generate the nybble C<0xa==10>. Do not use any characters
3875 but these with this format.
3877 Starting from the beginning of the template to pack(), each pair
3878 of characters is converted to 1 character of output. With format C<h>, the
3879 first character of the pair determines the least-significant nybble of the
3880 output character; with format C<H>, it determines the most-significant
3883 If the length of the input string is not even, it behaves as if padded by
3884 a null character at the end. Similarly, "extra" nybbles are ignored during
3887 If the input string is longer than needed, extra characters are ignored.
3889 A C<*> for the repeat count uses all characters of the input field. For
3890 unpack(), nybbles are converted to a string of hexadecimal digits.
3894 The C<p> format packs a pointer to a null-terminated string. You are
3895 responsible for ensuring that the string is not a temporary value, as that
3896 could potentially get deallocated before you got around to using the packed
3897 result. The C<P> format packs a pointer to a structure of the size indicated
3898 by the length. A null pointer is created if the corresponding value for
3899 C<p> or C<P> is C<undef>; similarly with unpack(), where a null pointer
3900 unpacks into C<undef>.
3902 If your system has a strange pointer size--meaning a pointer is neither as
3903 big as an int nor as big as a long--it may not be possible to pack or
3904 unpack pointers in big- or little-endian byte order. Attempting to do
3905 so raises an exception.
3909 The C</> template character allows packing and unpacking of a sequence of
3910 items where the packed structure contains a packed item count followed by
3911 the packed items themselves. This is useful when the structure you're
3912 unpacking has encoded the sizes or repeat counts for some of its fields
3913 within the structure itself as separate fields.
3915 For C<pack>, you write I<length-item>C</>I<sequence-item>, and the
3916 I<length-item> describes how the length value is packed. Formats likely
3917 to be of most use are integer-packing ones like C<n> for Java strings,
3918 C<w> for ASN.1 or SNMP, and C<N> for Sun XDR.
3920 For C<pack>, I<sequence-item> may have a repeat count, in which case
3921 the minimum of that and the number of available items is used as the argument
3922 for I<length-item>. If it has no repeat count or uses a '*', the number
3923 of available items is used.
3925 For C<unpack>, an internal stack of integer arguments unpacked so far is
3926 used. You write C</>I<sequence-item> and the repeat count is obtained by
3927 popping off the last element from the stack. The I<sequence-item> must not
3928 have a repeat count.
3930 If I<sequence-item> refers to a string type (C<"A">, C<"a">, or C<"Z">),
3931 the I<length-item> is the string length, not the number of strings. With
3932 an explicit repeat count for pack, the packed string is adjusted to that
3933 length. For example:
3935 unpack("W/a", "\04Gurusamy") gives ("Guru")
3936 unpack("a3/A A*", "007 Bond J ") gives (" Bond", "J")
3937 unpack("a3 x2 /A A*", "007: Bond, J.") gives ("Bond, J", ".")
3939 pack("n/a* w/a","hello,","world") gives "\000\006hello,\005world"
3940 pack("a/W2", ord("a") .. ord("z")) gives "2ab"
3942 The I<length-item> is not returned explicitly from C<unpack>.
3944 Supplying a count to the I<length-item> format letter is only useful with
3945 C<A>, C<a>, or C<Z>. Packing with a I<length-item> of C<a> or C<Z> may
3946 introduce C<"\000"> characters, which Perl does not regard as legal in
3951 The integer types C<s>, C<S>, C<l>, and C<L> may be
3952 followed by a C<!> modifier to specify native shorts or
3953 longs. As shown in the example above, a bare C<l> means
3954 exactly 32 bits, although the native C<long> as seen by the local C compiler
3955 may be larger. This is mainly an issue on 64-bit platforms. You can
3956 see whether using C<!> makes any difference this way:
3958 printf "format s is %d, s! is %d\n",
3959 length pack("s"), length pack("s!");
3961 printf "format l is %d, l! is %d\n",
3962 length pack("l"), length pack("l!");
3965 C<i!> and C<I!> are also allowed, but only for completeness' sake:
3966 they are identical to C<i> and C<I>.
3968 The actual sizes (in bytes) of native shorts, ints, longs, and long
3969 longs on the platform where Perl was built are also available from
3972 $ perl -V:{short,int,long{,long}}size
3978 or programmatically via the C<Config> module:
3981 print $Config{shortsize}, "\n";
3982 print $Config{intsize}, "\n";
3983 print $Config{longsize}, "\n";
3984 print $Config{longlongsize}, "\n";
3986 C<$Config{longlongsize}> is undefined on systems without
3991 The integer formats C<s>, C<S>, C<i>, C<I>, C<l>, C<L>, C<j>, and C<J> are
3992 inherently non-portable between processors and operating systems because
3993 they obey native byteorder and endianness. For example, a 4-byte integer
3994 0x12345678 (305419896 decimal) would be ordered natively (arranged in and
3995 handled by the CPU registers) into bytes as
3997 0x12 0x34 0x56 0x78 # big-endian
3998 0x78 0x56 0x34 0x12 # little-endian
4000 Basically, Intel and VAX CPUs are little-endian, while everybody else,
4001 including Motorola m68k/88k, PPC, Sparc, HP PA, Power, and Cray, are
4002 big-endian. Alpha and MIPS can be either: Digital/Compaq used/uses them in
4003 little-endian mode, but SGI/Cray uses them in big-endian mode.
4005 The names I<big-endian> and I<little-endian> are comic references to the
4006 egg-eating habits of the little-endian Lilliputians and the big-endian
4007 Blefuscudians from the classic Jonathan Swift satire, I<Gulliver's Travels>.
4008 This entered computer lingo via the paper "On Holy Wars and a Plea for
4009 Peace" by Danny Cohen, USC/ISI IEN 137, April 1, 1980.
4011 Some systems may have even weirder byte orders such as
4016 You can determine your system endianness with this incantation:
4018 printf("%#02x ", $_) for unpack("W*", pack L=>0x12345678);
4020 The byteorder on the platform where Perl was built is also available
4024 print "$Config{byteorder}\n";
4026 or from the command line:
4030 Byteorders C<"1234"> and C<"12345678"> are little-endian; C<"4321">
4031 and C<"87654321"> are big-endian.
4033 For portably packed integers, either use the formats C<n>, C<N>, C<v>,
4034 and C<V> or else use the C<< > >> and C<< < >> modifiers described
4035 immediately below. See also L<perlport>.
4039 Starting with Perl 5.9.2, integer and floating-point formats, along with
4040 the C<p> and C<P> formats and C<()> groups, may all be followed by the
4041 C<< > >> or C<< < >> endianness modifiers to respectively enforce big-
4042 or little-endian byte-order. These modifiers are especially useful
4043 given how C<n>, C<N>, C<v> and C<V> don't cover signed integers,
4044 64-bit integers, or floating-point values.
4046 Here are some concerns to keep in mind when using endianness modifier:
4052 Exchanging signed integers between different platforms works only
4053 when all platforms store them in the same format. Most platforms store
4054 signed integers in two's-complement notation, so usually this is not an issue.
4058 The C<< > >> or C<< < >> modifiers can only be used on floating-point
4059 formats on big- or little-endian machines. Otherwise, attempting to
4060 use them raises an exception.
4064 Forcing big- or little-endian byte-order on floating-point values for
4065 data exchange can work only if all platforms use the same
4066 binary representation such as IEEE floating-point. Even if all
4067 platforms are using IEEE, there may still be subtle differences. Being able
4068 to use C<< > >> or C<< < >> on floating-point values can be useful,
4069 but also dangerous if you don't know exactly what you're doing.
4070 It is not a general way to portably store floating-point values.
4074 When using C<< > >> or C<< < >> on a C<()> group, this affects
4075 all types inside the group that accept byte-order modifiers,
4076 including all subgroups. It is silently ignored for all other
4077 types. You are not allowed to override the byte-order within a group
4078 that already has a byte-order modifier suffix.
4084 Real numbers (floats and doubles) are in native machine format only.
4085 Due to the multiplicity of floating-point formats and the lack of a
4086 standard "network" representation for them, no facility for interchange has been
4087 made. This means that packed floating-point data written on one machine
4088 may not be readable on another, even if both use IEEE floating-point
4089 arithmetic (because the endianness of the memory representation is not part
4090 of the IEEE spec). See also L<perlport>.
4092 If you know I<exactly> what you're doing, you can use the C<< > >> or C<< < >>
4093 modifiers to force big- or little-endian byte-order on floating-point values.
4095 Because Perl uses doubles (or long doubles, if configured) internally for
4096 all numeric calculation, converting from double into float and thence
4097 to double again loses precision, so C<unpack("f", pack("f", $foo)>)
4098 will not in general equal $foo.
4102 Pack and unpack can operate in two modes: character mode (C<C0> mode) where
4103 the packed string is processed per character, and UTF-8 mode (C<U0> mode)
4104 where the packed string is processed in its UTF-8-encoded Unicode form on
4105 a byte-by-byte basis. Character mode is the default unless the format string
4106 starts with C<U>. You can always switch mode mid-format with an explicit
4107 C<C0> or C<U0> in the format. This mode remains in effect until the next
4108 mode change, or until the end of the C<()> group it (directly) applies to.
4112 You must yourself do any alignment or padding by inserting, for example,
4113 enough C<"x">es while packing. There is no way for pack() and unpack()
4114 to know where characters are going to or coming from, so they
4115 handle their output and input as flat sequences of characters.
4119 A C<()> group is a sub-TEMPLATE enclosed in parentheses. A group may
4120 take a repeat count either as postfix, or for unpack(), also via the C</>
4121 template character. Within each repetition of a group, positioning with
4122 C<@> starts over at 0. Therefore, the result of
4124 pack("@1A((@2A)@3A)", qw[X Y Z])
4126 is the string C<"\0X\0\0YZ">.
4130 C<x> and C<X> accept the C<!> modifier to act as alignment commands: they
4131 jump forward or back to the closest position aligned at a multiple of C<count>
4132 characters. For example, to pack() or unpack() a C structure like
4135 char c; /* one signed, 8-bit character */
4140 one may need to use the template C<c x![d] d c[2]>. This assumes that
4141 doubles must be aligned to the size of double.
4143 For alignment commands, a C<count> of 0 is equivalent to a C<count> of 1;
4148 C<n>, C<N>, C<v> and C<V> accept the C<!> modifier to
4149 represent signed 16-/32-bit integers in big-/little-endian order.
4150 This is portable only when all platforms sharing packed data use the
4151 same binary representation for signed integers; for example, when all
4152 platforms use two's-complement representation.
4156 Comments can be embedded in a TEMPLATE using C<#> through the end of line.
4157 White space can separate pack codes from each other, but modifiers and
4158 repeat counts must follow immediately. Breaking complex templates into
4159 individual line-by-line components, suitably annotated, can do as much to
4160 improve legibility and maintainability of pack/unpack formats as C</x> can
4161 for complicated pattern matches.
4165 If TEMPLATE requires more arguments that pack() is given, pack()
4166 assumes additional C<""> arguments. If TEMPLATE requires fewer arguments
4167 than given, extra arguments are ignored.
4173 $foo = pack("WWWW",65,66,67,68);
4175 $foo = pack("W4",65,66,67,68);
4177 $foo = pack("W4",0x24b6,0x24b7,0x24b8,0x24b9);
4178 # same thing with Unicode circled letters.
4179 $foo = pack("U4",0x24b6,0x24b7,0x24b8,0x24b9);
4180 # same thing with Unicode circled letters. You don't get the UTF-8
4181 # bytes because the U at the start of the format caused a switch to
4182 # U0-mode, so the UTF-8 bytes get joined into characters
4183 $foo = pack("C0U4",0x24b6,0x24b7,0x24b8,0x24b9);
4184 # foo eq "\xe2\x92\xb6\xe2\x92\xb7\xe2\x92\xb8\xe2\x92\xb9"
4185 # This is the UTF-8 encoding of the string in the previous example
4187 $foo = pack("ccxxcc",65,66,67,68);
4190 # NOTE: The examples above featuring "W" and "c" are true
4191 # only on ASCII and ASCII-derived systems such as ISO Latin 1
4192 # and UTF-8. On EBCDIC systems, the first example would be
4193 # $foo = pack("WWWW",193,194,195,196);
4195 $foo = pack("s2",1,2);
4196 # "\1\0\2\0" on little-endian
4197 # "\0\1\0\2" on big-endian
4199 $foo = pack("a4","abcd","x","y","z");
4202 $foo = pack("aaaa","abcd","x","y","z");
4205 $foo = pack("a14","abcdefg");
4206 # "abcdefg\0\0\0\0\0\0\0"
4208 $foo = pack("i9pl", gmtime);
4209 # a real struct tm (on my system anyway)
4211 $utmp_template = "Z8 Z8 Z16 L";
4212 $utmp = pack($utmp_template, @utmp1);
4213 # a struct utmp (BSDish)
4215 @utmp2 = unpack($utmp_template, $utmp);
4216 # "@utmp1" eq "@utmp2"
4219 unpack("N", pack("B32", substr("0" x 32 . shift, -32)));
4222 $foo = pack('sx2l', 12, 34);
4223 # short 12, two zero bytes padding, long 34
4224 $bar = pack('s@4l', 12, 34);
4225 # short 12, zero fill to position 4, long 34
4227 $baz = pack('s.l', 12, 4, 34);
4228 # short 12, zero fill to position 4, long 34
4230 $foo = pack('nN', 42, 4711);
4231 # pack big-endian 16- and 32-bit unsigned integers
4232 $foo = pack('S>L>', 42, 4711);
4234 $foo = pack('s<l<', -42, 4711);
4235 # pack little-endian 16- and 32-bit signed integers
4236 $foo = pack('(sl)<', -42, 4711);
4239 The same template may generally also be used in unpack().
4241 =item package NAMESPACE VERSION
4242 X<package> X<module> X<namespace> X<version>
4244 =item package NAMESPACE
4246 Declares the compilation unit as being in the given namespace. The scope
4247 of the package declaration is from the declaration itself through the end
4248 of the enclosing block, file, or eval (the same as the C<my> operator).
4249 All further unqualified dynamic identifiers will be in this namespace.
4250 A package statement affects dynamic variables only, including those
4251 you've used C<local> on, but I<not> lexical variables, which are created
4252 with C<my> (or C<our> (or C<state>)). Typically it would be the first
4253 declaration in a file included by C<require> or C<use>. You can switch into a
4254 package in more than one place, since this only determines which default
4255 symbol table the compiler uses for the rest of that block. You can refer to
4256 identifiers in other packages than the current one by prefixing the identifier
4257 with the package name and a double colon, as in C<$SomePack::var>
4258 or C<ThatPack::INPUT_HANDLE>. If package name is omitted, the C<main>
4259 package as assumed. That is, C<$::sail> is equivalent to
4260 C<$main::sail> (as well as to C<$main'sail>, still seen in ancient
4261 code, mostly from Perl 4).
4263 If VERSION is provided, C<package> sets the C<$VERSION> variable in the given
4264 namespace to a L<version> object with the VERSION provided. VERSION must be a
4265 "strict" style version number as defined by the L<version> module: a positive
4266 decimal number (integer or decimal-fraction) without exponentiation or else a
4267 dotted-decimal v-string with a leading 'v' character and at least three
4268 components. You should set C<$VERSION> only once per package.
4270 See L<perlmod/"Packages"> for more information about packages, modules,
4271 and classes. See L<perlsub> for other scoping issues.
4273 =item pipe READHANDLE,WRITEHANDLE
4276 Opens a pair of connected pipes like the corresponding system call.
4277 Note that if you set up a loop of piped processes, deadlock can occur
4278 unless you are very careful. In addition, note that Perl's pipes use
4279 IO buffering, so you may need to set C<$|> to flush your WRITEHANDLE
4280 after each command, depending on the application.
4282 See L<IPC::Open2>, L<IPC::Open3>, and L<perlipc/"Bidirectional Communication">
4283 for examples of such things.
4285 On systems that support a close-on-exec flag on files, that flag is set
4286 on all newly opened file descriptors whose C<fileno>s are I<higher> than
4287 the current value of $^F (by default 2 for C<STDERR>). See L<perlvar/$^F>.
4294 Pops and returns the last value of the array, shortening the array by
4297 Returns the undefined value if the array is empty, although this may also
4298 happen at other times. If ARRAY is omitted, pops the C<@ARGV> array in the
4299 main program, but the C<@_> array in subroutines, just like C<shift>.
4302 X<pos> X<match, position>
4306 Returns the offset of where the last C<m//g> search left off for the variable
4307 in question (C<$_> is used when the variable is not specified). Note that
4308 0 is a valid match offset. C<undef> indicates that the search position
4309 is reset (usually due to match failure, but can also be because no match has
4310 yet been run on the scalar). C<pos> directly accesses the location used
4311 by the regexp engine to store the offset, so assigning to C<pos> will change
4312 that offset, and so will also influence the C<\G> zero-width assertion in
4313 regular expressions. Because a failed C<m//gc> match doesn't reset the offset,
4314 the return from C<pos> won't change either in this case. See L<perlre> and
4317 =item print FILEHANDLE LIST
4324 Prints a string or a list of strings. Returns true if successful.
4325 FILEHANDLE may be a scalar variable containing
4326 the name of or a reference to the filehandle, thus introducing
4327 one level of indirection. (NOTE: If FILEHANDLE is a variable and
4328 the next token is a term, it may be misinterpreted as an operator
4329 unless you interpose a C<+> or put parentheses around the arguments.)
4330 If FILEHANDLE is omitted, prints to standard output by default, or
4331 to the last selected output channel; see L</select>. If LIST is
4332 also omitted, prints C<$_> to the currently selected output handle.
4333 To set the default output handle to something other than STDOUT
4334 use the select operation. The current value of C<$,> (if any) is
4335 printed between each LIST item. The current value of C<$\> (if
4336 any) is printed after the entire LIST has been printed. Because
4337 print takes a LIST, anything in the LIST is evaluated in list
4338 context, and any subroutine that you call will have one or more of
4339 its expressions evaluated in list context. Also be careful not to
4340 follow the print keyword with a left parenthesis unless you want
4341 the corresponding right parenthesis to terminate the arguments to
4342 the print; put parentheses around all the arguments
4343 (or interpose a C<+>, but that doesn't look as good).
