4 perlfunc - Perl builtin functions
8 The functions in this section can serve as terms in an expression.
9 They fall into two major categories: list operators and named unary
10 operators. These differ in their precedence relationship with a
11 following comma. (See the precedence table in L<perlop>.) List
12 operators take more than one argument, while unary operators can never
13 take more than one argument. Thus, a comma terminates the argument of
14 a unary operator, but merely separates the arguments of a list
15 operator. A unary operator generally provides a scalar context to its
16 argument, while a list operator may provide either scalar or list
17 contexts for its arguments. If it does both, the scalar arguments will
18 be first, and the list argument will follow. (Note that there can ever
19 be only one such list argument.) For instance, splice() has three scalar
20 arguments followed by a list, whereas gethostbyname() has four scalar
23 In the syntax descriptions that follow, list operators that expect a
24 list (and provide list context for the elements of the list) are shown
25 with LIST as an argument. Such a list may consist of any combination
26 of scalar arguments or list values; the list values will be included
27 in the list as if each individual element were interpolated at that
28 point in the list, forming a longer single-dimensional list value.
29 Commas should separate elements of the LIST.
31 Any function in the list below may be used either with or without
32 parentheses around its arguments. (The syntax descriptions omit the
33 parentheses.) If you use the parentheses, the simple (but occasionally
34 surprising) rule is this: It I<looks> like a function, therefore it I<is> a
35 function, and precedence doesn't matter. Otherwise it's a list
36 operator or unary operator, and precedence does matter. And whitespace
37 between the function and left parenthesis doesn't count--so you need to
40 print 1+2+4; # Prints 7.
41 print(1+2) + 4; # Prints 3.
42 print (1+2)+4; # Also prints 3!
43 print +(1+2)+4; # Prints 7.
44 print ((1+2)+4); # Prints 7.
46 If you run Perl with the B<-w> switch it can warn you about this. For
47 example, the third line above produces:
49 print (...) interpreted as function at - line 1.
50 Useless use of integer addition in void context at - line 1.
52 A few functions take no arguments at all, and therefore work as neither
53 unary nor list operators. These include such functions as C<time>
54 and C<endpwent>. For example, C<time+86_400> always means
57 For functions that can be used in either a scalar or list context,
58 nonabortive failure is generally indicated in a scalar context by
59 returning the undefined value, and in a list context by returning the
62 Remember the following important rule: There is B<no rule> that relates
63 the behavior of an expression in list context to its behavior in scalar
64 context, or vice versa. It might do two totally different things.
65 Each operator and function decides which sort of value it would be most
66 appropriate to return in scalar context. Some operators return the
67 length of the list that would have been returned in list context. Some
68 operators return the first value in the list. Some operators return the
69 last value in the list. Some operators return a count of successful
70 operations. In general, they do what you want, unless you want
74 A named array in scalar context is quite different from what would at
75 first glance appear to be a list in scalar context. You can't get a list
76 like C<(1,2,3)> into being in scalar context, because the compiler knows
77 the context at compile time. It would generate the scalar comma operator
78 there, not the list construction version of the comma. That means it
79 was never a list to start with.
81 In general, functions in Perl that serve as wrappers for system calls
82 of the same name (like chown(2), fork(2), closedir(2), etc.) all return
83 true when they succeed and C<undef> otherwise, as is usually mentioned
84 in the descriptions below. This is different from the C interfaces,
85 which return C<-1> on failure. Exceptions to this rule are C<wait>,
86 C<waitpid>, and C<syscall>. System calls also set the special C<$!>
87 variable on failure. Other functions do not, except accidentally.
89 =head2 Perl Functions by Category
92 Here are Perl's functions (including things that look like
93 functions, like some keywords and named operators)
94 arranged by category. Some functions appear in more
99 =item Functions for SCALARs or strings
100 X<scalar> X<string> X<character>
102 C<chomp>, C<chop>, C<chr>, C<crypt>, C<hex>, C<index>, C<lc>, C<lcfirst>,
103 C<length>, C<oct>, C<ord>, C<pack>, C<q//>, C<qq//>, C<reverse>,
104 C<rindex>, C<sprintf>, C<substr>, C<tr///>, C<uc>, C<ucfirst>, C<y///>
106 =item Regular expressions and pattern matching
107 X<regular expression> X<regex> X<regexp>
109 C<m//>, C<pos>, C<quotemeta>, C<s///>, C<split>, C<study>, C<qr//>
111 =item Numeric functions
112 X<numeric> X<number> X<trigonometric> X<trigonometry>
114 C<abs>, C<atan2>, C<cos>, C<exp>, C<hex>, C<int>, C<log>, C<oct>, C<rand>,
115 C<sin>, C<sqrt>, C<srand>
117 =item Functions for real @ARRAYs
120 C<pop>, C<push>, C<shift>, C<splice>, C<unshift>
122 =item Functions for list data
125 C<grep>, C<join>, C<map>, C<qw//>, C<reverse>, C<sort>, C<unpack>
127 =item Functions for real %HASHes
130 C<delete>, C<each>, C<exists>, C<keys>, C<values>
132 =item Input and output functions
133 X<I/O> X<input> X<output> X<dbm>
135 C<binmode>, C<close>, C<closedir>, C<dbmclose>, C<dbmopen>, C<die>, C<eof>,
136 C<fileno>, C<flock>, C<format>, C<getc>, C<print>, C<printf>, C<read>,
137 C<readdir>, C<rewinddir>, C<say>, C<seek>, C<seekdir>, C<select>, C<syscall>,
138 C<sysread>, C<sysseek>, C<syswrite>, C<tell>, C<telldir>, C<truncate>,
141 =item Functions for fixed length data or records
143 C<pack>, C<read>, C<syscall>, C<sysread>, C<syswrite>, C<unpack>, C<vec>
145 =item Functions for filehandles, files, or directories
146 X<file> X<filehandle> X<directory> X<pipe> X<link> X<symlink>
148 C<-I<X>>, C<chdir>, C<chmod>, C<chown>, C<chroot>, C<fcntl>, C<glob>,
149 C<ioctl>, C<link>, C<lstat>, C<mkdir>, C<open>, C<opendir>,
150 C<readlink>, C<rename>, C<rmdir>, C<stat>, C<symlink>, C<sysopen>,
151 C<umask>, C<unlink>, C<utime>
153 =item Keywords related to the control flow of your Perl program
156 C<caller>, C<continue>, C<die>, C<do>, C<dump>, C<eval>, C<exit>,
157 C<goto>, C<last>, C<next>, C<redo>, C<return>, C<sub>, C<wantarray>
159 =item Keywords related to switch
161 C<break>, C<continue>, C<given>, C<when>, C<default>
163 (These are only available if you enable the "switch" feature.
164 See L<feature> and L<perlsyn/"Switch statements">.)
166 =item Keywords related to scoping
168 C<caller>, C<import>, C<local>, C<my>, C<our>, C<state>, C<package>,
171 (C<state> is only available if the "state" feature is enabled. See
174 =item Miscellaneous functions
176 C<defined>, C<dump>, C<eval>, C<formline>, C<local>, C<my>, C<our>,
177 C<reset>, C<scalar>, C<state>, C<undef>, C<wantarray>
179 =item Functions for processes and process groups
180 X<process> X<pid> X<process id>
182 C<alarm>, C<exec>, C<fork>, C<getpgrp>, C<getppid>, C<getpriority>, C<kill>,
183 C<pipe>, C<qx//>, C<setpgrp>, C<setpriority>, C<sleep>, C<system>,
184 C<times>, C<wait>, C<waitpid>
186 =item Keywords related to perl modules
189 C<do>, C<import>, C<no>, C<package>, C<require>, C<use>
191 =item Keywords related to classes and object-orientation
192 X<object> X<class> X<package>
194 C<bless>, C<dbmclose>, C<dbmopen>, C<package>, C<ref>, C<tie>, C<tied>,
197 =item Low-level socket functions
200 C<accept>, C<bind>, C<connect>, C<getpeername>, C<getsockname>,
201 C<getsockopt>, C<listen>, C<recv>, C<send>, C<setsockopt>, C<shutdown>,
202 C<socket>, C<socketpair>
204 =item System V interprocess communication functions
205 X<IPC> X<System V> X<semaphore> X<shared memory> X<memory> X<message>
207 C<msgctl>, C<msgget>, C<msgrcv>, C<msgsnd>, C<semctl>, C<semget>, C<semop>,
208 C<shmctl>, C<shmget>, C<shmread>, C<shmwrite>
210 =item Fetching user and group info
211 X<user> X<group> X<password> X<uid> X<gid> X<passwd> X</etc/passwd>
213 C<endgrent>, C<endhostent>, C<endnetent>, C<endpwent>, C<getgrent>,
214 C<getgrgid>, C<getgrnam>, C<getlogin>, C<getpwent>, C<getpwnam>,
215 C<getpwuid>, C<setgrent>, C<setpwent>
217 =item Fetching network info
218 X<network> X<protocol> X<host> X<hostname> X<IP> X<address> X<service>
220 C<endprotoent>, C<endservent>, C<gethostbyaddr>, C<gethostbyname>,
221 C<gethostent>, C<getnetbyaddr>, C<getnetbyname>, C<getnetent>,
222 C<getprotobyname>, C<getprotobynumber>, C<getprotoent>,
223 C<getservbyname>, C<getservbyport>, C<getservent>, C<sethostent>,
224 C<setnetent>, C<setprotoent>, C<setservent>
226 =item Time-related functions
229 C<gmtime>, C<localtime>, C<time>, C<times>
231 =item Functions new in perl5
234 C<abs>, C<bless>, C<break>, C<chomp>, C<chr>, C<continue>, C<default>,
235 C<exists>, C<formline>, C<given>, C<glob>, C<import>, C<lc>, C<lcfirst>,
236 C<lock>, C<map>, C<my>, C<no>, C<our>, C<prototype>, C<qr//>, C<qw//>, C<qx//>,
237 C<readline>, C<readpipe>, C<ref>, C<sub>*, C<sysopen>, C<tie>, C<tied>, C<uc>,
238 C<ucfirst>, C<untie>, C<use>, C<when>
240 * - C<sub> was a keyword in perl4, but in perl5 it is an
241 operator, which can be used in expressions.
243 =item Functions obsoleted in perl5
245 C<dbmclose>, C<dbmopen>
250 X<portability> X<Unix> X<portable>
252 Perl was born in Unix and can therefore access all common Unix
253 system calls. In non-Unix environments, the functionality of some
254 Unix system calls may not be available, or details of the available
255 functionality may differ slightly. The Perl functions affected
258 C<-X>, C<binmode>, C<chmod>, C<chown>, C<chroot>, C<crypt>,
259 C<dbmclose>, C<dbmopen>, C<dump>, C<endgrent>, C<endhostent>,
260 C<endnetent>, C<endprotoent>, C<endpwent>, C<endservent>, C<exec>,
261 C<fcntl>, C<flock>, C<fork>, C<getgrent>, C<getgrgid>, C<gethostbyname>,
262 C<gethostent>, C<getlogin>, C<getnetbyaddr>, C<getnetbyname>, C<getnetent>,
263 C<getppid>, C<getpgrp>, C<getpriority>, C<getprotobynumber>,
264 C<getprotoent>, C<getpwent>, C<getpwnam>, C<getpwuid>,
265 C<getservbyport>, C<getservent>, C<getsockopt>, C<glob>, C<ioctl>,
266 C<kill>, C<link>, C<lstat>, C<msgctl>, C<msgget>, C<msgrcv>,
267 C<msgsnd>, C<open>, C<pipe>, C<readlink>, C<rename>, C<select>, C<semctl>,
268 C<semget>, C<semop>, C<setgrent>, C<sethostent>, C<setnetent>,
269 C<setpgrp>, C<setpriority>, C<setprotoent>, C<setpwent>,
270 C<setservent>, C<setsockopt>, C<shmctl>, C<shmget>, C<shmread>,
271 C<shmwrite>, C<socket>, C<socketpair>,
272 C<stat>, C<symlink>, C<syscall>, C<sysopen>, C<system>,
273 C<times>, C<truncate>, C<umask>, C<unlink>,
274 C<utime>, C<wait>, C<waitpid>
276 For more information about the portability of these functions, see
277 L<perlport> and other available platform-specific documentation.
279 =head2 Alphabetical Listing of Perl Functions
284 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>
285 X<-S>X<-b>X<-c>X<-t>X<-u>X<-g>X<-k>X<-T>X<-B>X<-M>X<-A>X<-C>
293 A file test, where X is one of the letters listed below. This unary
294 operator takes one argument, either a filename, a filehandle, or a dirhandle,
295 and tests the associated file to see if something is true about it. If the
296 argument is omitted, tests C<$_>, except for C<-t>, which tests STDIN.
297 Unless otherwise documented, it returns C<1> for true and C<''> for false, or
298 the undefined value if the file doesn't exist. Despite the funny
299 names, precedence is the same as any other named unary operator. The
300 operator may be any of:
302 -r File is readable by effective uid/gid.
303 -w File is writable by effective uid/gid.
304 -x File is executable by effective uid/gid.
305 -o File is owned by effective uid.
307 -R File is readable by real uid/gid.
308 -W File is writable by real uid/gid.
309 -X File is executable by real uid/gid.
310 -O File is owned by real uid.
313 -z File has zero size (is empty).
314 -s File has nonzero size (returns size in bytes).
316 -f File is a plain file.
317 -d File is a directory.
318 -l File is a symbolic link.
319 -p File is a named pipe (FIFO), or Filehandle is a pipe.
321 -b File is a block special file.
322 -c File is a character special file.
323 -t Filehandle is opened to a tty.
325 -u File has setuid bit set.
326 -g File has setgid bit set.
327 -k File has sticky bit set.
329 -T File is an ASCII text file (heuristic guess).
330 -B File is a "binary" file (opposite of -T).
332 -M Script start time minus file modification time, in days.
333 -A Same for access time.
334 -C Same for inode change time (Unix, may differ for other platforms)
340 next unless -f $_; # ignore specials
344 The interpretation of the file permission operators C<-r>, C<-R>,
345 C<-w>, C<-W>, C<-x>, and C<-X> is by default based solely on the mode
346 of the file and the uids and gids of the user. There may be other
347 reasons you can't actually read, write, or execute the file: for
348 example network filesystem access controls, ACLs (access control lists),
349 read-only filesystems, and unrecognized executable formats. Note
350 that the use of these six specific operators to verify if some operation
351 is possible is usually a mistake, because it may be open to race
354 Also note that, for the superuser on the local filesystems, the C<-r>,
355 C<-R>, C<-w>, and C<-W> tests always return 1, and C<-x> and C<-X> return 1
356 if any execute bit is set in the mode. Scripts run by the superuser
357 may thus need to do a stat() to determine the actual mode of the file,
358 or temporarily set their effective uid to something else.
360 If you are using ACLs, there is a pragma called C<filetest> that may
361 produce more accurate results than the bare stat() mode bits.
362 When under the C<use filetest 'access'> the above-mentioned filetests
363 will test whether the permission can (not) be granted using the
364 access() family of system calls. Also note that the C<-x> and C<-X> may
365 under this pragma return true even if there are no execute permission
366 bits set (nor any extra execute permission ACLs). This strangeness is
367 due to the underlying system calls' definitions. Note also that, due to
368 the implementation of C<use filetest 'access'>, the C<_> special
369 filehandle won't cache the results of the file tests when this pragma is
370 in effect. Read the documentation for the C<filetest> pragma for more
373 Note that C<-s/a/b/> does not do a negated substitution. Saying
374 C<-exp($foo)> still works as expected, however--only single letters
375 following a minus are interpreted as file tests.
377 The C<-T> and C<-B> switches work as follows. The first block or so of the
378 file is examined for odd characters such as strange control codes or
379 characters with the high bit set. If too many strange characters (>30%)
380 are found, it's a C<-B> file; otherwise it's a C<-T> file. Also, any file
381 containing null in the first block is considered a binary file. If C<-T>
382 or C<-B> is used on a filehandle, the current IO buffer is examined
383 rather than the first block. Both C<-T> and C<-B> return true on a null
384 file, or a file at EOF when testing a filehandle. Because you have to
385 read a file to do the C<-T> test, on most occasions you want to use a C<-f>
386 against the file first, as in C<next unless -f $file && -T $file>.
388 If any of the file tests (or either the C<stat> or C<lstat> operators) are given
389 the special filehandle consisting of a solitary underline, then the stat
390 structure of the previous file test (or stat operator) is used, saving
391 a system call. (This doesn't work with C<-t>, and you need to remember
392 that lstat() and C<-l> will leave values in the stat structure for the
393 symbolic link, not the real file.) (Also, if the stat buffer was filled by
394 an C<lstat> call, C<-T> and C<-B> will reset it with the results of C<stat _>).
397 print "Can do.\n" if -r $a || -w _ || -x _;
400 print "Readable\n" if -r _;
401 print "Writable\n" if -w _;
402 print "Executable\n" if -x _;
403 print "Setuid\n" if -u _;
404 print "Setgid\n" if -g _;
405 print "Sticky\n" if -k _;
406 print "Text\n" if -T _;
407 print "Binary\n" if -B _;
409 As of Perl 5.9.1, as a form of purely syntactic sugar, you can stack file
410 test operators, in a way that C<-f -w -x $file> is equivalent to
411 C<-x $file && -w _ && -f _>. (This is only syntax fancy: if you use
412 the return value of C<-f $file> as an argument to another filetest
413 operator, no special magic will happen.)
420 Returns the absolute value of its argument.
421 If VALUE is omitted, uses C<$_>.
423 =item accept NEWSOCKET,GENERICSOCKET
426 Accepts an incoming socket connect, just as the accept(2) system call
427 does. Returns the packed address if it succeeded, false otherwise.
428 See the example in L<perlipc/"Sockets: Client/Server Communication">.
430 On systems that support a close-on-exec flag on files, the flag will
431 be set for the newly opened file descriptor, as determined by the
432 value of $^F. See L<perlvar/$^F>.
441 Arranges to have a SIGALRM delivered to this process after the
442 specified number of wallclock seconds has elapsed. If SECONDS is not
443 specified, the value stored in C<$_> is used. (On some machines,
444 unfortunately, the elapsed time may be up to one second less or more
445 than you specified because of how seconds are counted, and process
446 scheduling may delay the delivery of the signal even further.)
448 Only one timer may be counting at once. Each call disables the
449 previous timer, and an argument of C<0> may be supplied to cancel the
450 previous timer without starting a new one. The returned value is the
451 amount of time remaining on the previous timer.
453 For delays of finer granularity than one second, the Time::HiRes module
454 (from CPAN, and starting from Perl 5.8 part of the standard
455 distribution) provides ualarm(). You may also use Perl's four-argument
456 version of select() leaving the first three arguments undefined, or you
457 might be able to use the C<syscall> interface to access setitimer(2) if
458 your system supports it. See L<perlfaq8> for details.
460 It is usually a mistake to intermix C<alarm> and C<sleep> calls.
461 (C<sleep> may be internally implemented in your system with C<alarm>)
463 If you want to use C<alarm> to time out a system call you need to use an
464 C<eval>/C<die> pair. You can't rely on the alarm causing the system call to
465 fail with C<$!> set to C<EINTR> because Perl sets up signal handlers to
466 restart system calls on some systems. Using C<eval>/C<die> always works,
467 modulo the caveats given in L<perlipc/"Signals">.
470 local $SIG{ALRM} = sub { die "alarm\n" }; # NB: \n required
472 $nread = sysread SOCKET, $buffer, $size;
476 die unless $@ eq "alarm\n"; # propagate unexpected errors
483 For more information see L<perlipc>.
486 X<atan2> X<arctangent> X<tan> X<tangent>
488 Returns the arctangent of Y/X in the range -PI to PI.
490 For the tangent operation, you may use the C<Math::Trig::tan>
491 function, or use the familiar relation:
493 sub tan { sin($_[0]) / cos($_[0]) }
495 The return value for C<atan2(0,0)> is implementation-defined; consult
496 your atan2(3) manpage for more information.
498 =item bind SOCKET,NAME
501 Binds a network address to a socket, just as the bind system call
502 does. Returns true if it succeeded, false otherwise. NAME should be a
503 packed address of the appropriate type for the socket. See the examples in
504 L<perlipc/"Sockets: Client/Server Communication">.
506 =item binmode FILEHANDLE, LAYER
507 X<binmode> X<binary> X<text> X<DOS> X<Windows>
509 =item binmode FILEHANDLE
511 Arranges for FILEHANDLE to be read or written in "binary" or "text"
512 mode on systems where the run-time libraries distinguish between
513 binary and text files. If FILEHANDLE is an expression, the value is
514 taken as the name of the filehandle. Returns true on success,
515 otherwise it returns C<undef> and sets C<$!> (errno).
517 On some systems (in general, DOS and Windows-based systems) binmode()
518 is necessary when you're not working with a text file. For the sake
519 of portability it is a good idea to always use it when appropriate,
520 and to never use it when it isn't appropriate. Also, people can
521 set their I/O to be by default UTF-8 encoded Unicode, not bytes.
523 In other words: regardless of platform, use binmode() on binary data,
524 like for example images.
526 If LAYER is present it is a single string, but may contain multiple
527 directives. The directives alter the behaviour of the file handle.
528 When LAYER is present using binmode on a text file makes sense.
530 If LAYER is omitted or specified as C<:raw> the filehandle is made
531 suitable for passing binary data. This includes turning off possible CRLF
532 translation and marking it as bytes (as opposed to Unicode characters).
533 Note that, despite what may be implied in I<"Programming Perl"> (the
534 Camel) or elsewhere, C<:raw> is I<not> simply the inverse of C<:crlf>
535 -- other layers which would affect the binary nature of the stream are
536 I<also> disabled. See L<PerlIO>, L<perlrun> and the discussion about the
537 PERLIO environment variable.
539 The C<:bytes>, C<:crlf>, and C<:utf8>, and any other directives of the
540 form C<:...>, are called I/O I<layers>. The C<open> pragma can be used to
541 establish default I/O layers. See L<open>.
543 I<The LAYER parameter of the binmode() function is described as "DISCIPLINE"
544 in "Programming Perl, 3rd Edition". However, since the publishing of this
545 book, by many known as "Camel III", the consensus of the naming of this
546 functionality has moved from "discipline" to "layer". All documentation
547 of this version of Perl therefore refers to "layers" rather than to
548 "disciplines". Now back to the regularly scheduled documentation...>
550 To mark FILEHANDLE as UTF-8, use C<:utf8> or C<:encoding(utf8)>.
551 C<:utf8> just marks the data as UTF-8 without further checking,
552 while C<:encoding(utf8)> checks the data for actually being valid
553 UTF-8. More details can be found in L<PerlIO::encoding>.
555 In general, binmode() should be called after open() but before any I/O
556 is done on the filehandle. Calling binmode() will normally flush any
557 pending buffered output data (and perhaps pending input data) on the
558 handle. An exception to this is the C<:encoding> layer that
559 changes the default character encoding of the handle, see L<open>.
560 The C<:encoding> layer sometimes needs to be called in
561 mid-stream, and it doesn't flush the stream. The C<:encoding>
562 also implicitly pushes on top of itself the C<:utf8> layer because
563 internally Perl will operate on UTF-8 encoded Unicode characters.
565 The operating system, device drivers, C libraries, and Perl run-time
566 system all work together to let the programmer treat a single
567 character (C<\n>) as the line terminator, irrespective of the external
568 representation. On many operating systems, the native text file
569 representation matches the internal representation, but on some
570 platforms the external representation of C<\n> is made up of more than
573 Mac OS, all variants of Unix, and Stream_LF files on VMS use a single
574 character to end each line in the external representation of text (even
575 though that single character is CARRIAGE RETURN on Mac OS and LINE FEED
576 on Unix and most VMS files). In other systems like OS/2, DOS and the
577 various flavors of MS-Windows your program sees a C<\n> as a simple C<\cJ>,
578 but what's stored in text files are the two characters C<\cM\cJ>. That
579 means that, if you don't use binmode() on these systems, C<\cM\cJ>
580 sequences on disk will be converted to C<\n> on input, and any C<\n> in
581 your program will be converted back to C<\cM\cJ> on output. This is what
582 you want for text files, but it can be disastrous for binary files.
584 Another consequence of using binmode() (on some systems) is that
585 special end-of-file markers will be seen as part of the data stream.
586 For systems from the Microsoft family this means that if your binary
587 data contains C<\cZ>, the I/O subsystem will regard it as the end of
588 the file, unless you use binmode().
590 binmode() is not only important for readline() and print() operations,
591 but also when using read(), seek(), sysread(), syswrite() and tell()
592 (see L<perlport> for more details). See the C<$/> and C<$\> variables
593 in L<perlvar> for how to manually set your input and output
594 line-termination sequences.
596 =item bless REF,CLASSNAME
601 This function tells the thingy referenced by REF that it is now an object
602 in the CLASSNAME package. If CLASSNAME is omitted, the current package
603 is used. Because a C<bless> is often the last thing in a constructor,
604 it returns the reference for convenience. Always use the two-argument
605 version if a derived class might inherit the function doing the blessing.
606 See L<perltoot> and L<perlobj> for more about the blessing (and blessings)
609 Consider always blessing objects in CLASSNAMEs that are mixed case.
610 Namespaces with all lowercase names are considered reserved for
611 Perl pragmata. Builtin types have all uppercase names. To prevent
612 confusion, you may wish to avoid such package names as well. Make sure
613 that CLASSNAME is a true value.
615 See L<perlmod/"Perl Modules">.
619 Break out of a C<given()> block.
621 This keyword is enabled by the "switch" feature: see L<feature>
622 for more information.
625 X<caller> X<call stack> X<stack> X<stack trace>
629 Returns the context of the current subroutine call. In scalar context,
630 returns the caller's package name if there is a caller, that is, if
631 we're in a subroutine or C<eval> or C<require>, and the undefined value
632 otherwise. In list context, returns
635 ($package, $filename, $line) = caller;
637 With EXPR, it returns some extra information that the debugger uses to
638 print a stack trace. The value of EXPR indicates how many call frames
639 to go back before the current one.
642 ($package, $filename, $line, $subroutine, $hasargs,
645 $wantarray, $evaltext, $is_require, $hints, $bitmask, $hinthash)
648 Here $subroutine may be C<(eval)> if the frame is not a subroutine
649 call, but an C<eval>. In such a case additional elements $evaltext and
650 C<$is_require> are set: C<$is_require> is true if the frame is created by a
651 C<require> or C<use> statement, $evaltext contains the text of the
652 C<eval EXPR> statement. In particular, for an C<eval BLOCK> statement,
653 $subroutine is C<(eval)>, but $evaltext is undefined. (Note also that
654 each C<use> statement creates a C<require> frame inside an C<eval EXPR>
655 frame.) $subroutine may also be C<(unknown)> if this particular
656 subroutine happens to have been deleted from the symbol table.
657 C<$hasargs> is true if a new instance of C<@_> was set up for the frame.
658 C<$hints> and C<$bitmask> contain pragmatic hints that the caller was
659 compiled with. The C<$hints> and C<$bitmask> values are subject to change
660 between versions of Perl, and are not meant for external use.
662 C<$hinthash> is a reference to a hash containing the value of C<%^H> when the
663 caller was compiled, or C<undef> if C<%^H> was empty. Do not modify the values
664 of this hash, as they are the actual values stored in the optree.
666 Furthermore, when called from within the DB package, caller returns more
667 detailed information: it sets the list variable C<@DB::args> to be the
668 arguments with which the subroutine was invoked.
670 Be aware that the optimizer might have optimized call frames away before
671 C<caller> had a chance to get the information. That means that C<caller(N)>
672 might not return information about the call frame you expect it do, for
673 C<< N > 1 >>. In particular, C<@DB::args> might have information from the
674 previous time C<caller> was called.
681 =item chdir FILEHANDLE
683 =item chdir DIRHANDLE
687 Changes the working directory to EXPR, if possible. If EXPR is omitted,
688 changes to the directory specified by C<$ENV{HOME}>, if set; if not,
689 changes to the directory specified by C<$ENV{LOGDIR}>. (Under VMS, the
690 variable C<$ENV{SYS$LOGIN}> is also checked, and used if it is set.) If
691 neither is set, C<chdir> does nothing. It returns true upon success,
692 false otherwise. See the example under C<die>.
694 On systems that support fchdir, you might pass a file handle or
695 directory handle as argument. On systems that don't support fchdir,
696 passing handles produces a fatal error at run time.
699 X<chmod> X<permission> X<mode>
701 Changes the permissions of a list of files. The first element of the
702 list must be the numerical mode, which should probably be an octal
703 number, and which definitely should I<not> be a string of octal digits:
704 C<0644> is okay, C<'0644'> is not. Returns the number of files
705 successfully changed. See also L</oct>, if all you have is a string.