4345 Note that if you're storing FILEHANDLEs in an array, or if you're using
4346 any other expression more complex than a scalar variable to retrieve it,
4347 you will have to use a block returning the filehandle value instead:
4349 print { $files[$i] } "stuff\n";
4350 print { $OK ? STDOUT : STDERR } "stuff\n";
4352 Printing to a closed pipe or socket will generate a SIGPIPE signal. See
4353 L<perlipc> for more on signal handling.
4355 =item printf FILEHANDLE FORMAT, LIST
4358 =item printf FORMAT, LIST
4360 Equivalent to C<print FILEHANDLE sprintf(FORMAT, LIST)>, except that C<$\>
4361 (the output record separator) is not appended. The first argument
4362 of the list will be interpreted as the C<printf> format. See C<sprintf>
4363 for an explanation of the format argument. If C<use locale> is in effect,
4364 and POSIX::setlocale() has been called, the character used for the decimal
4365 separator in formatted floating-point numbers is affected by the LC_NUMERIC
4366 locale. See L<perllocale> and L<POSIX>.
4368 Don't fall into the trap of using a C<printf> when a simple
4369 C<print> would do. The C<print> is more efficient and less
4372 =item prototype FUNCTION
4375 Returns the prototype of a function as a string (or C<undef> if the
4376 function has no prototype). FUNCTION is a reference to, or the name of,
4377 the function whose prototype you want to retrieve.
4379 If FUNCTION is a string starting with C<CORE::>, the rest is taken as a
4380 name for a Perl builtin. If the builtin is not I<overridable> (such as
4381 C<qw//>) or if its arguments cannot be adequately expressed by a prototype
4382 (such as C<system>), prototype() returns C<undef>, because the builtin
4383 does not really behave like a Perl function. Otherwise, the string
4384 describing the equivalent prototype is returned.
4386 =item push ARRAY,LIST
4389 Treats ARRAY as a stack, and pushes the values of LIST
4390 onto the end of ARRAY. The length of ARRAY increases by the length of
4391 LIST. Has the same effect as
4394 $ARRAY[++$#ARRAY] = $value;
4397 but is more efficient. Returns the number of elements in the array following
4398 the completed C<push>.
4408 Generalized quotes. See L<perlop/"Quote-Like Operators">.
4412 Regexp-like quote. See L<perlop/"Regexp Quote-Like Operators">.
4414 =item quotemeta EXPR
4415 X<quotemeta> X<metacharacter>
4419 Returns the value of EXPR with all non-"word"
4420 characters backslashed. (That is, all characters not matching
4421 C</[A-Za-z_0-9]/> will be preceded by a backslash in the
4422 returned string, regardless of any locale settings.)
4423 This is the internal function implementing
4424 the C<\Q> escape in double-quoted strings.
4426 If EXPR is omitted, uses C<$_>.
4428 quotemeta (and C<\Q> ... C<\E>) are useful when interpolating strings into
4429 regular expressions, because by default an interpolated variable will be
4430 considered a mini-regular expression. For example:
4432 my $sentence = 'The quick brown fox jumped over the lazy dog';
4433 my $substring = 'quick.*?fox';
4434 $sentence =~ s{$substring}{big bad wolf};
4436 Will cause C<$sentence> to become C<'The big bad wolf jumped over...'>.
4440 my $sentence = 'The quick brown fox jumped over the lazy dog';
4441 my $substring = 'quick.*?fox';
4442 $sentence =~ s{\Q$substring\E}{big bad wolf};
4446 my $sentence = 'The quick brown fox jumped over the lazy dog';
4447 my $substring = 'quick.*?fox';
4448 my $quoted_substring = quotemeta($substring);
4449 $sentence =~ s{$quoted_substring}{big bad wolf};
4451 Will both leave the sentence as is. Normally, when accepting string input from
4452 the user, quotemeta() or C<\Q> must be used.
4459 Returns a random fractional number greater than or equal to C<0> and less
4460 than the value of EXPR. (EXPR should be positive.) If EXPR is
4461 omitted, the value C<1> is used. Currently EXPR with the value C<0> is
4462 also special-cased as C<1> (this was undocumented before Perl 5.8.0
4463 and is subject to change in future versions of Perl). Automatically calls
4464 C<srand> unless C<srand> has already been called. See also C<srand>.
4466 Apply C<int()> to the value returned by C<rand()> if you want random
4467 integers instead of random fractional numbers. For example,
4471 returns a random integer between C<0> and C<9>, inclusive.
4473 (Note: If your rand function consistently returns numbers that are too
4474 large or too small, then your version of Perl was probably compiled
4475 with the wrong number of RANDBITS.)
4477 =item read FILEHANDLE,SCALAR,LENGTH,OFFSET
4478 X<read> X<file, read>
4480 =item read FILEHANDLE,SCALAR,LENGTH
4482 Attempts to read LENGTH I<characters> of data into variable SCALAR
4483 from the specified FILEHANDLE. Returns the number of characters
4484 actually read, C<0> at end of file, or undef if there was an error (in
4485 the latter case C<$!> is also set). SCALAR will be grown or shrunk
4486 so that the last character actually read is the last character of the
4487 scalar after the read.
4489 An OFFSET may be specified to place the read data at some place in the
4490 string other than the beginning. A negative OFFSET specifies
4491 placement at that many characters counting backwards from the end of
4492 the string. A positive OFFSET greater than the length of SCALAR
4493 results in the string being padded to the required size with C<"\0">
4494 bytes before the result of the read is appended.
4496 The call is implemented in terms of either Perl's or your system's native
4497 fread(3) library function. To get a true read(2) system call, see C<sysread>.
4499 Note the I<characters>: depending on the status of the filehandle,
4500 either (8-bit) bytes or characters are read. By default all
4501 filehandles operate on bytes, but for example if the filehandle has
4502 been opened with the C<:utf8> I/O layer (see L</open>, and the C<open>
4503 pragma, L<open>), the I/O will operate on UTF-8 encoded Unicode
4504 characters, not bytes. Similarly for the C<:encoding> pragma:
4505 in that case pretty much any characters can be read.
4507 =item readdir DIRHANDLE
4510 Returns the next directory entry for a directory opened by C<opendir>.
4511 If used in list context, returns all the rest of the entries in the
4512 directory. If there are no more entries, returns the undefined value in
4513 scalar context and the empty list in list context.
4515 If you're planning to filetest the return values out of a C<readdir>, you'd
4516 better prepend the directory in question. Otherwise, because we didn't
4517 C<chdir> there, it would have been testing the wrong file.
4519 opendir(my $dh, $some_dir) || die "can't opendir $some_dir: $!";
4520 @dots = grep { /^\./ && -f "$some_dir/$_" } readdir($dh);
4523 As of Perl 5.11.2 you can use a bare C<readdir> in a C<while> loop,
4524 which will set C<$_> on every iteration.
4526 opendir(my $dh, $some_dir) || die;
4527 while(readdir $dh) {
4528 print "$some_dir/$_\n";
4535 X<readline> X<gets> X<fgets>
4537 Reads from the filehandle whose typeglob is contained in EXPR (or from
4538 *ARGV if EXPR is not provided). In scalar context, each call reads and
4539 returns the next line until end-of-file is reached, whereupon the
4540 subsequent call returns C<undef>. In list context, reads until end-of-file
4541 is reached and returns a list of lines. Note that the notion of "line"
4542 used here is whatever you may have defined with C<$/> or
4543 C<$INPUT_RECORD_SEPARATOR>). See L<perlvar/"$/">.
4545 When C<$/> is set to C<undef>, when C<readline> is in scalar
4546 context (i.e., file slurp mode), and when an empty file is read, it
4547 returns C<''> the first time, followed by C<undef> subsequently.
4549 This is the internal function implementing the C<< <EXPR> >>
4550 operator, but you can use it directly. The C<< <EXPR> >>
4551 operator is discussed in more detail in L<perlop/"I/O Operators">.
4554 $line = readline(*STDIN); # same thing
4556 If C<readline> encounters an operating system error, C<$!> will be set
4557 with the corresponding error message. It can be helpful to check
4558 C<$!> when you are reading from filehandles you don't trust, such as a
4559 tty or a socket. The following example uses the operator form of
4560 C<readline> and dies if the result is not defined.
4562 while ( ! eof($fh) ) {
4563 defined( $_ = <$fh> ) or die "readline failed: $!";
4567 Note that you have can't handle C<readline> errors that way with the
4568 C<ARGV> filehandle. In that case, you have to open each element of
4569 C<@ARGV> yourself since C<eof> handles C<ARGV> differently.
4571 foreach my $arg (@ARGV) {
4572 open(my $fh, $arg) or warn "Can't open $arg: $!";
4574 while ( ! eof($fh) ) {
4575 defined( $_ = <$fh> )
4576 or die "readline failed for $arg: $!";
4586 Returns the value of a symbolic link, if symbolic links are
4587 implemented. If not, raises an exception. If there is a system
4588 error, returns the undefined value and sets C<$!> (errno). If EXPR is
4589 omitted, uses C<$_>.
4596 EXPR is executed as a system command.
4597 The collected standard output of the command is returned.
4598 In scalar context, it comes back as a single (potentially
4599 multi-line) string. In list context, returns a list of lines
4600 (however you've defined lines with C<$/> or C<$INPUT_RECORD_SEPARATOR>).
4601 This is the internal function implementing the C<qx/EXPR/>
4602 operator, but you can use it directly. The C<qx/EXPR/>
4603 operator is discussed in more detail in L<perlop/"I/O Operators">.
4604 If EXPR is omitted, uses C<$_>.
4606 =item recv SOCKET,SCALAR,LENGTH,FLAGS
4609 Receives a message on a socket. Attempts to receive LENGTH characters
4610 of data into variable SCALAR from the specified SOCKET filehandle.
4611 SCALAR will be grown or shrunk to the length actually read. Takes the
4612 same flags as the system call of the same name. Returns the address
4613 of the sender if SOCKET's protocol supports this; returns an empty
4614 string otherwise. If there's an error, returns the undefined value.
4615 This call is actually implemented in terms of recvfrom(2) system call.
4616 See L<perlipc/"UDP: Message Passing"> for examples.
4618 Note the I<characters>: depending on the status of the socket, either
4619 (8-bit) bytes or characters are received. By default all sockets
4620 operate on bytes, but for example if the socket has been changed using
4621 binmode() to operate with the C<:encoding(utf8)> I/O layer (see the
4622 C<open> pragma, L<open>), the I/O will operate on UTF-8 encoded Unicode
4623 characters, not bytes. Similarly for the C<:encoding> pragma: in that
4624 case pretty much any characters can be read.
4631 The C<redo> command restarts the loop block without evaluating the
4632 conditional again. The C<continue> block, if any, is not executed. If
4633 the LABEL is omitted, the command refers to the innermost enclosing
4634 loop. Programs that want to lie to themselves about what was just input
4635 normally use this command:
4637 # a simpleminded Pascal comment stripper
4638 # (warning: assumes no { or } in strings)
4639 LINE: while (<STDIN>) {
4640 while (s|({.*}.*){.*}|$1 |) {}
4645 if (/}/) { # end of comment?
4654 C<redo> cannot be used to retry a block that returns a value such as
4655 C<eval {}>, C<sub {}> or C<do {}>, and should not be used to exit
4656 a grep() or map() operation.
4658 Note that a block by itself is semantically identical to a loop
4659 that executes once. Thus C<redo> inside such a block will effectively
4660 turn it into a looping construct.
4662 See also L</continue> for an illustration of how C<last>, C<next>, and
4670 Returns a non-empty string if EXPR is a reference, the empty
4671 string otherwise. If EXPR
4672 is not specified, C<$_> will be used. The value returned depends on the
4673 type of thing the reference is a reference to.
4674 Builtin types include:
4688 If the referenced object has been blessed into a package, then that package
4689 name is returned instead. You can think of C<ref> as a C<typeof> operator.
4691 if (ref($r) eq "HASH") {
4692 print "r is a reference to a hash.\n";
4695 print "r is not a reference at all.\n";
4698 The return value C<LVALUE> indicates a reference to an lvalue that is not
4699 a variable. You get this from taking the reference of function calls like
4700 C<pos()> or C<substr()>. C<VSTRING> is returned if the reference points
4701 to a L<version string|perldata/"Version Strings">.
4703 The result C<Regexp> indicates that the argument is a regular expression
4704 resulting from C<qr//>.
4706 See also L<perlref>.
4708 =item rename OLDNAME,NEWNAME
4709 X<rename> X<move> X<mv> X<ren>
4711 Changes the name of a file; an existing file NEWNAME will be
4712 clobbered. Returns true for success, false otherwise.
4714 Behavior of this function varies wildly depending on your system
4715 implementation. For example, it will usually not work across file system
4716 boundaries, even though the system I<mv> command sometimes compensates
4717 for this. Other restrictions include whether it works on directories,
4718 open files, or pre-existing files. Check L<perlport> and either the
4719 rename(2) manpage or equivalent system documentation for details.
4721 For a platform independent C<move> function look at the L<File::Copy>
4724 =item require VERSION
4731 Demands a version of Perl specified by VERSION, or demands some semantics
4732 specified by EXPR or by C<$_> if EXPR is not supplied.
4734 VERSION may be either a numeric argument such as 5.006, which will be
4735 compared to C<$]>, or a literal of the form v5.6.1, which will be compared
4736 to C<$^V> (aka $PERL_VERSION). An exception is raised if
4737 VERSION is greater than the version of the current Perl interpreter.
4738 Compare with L</use>, which can do a similar check at compile time.
4740 Specifying VERSION as a literal of the form v5.6.1 should generally be
4741 avoided, because it leads to misleading error messages under earlier
4742 versions of Perl that do not support this syntax. The equivalent numeric
4743 version should be used instead.
4745 require v5.6.1; # run time version check
4746 require 5.6.1; # ditto
4747 require 5.006_001; # ditto; preferred for backwards compatibility
4749 Otherwise, C<require> demands that a library file be included if it
4750 hasn't already been included. The file is included via the do-FILE
4751 mechanism, which is essentially just a variety of C<eval> with the
4752 caveat that lexical variables in the invoking script will be invisible
4753 to the included code. Has semantics similar to the following subroutine:
4756 my ($filename) = @_;
4757 if (exists $INC{$filename}) {
4758 return 1 if $INC{$filename};
4759 die "Compilation failed in require";
4761 my ($realfilename,$result);
4763 foreach $prefix (@INC) {
4764 $realfilename = "$prefix/$filename";
4765 if (-f $realfilename) {
4766 $INC{$filename} = $realfilename;
4767 $result = do $realfilename;
4771 die "Can't find $filename in \@INC";
4774 $INC{$filename} = undef;
4776 } elsif (!$result) {
4777 delete $INC{$filename};
4778 die "$filename did not return true value";
4784 Note that the file will not be included twice under the same specified
4787 The file must return true as the last statement to indicate
4788 successful execution of any initialization code, so it's customary to
4789 end such a file with C<1;> unless you're sure it'll return true
4790 otherwise. But it's better just to put the C<1;>, in case you add more
4793 If EXPR is a bareword, the require assumes a "F<.pm>" extension and
4794 replaces "F<::>" with "F</>" in the filename for you,
4795 to make it easy to load standard modules. This form of loading of
4796 modules does not risk altering your namespace.
4798 In other words, if you try this:
4800 require Foo::Bar; # a splendid bareword
4802 The require function will actually look for the "F<Foo/Bar.pm>" file in the
4803 directories specified in the C<@INC> array.
4805 But if you try this:
4807 $class = 'Foo::Bar';
4808 require $class; # $class is not a bareword
4810 require "Foo::Bar"; # not a bareword because of the ""
4812 The require function will look for the "F<Foo::Bar>" file in the @INC array and
4813 will complain about not finding "F<Foo::Bar>" there. In this case you can do:
4815 eval "require $class";
4817 Now that you understand how C<require> looks for files with a
4818 bareword argument, there is a little extra functionality going on behind
4819 the scenes. Before C<require> looks for a "F<.pm>" extension, it will
4820 first look for a similar filename with a "F<.pmc>" extension. If this file
4821 is found, it will be loaded in place of any file ending in a "F<.pm>"
4824 You can also insert hooks into the import facility, by putting Perl code
4825 directly into the @INC array. There are three forms of hooks: subroutine
4826 references, array references and blessed objects.
4828 Subroutine references are the simplest case. When the inclusion system
4829 walks through @INC and encounters a subroutine, this subroutine gets
4830 called with two parameters, the first a reference to itself, and the
4831 second the name of the file to be included (e.g., "F<Foo/Bar.pm>"). The
4832 subroutine should return either nothing or else a list of up to three
4833 values in the following order:
4839 A filehandle, from which the file will be read.
4843 A reference to a subroutine. If there is no filehandle (previous item),
4844 then this subroutine is expected to generate one line of source code per
4845 call, writing the line into C<$_> and returning 1, then returning 0 at
4846 end of file. If there is a filehandle, then the subroutine will be
4847 called to act as a simple source filter, with the line as read in C<$_>.
4848 Again, return 1 for each valid line, and 0 after all lines have been
4853 Optional state for the subroutine. The state is passed in as C<$_[1]>. A
4854 reference to the subroutine itself is passed in as C<$_[0]>.
4858 If an empty list, C<undef>, or nothing that matches the first 3 values above
4859 is returned, then C<require> looks at the remaining elements of @INC.
4860 Note that this filehandle must be a real filehandle (strictly a typeglob
4861 or reference to a typeglob, blessed or unblessed); tied filehandles will be
4862 ignored and return value processing will stop there.
4864 If the hook is an array reference, its first element must be a subroutine
4865 reference. This subroutine is called as above, but the first parameter is
4866 the array reference. This lets you indirectly pass arguments to
4869 In other words, you can write:
4871 push @INC, \&my_sub;
4873 my ($coderef, $filename) = @_; # $coderef is \&my_sub
4879 push @INC, [ \&my_sub, $x, $y, ... ];
4881 my ($arrayref, $filename) = @_;
4882 # Retrieve $x, $y, ...
4883 my @parameters = @$arrayref[1..$#$arrayref];
4887 If the hook is an object, it must provide an INC method that will be
4888 called as above, the first parameter being the object itself. (Note that
4889 you must fully qualify the sub's name, as unqualified C<INC> is always forced
4890 into package C<main>.) Here is a typical code layout:
4896 my ($self, $filename) = @_;
4900 # In the main program
4901 push @INC, Foo->new(...);
4903 These hooks are also permitted to set the %INC entry
4904 corresponding to the files they have loaded. See L<perlvar/%INC>.
4906 For a yet-more-powerful import facility, see L</use> and L<perlmod>.
4913 Generally used in a C<continue> block at the end of a loop to clear
4914 variables and reset C<??> searches so that they work again. The
4915 expression is interpreted as a list of single characters (hyphens
4916 allowed for ranges). All variables and arrays beginning with one of
4917 those letters are reset to their pristine state. If the expression is
4918 omitted, one-match searches (C<?pattern?>) are reset to match again.