707 $cnt = chmod 0755, 'foo', 'bar';
708 chmod 0755, @executables;
709 $mode = '0644'; chmod $mode, 'foo'; # !!! sets mode to
711 $mode = '0644'; chmod oct($mode), 'foo'; # this is better
712 $mode = 0644; chmod $mode, 'foo'; # this is best
714 On systems that support fchmod, you might pass file handles among the
715 files. On systems that don't support fchmod, passing file handles
716 produces a fatal error at run time. The file handles must be passed
717 as globs or references to be recognized. Barewords are considered
720 open(my $fh, "<", "foo");
721 my $perm = (stat $fh)[2] & 07777;
722 chmod($perm | 0600, $fh);
724 You can also import the symbolic C<S_I*> constants from the Fcntl
729 chmod S_IRWXU|S_IRGRP|S_IXGRP|S_IROTH|S_IXOTH, @executables;
730 # This is identical to the chmod 0755 of the above example.
733 X<chomp> X<INPUT_RECORD_SEPARATOR> X<$/> X<newline> X<eol>
739 This safer version of L</chop> removes any trailing string
740 that corresponds to the current value of C<$/> (also known as
741 $INPUT_RECORD_SEPARATOR in the C<English> module). It returns the total
742 number of characters removed from all its arguments. It's often used to
743 remove the newline from the end of an input record when you're worried
744 that the final record may be missing its newline. When in paragraph
745 mode (C<$/ = "">), it removes all trailing newlines from the string.
746 When in slurp mode (C<$/ = undef>) or fixed-length record mode (C<$/> is
747 a reference to an integer or the like, see L<perlvar>) chomp() won't
749 If VARIABLE is omitted, it chomps C<$_>. Example:
752 chomp; # avoid \n on last field
757 If VARIABLE is a hash, it chomps the hash's values, but not its keys.
759 You can actually chomp anything that's an lvalue, including an assignment:
762 chomp($answer = <STDIN>);
764 If you chomp a list, each element is chomped, and the total number of
765 characters removed is returned.
767 Note that parentheses are necessary when you're chomping anything
768 that is not a simple variable. This is because C<chomp $cwd = `pwd`;>
769 is interpreted as C<(chomp $cwd) = `pwd`;>, rather than as
770 C<chomp( $cwd = `pwd` )> which you might expect. Similarly,
771 C<chomp $a, $b> is interpreted as C<chomp($a), $b> rather than
781 Chops off the last character of a string and returns the character
782 chopped. It is much more efficient than C<s/.$//s> because it neither
783 scans nor copies the string. If VARIABLE is omitted, chops C<$_>.
784 If VARIABLE is a hash, it chops the hash's values, but not its keys.
786 You can actually chop anything that's an lvalue, including an assignment.
788 If you chop a list, each element is chopped. Only the value of the
789 last C<chop> is returned.
791 Note that C<chop> returns the last character. To return all but the last
792 character, use C<substr($string, 0, -1)>.
797 X<chown> X<owner> X<user> X<group>
799 Changes the owner (and group) of a list of files. The first two
800 elements of the list must be the I<numeric> uid and gid, in that
801 order. A value of -1 in either position is interpreted by most
802 systems to leave that value unchanged. Returns the number of files
803 successfully changed.
805 $cnt = chown $uid, $gid, 'foo', 'bar';
806 chown $uid, $gid, @filenames;
808 On systems that support fchown, you might pass file handles among the
809 files. On systems that don't support fchown, passing file handles
810 produces a fatal error at run time. The file handles must be passed
811 as globs or references to be recognized. Barewords are considered
814 Here's an example that looks up nonnumeric uids in the passwd file:
817 chomp($user = <STDIN>);
819 chomp($pattern = <STDIN>);
821 ($login,$pass,$uid,$gid) = getpwnam($user)
822 or die "$user not in passwd file";
824 @ary = glob($pattern); # expand filenames
825 chown $uid, $gid, @ary;
827 On most systems, you are not allowed to change the ownership of the
828 file unless you're the superuser, although you should be able to change
829 the group to any of your secondary groups. On insecure systems, these
830 restrictions may be relaxed, but this is not a portable assumption.
831 On POSIX systems, you can detect this condition this way:
833 use POSIX qw(sysconf _PC_CHOWN_RESTRICTED);
834 $can_chown_giveaway = not sysconf(_PC_CHOWN_RESTRICTED);
837 X<chr> X<character> X<ASCII> X<Unicode>
841 Returns the character represented by that NUMBER in the character set.
842 For example, C<chr(65)> is C<"A"> in either ASCII or Unicode, and
843 chr(0x263a) is a Unicode smiley face.
845 Negative values give the Unicode replacement character (chr(0xfffd)),
846 except under the L<bytes> pragma, where low eight bits of the value
847 (truncated to an integer) are used.
849 If NUMBER is omitted, uses C<$_>.
851 For the reverse, use L</ord>.
853 Note that characters from 128 to 255 (inclusive) are by default
854 internally not encoded as UTF-8 for backward compatibility reasons.
856 See L<perlunicode> for more about Unicode.
858 =item chroot FILENAME
863 This function works like the system call by the same name: it makes the
864 named directory the new root directory for all further pathnames that
865 begin with a C</> by your process and all its children. (It doesn't
866 change your current working directory, which is unaffected.) For security
867 reasons, this call is restricted to the superuser. If FILENAME is
868 omitted, does a C<chroot> to C<$_>.
870 =item close FILEHANDLE
875 Closes the file or pipe associated with the file handle, flushes the IO
876 buffers, and closes the system file descriptor. Returns true if those
877 operations have succeeded and if no error was reported by any PerlIO
878 layer. Closes the currently selected filehandle if the argument is
881 You don't have to close FILEHANDLE if you are immediately going to do
882 another C<open> on it, because C<open> will close it for you. (See
883 C<open>.) However, an explicit C<close> on an input file resets the line
884 counter (C<$.>), while the implicit close done by C<open> does not.
886 If the file handle came from a piped open, C<close> will additionally
887 return false if one of the other system calls involved fails, or if the
888 program exits with non-zero status. (If the only problem was that the
889 program exited non-zero, C<$!> will be set to C<0>.) Closing a pipe
890 also waits for the process executing on the pipe to complete, in case you
891 want to look at the output of the pipe afterwards, and
892 implicitly puts the exit status value of that command into C<$?> and
893 C<${^CHILD_ERROR_NATIVE}>.
895 Prematurely closing the read end of a pipe (i.e. before the process
896 writing to it at the other end has closed it) will result in a
897 SIGPIPE being delivered to the writer. If the other end can't
898 handle that, be sure to read all the data before closing the pipe.
902 open(OUTPUT, '|sort >foo') # pipe to sort
903 or die "Can't start sort: $!";
904 #... # print stuff to output
905 close OUTPUT # wait for sort to finish
906 or warn $! ? "Error closing sort pipe: $!"
907 : "Exit status $? from sort";
908 open(INPUT, 'foo') # get sort's results
909 or die "Can't open 'foo' for input: $!";
911 FILEHANDLE may be an expression whose value can be used as an indirect
912 filehandle, usually the real filehandle name.
914 =item closedir DIRHANDLE
917 Closes a directory opened by C<opendir> and returns the success of that
920 =item connect SOCKET,NAME
923 Attempts to connect to a remote socket, just as the connect system call
924 does. Returns true if it succeeded, false otherwise. NAME should be a
925 packed address of the appropriate type for the socket. See the examples in
926 L<perlipc/"Sockets: Client/Server Communication">.
933 C<continue> is actually a flow control statement rather than a function. If
934 there is a C<continue> BLOCK attached to a BLOCK (typically in a C<while> or
935 C<foreach>), it is always executed just before the conditional is about to
936 be evaluated again, just like the third part of a C<for> loop in C. Thus
937 it can be used to increment a loop variable, even when the loop has been
938 continued via the C<next> statement (which is similar to the C C<continue>
941 C<last>, C<next>, or C<redo> may appear within a C<continue>
942 block. C<last> and C<redo> will behave as if they had been executed within
943 the main block. So will C<next>, but since it will execute a C<continue>
944 block, it may be more entertaining.
947 ### redo always comes here
950 ### next always comes here
952 # then back the top to re-check EXPR
954 ### last always comes here
956 Omitting the C<continue> section is semantically equivalent to using an
957 empty one, logically enough. In that case, C<next> goes directly back
958 to check the condition at the top of the loop.
960 If the "switch" feature is enabled, C<continue> is also a
961 function that will break out of the current C<when> or C<default>
962 block, and fall through to the next case. See L<feature> and
963 L<perlsyn/"Switch statements"> for more information.
967 X<cos> X<cosine> X<acos> X<arccosine>
971 Returns the cosine of EXPR (expressed in radians). If EXPR is omitted,
972 takes cosine of C<$_>.
974 For the inverse cosine operation, you may use the C<Math::Trig::acos()>
975 function, or use this relation:
977 sub acos { atan2( sqrt(1 - $_[0] * $_[0]), $_[0] ) }
979 =item crypt PLAINTEXT,SALT
980 X<crypt> X<digest> X<hash> X<salt> X<plaintext> X<password>
981 X<decrypt> X<cryptography> X<passwd> X<encrypt>
983 Creates a digest string exactly like the crypt(3) function in the C
984 library (assuming that you actually have a version there that has not
985 been extirpated as a potential munition).
987 crypt() is a one-way hash function. The PLAINTEXT and SALT is turned
988 into a short string, called a digest, which is returned. The same
989 PLAINTEXT and SALT will always return the same string, but there is no
990 (known) way to get the original PLAINTEXT from the hash. Small
991 changes in the PLAINTEXT or SALT will result in large changes in the
994 There is no decrypt function. This function isn't all that useful for
995 cryptography (for that, look for F<Crypt> modules on your nearby CPAN
996 mirror) and the name "crypt" is a bit of a misnomer. Instead it is
997 primarily used to check if two pieces of text are the same without
998 having to transmit or store the text itself. An example is checking
999 if a correct password is given. The digest of the password is stored,
1000 not the password itself. The user types in a password that is
1001 crypt()'d with the same salt as the stored digest. If the two digests
1002 match the password is correct.
1004 When verifying an existing digest string you should use the digest as
1005 the salt (like C<crypt($plain, $digest) eq $digest>). The SALT used
1006 to create the digest is visible as part of the digest. This ensures
1007 crypt() will hash the new string with the same salt as the digest.
1008 This allows your code to work with the standard L<crypt|/crypt> and
1009 with more exotic implementations. In other words, do not assume
1010 anything about the returned string itself, or how many bytes in the
1013 Traditionally the result is a string of 13 bytes: two first bytes of
1014 the salt, followed by 11 bytes from the set C<[./0-9A-Za-z]>, and only
1015 the first eight bytes of PLAINTEXT mattered. But alternative
1016 hashing schemes (like MD5), higher level security schemes (like C2),
1017 and implementations on non-UNIX platforms may produce different
1020 When choosing a new salt create a random two character string whose
1021 characters come from the set C<[./0-9A-Za-z]> (like C<join '', ('.',
1022 '/', 0..9, 'A'..'Z', 'a'..'z')[rand 64, rand 64]>). This set of
1023 characters is just a recommendation; the characters allowed in
1024 the salt depend solely on your system's crypt library, and Perl can't
1025 restrict what salts C<crypt()> accepts.
1027 Here's an example that makes sure that whoever runs this program knows
1030 $pwd = (getpwuid($<))[1];
1032 system "stty -echo";
1034 chomp($word = <STDIN>);
1038 if (crypt($word, $pwd) ne $pwd) {
1044 Of course, typing in your own password to whoever asks you
1047 The L<crypt|/crypt> function is unsuitable for hashing large quantities
1048 of data, not least of all because you can't get the information
1049 back. Look at the L<Digest> module for more robust algorithms.
1051 If using crypt() on a Unicode string (which I<potentially> has
1052 characters with codepoints above 255), Perl tries to make sense
1053 of the situation by trying to downgrade (a copy of the string)
1054 the string back to an eight-bit byte string before calling crypt()
1055 (on that copy). If that works, good. If not, crypt() dies with
1056 C<Wide character in crypt>.
1061 [This function has been largely superseded by the C<untie> function.]
1063 Breaks the binding between a DBM file and a hash.
1065 =item dbmopen HASH,DBNAME,MASK
1066 X<dbmopen> X<dbm> X<ndbm> X<sdbm> X<gdbm>
1068 [This function has been largely superseded by the C<tie> function.]
1070 This binds a dbm(3), ndbm(3), sdbm(3), gdbm(3), or Berkeley DB file to a
1071 hash. HASH is the name of the hash. (Unlike normal C<open>, the first
1072 argument is I<not> a filehandle, even though it looks like one). DBNAME
1073 is the name of the database (without the F<.dir> or F<.pag> extension if
1074 any). If the database does not exist, it is created with protection
1075 specified by MASK (as modified by the C<umask>). If your system supports
1076 only the older DBM functions, you may perform only one C<dbmopen> in your
1077 program. In older versions of Perl, if your system had neither DBM nor
1078 ndbm, calling C<dbmopen> produced a fatal error; it now falls back to
1081 If you don't have write access to the DBM file, you can only read hash
1082 variables, not set them. If you want to test whether you can write,
1083 either use file tests or try setting a dummy hash entry inside an C<eval>,
1084 which will trap the error.
1086 Note that functions such as C<keys> and C<values> may return huge lists
1087 when used on large DBM files. You may prefer to use the C<each>
1088 function to iterate over large DBM files. Example:
1090 # print out history file offsets
1091 dbmopen(%HIST,'/usr/lib/news/history',0666);
1092 while (($key,$val) = each %HIST) {
1093 print $key, ' = ', unpack('L',$val), "\n";
1097 See also L<AnyDBM_File> for a more general description of the pros and
1098 cons of the various dbm approaches, as well as L<DB_File> for a particularly
1099 rich implementation.
1101 You can control which DBM library you use by loading that library
1102 before you call dbmopen():
1105 dbmopen(%NS_Hist, "$ENV{HOME}/.netscape/history.db")
1106 or die "Can't open netscape history file: $!";
1109 X<defined> X<undef> X<undefined>
1113 Returns a Boolean value telling whether EXPR has a value other than
1114 the undefined value C<undef>. If EXPR is not present, C<$_> will be
1117 Many operations return C<undef> to indicate failure, end of file,
1118 system error, uninitialized variable, and other exceptional
1119 conditions. This function allows you to distinguish C<undef> from
1120 other values. (A simple Boolean test will not distinguish among
1121 C<undef>, zero, the empty string, and C<"0">, which are all equally
1122 false.) Note that since C<undef> is a valid scalar, its presence
1123 doesn't I<necessarily> indicate an exceptional condition: C<pop>
1124 returns C<undef> when its argument is an empty array, I<or> when the
1125 element to return happens to be C<undef>.
1127 You may also use C<defined(&func)> to check whether subroutine C<&func>
1128 has ever been defined. The return value is unaffected by any forward
1129 declarations of C<&func>. Note that a subroutine which is not defined
1130 may still be callable: its package may have an C<AUTOLOAD> method that
1131 makes it spring into existence the first time that it is called -- see
1134 Use of C<defined> on aggregates (hashes and arrays) is deprecated. It
1135 used to report whether memory for that aggregate has ever been
1136 allocated. This behavior may disappear in future versions of Perl.
1137 You should instead use a simple test for size:
1139 if (@an_array) { print "has array elements\n" }
1140 if (%a_hash) { print "has hash members\n" }
1142 When used on a hash element, it tells you whether the value is defined,
1143 not whether the key exists in the hash. Use L</exists> for the latter
1148 print if defined $switch{'D'};
1149 print "$val\n" while defined($val = pop(@ary));
1150 die "Can't readlink $sym: $!"
1151 unless defined($value = readlink $sym);
1152 sub foo { defined &$bar ? &$bar(@_) : die "No bar"; }
1153 $debugging = 0 unless defined $debugging;
1155 Note: Many folks tend to overuse C<defined>, and then are surprised to
1156 discover that the number C<0> and C<""> (the zero-length string) are, in fact,
1157 defined values. For example, if you say
1161 The pattern match succeeds, and C<$1> is defined, despite the fact that it
1162 matched "nothing". It didn't really fail to match anything. Rather, it
1163 matched something that happened to be zero characters long. This is all
1164 very above-board and honest. When a function returns an undefined value,
1165 it's an admission that it couldn't give you an honest answer. So you
1166 should use C<defined> only when you're questioning the integrity of what
1167 you're trying to do. At other times, a simple comparison to C<0> or C<""> is
1170 See also L</undef>, L</exists>, L</ref>.
1175 Given an expression that specifies a hash element, array element, hash slice,
1176 or array slice, deletes the specified element(s) from the hash or array.
1177 In the case of an array, if the array elements happen to be at the end,
1178 the size of the array will shrink to the highest element that tests
1179 true for exists() (or 0 if no such element exists).
1181 Returns a list with the same number of elements as the number of elements
1182 for which deletion was attempted. Each element of that list consists of
1183 either the value of the element deleted, or the undefined value. In scalar
1184 context, this means that you get the value of the last element deleted (or
1185 the undefined value if that element did not exist).
1187 %hash = (foo => 11, bar => 22, baz => 33);
1188 $scalar = delete $hash{foo}; # $scalar is 11
1189 $scalar = delete @hash{qw(foo bar)}; # $scalar is 22
1190 @array = delete @hash{qw(foo bar baz)}; # @array is (undef,undef,33)
1192 Deleting from C<%ENV> modifies the environment. Deleting from
1193 a hash tied to a DBM file deletes the entry from the DBM file. Deleting
1194 from a C<tie>d hash or array may not necessarily return anything.
1196 Deleting an array element effectively returns that position of the array
1197 to its initial, uninitialized state. Subsequently testing for the same
1198 element with exists() will return false. Also, deleting array elements
1199 in the middle of an array will not shift the index of the elements
1200 after them down. Use splice() for that. See L</exists>.
1202 The following (inefficiently) deletes all the values of %HASH and @ARRAY:
1204 foreach $key (keys %HASH) {
1208 foreach $index (0 .. $#ARRAY) {
1209 delete $ARRAY[$index];
1214 delete @HASH{keys %HASH};
1216 delete @ARRAY[0 .. $#ARRAY];
1218 But both of these are slower than just assigning the empty list
1219 or undefining %HASH or @ARRAY:
1221 %HASH = (); # completely empty %HASH
1222 undef %HASH; # forget %HASH ever existed
1224 @ARRAY = (); # completely empty @ARRAY
1225 undef @ARRAY; # forget @ARRAY ever existed
1227 Note that the EXPR can be arbitrarily complicated as long as the final
1228 operation is a hash element, array element, hash slice, or array slice
1231 delete $ref->[$x][$y]{$key};
1232 delete @{$ref->[$x][$y]}{$key1, $key2, @morekeys};
1234 delete $ref->[$x][$y][$index];
1235 delete @{$ref->[$x][$y]}[$index1, $index2, @moreindices];
1237 The C<delete local EXPR> construct can also be used to localize the deletion
1238 of array/hash elements to the current block.
1239 See L<perlsub/"Localized deletion of elements of composite types">.
1242 X<die> X<throw> X<exception> X<raise> X<$@> X<abort>
1244 Outside an C<eval>, prints the value of LIST to C<STDERR> and
1245 exits with the current value of C<$!> (errno). If C<$!> is C<0>,
1246 exits with the value of C<<< ($? >> 8) >>> (backtick `command`
1247 status). If C<<< ($? >> 8) >>> is C<0>, exits with C<255>. Inside
1248 an C<eval(),> the error message is stuffed into C<$@> and the
1249 C<eval> is terminated with the undefined value. This makes
1250 C<die> the way to raise an exception.
1252 Equivalent examples:
1254 die "Can't cd to spool: $!\n" unless chdir '/usr/spool/news';
1255 chdir '/usr/spool/news' or die "Can't cd to spool: $!\n"
1257 If the last element of LIST does not end in a newline, the current
1258 script line number and input line number (if any) are also printed,
1259 and a newline is supplied. Note that the "input line number" (also
1260 known as "chunk") is subject to whatever notion of "line" happens to
1261 be currently in effect, and is also available as the special variable
1262 C<$.>. See L<perlvar/"$/"> and L<perlvar/"$.">.
1264 Hint: sometimes appending C<", stopped"> to your message will cause it
1265 to make better sense when the string C<"at foo line 123"> is appended.
1266 Suppose you are running script "canasta".
1268 die "/etc/games is no good";
1269 die "/etc/games is no good, stopped";
1271 produce, respectively
1273 /etc/games is no good at canasta line 123.
1274 /etc/games is no good, stopped at canasta line 123.
1276 See also exit(), warn(), and the Carp module.
1278 If the output is empty and C<$@> already contains a value (typically from a
1279 previous eval) that value is reused after appending C<"\t...propagated">.
1280 This is useful for propagating exceptions:
1283 die unless $@ =~ /Expected exception/;
1285 If the output is empty and C<$@> contains an object reference that has a
1286 C<PROPAGATE> method, that method will be called with additional file
1287 and line number parameters. The return value replaces the value in
1288 C<$@>. i.e. as if C<< $@ = eval { $@->PROPAGATE(__FILE__, __LINE__) }; >>
1291 If C<$@> is empty then the string C<"Died"> is used.
1293 die() can also be called with a reference argument. If this happens to be
1294 trapped within an eval(), $@ contains the reference. This behavior permits
1295 a more elaborate exception handling implementation using objects that
1296 maintain arbitrary state about the nature of the exception. Such a scheme
1297 is sometimes preferable to matching particular string values of $@ using
1298 regular expressions. Because $@ is a global variable, and eval() may be
1299 used within object implementations, care must be taken that analyzing the
1300 error object doesn't replace the reference in the global variable. The
1301 easiest solution is to make a local copy of the reference before doing
1302 other manipulations. Here's an example:
1304 use Scalar::Util 'blessed';
1306 eval { ... ; die Some::Module::Exception->new( FOO => "bar" ) };
1307 if (my $ev_err = $@) {
1308 if (blessed($ev_err) && $ev_err->isa("Some::Module::Exception")) {
1309 # handle Some::Module::Exception
1312 # handle all other possible exceptions
1316 Because perl will stringify uncaught exception messages before displaying
1317 them, you may want to overload stringification operations on such custom
1318 exception objects. See L<overload> for details about that.
1320 You can arrange for a callback to be run just before the C<die>
1321 does its deed, by setting the C<$SIG{__DIE__}> hook. The associated
1322 handler will be called with the error text and can change the error
1323 message, if it sees fit, by calling C<die> again. See
1324 L<perlvar/$SIG{expr}> for details on setting C<%SIG> entries, and
1325 L<"eval BLOCK"> for some examples. Although this feature was
1326 to be run only right before your program was to exit, this is not
1327 currently the case--the C<$SIG{__DIE__}> hook is currently called
1328 even inside eval()ed blocks/strings! If one wants the hook to do
1329 nothing in such situations, put
1333 as the first line of the handler (see L<perlvar/$^S>). Because
1334 this promotes strange action at a distance, this counterintuitive
1335 behavior may be fixed in a future release.
1340 Not really a function. Returns the value of the last command in the
1341 sequence of commands indicated by BLOCK. When modified by the C<while> or
1342 C<until> loop modifier, executes the BLOCK once before testing the loop
1343 condition. (On other statements the loop modifiers test the conditional
1346 C<do BLOCK> does I<not> count as a loop, so the loop control statements
1347 C<next>, C<last>, or C<redo> cannot be used to leave or restart the block.
1348 See L<perlsyn> for alternative strategies.
1350 =item do SUBROUTINE(LIST)
1353 This form of subroutine call is deprecated. See L<perlsub>.
1358 Uses the value of EXPR as a filename and executes the contents of the
1359 file as a Perl script.
1367 except that it's more efficient and concise, keeps track of the current
1368 filename for error messages, searches the @INC directories, and updates
1369 C<%INC> if the file is found. See L<perlvar/Predefined Names> for these
1370 variables. It also differs in that code evaluated with C<do FILENAME>
1371 cannot see lexicals in the enclosing scope; C<eval STRING> does. It's the
1372 same, however, in that it does reparse the file every time you call it,
1373 so you probably don't want to do this inside a loop.
1375 If C<do> cannot read the file, it returns undef and sets C<$!> to the
1376 error. If C<do> can read the file but cannot compile it, it
1377 returns undef and sets an error message in C<$@>. If the file is
1378 successfully compiled, C<do> returns the value of the last expression
1381 Note that inclusion of library modules is better done with the
1382 C<use> and C<require> operators, which also do automatic error checking
1383 and raise an exception if there's a problem.
1385 You might like to use C<do> to read in a program configuration
1386 file. Manual error checking can be done this way:
1388 # read in config files: system first, then user
1389 for $file ("/share/prog/defaults.rc",
1390 "$ENV{HOME}/.someprogrc")
1392 unless ($return = do $file) {
1393 warn "couldn't parse $file: $@" if $@;
1394 warn "couldn't do $file: $!" unless defined $return;
1395 warn "couldn't run $file" unless $return;
1400 X<dump> X<core> X<undump>
1404 This function causes an immediate core dump. See also the B<-u>
1405 command-line switch in L<perlrun>, which does the same thing.
1406 Primarily this is so that you can use the B<undump> program (not
1407 supplied) to turn your core dump into an executable binary after
1408 having initialized all your variables at the beginning of the
1409 program. When the new binary is executed it will begin by executing
1410 a C<goto LABEL> (with all the restrictions that C<goto> suffers).
1411 Think of it as a goto with an intervening core dump and reincarnation.
1412 If C<LABEL> is omitted, restarts the program from the top.
1414 B<WARNING>: Any files opened at the time of the dump will I<not>
1415 be open any more when the program is reincarnated, with possible
1416 resulting confusion on the part of Perl.
1418 This function is now largely obsolete, mostly because it's very hard to
1419 convert a core file into an executable. That's why you should now invoke
1420 it as C<CORE::dump()>, if you don't want to be warned against a possible
1424 X<each> X<hash, iterator>
1429 When called in list context, returns a 2-element list consisting of the
1430 key and value for the next element of a hash, or the index and value for
1431 the next element of an array, so that you can iterate over it. When called
1432 in scalar context, returns only the key for the next element in the hash
1433 (or the index for an array).
1435 Hash entries are returned in an apparently random order. The actual random
1436 order is subject to change in future versions of perl, but it is
1437 guaranteed to be in the same order as either the C<keys> or C<values>
1438 function would produce on the same (unmodified) hash. Since Perl
1439 5.8.2 the ordering can be different even between different runs of Perl
1440 for security reasons (see L<perlsec/"Algorithmic Complexity Attacks">).
1442 When the hash or array is entirely read, a null array is returned in list
1443 context (which when assigned produces a false (C<0>) value), and C<undef> in
1444 scalar context. The next call to C<each> after that will start iterating
1445 again. There is a single iterator for each hash or array, shared by all
1446 C<each>, C<keys>, and C<values> function calls in the program; it can be
1447 reset by reading all the elements from the hash or array, or by evaluating
1448 C<keys HASH>, C<values HASH>, C<keys ARRAY>, or C<values ARRAY>. If you add
1449 or delete elements of a hash while you're
1450 iterating over it, you may get entries skipped or duplicated, so
1451 don't. Exception: It is always safe to delete the item most recently
1452 returned by C<each()>, which means that the following code will work:
1454 while (($key, $value) = each %hash) {
1456 delete $hash{$key}; # This is safe
1459 The following prints out your environment like the printenv(1) program,
1460 only in a different order:
1462 while (($key,$value) = each %ENV) {
1463 print "$key=$value\n";
1466 See also C<keys>, C<values> and C<sort>.
1468 =item eof FILEHANDLE
1477 Returns 1 if the next read on FILEHANDLE will return end of file, or if
1478 FILEHANDLE is not open. FILEHANDLE may be an expression whose value
1479 gives the real filehandle. (Note that this function actually
1480 reads a character and then C<ungetc>s it, so isn't very useful in an
1481 interactive context.) Do not read from a terminal file (or call
1482 C<eof(FILEHANDLE)> on it) after end-of-file is reached. File types such
1483 as terminals may lose the end-of-file condition if you do.
1485 An C<eof> without an argument uses the last file read. Using C<eof()>
1486 with empty parentheses is very different. It refers to the pseudo file
1487 formed from the files listed on the command line and accessed via the
1488 C<< <> >> operator. Since C<< <> >> isn't explicitly opened,
1489 as a normal filehandle is, an C<eof()> before C<< <> >> has been
1490 used will cause C<@ARGV> to be examined to determine if input is
1491 available. Similarly, an C<eof()> after C<< <> >> has returned
1492 end-of-file will assume you are processing another C<@ARGV> list,
1493 and if you haven't set C<@ARGV>, will read input from C<STDIN>;
1494 see L<perlop/"I/O Operators">.
1496 In a C<< while (<>) >> loop, C<eof> or C<eof(ARGV)> can be used to
1497 detect the end of each file, C<eof()> will only detect the end of the
1498 last file. Examples:
1500 # reset line numbering on each input file
1502 next if /^\s*#/; # skip comments
1505 close ARGV if eof; # Not eof()!
1508 # insert dashes just before last line of last file
1510 if (eof()) { # check for end of last file
1511 print "--------------\n";
1514 last if eof(); # needed if we're reading from a terminal
1517 Practical hint: you almost never need to use C<eof> in Perl, because the
1518 input operators typically return C<undef> when they run out of data, or if
1522 X<eval> X<try> X<catch> X<evaluate> X<parse> X<execute>
1523 X<error, handling> X<exception, handling>
1529 In the first form, the return value of EXPR is parsed and executed as if it
1530 were a little Perl program. The value of the expression (which is itself
1531 determined within scalar context) is first parsed, and if there weren't any
1532 errors, executed in the lexical context of the current Perl program, so
1533 that any variable settings or subroutine and format definitions remain
1534 afterwards. Note that the value is parsed every time the C<eval> executes.
1535 If EXPR is omitted, evaluates C<$_>. This form is typically used to
1536 delay parsing and subsequent execution of the text of EXPR until run time.
1538 In the second form, the code within the BLOCK is parsed only once--at the
1539 same time the code surrounding the C<eval> itself was parsed--and executed
1540 within the context of the current Perl program. This form is typically
1541 used to trap exceptions more efficiently than the first (see below), while
1542 also providing the benefit of checking the code within BLOCK at compile
1545 The final semicolon, if any, may be omitted from the value of EXPR or within
1548 In both forms, the value returned is the value of the last expression
1549 evaluated inside the mini-program; a return statement may be also used, just
1550 as with subroutines. The expression providing the return value is evaluated
1551 in void, scalar, or list context, depending on the context of the C<eval>
1552 itself. See L</wantarray> for more on how the evaluation context can be
1555 If there is a syntax error or runtime error, or a C<die> statement is
1556 executed, C<eval> returns an undefined value in scalar context
1557 or an empty list in list context, and C<$@> is set to the
1558 error message. If there was no error, C<$@> is guaranteed to be a null
1559 string. Beware that using C<eval> neither silences perl from printing
1560 warnings to STDERR, nor does it stuff the text of warning messages into C<$@>.
1561 To do either of those, you have to use the C<$SIG{__WARN__}> facility, or
1562 turn off warnings inside the BLOCK or EXPR using S<C<no warnings 'all'>>.
1563 See L</warn>, L<perlvar>, L<warnings> and L<perllexwarn>.
1565 Note that, because C<eval> traps otherwise-fatal errors, it is useful for
1566 determining whether a particular feature (such as C<socket> or C<symlink>)
1567 is implemented. It is also Perl's exception trapping mechanism, where
1568 the die operator is used to raise exceptions.
1570 If you want to trap errors when loading an XS module, some problems with
1571 the binary interface (such as Perl version skew) may be fatal even with
1572 C<eval> unless C<$ENV{PERL_DL_NONLAZY}> is set. See L<perlrun>.
1574 If the code to be executed doesn't vary, you may use the eval-BLOCK
1575 form to trap run-time errors without incurring the penalty of
1576 recompiling each time. The error, if any, is still returned in C<$@>.
1579 # make divide-by-zero nonfatal
1580 eval { $answer = $a / $b; }; warn $@ if $@;
1582 # same thing, but less efficient
1583 eval '$answer = $a / $b'; warn $@ if $@;
1585 # a compile-time error
1586 eval { $answer = }; # WRONG
1589 eval '$answer ='; # sets $@
1591 Using the C<eval{}> form as an exception trap in libraries does have some
1592 issues. Due to the current arguably broken state of C<__DIE__> hooks, you
1593 may wish not to trigger any C<__DIE__> hooks that user code may have installed.
1594 You can use the C<local $SIG{__DIE__}> construct for this purpose,
1595 as shown in this example:
1597 # a very private exception trap for divide-by-zero
1598 eval { local $SIG{'__DIE__'}; $answer = $a / $b; };
1601 This is especially significant, given that C<__DIE__> hooks can call
1602 C<die> again, which has the effect of changing their error messages:
1604 # __DIE__ hooks may modify error messages
1606 local $SIG{'__DIE__'} =
1607 sub { (my $x = $_[0]) =~ s/foo/bar/g; die $x };
1608 eval { die "foo lives here" };
1609 print $@ if $@; # prints "bar lives here"
1612 Because this promotes action at a distance, this counterintuitive behavior
1613 may be fixed in a future release.
1615 With an C<eval>, you should be especially careful to remember what's
1616 being looked at when:
1622 eval { $x }; # CASE 4
1624 eval "\$$x++"; # CASE 5
1627 Cases 1 and 2 above behave identically: they run the code contained in
1628 the variable $x. (Although case 2 has misleading double quotes making
1629 the reader wonder what else might be happening (nothing is).) Cases 3
1630 and 4 likewise behave in the same way: they run the code C<'$x'>, which
1631 does nothing but return the value of $x. (Case 4 is preferred for
1632 purely visual reasons, but it also has the advantage of compiling at
1633 compile-time instead of at run-time.) Case 5 is a place where
1634 normally you I<would> like to use double quotes, except that in this
1635 particular situation, you can just use symbolic references instead, as
1638 The assignment to C<$@> occurs before restoration of localised variables,
1639 which means a temporary is required if you want to mask some but not all
1642 # alter $@ on nefarious repugnancy only
1646 local $@; # protect existing $@
1647 eval { test_repugnancy() };
1648 # $@ =~ /nefarious/ and die $@; # DOES NOT WORK
1649 $@ =~ /nefarious/ and $e = $@;
1651 die $e if defined $e
1654 C<eval BLOCK> does I<not> count as a loop, so the loop control statements
1655 C<next>, C<last>, or C<redo> cannot be used to leave or restart the block.
1657 Note that as a very special case, an C<eval ''> executed within the C<DB>
1658 package doesn't see the usual surrounding lexical scope, but rather the
1659 scope of the first non-DB piece of code that called it. You don't normally
1660 need to worry about this unless you are writing a Perl debugger.
1665 =item exec PROGRAM LIST
1667 The C<exec> function executes a system command I<and never returns>--
1668 use C<system> instead of C<exec> if you want it to return. It fails and
1669 returns false only if the command does not exist I<and> it is executed
1670 directly instead of via your system's command shell (see below).
1672 Since it's a common mistake to use C<exec> instead of C<system>, Perl
1673 warns you if there is a following statement which isn't C<die>, C<warn>,
1674 or C<exit> (if C<-w> is set - but you always do that). If you
1675 I<really> want to follow an C<exec> with some other statement, you
1676 can use one of these styles to avoid the warning:
1678 exec ('foo') or print STDERR "couldn't exec foo: $!";
1679 { exec ('foo') }; print STDERR "couldn't exec foo: $!";
1681 If there is more than one argument in LIST, or if LIST is an array
1682 with more than one value, calls execvp(3) with the arguments in LIST.
1683 If there is only one scalar argument or an array with one element in it,
1684 the argument is checked for shell metacharacters, and if there are any,
1685 the entire argument is passed to the system's command shell for parsing
1686 (this is C</bin/sh -c> on Unix platforms, but varies on other platforms).
1687 If there are no shell metacharacters in the argument, it is split into
1688 words and passed directly to C<execvp>, which is more efficient.
1691 exec '/bin/echo', 'Your arguments are: ', @ARGV;
1692 exec "sort $outfile | uniq";
1694 If you don't really want to execute the first argument, but want to lie
1695 to the program you are executing about its own name, you can specify
1696 the program you actually want to run as an "indirect object" (without a
1697 comma) in front of the LIST. (This always forces interpretation of the
1698 LIST as a multivalued list, even if there is only a single scalar in
1701 $shell = '/bin/csh';
1702 exec $shell '-sh'; # pretend it's a login shell
1706 exec {'/bin/csh'} '-sh'; # pretend it's a login shell
1708 When the arguments get executed via the system shell, results will
1709 be subject to its quirks and capabilities. See L<perlop/"`STRING`">
1712 Using an indirect object with C<exec> or C<system> is also more
1713 secure. This usage (which also works fine with system()) forces
1714 interpretation of the arguments as a multivalued list, even if the
1715 list had just one argument. That way you're safe from the shell
1716 expanding wildcards or splitting up words with whitespace in them.
1718 @args = ( "echo surprise" );
1720 exec @args; # subject to shell escapes
1722 exec { $args[0] } @args; # safe even with one-arg list
1724 The first version, the one without the indirect object, ran the I<echo>
1725 program, passing it C<"surprise"> an argument. The second version
1726 didn't--it tried to run a program literally called I<"echo surprise">,
1727 didn't find it, and set C<$?> to a non-zero value indicating failure.
1729 Beginning with v5.6.0, Perl will attempt to flush all files opened for
1730 output before the exec, but this may not be supported on some platforms
1731 (see L<perlport>). To be safe, you may need to set C<$|> ($AUTOFLUSH
1732 in English) or call the C<autoflush()> method of C<IO::Handle> on any
1733 open handles in order to avoid lost output.
1735 Note that C<exec> will not call your C<END> blocks, nor will it call
1736 any C<DESTROY> methods in your objects.
1739 X<exists> X<autovivification>
1741 Given an expression that specifies a hash element or array element,
1742 returns true if the specified element in the hash or array has ever
1743 been initialized, even if the corresponding value is undefined.
1745 print "Exists\n" if exists $hash{$key};
1746 print "Defined\n" if defined $hash{$key};
1747 print "True\n" if $hash{$key};
1749 print "Exists\n" if exists $array[$index];
1750 print "Defined\n" if defined $array[$index];
1751 print "True\n" if $array[$index];
1753 A hash or array element can be true only if it's defined, and defined if
1754 it exists, but the reverse doesn't necessarily hold true.
1756 Given an expression that specifies the name of a subroutine,
1757 returns true if the specified subroutine has ever been declared, even
1758 if it is undefined. Mentioning a subroutine name for exists or defined
1759 does not count as declaring it. Note that a subroutine which does not
1760 exist may still be callable: its package may have an C<AUTOLOAD>
1761 method that makes it spring into existence the first time that it is
1762 called -- see L<perlsub>.
1764 print "Exists\n" if exists &subroutine;
1765 print "Defined\n" if defined &subroutine;
1767 Note that the EXPR can be arbitrarily complicated as long as the final
1768 operation is a hash or array key lookup or subroutine name:
1770 if (exists $ref->{A}->{B}->{$key}) { }
1771 if (exists $hash{A}{B}{$key}) { }
1773 if (exists $ref->{A}->{B}->[$ix]) { }
1774 if (exists $hash{A}{B}[$ix]) { }
1776 if (exists &{$ref->{A}{B}{$key}}) { }
1778 Although the deepest nested array or hash will not spring into existence
1779 just because its existence was tested, any intervening ones will.
1780 Thus C<< $ref->{"A"} >> and C<< $ref->{"A"}->{"B"} >> will spring
1781 into existence due to the existence test for the $key element above.
1782 This happens anywhere the arrow operator is used, including even:
1785 if (exists $ref->{"Some key"}) { }
1786 print $ref; # prints HASH(0x80d3d5c)
1788 This surprising autovivification in what does not at first--or even
1789 second--glance appear to be an lvalue context may be fixed in a future
1792 Use of a subroutine call, rather than a subroutine name, as an argument
1793 to exists() is an error.
1796 exists &sub(); # Error
1799 X<exit> X<terminate> X<abort>
1803 Evaluates EXPR and exits immediately with that value. Example:
1806 exit 0 if $ans =~ /^[Xx]/;
1808 See also C<die>. If EXPR is omitted, exits with C<0> status. The only
1809 universally recognized values for EXPR are C<0> for success and C<1>
1810 for error; other values are subject to interpretation depending on the
1811 environment in which the Perl program is running. For example, exiting
1812 69 (EX_UNAVAILABLE) from a I<sendmail> incoming-mail filter will cause
1813 the mailer to return the item undelivered, but that's not true everywhere.
1815 Don't use C<exit> to abort a subroutine if there's any chance that
1816 someone might want to trap whatever error happened. Use C<die> instead,
1817 which can be trapped by an C<eval>.
1819 The exit() function does not always exit immediately. It calls any
1820 defined C<END> routines first, but these C<END> routines may not
1821 themselves abort the exit. Likewise any object destructors that need to
1822 be called are called before the real exit. If this is a problem, you
1823 can call C<POSIX:_exit($status)> to avoid END and destructor processing.
1824 See L<perlmod> for details.
1827 X<exp> X<exponential> X<antilog> X<antilogarithm> X<e>
1831 Returns I<e> (the natural logarithm base) to the power of EXPR.
1832 If EXPR is omitted, gives C<exp($_)>.
1834 =item fcntl FILEHANDLE,FUNCTION,SCALAR
1837 Implements the fcntl(2) function. You'll probably have to say
1841 first to get the correct constant definitions. Argument processing and
1842 value return works just like C<ioctl> below.
1846 fcntl($filehandle, F_GETFL, $packed_return_buffer)
1847 or die "can't fcntl F_GETFL: $!";
1849 You don't have to check for C<defined> on the return from C<fcntl>.
1850 Like C<ioctl>, it maps a C<0> return from the system call into
1851 C<"0 but true"> in Perl. This string is true in boolean context and C<0>
1852 in numeric context. It is also exempt from the normal B<-w> warnings
1853 on improper numeric conversions.
1855 Note that C<fcntl> will produce a fatal error if used on a machine that
1856 doesn't implement fcntl(2). See the Fcntl module or your fcntl(2)
1857 manpage to learn what functions are available on your system.
1859 Here's an example of setting a filehandle named C<REMOTE> to be
1860 non-blocking at the system level. You'll have to negotiate C<$|>
1861 on your own, though.
1863 use Fcntl qw(F_GETFL F_SETFL O_NONBLOCK);
1865 $flags = fcntl(REMOTE, F_GETFL, 0)
1866 or die "Can't get flags for the socket: $!\n";
1868 $flags = fcntl(REMOTE, F_SETFL, $flags | O_NONBLOCK)
1869 or die "Can't set flags for the socket: $!\n";
1871 =item fileno FILEHANDLE
1874 Returns the file descriptor for a filehandle, or undefined if the
1875 filehandle is not open. This is mainly useful for constructing
1876 bitmaps for C<select> and low-level POSIX tty-handling operations.
1877 If FILEHANDLE is an expression, the value is taken as an indirect
1878 filehandle, generally its name.
1880 You can use this to find out whether two handles refer to the
1881 same underlying descriptor:
1883 if (fileno(THIS) == fileno(THAT)) {
1884 print "THIS and THAT are dups\n";
1887 (Filehandles connected to memory objects via new features of C<open> may
1888 return undefined even though they are open.)
1891 =item flock FILEHANDLE,OPERATION
1892 X<flock> X<lock> X<locking>
1894 Calls flock(2), or an emulation of it, on FILEHANDLE. Returns true
1895 for success, false on failure. Produces a fatal error if used on a
1896 machine that doesn't implement flock(2), fcntl(2) locking, or lockf(3).
1897 C<flock> is Perl's portable file locking interface, although it locks
1898 only entire files, not records.
1900 Two potentially non-obvious but traditional C<flock> semantics are
1901 that it waits indefinitely until the lock is granted, and that its locks
1902 B<merely advisory>. Such discretionary locks are more flexible, but offer
1903 fewer guarantees. This means that programs that do not also use C<flock>
1904 may modify files locked with C<flock>. See L<perlport>,
1905 your port's specific documentation, or your system-specific local manpages
1906 for details. It's best to assume traditional behavior if you're writing
1907 portable programs. (But if you're not, you should as always feel perfectly
1908 free to write for your own system's idiosyncrasies (sometimes called
1909 "features"). Slavish adherence to portability concerns shouldn't get
1910 in the way of your getting your job done.)
1912 OPERATION is one of LOCK_SH, LOCK_EX, or LOCK_UN, possibly combined with
1913 LOCK_NB. These constants are traditionally valued 1, 2, 8 and 4, but
1914 you can use the symbolic names if you import them from the Fcntl module,
1915 either individually, or as a group using the ':flock' tag. LOCK_SH
1916 requests a shared lock, LOCK_EX requests an exclusive lock, and LOCK_UN
1917 releases a previously requested lock. If LOCK_NB is bitwise-or'ed with
1918 LOCK_SH or LOCK_EX then C<flock> will return immediately rather than blocking
1919 waiting for the lock (check the return status to see if you got it).
1921 To avoid the possibility of miscoordination, Perl now flushes FILEHANDLE
1922 before locking or unlocking it.
1924 Note that the emulation built with lockf(3) doesn't provide shared
1925 locks, and it requires that FILEHANDLE be open with write intent. These
1926 are the semantics that lockf(3) implements. Most if not all systems
1927 implement lockf(3) in terms of fcntl(2) locking, though, so the
1928 differing semantics shouldn't bite too many people.
1930 Note that the fcntl(2) emulation of flock(3) requires that FILEHANDLE
1931 be open with read intent to use LOCK_SH and requires that it be open
1932 with write intent to use LOCK_EX.
1934 Note also that some versions of C<flock> cannot lock things over the
1935 network; you would need to use the more system-specific C<fcntl> for
1936 that. If you like you can force Perl to ignore your system's flock(2)
1937 function, and so provide its own fcntl(2)-based emulation, by passing
1938 the switch C<-Ud_flock> to the F<Configure> program when you configure
1941 Here's a mailbox appender for BSD systems.
1943 use Fcntl qw(:flock SEEK_END); # import LOCK_* and SEEK_END constants
1947 flock($fh, LOCK_EX) or die "Cannot lock mailbox - $!\n";
1949 # and, in case someone appended while we were waiting...
1950 seek($fh, 0, SEEK_END) or die "Cannot seek - $!\n";
1955 flock($fh, LOCK_UN) or die "Cannot unlock mailbox - $!\n";
1958 open(my $mbox, ">>", "/usr/spool/mail/$ENV{'USER'}")
1959 or die "Can't open mailbox: $!";
1962 print $mbox $msg,"\n\n";
1965 On systems that support a real flock(), locks are inherited across fork()
1966 calls, whereas those that must resort to the more capricious fcntl()
1967 function lose the locks, making it harder to write servers.
1969 See also L<DB_File> for other flock() examples.
1972 X<fork> X<child> X<parent>
1974 Does a fork(2) system call to create a new process running the
1975 same program at the same point. It returns the child pid to the
1976 parent process, C<0> to the child process, or C<undef> if the fork is
1977 unsuccessful. File descriptors (and sometimes locks on those descriptors)
1978 are shared, while everything else is copied. On most systems supporting
1979 fork(), great care has gone into making it extremely efficient (for
1980 example, using copy-on-write technology on data pages), making it the
1981 dominant paradigm for multitasking over the last few decades.
1983 Beginning with v5.6.0, Perl will attempt to flush all files opened for
1984 output before forking the child process, but this may not be supported
1985 on some platforms (see L<perlport>). To be safe, you may need to set
1986 C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method of
1987 C<IO::Handle> on any open handles in order to avoid duplicate output.
1989 If you C<fork> without ever waiting on your children, you will
1990 accumulate zombies. On some systems, you can avoid this by setting
1991 C<$SIG{CHLD}> to C<"IGNORE">. See also L<perlipc> for more examples of
1992 forking and reaping moribund children.
1994 Note that if your forked child inherits system file descriptors like
1995 STDIN and STDOUT that are actually connected by a pipe or socket, even
1996 if you exit, then the remote server (such as, say, a CGI script or a
1997 backgrounded job launched from a remote shell) won't think you're done.
1998 You should reopen those to F</dev/null> if it's any issue.
2003 Declare a picture format for use by the C<write> function. For
2007 Test: @<<<<<<<< @||||| @>>>>>
2008 $str, $%, '$' . int($num)
2012 $num = $cost/$quantity;
2016 See L<perlform> for many details and examples.
2018 =item formline PICTURE,LIST
2021 This is an internal function used by C<format>s, though you may call it,
2022 too. It formats (see L<perlform>) a list of values according to the
2023 contents of PICTURE, placing the output into the format output
2024 accumulator, C<$^A> (or C<$ACCUMULATOR> in English).
2025 Eventually, when a C<write> is done, the contents of
2026 C<$^A> are written to some filehandle. You could also read C<$^A>
2027 and then set C<$^A> back to C<"">. Note that a format typically
2028 does one C<formline> per line of form, but the C<formline> function itself
2029 doesn't care how many newlines are embedded in the PICTURE. This means
2030 that the C<~> and C<~~> tokens will treat the entire PICTURE as a single line.
2031 You may therefore need to use multiple formlines to implement a single
2032 record format, just like the format compiler.
2034 Be careful if you put double quotes around the picture, because an C<@>
2035 character may be taken to mean the beginning of an array name.
2036 C<formline> always returns true. See L<perlform> for other examples.
2038 =item getc FILEHANDLE
2039 X<getc> X<getchar> X<character> X<file, read>
2043 Returns the next character from the input file attached to FILEHANDLE,
2044 or the undefined value at end of file, or if there was an error (in
2045 the latter case C<$!> is set). If FILEHANDLE is omitted, reads from
2046 STDIN. This is not particularly efficient. However, it cannot be
2047 used by itself to fetch single characters without waiting for the user
2048 to hit enter. For that, try something more like:
2051 system "stty cbreak </dev/tty >/dev/tty 2>&1";
2054 system "stty", '-icanon', 'eol', "\001";
2060 system "stty -cbreak </dev/tty >/dev/tty 2>&1";
2063 system "stty", 'icanon', 'eol', '^@'; # ASCII null
2067 Determination of whether $BSD_STYLE should be set
2068 is left as an exercise to the reader.
2070 The C<POSIX::getattr> function can do this more portably on
2071 systems purporting POSIX compliance. See also the C<Term::ReadKey>
2072 module from your nearest CPAN site; details on CPAN can be found on
2076 X<getlogin> X<login>
2078 This implements the C library function of the same name, which on most
2079 systems returns the current login from F</etc/utmp>, if any. If null,
2082 $login = getlogin || getpwuid($<) || "Kilroy";
2084 Do not consider C<getlogin> for authentication: it is not as
2085 secure as C<getpwuid>.
2087 =item getpeername SOCKET
2088 X<getpeername> X<peer>
2090 Returns the packed sockaddr address of other end of the SOCKET connection.
2093 $hersockaddr = getpeername(SOCK);
2094 ($port, $iaddr) = sockaddr_in($hersockaddr);
2095 $herhostname = gethostbyaddr($iaddr, AF_INET);
2096 $herstraddr = inet_ntoa($iaddr);
2101 Returns the current process group for the specified PID. Use
2102 a PID of C<0> to get the current process group for the
2103 current process. Will raise an exception if used on a machine that
2104 doesn't implement getpgrp(2). If PID is omitted, returns process
2105 group of current process. Note that the POSIX version of C<getpgrp>
2106 does not accept a PID argument, so only C<PID==0> is truly portable.
2109 X<getppid> X<parent> X<pid>
2111 Returns the process id of the parent process.
2113 Note for Linux users: on Linux, the C functions C<getpid()> and
2114 C<getppid()> return different values from different threads. In order to
2115 be portable, this behavior is not reflected by the perl-level function
2116 C<getppid()>, that returns a consistent value across threads. If you want
2117 to call the underlying C<getppid()>, you may use the CPAN module
2120 =item getpriority WHICH,WHO
2121 X<getpriority> X<priority> X<nice>
2123 Returns the current priority for a process, a process group, or a user.
2124 (See C<getpriority(2)>.) Will raise a fatal exception if used on a
2125 machine that doesn't implement getpriority(2).
2128 X<getpwnam> X<getgrnam> X<gethostbyname> X<getnetbyname> X<getprotobyname>
2129 X<getpwuid> X<getgrgid> X<getservbyname> X<gethostbyaddr> X<getnetbyaddr>
2130 X<getprotobynumber> X<getservbyport> X<getpwent> X<getgrent> X<gethostent>
2131 X<getnetent> X<getprotoent> X<getservent> X<setpwent> X<setgrent> X<sethostent>
2132 X<setnetent> X<setprotoent> X<setservent> X<endpwent> X<endgrent> X<endhostent>
2133 X<endnetent> X<endprotoent> X<endservent>
2137 =item gethostbyname NAME
2139 =item getnetbyname NAME
2141 =item getprotobyname NAME
2147 =item getservbyname NAME,PROTO
2149 =item gethostbyaddr ADDR,ADDRTYPE
2151 =item getnetbyaddr ADDR,ADDRTYPE
2153 =item getprotobynumber NUMBER
2155 =item getservbyport PORT,PROTO
2173 =item sethostent STAYOPEN
2175 =item setnetent STAYOPEN
2177 =item setprotoent STAYOPEN
2179 =item setservent STAYOPEN
2193 These routines perform the same functions as their counterparts in the
2194 system library. In list context, the return values from the
2195 various get routines are as follows:
2197 ($name,$passwd,$uid,$gid,
2198 $quota,$comment,$gcos,$dir,$shell,$expire) = getpw*
2199 ($name,$passwd,$gid,$members) = getgr*
2200 ($name,$aliases,$addrtype,$length,@addrs) = gethost*
2201 ($name,$aliases,$addrtype,$net) = getnet*
2202 ($name,$aliases,$proto) = getproto*
2203 ($name,$aliases,$port,$proto) = getserv*
2205 (If the entry doesn't exist you get a null list.)
2207 The exact meaning of the $gcos field varies but it usually contains
2208 the real name of the user (as opposed to the login name) and other
2209 information pertaining to the user. Beware, however, that in many
2210 system users are able to change this information and therefore it
2211 cannot be trusted and therefore the $gcos is tainted (see
2212 L<perlsec>). The $passwd and $shell, user's encrypted password and
2213 login shell, are also tainted, because of the same reason.
2215 In scalar context, you get the name, unless the function was a
2216 lookup by name, in which case you get the other thing, whatever it is.
2217 (If the entry doesn't exist you get the undefined value.) For example:
2219 $uid = getpwnam($name);
2220 $name = getpwuid($num);
2222 $gid = getgrnam($name);
2223 $name = getgrgid($num);
2227 In I<getpw*()> the fields $quota, $comment, and $expire are special
2228 cases in the sense that in many systems they are unsupported. If the
2229 $quota is unsupported, it is an empty scalar. If it is supported, it
2230 usually encodes the disk quota. If the $comment field is unsupported,
2231 it is an empty scalar. If it is supported it usually encodes some
2232 administrative comment about the user. In some systems the $quota
2233 field may be $change or $age, fields that have to do with password
2234 aging. In some systems the $comment field may be $class. The $expire
2235 field, if present, encodes the expiration period of the account or the
2236 password. For the availability and the exact meaning of these fields
2237 in your system, please consult your getpwnam(3) documentation and your
2238 F<pwd.h> file. You can also find out from within Perl what your
2239 $quota and $comment fields mean and whether you have the $expire field
2240 by using the C<Config> module and the values C<d_pwquota>, C<d_pwage>,
2241 C<d_pwchange>, C<d_pwcomment>, and C<d_pwexpire>. Shadow password
2242 files are only supported if your vendor has implemented them in the
2243 intuitive fashion that calling the regular C library routines gets the
2244 shadow versions if you're running under privilege or if there exists
2245 the shadow(3) functions as found in System V (this includes Solaris
2246 and Linux.) Those systems that implement a proprietary shadow password
2247 facility are unlikely to be supported.
2249 The $members value returned by I<getgr*()> is a space separated list of
2250 the login names of the members of the group.
2252 For the I<gethost*()> functions, if the C<h_errno> variable is supported in
2253 C, it will be returned to you via C<$?> if the function call fails. The
2254 C<@addrs> value returned by a successful call is a list of the raw
2255 addresses returned by the corresponding system library call. In the
2256 Internet domain, each address is four bytes long and you can unpack it
2257 by saying something like:
2259 ($a,$b,$c,$d) = unpack('W4',$addr[0]);
2261 The Socket library makes this slightly easier:
2264 $iaddr = inet_aton("127.1"); # or whatever address
2265 $name = gethostbyaddr($iaddr, AF_INET);
2267 # or going the other way
2268 $straddr = inet_ntoa($iaddr);
2270 In the opposite way, to resolve a hostname to the IP address
2274 $packed_ip = gethostbyname("www.perl.org");
2275 if (defined $packed_ip) {
2276 $ip_address = inet_ntoa($packed_ip);
2279 Make sure <gethostbyname()> is called in SCALAR context and that
2280 its return value is checked for definedness.