4919 Only resets variables or searches in the current package. Always returns
4922 reset 'X'; # reset all X variables
4923 reset 'a-z'; # reset lower case variables
4924 reset; # just reset ?one-time? searches
4926 Resetting C<"A-Z"> is not recommended because you'll wipe out your
4927 C<@ARGV> and C<@INC> arrays and your C<%ENV> hash. Resets only package
4928 variables; lexical variables are unaffected, but they clean themselves
4929 up on scope exit anyway, so you'll probably want to use them instead.
4937 Returns from a subroutine, C<eval>, or C<do FILE> with the value
4938 given in EXPR. Evaluation of EXPR may be in list, scalar, or void
4939 context, depending on how the return value will be used, and the context
4940 may vary from one execution to the next (see C<wantarray>). If no EXPR
4941 is given, returns an empty list in list context, the undefined value in
4942 scalar context, and (of course) nothing at all in void context.
4944 (In the absence of an explicit C<return>, a subroutine, eval,
4945 or do FILE automatically returns the value of the last expression
4949 X<reverse> X<rev> X<invert>
4951 In list context, returns a list value consisting of the elements
4952 of LIST in the opposite order. In scalar context, concatenates the
4953 elements of LIST and returns a string value with all characters
4954 in the opposite order.
4956 print join(", ", reverse "world", "Hello"); # Hello, world
4958 print scalar reverse "dlrow ,", "olleH"; # Hello, world
4960 Used without arguments in scalar context, reverse() reverses C<$_>.
4962 $_ = "dlrow ,olleH";
4963 print reverse; # No output, list context
4964 print scalar reverse; # Hello, world
4966 Note that reversing an array to itself (as in C<@a = reverse @a>) will
4967 preserve non-existent elements whenever possible, i.e., for non magical
4968 arrays or tied arrays with C<EXISTS> and C<DELETE> methods.
4970 This operator is also handy for inverting a hash, although there are some
4971 caveats. If a value is duplicated in the original hash, only one of those
4972 can be represented as a key in the inverted hash. Also, this has to
4973 unwind one hash and build a whole new one, which may take some time
4974 on a large hash, such as from a DBM file.
4976 %by_name = reverse %by_address; # Invert the hash
4978 =item rewinddir DIRHANDLE
4981 Sets the current position to the beginning of the directory for the
4982 C<readdir> routine on DIRHANDLE.
4984 =item rindex STR,SUBSTR,POSITION
4987 =item rindex STR,SUBSTR
4989 Works just like index() except that it returns the position of the I<last>
4990 occurrence of SUBSTR in STR. If POSITION is specified, returns the
4991 last occurrence beginning at or before that position.
4993 =item rmdir FILENAME
4994 X<rmdir> X<rd> X<directory, remove>
4998 Deletes the directory specified by FILENAME if that directory is
4999 empty. If it succeeds it returns true, otherwise it returns false and
5000 sets C<$!> (errno). If FILENAME is omitted, uses C<$_>.
5002 To remove a directory tree recursively (C<rm -rf> on Unix) look at
5003 the C<rmtree> function of the L<File::Path> module.
5007 The substitution operator. See L<perlop/"Regexp Quote-Like Operators">.
5009 =item say FILEHANDLE LIST
5016 Just like C<print>, but implicitly appends a newline.
5017 C<say LIST> is simply an abbreviation for C<{ local $\ = "\n"; print
5020 This keyword is available only when the "say" feature is
5021 enabled: see L<feature>.
5024 X<scalar> X<context>
5026 Forces EXPR to be interpreted in scalar context and returns the value
5029 @counts = ( scalar @a, scalar @b, scalar @c );
5031 There is no equivalent operator to force an expression to
5032 be interpolated in list context because in practice, this is never
5033 needed. If you really wanted to do so, however, you could use
5034 the construction C<@{[ (some expression) ]}>, but usually a simple
5035 C<(some expression)> suffices.
5037 Because C<scalar> is a unary operator, if you accidentally use for EXPR a
5038 parenthesized list, this behaves as a scalar comma expression, evaluating
5039 all but the last element in void context and returning the final element
5040 evaluated in scalar context. This is seldom what you want.
5042 The following single statement:
5044 print uc(scalar(&foo,$bar)),$baz;
5046 is the moral equivalent of these two:
5049 print(uc($bar),$baz);
5051 See L<perlop> for more details on unary operators and the comma operator.
5053 =item seek FILEHANDLE,POSITION,WHENCE
5054 X<seek> X<fseek> X<filehandle, position>
5056 Sets FILEHANDLE's position, just like the C<fseek> call of C<stdio>.
5057 FILEHANDLE may be an expression whose value gives the name of the
5058 filehandle. The values for WHENCE are C<0> to set the new position
5059 I<in bytes> to POSITION, C<1> to set it to the current position plus
5060 POSITION, and C<2> to set it to EOF plus POSITION (typically
5061 negative). For WHENCE you may use the constants C<SEEK_SET>,
5062 C<SEEK_CUR>, and C<SEEK_END> (start of the file, current position, end
5063 of the file) from the Fcntl module. Returns C<1> on success, C<0>
5066 Note the I<in bytes>: even if the filehandle has been set to
5067 operate on characters (for example by using the C<:encoding(utf8)> open
5068 layer), tell() will return byte offsets, not character offsets
5069 (because implementing that would render seek() and tell() rather slow).
5071 If you want to position the file for C<sysread> or C<syswrite>, don't use
5072 C<seek>, because buffering makes its effect on the file's read-write position
5073 unpredictable and non-portable. Use C<sysseek> instead.
5075 Due to the rules and rigors of ANSI C, on some systems you have to do a
5076 seek whenever you switch between reading and writing. Amongst other
5077 things, this may have the effect of calling stdio's clearerr(3).
5078 A WHENCE of C<1> (C<SEEK_CUR>) is useful for not moving the file position:
5082 This is also useful for applications emulating C<tail -f>. Once you hit
5083 EOF on your read and then sleep for a while, you (probably) have to stick in a
5084 dummy seek() to reset things. The C<seek> doesn't change the position,
5085 but it I<does> clear the end-of-file condition on the handle, so that the
5086 next C<< <FILE> >> makes Perl try again to read something. (We hope.)
5088 If that doesn't work (some I/O implementations are particularly
5089 cantankerous), you might need something like this:
5092 for ($curpos = tell(FILE); $_ = <FILE>;
5093 $curpos = tell(FILE)) {
5094 # search for some stuff and put it into files
5096 sleep($for_a_while);
5097 seek(FILE, $curpos, 0);
5100 =item seekdir DIRHANDLE,POS
5103 Sets the current position for the C<readdir> routine on DIRHANDLE. POS
5104 must be a value returned by C<telldir>. C<seekdir> also has the same caveats
5105 about possible directory compaction as the corresponding system library
5108 =item select FILEHANDLE
5109 X<select> X<filehandle, default>
5113 Returns the currently selected filehandle. If FILEHANDLE is supplied,
5114 sets the new current default filehandle for output. This has two
5115 effects: first, a C<write> or a C<print> without a filehandle will
5116 default to this FILEHANDLE. Second, references to variables related to
5117 output will refer to this output channel. For example, if you have to
5118 set the top of form format for more than one output channel, you might
5126 FILEHANDLE may be an expression whose value gives the name of the
5127 actual filehandle. Thus:
5129 $oldfh = select(STDERR); $| = 1; select($oldfh);
5131 Some programmers may prefer to think of filehandles as objects with
5132 methods, preferring to write the last example as:
5135 STDERR->autoflush(1);
5137 =item select RBITS,WBITS,EBITS,TIMEOUT
5140 This calls the select(2) syscall with the bit masks specified, which
5141 can be constructed using C<fileno> and C<vec>, along these lines:
5143 $rin = $win = $ein = '';
5144 vec($rin,fileno(STDIN),1) = 1;
5145 vec($win,fileno(STDOUT),1) = 1;
5148 If you want to select on many filehandles, you may wish to write a
5149 subroutine like this:
5152 my(@fhlist) = split(' ',$_[0]);
5155 vec($bits,fileno($_),1) = 1;
5159 $rin = fhbits('STDIN TTY SOCK');
5163 ($nfound,$timeleft) =
5164 select($rout=$rin, $wout=$win, $eout=$ein, $timeout);
5166 or to block until something becomes ready just do this
5168 $nfound = select($rout=$rin, $wout=$win, $eout=$ein, undef);
5170 Most systems do not bother to return anything useful in $timeleft, so
5171 calling select() in scalar context just returns $nfound.
5173 Any of the bit masks can also be undef. The timeout, if specified, is
5174 in seconds, which may be fractional. Note: not all implementations are
5175 capable of returning the $timeleft. If not, they always return
5176 $timeleft equal to the supplied $timeout.
5178 You can effect a sleep of 250 milliseconds this way:
5180 select(undef, undef, undef, 0.25);
5182 Note that whether C<select> gets restarted after signals (say, SIGALRM)
5183 is implementation-dependent. See also L<perlport> for notes on the
5184 portability of C<select>.
5186 On error, C<select> behaves like select(2): it returns
5189 On some Unixes, select(2) may report a socket file
5190 descriptor as "ready for reading" when no data is available, and
5191 thus a subsequent read blocks. This can be avoided if you always use
5192 O_NONBLOCK on the socket. See select(2) and fcntl(2) for further
5195 B<WARNING>: One should not attempt to mix buffered I/O (like C<read>
5196 or <FH>) with C<select>, except as permitted by POSIX, and even
5197 then only on POSIX systems. You have to use C<sysread> instead.
5199 =item semctl ID,SEMNUM,CMD,ARG
5202 Calls the System V IPC function semctl(2). You'll probably have to say
5206 first to get the correct constant definitions. If CMD is IPC_STAT or
5207 GETALL, then ARG must be a variable that will hold the returned
5208 semid_ds structure or semaphore value array. Returns like C<ioctl>:
5209 the undefined value for error, "C<0 but true>" for zero, or the actual
5210 return value otherwise. The ARG must consist of a vector of native
5211 short integers, which may be created with C<pack("s!",(0)x$nsem)>.
5212 See also L<perlipc/"SysV IPC">, C<IPC::SysV>, C<IPC::Semaphore>
5215 =item semget KEY,NSEMS,FLAGS
5218 Calls the System V IPC function semget(2). Returns the semaphore id, or
5219 the undefined value if there is an error. See also
5220 L<perlipc/"SysV IPC">, C<IPC::SysV>, C<IPC::SysV::Semaphore>
5223 =item semop KEY,OPSTRING
5226 Calls the System V IPC function semop(2) for semaphore operations
5227 such as signalling and waiting. OPSTRING must be a packed array of
5228 semop structures. Each semop structure can be generated with
5229 C<pack("s!3", $semnum, $semop, $semflag)>. The length of OPSTRING
5230 implies the number of semaphore operations. Returns true if
5231 successful, or false if there is an error. As an example, the
5232 following code waits on semaphore $semnum of semaphore id $semid:
5234 $semop = pack("s!3", $semnum, -1, 0);
5235 die "Semaphore trouble: $!\n" unless semop($semid, $semop);
5237 To signal the semaphore, replace C<-1> with C<1>. See also
5238 L<perlipc/"SysV IPC">, C<IPC::SysV>, and C<IPC::SysV::Semaphore>
5241 =item send SOCKET,MSG,FLAGS,TO
5244 =item send SOCKET,MSG,FLAGS
5246 Sends a message on a socket. Attempts to send the scalar MSG to the SOCKET
5247 filehandle. Takes the same flags as the system call of the same name. On
5248 unconnected sockets, you must specify a destination to I<send to>, in which
5249 case it does a sendto(2) syscall. Returns the number of characters sent,
5250 or the undefined value on error. The sendmsg(2) syscall is currently
5251 unimplemented. See L<perlipc/"UDP: Message Passing"> for examples.
5253 Note the I<characters>: depending on the status of the socket, either
5254 (8-bit) bytes or characters are sent. By default all sockets operate
5255 on bytes, but for example if the socket has been changed using
5256 binmode() to operate with the C<:encoding(utf8)> I/O layer (see
5257 L</open>, or the C<open> pragma, L<open>), the I/O will operate on UTF-8
5258 encoded Unicode characters, not bytes. Similarly for the C<:encoding>
5259 pragma: in that case pretty much any characters can be sent.
5261 =item setpgrp PID,PGRP
5264 Sets the current process group for the specified PID, C<0> for the current
5265 process. Raises an exception when used on a machine that doesn't
5266 implement POSIX setpgid(2) or BSD setpgrp(2). If the arguments are omitted,
5267 it defaults to C<0,0>. Note that the BSD 4.2 version of C<setpgrp> does not
5268 accept any arguments, so only C<setpgrp(0,0)> is portable. See also
5271 =item setpriority WHICH,WHO,PRIORITY
5272 X<setpriority> X<priority> X<nice> X<renice>
5274 Sets the current priority for a process, a process group, or a user.
5275 (See setpriority(2).) Raises an exception when used on a machine
5276 that doesn't implement setpriority(2).
5278 =item setsockopt SOCKET,LEVEL,OPTNAME,OPTVAL
5281 Sets the socket option requested. Returns undefined if there is an
5282 error. Use integer constants provided by the C<Socket> module for
5283 LEVEL and OPNAME. Values for LEVEL can also be obtained from
5284 getprotobyname. OPTVAL might either be a packed string or an integer.
5285 An integer OPTVAL is shorthand for pack("i", OPTVAL).
5287 An example disabling Nagle's algorithm on a socket:
5289 use Socket qw(IPPROTO_TCP TCP_NODELAY);
5290 setsockopt($socket, IPPROTO_TCP, TCP_NODELAY, 1);
5297 Shifts the first value of the array off and returns it, shortening the
5298 array by 1 and moving everything down. If there are no elements in the
5299 array, returns the undefined value. If ARRAY is omitted, shifts the
5300 C<@_> array within the lexical scope of subroutines and formats, and the
5301 C<@ARGV> array outside a subroutine and also within the lexical scopes
5302 established by the C<eval STRING>, C<BEGIN {}>, C<INIT {}>, C<CHECK {}>,
5303 C<UNITCHECK {}> and C<END {}> constructs.
5305 See also C<unshift>, C<push>, and C<pop>. C<shift> and C<unshift> do the
5306 same thing to the left end of an array that C<pop> and C<push> do to the
5309 =item shmctl ID,CMD,ARG
5312 Calls the System V IPC function shmctl. You'll probably have to say
5316 first to get the correct constant definitions. If CMD is C<IPC_STAT>,
5317 then ARG must be a variable that will hold the returned C<shmid_ds>
5318 structure. Returns like ioctl: the undefined value for error, "C<0> but
5319 true" for zero, or the actual return value otherwise.
5320 See also L<perlipc/"SysV IPC"> and C<IPC::SysV> documentation.
5322 =item shmget KEY,SIZE,FLAGS
5325 Calls the System V IPC function shmget. Returns the shared memory
5326 segment id, or the undefined value if there is an error.
5327 See also L<perlipc/"SysV IPC"> and C<IPC::SysV> documentation.
5329 =item shmread ID,VAR,POS,SIZE
5333 =item shmwrite ID,STRING,POS,SIZE
5335 Reads or writes the System V shared memory segment ID starting at
5336 position POS for size SIZE by attaching to it, copying in/out, and
5337 detaching from it. When reading, VAR must be a variable that will
5338 hold the data read. When writing, if STRING is too long, only SIZE
5339 bytes are used; if STRING is too short, nulls are written to fill out
5340 SIZE bytes. Return true if successful, or false if there is an error.
5341 shmread() taints the variable. See also L<perlipc/"SysV IPC">,
5342 C<IPC::SysV> documentation, and the C<IPC::Shareable> module from CPAN.
5344 =item shutdown SOCKET,HOW
5347 Shuts down a socket connection in the manner indicated by HOW, which
5348 has the same interpretation as in the syscall of the same name.
5350 shutdown(SOCKET, 0); # I/we have stopped reading data
5351 shutdown(SOCKET, 1); # I/we have stopped writing data
5352 shutdown(SOCKET, 2); # I/we have stopped using this socket
5354 This is useful with sockets when you want to tell the other
5355 side you're done writing but not done reading, or vice versa.
5356 It's also a more insistent form of close because it also
5357 disables the file descriptor in any forked copies in other
5360 Returns C<1> for success; on error, returns C<undef> if
5361 the first argument is not a valid filehandle, or returns C<0> and sets
5362 C<$!> for any other failure.
5365 X<sin> X<sine> X<asin> X<arcsine>
5369 Returns the sine of EXPR (expressed in radians). If EXPR is omitted,
5370 returns sine of C<$_>.
5372 For the inverse sine operation, you may use the C<Math::Trig::asin>
5373 function, or use this relation:
5375 sub asin { atan2($_[0], sqrt(1 - $_[0] * $_[0])) }
5382 Causes the script to sleep for (integer) EXPR seconds, or forever if no
5383 argument is given. Returns the integer number of seconds actually slept.
5385 May be interrupted if the process receives a signal such as C<SIGALRM>.
5388 local $SIG{ALARM} = sub { die "Alarm!\n" };
5391 die $@ unless $@ eq "Alarm!\n";
5393 You probably cannot mix C<alarm> and C<sleep> calls, because C<sleep>
5394 is often implemented using C<alarm>.
5396 On some older systems, it may sleep up to a full second less than what
5397 you requested, depending on how it counts seconds. Most modern systems
5398 always sleep the full amount. They may appear to sleep longer than that,
5399 however, because your process might not be scheduled right away in a
5400 busy multitasking system.
5402 For delays of finer granularity than one second, the Time::HiRes module
5403 (from CPAN, and starting from Perl 5.8 part of the standard
5404 distribution) provides usleep(). You may also use Perl's four-argument
5405 version of select() leaving the first three arguments undefined, or you
5406 might be able to use the C<syscall> interface to access setitimer(2) if
5407 your system supports it. See L<perlfaq8> for details.
5409 See also the POSIX module's C<pause> function.
5411 =item socket SOCKET,DOMAIN,TYPE,PROTOCOL
5414 Opens a socket of the specified kind and attaches it to filehandle
5415 SOCKET. DOMAIN, TYPE, and PROTOCOL are specified the same as for
5416 the syscall of the same name. You should C<use Socket> first
5417 to get the proper definitions imported. See the examples in
5418 L<perlipc/"Sockets: Client/Server Communication">.
5420 On systems that support a close-on-exec flag on files, the flag will
5421 be set for the newly opened file descriptor, as determined by the
5422 value of $^F. See L<perlvar/$^F>.