2282 If you get tired of remembering which element of the return list
2283 contains which return value, by-name interfaces are provided
2284 in standard modules: C<File::stat>, C<Net::hostent>, C<Net::netent>,
2285 C<Net::protoent>, C<Net::servent>, C<Time::gmtime>, C<Time::localtime>,
2286 and C<User::grent>. These override the normal built-ins, supplying
2287 versions that return objects with the appropriate names
2288 for each field. For example:
2292 $is_his = (stat($filename)->uid == pwent($whoever)->uid);
2294 Even though it looks like they're the same method calls (uid),
2295 they aren't, because a C<File::stat> object is different from
2296 a C<User::pwent> object.
2298 =item getsockname SOCKET
2301 Returns the packed sockaddr address of this end of the SOCKET connection,
2302 in case you don't know the address because you have several different
2303 IPs that the connection might have come in on.
2306 $mysockaddr = getsockname(SOCK);
2307 ($port, $myaddr) = sockaddr_in($mysockaddr);
2308 printf "Connect to %s [%s]\n",
2309 scalar gethostbyaddr($myaddr, AF_INET),
2312 =item getsockopt SOCKET,LEVEL,OPTNAME
2315 Queries the option named OPTNAME associated with SOCKET at a given LEVEL.
2316 Options may exist at multiple protocol levels depending on the socket
2317 type, but at least the uppermost socket level SOL_SOCKET (defined in the
2318 C<Socket> module) will exist. To query options at another level the
2319 protocol number of the appropriate protocol controlling the option
2320 should be supplied. For example, to indicate that an option is to be
2321 interpreted by the TCP protocol, LEVEL should be set to the protocol
2322 number of TCP, which you can get using getprotobyname.
2324 The call returns a packed string representing the requested socket option,
2325 or C<undef> if there is an error (the error reason will be in $!). What
2326 exactly is in the packed string depends in the LEVEL and OPTNAME, consult
2327 your system documentation for details. A very common case however is that
2328 the option is an integer, in which case the result will be a packed
2329 integer which you can decode using unpack with the C<i> (or C<I>) format.
2331 An example testing if Nagle's algorithm is turned on on a socket:
2333 use Socket qw(:all);
2335 defined(my $tcp = getprotobyname("tcp"))
2336 or die "Could not determine the protocol number for tcp";
2337 # my $tcp = IPPROTO_TCP; # Alternative
2338 my $packed = getsockopt($socket, $tcp, TCP_NODELAY)
2339 or die "Could not query TCP_NODELAY socket option: $!";
2340 my $nodelay = unpack("I", $packed);
2341 print "Nagle's algorithm is turned ", $nodelay ? "off\n" : "on\n";
2345 X<glob> X<wildcard> X<filename, expansion> X<expand>
2349 In list context, returns a (possibly empty) list of filename expansions on
2350 the value of EXPR such as the standard Unix shell F</bin/csh> would do. In
2351 scalar context, glob iterates through such filename expansions, returning
2352 undef when the list is exhausted. This is the internal function
2353 implementing the C<< <*.c> >> operator, but you can use it directly. If
2354 EXPR is omitted, C<$_> is used. The C<< <*.c> >> operator is discussed in
2355 more detail in L<perlop/"I/O Operators">.
2357 Note that C<glob> will split its arguments on whitespace, treating
2358 each segment as separate pattern. As such, C<glob('*.c *.h')> would
2359 match all files with a F<.c> or F<.h> extension. The expression
2360 C<glob('.* *')> would match all files in the current working directory.
2362 Beginning with v5.6.0, this operator is implemented using the standard
2363 C<File::Glob> extension. See L<File::Glob> for details, including
2364 C<bsd_glob> which does not treat whitespace as a pattern separator.
2367 X<gmtime> X<UTC> X<Greenwich>
2371 Works just like L<localtime> but the returned values are
2372 localized for the standard Greenwich time zone.
2374 Note: when called in list context, $isdst, the last value
2375 returned by gmtime is always C<0>. There is no
2376 Daylight Saving Time in GMT.
2378 See L<perlport/gmtime> for portability concerns.
2381 X<goto> X<jump> X<jmp>
2387 The C<goto-LABEL> form finds the statement labeled with LABEL and resumes
2388 execution there. It may not be used to go into any construct that
2389 requires initialization, such as a subroutine or a C<foreach> loop. It
2390 also can't be used to go into a construct that is optimized away,
2391 or to get out of a block or subroutine given to C<sort>.
2392 It can be used to go almost anywhere else within the dynamic scope,
2393 including out of subroutines, but it's usually better to use some other
2394 construct such as C<last> or C<die>. The author of Perl has never felt the
2395 need to use this form of C<goto> (in Perl, that is--C is another matter).
2396 (The difference being that C does not offer named loops combined with
2397 loop control. Perl does, and this replaces most structured uses of C<goto>
2398 in other languages.)
2400 The C<goto-EXPR> form expects a label name, whose scope will be resolved
2401 dynamically. This allows for computed C<goto>s per FORTRAN, but isn't
2402 necessarily recommended if you're optimizing for maintainability:
2404 goto ("FOO", "BAR", "GLARCH")[$i];
2406 The C<goto-&NAME> form is quite different from the other forms of
2407 C<goto>. In fact, it isn't a goto in the normal sense at all, and
2408 doesn't have the stigma associated with other gotos. Instead, it
2409 exits the current subroutine (losing any changes set by local()) and
2410 immediately calls in its place the named subroutine using the current
2411 value of @_. This is used by C<AUTOLOAD> subroutines that wish to
2412 load another subroutine and then pretend that the other subroutine had
2413 been called in the first place (except that any modifications to C<@_>
2414 in the current subroutine are propagated to the other subroutine.)
2415 After the C<goto>, not even C<caller> will be able to tell that this
2416 routine was called first.
2418 NAME needn't be the name of a subroutine; it can be a scalar variable
2419 containing a code reference, or a block that evaluates to a code
2422 =item grep BLOCK LIST
2425 =item grep EXPR,LIST
2427 This is similar in spirit to, but not the same as, grep(1) and its
2428 relatives. In particular, it is not limited to using regular expressions.
2430 Evaluates the BLOCK or EXPR for each element of LIST (locally setting
2431 C<$_> to each element) and returns the list value consisting of those
2432 elements for which the expression evaluated to true. In scalar
2433 context, returns the number of times the expression was true.
2435 @foo = grep(!/^#/, @bar); # weed out comments
2439 @foo = grep {!/^#/} @bar; # weed out comments
2441 Note that C<$_> is an alias to the list value, so it can be used to
2442 modify the elements of the LIST. While this is useful and supported,
2443 it can cause bizarre results if the elements of LIST are not variables.
2444 Similarly, grep returns aliases into the original list, much as a for
2445 loop's index variable aliases the list elements. That is, modifying an
2446 element of a list returned by grep (for example, in a C<foreach>, C<map>
2447 or another C<grep>) actually modifies the element in the original list.
2448 This is usually something to be avoided when writing clear code.
2450 If C<$_> is lexical in the scope where the C<grep> appears (because it has
2451 been declared with C<my $_>) then, in addition to being locally aliased to
2452 the list elements, C<$_> keeps being lexical inside the block; i.e. it
2453 can't be seen from the outside, avoiding any potential side-effects.
2455 See also L</map> for a list composed of the results of the BLOCK or EXPR.
2458 X<hex> X<hexadecimal>
2462 Interprets EXPR as a hex string and returns the corresponding value.
2463 (To convert strings that might start with either C<0>, C<0x>, or C<0b>, see
2464 L</oct>.) If EXPR is omitted, uses C<$_>.
2466 print hex '0xAf'; # prints '175'
2467 print hex 'aF'; # same
2469 Hex strings may only represent integers. Strings that would cause
2470 integer overflow trigger a warning. Leading whitespace is not stripped,
2471 unlike oct(). To present something as hex, look into L</printf>,
2472 L</sprintf>, or L</unpack>.
2477 There is no builtin C<import> function. It is just an ordinary
2478 method (subroutine) defined (or inherited) by modules that wish to export
2479 names to another module. The C<use> function calls the C<import> method
2480 for the package used. See also L</use>, L<perlmod>, and L<Exporter>.
2482 =item index STR,SUBSTR,POSITION
2483 X<index> X<indexOf> X<InStr>
2485 =item index STR,SUBSTR
2487 The index function searches for one string within another, but without
2488 the wildcard-like behavior of a full regular-expression pattern match.
2489 It returns the position of the first occurrence of SUBSTR in STR at
2490 or after POSITION. If POSITION is omitted, starts searching from the
2491 beginning of the string. POSITION before the beginning of the string
2492 or after its end is treated as if it were the beginning or the end,
2493 respectively. POSITION and the return value are based at C<0> (or whatever
2494 you've set the C<$[> variable to--but don't do that). If the substring
2495 is not found, C<index> returns one less than the base, ordinarily C<-1>.
2498 X<int> X<integer> X<truncate> X<trunc> X<floor>
2502 Returns the integer portion of EXPR. If EXPR is omitted, uses C<$_>.
2503 You should not use this function for rounding: one because it truncates
2504 towards C<0>, and two because machine representations of floating point
2505 numbers can sometimes produce counterintuitive results. For example,
2506 C<int(-6.725/0.025)> produces -268 rather than the correct -269; that's
2507 because it's really more like -268.99999999999994315658 instead. Usually,
2508 the C<sprintf>, C<printf>, or the C<POSIX::floor> and C<POSIX::ceil>
2509 functions will serve you better than will int().
2511 =item ioctl FILEHANDLE,FUNCTION,SCALAR
2514 Implements the ioctl(2) function. You'll probably first have to say
2516 require "sys/ioctl.ph"; # probably in $Config{archlib}/sys/ioctl.ph
2518 to get the correct function definitions. If F<sys/ioctl.ph> doesn't
2519 exist or doesn't have the correct definitions you'll have to roll your
2520 own, based on your C header files such as F<< <sys/ioctl.h> >>.
2521 (There is a Perl script called B<h2ph> that comes with the Perl kit that
2522 may help you in this, but it's nontrivial.) SCALAR will be read and/or
2523 written depending on the FUNCTION--a pointer to the string value of SCALAR
2524 will be passed as the third argument of the actual C<ioctl> call. (If SCALAR
2525 has no string value but does have a numeric value, that value will be
2526 passed rather than a pointer to the string value. To guarantee this to be
2527 true, add a C<0> to the scalar before using it.) The C<pack> and C<unpack>
2528 functions may be needed to manipulate the values of structures used by
2531 The return value of C<ioctl> (and C<fcntl>) is as follows:
2533 if OS returns: then Perl returns:
2535 0 string "0 but true"
2536 anything else that number
2538 Thus Perl returns true on success and false on failure, yet you can
2539 still easily determine the actual value returned by the operating
2542 $retval = ioctl(...) || -1;
2543 printf "System returned %d\n", $retval;
2545 The special string C<"0 but true"> is exempt from B<-w> complaints
2546 about improper numeric conversions.
2548 =item join EXPR,LIST
2551 Joins the separate strings of LIST into a single string with fields
2552 separated by the value of EXPR, and returns that new string. Example:
2554 $rec = join(':', $login,$passwd,$uid,$gid,$gcos,$home,$shell);
2556 Beware that unlike C<split>, C<join> doesn't take a pattern as its
2557 first argument. Compare L</split>.
2564 Returns a list consisting of all the keys of the named hash, or the indices
2565 of an array. (In scalar context, returns the number of keys or indices.)
2567 The keys of a hash are returned in an apparently random order. The actual
2568 random order is subject to change in future versions of perl, but it
2569 is guaranteed to be the same order as either the C<values> or C<each>
2570 function produces (given that the hash has not been modified). Since
2571 Perl 5.8.1 the ordering is different even between different runs of
2572 Perl for security reasons (see L<perlsec/"Algorithmic Complexity
2575 As a side effect, calling keys() resets the HASH or ARRAY's internal iterator
2576 (see L</each>). In particular, calling keys() in void context resets
2577 the iterator with no other overhead.
2579 Here is yet another way to print your environment:
2582 @values = values %ENV;
2584 print pop(@keys), '=', pop(@values), "\n";
2587 or how about sorted by key:
2589 foreach $key (sort(keys %ENV)) {
2590 print $key, '=', $ENV{$key}, "\n";
2593 The returned values are copies of the original keys in the hash, so
2594 modifying them will not affect the original hash. Compare L</values>.
2596 To sort a hash by value, you'll need to use a C<sort> function.
2597 Here's a descending numeric sort of a hash by its values:
2599 foreach $key (sort { $hash{$b} <=> $hash{$a} } keys %hash) {
2600 printf "%4d %s\n", $hash{$key}, $key;
2603 As an lvalue C<keys> allows you to increase the number of hash buckets
2604 allocated for the given hash. This can gain you a measure of efficiency if
2605 you know the hash is going to get big. (This is similar to pre-extending
2606 an array by assigning a larger number to $#array.) If you say
2610 then C<%hash> will have at least 200 buckets allocated for it--256 of them,
2611 in fact, since it rounds up to the next power of two. These
2612 buckets will be retained even if you do C<%hash = ()>, use C<undef
2613 %hash> if you want to free the storage while C<%hash> is still in scope.
2614 You can't shrink the number of buckets allocated for the hash using
2615 C<keys> in this way (but you needn't worry about doing this by accident,
2616 as trying has no effect). C<keys @array> in an lvalue context is a syntax
2619 See also C<each>, C<values> and C<sort>.
2621 =item kill SIGNAL, LIST
2624 Sends a signal to a list of processes. Returns the number of
2625 processes successfully signaled (which is not necessarily the
2626 same as the number actually killed).
2628 $cnt = kill 1, $child1, $child2;
2631 If SIGNAL is zero, no signal is sent to the process, but the kill(2)
2632 system call will check whether it's possible to send a signal to it (that
2633 means, to be brief, that the process is owned by the same user, or we are
2634 the super-user). This is a useful way to check that a child process is
2635 alive (even if only as a zombie) and hasn't changed its UID. See
2636 L<perlport> for notes on the portability of this construct.
2638 Unlike in the shell, if SIGNAL is negative, it kills process groups instead
2639 of processes. That means you usually want to use positive not negative signals.
2640 You may also use a signal name in quotes.
2642 The behavior of kill when a I<PROCESS> number is zero or negative depends on
2643 the operating system. For example, on POSIX-conforming systems, zero will
2644 signal the current process group and -1 will signal all processes.
2646 See L<perlipc/"Signals"> for more details.
2653 The C<last> command is like the C<break> statement in C (as used in
2654 loops); it immediately exits the loop in question. If the LABEL is
2655 omitted, the command refers to the innermost enclosing loop. The
2656 C<continue> block, if any, is not executed:
2658 LINE: while (<STDIN>) {
2659 last LINE if /^$/; # exit when done with header
2663 C<last> cannot be used to exit a block which returns a value such as
2664 C<eval {}>, C<sub {}> or C<do {}>, and should not be used to exit
2665 a grep() or map() operation.
2667 Note that a block by itself is semantically identical to a loop
2668 that executes once. Thus C<last> can be used to effect an early
2669 exit out of such a block.
2671 See also L</continue> for an illustration of how C<last>, C<next>, and
2679 Returns a lowercased version of EXPR. This is the internal function
2680 implementing the C<\L> escape in double-quoted strings. Respects
2681 current LC_CTYPE locale if C<use locale> in force. See L<perllocale>
2682 and L<perlunicode> for more details about locale and Unicode support.
2684 If EXPR is omitted, uses C<$_>.
2687 X<lcfirst> X<lowercase>
2691 Returns the value of EXPR with the first character lowercased. This
2692 is the internal function implementing the C<\l> escape in
2693 double-quoted strings. Respects current LC_CTYPE locale if C<use
2694 locale> in force. See L<perllocale> and L<perlunicode> for more
2695 details about locale and Unicode support.
2697 If EXPR is omitted, uses C<$_>.
2704 Returns the length in I<characters> of the value of EXPR. If EXPR is
2705 omitted, returns length of C<$_>. If EXPR is undefined, returns C<undef>.
2706 Note that this cannot be used on an entire array or hash to find out how
2707 many elements these have. For that, use C<scalar @array> and C<scalar keys
2708 %hash> respectively.
2710 Note the I<characters>: if the EXPR is in Unicode, you will get the
2711 number of characters, not the number of bytes. To get the length
2712 of the internal string in bytes, use C<bytes::length(EXPR)>, see
2713 L<bytes>. Note that the internal encoding is variable, and the number
2714 of bytes usually meaningless. To get the number of bytes that the
2715 string would have when encoded as UTF-8, use
2716 C<length(Encoding::encode_utf8(EXPR))>.
2718 =item link OLDFILE,NEWFILE
2721 Creates a new filename linked to the old filename. Returns true for
2722 success, false otherwise.
2724 =item listen SOCKET,QUEUESIZE
2727 Does the same thing that the listen system call does. Returns true if
2728 it succeeded, false otherwise. See the example in
2729 L<perlipc/"Sockets: Client/Server Communication">.
2734 You really probably want to be using C<my> instead, because C<local> isn't
2735 what most people think of as "local". See
2736 L<perlsub/"Private Variables via my()"> for details.
2738 A local modifies the listed variables to be local to the enclosing
2739 block, file, or eval. If more than one value is listed, the list must
2740 be placed in parentheses. See L<perlsub/"Temporary Values via local()">
2741 for details, including issues with tied arrays and hashes.
2743 The C<delete local EXPR> construct can also be used to localize the deletion
2744 of array/hash elements to the current block.
2745 See L<perlsub/"Localized deletion of elements of composite types">.
2747 =item localtime EXPR
2748 X<localtime> X<ctime>
2752 Converts a time as returned by the time function to a 9-element list
2753 with the time analyzed for the local time zone. Typically used as
2757 ($sec,$min,$hour,$mday,$mon,$year,$wday,$yday,$isdst) =
2760 All list elements are numeric, and come straight out of the C `struct
2761 tm'. C<$sec>, C<$min>, and C<$hour> are the seconds, minutes, and hours
2762 of the specified time.
2764 C<$mday> is the day of the month, and C<$mon> is the month itself, in
2765 the range C<0..11> with 0 indicating January and 11 indicating December.
2766 This makes it easy to get a month name from a list:
2768 my @abbr = qw( Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec );
2769 print "$abbr[$mon] $mday";
2770 # $mon=9, $mday=18 gives "Oct 18"
2772 C<$year> is the number of years since 1900, not just the last two digits
2773 of the year. That is, C<$year> is C<123> in year 2023. The proper way
2774 to get a complete 4-digit year is simply:
2778 Otherwise you create non-Y2K-compliant programs--and you wouldn't want
2779 to do that, would you?
2781 To get the last two digits of the year (e.g., '01' in 2001) do:
2783 $year = sprintf("%02d", $year % 100);
2785 C<$wday> is the day of the week, with 0 indicating Sunday and 3 indicating
2786 Wednesday. C<$yday> is the day of the year, in the range C<0..364>
2787 (or C<0..365> in leap years.)
2789 C<$isdst> is true if the specified time occurs during Daylight Saving
2790 Time, false otherwise.
2792 If EXPR is omitted, C<localtime()> uses the current time (as returned
2795 In scalar context, C<localtime()> returns the ctime(3) value:
2797 $now_string = localtime; # e.g., "Thu Oct 13 04:54:34 1994"
2799 This scalar value is B<not> locale dependent but is a Perl builtin. For GMT
2800 instead of local time use the L</gmtime> builtin. See also the
2801 C<Time::Local> module (to convert the second, minutes, hours, ... back to
2802 the integer value returned by time()), and the L<POSIX> module's strftime(3)
2803 and mktime(3) functions.
2805 To get somewhat similar but locale dependent date strings, set up your
2806 locale environment variables appropriately (please see L<perllocale>) and
2809 use POSIX qw(strftime);
2810 $now_string = strftime "%a %b %e %H:%M:%S %Y", localtime;
2811 # or for GMT formatted appropriately for your locale:
2812 $now_string = strftime "%a %b %e %H:%M:%S %Y", gmtime;
2814 Note that the C<%a> and C<%b>, the short forms of the day of the week
2815 and the month of the year, may not necessarily be three characters wide.
2817 See L<perlport/localtime> for portability concerns.
2819 The L<Time::gmtime> and L<Time::localtime> modules provides a convenient,
2820 by-name access mechanism to the gmtime() and localtime() functions,
2823 For a comprehensive date and time representation look at the
2824 L<DateTime> module on CPAN.
2829 This function places an advisory lock on a shared variable, or referenced
2830 object contained in I<THING> until the lock goes out of scope.
2832 lock() is a "weak keyword" : this means that if you've defined a function
2833 by this name (before any calls to it), that function will be called
2834 instead. (However, if you've said C<use threads>, lock() is always a
2835 keyword.) See L<threads>.
2838 X<log> X<logarithm> X<e> X<ln> X<base>
2842 Returns the natural logarithm (base I<e>) of EXPR. If EXPR is omitted,
2843 returns log of C<$_>. To get the log of another base, use basic algebra:
2844 The base-N log of a number is equal to the natural log of that number
2845 divided by the natural log of N. For example:
2849 return log($n)/log(10);
2852 See also L</exp> for the inverse operation.
2859 Does the same thing as the C<stat> function (including setting the
2860 special C<_> filehandle) but stats a symbolic link instead of the file
2861 the symbolic link points to. If symbolic links are unimplemented on
2862 your system, a normal C<stat> is done. For much more detailed
2863 information, please see the documentation for C<stat>.
2865 If EXPR is omitted, stats C<$_>.
2869 The match operator. See L<perlop/"Regexp Quote-Like Operators">.
2871 =item map BLOCK LIST
2876 Evaluates the BLOCK or EXPR for each element of LIST (locally setting
2877 C<$_> to each element) and returns the list value composed of the
2878 results of each such evaluation. In scalar context, returns the
2879 total number of elements so generated. Evaluates BLOCK or EXPR in
2880 list context, so each element of LIST may produce zero, one, or
2881 more elements in the returned value.
2883 @chars = map(chr, @nums);
2885 translates a list of numbers to the corresponding characters. And
2887 %hash = map { get_a_key_for($_) => $_ } @array;
2889 is just a funny way to write
2893 $hash{get_a_key_for($_)} = $_;
2896 Note that C<$_> is an alias to the list value, so it can be used to
2897 modify the elements of the LIST. While this is useful and supported,
2898 it can cause bizarre results if the elements of LIST are not variables.
2899 Using a regular C<foreach> loop for this purpose would be clearer in
2900 most cases. See also L</grep> for an array composed of those items of
2901 the original list for which the BLOCK or EXPR evaluates to true.
2903 If C<$_> is lexical in the scope where the C<map> appears (because it has
2904 been declared with C<my $_>), then, in addition to being locally aliased to
2905 the list elements, C<$_> keeps being lexical inside the block; that is, it
2906 can't be seen from the outside, avoiding any potential side-effects.
2908 C<{> starts both hash references and blocks, so C<map { ...> could be either
2909 the start of map BLOCK LIST or map EXPR, LIST. Because perl doesn't look
2910 ahead for the closing C<}> it has to take a guess at which its dealing with
2911 based what it finds just after the C<{>. Usually it gets it right, but if it
2912 doesn't it won't realize something is wrong until it gets to the C<}> and
2913 encounters the missing (or unexpected) comma. The syntax error will be
2914 reported close to the C<}> but you'll need to change something near the C<{>
2915 such as using a unary C<+> to give perl some help:
2917 %hash = map { "\L$_", 1 } @array # perl guesses EXPR. wrong
2918 %hash = map { +"\L$_", 1 } @array # perl guesses BLOCK. right
2919 %hash = map { ("\L$_", 1) } @array # this also works
2920 %hash = map { lc($_), 1 } @array # as does this.
2921 %hash = map +( lc($_), 1 ), @array # this is EXPR and works!
2923 %hash = map ( lc($_), 1 ), @array # evaluates to (1, @array)
2925 or to force an anon hash constructor use C<+{>:
2927 @hashes = map +{ lc($_), 1 }, @array # EXPR, so needs , at end
2929 and you get list of anonymous hashes each with only 1 entry.
2931 =item mkdir FILENAME,MASK
2932 X<mkdir> X<md> X<directory, create>
2934 =item mkdir FILENAME
2938 Creates the directory specified by FILENAME, with permissions
2939 specified by MASK (as modified by C<umask>). If it succeeds it
2940 returns true, otherwise it returns false and sets C<$!> (errno).
2941 If omitted, MASK defaults to 0777. If omitted, FILENAME defaults
2944 In general, it is better to create directories with permissive MASK,
2945 and let the user modify that with their C<umask>, than it is to supply
2946 a restrictive MASK and give the user no way to be more permissive.
2947 The exceptions to this rule are when the file or directory should be
2948 kept private (mail files, for instance). The perlfunc(1) entry on
2949 C<umask> discusses the choice of MASK in more detail.
2951 Note that according to the POSIX 1003.1-1996 the FILENAME may have any
2952 number of trailing slashes. Some operating and filesystems do not get
2953 this right, so Perl automatically removes all trailing slashes to keep
2956 In order to recursively create a directory structure look at
2957 the C<mkpath> function of the L<File::Path> module.
2959 =item msgctl ID,CMD,ARG
2962 Calls the System V IPC function msgctl(2). You'll probably have to say
2966 first to get the correct constant definitions. If CMD is C<IPC_STAT>,
2967 then ARG must be a variable that will hold the returned C<msqid_ds>
2968 structure. Returns like C<ioctl>: the undefined value for error,
2969 C<"0 but true"> for zero, or the actual return value otherwise. See also
2970 L<perlipc/"SysV IPC">, C<IPC::SysV>, and C<IPC::Semaphore> documentation.
2972 =item msgget KEY,FLAGS
2975 Calls the System V IPC function msgget(2). Returns the message queue
2976 id, or the undefined value if there is an error. See also
2977 L<perlipc/"SysV IPC"> and C<IPC::SysV> and C<IPC::Msg> documentation.
2979 =item msgrcv ID,VAR,SIZE,TYPE,FLAGS
2982 Calls the System V IPC function msgrcv to receive a message from
2983 message queue ID into variable VAR with a maximum message size of
2984 SIZE. Note that when a message is received, the message type as a
2985 native long integer will be the first thing in VAR, followed by the
2986 actual message. This packing may be opened with C<unpack("l! a*")>.
2987 Taints the variable. Returns true if successful, or false if there is
2988 an error. See also L<perlipc/"SysV IPC">, C<IPC::SysV>, and
2989 C<IPC::SysV::Msg> documentation.
2991 =item msgsnd ID,MSG,FLAGS
2994 Calls the System V IPC function msgsnd to send the message MSG to the
2995 message queue ID. MSG must begin with the native long integer message
2996 type, and be followed by the length of the actual message, and finally
2997 the message itself. This kind of packing can be achieved with
2998 C<pack("l! a*", $type, $message)>. Returns true if successful,
2999 or false if there is an error. See also C<IPC::SysV>
3000 and C<IPC::SysV::Msg> documentation.
3007 =item my EXPR : ATTRS
3009 =item my TYPE EXPR : ATTRS
3011 A C<my> declares the listed variables to be local (lexically) to the
3012 enclosing block, file, or C<eval>. If more than one value is listed,
3013 the list must be placed in parentheses.
3015 The exact semantics and interface of TYPE and ATTRS are still
3016 evolving. TYPE is currently bound to the use of C<fields> pragma,
3017 and attributes are handled using the C<attributes> pragma, or starting
3018 from Perl 5.8.0 also via the C<Attribute::Handlers> module. See
3019 L<perlsub/"Private Variables via my()"> for details, and L<fields>,
3020 L<attributes>, and L<Attribute::Handlers>.
3027 The C<next> command is like the C<continue> statement in C; it starts
3028 the next iteration of the loop:
3030 LINE: while (<STDIN>) {
3031 next LINE if /^#/; # discard comments
3035 Note that if there were a C<continue> block on the above, it would get
3036 executed even on discarded lines. If the LABEL is omitted, the command
3037 refers to the innermost enclosing loop.
3039 C<next> cannot be used to exit a block which returns a value such as
3040 C<eval {}>, C<sub {}> or C<do {}>, and should not be used to exit
3041 a grep() or map() operation.
3043 Note that a block by itself is semantically identical to a loop
3044 that executes once. Thus C<next> will exit such a block early.
3046 See also L</continue> for an illustration of how C<last>, C<next>, and
3049 =item no Module VERSION LIST
3052 =item no Module VERSION
3054 =item no Module LIST
3060 See the C<use> function, of which C<no> is the opposite.
3063 X<oct> X<octal> X<hex> X<hexadecimal> X<binary> X<bin>
3067 Interprets EXPR as an octal string and returns the corresponding
3068 value. (If EXPR happens to start off with C<0x>, interprets it as a
3069 hex string. If EXPR starts off with C<0b>, it is interpreted as a
3070 binary string. Leading whitespace is ignored in all three cases.)