5424 =item socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL
5427 Creates an unnamed pair of sockets in the specified domain, of the
5428 specified type. DOMAIN, TYPE, and PROTOCOL are specified the same as
5429 for the syscall of the same name. If unimplemented, raises an exception.
5430 Returns true if successful.
5432 On systems that support a close-on-exec flag on files, the flag will
5433 be set for the newly opened file descriptors, as determined by the value
5434 of $^F. See L<perlvar/$^F>.
5436 Some systems defined C<pipe> in terms of C<socketpair>, in which a call
5437 to C<pipe(Rdr, Wtr)> is essentially:
5440 socketpair(Rdr, Wtr, AF_UNIX, SOCK_STREAM, PF_UNSPEC);
5441 shutdown(Rdr, 1); # no more writing for reader
5442 shutdown(Wtr, 0); # no more reading for writer
5444 See L<perlipc> for an example of socketpair use. Perl 5.8 and later will
5445 emulate socketpair using IP sockets to localhost if your system implements
5446 sockets but not socketpair.
5448 =item sort SUBNAME LIST
5449 X<sort> X<qsort> X<quicksort> X<mergesort>
5451 =item sort BLOCK LIST
5455 In list context, this sorts the LIST and returns the sorted list value.
5456 In scalar context, the behaviour of C<sort()> is undefined.
5458 If SUBNAME or BLOCK is omitted, C<sort>s in standard string comparison
5459 order. If SUBNAME is specified, it gives the name of a subroutine
5460 that returns an integer less than, equal to, or greater than C<0>,
5461 depending on how the elements of the list are to be ordered. (The
5462 C<< <=> >> and C<cmp> operators are extremely useful in such routines.)
5463 SUBNAME may be a scalar variable name (unsubscripted), in which case
5464 the value provides the name of (or a reference to) the actual
5465 subroutine to use. In place of a SUBNAME, you can provide a BLOCK as
5466 an anonymous, in-line sort subroutine.
5468 If the subroutine's prototype is C<($$)>, the elements to be compared
5469 are passed by reference in C<@_>, as for a normal subroutine. This is
5470 slower than unprototyped subroutines, where the elements to be
5471 compared are passed into the subroutine
5472 as the package global variables $a and $b (see example below). Note that
5473 in the latter case, it is usually counter-productive to declare $a and
5476 The values to be compared are always passed by reference and should not
5479 You also cannot exit out of the sort block or subroutine using any of the
5480 loop control operators described in L<perlsyn> or with C<goto>.
5482 When C<use locale> is in effect, C<sort LIST> sorts LIST according to the
5483 current collation locale. See L<perllocale>.
5485 sort() returns aliases into the original list, much as a for loop's index
5486 variable aliases the list elements. That is, modifying an element of a
5487 list returned by sort() (for example, in a C<foreach>, C<map> or C<grep>)
5488 actually modifies the element in the original list. This is usually
5489 something to be avoided when writing clear code.
5491 Perl 5.6 and earlier used a quicksort algorithm to implement sort.
5492 That algorithm was not stable, and I<could> go quadratic. (A I<stable> sort
5493 preserves the input order of elements that compare equal. Although
5494 quicksort's run time is O(NlogN) when averaged over all arrays of
5495 length N, the time can be O(N**2), I<quadratic> behavior, for some
5496 inputs.) In 5.7, the quicksort implementation was replaced with
5497 a stable mergesort algorithm whose worst-case behavior is O(NlogN).
5498 But benchmarks indicated that for some inputs, on some platforms,
5499 the original quicksort was faster. 5.8 has a sort pragma for
5500 limited control of the sort. Its rather blunt control of the
5501 underlying algorithm may not persist into future Perls, but the
5502 ability to characterize the input or output in implementation
5503 independent ways quite probably will. See L<the sort pragma|sort>.
5508 @articles = sort @files;
5510 # same thing, but with explicit sort routine
5511 @articles = sort {$a cmp $b} @files;
5513 # now case-insensitively
5514 @articles = sort {uc($a) cmp uc($b)} @files;
5516 # same thing in reversed order
5517 @articles = sort {$b cmp $a} @files;
5519 # sort numerically ascending
5520 @articles = sort {$a <=> $b} @files;
5522 # sort numerically descending
5523 @articles = sort {$b <=> $a} @files;
5525 # this sorts the %age hash by value instead of key
5526 # using an in-line function
5527 @eldest = sort { $age{$b} <=> $age{$a} } keys %age;
5529 # sort using explicit subroutine name
5531 $age{$a} <=> $age{$b}; # presuming numeric
5533 @sortedclass = sort byage @class;
5535 sub backwards { $b cmp $a }
5536 @harry = qw(dog cat x Cain Abel);
5537 @george = qw(gone chased yz Punished Axed);
5539 # prints AbelCaincatdogx
5540 print sort backwards @harry;
5541 # prints xdogcatCainAbel
5542 print sort @george, 'to', @harry;
5543 # prints AbelAxedCainPunishedcatchaseddoggonetoxyz
5545 # inefficiently sort by descending numeric compare using
5546 # the first integer after the first = sign, or the
5547 # whole record case-insensitively otherwise
5550 ($b =~ /=(\d+)/)[0] <=> ($a =~ /=(\d+)/)[0]
5555 # same thing, but much more efficiently;
5556 # we'll build auxiliary indices instead
5558 my @nums = @caps = ();
5560 push @nums, ( /=(\d+)/ ? $1 : undef );
5564 my @new = @old[ sort {
5565 $nums[$b] <=> $nums[$a]
5567 $caps[$a] cmp $caps[$b]
5571 # same thing, but without any temps
5572 @new = map { $_->[0] }
5573 sort { $b->[1] <=> $a->[1]
5576 } map { [$_, /=(\d+)/, uc($_)] } @old;
5578 # using a prototype allows you to use any comparison subroutine
5579 # as a sort subroutine (including other package's subroutines)
5581 sub backwards ($$) { $_[1] cmp $_[0]; } # $a and $b are not set here
5584 @new = sort other::backwards @old;
5586 # guarantee stability, regardless of algorithm
5588 @new = sort { substr($a, 3, 5) cmp substr($b, 3, 5) } @old;
5590 # force use of mergesort (not portable outside Perl 5.8)
5591 use sort '_mergesort'; # note discouraging _
5592 @new = sort { substr($a, 3, 5) cmp substr($b, 3, 5) } @old;
5594 Warning: syntactical care is required when sorting the list returned from
5595 a function. If you want to sort the list returned by the function call
5596 C<find_records(@key)>, you can use:
5598 @contact = sort { $a cmp $b } find_records @key;
5599 @contact = sort +find_records(@key);
5600 @contact = sort &find_records(@key);
5601 @contact = sort(find_records(@key));
5603 If instead you want to sort the array @key with the comparison routine
5604 C<find_records()> then you can use:
5606 @contact = sort { find_records() } @key;
5607 @contact = sort find_records(@key);
5608 @contact = sort(find_records @key);
5609 @contact = sort(find_records (@key));
5611 If you're using strict, you I<must not> declare $a
5612 and $b as lexicals. They are package globals. That means
5613 that if you're in the C<main> package and type
5615 @articles = sort {$b <=> $a} @files;
5617 then C<$a> and C<$b> are C<$main::a> and C<$main::b> (or C<$::a> and C<$::b>),
5618 but if you're in the C<FooPack> package, it's the same as typing
5620 @articles = sort {$FooPack::b <=> $FooPack::a} @files;
5622 The comparison function is required to behave. If it returns
5623 inconsistent results (sometimes saying C<$x[1]> is less than C<$x[2]> and
5624 sometimes saying the opposite, for example) the results are not
5627 Because C<< <=> >> returns C<undef> when either operand is C<NaN>
5628 (not-a-number), and because C<sort> raises an exception unless the
5629 result of a comparison is defined, when sorting with a comparison function
5630 like C<< $a <=> $b >>, be careful about lists that might contain a C<NaN>.
5631 The following example takes advantage that C<NaN != NaN> to
5632 eliminate any C<NaN>s from the input list.
5634 @result = sort { $a <=> $b } grep { $_ == $_ } @input;
5636 =item splice ARRAY,OFFSET,LENGTH,LIST
5639 =item splice ARRAY,OFFSET,LENGTH
5641 =item splice ARRAY,OFFSET
5645 Removes the elements designated by OFFSET and LENGTH from an array, and
5646 replaces them with the elements of LIST, if any. In list context,
5647 returns the elements removed from the array. In scalar context,
5648 returns the last element removed, or C<undef> if no elements are
5649 removed. The array grows or shrinks as necessary.
5650 If OFFSET is negative then it starts that far from the end of the array.
5651 If LENGTH is omitted, removes everything from OFFSET onward.
5652 If LENGTH is negative, removes the elements from OFFSET onward
5653 except for -LENGTH elements at the end of the array.
5654 If both OFFSET and LENGTH are omitted, removes everything. If OFFSET is
5655 past the end of the array, Perl issues a warning, and splices at the
5658 The following equivalences hold (assuming C<< $[ == 0 and $#a >= $i >> )
5660 push(@a,$x,$y) splice(@a,@a,0,$x,$y)
5661 pop(@a) splice(@a,-1)
5662 shift(@a) splice(@a,0,1)
5663 unshift(@a,$x,$y) splice(@a,0,0,$x,$y)
5664 $a[$i] = $y splice(@a,$i,1,$y)
5666 Example, assuming array lengths are passed before arrays:
5668 sub aeq { # compare two list values
5669 my(@a) = splice(@_,0,shift);
5670 my(@b) = splice(@_,0,shift);
5671 return 0 unless @a == @b; # same len?
5673 return 0 if pop(@a) ne pop(@b);
5677 if (&aeq($len,@foo[1..$len],0+@bar,@bar)) { ... }
5679 =item split /PATTERN/,EXPR,LIMIT
5682 =item split /PATTERN/,EXPR
5684 =item split /PATTERN/
5688 Splits the string EXPR into a list of strings and returns that list. By
5689 default, empty leading fields are preserved, and empty trailing ones are
5690 deleted. (If all fields are empty, they are considered to be trailing.)
5692 In scalar context, returns the number of fields found.
5694 If EXPR is omitted, splits the C<$_> string. If PATTERN is also omitted,
5695 splits on whitespace (after skipping any leading whitespace). Anything
5696 matching PATTERN is taken to be a delimiter separating the fields. (Note
5697 that the delimiter may be longer than one character.)
5699 If LIMIT is specified and positive, it represents the maximum number
5700 of fields the EXPR will be split into, though the actual number of
5701 fields returned depends on the number of times PATTERN matches within
5702 EXPR. If LIMIT is unspecified or zero, trailing null fields are
5703 stripped (which potential users of C<pop> would do well to remember).
5704 If LIMIT is negative, it is treated as if an arbitrarily large LIMIT
5705 had been specified. Note that splitting an EXPR that evaluates to the
5706 empty string always returns the empty list, regardless of the LIMIT
5709 A pattern matching the empty string (not to be confused with
5710 an empty pattern C<//>, which is just one member of the set of patterns
5711 matching the epmty string), splits EXPR into individual
5712 characters. For example:
5714 print join(':', split(/ */, 'hi there')), "\n";
5716 produces the output 'h:i:t:h:e:r:e'.
5718 As a special case for C<split>, the empty pattern C<//> specifically
5719 matches the empty string; this is not be confused with the normal use
5720 of an empty pattern to mean the last successful match. So to split
5721 a string into individual characters, the following:
5723 print join(':', split(//, 'hi there')), "\n";
5725 produces the output 'h:i: :t:h:e:r:e'.
5727 Empty leading fields are produced when there are positive-width matches at
5728 the beginning of the string; a zero-width match at the beginning of
5729 the string does not produce an empty field. For example:
5731 print join(':', split(/(?=\w)/, 'hi there!'));
5733 produces the output 'h:i :t:h:e:r:e!'. Empty trailing fields, on the other
5734 hand, are produced when there is a match at the end of the string (and
5735 when LIMIT is given and is not 0), regardless of the length of the match.
5738 print join(':', split(//, 'hi there!', -1)), "\n";
5739 print join(':', split(/\W/, 'hi there!', -1)), "\n";
5741 produce the output 'h:i: :t:h:e:r:e:!:' and 'hi:there:', respectively,
5742 both with an empty trailing field.
5744 The LIMIT parameter can be used to split a line partially
5746 ($login, $passwd, $remainder) = split(/:/, $_, 3);
5748 When assigning to a list, if LIMIT is omitted, or zero, Perl supplies
5749 a LIMIT one larger than the number of variables in the list, to avoid
5750 unnecessary work. For the list above LIMIT would have been 4 by
5751 default. In time critical applications it behooves you not to split
5752 into more fields than you really need.
5754 If the PATTERN contains parentheses, additional list elements are
5755 created from each matching substring in the delimiter.
5757 split(/([,-])/, "1-10,20", 3);
5759 produces the list value
5761 (1, '-', 10, ',', 20)
5763 If you had the entire header of a normal Unix email message in $header,
5764 you could split it up into fields and their values this way:
5766 $header =~ s/\n(?=\s)//g; # fix continuation lines
5767 %hdrs = (UNIX_FROM => split /^(\S*?):\s*/m, $header);
5769 The pattern C</PATTERN/> may be replaced with an expression to specify
5770 patterns that vary at runtime. (To do runtime compilation only once,
5771 use C</$variable/o>.)
5773 As a special case, specifying a PATTERN of space (S<C<' '>>) will split on
5774 white space just as C<split> with no arguments does. Thus, S<C<split(' ')>> can
5775 be used to emulate B<awk>'s default behavior, whereas S<C<split(/ /)>>
5776 will give you as many initial null fields (empty string) as there are leading spaces.
5777 A C<split> on C</\s+/> is like a S<C<split(' ')>> except that any leading
5778 whitespace produces a null first field. A C<split> with no arguments
5779 really does a S<C<split(' ', $_)>> internally.
5781 A PATTERN of C</^/> is treated as if it were C</^/m>, since it isn't
5786 open(PASSWD, '/etc/passwd');
5789 ($login, $passwd, $uid, $gid,
5790 $gcos, $home, $shell) = split(/:/);
5794 As with regular pattern matching, any capturing parentheses that are not
5795 matched in a C<split()> will be set to C<undef> when returned:
5797 @fields = split /(A)|B/, "1A2B3";
5798 # @fields is (1, 'A', 2, undef, 3)
5800 =item sprintf FORMAT, LIST
5803 Returns a string formatted by the usual C<printf> conventions of the C
5804 library function C<sprintf>. See below for more details
5805 and see C<sprintf(3)> or C<printf(3)> on your system for an explanation of
5806 the general principles.
5810 # Format number with up to 8 leading zeroes
5811 $result = sprintf("%08d", $number);
5813 # Round number to 3 digits after decimal point
5814 $rounded = sprintf("%.3f", $number);
5816 Perl does its own C<sprintf> formatting: it emulates the C
5817 function sprintf(3), but doesn't use it except for floating-point
5818 numbers, and even then only standard modifiers are allowed.
5819 Non-standard extensions in your local sprintf(3) are
5820 therefore unavailable from Perl.
5822 Unlike C<printf>, C<sprintf> does not do what you probably mean when you
5823 pass it an array as your first argument. The array is given scalar context,
5824 and instead of using the 0th element of the array as the format, Perl will
5825 use the count of elements in the array as the format, which is almost never
5828 Perl's C<sprintf> permits the following universally-known conversions:
5831 %c a character with the given number
5833 %d a signed integer, in decimal
5834 %u an unsigned integer, in decimal
5835 %o an unsigned integer, in octal
5836 %x an unsigned integer, in hexadecimal
5837 %e a floating-point number, in scientific notation
5838 %f a floating-point number, in fixed decimal notation
5839 %g a floating-point number, in %e or %f notation
5841 In addition, Perl permits the following widely-supported conversions:
5843 %X like %x, but using upper-case letters
5844 %E like %e, but using an upper-case "E"
5845 %G like %g, but with an upper-case "E" (if applicable)
5846 %b an unsigned integer, in binary
5847 %B like %b, but using an upper-case "B" with the # flag
5848 %p a pointer (outputs the Perl value's address in hexadecimal)
5849 %n special: *stores* the number of characters output so far
5850 into the next variable in the parameter list
5852 Finally, for backward (and we do mean "backward") compatibility, Perl
5853 permits these unnecessary but widely-supported conversions:
5856 %D a synonym for %ld
5857 %U a synonym for %lu
5858 %O a synonym for %lo
5861 Note that the number of exponent digits in the scientific notation produced
5862 by C<%e>, C<%E>, C<%g> and C<%G> for numbers with the modulus of the
5863 exponent less than 100 is system-dependent: it may be three or less
5864 (zero-padded as necessary). In other words, 1.23 times ten to the
5865 99th may be either "1.23e99" or "1.23e099".
5867 Between the C<%> and the format letter, you may specify several
5868 additional attributes controlling the interpretation of the format.
5869 In order, these are:
5873 =item format parameter index
5875 An explicit format parameter index, such as C<2$>. By default sprintf
5876 will format the next unused argument in the list, but this allows you
5877 to take the arguments out of order:
5879 printf '%2$d %1$d', 12, 34; # prints "34 12"
5880 printf '%3$d %d %1$d', 1, 2, 3; # prints "3 1 1"
5886 space prefix non-negative number with a space
5887 + prefix non-negative number with a plus sign
5888 - left-justify within the field
5889 0 use zeros, not spaces, to right-justify
5890 # ensure the leading "0" for any octal,
5891 prefix non-zero hexadecimal with "0x" or "0X",
5892 prefix non-zero binary with "0b" or "0B"
5896 printf '<% d>', 12; # prints "< 12>"
5897 printf '<%+d>', 12; # prints "<+12>"
5898 printf '<%6s>', 12; # prints "< 12>"
5899 printf '<%-6s>', 12; # prints "<12 >"
5900 printf '<%06s>', 12; # prints "<000012>"
5901 printf '<%#o>', 12; # prints "<014>"
5902 printf '<%#x>', 12; # prints "<0xc>"
5903 printf '<%#X>', 12; # prints "<0XC>"
5904 printf '<%#b>', 12; # prints "<0b1100>"
5905 printf '<%#B>', 12; # prints "<0B1100>"
5907 When a space and a plus sign are given as the flags at once,
5908 a plus sign is used to prefix a positive number.
5910 printf '<%+ d>', 12; # prints "<+12>"
5911 printf '<% +d>', 12; # prints "<+12>"
5913 When the # flag and a precision are given in the %o conversion,
5914 the precision is incremented if it's necessary for the leading "0".
5916 printf '<%#.5o>', 012; # prints "<00012>"
5917 printf '<%#.5o>', 012345; # prints "<012345>"
5918 printf '<%#.0o>', 0; # prints "<0>"
5922 This flag tells Perl to interpret the supplied string as a vector of
5923 integers, one for each character in the string. Perl applies the format to
5924 each integer in turn, then joins the resulting strings with a separator (a
5925 dot C<.> by default). This can be useful for displaying ordinal values of
5926 characters in arbitrary strings:
5928 printf "%vd", "AB\x{100}"; # prints "65.66.256"
5929 printf "version is v%vd\n", $^V; # Perl's version
5931 Put an asterisk C<*> before the C<v> to override the string to
5932 use to separate the numbers:
5934 printf "address is %*vX\n", ":", $addr; # IPv6 address
5935 printf "bits are %0*v8b\n", " ", $bits; # random bitstring
5937 You can also explicitly specify the argument number to use for
5938 the join string using something like C<*2$v>; for example:
5940 printf '%*4$vX %*4$vX %*4$vX', @addr[1..3], ":"; # 3 IPv6 addresses
5942 =item (minimum) width
5944 Arguments are usually formatted to be only as wide as required to
5945 display the given value. You can override the width by putting
5946 a number here, or get the width from the next argument (with C<*>)
5947 or from a specified argument (e.g., with C<*2$>):
5949 printf '<%s>', "a"; # prints "<a>"
5950 printf '<%6s>', "a"; # prints "< a>"
5951 printf '<%*s>', 6, "a"; # prints "< a>"
5952 printf '<%*2$s>', "a", 6; # prints "< a>"
5953 printf '<%2s>', "long"; # prints "<long>" (does not truncate)
5955 If a field width obtained through C<*> is negative, it has the same
5956 effect as the C<-> flag: left-justification.
5958 =item precision, or maximum width
5961 You can specify a precision (for numeric conversions) or a maximum
5962 width (for string conversions) by specifying a C<.> followed by a number.
5963 For floating-point formats except 'g' and 'G', this specifies
5964 how many places right of the decimal point to show (the default being 6).
5967 # these examples are subject to system-specific variation
5968 printf '<%f>', 1; # prints "<1.000000>"
5969 printf '<%.1f>', 1; # prints "<1.0>"
5970 printf '<%.0f>', 1; # prints "<1>"
5971 printf '<%e>', 10; # prints "<1.000000e+01>"
5972 printf '<%.1e>', 10; # prints "<1.0e+01>"
5974 For "g" and "G", this specifies the maximum number of digits to show,
5975 including thoe prior to the decimal point and those after it; for
5978 # These examples are subject to system-specific variation.
5979 printf '<%g>', 1; # prints "<1>"
5980 printf '<%.10g>', 1; # prints "<1>"
5981 printf '<%g>', 100; # prints "<100>"
5982 printf '<%.1g>', 100; # prints "<1e+02>"
5983 printf '<%.2g>', 100.01; # prints "<1e+02>"
5984 printf '<%.5g>', 100.01; # prints "<100.01>"
5985 printf '<%.4g>', 100.01; # prints "<100>"
5987 For integer conversions, specifying a precision implies that the
5988 output of the number itself should be zero-padded to this width,
5989 where the 0 flag is ignored:
5991 printf '<%.6d>', 1; # prints "<000001>"
5992 printf '<%+.6d>', 1; # prints "<+000001>"
5993 printf '<%-10.6d>', 1; # prints "<000001 >"
5994 printf '<%10.6d>', 1; # prints "< 000001>"
5995 printf '<%010.6d>', 1; # prints "< 000001>"
5996 printf '<%+10.6d>', 1; # prints "< +000001>"
5998 printf '<%.6x>', 1; # prints "<000001>"
5999 printf '<%#.6x>', 1; # prints "<0x000001>"
6000 printf '<%-10.6x>', 1; # prints "<000001 >"
6001 printf '<%10.6x>', 1; # prints "< 000001>"
6002 printf '<%010.6x>', 1; # prints "< 000001>"
6003 printf '<%#10.6x>', 1; # prints "< 0x000001>"
6005 For string conversions, specifying a precision truncates the string
6006 to fit the specified width:
6008 printf '<%.5s>', "truncated"; # prints "<trunc>"
6009 printf '<%10.5s>', "truncated"; # prints "< trunc>"
6011 You can also get the precision from the next argument using C<.*>:
6013 printf '<%.6x>', 1; # prints "<000001>"
6014 printf '<%.*x>', 6, 1; # prints "<000001>"
6016 If a precision obtained through C<*> is negative, it counts
6017 as having no precision at all.
6019 printf '<%.*s>', 7, "string"; # prints "<string>"
6020 printf '<%.*s>', 3, "string"; # prints "<str>"
6021 printf '<%.*s>', 0, "string"; # prints "<>"
6022 printf '<%.*s>', -1, "string"; # prints "<string>"
6024 printf '<%.*d>', 1, 0; # prints "<0>"
6025 printf '<%.*d>', 0, 0; # prints "<>"
6026 printf '<%.*d>', -1, 0; # prints "<0>"
6028 You cannot currently get the precision from a specified number,
6029 but it is intended that this will be possible in the future, for
6030 example using C<.*2$>:
6032 printf "<%.*2$x>", 1, 6; # INVALID, but in future will print "<000001>"
6036 For numeric conversions, you can specify the size to interpret the
6037 number as using C<l>, C<h>, C<V>, C<q>, C<L>, or C<ll>. For integer
6038 conversions (C<d u o x X b i D U O>), numbers are usually assumed to be
6039 whatever the default integer size is on your platform (usually 32 or 64
6040 bits), but you can override this to use instead one of the standard C types,
6041 as supported by the compiler used to build Perl:
6043 l interpret integer as C type "long" or "unsigned long"
6044 h interpret integer as C type "short" or "unsigned short"
6045 q, L or ll interpret integer as C type "long long", "unsigned long long".
6046 or "quads" (typically 64-bit integers)
6048 The last will raise an exception if Perl does not understand "quads" in your
6049 installation. (This requires either that the platform natively support quads,
6050 or that Perl were specifically compiled to support quads.) You can find out
6051 whether your Perl supports quads via L<Config>:
6054 if ($Config{use64bitint} eq "define" || $Config{longsize} >= 8) {
6055 print "Nice quads!\n";
6058 For floating-point conversions (C<e f g E F G>), numbers are usually assumed
6059 to be the default floating-point size on your platform (double or long double),
6060 but you can force "long double" with C<q>, C<L>, or C<ll> if your
6061 platform supports them. You can find out whether your Perl supports long
6062 doubles via L<Config>:
6065 print "long doubles\n" if $Config{d_longdbl} eq "define";
6067 You can find out whether Perl considers "long double" to be the default
6068 floating-point size to use on your platform via L<Config>:
6071 if ($Config{uselongdouble} eq "define") {
6072 print "long doubles by default\n";
6075 It can also be that long doubles and doubles are the same thing:
6078 ($Config{doublesize} == $Config{longdblsize}) &&
6079 print "doubles are long doubles\n";
6081 The size specifier C<V> has no effect for Perl code, but is supported for
6082 compatibility with XS code. It means "use the standard size for a Perl
6083 integer or floating-point number", which is the default.
6085 =item order of arguments
6087 Normally, sprintf() takes the next unused argument as the value to
6088 format for each format specification. If the format specification
6089 uses C<*> to require additional arguments, these are consumed from
6090 the argument list in the order they appear in the format
6091 specification I<before> the value to format. Where an argument is
6092 specified by an explicit index, this does not affect the normal
6093 order for the arguments, even when the explicitly specified index
6094 would have been the next argument.
6098 printf "<%*.*s>", $a, $b, $c;
6100 uses C<$a> for the width, C<$b> for the precision, and C<$c>
6101 as the value to format; while:
6103 printf "<%*1$.*s>", $a, $b;
6105 would use C<$a> for the width and precision, and C<$b> as the
6108 Here are some more examples; be aware that when using an explicit
6109 index, the C<$> may need escaping:
6111 printf "%2\$d %d\n", 12, 34; # will print "34 12\n"
6112 printf "%2\$d %d %d\n", 12, 34; # will print "34 12 34\n"
6113 printf "%3\$d %d %d\n", 12, 34, 56; # will print "56 12 34\n"
6114 printf "%2\$*3\$d %d\n", 12, 34, 3; # will print " 34 12\n"
6118 If C<use locale> is in effect and POSIX::setlocale() has been called,
6119 the character used for the decimal separator in formatted floating-point
6120 numbers is affected by the LC_NUMERIC locale. See L<perllocale>
6124 X<sqrt> X<root> X<square root>
6128 Return the positive square root of EXPR. If EXPR is omitted, uses
6129 C<$_>. Works only for non-negative operands unless you've
6130 loaded the C<Math::Complex> module.
6133 print sqrt(-4); # prints 2i
6136 X<srand> X<seed> X<randseed>
6140 Sets the random number seed for the C<rand> operator.
6142 The point of the function is to "seed" the C<rand> function so that
6143 C<rand> can produce a different sequence each time you run your
6146 If srand() is not called explicitly, it is called implicitly at the
6147 first use of the C<rand> operator. However, this was not true of
6148 versions of Perl before 5.004, so if your script will run under older
6149 Perl versions, it should call C<srand>.
6151 Most programs won't even call srand() at all, except those that
6152 need a cryptographically-strong starting point rather than the
6153 generally acceptable default, which is based on time of day,
6154 process ID, and memory allocation, or the F</dev/urandom> device
6157 You can call srand($seed) with the same $seed to reproduce the
6158 I<same> sequence from rand(), but this is usually reserved for
6159 generating predictable results for testing or debugging.
6160 Otherwise, don't call srand() more than once in your program.
6162 Do B<not> call srand() (i.e., without an argument) more than once in
6163 a script. The internal state of the random number generator should
6164 contain more entropy than can be provided by any seed, so calling
6165 srand() again actually I<loses> randomness.
6167 Most implementations of C<srand> take an integer and will silently
6168 truncate decimal numbers. This means C<srand(42)> will usually
6169 produce the same results as C<srand(42.1)>. To be safe, always pass
6170 C<srand> an integer.
6172 In versions of Perl prior to 5.004 the default seed was just the
6173 current C<time>. This isn't a particularly good seed, so many old
6174 programs supply their own seed value (often C<time ^ $$> or C<time ^
6175 ($$ + ($$ << 15))>), but that isn't necessary any more.
6177 For cryptographic purposes, however, you need something much more random
6178 than the default seed. Checksumming the compressed output of one or more
6179 rapidly changing operating system status programs is the usual method. For
6182 srand (time ^ $$ ^ unpack "%L*", `ps axww | gzip -f`);
6184 If you're particularly concerned with this, see the C<Math::TrulyRandom>
6187 Frequently called programs (like CGI scripts) that simply use
6191 for a seed can fall prey to the mathematical property that
6195 one-third of the time. So don't do that.
6197 =item stat FILEHANDLE
6198 X<stat> X<file, status> X<ctime>
6202 =item stat DIRHANDLE
6206 Returns a 13-element list giving the status info for a file, either
6207 the file opened via FILEHANDLE or DIRHANDLE, or named by EXPR. If EXPR is
6208 omitted, it stats C<$_>. Returns the empty list if C<stat> fails. Typically
6211 ($dev,$ino,$mode,$nlink,$uid,$gid,$rdev,$size,
6212 $atime,$mtime,$ctime,$blksize,$blocks)
6215 Not all fields are supported on all filesystem types. Here are the
6216 meanings of the fields:
6218 0 dev device number of filesystem
6220 2 mode file mode (type and permissions)
6221 3 nlink number of (hard) links to the file
6222 4 uid numeric user ID of file's owner
6223 5 gid numeric group ID of file's owner
6224 6 rdev the device identifier (special files only)
6225 7 size total size of file, in bytes
6226 8 atime last access time in seconds since the epoch
6227 9 mtime last modify time in seconds since the epoch
6228 10 ctime inode change time in seconds since the epoch (*)
6229 11 blksize preferred block size for file system I/O
6230 12 blocks actual number of blocks allocated
6232 (The epoch was at 00:00 January 1, 1970 GMT.)
6234 (*) Not all fields are supported on all filesystem types. Notably, the
6235 ctime field is non-portable. In particular, you cannot expect it to be a
6236 "creation time", see L<perlport/"Files and Filesystems"> for details.
6238 If C<stat> is passed the special filehandle consisting of an underline, no
6239 stat is done, but the current contents of the stat structure from the
6240 last C<stat>, C<lstat>, or filetest are returned. Example:
6242 if (-x $file && (($d) = stat(_)) && $d < 0) {
6243 print "$file is executable NFS file\n";
6246 (This works on machines only for which the device number is negative
6249 Because the mode contains both the file type and its permissions, you
6250 should mask off the file type portion and (s)printf using a C<"%o">
6251 if you want to see the real permissions.
6253 $mode = (stat($filename))[2];
6254 printf "Permissions are %04o\n", $mode & 07777;
6256 In scalar context, C<stat> returns a boolean value indicating success
6257 or failure, and, if successful, sets the information associated with
6258 the special filehandle C<_>.
6260 The L<File::stat> module provides a convenient, by-name access mechanism:
6263 $sb = stat($filename);
6264 printf "File is %s, size is %s, perm %04o, mtime %s\n",
6265 $filename, $sb->size, $sb->mode & 07777,
6266 scalar localtime $sb->mtime;
6268 You can import symbolic mode constants (C<S_IF*>) and functions
6269 (C<S_IS*>) from the Fcntl module:
6273 $mode = (stat($filename))[2];
6275 $user_rwx = ($mode & S_IRWXU) >> 6;
6276 $group_read = ($mode & S_IRGRP) >> 3;
6277 $other_execute = $mode & S_IXOTH;
6279 printf "Permissions are %04o\n", S_IMODE($mode), "\n";
6281 $is_setuid = $mode & S_ISUID;
6282 $is_directory = S_ISDIR($mode);
6284 You could write the last two using the C<-u> and C<-d> operators.
6285 Commonly available C<S_IF*> constants are:
6287 # Permissions: read, write, execute, for user, group, others.
6289 S_IRWXU S_IRUSR S_IWUSR S_IXUSR
6290 S_IRWXG S_IRGRP S_IWGRP S_IXGRP
6291 S_IRWXO S_IROTH S_IWOTH S_IXOTH
6293 # Setuid/Setgid/Stickiness/SaveText.
6294 # Note that the exact meaning of these is system dependent.
6296 S_ISUID S_ISGID S_ISVTX S_ISTXT
6298 # File types. Not necessarily all are available on your system.
6300 S_IFREG S_IFDIR S_IFLNK S_IFBLK S_IFCHR S_IFIFO S_IFSOCK S_IFWHT S_ENFMT
6302 # The following are compatibility aliases for S_IRUSR, S_IWUSR, S_IXUSR.
6304 S_IREAD S_IWRITE S_IEXEC
6306 and the C<S_IF*> functions are
6308 S_IMODE($mode) the part of $mode containing the permission bits
6309 and the setuid/setgid/sticky bits
6311 S_IFMT($mode) the part of $mode containing the file type
6312 which can be bit-anded with (for example) S_IFREG
6313 or with the following functions
6315 # The operators -f, -d, -l, -b, -c, -p, and -S.
6317 S_ISREG($mode) S_ISDIR($mode) S_ISLNK($mode)
6318 S_ISBLK($mode) S_ISCHR($mode) S_ISFIFO($mode) S_ISSOCK($mode)
6320 # No direct -X operator counterpart, but for the first one
6321 # the -g operator is often equivalent. The ENFMT stands for
6322 # record flocking enforcement, a platform-dependent feature.
6324 S_ISENFMT($mode) S_ISWHT($mode)
6326 See your native chmod(2) and stat(2) documentation for more details
6327 about the C<S_*> constants. To get status info for a symbolic link
6328 instead of the target file behind the link, use the C<lstat> function.
6333 =item state TYPE EXPR
6335 =item state EXPR : ATTRS
6337 =item state TYPE EXPR : ATTRS
6339 C<state> declares a lexically scoped variable, just like C<my> does.
6340 However, those variables will never be reinitialized, contrary to
6341 lexical variables that are reinitialized each time their enclosing block
6344 C<state> variables are enabled only when the C<use feature "state"> pragma
6345 is in effect. See L<feature>.
6352 Takes extra time to study SCALAR (C<$_> if unspecified) in anticipation of
6353 doing many pattern matches on the string before it is next modified.
6354 This may or may not save time, depending on the nature and number of
6355 patterns you are searching on, and on the distribution of character
6356 frequencies in the string to be searched; you probably want to compare
6357 run times with and without it to see which runs faster. Those loops
6358 that scan for many short constant strings (including the constant
6359 parts of more complex patterns) will benefit most. You may have only
6360 one C<study> active at a time: if you study a different scalar the first
6361 is "unstudied". (The way C<study> works is this: a linked list of every
6362 character in the string to be searched is made, so we know, for
6363 example, where all the C<'k'> characters are. From each search string,
6364 the rarest character is selected, based on some static frequency tables
6365 constructed from some C programs and English text. Only those places
6366 that contain this "rarest" character are examined.)
6368 For example, here is a loop that inserts index producing entries
6369 before any line containing a certain pattern:
6373 print ".IX foo\n" if /\bfoo\b/;
6374 print ".IX bar\n" if /\bbar\b/;
6375 print ".IX blurfl\n" if /\bblurfl\b/;
6380 In searching for C</\bfoo\b/>, only locations in C<$_> that contain C<f>
6381 will be looked at, because C<f> is rarer than C<o>. In general, this is
6382 a big win except in pathological cases. The only question is whether
6383 it saves you more time than it took to build the linked list in the
6386 Note that if you have to look for strings that you don't know till
6387 runtime, you can build an entire loop as a string and C<eval> that to
6388 avoid recompiling all your patterns all the time. Together with
6389 undefining C<$/> to input entire files as one record, this can be quite
6390 fast, often faster than specialized programs like fgrep(1). The following
6391 scans a list of files (C<@files>) for a list of words (C<@words>), and prints
6392 out the names of those files that contain a match:
6394 $search = 'while (<>) { study;';
6395 foreach $word (@words) {
6396 $search .= "++\$seen{\$ARGV} if /\\b$word\\b/;\n";
6401 eval $search; # this screams
6402 $/ = "\n"; # put back to normal input delimiter
6403 foreach $file (sort keys(%seen)) {
6407 =item sub NAME BLOCK
6410 =item sub NAME (PROTO) BLOCK
6412 =item sub NAME : ATTRS BLOCK
6414 =item sub NAME (PROTO) : ATTRS BLOCK
6416 This is subroutine definition, not a real function I<per se>.
6417 Without a BLOCK it's just a forward declaration. Without a NAME,
6418 it's an anonymous function declaration, and does actually return
6419 a value: the CODE ref of the closure you just created.