3071 The following will handle decimal, binary, octal, and hex in the standard
3074 $val = oct($val) if $val =~ /^0/;
3076 If EXPR is omitted, uses C<$_>. To go the other way (produce a number
3077 in octal), use sprintf() or printf():
3079 $perms = (stat("filename"))[2] & 07777;
3080 $oct_perms = sprintf "%lo", $perms;
3082 The oct() function is commonly used when a string such as C<644> needs
3083 to be converted into a file mode, for example. (Although perl will
3084 automatically convert strings into numbers as needed, this automatic
3085 conversion assumes base 10.)
3087 =item open FILEHANDLE,EXPR
3088 X<open> X<pipe> X<file, open> X<fopen>
3090 =item open FILEHANDLE,MODE,EXPR
3092 =item open FILEHANDLE,MODE,EXPR,LIST
3094 =item open FILEHANDLE,MODE,REFERENCE
3096 =item open FILEHANDLE
3098 Opens the file whose filename is given by EXPR, and associates it with
3101 Simple examples to open a file for reading:
3103 open(my $fh, '<', "input.txt") or die $!;
3107 open(my $fh, '>', "output.txt") or die $!;
3109 (The following is a comprehensive reference to open(): for a gentler
3110 introduction you may consider L<perlopentut>.)
3112 If FILEHANDLE is an undefined scalar variable (or array or hash element)
3113 the variable is assigned a reference to a new anonymous filehandle,
3114 otherwise if FILEHANDLE is an expression, its value is used as the name of
3115 the real filehandle wanted. (This is considered a symbolic reference, so
3116 C<use strict 'refs'> should I<not> be in effect.)
3118 If EXPR is omitted, the scalar variable of the same name as the
3119 FILEHANDLE contains the filename. (Note that lexical variables--those
3120 declared with C<my>--will not work for this purpose; so if you're
3121 using C<my>, specify EXPR in your call to open.)
3123 If three or more arguments are specified then the mode of opening and
3124 the file name are separate. If MODE is C<< '<' >> or nothing, the file
3125 is opened for input. If MODE is C<< '>' >>, the file is truncated and
3126 opened for output, being created if necessary. If MODE is C<<< '>>' >>>,
3127 the file is opened for appending, again being created if necessary.
3129 You can put a C<'+'> in front of the C<< '>' >> or C<< '<' >> to
3130 indicate that you want both read and write access to the file; thus
3131 C<< '+<' >> is almost always preferred for read/write updates--the C<<
3132 '+>' >> mode would clobber the file first. You can't usually use
3133 either read-write mode for updating textfiles, since they have
3134 variable length records. See the B<-i> switch in L<perlrun> for a
3135 better approach. The file is created with permissions of C<0666>
3136 modified by the process' C<umask> value.
3138 These various prefixes correspond to the fopen(3) modes of C<'r'>,
3139 C<'r+'>, C<'w'>, C<'w+'>, C<'a'>, and C<'a+'>.
3141 In the 2-arguments (and 1-argument) form of the call the mode and
3142 filename should be concatenated (in this order), possibly separated by
3143 spaces. It is possible to omit the mode in these forms if the mode is
3146 If the filename begins with C<'|'>, the filename is interpreted as a
3147 command to which output is to be piped, and if the filename ends with a
3148 C<'|'>, the filename is interpreted as a command which pipes output to
3149 us. See L<perlipc/"Using open() for IPC">
3150 for more examples of this. (You are not allowed to C<open> to a command
3151 that pipes both in I<and> out, but see L<IPC::Open2>, L<IPC::Open3>,
3152 and L<perlipc/"Bidirectional Communication with Another Process">
3155 For three or more arguments if MODE is C<'|-'>, the filename is
3156 interpreted as a command to which output is to be piped, and if MODE
3157 is C<'-|'>, the filename is interpreted as a command which pipes
3158 output to us. In the 2-arguments (and 1-argument) form one should
3159 replace dash (C<'-'>) with the command.
3160 See L<perlipc/"Using open() for IPC"> for more examples of this.
3161 (You are not allowed to C<open> to a command that pipes both in I<and>
3162 out, but see L<IPC::Open2>, L<IPC::Open3>, and
3163 L<perlipc/"Bidirectional Communication"> for alternatives.)
3165 In the three-or-more argument form of pipe opens, if LIST is specified
3166 (extra arguments after the command name) then LIST becomes arguments
3167 to the command invoked if the platform supports it. The meaning of
3168 C<open> with more than three arguments for non-pipe modes is not yet
3169 specified. Experimental "layers" may give extra LIST arguments
3172 In the 2-arguments (and 1-argument) form opening C<'-'> opens STDIN
3173 and opening C<< '>-' >> opens STDOUT.
3175 You may use the three-argument form of open to specify IO "layers"
3176 (sometimes also referred to as "disciplines") to be applied to the handle
3177 that affect how the input and output are processed (see L<open> and
3178 L<PerlIO> for more details). For example
3180 open(my $fh, "<:encoding(UTF-8)", "file")
3182 will open the UTF-8 encoded file containing Unicode characters,
3183 see L<perluniintro>. Note that if layers are specified in the
3184 three-arg form then default layers stored in ${^OPEN} (see L<perlvar>;
3185 usually set by the B<open> pragma or the switch B<-CioD>) are ignored.
3187 Open returns nonzero upon success, the undefined value otherwise. If
3188 the C<open> involved a pipe, the return value happens to be the pid of
3191 If you're running Perl on a system that distinguishes between text
3192 files and binary files, then you should check out L</binmode> for tips
3193 for dealing with this. The key distinction between systems that need
3194 C<binmode> and those that don't is their text file formats. Systems
3195 like Unix, Mac OS, and Plan 9, which delimit lines with a single
3196 character, and which encode that character in C as C<"\n">, do not
3197 need C<binmode>. The rest need it.
3199 When opening a file, it's usually a bad idea to continue normal execution
3200 if the request failed, so C<open> is frequently used in connection with
3201 C<die>. Even if C<die> won't do what you want (say, in a CGI script,
3202 where you want to make a nicely formatted error message (but there are
3203 modules that can help with that problem)) you should always check
3204 the return value from opening a file. The infrequent exception is when
3205 working with an unopened filehandle is actually what you want to do.
3207 As a special case the 3-arg form with a read/write mode and the third
3208 argument being C<undef>:
3210 open(my $tmp, "+>", undef) or die ...
3212 opens a filehandle to an anonymous temporary file. Also using "+<"
3213 works for symmetry, but you really should consider writing something
3214 to the temporary file first. You will need to seek() to do the
3217 Since v5.8.0, perl has built using PerlIO by default. Unless you've
3218 changed this (i.e. Configure -Uuseperlio), you can open file handles to
3219 "in memory" files held in Perl scalars via:
3221 open($fh, '>', \$variable) || ..
3223 Though if you try to re-open C<STDOUT> or C<STDERR> as an "in memory"
3224 file, you have to close it first:
3227 open STDOUT, '>', \$variable or die "Can't open STDOUT: $!";
3232 open ARTICLE or die "Can't find article $ARTICLE: $!\n";
3233 while (<ARTICLE>) {...
3235 open(LOG, '>>/usr/spool/news/twitlog'); # (log is reserved)
3236 # if the open fails, output is discarded
3238 open(my $dbase, '+<', 'dbase.mine') # open for update
3239 or die "Can't open 'dbase.mine' for update: $!";
3241 open(my $dbase, '+<dbase.mine') # ditto
3242 or die "Can't open 'dbase.mine' for update: $!";
3244 open(ARTICLE, '-|', "caesar <$article") # decrypt article
3245 or die "Can't start caesar: $!";
3247 open(ARTICLE, "caesar <$article |") # ditto
3248 or die "Can't start caesar: $!";
3250 open(EXTRACT, "|sort >Tmp$$") # $$ is our process id
3251 or die "Can't start sort: $!";
3254 open(MEMORY,'>', \$var)
3255 or die "Can't open memory file: $!";
3256 print MEMORY "foo!\n"; # output will end up in $var
3258 # process argument list of files along with any includes
3260 foreach $file (@ARGV) {
3261 process($file, 'fh00');
3265 my($filename, $input) = @_;
3266 $input++; # this is a string increment
3267 unless (open($input, $filename)) {
3268 print STDERR "Can't open $filename: $!\n";
3273 while (<$input>) { # note use of indirection
3274 if (/^#include "(.*)"/) {
3275 process($1, $input);
3282 See L<perliol> for detailed info on PerlIO.
3284 You may also, in the Bourne shell tradition, specify an EXPR beginning
3285 with C<< '>&' >>, in which case the rest of the string is interpreted
3286 as the name of a filehandle (or file descriptor, if numeric) to be
3287 duped (as C<dup(2)>) and opened. You may use C<&> after C<< > >>,
3288 C<<< >> >>>, C<< < >>, C<< +> >>, C<<< +>> >>>, and C<< +< >>.
3289 The mode you specify should match the mode of the original filehandle.
3290 (Duping a filehandle does not take into account any existing contents
3291 of IO buffers.) If you use the 3-arg form then you can pass either a
3292 number, the name of a filehandle or the normal "reference to a glob".
3294 Here is a script that saves, redirects, and restores C<STDOUT> and
3295 C<STDERR> using various methods:
3298 open my $oldout, ">&STDOUT" or die "Can't dup STDOUT: $!";
3299 open OLDERR, ">&", \*STDERR or die "Can't dup STDERR: $!";
3301 open STDOUT, '>', "foo.out" or die "Can't redirect STDOUT: $!";
3302 open STDERR, ">&STDOUT" or die "Can't dup STDOUT: $!";
3304 select STDERR; $| = 1; # make unbuffered
3305 select STDOUT; $| = 1; # make unbuffered
3307 print STDOUT "stdout 1\n"; # this works for
3308 print STDERR "stderr 1\n"; # subprocesses too
3310 open STDOUT, ">&", $oldout or die "Can't dup \$oldout: $!";
3311 open STDERR, ">&OLDERR" or die "Can't dup OLDERR: $!";
3313 print STDOUT "stdout 2\n";
3314 print STDERR "stderr 2\n";
3316 If you specify C<< '<&=X' >>, where C<X> is a file descriptor number
3317 or a filehandle, then Perl will do an equivalent of C's C<fdopen> of
3318 that file descriptor (and not call C<dup(2)>); this is more
3319 parsimonious of file descriptors. For example:
3321 # open for input, reusing the fileno of $fd
3322 open(FILEHANDLE, "<&=$fd")
3326 open(FILEHANDLE, "<&=", $fd)
3330 # open for append, using the fileno of OLDFH
3331 open(FH, ">>&=", OLDFH)
3335 open(FH, ">>&=OLDFH")
3337 Being parsimonious on filehandles is also useful (besides being
3338 parsimonious) for example when something is dependent on file
3339 descriptors, like for example locking using flock(). If you do just
3340 C<< open(A, '>>&B') >>, the filehandle A will not have the same file
3341 descriptor as B, and therefore flock(A) will not flock(B), and vice
3342 versa. But with C<< open(A, '>>&=B') >> the filehandles will share
3343 the same file descriptor.
3345 Note that if you are using Perls older than 5.8.0, Perl will be using
3346 the standard C libraries' fdopen() to implement the "=" functionality.
3347 On many UNIX systems fdopen() fails when file descriptors exceed a
3348 certain value, typically 255. For Perls 5.8.0 and later, PerlIO is
3349 most often the default.
3351 You can see whether Perl has been compiled with PerlIO or not by
3352 running C<perl -V> and looking for C<useperlio=> line. If C<useperlio>
3353 is C<define>, you have PerlIO, otherwise you don't.
3355 If you open a pipe on the command C<'-'>, i.e., either C<'|-'> or C<'-|'>
3356 with 2-arguments (or 1-argument) form of open(), then
3357 there is an implicit fork done, and the return value of open is the pid
3358 of the child within the parent process, and C<0> within the child
3359 process. (Use C<defined($pid)> to determine whether the open was successful.)
3360 The filehandle behaves normally for the parent, but i/o to that
3361 filehandle is piped from/to the STDOUT/STDIN of the child process.
3362 In the child process the filehandle isn't opened--i/o happens from/to
3363 the new STDOUT or STDIN. Typically this is used like the normal
3364 piped open when you want to exercise more control over just how the
3365 pipe command gets executed, such as when you are running setuid, and
3366 don't want to have to scan shell commands for metacharacters.
3367 The following triples are more or less equivalent:
3369 open(FOO, "|tr '[a-z]' '[A-Z]'");
3370 open(FOO, '|-', "tr '[a-z]' '[A-Z]'");
3371 open(FOO, '|-') || exec 'tr', '[a-z]', '[A-Z]';
3372 open(FOO, '|-', "tr", '[a-z]', '[A-Z]');
3374 open(FOO, "cat -n '$file'|");
3375 open(FOO, '-|', "cat -n '$file'");
3376 open(FOO, '-|') || exec 'cat', '-n', $file;
3377 open(FOO, '-|', "cat", '-n', $file);
3379 The last example in each block shows the pipe as "list form", which is
3380 not yet supported on all platforms. A good rule of thumb is that if
3381 your platform has true C<fork()> (in other words, if your platform is
3382 UNIX) you can use the list form.
3384 See L<perlipc/"Safe Pipe Opens"> for more examples of this.
3386 Beginning with v5.6.0, Perl will attempt to flush all files opened for
3387 output before any operation that may do a fork, but this may not be
3388 supported on some platforms (see L<perlport>). To be safe, you may need
3389 to set C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method
3390 of C<IO::Handle> on any open handles.
3392 On systems that support a close-on-exec flag on files, the flag will
3393 be set for the newly opened file descriptor as determined by the value
3394 of $^F. See L<perlvar/$^F>.
3396 Closing any piped filehandle causes the parent process to wait for the
3397 child to finish, and returns the status value in C<$?> and
3398 C<${^CHILD_ERROR_NATIVE}>.
3400 The filename passed to 2-argument (or 1-argument) form of open() will
3401 have leading and trailing whitespace deleted, and the normal
3402 redirection characters honored. This property, known as "magic open",
3403 can often be used to good effect. A user could specify a filename of
3404 F<"rsh cat file |">, or you could change certain filenames as needed:
3406 $filename =~ s/(.*\.gz)\s*$/gzip -dc < $1|/;
3407 open(FH, $filename) or die "Can't open $filename: $!";
3409 Use 3-argument form to open a file with arbitrary weird characters in it,
3411 open(FOO, '<', $file);
3413 otherwise it's necessary to protect any leading and trailing whitespace:
3415 $file =~ s#^(\s)#./$1#;
3416 open(FOO, "< $file\0");
3418 (this may not work on some bizarre filesystems). One should
3419 conscientiously choose between the I<magic> and 3-arguments form
3424 will allow the user to specify an argument of the form C<"rsh cat file |">,
3425 but will not work on a filename which happens to have a trailing space, while
3427 open IN, '<', $ARGV[0];
3429 will have exactly the opposite restrictions.
3431 If you want a "real" C C<open> (see C<open(2)> on your system), then you
3432 should use the C<sysopen> function, which involves no such magic (but
3433 may use subtly different filemodes than Perl open(), which is mapped
3434 to C fopen()). This is
3435 another way to protect your filenames from interpretation. For example:
3438 sysopen(HANDLE, $path, O_RDWR|O_CREAT|O_EXCL)
3439 or die "sysopen $path: $!";
3440 $oldfh = select(HANDLE); $| = 1; select($oldfh);
3441 print HANDLE "stuff $$\n";
3443 print "File contains: ", <HANDLE>;
3445 Using the constructor from the C<IO::Handle> package (or one of its
3446 subclasses, such as C<IO::File> or C<IO::Socket>), you can generate anonymous
3447 filehandles that have the scope of whatever variables hold references to
3448 them, and automatically close whenever and however you leave that scope:
3452 sub read_myfile_munged {
3454 my $handle = IO::File->new;
3455 open($handle, "myfile") or die "myfile: $!";
3457 or return (); # Automatically closed here.
3458 mung $first or die "mung failed"; # Or here.
3459 return $first, <$handle> if $ALL; # Or here.
3463 See L</seek> for some details about mixing reading and writing.
3465 =item opendir DIRHANDLE,EXPR
3468 Opens a directory named EXPR for processing by C<readdir>, C<telldir>,
3469 C<seekdir>, C<rewinddir>, and C<closedir>. Returns true if successful.
3470 DIRHANDLE may be an expression whose value can be used as an indirect
3471 dirhandle, usually the real dirhandle name. If DIRHANDLE is an undefined
3472 scalar variable (or array or hash element), the variable is assigned a
3473 reference to a new anonymous dirhandle.
3474 DIRHANDLEs have their own namespace separate from FILEHANDLEs.
3476 See example at C<readdir>.
3483 Returns the numeric (the native 8-bit encoding, like ASCII or EBCDIC,
3484 or Unicode) value of the first character of EXPR. If EXPR is omitted,
3487 For the reverse, see L</chr>.
3488 See L<perlunicode> for more about Unicode.
3495 =item our EXPR : ATTRS
3497 =item our TYPE EXPR : ATTRS
3499 C<our> associates a simple name with a package variable in the current
3500 package for use within the current scope. When C<use strict 'vars'> is in
3501 effect, C<our> lets you use declared global variables without qualifying
3502 them with package names, within the lexical scope of the C<our> declaration.
3503 In this way C<our> differs from C<use vars>, which is package scoped.
3505 Unlike C<my>, which both allocates storage for a variable and associates
3506 a simple name with that storage for use within the current scope, C<our>
3507 associates a simple name with a package variable in the current package,
3508 for use within the current scope. In other words, C<our> has the same
3509 scoping rules as C<my>, but does not necessarily create a
3512 If more than one value is listed, the list must be placed
3518 An C<our> declaration declares a global variable that will be visible
3519 across its entire lexical scope, even across package boundaries. The
3520 package in which the variable is entered is determined at the point
3521 of the declaration, not at the point of use. This means the following
3525 our $bar; # declares $Foo::bar for rest of lexical scope
3529 print $bar; # prints 20, as it refers to $Foo::bar
3531 Multiple C<our> declarations with the same name in the same lexical
3532 scope are allowed if they are in different packages. If they happen
3533 to be in the same package, Perl will emit warnings if you have asked
3534 for them, just like multiple C<my> declarations. Unlike a second
3535 C<my> declaration, which will bind the name to a fresh variable, a
3536 second C<our> declaration in the same package, in the same scope, is
3541 our $bar; # declares $Foo::bar for rest of lexical scope
3545 our $bar = 30; # declares $Bar::bar for rest of lexical scope
3546 print $bar; # prints 30
3548 our $bar; # emits warning but has no other effect
3549 print $bar; # still prints 30
3551 An C<our> declaration may also have a list of attributes associated
3554 The exact semantics and interface of TYPE and ATTRS are still
3555 evolving. TYPE is currently bound to the use of C<fields> pragma,
3556 and attributes are handled using the C<attributes> pragma, or starting
3557 from Perl 5.8.0 also via the C<Attribute::Handlers> module. See
3558 L<perlsub/"Private Variables via my()"> for details, and L<fields>,
3559 L<attributes>, and L<Attribute::Handlers>.
3561 =item pack TEMPLATE,LIST
3564 Takes a LIST of values and converts it into a string using the rules
3565 given by the TEMPLATE. The resulting string is the concatenation of
3566 the converted values. Typically, each converted value looks
3567 like its machine-level representation. For example, on 32-bit machines
3568 an integer may be represented by a sequence of 4 bytes that will be
3569 converted to a sequence of 4 characters.
3571 The TEMPLATE is a sequence of characters that give the order and type
3572 of values, as follows:
3574 a A string with arbitrary binary data, will be null padded.
3575 A A text (ASCII) string, will be space padded.
3576 Z A null terminated (ASCIZ) string, will be null padded.
3578 b A bit string (ascending bit order inside each byte, like vec()).
3579 B A bit string (descending bit order inside each byte).
3580 h A hex string (low nybble first).
3581 H A hex string (high nybble first).
3583 c A signed char (8-bit) value.
3584 C An unsigned char (octet) value.
3585 W An unsigned char value (can be greater than 255).
3587 s A signed short (16-bit) value.
3588 S An unsigned short value.
3590 l A signed long (32-bit) value.
3591 L An unsigned long value.
3593 q A signed quad (64-bit) value.
3594 Q An unsigned quad value.
3595 (Quads are available only if your system supports 64-bit
3596 integer values _and_ if Perl has been compiled to support those.
3597 Causes a fatal error otherwise.)
3599 i A signed integer value.
3600 I A unsigned integer value.
3601 (This 'integer' is _at_least_ 32 bits wide. Its exact
3602 size depends on what a local C compiler calls 'int'.)
3604 n An unsigned short (16-bit) in "network" (big-endian) order.
3605 N An unsigned long (32-bit) in "network" (big-endian) order.
3606 v An unsigned short (16-bit) in "VAX" (little-endian) order.
3607 V An unsigned long (32-bit) in "VAX" (little-endian) order.
3609 j A Perl internal signed integer value (IV).
3610 J A Perl internal unsigned integer value (UV).
3612 f A single-precision float in the native format.
3613 d A double-precision float in the native format.
3615 F A Perl internal floating point value (NV) in the native format
3616 D A long double-precision float in the native format.
3617 (Long doubles are available only if your system supports long
3618 double values _and_ if Perl has been compiled to support those.
3619 Causes a fatal error otherwise.)
3621 p A pointer to a null-terminated string.
3622 P A pointer to a structure (fixed-length string).
3624 u A uuencoded string.
3625 U A Unicode character number. Encodes to a character in character mode
3626 and UTF-8 (or UTF-EBCDIC in EBCDIC platforms) in byte mode.
3628 w A BER compressed integer (not an ASN.1 BER, see perlpacktut for
3629 details). Its bytes represent an unsigned integer in base 128,
3630 most significant digit first, with as few digits as possible. Bit
3631 eight (the high bit) is set on each byte except the last.
3635 @ Null fill or truncate to absolute position, counted from the
3636 start of the innermost ()-group.
3637 . Null fill or truncate to absolute position specified by value.
3638 ( Start of a ()-group.
3640 One or more of the modifiers below may optionally follow some letters in the
3641 TEMPLATE (the second column lists the letters for which the modifier is
3644 ! sSlLiI Forces native (short, long, int) sizes instead
3645 of fixed (16-/32-bit) sizes.
3647 xX Make x and X act as alignment commands.
3649 nNvV Treat integers as signed instead of unsigned.
3651 @. Specify position as byte offset in the internal
3652 representation of the packed string. Efficient but
3655 > sSiIlLqQ Force big-endian byte-order on the type.
3656 jJfFdDpP (The "big end" touches the construct.)
3658 < sSiIlLqQ Force little-endian byte-order on the type.
3659 jJfFdDpP (The "little end" touches the construct.)
3661 The C<E<gt>> and C<E<lt>> modifiers can also be used on C<()>-groups,
3662 in which case they force a certain byte-order on all components of
3663 that group, including subgroups.
3665 The following rules apply:
3671 Each letter may optionally be followed by a number giving a repeat
3672 count. With all types except C<a>, C<A>, C<Z>, C<b>, C<B>, C<h>,
3673 C<H>, C<@>, C<.>, C<x>, C<X> and C<P> the pack function will gobble up
3674 that many values from the LIST. A C<*> for the repeat count means to
3675 use however many items are left, except for C<@>, C<x>, C<X>, where it
3676 is equivalent to C<0>, for <.> where it means relative to string start
3677 and C<u>, where it is equivalent to 1 (or 45, which is the same).
3678 A numeric repeat count may optionally be enclosed in brackets, as in
3679 C<pack 'C[80]', @arr>.
3681 One can replace the numeric repeat count by a template enclosed in brackets;
3682 then the packed length of this template in bytes is used as a count.
3683 For example, C<x[L]> skips a long (it skips the number of bytes in a long);
3684 the template C<$t X[$t] $t> unpack()s twice what $t unpacks.
3685 If the template in brackets contains alignment commands (such as C<x![d]>),
3686 its packed length is calculated as if the start of the template has the maximal
3689 When used with C<Z>, C<*> results in the addition of a trailing null
3690 byte (so the packed result will be one longer than the byte C<length>
3693 When used with C<@>, the repeat count represents an offset from the start
3694 of the innermost () group.
3696 When used with C<.>, the repeat count is used to determine the starting
3697 position from where the value offset is calculated. If the repeat count
3698 is 0, it's relative to the current position. If the repeat count is C<*>,
3699 the offset is relative to the start of the packed string. And if its an
3700 integer C<n> the offset is relative to the start of the n-th innermost
3701 () group (or the start of the string if C<n> is bigger then the group
3704 The repeat count for C<u> is interpreted as the maximal number of bytes
3705 to encode per line of output, with 0, 1 and 2 replaced by 45. The repeat
3706 count should not be more than 65.
3710 The C<a>, C<A>, and C<Z> types gobble just one value, but pack it as a
3711 string of length count, padding with nulls or spaces as necessary. When
3712 unpacking, C<A> strips trailing whitespace and nulls, C<Z> strips everything
3713 after the first null, and C<a> returns data verbatim.
3715 If the value-to-pack is too long, it is truncated. If too long and an
3716 explicit count is provided, C<Z> packs only C<$count-1> bytes, followed
3717 by a null byte. Thus C<Z> always packs a trailing null (except when the
3722 Likewise, the C<b> and C<B> fields pack a string that many bits long.
3723 Each character of the input field of pack() generates 1 bit of the result.
3724 Each result bit is based on the least-significant bit of the corresponding
3725 input character, i.e., on C<ord($char)%2>. In particular, characters C<"0">
3726 and C<"1"> generate bits 0 and 1, as do characters C<"\0"> and C<"\1">.
3728 Starting from the beginning of the input string of pack(), each 8-tuple
3729 of characters is converted to 1 character of output. With format C<b>
3730 the first character of the 8-tuple determines the least-significant bit of a
3731 character, and with format C<B> it determines the most-significant bit of
3734 If the length of the input string is not exactly divisible by 8, the
3735 remainder is packed as if the input string were padded by null characters
3736 at the end. Similarly, during unpack()ing the "extra" bits are ignored.
3738 If the input string of pack() is longer than needed, extra characters are
3739 ignored. A C<*> for the repeat count of pack() means to use all the
3740 characters of the input field. On unpack()ing the bits are converted to a
3741 string of C<"0">s and C<"1">s.
3745 The C<h> and C<H> fields pack a string that many nybbles (4-bit groups,
3746 representable as hexadecimal digits, 0-9a-f) long.
3748 Each character of the input field of pack() generates 4 bits of the result.
3749 For non-alphabetical characters the result is based on the 4 least-significant
3750 bits of the input character, i.e., on C<ord($char)%16>. In particular,
3751 characters C<"0"> and C<"1"> generate nybbles 0 and 1, as do bytes
3752 C<"\0"> and C<"\1">. For characters C<"a".."f"> and C<"A".."F"> the result
3753 is compatible with the usual hexadecimal digits, so that C<"a"> and
3754 C<"A"> both generate the nybble C<0xa==10>. The result for characters
3755 C<"g".."z"> and C<"G".."Z"> is not well-defined.
3757 Starting from the beginning of the input string of pack(), each pair
3758 of characters is converted to 1 character of output. With format C<h> the
3759 first character of the pair determines the least-significant nybble of the
3760 output character, and with format C<H> it determines the most-significant
3763 If the length of the input string is not even, it behaves as if padded
3764 by a null character at the end. Similarly, during unpack()ing the "extra"
3765 nybbles are ignored.
3767 If the input string of pack() is longer than needed, extra characters are
3769 A C<*> for the repeat count of pack() means to use all the characters of
3770 the input field. On unpack()ing the nybbles are converted to a string
3771 of hexadecimal digits.
3775 The C<p> type packs a pointer to a null-terminated string. You are
3776 responsible for ensuring the string is not a temporary value (which can
3777 potentially get deallocated before you get around to using the packed result).
3778 The C<P> type packs a pointer to a structure of the size indicated by the
3779 length. A NULL pointer is created if the corresponding value for C<p> or
3780 C<P> is C<undef>, similarly for unpack().
3782 If your system has a strange pointer size (i.e. a pointer is neither as
3783 big as an int nor as big as a long), it may not be possible to pack or
3784 unpack pointers in big- or little-endian byte order. Attempting to do
3785 so will result in a fatal error.
3789 The C</> template character allows packing and unpacking of a sequence of
3790 items where the packed structure contains a packed item count followed by
3791 the packed items themselves.