6421 See L<perlsub> and L<perlref> for details about subroutines and
6422 references, and L<attributes> and L<Attribute::Handlers> for more
6423 information about attributes.
6425 =item substr EXPR,OFFSET,LENGTH,REPLACEMENT
6426 X<substr> X<substring> X<mid> X<left> X<right>
6428 =item substr EXPR,OFFSET,LENGTH
6430 =item substr EXPR,OFFSET
6432 Extracts a substring out of EXPR and returns it. First character is at
6433 offset C<0>, or whatever you've set C<$[> to (but don't do that).
6434 If OFFSET is negative (or more precisely, less than C<$[>), starts
6435 that far from the end of the string. If LENGTH is omitted, returns
6436 everything to the end of the string. If LENGTH is negative, leaves that
6437 many characters off the end of the string.
6439 my $s = "The black cat climbed the green tree";
6440 my $color = substr $s, 4, 5; # black
6441 my $middle = substr $s, 4, -11; # black cat climbed the
6442 my $end = substr $s, 14; # climbed the green tree
6443 my $tail = substr $s, -4; # tree
6444 my $z = substr $s, -4, 2; # tr
6446 You can use the substr() function as an lvalue, in which case EXPR
6447 must itself be an lvalue. If you assign something shorter than LENGTH,
6448 the string will shrink, and if you assign something longer than LENGTH,
6449 the string will grow to accommodate it. To keep the string the same
6450 length, you may need to pad or chop your value using C<sprintf>.
6452 If OFFSET and LENGTH specify a substring that is partly outside the
6453 string, only the part within the string is returned. If the substring
6454 is beyond either end of the string, substr() returns the undefined
6455 value and produces a warning. When used as an lvalue, specifying a
6456 substring that is entirely outside the string raises an exception.
6457 Here's an example showing the behavior for boundary cases:
6460 substr($name, 4) = 'dy'; # $name is now 'freddy'
6461 my $null = substr $name, 6, 2; # returns "" (no warning)
6462 my $oops = substr $name, 7; # returns undef, with warning
6463 substr($name, 7) = 'gap'; # raises an exception
6465 An alternative to using substr() as an lvalue is to specify the
6466 replacement string as the 4th argument. This allows you to replace
6467 parts of the EXPR and return what was there before in one operation,
6468 just as you can with splice().
6470 my $s = "The black cat climbed the green tree";
6471 my $z = substr $s, 14, 7, "jumped from"; # climbed
6472 # $s is now "The black cat jumped from the green tree"
6474 Note that the lvalue returned by the 3-arg version of substr() acts as
6475 a 'magic bullet'; each time it is assigned to, it remembers which part
6476 of the original string is being modified; for example:
6479 for (substr($x,1,2)) {
6480 $_ = 'a'; print $x,"\n"; # prints 1a4
6481 $_ = 'xyz'; print $x,"\n"; # prints 1xyz4
6483 $_ = 'pq'; print $x,"\n"; # prints 5pq9
6486 Prior to Perl version 5.9.1, the result of using an lvalue multiple times was
6489 =item symlink OLDFILE,NEWFILE
6490 X<symlink> X<link> X<symbolic link> X<link, symbolic>
6492 Creates a new filename symbolically linked to the old filename.
6493 Returns C<1> for success, C<0> otherwise. On systems that don't support
6494 symbolic links, raises an exception. To check for that,
6497 $symlink_exists = eval { symlink("",""); 1 };
6499 =item syscall NUMBER, LIST
6500 X<syscall> X<system call>
6502 Calls the system call specified as the first element of the list,
6503 passing the remaining elements as arguments to the system call. If
6504 unimplemented, raises an exception. The arguments are interpreted
6505 as follows: if a given argument is numeric, the argument is passed as
6506 an int. If not, the pointer to the string value is passed. You are
6507 responsible to make sure a string is pre-extended long enough to
6508 receive any result that might be written into a string. You can't use a
6509 string literal (or other read-only string) as an argument to C<syscall>
6510 because Perl has to assume that any string pointer might be written
6512 integer arguments are not literals and have never been interpreted in a
6513 numeric context, you may need to add C<0> to them to force them to look
6514 like numbers. This emulates the C<syswrite> function (or vice versa):
6516 require 'syscall.ph'; # may need to run h2ph
6518 syscall(&SYS_write, fileno(STDOUT), $s, length $s);
6520 Note that Perl supports passing of up to only 14 arguments to your syscall,
6521 which in practice should (usually) suffice.
6523 Syscall returns whatever value returned by the system call it calls.
6524 If the system call fails, C<syscall> returns C<-1> and sets C<$!> (errno).
6525 Note that some system calls can legitimately return C<-1>. The proper
6526 way to handle such calls is to assign C<$!=0;> before the call and
6527 check the value of C<$!> if syscall returns C<-1>.
6529 There's a problem with C<syscall(&SYS_pipe)>: it returns the file
6530 number of the read end of the pipe it creates. There is no way
6531 to retrieve the file number of the other end. You can avoid this
6532 problem by using C<pipe> instead.
6534 =item sysopen FILEHANDLE,FILENAME,MODE
6537 =item sysopen FILEHANDLE,FILENAME,MODE,PERMS
6539 Opens the file whose filename is given by FILENAME, and associates it
6540 with FILEHANDLE. If FILEHANDLE is an expression, its value is used as
6541 the name of the real filehandle wanted. This function calls the
6542 underlying operating system's C<open> function with the parameters
6543 FILENAME, MODE, PERMS.
6545 The possible values and flag bits of the MODE parameter are
6546 system-dependent; they are available via the standard module C<Fcntl>.
6547 See the documentation of your operating system's C<open> to see which
6548 values and flag bits are available. You may combine several flags
6549 using the C<|>-operator.
6551 Some of the most common values are C<O_RDONLY> for opening the file in
6552 read-only mode, C<O_WRONLY> for opening the file in write-only mode,
6553 and C<O_RDWR> for opening the file in read-write mode.
6554 X<O_RDONLY> X<O_RDWR> X<O_WRONLY>
6556 For historical reasons, some values work on almost every system
6557 supported by Perl: 0 means read-only, 1 means write-only, and 2
6558 means read/write. We know that these values do I<not> work under
6559 OS/390 & VM/ESA Unix and on the Macintosh; you probably don't want to
6560 use them in new code.
6562 If the file named by FILENAME does not exist and the C<open> call creates
6563 it (typically because MODE includes the C<O_CREAT> flag), then the value of
6564 PERMS specifies the permissions of the newly created file. If you omit
6565 the PERMS argument to C<sysopen>, Perl uses the octal value C<0666>.
6566 These permission values need to be in octal, and are modified by your
6567 process's current C<umask>.
6570 In many systems the C<O_EXCL> flag is available for opening files in
6571 exclusive mode. This is B<not> locking: exclusiveness means here that
6572 if the file already exists, sysopen() fails. C<O_EXCL> may not work
6573 on network filesystems, and has no effect unless the C<O_CREAT> flag
6574 is set as well. Setting C<O_CREAT|O_EXCL> prevents the file from
6575 being opened if it is a symbolic link. It does not protect against
6576 symbolic links in the file's path.
6579 Sometimes you may want to truncate an already-existing file. This
6580 can be done using the C<O_TRUNC> flag. The behavior of
6581 C<O_TRUNC> with C<O_RDONLY> is undefined.
6584 You should seldom if ever use C<0644> as argument to C<sysopen>, because
6585 that takes away the user's option to have a more permissive umask.
6586 Better to omit it. See the perlfunc(1) entry on C<umask> for more
6589 Note that C<sysopen> depends on the fdopen() C library function.
6590 On many Unix systems, fdopen() is known to fail when file descriptors
6591 exceed a certain value, typically 255. If you need more file
6592 descriptors than that, consider rebuilding Perl to use the C<sfio>
6593 library, or perhaps using the POSIX::open() function.
6595 See L<perlopentut> for a kinder, gentler explanation of opening files.
6597 =item sysread FILEHANDLE,SCALAR,LENGTH,OFFSET
6600 =item sysread FILEHANDLE,SCALAR,LENGTH
6602 Attempts to read LENGTH bytes of data into variable SCALAR from the
6603 specified FILEHANDLE, using the read(2). It bypasses
6604 buffered IO, so mixing this with other kinds of reads, C<print>,
6605 C<write>, C<seek>, C<tell>, or C<eof> can cause confusion because the
6606 perlio or stdio layers usually buffers data. Returns the number of
6607 bytes actually read, C<0> at end of file, or undef if there was an
6608 error (in the latter case C<$!> is also set). SCALAR will be grown or
6609 shrunk so that the last byte actually read is the last byte of the
6610 scalar after the read.
6612 An OFFSET may be specified to place the read data at some place in the
6613 string other than the beginning. A negative OFFSET specifies
6614 placement at that many characters counting backwards from the end of
6615 the string. A positive OFFSET greater than the length of SCALAR
6616 results in the string being padded to the required size with C<"\0">
6617 bytes before the result of the read is appended.
6619 There is no syseof() function, which is ok, since eof() doesn't work
6620 well on device files (like ttys) anyway. Use sysread() and check
6621 for a return value for 0 to decide whether you're done.
6623 Note that if the filehandle has been marked as C<:utf8> Unicode
6624 characters are read instead of bytes (the LENGTH, OFFSET, and the
6625 return value of sysread() are in Unicode characters).
6626 The C<:encoding(...)> layer implicitly introduces the C<:utf8> layer.
6627 See L</binmode>, L</open>, and the C<open> pragma, L<open>.
6629 =item sysseek FILEHANDLE,POSITION,WHENCE
6632 Sets FILEHANDLE's system position in bytes using
6633 lseek(2). FILEHANDLE may be an expression whose value gives the name
6634 of the filehandle. The values for WHENCE are C<0> to set the new
6635 position to POSITION, C<1> to set the it to the current position plus
6636 POSITION, and C<2> to set it to EOF plus POSITION (typically
6639 Note the I<in bytes>: even if the filehandle has been set to operate
6640 on characters (for example by using the C<:encoding(utf8)> I/O layer),
6641 tell() will return byte offsets, not character offsets (because
6642 implementing that would render sysseek() unacceptably slow).
6644 sysseek() bypasses normal buffered IO, so mixing this with reads (other
6645 than C<sysread>, for example C<< <> >> or read()) C<print>, C<write>,
6646 C<seek>, C<tell>, or C<eof> may cause confusion.
6648 For WHENCE, you may also use the constants C<SEEK_SET>, C<SEEK_CUR>,
6649 and C<SEEK_END> (start of the file, current position, end of the file)
6650 from the Fcntl module. Use of the constants is also more portable
6651 than relying on 0, 1, and 2. For example to define a "systell" function:
6653 use Fcntl 'SEEK_CUR';
6654 sub systell { sysseek($_[0], 0, SEEK_CUR) }
6656 Returns the new position, or the undefined value on failure. A position
6657 of zero is returned as the string C<"0 but true">; thus C<sysseek> returns
6658 true on success and false on failure, yet you can still easily determine
6664 =item system PROGRAM LIST
6666 Does exactly the same thing as C<exec LIST>, except that a fork is
6667 done first, and the parent process waits for the child process to
6668 exit. Note that argument processing varies depending on the
6669 number of arguments. If there is more than one argument in LIST,
6670 or if LIST is an array with more than one value, starts the program
6671 given by the first element of the list with arguments given by the
6672 rest of the list. If there is only one scalar argument, the argument
6673 is checked for shell metacharacters, and if there are any, the
6674 entire argument is passed to the system's command shell for parsing
6675 (this is C</bin/sh -c> on Unix platforms, but varies on other
6676 platforms). If there are no shell metacharacters in the argument,
6677 it is split into words and passed directly to C<execvp>, which is
6680 Beginning with v5.6.0, Perl will attempt to flush all files opened for
6681 output before any operation that may do a fork, but this may not be
6682 supported on some platforms (see L<perlport>). To be safe, you may need
6683 to set C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method
6684 of C<IO::Handle> on any open handles.
6686 The return value is the exit status of the program as returned by the
6687 C<wait> call. To get the actual exit value, shift right by eight (see
6688 below). See also L</exec>. This is I<not> what you want to use to capture
6689 the output from a command, for that you should use merely backticks or
6690 C<qx//>, as described in L<perlop/"`STRING`">. Return value of -1
6691 indicates a failure to start the program or an error of the wait(2) system
6692 call (inspect $! for the reason).
6694 If you'd like to make C<system> (and many other bits of Perl) die on error,
6695 have a look at the L<autodie> pragma.
6697 Like C<exec>, C<system> allows you to lie to a program about its name if
6698 you use the C<system PROGRAM LIST> syntax. Again, see L</exec>.
6700 Since C<SIGINT> and C<SIGQUIT> are ignored during the execution of
6701 C<system>, if you expect your program to terminate on receipt of these
6702 signals you will need to arrange to do so yourself based on the return
6705 @args = ("command", "arg1", "arg2");
6707 or die "system @args failed: $?"
6709 If you'd like to manually inspect C<system>'s failure, you can check all
6710 possible failure modes by inspecting C<$?> like this:
6713 print "failed to execute: $!\n";
6716 printf "child died with signal %d, %s coredump\n",
6717 ($? & 127), ($? & 128) ? 'with' : 'without';
6720 printf "child exited with value %d\n", $? >> 8;
6723 Alternatively, you may inspect the value of C<${^CHILD_ERROR_NATIVE}>
6724 with the C<W*()> calls from the POSIX module.
6726 When C<system>'s arguments are executed indirectly by the shell,
6727 results and return codes are subject to its quirks.
6728 See L<perlop/"`STRING`"> and L</exec> for details.
6730 =item syswrite FILEHANDLE,SCALAR,LENGTH,OFFSET
6733 =item syswrite FILEHANDLE,SCALAR,LENGTH
6735 =item syswrite FILEHANDLE,SCALAR
6737 Attempts to write LENGTH bytes of data from variable SCALAR to the
6738 specified FILEHANDLE, using write(2). If LENGTH is
6739 not specified, writes whole SCALAR. It bypasses buffered IO, so
6740 mixing this with reads (other than C<sysread())>, C<print>, C<write>,
6741 C<seek>, C<tell>, or C<eof> may cause confusion because the perlio and
6742 stdio layers usually buffers data. Returns the number of bytes
6743 actually written, or C<undef> if there was an error (in this case the
6744 errno variable C<$!> is also set). If the LENGTH is greater than the
6745 data available in the SCALAR after the OFFSET, only as much data as is
6746 available will be written.
6748 An OFFSET may be specified to write the data from some part of the
6749 string other than the beginning. A negative OFFSET specifies writing
6750 that many characters counting backwards from the end of the string.
6751 If SCALAR is of length zero, you can only use an OFFSET of 0.
6753 B<Warning>: If the filehandle is marked C<:utf8>, Unicode characters
6754 encoded in UTF-8 are written instead of bytes, and the LENGTH, OFFSET, and
6755 return value of syswrite() are in (UTF-8 encoded Unicode) characters.
6756 The C<:encoding(...)> layer implicitly introduces the C<:utf8> layer.
6757 See L</binmode>, L</open>, and the C<open> pragma, L<open>.
6759 =item tell FILEHANDLE
6764 Returns the current position I<in bytes> for FILEHANDLE, or -1 on
6765 error. FILEHANDLE may be an expression whose value gives the name of
6766 the actual filehandle. If FILEHANDLE is omitted, assumes the file
6769 Note the I<in bytes>: even if the filehandle has been set to
6770 operate on characters (for example by using the C<:encoding(utf8)> open
6771 layer), tell() will return byte offsets, not character offsets (because
6772 that would render seek() and tell() rather slow).
6774 The return value of tell() for the standard streams like the STDIN
6775 depends on the operating system: it may return -1 or something else.
6776 tell() on pipes, fifos, and sockets usually returns -1.
6778 There is no C<systell> function. Use C<sysseek(FH, 0, 1)> for that.
6780 Do not use tell() (or other buffered I/O operations) on a filehandle
6781 that has been manipulated by sysread(), syswrite() or sysseek().
6782 Those functions ignore the buffering, while tell() does not.
6784 =item telldir DIRHANDLE
6787 Returns the current position of the C<readdir> routines on DIRHANDLE.
6788 Value may be given to C<seekdir> to access a particular location in a
6789 directory. C<telldir> has the same caveats about possible directory
6790 compaction as the corresponding system library routine.
6792 =item tie VARIABLE,CLASSNAME,LIST
6795 This function binds a variable to a package class that will provide the
6796 implementation for the variable. VARIABLE is the name of the variable
6797 to be enchanted. CLASSNAME is the name of a class implementing objects
6798 of correct type. Any additional arguments are passed to the C<new>
6799 method of the class (meaning C<TIESCALAR>, C<TIEHANDLE>, C<TIEARRAY>,
6800 or C<TIEHASH>). Typically these are arguments such as might be passed
6801 to the C<dbm_open()> function of C. The object returned by the C<new>
6802 method is also returned by the C<tie> function, which would be useful
6803 if you want to access other methods in CLASSNAME.
6805 Note that functions such as C<keys> and C<values> may return huge lists
6806 when used on large objects, like DBM files. You may prefer to use the
6807 C<each> function to iterate over such. Example:
6809 # print out history file offsets
6811 tie(%HIST, 'NDBM_File', '/usr/lib/news/history', 1, 0);
6812 while (($key,$val) = each %HIST) {
6813 print $key, ' = ', unpack('L',$val), "\n";
6817 A class implementing a hash should have the following methods:
6819 TIEHASH classname, LIST
6821 STORE this, key, value
6826 NEXTKEY this, lastkey
6831 A class implementing an ordinary array should have the following methods:
6833 TIEARRAY classname, LIST
6835 STORE this, key, value
6837 STORESIZE this, count
6843 SPLICE this, offset, length, LIST
6848 A class implementing a filehandle should have the following methods:
6850 TIEHANDLE classname, LIST
6851 READ this, scalar, length, offset
6854 WRITE this, scalar, length, offset
6856 PRINTF this, format, LIST
6860 SEEK this, position, whence
6862 OPEN this, mode, LIST
6867 A class implementing a scalar should have the following methods:
6869 TIESCALAR classname, LIST
6875 Not all methods indicated above need be implemented. See L<perltie>,
6876 L<Tie::Hash>, L<Tie::Array>, L<Tie::Scalar>, and L<Tie::Handle>.
6878 Unlike C<dbmopen>, the C<tie> function will not C<use> or C<require> a module
6879 for you; you need to do that explicitly yourself. See L<DB_File>
6880 or the F<Config> module for interesting C<tie> implementations.
6882 For further details see L<perltie>, L<"tied VARIABLE">.
6887 Returns a reference to the object underlying VARIABLE (the same value
6888 that was originally returned by the C<tie> call that bound the variable
6889 to a package.) Returns the undefined value if VARIABLE isn't tied to a
6895 Returns the number of non-leap seconds since whatever time the system
6896 considers to be the epoch, suitable for feeding to C<gmtime> and
6897 C<localtime>. On most systems the epoch is 00:00:00 UTC, January 1, 1970;
6898 a prominent exception being Mac OS Classic which uses 00:00:00, January 1,
6899 1904 in the current local time zone for its epoch.
6901 For measuring time in better granularity than one second,
6902 you may use either the L<Time::HiRes> module (from CPAN, and starting from
6903 Perl 5.8 part of the standard distribution), or if you have
6904 gettimeofday(2), you may be able to use the C<syscall> interface of Perl.
6905 See L<perlfaq8> for details.
6907 For date and time processing look at the many related modules on CPAN.
6908 For a comprehensive date and time representation look at the
6914 Returns a four-element list giving the user and system times, in
6915 seconds, for this process and the children of this process.
6917 ($user,$system,$cuser,$csystem) = times;
6919 In scalar context, C<times> returns C<$user>.
6921 Children's times are only included for terminated children.
6925 The transliteration operator. Same as C<y///>. See
6926 L<perlop/"Quote and Quote-like Operators">.
6928 =item truncate FILEHANDLE,LENGTH
6931 =item truncate EXPR,LENGTH
6933 Truncates the file opened on FILEHANDLE, or named by EXPR, to the
6934 specified length. Raises an exception if truncate isn't implemented
6935 on your system. Returns true if successful, the undefined value
6938 The behavior is undefined if LENGTH is greater than the length of the
6941 The position in the file of FILEHANDLE is left unchanged. You may want to
6942 call L<seek> before writing to the file.
6945 X<uc> X<uppercase> X<toupper>
6949 Returns an uppercased version of EXPR. This is the internal function
6950 implementing the C<\U> escape in double-quoted strings.
6951 It does not attempt to do titlecase mapping on initial letters. See
6952 L</ucfirst> for that.
6954 If EXPR is omitted, uses C<$_>.
6956 This function behaves the same way under various pragma, such as in a locale,
6960 X<ucfirst> X<uppercase>
6964 Returns the value of EXPR with the first character in uppercase
6965 (titlecase in Unicode). This is the internal function implementing
6966 the C<\u> escape in double-quoted strings.
6968 If EXPR is omitted, uses C<$_>.
6970 This function behaves the same way under various pragma, such as in a locale,
6978 Sets the umask for the process to EXPR and returns the previous value.
6979 If EXPR is omitted, merely returns the current umask.
6981 The Unix permission C<rwxr-x---> is represented as three sets of three
6982 bits, or three octal digits: C<0750> (the leading 0 indicates octal
6983 and isn't one of the digits). The C<umask> value is such a number
6984 representing disabled permissions bits. The permission (or "mode")
6985 values you pass C<mkdir> or C<sysopen> are modified by your umask, so
6986 even if you tell C<sysopen> to create a file with permissions C<0777>,
6987 if your umask is C<0022> then the file will actually be created with
6988 permissions C<0755>. If your C<umask> were C<0027> (group can't
6989 write; others can't read, write, or execute), then passing
6990 C<sysopen> C<0666> would create a file with mode C<0640> (C<0666 &~
6993 Here's some advice: supply a creation mode of C<0666> for regular
6994 files (in C<sysopen>) and one of C<0777> for directories (in
6995 C<mkdir>) and executable files. This gives users the freedom of
6996 choice: if they want protected files, they might choose process umasks
6997 of C<022>, C<027>, or even the particularly antisocial mask of C<077>.
6998 Programs should rarely if ever make policy decisions better left to
6999 the user. The exception to this is when writing files that should be
7000 kept private: mail files, web browser cookies, I<.rhosts> files, and
7003 If umask(2) is not implemented on your system and you are trying to
7004 restrict access for I<yourself> (i.e., C<< (EXPR & 0700) > 0 >>),
7005 raises an exception. If umask(2) is not implemented and you are
7006 not trying to restrict access for yourself, returns C<undef>.
7008 Remember that a umask is a number, usually given in octal; it is I<not> a
7009 string of octal digits. See also L</oct>, if all you have is a string.
7012 X<undef> X<undefine>
7016 Undefines the value of EXPR, which must be an lvalue. Use only on a
7017 scalar value, an array (using C<@>), a hash (using C<%>), a subroutine
7018 (using C<&>), or a typeglob (using C<*>). Saying C<undef $hash{$key}>
7019 will probably not do what you expect on most predefined variables or
7020 DBM list values, so don't do that; see L<delete>. Always returns the
7021 undefined value. You can omit the EXPR, in which case nothing is
7022 undefined, but you still get an undefined value that you could, for
7023 instance, return from a subroutine, assign to a variable, or pass as a
7024 parameter. Examples:
7027 undef $bar{'blurfl'}; # Compare to: delete $bar{'blurfl'};
7031 undef *xyz; # destroys $xyz, @xyz, %xyz, &xyz, etc.
7032 return (wantarray ? (undef, $errmsg) : undef) if $they_blew_it;
7033 select undef, undef, undef, 0.25;
7034 ($a, $b, undef, $c) = &foo; # Ignore third value returned
7036 Note that this is a unary operator, not a list operator.
7039 X<unlink> X<delete> X<remove> X<rm> X<del>
7043 Deletes a list of files. On success, it returns the number of files
7044 it successfully deleted. On failure, it returns false and sets C<$!>
7047 my $unlinked = unlink 'a', 'b', 'c';
7049 unlink glob "*.bak";
7051 On error, C<unlink> will not tell you which files it could not remove.
7052 If you want to know which files you could not remove, try them one
7055 foreach my $file ( @goners ) {
7056 unlink $file or warn "Could not unlink $file: $!";
7059 Note: C<unlink> will not attempt to delete directories unless you are
7060 superuser and the B<-U> flag is supplied to Perl. Even if these
7061 conditions are met, be warned that unlinking a directory can inflict
7062 damage on your filesystem. Finally, using C<unlink> on directories is
7063 not supported on many operating systems. Use C<rmdir> instead.
7065 If LIST is omitted, C<unlink> uses C<$_>.
7067 =item unpack TEMPLATE,EXPR
7070 =item unpack TEMPLATE
7072 C<unpack> does the reverse of C<pack>: it takes a string
7073 and expands it out into a list of values.
7074 (In scalar context, it returns merely the first value produced.)
7076 If EXPR is omitted, unpacks the C<$_> string.
7077 See L<perlpacktut> for an introduction to this function.
7079 The string is broken into chunks described by the TEMPLATE. Each chunk
7080 is converted separately to a value. Typically, either the string is a result
7081 of C<pack>, or the characters of the string represent a C structure of some
7084 The TEMPLATE has the same format as in the C<pack> function.
7085 Here's a subroutine that does substring:
7088 my($what,$where,$howmuch) = @_;
7089 unpack("x$where a$howmuch", $what);
7094 sub ordinal { unpack("W",$_[0]); } # same as ord()
7096 In addition to fields allowed in pack(), you may prefix a field with
7097 a %<number> to indicate that
7098 you want a <number>-bit checksum of the items instead of the items
7099 themselves. Default is a 16-bit checksum. Checksum is calculated by
7100 summing numeric values of expanded values (for string fields the sum of
7101 C<ord($char)> is taken, for bit fields the sum of zeroes and ones).
7103 For example, the following
7104 computes the same number as the System V sum program:
7108 unpack("%32W*",<>) % 65535;
7111 The following efficiently counts the number of set bits in a bit vector:
7113 $setbits = unpack("%32b*", $selectmask);
7115 The C<p> and C<P> formats should be used with care. Since Perl
7116 has no way of checking whether the value passed to C<unpack()>
7117 corresponds to a valid memory location, passing a pointer value that's
7118 not known to be valid is likely to have disastrous consequences.
7120 If there are more pack codes or if the repeat count of a field or a group
7121 is larger than what the remainder of the input string allows, the result
7122 is not well defined: the repeat count may be decreased, or
7123 C<unpack()> may produce empty strings or zeros, or it may raise an exception.
7124 If the input string is longer than one described by the TEMPLATE,
7125 the remainder of that input string is ignored.
7127 See L</pack> for more examples and notes.
7129 =item untie VARIABLE
7132 Breaks the binding between a variable and a package. (See C<tie>.)
7133 Has no effect if the variable is not tied.
7135 =item unshift ARRAY,LIST
7138 Does the opposite of a C<shift>. Or the opposite of a C<push>,
7139 depending on how you look at it. Prepends list to the front of the
7140 array, and returns the new number of elements in the array.
7142 unshift(@ARGV, '-e') unless $ARGV[0] =~ /^-/;
7144 Note the LIST is prepended whole, not one element at a time, so the
7145 prepended elements stay in the same order. Use C<reverse> to do the
7148 =item use Module VERSION LIST
7149 X<use> X<module> X<import>
7151 =item use Module VERSION
7153 =item use Module LIST
7159 Imports some semantics into the current package from the named module,
7160 generally by aliasing certain subroutine or variable names into your
7161 package. It is exactly equivalent to
7163 BEGIN { require Module; Module->import( LIST ); }
7165 except that Module I<must> be a bareword.
7167 In the peculiar C<use VERSION> form, VERSION may be either a positive
7168 decimal fraction such as 5.006, which will be compared to C<$]>, or a v-string
7169 of the form v5.6.1, which will be compared to C<$^V> (aka $PERL_VERSION). An
7170 exception is raised if VERSION is greater than the version of the
7171 current Perl interpreter; Perl will not attempt to parse the rest of the
7172 file. Compare with L</require>, which can do a similar check at run time.
7173 Symmetrically, C<no VERSION> allows you to specify that you want a version
7174 of Perl older than the specified one.
7176 Specifying VERSION as a literal of the form v5.6.1 should generally be
7177 avoided, because it leads to misleading error messages under earlier
7178 versions of Perl (that is, prior to 5.6.0) that do not support this
7179 syntax. The equivalent numeric version should be used instead.
7181 use v5.6.1; # compile time version check
7183 use 5.006_001; # ditto; preferred for backwards compatibility
7185 This is often useful if you need to check the current Perl version before
7186 C<use>ing library modules that won't work with older versions of Perl.
7187 (We try not to do this more than we have to.)
7189 Also, if the specified Perl version is greater than or equal to 5.9.5,
7190 C<use VERSION> will also load the C<feature> pragma and enable all
7191 features available in the requested version. See L<feature>.
7192 Similarly, if the specified Perl version is greater than or equal to
7193 5.11.0, strictures are enabled lexically as with C<use strict> (except
7194 that the F<strict.pm> file is not actually loaded).
7196 The C<BEGIN> forces the C<require> and C<import> to happen at compile time. The
7197 C<require> makes sure the module is loaded into memory if it hasn't been
7198 yet. The C<import> is not a builtin; it's just an ordinary static method
7199 call into the C<Module> package to tell the module to import the list of
7200 features back into the current package. The module can implement its
7201 C<import> method any way it likes, though most modules just choose to
7202 derive their C<import> method via inheritance from the C<Exporter> class that
7203 is defined in the C<Exporter> module. See L<Exporter>. If no C<import>
7204 method can be found then the call is skipped, even if there is an AUTOLOAD
7207 If you do not want to call the package's C<import> method (for instance,
7208 to stop your namespace from being altered), explicitly supply the empty list:
7212 That is exactly equivalent to
7214 BEGIN { require Module }
7216 If the VERSION argument is present between Module and LIST, then the
7217 C<use> will call the VERSION method in class Module with the given
7218 version as an argument. The default VERSION method, inherited from
7219 the UNIVERSAL class, croaks if the given version is larger than the
7220 value of the variable C<$Module::VERSION>.
7222 Again, there is a distinction between omitting LIST (C<import> called
7223 with no arguments) and an explicit empty LIST C<()> (C<import> not
7224 called). Note that there is no comma after VERSION!
7226 Because this is a wide-open interface, pragmas (compiler directives)
7227 are also implemented this way. Currently implemented pragmas are:
7232 use sigtrap qw(SEGV BUS);
7233 use strict qw(subs vars refs);
7234 use subs qw(afunc blurfl);
7235 use warnings qw(all);
7236 use sort qw(stable _quicksort _mergesort);
7238 Some of these pseudo-modules import semantics into the current
7239 block scope (like C<strict> or C<integer>, unlike ordinary modules,
7240 which import symbols into the current package (which are effective
7241 through the end of the file).
7243 Because C<use> takes effect at compile time, it doesn't respect the
7244 ordinary flow control of the code being compiled. In particular, putting
7245 a C<use> inside the false branch of a conditional doesn't prevent it
7246 from being processed. If a module or pragma only needs to be loaded
7247 conditionally, this can be done using the L<if> pragma:
7249 use if $] < 5.008, "utf8";
7250 use if WANT_WARNINGS, warnings => qw(all);
7252 There's a corresponding C<no> command that unimports meanings imported
7253 by C<use>, i.e., it calls C<unimport Module LIST> instead of C<import>.
7254 It behaves just as C<import> does with VERSION, an omitted or empty LIST,
7255 or no unimport method being found.
7261 See L<perlmodlib> for a list of standard modules and pragmas. See L<perlrun>
7262 for the C<-M> and C<-m> command-line options to Perl that give C<use>
7263 functionality from the command-line.
7268 Changes the access and modification times on each file of a list of
7269 files. The first two elements of the list must be the NUMERICAL access
7270 and modification times, in that order. Returns the number of files
7271 successfully changed. The inode change time of each file is set
7272 to the current time. For example, this code has the same effect as the
7273 Unix touch(1) command when the files I<already exist> and belong to
7274 the user running the program:
7277 $atime = $mtime = time;
7278 utime $atime, $mtime, @ARGV;
7280 Since Perl 5.7.2, if the first two elements of the list are C<undef>,
7281 the utime(2) syscall from your C library is called with a null second
7282 argument. On most systems, this will set the file's access and
7283 modification times to the current time (i.e., equivalent to the example
7284 above) and will work even on files you don't own provided you have write
7288 utime(undef, undef, $file)
7289 || warn "couldn't touch $file: $!";
7292 Under NFS this will use the time of the NFS server, not the time of
7293 the local machine. If there is a time synchronization problem, the
7294 NFS server and local machine will have different times. The Unix
7295 touch(1) command will in fact normally use this form instead of the
7296 one shown in the first example.
7298 Passing only one of the first two elements as C<undef> is
7299 equivalent to passing a 0 and will not have the effect
7300 described when both are C<undef>. This also triggers an
7301 uninitialized warning.
7303 On systems that support futimes(2), you may pass filehandles among the
7304 files. On systems that don't support futimes(2), passing filehandles raises
7305 an exception. Filehandles must be passed as globs or glob references to be
7306 recognized; barewords are considered filenames.
7313 Returns a list consisting of all the values of the named hash, or the values
7314 of an array. (In a scalar context, returns the number of values.)
7316 The values are returned in an apparently random order. The actual
7317 random order is subject to change in future versions of Perl, but it
7318 is guaranteed to be the same order as either the C<keys> or C<each>
7319 function would produce on the same (unmodified) hash. Since Perl
7320 5.8.1 the ordering is different even between different runs of Perl
7321 for security reasons (see L<perlsec/"Algorithmic Complexity Attacks">).
7323 As a side effect, calling values() resets the HASH or ARRAY's internal
7325 see L</each>. (In particular, calling values() in void context resets
7326 the iterator with no other overhead. Apart from resetting the iterator,
7327 C<values @array> in list context is the same as plain C<@array>.
7328 We recommend that you use void context C<keys @array> for this, but reasoned
7329 that it taking C<values @array> out would require more documentation than
7333 Note that the values are not copied, which means modifying them will
7334 modify the contents of the hash:
7336 for (values %hash) { s/foo/bar/g } # modifies %hash values
7337 for (@hash{keys %hash}) { s/foo/bar/g } # same
7339 See also C<keys>, C<each>, and C<sort>.
7341 =item vec EXPR,OFFSET,BITS
7342 X<vec> X<bit> X<bit vector>
7344 Treats the string in EXPR as a bit vector made up of elements of
7345 width BITS, and returns the value of the element specified by OFFSET
7346 as an unsigned integer. BITS therefore specifies the number of bits
7347 that are reserved for each element in the bit vector. This must
7348 be a power of two from 1 to 32 (or 64, if your platform supports
7351 If BITS is 8, "elements" coincide with bytes of the input string.
7353 If BITS is 16 or more, bytes of the input string are grouped into chunks
7354 of size BITS/8, and each group is converted to a number as with
7355 pack()/unpack() with big-endian formats C<n>/C<N> (and analogously
7356 for BITS==64). See L<"pack"> for details.
7358 If bits is 4 or less, the string is broken into bytes, then the bits
7359 of each byte are broken into 8/BITS groups. Bits of a byte are
7360 numbered in a little-endian-ish way, as in C<0x01>, C<0x02>,
7361 C<0x04>, C<0x08>, C<0x10>, C<0x20>, C<0x40>, C<0x80>. For example,
7362 breaking the single input byte C<chr(0x36)> into two groups gives a list
7363 C<(0x6, 0x3)>; breaking it into 4 groups gives C<(0x2, 0x1, 0x3, 0x0)>.
7365 C<vec> may also be assigned to, in which case parentheses are needed
7366 to give the expression the correct precedence as in
7368 vec($image, $max_x * $x + $y, 8) = 3;
7370 If the selected element is outside the string, the value 0 is returned.
7371 If an element off the end of the string is written to, Perl will first
7372 extend the string with sufficiently many zero bytes. It is an error
7373 to try to write off the beginning of the string (i.e., negative OFFSET).
7375 If the string happens to be encoded as UTF-8 internally (and thus has
7376 the UTF8 flag set), this is ignored by C<vec>, and it operates on the
7377 internal byte string, not the conceptual character string, even if you
7378 only have characters with values less than 256.