3793 For C<pack> you write I<length-item>C</>I<sequence-item> and the
3794 I<length-item> describes how the length value is packed. The ones likely
3795 to be of most use are integer-packing ones like C<n> (for Java strings),
3796 C<w> (for ASN.1 or SNMP) and C<N> (for Sun XDR).
3798 For C<pack>, the I<sequence-item> may have a repeat count, in which case
3799 the minimum of that and the number of available items is used as argument
3800 for the I<length-item>. If it has no repeat count or uses a '*', the number
3801 of available items is used.
3803 For C<unpack> an internal stack of integer arguments unpacked so far is
3804 used. You write C</>I<sequence-item> and the repeat count is obtained by
3805 popping off the last element from the stack. The I<sequence-item> must not
3806 have a repeat count.
3808 If the I<sequence-item> refers to a string type (C<"A">, C<"a"> or C<"Z">),
3809 the I<length-item> is a string length, not a number of strings. If there is
3810 an explicit repeat count for pack, the packed string will be adjusted to that
3813 unpack 'W/a', "\04Gurusamy"; gives ('Guru')
3814 unpack 'a3/A A*', '007 Bond J '; gives (' Bond', 'J')
3815 unpack 'a3 x2 /A A*', '007: Bond, J.'; gives ('Bond, J', '.')
3816 pack 'n/a* w/a','hello,','world'; gives "\000\006hello,\005world"
3817 pack 'a/W2', ord('a') .. ord('z'); gives '2ab'
3819 The I<length-item> is not returned explicitly from C<unpack>.
3821 Adding a count to the I<length-item> letter is unlikely to do anything
3822 useful, unless that letter is C<A>, C<a> or C<Z>. Packing with a
3823 I<length-item> of C<a> or C<Z> may introduce C<"\000"> characters,
3824 which Perl does not regard as legal in numeric strings.
3828 The integer types C<s>, C<S>, C<l>, and C<L> may be
3829 followed by a C<!> modifier to signify native shorts or
3830 longs--as you can see from above for example a bare C<l> does mean
3831 exactly 32 bits, the native C<long> (as seen by the local C compiler)
3832 may be larger. This is an issue mainly in 64-bit platforms. You can
3833 see whether using C<!> makes any difference by
3835 print length(pack("s")), " ", length(pack("s!")), "\n";
3836 print length(pack("l")), " ", length(pack("l!")), "\n";
3838 C<i!> and C<I!> also work but only because of completeness;
3839 they are identical to C<i> and C<I>.
3841 The actual sizes (in bytes) of native shorts, ints, longs, and long
3842 longs on the platform where Perl was built are also available via
3846 print $Config{shortsize}, "\n";
3847 print $Config{intsize}, "\n";
3848 print $Config{longsize}, "\n";
3849 print $Config{longlongsize}, "\n";
3851 (The C<$Config{longlongsize}> will be undefined if your system does
3852 not support long longs.)
3856 The integer formats C<s>, C<S>, C<i>, C<I>, C<l>, C<L>, C<j>, and C<J>
3857 are inherently non-portable between processors and operating systems
3858 because they obey the native byteorder and endianness. For example a
3859 4-byte integer 0x12345678 (305419896 decimal) would be ordered natively
3860 (arranged in and handled by the CPU registers) into bytes as
3862 0x12 0x34 0x56 0x78 # big-endian
3863 0x78 0x56 0x34 0x12 # little-endian
3865 Basically, the Intel and VAX CPUs are little-endian, while everybody
3866 else, for example Motorola m68k/88k, PPC, Sparc, HP PA, Power, and
3867 Cray are big-endian. Alpha and MIPS can be either: Digital/Compaq
3868 used/uses them in little-endian mode; SGI/Cray uses them in big-endian
3871 The names `big-endian' and `little-endian' are comic references to
3872 the classic "Gulliver's Travels" (via the paper "On Holy Wars and a
3873 Plea for Peace" by Danny Cohen, USC/ISI IEN 137, April 1, 1980) and
3874 the egg-eating habits of the Lilliputians.
3876 Some systems may have even weirder byte orders such as
3881 You can see your system's preference with
3883 print join(" ", map { sprintf "%#02x", $_ }
3884 unpack("W*",pack("L",0x12345678))), "\n";
3886 The byteorder on the platform where Perl was built is also available
3890 print $Config{byteorder}, "\n";
3892 Byteorders C<'1234'> and C<'12345678'> are little-endian, C<'4321'>
3893 and C<'87654321'> are big-endian.
3895 If you want portable packed integers you can either use the formats
3896 C<n>, C<N>, C<v>, and C<V>, or you can use the C<E<gt>> and C<E<lt>>
3897 modifiers. These modifiers are only available as of perl 5.9.2.
3898 See also L<perlport>.
3902 All integer and floating point formats as well as C<p> and C<P> and
3903 C<()>-groups may be followed by the C<E<gt>> or C<E<lt>> modifiers
3904 to force big- or little- endian byte-order, respectively.
3905 This is especially useful, since C<n>, C<N>, C<v> and C<V> don't cover
3906 signed integers, 64-bit integers and floating point values. However,
3907 there are some things to keep in mind.
3909 Exchanging signed integers between different platforms only works
3910 if all platforms store them in the same format. Most platforms store
3911 signed integers in two's complement, so usually this is not an issue.
3913 The C<E<gt>> or C<E<lt>> modifiers can only be used on floating point
3914 formats on big- or little-endian machines. Otherwise, attempting to
3915 do so will result in a fatal error.
3917 Forcing big- or little-endian byte-order on floating point values for
3918 data exchange can only work if all platforms are using the same
3919 binary representation (e.g. IEEE floating point format). Even if all
3920 platforms are using IEEE, there may be subtle differences. Being able
3921 to use C<E<gt>> or C<E<lt>> on floating point values can be very useful,
3922 but also very dangerous if you don't know exactly what you're doing.
3923 It is definitely not a general way to portably store floating point
3926 When using C<E<gt>> or C<E<lt>> on an C<()>-group, this will affect
3927 all types inside the group that accept the byte-order modifiers,
3928 including all subgroups. It will silently be ignored for all other
3929 types. You are not allowed to override the byte-order within a group
3930 that already has a byte-order modifier suffix.
3934 Real numbers (floats and doubles) are in the native machine format only;
3935 due to the multiplicity of floating formats around, and the lack of a
3936 standard "network" representation, no facility for interchange has been
3937 made. This means that packed floating point data written on one machine
3938 may not be readable on another - even if both use IEEE floating point
3939 arithmetic (as the endian-ness of the memory representation is not part
3940 of the IEEE spec). See also L<perlport>.
3942 If you know exactly what you're doing, you can use the C<E<gt>> or C<E<lt>>
3943 modifiers to force big- or little-endian byte-order on floating point values.
3945 Note that Perl uses doubles (or long doubles, if configured) internally for
3946 all numeric calculation, and converting from double into float and thence back
3947 to double again will lose precision (i.e., C<unpack("f", pack("f", $foo)>)
3948 will not in general equal $foo).
3952 Pack and unpack can operate in two modes, character mode (C<C0> mode) where
3953 the packed string is processed per character and UTF-8 mode (C<U0> mode)
3954 where the packed string is processed in its UTF-8-encoded Unicode form on
3955 a byte by byte basis. Character mode is the default unless the format string
3956 starts with an C<U>. You can switch mode at any moment with an explicit
3957 C<C0> or C<U0> in the format. A mode is in effect until the next mode switch
3958 or until the end of the ()-group in which it was entered.
3962 You must yourself do any alignment or padding by inserting for example
3963 enough C<'x'>es while packing. There is no way to pack() and unpack()
3964 could know where the characters are going to or coming from. Therefore
3965 C<pack> (and C<unpack>) handle their output and input as flat
3966 sequences of characters.
3970 A ()-group is a sub-TEMPLATE enclosed in parentheses. A group may
3971 take a repeat count, both as postfix, and for unpack() also via the C</>
3972 template character. Within each repetition of a group, positioning with
3973 C<@> starts again at 0. Therefore, the result of
3975 pack( '@1A((@2A)@3A)', 'a', 'b', 'c' )
3977 is the string "\0a\0\0bc".
3981 C<x> and C<X> accept C<!> modifier. In this case they act as
3982 alignment commands: they jump forward/back to the closest position
3983 aligned at a multiple of C<count> characters. For example, to pack() or
3984 unpack() C's C<struct {char c; double d; char cc[2]}> one may need to
3985 use the template C<W x![d] d W[2]>; this assumes that doubles must be
3986 aligned on the double's size.
3988 For alignment commands C<count> of 0 is equivalent to C<count> of 1;
3989 both result in no-ops.
3993 C<n>, C<N>, C<v> and C<V> accept the C<!> modifier. In this case they
3994 will represent signed 16-/32-bit integers in big-/little-endian order.
3995 This is only portable if all platforms sharing the packed data use the
3996 same binary representation for signed integers (e.g. all platforms are
3997 using two's complement representation).
4001 A comment in a TEMPLATE starts with C<#> and goes to the end of line.
4002 White space may be used to separate pack codes from each other, but
4003 modifiers and a repeat count must follow immediately.
4007 If TEMPLATE requires more arguments to pack() than actually given, pack()
4008 assumes additional C<""> arguments. If TEMPLATE requires fewer arguments
4009 to pack() than actually given, extra arguments are ignored.
4015 $foo = pack("WWWW",65,66,67,68);
4017 $foo = pack("W4",65,66,67,68);
4019 $foo = pack("W4",0x24b6,0x24b7,0x24b8,0x24b9);
4020 # same thing with Unicode circled letters.
4021 $foo = pack("U4",0x24b6,0x24b7,0x24b8,0x24b9);
4022 # same thing with Unicode circled letters. You don't get the UTF-8
4023 # bytes because the U at the start of the format caused a switch to
4024 # U0-mode, so the UTF-8 bytes get joined into characters
4025 $foo = pack("C0U4",0x24b6,0x24b7,0x24b8,0x24b9);
4026 # foo eq "\xe2\x92\xb6\xe2\x92\xb7\xe2\x92\xb8\xe2\x92\xb9"
4027 # This is the UTF-8 encoding of the string in the previous example
4029 $foo = pack("ccxxcc",65,66,67,68);
4032 # note: the above examples featuring "W" and "c" are true
4033 # only on ASCII and ASCII-derived systems such as ISO Latin 1
4034 # and UTF-8. In EBCDIC the first example would be
4035 # $foo = pack("WWWW",193,194,195,196);
4037 $foo = pack("s2",1,2);
4038 # "\1\0\2\0" on little-endian
4039 # "\0\1\0\2" on big-endian
4041 $foo = pack("a4","abcd","x","y","z");
4044 $foo = pack("aaaa","abcd","x","y","z");
4047 $foo = pack("a14","abcdefg");
4048 # "abcdefg\0\0\0\0\0\0\0"
4050 $foo = pack("i9pl", gmtime);
4051 # a real struct tm (on my system anyway)
4053 $utmp_template = "Z8 Z8 Z16 L";
4054 $utmp = pack($utmp_template, @utmp1);
4055 # a struct utmp (BSDish)
4057 @utmp2 = unpack($utmp_template, $utmp);
4058 # "@utmp1" eq "@utmp2"
4061 unpack("N", pack("B32", substr("0" x 32 . shift, -32)));
4064 $foo = pack('sx2l', 12, 34);
4065 # short 12, two zero bytes padding, long 34
4066 $bar = pack('s@4l', 12, 34);
4067 # short 12, zero fill to position 4, long 34
4069 $baz = pack('s.l', 12, 4, 34);
4070 # short 12, zero fill to position 4, long 34
4072 $foo = pack('nN', 42, 4711);
4073 # pack big-endian 16- and 32-bit unsigned integers
4074 $foo = pack('S>L>', 42, 4711);
4076 $foo = pack('s<l<', -42, 4711);
4077 # pack little-endian 16- and 32-bit signed integers
4078 $foo = pack('(sl)<', -42, 4711);
4081 The same template may generally also be used in unpack().
4083 =item package NAMESPACE VERSION
4084 X<package> X<module> X<namespace> X<version>
4086 =item package NAMESPACE
4088 Declares the compilation unit as being in the given namespace. The scope
4089 of the package declaration is from the declaration itself through the end
4090 of the enclosing block, file, or eval (the same as the C<my> operator).
4091 All further unqualified dynamic identifiers will be in this namespace.
4092 A package statement affects only dynamic variables--including those
4093 you've used C<local> on--but I<not> lexical variables, which are created
4094 with C<my>. Typically it would be the first declaration in a file to
4095 be included by the C<require> or C<use> operator. You can switch into a
4096 package in more than one place; it merely influences which symbol table
4097 is used by the compiler for the rest of that block. You can refer to
4098 variables and filehandles in other packages by prefixing the identifier
4099 with the package name and a double colon: C<$Package::Variable>.
4100 If the package name is null, the C<main> package as assumed. That is,
4101 C<$::sail> is equivalent to C<$main::sail> (as well as to C<$main'sail>,
4102 still seen in older code).
4104 If VERSION is provided, C<package> also sets the C<$VERSION> variable in the
4105 given namespace. VERSION must be be a numeric literal or v-string; it is
4106 parsed exactly the same way as a VERSION argument to C<use MODULE VERSION>.
4107 C<$VERSION> should only be set once per package.
4109 See L<perlmod/"Packages"> for more information about packages, modules,
4110 and classes. See L<perlsub> for other scoping issues.
4112 =item pipe READHANDLE,WRITEHANDLE
4115 Opens a pair of connected pipes like the corresponding system call.
4116 Note that if you set up a loop of piped processes, deadlock can occur
4117 unless you are very careful. In addition, note that Perl's pipes use
4118 IO buffering, so you may need to set C<$|> to flush your WRITEHANDLE
4119 after each command, depending on the application.
4121 See L<IPC::Open2>, L<IPC::Open3>, and L<perlipc/"Bidirectional Communication">
4122 for examples of such things.
4124 On systems that support a close-on-exec flag on files, the flag will be set
4125 for the newly opened file descriptors as determined by the value of $^F.
4133 Pops and returns the last value of the array, shortening the array by
4136 If there are no elements in the array, returns the undefined value
4137 (although this may happen at other times as well). If ARRAY is
4138 omitted, pops the C<@ARGV> array in the main program, and the C<@_>
4139 array in subroutines, just like C<shift>.
4142 X<pos> X<match, position>
4146 Returns the offset of where the last C<m//g> search left off for the variable
4147 in question (C<$_> is used when the variable is not specified). Note that
4148 0 is a valid match offset. C<undef> indicates that the search position
4149 is reset (usually due to match failure, but can also be because no match has
4150 yet been performed on the scalar). C<pos> directly accesses the location used
4151 by the regexp engine to store the offset, so assigning to C<pos> will change
4152 that offset, and so will also influence the C<\G> zero-width assertion in
4153 regular expressions. Because a failed C<m//gc> match doesn't reset the offset,
4154 the return from C<pos> won't change either in this case. See L<perlre> and
4157 =item print FILEHANDLE LIST
4164 Prints a string or a list of strings. Returns true if successful.
4165 FILEHANDLE may be a scalar variable name, in which case the variable
4166 contains the name of or a reference to the filehandle, thus introducing
4167 one level of indirection. (NOTE: If FILEHANDLE is a variable and
4168 the next token is a term, it may be misinterpreted as an operator
4169 unless you interpose a C<+> or put parentheses around the arguments.)
4170 If FILEHANDLE is omitted, prints by default to standard output (or
4171 to the last selected output channel--see L</select>). If LIST is
4172 also omitted, prints C<$_> to the currently selected output channel.
4173 To set the default output channel to something other than STDOUT
4174 use the select operation. The current value of C<$,> (if any) is
4175 printed between each LIST item. The current value of C<$\> (if
4176 any) is printed after the entire LIST has been printed. Because
4177 print takes a LIST, anything in the LIST is evaluated in list
4178 context, and any subroutine that you call will have one or more of
4179 its expressions evaluated in list context. Also be careful not to
4180 follow the print keyword with a left parenthesis unless you want
4181 the corresponding right parenthesis to terminate the arguments to
4182 the print--interpose a C<+> or put parentheses around all the
4185 Note that if you're storing FILEHANDLEs in an array, or if you're using
4186 any other expression more complex than a scalar variable to retrieve it,
4187 you will have to use a block returning the filehandle value instead:
4189 print { $files[$i] } "stuff\n";
4190 print { $OK ? STDOUT : STDERR } "stuff\n";
4192 =item printf FILEHANDLE FORMAT, LIST
4195 =item printf FORMAT, LIST
4197 Equivalent to C<print FILEHANDLE sprintf(FORMAT, LIST)>, except that C<$\>
4198 (the output record separator) is not appended. The first argument
4199 of the list will be interpreted as the C<printf> format. See C<sprintf>
4200 for an explanation of the format argument. If C<use locale> is in effect,
4201 and POSIX::setlocale() has been called, the character used for the decimal
4202 separator in formatted floating point numbers is affected by the LC_NUMERIC
4203 locale. See L<perllocale> and L<POSIX>.
4205 Don't fall into the trap of using a C<printf> when a simple
4206 C<print> would do. The C<print> is more efficient and less
4209 =item prototype FUNCTION
4212 Returns the prototype of a function as a string (or C<undef> if the
4213 function has no prototype). FUNCTION is a reference to, or the name of,
4214 the function whose prototype you want to retrieve.
4216 If FUNCTION is a string starting with C<CORE::>, the rest is taken as a
4217 name for Perl builtin. If the builtin is not I<overridable> (such as
4218 C<qw//>) or if its arguments cannot be adequately expressed by a prototype
4219 (such as C<system>), prototype() returns C<undef>, because the builtin
4220 does not really behave like a Perl function. Otherwise, the string
4221 describing the equivalent prototype is returned.
4223 =item push ARRAY,LIST
4226 Treats ARRAY as a stack, and pushes the values of LIST
4227 onto the end of ARRAY. The length of ARRAY increases by the length of
4228 LIST. Has the same effect as
4231 $ARRAY[++$#ARRAY] = $value;
4234 but is more efficient. Returns the number of elements in the array following
4235 the completed C<push>.
4245 Generalized quotes. See L<perlop/"Quote-Like Operators">.
4249 Regexp-like quote. See L<perlop/"Regexp Quote-Like Operators">.
4251 =item quotemeta EXPR
4252 X<quotemeta> X<metacharacter>
4256 Returns the value of EXPR with all non-"word"
4257 characters backslashed. (That is, all characters not matching
4258 C</[A-Za-z_0-9]/> will be preceded by a backslash in the
4259 returned string, regardless of any locale settings.)
4260 This is the internal function implementing
4261 the C<\Q> escape in double-quoted strings.
4263 If EXPR is omitted, uses C<$_>.
4270 Returns a random fractional number greater than or equal to C<0> and less
4271 than the value of EXPR. (EXPR should be positive.) If EXPR is
4272 omitted, the value C<1> is used. Currently EXPR with the value C<0> is
4273 also special-cased as C<1> - this has not been documented before perl 5.8.0
4274 and is subject to change in future versions of perl. Automatically calls
4275 C<srand> unless C<srand> has already been called. See also C<srand>.
4277 Apply C<int()> to the value returned by C<rand()> if you want random
4278 integers instead of random fractional numbers. For example,
4282 returns a random integer between C<0> and C<9>, inclusive.
4284 (Note: If your rand function consistently returns numbers that are too
4285 large or too small, then your version of Perl was probably compiled
4286 with the wrong number of RANDBITS.)
4288 =item read FILEHANDLE,SCALAR,LENGTH,OFFSET
4289 X<read> X<file, read>
4291 =item read FILEHANDLE,SCALAR,LENGTH
4293 Attempts to read LENGTH I<characters> of data into variable SCALAR
4294 from the specified FILEHANDLE. Returns the number of characters
4295 actually read, C<0> at end of file, or undef if there was an error (in
4296 the latter case C<$!> is also set). SCALAR will be grown or shrunk
4297 so that the last character actually read is the last character of the
4298 scalar after the read.
4300 An OFFSET may be specified to place the read data at some place in the
4301 string other than the beginning. A negative OFFSET specifies
4302 placement at that many characters counting backwards from the end of
4303 the string. A positive OFFSET greater than the length of SCALAR
4304 results in the string being padded to the required size with C<"\0">
4305 bytes before the result of the read is appended.
4307 The call is actually implemented in terms of either Perl's or system's
4308 fread() call. To get a true read(2) system call, see C<sysread>.
4310 Note the I<characters>: depending on the status of the filehandle,
4311 either (8-bit) bytes or characters are read. By default all
4312 filehandles operate on bytes, but for example if the filehandle has
4313 been opened with the C<:utf8> I/O layer (see L</open>, and the C<open>
4314 pragma, L<open>), the I/O will operate on UTF-8 encoded Unicode
4315 characters, not bytes. Similarly for the C<:encoding> pragma:
4316 in that case pretty much any characters can be read.
4318 =item readdir DIRHANDLE
4321 Returns the next directory entry for a directory opened by C<opendir>.
4322 If used in list context, returns all the rest of the entries in the
4323 directory. If there are no more entries, returns an undefined value in
4324 scalar context or a null list in list context.
4326 If you're planning to filetest the return values out of a C<readdir>, you'd
4327 better prepend the directory in question. Otherwise, because we didn't
4328 C<chdir> there, it would have been testing the wrong file.
4330 opendir(my $dh, $some_dir) || die "can't opendir $some_dir: $!";
4331 @dots = grep { /^\./ && -f "$some_dir/$_" } readdir($dh);
4337 X<readline> X<gets> X<fgets>
4339 Reads from the filehandle whose typeglob is contained in EXPR (or from
4340 *ARGV if EXPR is not provided). In scalar context, each call reads and
4341 returns the next line, until end-of-file is reached, whereupon the
4342 subsequent call returns C<undef>. In list context, reads until end-of-file
4343 is reached and returns a list of lines. Note that the notion of "line"
4344 used here is however you may have defined it with C<$/> or
4345 C<$INPUT_RECORD_SEPARATOR>). See L<perlvar/"$/">.
4347 When C<$/> is set to C<undef>, when C<readline> is in scalar
4348 context (i.e. file slurp mode), and when an empty file is read, it
4349 returns C<''> the first time, followed by C<undef> subsequently.
4351 This is the internal function implementing the C<< <EXPR> >>
4352 operator, but you can use it directly. The C<< <EXPR> >>
4353 operator is discussed in more detail in L<perlop/"I/O Operators">.
4356 $line = readline(*STDIN); # same thing
4358 If C<readline> encounters an operating system error, C<$!> will be set
4359 with the corresponding error message. It can be helpful to check
4360 C<$!> when you are reading from filehandles you don't trust, such as a
4361 tty or a socket. The following example uses the operator form of
4362 C<readline> and dies if the result is not defined.
4364 while ( ! eof($fh) ) {
4365 defined( $_ = <$fh> ) or die "readline failed: $!";
4369 Note that you have can't handle C<readline> errors that way with the
4370 C<ARGV> filehandle. In that case, you have to open each element of
4371 C<@ARGV> yourself since C<eof> handles C<ARGV> differently.
4373 foreach my $arg (@ARGV) {
4374 open(my $fh, $arg) or warn "Can't open $arg: $!";
4376 while ( ! eof($fh) ) {
4377 defined( $_ = <$fh> )
4378 or die "readline failed for $arg: $!";
4388 Returns the value of a symbolic link, if symbolic links are
4389 implemented. If not, gives a fatal error. If there is some system
4390 error, returns the undefined value and sets C<$!> (errno). If EXPR is
4391 omitted, uses C<$_>.
4398 EXPR is executed as a system command.
4399 The collected standard output of the command is returned.
4400 In scalar context, it comes back as a single (potentially
4401 multi-line) string. In list context, returns a list of lines
4402 (however you've defined lines with C<$/> or C<$INPUT_RECORD_SEPARATOR>).
4403 This is the internal function implementing the C<qx/EXPR/>
4404 operator, but you can use it directly. The C<qx/EXPR/>
4405 operator is discussed in more detail in L<perlop/"I/O Operators">.
4406 If EXPR is omitted, uses C<$_>.
4408 =item recv SOCKET,SCALAR,LENGTH,FLAGS
4411 Receives a message on a socket. Attempts to receive LENGTH characters
4412 of data into variable SCALAR from the specified SOCKET filehandle.
4413 SCALAR will be grown or shrunk to the length actually read. Takes the
4414 same flags as the system call of the same name. Returns the address
4415 of the sender if SOCKET's protocol supports this; returns an empty
4416 string otherwise. If there's an error, returns the undefined value.
4417 This call is actually implemented in terms of recvfrom(2) system call.
4418 See L<perlipc/"UDP: Message Passing"> for examples.
4420 Note the I<characters>: depending on the status of the socket, either
4421 (8-bit) bytes or characters are received. By default all sockets
4422 operate on bytes, but for example if the socket has been changed using
4423 binmode() to operate with the C<:encoding(utf8)> I/O layer (see the
4424 C<open> pragma, L<open>), the I/O will operate on UTF-8 encoded Unicode
4425 characters, not bytes. Similarly for the C<:encoding> pragma: in that
4426 case pretty much any characters can be read.
4433 The C<redo> command restarts the loop block without evaluating the
4434 conditional again. The C<continue> block, if any, is not executed. If
4435 the LABEL is omitted, the command refers to the innermost enclosing
4436 loop. Programs that want to lie to themselves about what was just input
4437 normally use this command:
4439 # a simpleminded Pascal comment stripper
4440 # (warning: assumes no { or } in strings)
4441 LINE: while (<STDIN>) {
4442 while (s|({.*}.*){.*}|$1 |) {}
4447 if (/}/) { # end of comment?
4456 C<redo> cannot be used to retry a block which returns a value such as
4457 C<eval {}>, C<sub {}> or C<do {}>, and should not be used to exit
4458 a grep() or map() operation.
4460 Note that a block by itself is semantically identical to a loop
4461 that executes once. Thus C<redo> inside such a block will effectively
4462 turn it into a looping construct.
4464 See also L</continue> for an illustration of how C<last>, C<next>, and
4472 Returns a non-empty string if EXPR is a reference, the empty
4473 string otherwise. If EXPR
4474 is not specified, C<$_> will be used. The value returned depends on the
4475 type of thing the reference is a reference to.
4476 Builtin types include:
4490 If the referenced object has been blessed into a package, then that package
4491 name is returned instead. You can think of C<ref> as a C<typeof> operator.
4493 if (ref($r) eq "HASH") {
4494 print "r is a reference to a hash.\n";
4497 print "r is not a reference at all.\n";
4500 The return value C<LVALUE> indicates a reference to an lvalue that is not
4501 a variable. You get this from taking the reference of function calls like
4502 C<pos()> or C<substr()>. C<VSTRING> is returned if the reference points
4503 to a L<version string|perldata/"Version Strings">.
4505 The result C<Regexp> indicates that the argument is a regular expression
4506 resulting from C<qr//>.
4508 See also L<perlref>.
4510 =item rename OLDNAME,NEWNAME
4511 X<rename> X<move> X<mv> X<ren>
4513 Changes the name of a file; an existing file NEWNAME will be
4514 clobbered. Returns true for success, false otherwise.
4516 Behavior of this function varies wildly depending on your system
4517 implementation. For example, it will usually not work across file system
4518 boundaries, even though the system I<mv> command sometimes compensates
4519 for this. Other restrictions include whether it works on directories,
4520 open files, or pre-existing files. Check L<perlport> and either the
4521 rename(2) manpage or equivalent system documentation for details.
4523 For a platform independent C<move> function look at the L<File::Copy>
4526 =item require VERSION
4533 Demands a version of Perl specified by VERSION, or demands some semantics
4534 specified by EXPR or by C<$_> if EXPR is not supplied.
4536 VERSION may be either a numeric argument such as 5.006, which will be
4537 compared to C<$]>, or a literal of the form v5.6.1, which will be compared
4538 to C<$^V> (aka $PERL_VERSION). A fatal error is produced at run time if
4539 VERSION is greater than the version of the current Perl interpreter.
4540 Compare with L</use>, which can do a similar check at compile time.
4542 Specifying VERSION as a literal of the form v5.6.1 should generally be
4543 avoided, because it leads to misleading error messages under earlier
4544 versions of Perl that do not support this syntax. The equivalent numeric
4545 version should be used instead.