7380 Strings created with C<vec> can also be manipulated with the logical
7381 operators C<|>, C<&>, C<^>, and C<~>. These operators will assume a bit
7382 vector operation is desired when both operands are strings.
7383 See L<perlop/"Bitwise String Operators">.
7385 The following code will build up an ASCII string saying C<'PerlPerlPerl'>.
7386 The comments show the string after each step. Note that this code works
7387 in the same way on big-endian or little-endian machines.
7390 vec($foo, 0, 32) = 0x5065726C; # 'Perl'
7392 # $foo eq "Perl" eq "\x50\x65\x72\x6C", 32 bits
7393 print vec($foo, 0, 8); # prints 80 == 0x50 == ord('P')
7395 vec($foo, 2, 16) = 0x5065; # 'PerlPe'
7396 vec($foo, 3, 16) = 0x726C; # 'PerlPerl'
7397 vec($foo, 8, 8) = 0x50; # 'PerlPerlP'
7398 vec($foo, 9, 8) = 0x65; # 'PerlPerlPe'
7399 vec($foo, 20, 4) = 2; # 'PerlPerlPe' . "\x02"
7400 vec($foo, 21, 4) = 7; # 'PerlPerlPer'
7402 vec($foo, 45, 2) = 3; # 'PerlPerlPer' . "\x0c"
7403 vec($foo, 93, 1) = 1; # 'PerlPerlPer' . "\x2c"
7404 vec($foo, 94, 1) = 1; # 'PerlPerlPerl'
7407 To transform a bit vector into a string or list of 0's and 1's, use these:
7409 $bits = unpack("b*", $vector);
7410 @bits = split(//, unpack("b*", $vector));
7412 If you know the exact length in bits, it can be used in place of the C<*>.
7414 Here is an example to illustrate how the bits actually fall in place:
7420 unpack("V",$_) 01234567890123456789012345678901
7421 ------------------------------------------------------------------
7426 for ($shift=0; $shift < $width; ++$shift) {
7427 for ($off=0; $off < 32/$width; ++$off) {
7428 $str = pack("B*", "0"x32);
7429 $bits = (1<<$shift);
7430 vec($str, $off, $width) = $bits;
7431 $res = unpack("b*",$str);
7432 $val = unpack("V", $str);
7439 vec($_,@#,@#) = @<< == @######### @>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
7440 $off, $width, $bits, $val, $res
7444 Regardless of the machine architecture on which it runs, the
7445 example above should print the following table:
7448 unpack("V",$_) 01234567890123456789012345678901
7449 ------------------------------------------------------------------
7450 vec($_, 0, 1) = 1 == 1 10000000000000000000000000000000
7451 vec($_, 1, 1) = 1 == 2 01000000000000000000000000000000
7452 vec($_, 2, 1) = 1 == 4 00100000000000000000000000000000
7453 vec($_, 3, 1) = 1 == 8 00010000000000000000000000000000
7454 vec($_, 4, 1) = 1 == 16 00001000000000000000000000000000
7455 vec($_, 5, 1) = 1 == 32 00000100000000000000000000000000
7456 vec($_, 6, 1) = 1 == 64 00000010000000000000000000000000
7457 vec($_, 7, 1) = 1 == 128 00000001000000000000000000000000
7458 vec($_, 8, 1) = 1 == 256 00000000100000000000000000000000
7459 vec($_, 9, 1) = 1 == 512 00000000010000000000000000000000
7460 vec($_,10, 1) = 1 == 1024 00000000001000000000000000000000
7461 vec($_,11, 1) = 1 == 2048 00000000000100000000000000000000
7462 vec($_,12, 1) = 1 == 4096 00000000000010000000000000000000
7463 vec($_,13, 1) = 1 == 8192 00000000000001000000000000000000
7464 vec($_,14, 1) = 1 == 16384 00000000000000100000000000000000
7465 vec($_,15, 1) = 1 == 32768 00000000000000010000000000000000
7466 vec($_,16, 1) = 1 == 65536 00000000000000001000000000000000
7467 vec($_,17, 1) = 1 == 131072 00000000000000000100000000000000
7468 vec($_,18, 1) = 1 == 262144 00000000000000000010000000000000
7469 vec($_,19, 1) = 1 == 524288 00000000000000000001000000000000
7470 vec($_,20, 1) = 1 == 1048576 00000000000000000000100000000000
7471 vec($_,21, 1) = 1 == 2097152 00000000000000000000010000000000
7472 vec($_,22, 1) = 1 == 4194304 00000000000000000000001000000000
7473 vec($_,23, 1) = 1 == 8388608 00000000000000000000000100000000
7474 vec($_,24, 1) = 1 == 16777216 00000000000000000000000010000000
7475 vec($_,25, 1) = 1 == 33554432 00000000000000000000000001000000
7476 vec($_,26, 1) = 1 == 67108864 00000000000000000000000000100000
7477 vec($_,27, 1) = 1 == 134217728 00000000000000000000000000010000
7478 vec($_,28, 1) = 1 == 268435456 00000000000000000000000000001000
7479 vec($_,29, 1) = 1 == 536870912 00000000000000000000000000000100
7480 vec($_,30, 1) = 1 == 1073741824 00000000000000000000000000000010
7481 vec($_,31, 1) = 1 == 2147483648 00000000000000000000000000000001
7482 vec($_, 0, 2) = 1 == 1 10000000000000000000000000000000
7483 vec($_, 1, 2) = 1 == 4 00100000000000000000000000000000
7484 vec($_, 2, 2) = 1 == 16 00001000000000000000000000000000
7485 vec($_, 3, 2) = 1 == 64 00000010000000000000000000000000
7486 vec($_, 4, 2) = 1 == 256 00000000100000000000000000000000
7487 vec($_, 5, 2) = 1 == 1024 00000000001000000000000000000000
7488 vec($_, 6, 2) = 1 == 4096 00000000000010000000000000000000
7489 vec($_, 7, 2) = 1 == 16384 00000000000000100000000000000000
7490 vec($_, 8, 2) = 1 == 65536 00000000000000001000000000000000
7491 vec($_, 9, 2) = 1 == 262144 00000000000000000010000000000000
7492 vec($_,10, 2) = 1 == 1048576 00000000000000000000100000000000
7493 vec($_,11, 2) = 1 == 4194304 00000000000000000000001000000000
7494 vec($_,12, 2) = 1 == 16777216 00000000000000000000000010000000
7495 vec($_,13, 2) = 1 == 67108864 00000000000000000000000000100000
7496 vec($_,14, 2) = 1 == 268435456 00000000000000000000000000001000
7497 vec($_,15, 2) = 1 == 1073741824 00000000000000000000000000000010
7498 vec($_, 0, 2) = 2 == 2 01000000000000000000000000000000
7499 vec($_, 1, 2) = 2 == 8 00010000000000000000000000000000
7500 vec($_, 2, 2) = 2 == 32 00000100000000000000000000000000
7501 vec($_, 3, 2) = 2 == 128 00000001000000000000000000000000
7502 vec($_, 4, 2) = 2 == 512 00000000010000000000000000000000
7503 vec($_, 5, 2) = 2 == 2048 00000000000100000000000000000000
7504 vec($_, 6, 2) = 2 == 8192 00000000000001000000000000000000
7505 vec($_, 7, 2) = 2 == 32768 00000000000000010000000000000000
7506 vec($_, 8, 2) = 2 == 131072 00000000000000000100000000000000
7507 vec($_, 9, 2) = 2 == 524288 00000000000000000001000000000000
7508 vec($_,10, 2) = 2 == 2097152 00000000000000000000010000000000
7509 vec($_,11, 2) = 2 == 8388608 00000000000000000000000100000000
7510 vec($_,12, 2) = 2 == 33554432 00000000000000000000000001000000
7511 vec($_,13, 2) = 2 == 134217728 00000000000000000000000000010000
7512 vec($_,14, 2) = 2 == 536870912 00000000000000000000000000000100
7513 vec($_,15, 2) = 2 == 2147483648 00000000000000000000000000000001
7514 vec($_, 0, 4) = 1 == 1 10000000000000000000000000000000
7515 vec($_, 1, 4) = 1 == 16 00001000000000000000000000000000
7516 vec($_, 2, 4) = 1 == 256 00000000100000000000000000000000
7517 vec($_, 3, 4) = 1 == 4096 00000000000010000000000000000000
7518 vec($_, 4, 4) = 1 == 65536 00000000000000001000000000000000
7519 vec($_, 5, 4) = 1 == 1048576 00000000000000000000100000000000
7520 vec($_, 6, 4) = 1 == 16777216 00000000000000000000000010000000
7521 vec($_, 7, 4) = 1 == 268435456 00000000000000000000000000001000
7522 vec($_, 0, 4) = 2 == 2 01000000000000000000000000000000
7523 vec($_, 1, 4) = 2 == 32 00000100000000000000000000000000
7524 vec($_, 2, 4) = 2 == 512 00000000010000000000000000000000
7525 vec($_, 3, 4) = 2 == 8192 00000000000001000000000000000000
7526 vec($_, 4, 4) = 2 == 131072 00000000000000000100000000000000
7527 vec($_, 5, 4) = 2 == 2097152 00000000000000000000010000000000
7528 vec($_, 6, 4) = 2 == 33554432 00000000000000000000000001000000
7529 vec($_, 7, 4) = 2 == 536870912 00000000000000000000000000000100
7530 vec($_, 0, 4) = 4 == 4 00100000000000000000000000000000
7531 vec($_, 1, 4) = 4 == 64 00000010000000000000000000000000
7532 vec($_, 2, 4) = 4 == 1024 00000000001000000000000000000000
7533 vec($_, 3, 4) = 4 == 16384 00000000000000100000000000000000
7534 vec($_, 4, 4) = 4 == 262144 00000000000000000010000000000000
7535 vec($_, 5, 4) = 4 == 4194304 00000000000000000000001000000000
7536 vec($_, 6, 4) = 4 == 67108864 00000000000000000000000000100000
7537 vec($_, 7, 4) = 4 == 1073741824 00000000000000000000000000000010
7538 vec($_, 0, 4) = 8 == 8 00010000000000000000000000000000
7539 vec($_, 1, 4) = 8 == 128 00000001000000000000000000000000
7540 vec($_, 2, 4) = 8 == 2048 00000000000100000000000000000000
7541 vec($_, 3, 4) = 8 == 32768 00000000000000010000000000000000
7542 vec($_, 4, 4) = 8 == 524288 00000000000000000001000000000000
7543 vec($_, 5, 4) = 8 == 8388608 00000000000000000000000100000000
7544 vec($_, 6, 4) = 8 == 134217728 00000000000000000000000000010000
7545 vec($_, 7, 4) = 8 == 2147483648 00000000000000000000000000000001
7546 vec($_, 0, 8) = 1 == 1 10000000000000000000000000000000
7547 vec($_, 1, 8) = 1 == 256 00000000100000000000000000000000
7548 vec($_, 2, 8) = 1 == 65536 00000000000000001000000000000000
7549 vec($_, 3, 8) = 1 == 16777216 00000000000000000000000010000000
7550 vec($_, 0, 8) = 2 == 2 01000000000000000000000000000000
7551 vec($_, 1, 8) = 2 == 512 00000000010000000000000000000000
7552 vec($_, 2, 8) = 2 == 131072 00000000000000000100000000000000
7553 vec($_, 3, 8) = 2 == 33554432 00000000000000000000000001000000
7554 vec($_, 0, 8) = 4 == 4 00100000000000000000000000000000
7555 vec($_, 1, 8) = 4 == 1024 00000000001000000000000000000000
7556 vec($_, 2, 8) = 4 == 262144 00000000000000000010000000000000
7557 vec($_, 3, 8) = 4 == 67108864 00000000000000000000000000100000
7558 vec($_, 0, 8) = 8 == 8 00010000000000000000000000000000
7559 vec($_, 1, 8) = 8 == 2048 00000000000100000000000000000000
7560 vec($_, 2, 8) = 8 == 524288 00000000000000000001000000000000
7561 vec($_, 3, 8) = 8 == 134217728 00000000000000000000000000010000
7562 vec($_, 0, 8) = 16 == 16 00001000000000000000000000000000
7563 vec($_, 1, 8) = 16 == 4096 00000000000010000000000000000000
7564 vec($_, 2, 8) = 16 == 1048576 00000000000000000000100000000000
7565 vec($_, 3, 8) = 16 == 268435456 00000000000000000000000000001000
7566 vec($_, 0, 8) = 32 == 32 00000100000000000000000000000000
7567 vec($_, 1, 8) = 32 == 8192 00000000000001000000000000000000
7568 vec($_, 2, 8) = 32 == 2097152 00000000000000000000010000000000
7569 vec($_, 3, 8) = 32 == 536870912 00000000000000000000000000000100
7570 vec($_, 0, 8) = 64 == 64 00000010000000000000000000000000
7571 vec($_, 1, 8) = 64 == 16384 00000000000000100000000000000000
7572 vec($_, 2, 8) = 64 == 4194304 00000000000000000000001000000000
7573 vec($_, 3, 8) = 64 == 1073741824 00000000000000000000000000000010
7574 vec($_, 0, 8) = 128 == 128 00000001000000000000000000000000
7575 vec($_, 1, 8) = 128 == 32768 00000000000000010000000000000000
7576 vec($_, 2, 8) = 128 == 8388608 00000000000000000000000100000000
7577 vec($_, 3, 8) = 128 == 2147483648 00000000000000000000000000000001
7582 Behaves like wait(2) on your system: it waits for a child
7583 process to terminate and returns the pid of the deceased process, or
7584 C<-1> if there are no child processes. The status is returned in C<$?>
7585 and C<${^CHILD_ERROR_NATIVE}>.
7586 Note that a return value of C<-1> could mean that child processes are
7587 being automatically reaped, as described in L<perlipc>.
7589 =item waitpid PID,FLAGS
7592 Waits for a particular child process to terminate and returns the pid of
7593 the deceased process, or C<-1> if there is no such child process. On some
7594 systems, a value of 0 indicates that there are processes still running.
7595 The status is returned in C<$?> and C<${^CHILD_ERROR_NATIVE}>. If you say
7597 use POSIX ":sys_wait_h";
7600 $kid = waitpid(-1, WNOHANG);
7603 then you can do a non-blocking wait for all pending zombie processes.
7604 Non-blocking wait is available on machines supporting either the
7605 waitpid(2) or wait4(2) syscalls. However, waiting for a particular
7606 pid with FLAGS of C<0> is implemented everywhere. (Perl emulates the
7607 system call by remembering the status values of processes that have
7608 exited but have not been harvested by the Perl script yet.)
7610 Note that on some systems, a return value of C<-1> could mean that child
7611 processes are being automatically reaped. See L<perlipc> for details,
7612 and for other examples.
7615 X<wantarray> X<context>
7617 Returns true if the context of the currently executing subroutine or
7618 C<eval> is looking for a list value. Returns false if the context is
7619 looking for a scalar. Returns the undefined value if the context is
7620 looking for no value (void context).
7622 return unless defined wantarray; # don't bother doing more
7623 my @a = complex_calculation();
7624 return wantarray ? @a : "@a";
7626 C<wantarray()>'s result is unspecified in the top level of a file,
7627 in a C<BEGIN>, C<UNITCHECK>, C<CHECK>, C<INIT> or C<END> block, or
7628 in a C<DESTROY> method.
7630 This function should have been named wantlist() instead.
7633 X<warn> X<warning> X<STDERR>
7635 Prints the value of LIST to STDERR. If the last element of LIST does
7636 not end in a newline, it appends the same file/line number text as C<die>
7639 If the output is empty and C<$@> already contains a value (typically from a
7640 previous eval) that value is used after appending C<"\t...caught">
7641 to C<$@>. This is useful for staying almost, but not entirely similar to
7644 If C<$@> is empty then the string C<"Warning: Something's wrong"> is used.
7646 No message is printed if there is a C<$SIG{__WARN__}> handler
7647 installed. It is the handler's responsibility to deal with the message
7648 as it sees fit (like, for instance, converting it into a C<die>). Most
7649 handlers must therefore arrange to actually display the
7650 warnings that they are not prepared to deal with, by calling C<warn>
7651 again in the handler. Note that this is quite safe and will not
7652 produce an endless loop, since C<__WARN__> hooks are not called from
7655 You will find this behavior is slightly different from that of
7656 C<$SIG{__DIE__}> handlers (which don't suppress the error text, but can
7657 instead call C<die> again to change it).
7659 Using a C<__WARN__> handler provides a powerful way to silence all
7660 warnings (even the so-called mandatory ones). An example:
7662 # wipe out *all* compile-time warnings
7663 BEGIN { $SIG{'__WARN__'} = sub { warn $_[0] if $DOWARN } }
7665 my $foo = 20; # no warning about duplicate my $foo,
7666 # but hey, you asked for it!
7667 # no compile-time or run-time warnings before here
7670 # run-time warnings enabled after here
7671 warn "\$foo is alive and $foo!"; # does show up
7673 See L<perlvar> for details on setting C<%SIG> entries, and for more
7674 examples. See the Carp module for other kinds of warnings using its
7675 carp() and cluck() functions.
7677 =item write FILEHANDLE
7684 Writes a formatted record (possibly multi-line) to the specified FILEHANDLE,
7685 using the format associated with that file. By default the format for
7686 a file is the one having the same name as the filehandle, but the
7687 format for the current output channel (see the C<select> function) may be set
7688 explicitly by assigning the name of the format to the C<$~> variable.
7690 Top of form processing is handled automatically: if there is
7691 insufficient room on the current page for the formatted record, the
7692 page is advanced by writing a form feed, a special top-of-page format
7693 is used to format the new page header, and then the record is written.
7694 By default the top-of-page format is the name of the filehandle with
7695 "_TOP" appended, but it may be dynamically set to the format of your
7696 choice by assigning the name to the C<$^> variable while the filehandle is
7697 selected. The number of lines remaining on the current page is in
7698 variable C<$->, which can be set to C<0> to force a new page.
7700 If FILEHANDLE is unspecified, output goes to the current default output
7701 channel, which starts out as STDOUT but may be changed by the
7702 C<select> operator. If the FILEHANDLE is an EXPR, then the expression
7703 is evaluated and the resulting string is used to look up the name of
7704 the FILEHANDLE at run time. For more on formats, see L<perlform>.
7706 Note that write is I<not> the opposite of C<read>. Unfortunately.
7710 The transliteration operator. Same as C<tr///>. See
7711 L<perlop/"Quote and Quote-like Operators">.