4547 require v5.6.1; # run time version check
4548 require 5.6.1; # ditto
4549 require 5.006_001; # ditto; preferred for backwards compatibility
4551 Otherwise, C<require> demands that a library file be included if it
4552 hasn't already been included. The file is included via the do-FILE
4553 mechanism, which is essentially just a variety of C<eval> with the
4554 caveat that lexical variables in the invoking script will be invisible
4555 to the included code. Has semantics similar to the following subroutine:
4558 my ($filename) = @_;
4559 if (exists $INC{$filename}) {
4560 return 1 if $INC{$filename};
4561 die "Compilation failed in require";
4563 my ($realfilename,$result);
4565 foreach $prefix (@INC) {
4566 $realfilename = "$prefix/$filename";
4567 if (-f $realfilename) {
4568 $INC{$filename} = $realfilename;
4569 $result = do $realfilename;
4573 die "Can't find $filename in \@INC";
4576 $INC{$filename} = undef;
4578 } elsif (!$result) {
4579 delete $INC{$filename};
4580 die "$filename did not return true value";
4586 Note that the file will not be included twice under the same specified
4589 The file must return true as the last statement to indicate
4590 successful execution of any initialization code, so it's customary to
4591 end such a file with C<1;> unless you're sure it'll return true
4592 otherwise. But it's better just to put the C<1;>, in case you add more
4595 If EXPR is a bareword, the require assumes a "F<.pm>" extension and
4596 replaces "F<::>" with "F</>" in the filename for you,
4597 to make it easy to load standard modules. This form of loading of
4598 modules does not risk altering your namespace.
4600 In other words, if you try this:
4602 require Foo::Bar; # a splendid bareword
4604 The require function will actually look for the "F<Foo/Bar.pm>" file in the
4605 directories specified in the C<@INC> array.
4607 But if you try this:
4609 $class = 'Foo::Bar';
4610 require $class; # $class is not a bareword
4612 require "Foo::Bar"; # not a bareword because of the ""
4614 The require function will look for the "F<Foo::Bar>" file in the @INC array and
4615 will complain about not finding "F<Foo::Bar>" there. In this case you can do:
4617 eval "require $class";
4619 Now that you understand how C<require> looks for files in the case of a
4620 bareword argument, there is a little extra functionality going on behind
4621 the scenes. Before C<require> looks for a "F<.pm>" extension, it will
4622 first look for a similar filename with a "F<.pmc>" extension. If this file
4623 is found, it will be loaded in place of any file ending in a "F<.pm>"
4626 You can also insert hooks into the import facility, by putting directly
4627 Perl code into the @INC array. There are three forms of hooks: subroutine
4628 references, array references and blessed objects.
4630 Subroutine references are the simplest case. When the inclusion system
4631 walks through @INC and encounters a subroutine, this subroutine gets
4632 called with two parameters, the first being a reference to itself, and the
4633 second the name of the file to be included (e.g. "F<Foo/Bar.pm>"). The
4634 subroutine should return nothing, or a list of up to three values in the
4641 A filehandle, from which the file will be read.
4645 A reference to a subroutine. If there is no filehandle (previous item),
4646 then this subroutine is expected to generate one line of source code per
4647 call, writing the line into C<$_> and returning 1, then returning 0 at
4648 "end of file". If there is a filehandle, then the subroutine will be
4649 called to act as a simple source filter, with the line as read in C<$_>.
4650 Again, return 1 for each valid line, and 0 after all lines have been
4655 Optional state for the subroutine. The state is passed in as C<$_[1]>. A
4656 reference to the subroutine itself is passed in as C<$_[0]>.
4660 If an empty list, C<undef>, or nothing that matches the first 3 values above
4661 is returned then C<require> will look at the remaining elements of @INC.
4662 Note that this file handle must be a real file handle (strictly a typeglob,
4663 or reference to a typeglob, blessed or unblessed) - tied file handles will be
4664 ignored and return value processing will stop there.
4666 If the hook is an array reference, its first element must be a subroutine
4667 reference. This subroutine is called as above, but the first parameter is
4668 the array reference. This enables to pass indirectly some arguments to
4671 In other words, you can write:
4673 push @INC, \&my_sub;
4675 my ($coderef, $filename) = @_; # $coderef is \&my_sub
4681 push @INC, [ \&my_sub, $x, $y, ... ];
4683 my ($arrayref, $filename) = @_;
4684 # Retrieve $x, $y, ...
4685 my @parameters = @$arrayref[1..$#$arrayref];
4689 If the hook is an object, it must provide an INC method that will be
4690 called as above, the first parameter being the object itself. (Note that
4691 you must fully qualify the sub's name, as unqualified C<INC> is always forced
4692 into package C<main>.) Here is a typical code layout:
4698 my ($self, $filename) = @_;
4702 # In the main program
4703 push @INC, Foo->new(...);
4705 Note that these hooks are also permitted to set the %INC entry
4706 corresponding to the files they have loaded. See L<perlvar/%INC>.
4708 For a yet-more-powerful import facility, see L</use> and L<perlmod>.
4715 Generally used in a C<continue> block at the end of a loop to clear
4716 variables and reset C<??> searches so that they work again. The
4717 expression is interpreted as a list of single characters (hyphens
4718 allowed for ranges). All variables and arrays beginning with one of
4719 those letters are reset to their pristine state. If the expression is
4720 omitted, one-match searches (C<?pattern?>) are reset to match again. Resets
4721 only variables or searches in the current package. Always returns
4724 reset 'X'; # reset all X variables
4725 reset 'a-z'; # reset lower case variables
4726 reset; # just reset ?one-time? searches
4728 Resetting C<"A-Z"> is not recommended because you'll wipe out your
4729 C<@ARGV> and C<@INC> arrays and your C<%ENV> hash. Resets only package
4730 variables--lexical variables are unaffected, but they clean themselves
4731 up on scope exit anyway, so you'll probably want to use them instead.
4739 Returns from a subroutine, C<eval>, or C<do FILE> with the value
4740 given in EXPR. Evaluation of EXPR may be in list, scalar, or void
4741 context, depending on how the return value will be used, and the context
4742 may vary from one execution to the next (see C<wantarray>). If no EXPR
4743 is given, returns an empty list in list context, the undefined value in
4744 scalar context, and (of course) nothing at all in a void context.
4746 (Note that in the absence of an explicit C<return>, a subroutine, eval,
4747 or do FILE will automatically return the value of the last expression
4751 X<reverse> X<rev> X<invert>
4753 In list context, returns a list value consisting of the elements
4754 of LIST in the opposite order. In scalar context, concatenates the
4755 elements of LIST and returns a string value with all characters
4756 in the opposite order.
4758 print join(", ", reverse "world", "Hello"); # Hello, world
4760 print scalar reverse "dlrow ,", "olleH"; # Hello, world
4762 Used without arguments in scalar context, reverse() reverses C<$_>.
4764 $_ = "dlrow ,olleH";
4765 print reverse; # No output, list context
4766 print scalar reverse; # Hello, world
4768 This operator is also handy for inverting a hash, although there are some
4769 caveats. If a value is duplicated in the original hash, only one of those
4770 can be represented as a key in the inverted hash. Also, this has to
4771 unwind one hash and build a whole new one, which may take some time
4772 on a large hash, such as from a DBM file.
4774 %by_name = reverse %by_address; # Invert the hash
4776 =item rewinddir DIRHANDLE
4779 Sets the current position to the beginning of the directory for the
4780 C<readdir> routine on DIRHANDLE.
4782 =item rindex STR,SUBSTR,POSITION
4785 =item rindex STR,SUBSTR
4787 Works just like index() except that it returns the position of the I<last>
4788 occurrence of SUBSTR in STR. If POSITION is specified, returns the
4789 last occurrence beginning at or before that position.
4791 =item rmdir FILENAME
4792 X<rmdir> X<rd> X<directory, remove>
4796 Deletes the directory specified by FILENAME if that directory is
4797 empty. If it succeeds it returns true, otherwise it returns false and
4798 sets C<$!> (errno). If FILENAME is omitted, uses C<$_>.
4800 To remove a directory tree recursively (C<rm -rf> on unix) look at
4801 the C<rmtree> function of the L<File::Path> module.
4805 The substitution operator. See L<perlop/"Regexp Quote-Like Operators">.
4807 =item say FILEHANDLE LIST
4814 Just like C<print>, but implicitly appends a newline.
4815 C<say LIST> is simply an abbreviation for C<{ local $\ = "\n"; print
4818 This keyword is only available when the "say" feature is
4819 enabled: see L<feature>.
4822 X<scalar> X<context>
4824 Forces EXPR to be interpreted in scalar context and returns the value
4827 @counts = ( scalar @a, scalar @b, scalar @c );
4829 There is no equivalent operator to force an expression to
4830 be interpolated in list context because in practice, this is never
4831 needed. If you really wanted to do so, however, you could use
4832 the construction C<@{[ (some expression) ]}>, but usually a simple
4833 C<(some expression)> suffices.
4835 Because C<scalar> is unary operator, if you accidentally use for EXPR a
4836 parenthesized list, this behaves as a scalar comma expression, evaluating
4837 all but the last element in void context and returning the final element
4838 evaluated in scalar context. This is seldom what you want.
4840 The following single statement:
4842 print uc(scalar(&foo,$bar)),$baz;
4844 is the moral equivalent of these two:
4847 print(uc($bar),$baz);
4849 See L<perlop> for more details on unary operators and the comma operator.
4851 =item seek FILEHANDLE,POSITION,WHENCE
4852 X<seek> X<fseek> X<filehandle, position>
4854 Sets FILEHANDLE's position, just like the C<fseek> call of C<stdio>.
4855 FILEHANDLE may be an expression whose value gives the name of the
4856 filehandle. The values for WHENCE are C<0> to set the new position
4857 I<in bytes> to POSITION, C<1> to set it to the current position plus
4858 POSITION, and C<2> to set it to EOF plus POSITION (typically
4859 negative). For WHENCE you may use the constants C<SEEK_SET>,
4860 C<SEEK_CUR>, and C<SEEK_END> (start of the file, current position, end
4861 of the file) from the Fcntl module. Returns C<1> upon success, C<0>
4864 Note the I<in bytes>: even if the filehandle has been set to
4865 operate on characters (for example by using the C<:encoding(utf8)> open
4866 layer), tell() will return byte offsets, not character offsets
4867 (because implementing that would render seek() and tell() rather slow).
4869 If you want to position file for C<sysread> or C<syswrite>, don't use
4870 C<seek>--buffering makes its effect on the file's system position
4871 unpredictable and non-portable. Use C<sysseek> instead.
4873 Due to the rules and rigors of ANSI C, on some systems you have to do a
4874 seek whenever you switch between reading and writing. Amongst other
4875 things, this may have the effect of calling stdio's clearerr(3).
4876 A WHENCE of C<1> (C<SEEK_CUR>) is useful for not moving the file position:
4880 This is also useful for applications emulating C<tail -f>. Once you hit
4881 EOF on your read, and then sleep for a while, you might have to stick in a
4882 seek() to reset things. The C<seek> doesn't change the current position,
4883 but it I<does> clear the end-of-file condition on the handle, so that the
4884 next C<< <FILE> >> makes Perl try again to read something. We hope.
4886 If that doesn't work (some IO implementations are particularly
4887 cantankerous), then you may need something more like this:
4890 for ($curpos = tell(FILE); $_ = <FILE>;
4891 $curpos = tell(FILE)) {
4892 # search for some stuff and put it into files
4894 sleep($for_a_while);
4895 seek(FILE, $curpos, 0);
4898 =item seekdir DIRHANDLE,POS
4901 Sets the current position for the C<readdir> routine on DIRHANDLE. POS
4902 must be a value returned by C<telldir>. C<seekdir> also has the same caveats
4903 about possible directory compaction as the corresponding system library
4906 =item select FILEHANDLE
4907 X<select> X<filehandle, default>
4911 Returns the currently selected filehandle. If FILEHANDLE is supplied,
4912 sets the new current default filehandle for output. This has two
4913 effects: first, a C<write> or a C<print> without a filehandle will
4914 default to this FILEHANDLE. Second, references to variables related to
4915 output will refer to this output channel. For example, if you have to
4916 set the top of form format for more than one output channel, you might
4924 FILEHANDLE may be an expression whose value gives the name of the
4925 actual filehandle. Thus:
4927 $oldfh = select(STDERR); $| = 1; select($oldfh);
4929 Some programmers may prefer to think of filehandles as objects with
4930 methods, preferring to write the last example as:
4933 STDERR->autoflush(1);
4935 =item select RBITS,WBITS,EBITS,TIMEOUT
4938 This calls the select(2) system call with the bit masks specified, which
4939 can be constructed using C<fileno> and C<vec>, along these lines:
4941 $rin = $win = $ein = '';
4942 vec($rin,fileno(STDIN),1) = 1;
4943 vec($win,fileno(STDOUT),1) = 1;
4946 If you want to select on many filehandles you might wish to write a
4950 my(@fhlist) = split(' ',$_[0]);
4953 vec($bits,fileno($_),1) = 1;
4957 $rin = fhbits('STDIN TTY SOCK');
4961 ($nfound,$timeleft) =
4962 select($rout=$rin, $wout=$win, $eout=$ein, $timeout);
4964 or to block until something becomes ready just do this
4966 $nfound = select($rout=$rin, $wout=$win, $eout=$ein, undef);
4968 Most systems do not bother to return anything useful in $timeleft, so
4969 calling select() in scalar context just returns $nfound.
4971 Any of the bit masks can also be undef. The timeout, if specified, is
4972 in seconds, which may be fractional. Note: not all implementations are
4973 capable of returning the $timeleft. If not, they always return
4974 $timeleft equal to the supplied $timeout.
4976 You can effect a sleep of 250 milliseconds this way:
4978 select(undef, undef, undef, 0.25);
4980 Note that whether C<select> gets restarted after signals (say, SIGALRM)
4981 is implementation-dependent. See also L<perlport> for notes on the
4982 portability of C<select>.
4984 On error, C<select> behaves like the select(2) system call : it returns
4987 Note: on some Unixes, the select(2) system call may report a socket file
4988 descriptor as "ready for reading", when actually no data is available,
4989 thus a subsequent read blocks. It can be avoided using always the
4990 O_NONBLOCK flag on the socket. See select(2) and fcntl(2) for further
4993 B<WARNING>: One should not attempt to mix buffered I/O (like C<read>
4994 or <FH>) with C<select>, except as permitted by POSIX, and even
4995 then only on POSIX systems. You have to use C<sysread> instead.
4997 =item semctl ID,SEMNUM,CMD,ARG
5000 Calls the System V IPC function C<semctl>. You'll probably have to say
5004 first to get the correct constant definitions. If CMD is IPC_STAT or
5005 GETALL, then ARG must be a variable that will hold the returned
5006 semid_ds structure or semaphore value array. Returns like C<ioctl>:
5007 the undefined value for error, "C<0 but true>" for zero, or the actual
5008 return value otherwise. The ARG must consist of a vector of native
5009 short integers, which may be created with C<pack("s!",(0)x$nsem)>.
5010 See also L<perlipc/"SysV IPC">, C<IPC::SysV>, C<IPC::Semaphore>
5013 =item semget KEY,NSEMS,FLAGS
5016 Calls the System V IPC function semget. Returns the semaphore id, or
5017 the undefined value if there is an error. See also
5018 L<perlipc/"SysV IPC">, C<IPC::SysV>, C<IPC::SysV::Semaphore>
5021 =item semop KEY,OPSTRING
5024 Calls the System V IPC function semop to perform semaphore operations
5025 such as signalling and waiting. OPSTRING must be a packed array of
5026 semop structures. Each semop structure can be generated with
5027 C<pack("s!3", $semnum, $semop, $semflag)>. The length of OPSTRING
5028 implies the number of semaphore operations. Returns true if
5029 successful, or false if there is an error. As an example, the
5030 following code waits on semaphore $semnum of semaphore id $semid:
5032 $semop = pack("s!3", $semnum, -1, 0);
5033 die "Semaphore trouble: $!\n" unless semop($semid, $semop);
5035 To signal the semaphore, replace C<-1> with C<1>. See also
5036 L<perlipc/"SysV IPC">, C<IPC::SysV>, and C<IPC::SysV::Semaphore>
5039 =item send SOCKET,MSG,FLAGS,TO
5042 =item send SOCKET,MSG,FLAGS
5044 Sends a message on a socket. Attempts to send the scalar MSG to the
5045 SOCKET filehandle. Takes the same flags as the system call of the
5046 same name. On unconnected sockets you must specify a destination to
5047 send TO, in which case it does a C C<sendto>. Returns the number of
5048 characters sent, or the undefined value if there is an error. The C
5049 system call sendmsg(2) is currently unimplemented. See
5050 L<perlipc/"UDP: Message Passing"> for examples.
5052 Note the I<characters>: depending on the status of the socket, either
5053 (8-bit) bytes or characters are sent. By default all sockets operate
5054 on bytes, but for example if the socket has been changed using
5055 binmode() to operate with the C<:encoding(utf8)> I/O layer (see
5056 L</open>, or the C<open> pragma, L<open>), the I/O will operate on UTF-8
5057 encoded Unicode characters, not bytes. Similarly for the C<:encoding>
5058 pragma: in that case pretty much any characters can be sent.
5060 =item setpgrp PID,PGRP
5063 Sets the current process group for the specified PID, C<0> for the current
5064 process. Will produce a fatal error if used on a machine that doesn't
5065 implement POSIX setpgid(2) or BSD setpgrp(2). If the arguments are omitted,
5066 it defaults to C<0,0>. Note that the BSD 4.2 version of C<setpgrp> does not
5067 accept any arguments, so only C<setpgrp(0,0)> is portable. See also
5070 =item setpriority WHICH,WHO,PRIORITY
5071 X<setpriority> X<priority> X<nice> X<renice>
5073 Sets the current priority for a process, a process group, or a user.
5074 (See setpriority(2).) Will produce a fatal error if used on a machine
5075 that doesn't implement setpriority(2).
5077 =item setsockopt SOCKET,LEVEL,OPTNAME,OPTVAL
5080 Sets the socket option requested. Returns undefined if there is an
5081 error. Use integer constants provided by the C<Socket> module for
5082 LEVEL and OPNAME. Values for LEVEL can also be obtained from
5083 getprotobyname. OPTVAL might either be a packed string or an integer.
5084 An integer OPTVAL is shorthand for pack("i", OPTVAL).
5086 An example disabling the Nagle's algorithm for a socket:
5088 use Socket qw(IPPROTO_TCP TCP_NODELAY);
5089 setsockopt($socket, IPPROTO_TCP, TCP_NODELAY, 1);
5096 Shifts the first value of the array off and returns it, shortening the
5097 array by 1 and moving everything down. If there are no elements in the
5098 array, returns the undefined value. If ARRAY is omitted, shifts the
5099 C<@_> array within the lexical scope of subroutines and formats, and the
5100 C<@ARGV> array outside of a subroutine and also within the lexical scopes
5101 established by the C<eval STRING>, C<BEGIN {}>, C<INIT {}>, C<CHECK {}>,
5102 C<UNITCHECK {}> and C<END {}> constructs.
5104 See also C<unshift>, C<push>, and C<pop>. C<shift> and C<unshift> do the
5105 same thing to the left end of an array that C<pop> and C<push> do to the
5108 =item shmctl ID,CMD,ARG
5111 Calls the System V IPC function shmctl. You'll probably have to say
5115 first to get the correct constant definitions. If CMD is C<IPC_STAT>,
5116 then ARG must be a variable that will hold the returned C<shmid_ds>
5117 structure. Returns like ioctl: the undefined value for error, "C<0> but
5118 true" for zero, or the actual return value otherwise.
5119 See also L<perlipc/"SysV IPC"> and C<IPC::SysV> documentation.
5121 =item shmget KEY,SIZE,FLAGS
5124 Calls the System V IPC function shmget. Returns the shared memory
5125 segment id, or the undefined value if there is an error.
5126 See also L<perlipc/"SysV IPC"> and C<IPC::SysV> documentation.
5128 =item shmread ID,VAR,POS,SIZE
5132 =item shmwrite ID,STRING,POS,SIZE
5134 Reads or writes the System V shared memory segment ID starting at
5135 position POS for size SIZE by attaching to it, copying in/out, and
5136 detaching from it. When reading, VAR must be a variable that will
5137 hold the data read. When writing, if STRING is too long, only SIZE
5138 bytes are used; if STRING is too short, nulls are written to fill out
5139 SIZE bytes. Return true if successful, or false if there is an error.
5140 shmread() taints the variable. See also L<perlipc/"SysV IPC">,
5141 C<IPC::SysV> documentation, and the C<IPC::Shareable> module from CPAN.
5143 =item shutdown SOCKET,HOW
5146 Shuts down a socket connection in the manner indicated by HOW, which
5147 has the same interpretation as in the system call of the same name.
5149 shutdown(SOCKET, 0); # I/we have stopped reading data
5150 shutdown(SOCKET, 1); # I/we have stopped writing data
5151 shutdown(SOCKET, 2); # I/we have stopped using this socket
5153 This is useful with sockets when you want to tell the other
5154 side you're done writing but not done reading, or vice versa.
5155 It's also a more insistent form of close because it also
5156 disables the file descriptor in any forked copies in other
5159 Returns C<1> for success. In the case of error, returns C<undef> if
5160 the first argument is not a valid filehandle, or returns C<0> and sets
5161 C<$!> for any other failure.
5164 X<sin> X<sine> X<asin> X<arcsine>
5168 Returns the sine of EXPR (expressed in radians). If EXPR is omitted,
5169 returns sine of C<$_>.
5171 For the inverse sine operation, you may use the C<Math::Trig::asin>
5172 function, or use this relation:
5174 sub asin { atan2($_[0], sqrt(1 - $_[0] * $_[0])) }
5181 Causes the script to sleep for EXPR seconds, or forever if no EXPR.
5182 Returns the number of seconds actually slept.
5184 May be interrupted if the process receives a signal such as C<SIGALRM>.
5187 local $SIG{ALARM} = sub { die "Alarm!\n" };
5190 die $@ unless $@ eq "Alarm!\n";
5192 You probably cannot mix C<alarm> and C<sleep> calls, because C<sleep>
5193 is often implemented using C<alarm>.
5195 On some older systems, it may sleep up to a full second less than what
5196 you requested, depending on how it counts seconds. Most modern systems
5197 always sleep the full amount. They may appear to sleep longer than that,
5198 however, because your process might not be scheduled right away in a
5199 busy multitasking system.
5201 For delays of finer granularity than one second, the Time::HiRes module
5202 (from CPAN, and starting from Perl 5.8 part of the standard
5203 distribution) provides usleep(). You may also use Perl's four-argument
5204 version of select() leaving the first three arguments undefined, or you
5205 might be able to use the C<syscall> interface to access setitimer(2) if
5206 your system supports it. See L<perlfaq8> for details.
5208 See also the POSIX module's C<pause> function.
5210 =item socket SOCKET,DOMAIN,TYPE,PROTOCOL
5213 Opens a socket of the specified kind and attaches it to filehandle
5214 SOCKET. DOMAIN, TYPE, and PROTOCOL are specified the same as for
5215 the system call of the same name. You should C<use Socket> first
5216 to get the proper definitions imported. See the examples in
5217 L<perlipc/"Sockets: Client/Server Communication">.
5219 On systems that support a close-on-exec flag on files, the flag will
5220 be set for the newly opened file descriptor, as determined by the
5221 value of $^F. See L<perlvar/$^F>.
5223 =item socketpair SOCKET1,SOCKET2,DOMAIN,TYPE,PROTOCOL
5226 Creates an unnamed pair of sockets in the specified domain, of the
5227 specified type. DOMAIN, TYPE, and PROTOCOL are specified the same as
5228 for the system call of the same name. If unimplemented, yields a fatal
5229 error. Returns true if successful.
5231 On systems that support a close-on-exec flag on files, the flag will
5232 be set for the newly opened file descriptors, as determined by the value
5233 of $^F. See L<perlvar/$^F>.
5235 Some systems defined C<pipe> in terms of C<socketpair>, in which a call
5236 to C<pipe(Rdr, Wtr)> is essentially:
5239 socketpair(Rdr, Wtr, AF_UNIX, SOCK_STREAM, PF_UNSPEC);
5240 shutdown(Rdr, 1); # no more writing for reader
5241 shutdown(Wtr, 0); # no more reading for writer
5243 See L<perlipc> for an example of socketpair use. Perl 5.8 and later will
5244 emulate socketpair using IP sockets to localhost if your system implements
5245 sockets but not socketpair.
5247 =item sort SUBNAME LIST
5248 X<sort> X<qsort> X<quicksort> X<mergesort>
5250 =item sort BLOCK LIST
5254 In list context, this sorts the LIST and returns the sorted list value.
5255 In scalar context, the behaviour of C<sort()> is undefined.
5257 If SUBNAME or BLOCK is omitted, C<sort>s in standard string comparison
5258 order. If SUBNAME is specified, it gives the name of a subroutine
5259 that returns an integer less than, equal to, or greater than C<0>,
5260 depending on how the elements of the list are to be ordered. (The C<<
5261 <=> >> and C<cmp> operators are extremely useful in such routines.)
5262 SUBNAME may be a scalar variable name (unsubscripted), in which case
5263 the value provides the name of (or a reference to) the actual
5264 subroutine to use. In place of a SUBNAME, you can provide a BLOCK as
5265 an anonymous, in-line sort subroutine.
5267 If the subroutine's prototype is C<($$)>, the elements to be compared
5268 are passed by reference in C<@_>, as for a normal subroutine. This is
5269 slower than unprototyped subroutines, where the elements to be
5270 compared are passed into the subroutine
5271 as the package global variables $a and $b (see example below). Note that
5272 in the latter case, it is usually counter-productive to declare $a and
5275 The values to be compared are always passed by reference and should not
5278 You also cannot exit out of the sort block or subroutine using any of the
5279 loop control operators described in L<perlsyn> or with C<goto>.
5281 When C<use locale> is in effect, C<sort LIST> sorts LIST according to the
5282 current collation locale. See L<perllocale>.
5284 sort() returns aliases into the original list, much as a for loop's index
5285 variable aliases the list elements. That is, modifying an element of a
5286 list returned by sort() (for example, in a C<foreach>, C<map> or C<grep>)
5287 actually modifies the element in the original list. This is usually
5288 something to be avoided when writing clear code.
5290 Perl 5.6 and earlier used a quicksort algorithm to implement sort.
5291 That algorithm was not stable, and I<could> go quadratic. (A I<stable> sort
5292 preserves the input order of elements that compare equal. Although
5293 quicksort's run time is O(NlogN) when averaged over all arrays of
5294 length N, the time can be O(N**2), I<quadratic> behavior, for some
5295 inputs.) In 5.7, the quicksort implementation was replaced with
5296 a stable mergesort algorithm whose worst-case behavior is O(NlogN).
5297 But benchmarks indicated that for some inputs, on some platforms,
5298 the original quicksort was faster. 5.8 has a sort pragma for
5299 limited control of the sort. Its rather blunt control of the
5300 underlying algorithm may not persist into future Perls, but the
5301 ability to characterize the input or output in implementation
5302 independent ways quite probably will. See L<the sort pragma|sort>.
5307 @articles = sort @files;
5309 # same thing, but with explicit sort routine
5310 @articles = sort {$a cmp $b} @files;
5312 # now case-insensitively
5313 @articles = sort {uc($a) cmp uc($b)} @files;
5315 # same thing in reversed order
5316 @articles = sort {$b cmp $a} @files;
5318 # sort numerically ascending
5319 @articles = sort {$a <=> $b} @files;
5321 # sort numerically descending
5322 @articles = sort {$b <=> $a} @files;
5324 # this sorts the %age hash by value instead of key
5325 # using an in-line function
5326 @eldest = sort { $age{$b} <=> $age{$a} } keys %age;
5328 # sort using explicit subroutine name
5330 $age{$a} <=> $age{$b}; # presuming numeric
5332 @sortedclass = sort byage @class;
5334 sub backwards { $b cmp $a }
5335 @harry = qw(dog cat x Cain Abel);
5336 @george = qw(gone chased yz Punished Axed);
5338 # prints AbelCaincatdogx
5339 print sort backwards @harry;
5340 # prints xdogcatCainAbel
5341 print sort @george, 'to', @harry;
5342 # prints AbelAxedCainPunishedcatchaseddoggonetoxyz
5344 # inefficiently sort by descending numeric compare using
5345 # the first integer after the first = sign, or the
5346 # whole record case-insensitively otherwise
5349 ($b =~ /=(\d+)/)[0] <=> ($a =~ /=(\d+)/)[0]
5354 # same thing, but much more efficiently;
5355 # we'll build auxiliary indices instead
5359 push @nums, /=(\d+)/;
5364 $nums[$b] <=> $nums[$a]
5366 $caps[$a] cmp $caps[$b]
5370 # same thing, but without any temps
5371 @new = map { $_->[0] }
5372 sort { $b->[1] <=> $a->[1]
5375 } map { [$_, /=(\d+)/, uc($_)] } @old;
5377 # using a prototype allows you to use any comparison subroutine
5378 # as a sort subroutine (including other package's subroutines)
5380 sub backwards ($$) { $_[1] cmp $_[0]; } # $a and $b are not set here
5383 @new = sort other::backwards @old;
5385 # guarantee stability, regardless of algorithm
5387 @new = sort { substr($a, 3, 5) cmp substr($b, 3, 5) } @old;
5389 # force use of mergesort (not portable outside Perl 5.8)
5390 use sort '_mergesort'; # note discouraging _
5391 @new = sort { substr($a, 3, 5) cmp substr($b, 3, 5) } @old;
5393 Warning: syntactical care is required when sorting the list returned from
5394 a function. If you want to sort the list returned by the function call
5395 C<find_records(@key)>, you can use:
5397 @contact = sort { $a cmp $b } find_records @key;
5398 @contact = sort +find_records(@key);
5399 @contact = sort &find_records(@key);
5400 @contact = sort(find_records(@key));
5402 If instead you want to sort the array @key with the comparison routine
5403 C<find_records()> then you can use:
5405 @contact = sort { find_records() } @key;
5406 @contact = sort find_records(@key);
5407 @contact = sort(find_records @key);
5408 @contact = sort(find_records (@key));
5410 If you're using strict, you I<must not> declare $a
5411 and $b as lexicals. They are package globals. That means
5412 that if you're in the C<main> package and type
5414 @articles = sort {$b <=> $a} @files;
5416 then C<$a> and C<$b> are C<$main::a> and C<$main::b> (or C<$::a> and C<$::b>),
5417 but if you're in the C<FooPack> package, it's the same as typing
5419 @articles = sort {$FooPack::b <=> $FooPack::a} @files;
5421 The comparison function is required to behave. If it returns
5422 inconsistent results (sometimes saying C<$x[1]> is less than C<$x[2]> and
5423 sometimes saying the opposite, for example) the results are not
5426 Because C<< <=> >> returns C<undef> when either operand is C<NaN>
5427 (not-a-number), and because C<sort> will trigger a fatal error unless the
5428 result of a comparison is defined, when sorting with a comparison function
5429 like C<< $a <=> $b >>, be careful about lists that might contain a C<NaN>.
5430 The following example takes advantage of the fact that C<NaN != NaN> to
5431 eliminate any C<NaN>s from the input.
5433 @result = sort { $a <=> $b } grep { $_ == $_ } @input;
5435 =item splice ARRAY,OFFSET,LENGTH,LIST
5438 =item splice ARRAY,OFFSET,LENGTH
5440 =item splice ARRAY,OFFSET
5444 Removes the elements designated by OFFSET and LENGTH from an array, and
5445 replaces them with the elements of LIST, if any. In list context,
5446 returns the elements removed from the array. In scalar context,
5447 returns the last element removed, or C<undef> if no elements are
5448 removed. The array grows or shrinks as necessary.
5449 If OFFSET is negative then it starts that far from the end of the array.
5450 If LENGTH is omitted, removes everything from OFFSET onward.
5451 If LENGTH is negative, removes the elements from OFFSET onward
5452 except for -LENGTH elements at the end of the array.
5453 If both OFFSET and LENGTH are omitted, removes everything. If OFFSET is
5454 past the end of the array, perl issues a warning, and splices at the
5457 The following equivalences hold (assuming C<< $[ == 0 and $#a >= $i >> )
5459 push(@a,$x,$y) splice(@a,@a,0,$x,$y)
5460 pop(@a) splice(@a,-1)
5461 shift(@a) splice(@a,0,1)
5462 unshift(@a,$x,$y) splice(@a,0,0,$x,$y)
5463 $a[$i] = $y splice(@a,$i,1,$y)
5465 Example, assuming array lengths are passed before arrays:
5467 sub aeq { # compare two list values
5468 my(@a) = splice(@_,0,shift);
5469 my(@b) = splice(@_,0,shift);
5470 return 0 unless @a == @b; # same len?
5472 return 0 if pop(@a) ne pop(@b);
5476 if (&aeq($len,@foo[1..$len],0+@bar,@bar)) { ... }
5478 =item split /PATTERN/,EXPR,LIMIT
5481 =item split /PATTERN/,EXPR
5483 =item split /PATTERN/
5487 Splits the string EXPR into a list of strings and returns that list. By
5488 default, empty leading fields are preserved, and empty trailing ones are
5489 deleted. (If all fields are empty, they are considered to be trailing.)
5491 In scalar context, returns the number of fields found.
5493 If EXPR is omitted, splits the C<$_> string. If PATTERN is also omitted,
5494 splits on whitespace (after skipping any leading whitespace). Anything
5495 matching PATTERN is taken to be a delimiter separating the fields. (Note
5496 that the delimiter may be longer than one character.)
5498 If LIMIT is specified and positive, it represents the maximum number
5499 of fields the EXPR will be split into, though the actual number of
5500 fields returned depends on the number of times PATTERN matches within
5501 EXPR. If LIMIT is unspecified or zero, trailing null fields are
5502 stripped (which potential users of C<pop> would do well to remember).
5503 If LIMIT is negative, it is treated as if an arbitrarily large LIMIT
5504 had been specified. Note that splitting an EXPR that evaluates to the
5505 empty string always returns the empty list, regardless of the LIMIT
5508 A pattern matching the null string (not to be confused with
5509 a null pattern C<//>, which is just one member of the set of patterns
5510 matching a null string) will split the value of EXPR into separate
5511 characters at each point it matches that way. For example:
5513 print join(':', split(/ */, 'hi there')), "\n";
5515 produces the output 'h:i:t:h:e:r:e'.
5517 As a special case for C<split>, using the empty pattern C<//> specifically
5518 matches only the null string, and is not be confused with the regular use
5519 of C<//> to mean "the last successful pattern match". So, for C<split>,
5522 print join(':', split(//, 'hi there')), "\n";
5524 produces the output 'h:i: :t:h:e:r:e'.
5526 Empty leading fields are produced when there are positive-width matches at
5527 the beginning of the string; a zero-width match at the beginning of
5528 the string does not produce an empty field. For example:
5530 print join(':', split(/(?=\w)/, 'hi there!'));
5532 produces the output 'h:i :t:h:e:r:e!'. Empty trailing fields, on the other
5533 hand, are produced when there is a match at the end of the string (and
5534 when LIMIT is given and is not 0), regardless of the length of the match.
5537 print join(':', split(//, 'hi there!', -1)), "\n";
5538 print join(':', split(/\W/, 'hi there!', -1)), "\n";
5540 produce the output 'h:i: :t:h:e:r:e:!:' and 'hi:there:', respectively,
5541 both with an empty trailing field.
5543 The LIMIT parameter can be used to split a line partially
5545 ($login, $passwd, $remainder) = split(/:/, $_, 3);
5547 When assigning to a list, if LIMIT is omitted, or zero, Perl supplies
5548 a LIMIT one larger than the number of variables in the list, to avoid
5549 unnecessary work. For the list above LIMIT would have been 4 by
5550 default. In time critical applications it behooves you not to split
5551 into more fields than you really need.
5553 If the PATTERN contains parentheses, additional list elements are
5554 created from each matching substring in the delimiter.
5556 split(/([,-])/, "1-10,20", 3);
5558 produces the list value
5560 (1, '-', 10, ',', 20)
5562 If you had the entire header of a normal Unix email message in $header,
5563 you could split it up into fields and their values this way:
5565 $header =~ s/\n(?=\s)//g; # fix continuation lines
5566 %hdrs = (UNIX_FROM => split /^(\S*?):\s*/m, $header);
5568 The pattern C</PATTERN/> may be replaced with an expression to specify
5569 patterns that vary at runtime. (To do runtime compilation only once,
5570 use C</$variable/o>.)
5572 As a special case, specifying a PATTERN of space (S<C<' '>>) will split on
5573 white space just as C<split> with no arguments does. Thus, S<C<split(' ')>> can
5574 be used to emulate B<awk>'s default behavior, whereas S<C<split(/ /)>>
5575 will give you as many null initial fields as there are leading spaces.
5576 A C<split> on C</\s+/> is like a S<C<split(' ')>> except that any leading
5577 whitespace produces a null first field. A C<split> with no arguments
5578 really does a S<C<split(' ', $_)>> internally.
5580 A PATTERN of C</^/> is treated as if it were C</^/m>, since it isn't
5585 open(PASSWD, '/etc/passwd');
5588 ($login, $passwd, $uid, $gid,
5589 $gcos, $home, $shell) = split(/:/);
5593 As with regular pattern matching, any capturing parentheses that are not
5594 matched in a C<split()> will be set to C<undef> when returned:
5596 @fields = split /(A)|B/, "1A2B3";
5597 # @fields is (1, 'A', 2, undef, 3)
5599 =item sprintf FORMAT, LIST
5602 Returns a string formatted by the usual C<printf> conventions of the C
5603 library function C<sprintf>. See below for more details
5604 and see C<sprintf(3)> or C<printf(3)> on your system for an explanation of
5605 the general principles.
5609 # Format number with up to 8 leading zeroes
5610 $result = sprintf("%08d", $number);
5612 # Round number to 3 digits after decimal point
5613 $rounded = sprintf("%.3f", $number);
5615 Perl does its own C<sprintf> formatting--it emulates the C
5616 function C<sprintf>, but it doesn't use it (except for floating-point
5617 numbers, and even then only the standard modifiers are allowed). As a
5618 result, any non-standard extensions in your local C<sprintf> are not
5619 available from Perl.
5621 Unlike C<printf>, C<sprintf> does not do what you probably mean when you
5622 pass it an array as your first argument. The array is given scalar context,
5623 and instead of using the 0th element of the array as the format, Perl will
5624 use the count of elements in the array as the format, which is almost never
5627 Perl's C<sprintf> permits the following universally-known conversions:
5630 %c a character with the given number
5632 %d a signed integer, in decimal
5633 %u an unsigned integer, in decimal
5634 %o an unsigned integer, in octal
5635 %x an unsigned integer, in hexadecimal
5636 %e a floating-point number, in scientific notation
5637 %f a floating-point number, in fixed decimal notation
5638 %g a floating-point number, in %e or %f notation
5640 In addition, Perl permits the following widely-supported conversions:
5642 %X like %x, but using upper-case letters
5643 %E like %e, but using an upper-case "E"
5644 %G like %g, but with an upper-case "E" (if applicable)
5645 %b an unsigned integer, in binary
5646 %B like %b, but using an upper-case "B" with the # flag
5647 %p a pointer (outputs the Perl value's address in hexadecimal)
5648 %n special: *stores* the number of characters output so far
5649 into the next variable in the parameter list
5651 Finally, for backward (and we do mean "backward") compatibility, Perl
5652 permits these unnecessary but widely-supported conversions:
5655 %D a synonym for %ld
5656 %U a synonym for %lu
5657 %O a synonym for %lo
5660 Note that the number of exponent digits in the scientific notation produced
5661 by C<%e>, C<%E>, C<%g> and C<%G> for numbers with the modulus of the
5662 exponent less than 100 is system-dependent: it may be three or less
5663 (zero-padded as necessary). In other words, 1.23 times ten to the
5664 99th may be either "1.23e99" or "1.23e099".
5666 Between the C<%> and the format letter, you may specify a number of
5667 additional attributes controlling the interpretation of the format.
5668 In order, these are:
5672 =item format parameter index
5674 An explicit format parameter index, such as C<2$>. By default sprintf
5675 will format the next unused argument in the list, but this allows you
5676 to take the arguments out of order, e.g.:
5678 printf '%2$d %1$d', 12, 34; # prints "34 12"
5679 printf '%3$d %d %1$d', 1, 2, 3; # prints "3 1 1"
5685 space prefix non-negative number with a space
5686 + prefix non-negative number with a plus sign
5687 - left-justify within the field
5688 0 use zeros, not spaces, to right-justify
5689 # ensure the leading "0" for any octal,
5690 prefix non-zero hexadecimal with "0x" or "0X",
5691 prefix non-zero binary with "0b" or "0B"
5695 printf '<% d>', 12; # prints "< 12>"
5696 printf '<%+d>', 12; # prints "<+12>"
5697 printf '<%6s>', 12; # prints "< 12>"
5698 printf '<%-6s>', 12; # prints "<12 >"
5699 printf '<%06s>', 12; # prints "<000012>"
5700 printf '<%#o>', 12; # prints "<014>"
5701 printf '<%#x>', 12; # prints "<0xc>"
5702 printf '<%#X>', 12; # prints "<0XC>"
5703 printf '<%#b>', 12; # prints "<0b1100>"
5704 printf '<%#B>', 12; # prints "<0B1100>"
5706 When a space and a plus sign are given as the flags at once,
5707 a plus sign is used to prefix a positive number.
5709 printf '<%+ d>', 12; # prints "<+12>"
5710 printf '<% +d>', 12; # prints "<+12>"
5712 When the # flag and a precision are given in the %o conversion,
5713 the precision is incremented if it's necessary for the leading "0".
5715 printf '<%#.5o>', 012; # prints "<00012>"
5716 printf '<%#.5o>', 012345; # prints "<012345>"
5717 printf '<%#.0o>', 0; # prints "<0>"
5721 This flag tells perl to interpret the supplied string as a vector of
5722 integers, one for each character in the string. Perl applies the format to
5723 each integer in turn, then joins the resulting strings with a separator (a
5724 dot C<.> by default). This can be useful for displaying ordinal values of
5725 characters in arbitrary strings:
5727 printf "%vd", "AB\x{100}"; # prints "65.66.256"
5728 printf "version is v%vd\n", $^V; # Perl's version
5730 Put an asterisk C<*> before the C<v> to override the string to
5731 use to separate the numbers:
5733 printf "address is %*vX\n", ":", $addr; # IPv6 address
5734 printf "bits are %0*v8b\n", " ", $bits; # random bitstring
5736 You can also explicitly specify the argument number to use for
5737 the join string using e.g. C<*2$v>:
5739 printf '%*4$vX %*4$vX %*4$vX', @addr[1..3], ":"; # 3 IPv6 addresses
5741 =item (minimum) width
5743 Arguments are usually formatted to be only as wide as required to
5744 display the given value. You can override the width by putting
5745 a number here, or get the width from the next argument (with C<*>)
5746 or from a specified argument (with e.g. C<*2$>):
5748 printf '<%s>', "a"; # prints "<a>"
5749 printf '<%6s>', "a"; # prints "< a>"
5750 printf '<%*s>', 6, "a"; # prints "< a>"
5751 printf '<%*2$s>', "a", 6; # prints "< a>"
5752 printf '<%2s>', "long"; # prints "<long>" (does not truncate)
5754 If a field width obtained through C<*> is negative, it has the same
5755 effect as the C<-> flag: left-justification.
5757 =item precision, or maximum width
5760 You can specify a precision (for numeric conversions) or a maximum
5761 width (for string conversions) by specifying a C<.> followed by a number.
5762 For floating point formats, with the exception of 'g' and 'G', this specifies
5763 the number of decimal places to show (the default being 6), e.g.:
5765 # these examples are subject to system-specific variation
5766 printf '<%f>', 1; # prints "<1.000000>"
5767 printf '<%.1f>', 1; # prints "<1.0>"
5768 printf '<%.0f>', 1; # prints "<1>"
5769 printf '<%e>', 10; # prints "<1.000000e+01>"
5770 printf '<%.1e>', 10; # prints "<1.0e+01>"
5772 For 'g' and 'G', this specifies the maximum number of digits to show,
5773 including prior to the decimal point as well as after it, e.g.:
5775 # these examples are subject to system-specific variation
5776 printf '<%g>', 1; # prints "<1>"
5777 printf '<%.10g>', 1; # prints "<1>"
5778 printf '<%g>', 100; # prints "<100>"
5779 printf '<%.1g>', 100; # prints "<1e+02>"
5780 printf '<%.2g>', 100.01; # prints "<1e+02>"
5781 printf '<%.5g>', 100.01; # prints "<100.01>"
5782 printf '<%.4g>', 100.01; # prints "<100>"
5784 For integer conversions, specifying a precision implies that the
5785 output of the number itself should be zero-padded to this width,
5786 where the 0 flag is ignored:
5788 printf '<%.6d>', 1; # prints "<000001>"
5789 printf '<%+.6d>', 1; # prints "<+000001>"
5790 printf '<%-10.6d>', 1; # prints "<000001 >"
5791 printf '<%10.6d>', 1; # prints "< 000001>"
5792 printf '<%010.6d>', 1; # prints "< 000001>"
5793 printf '<%+10.6d>', 1; # prints "< +000001>"
5795 printf '<%.6x>', 1; # prints "<000001>"
5796 printf '<%#.6x>', 1; # prints "<0x000001>"
5797 printf '<%-10.6x>', 1; # prints "<000001 >"
5798 printf '<%10.6x>', 1; # prints "< 000001>"
5799 printf '<%010.6x>', 1; # prints "< 000001>"
5800 printf '<%#10.6x>', 1; # prints "< 0x000001>"
5802 For string conversions, specifying a precision truncates the string
5803 to fit in the specified width:
5805 printf '<%.5s>', "truncated"; # prints "<trunc>"
5806 printf '<%10.5s>', "truncated"; # prints "< trunc>"
5808 You can also get the precision from the next argument using C<.*>:
5810 printf '<%.6x>', 1; # prints "<000001>"
5811 printf '<%.*x>', 6, 1; # prints "<000001>"
5813 If a precision obtained through C<*> is negative, it has the same
5814 effect as no precision.
5816 printf '<%.*s>', 7, "string"; # prints "<string>"
5817 printf '<%.*s>', 3, "string"; # prints "<str>"
5818 printf '<%.*s>', 0, "string"; # prints "<>"
5819 printf '<%.*s>', -1, "string"; # prints "<string>"
5821 printf '<%.*d>', 1, 0; # prints "<0>"
5822 printf '<%.*d>', 0, 0; # prints "<>"
5823 printf '<%.*d>', -1, 0; # prints "<0>"
5825 You cannot currently get the precision from a specified number,
5826 but it is intended that this will be possible in the future using
5829 printf '<%.*2$x>', 1, 6; # INVALID, but in future will print "<000001>"
5833 For numeric conversions, you can specify the size to interpret the
5834 number as using C<l>, C<h>, C<V>, C<q>, C<L>, or C<ll>. For integer
5835 conversions (C<d u o x X b i D U O>), numbers are usually assumed to be
5836 whatever the default integer size is on your platform (usually 32 or 64
5837 bits), but you can override this to use instead one of the standard C types,
5838 as supported by the compiler used to build Perl:
5840 l interpret integer as C type "long" or "unsigned long"
5841 h interpret integer as C type "short" or "unsigned short"
5842 q, L or ll interpret integer as C type "long long", "unsigned long long".
5843 or "quads" (typically 64-bit integers)
5845 The last will produce errors if Perl does not understand "quads" in your
5846 installation. (This requires that either the platform natively supports quads
5847 or Perl was specifically compiled to support quads.) You can find out
5848 whether your Perl supports quads via L<Config>:
5851 ($Config{use64bitint} eq 'define' || $Config{longsize} >= 8) &&
5854 For floating point conversions (C<e f g E F G>), numbers are usually assumed
5855 to be the default floating point size on your platform (double or long double),
5856 but you can force 'long double' with C<q>, C<L>, or C<ll> if your
5857 platform supports them. You can find out whether your Perl supports long
5858 doubles via L<Config>:
5861 $Config{d_longdbl} eq 'define' && print "long doubles\n";
5863 You can find out whether Perl considers 'long double' to be the default
5864 floating point size to use on your platform via L<Config>:
5867 ($Config{uselongdouble} eq 'define') &&
5868 print "long doubles by default\n";
5870 It can also be the case that long doubles and doubles are the same thing:
5873 ($Config{doublesize} == $Config{longdblsize}) &&
5874 print "doubles are long doubles\n";
5876 The size specifier C<V> has no effect for Perl code, but it is supported
5877 for compatibility with XS code; it means 'use the standard size for
5878 a Perl integer (or floating-point number)', which is already the
5879 default for Perl code.
5881 =item order of arguments
5883 Normally, sprintf takes the next unused argument as the value to
5884 format for each format specification. If the format specification
5885 uses C<*> to require additional arguments, these are consumed from
5886 the argument list in the order in which they appear in the format
5887 specification I<before> the value to format. Where an argument is
5888 specified using an explicit index, this does not affect the normal
5889 order for the arguments (even when the explicitly specified index
5890 would have been the next argument in any case).
5894 printf '<%*.*s>', $a, $b, $c;
5896 would use C<$a> for the width, C<$b> for the precision and C<$c>
5897 as the value to format, while:
5899 printf '<%*1$.*s>', $a, $b;
5901 would use C<$a> for the width and the precision, and C<$b> as the
5904 Here are some more examples - beware that when using an explicit
5905 index, the C<$> may need to be escaped:
5907 printf "%2\$d %d\n", 12, 34; # will print "34 12\n"
5908 printf "%2\$d %d %d\n", 12, 34; # will print "34 12 34\n"
5909 printf "%3\$d %d %d\n", 12, 34, 56; # will print "56 12 34\n"
5910 printf "%2\$*3\$d %d\n", 12, 34, 3; # will print " 34 12\n"
5914 If C<use locale> is in effect, and POSIX::setlocale() has been called,
5915 the character used for the decimal separator in formatted floating
5916 point numbers is affected by the LC_NUMERIC locale. See L<perllocale>
5920 X<sqrt> X<root> X<square root>
5924 Return the square root of EXPR. If EXPR is omitted, returns square
5925 root of C<$_>. Only works on non-negative operands, unless you've
5926 loaded the standard Math::Complex module.
5929 print sqrt(-2); # prints 1.4142135623731i
5932 X<srand> X<seed> X<randseed>
5936 Sets the random number seed for the C<rand> operator.
5938 The point of the function is to "seed" the C<rand> function so that
5939 C<rand> can produce a different sequence each time you run your
5942 If srand() is not called explicitly, it is called implicitly at the
5943 first use of the C<rand> operator. However, this was not the case in
5944 versions of Perl before 5.004, so if your script will run under older
5945 Perl versions, it should call C<srand>.
5947 Most programs won't even call srand() at all, except those that
5948 need a cryptographically-strong starting point rather than the
5949 generally acceptable default, which is based on time of day,
5950 process ID, and memory allocation, or the F</dev/urandom> device,
5953 You can call srand($seed) with the same $seed to reproduce the
5954 I<same> sequence from rand(), but this is usually reserved for
5955 generating predictable results for testing or debugging.
5956 Otherwise, don't call srand() more than once in your program.
5958 Do B<not> call srand() (i.e. without an argument) more than once in
5959 a script. The internal state of the random number generator should
5960 contain more entropy than can be provided by any seed, so calling
5961 srand() again actually I<loses> randomness.
5963 Most implementations of C<srand> take an integer and will silently
5964 truncate decimal numbers. This means C<srand(42)> will usually
5965 produce the same results as C<srand(42.1)>. To be safe, always pass
5966 C<srand> an integer.
5968 In versions of Perl prior to 5.004 the default seed was just the
5969 current C<time>. This isn't a particularly good seed, so many old
5970 programs supply their own seed value (often C<time ^ $$> or C<time ^
5971 ($$ + ($$ << 15))>), but that isn't necessary any more.
5973 For cryptographic purposes, however, you need something much more random
5974 than the default seed. Checksumming the compressed output of one or more
5975 rapidly changing operating system status programs is the usual method. For
5978 srand (time ^ $$ ^ unpack "%L*", `ps axww | gzip -f`);
5980 If you're particularly concerned with this, see the C<Math::TrulyRandom>
5983 Frequently called programs (like CGI scripts) that simply use
5987 for a seed can fall prey to the mathematical property that
5991 one-third of the time. So don't do that.
5993 =item stat FILEHANDLE
5994 X<stat> X<file, status> X<ctime>
5998 =item stat DIRHANDLE
6002 Returns a 13-element list giving the status info for a file, either
6003 the file opened via FILEHANDLE or DIRHANDLE, or named by EXPR. If EXPR is
6004 omitted, it stats C<$_>. Returns a null list if the stat fails. Typically
6007 ($dev,$ino,$mode,$nlink,$uid,$gid,$rdev,$size,
6008 $atime,$mtime,$ctime,$blksize,$blocks)
6011 Not all fields are supported on all filesystem types. Here are the
6012 meanings of the fields:
6014 0 dev device number of filesystem
6016 2 mode file mode (type and permissions)
6017 3 nlink number of (hard) links to the file
6018 4 uid numeric user ID of file's owner
6019 5 gid numeric group ID of file's owner
6020 6 rdev the device identifier (special files only)
6021 7 size total size of file, in bytes
6022 8 atime last access time in seconds since the epoch
6023 9 mtime last modify time in seconds since the epoch
6024 10 ctime inode change time in seconds since the epoch (*)
6025 11 blksize preferred block size for file system I/O
6026 12 blocks actual number of blocks allocated
6028 (The epoch was at 00:00 January 1, 1970 GMT.)
6030 (*) Not all fields are supported on all filesystem types. Notably, the
6031 ctime field is non-portable. In particular, you cannot expect it to be a
6032 "creation time", see L<perlport/"Files and Filesystems"> for details.
6034 If C<stat> is passed the special filehandle consisting of an underline, no
6035 stat is done, but the current contents of the stat structure from the
6036 last C<stat>, C<lstat>, or filetest are returned. Example:
6038 if (-x $file && (($d) = stat(_)) && $d < 0) {
6039 print "$file is executable NFS file\n";
6042 (This works on machines only for which the device number is negative
6045 Because the mode contains both the file type and its permissions, you
6046 should mask off the file type portion and (s)printf using a C<"%o">
6047 if you want to see the real permissions.
6049 $mode = (stat($filename))[2];
6050 printf "Permissions are %04o\n", $mode & 07777;
6052 In scalar context, C<stat> returns a boolean value indicating success
6053 or failure, and, if successful, sets the information associated with
6054 the special filehandle C<_>.
6056 The L<File::stat> module provides a convenient, by-name access mechanism:
6059 $sb = stat($filename);
6060 printf "File is %s, size is %s, perm %04o, mtime %s\n",
6061 $filename, $sb->size, $sb->mode & 07777,
6062 scalar localtime $sb->mtime;
6064 You can import symbolic mode constants (C<S_IF*>) and functions
6065 (C<S_IS*>) from the Fcntl module:
6069 $mode = (stat($filename))[2];
6071 $user_rwx = ($mode & S_IRWXU) >> 6;
6072 $group_read = ($mode & S_IRGRP) >> 3;
6073 $other_execute = $mode & S_IXOTH;
6075 printf "Permissions are %04o\n", S_IMODE($mode), "\n";
6077 $is_setuid = $mode & S_ISUID;
6078 $is_directory = S_ISDIR($mode);
6080 You could write the last two using the C<-u> and C<-d> operators.
6081 The commonly available C<S_IF*> constants are
6083 # Permissions: read, write, execute, for user, group, others.
6085 S_IRWXU S_IRUSR S_IWUSR S_IXUSR
6086 S_IRWXG S_IRGRP S_IWGRP S_IXGRP
6087 S_IRWXO S_IROTH S_IWOTH S_IXOTH
6089 # Setuid/Setgid/Stickiness/SaveText.
6090 # Note that the exact meaning of these is system dependent.
6092 S_ISUID S_ISGID S_ISVTX S_ISTXT
6094 # File types. Not necessarily all are available on your system.
6096 S_IFREG S_IFDIR S_IFLNK S_IFBLK S_IFCHR S_IFIFO S_IFSOCK S_IFWHT S_ENFMT
6098 # The following are compatibility aliases for S_IRUSR, S_IWUSR, S_IXUSR.
6100 S_IREAD S_IWRITE S_IEXEC
6102 and the C<S_IF*> functions are
6104 S_IMODE($mode) the part of $mode containing the permission bits
6105 and the setuid/setgid/sticky bits
6107 S_IFMT($mode) the part of $mode containing the file type
6108 which can be bit-anded with e.g. S_IFREG
6109 or with the following functions
6111 # The operators -f, -d, -l, -b, -c, -p, and -S.
6113 S_ISREG($mode) S_ISDIR($mode) S_ISLNK($mode)
6114 S_ISBLK($mode) S_ISCHR($mode) S_ISFIFO($mode) S_ISSOCK($mode)
6116 # No direct -X operator counterpart, but for the first one
6117 # the -g operator is often equivalent. The ENFMT stands for
6118 # record flocking enforcement, a platform-dependent feature.
6120 S_ISENFMT($mode) S_ISWHT($mode)
6122 See your native chmod(2) and stat(2) documentation for more details
6123 about the C<S_*> constants. To get status info for a symbolic link
6124 instead of the target file behind the link, use the C<lstat> function.
6129 =item state TYPE EXPR
6131 =item state EXPR : ATTRS
6133 =item state TYPE EXPR : ATTRS
6135 C<state> declares a lexically scoped variable, just like C<my> does.
6136 However, those variables will never be reinitialized, contrary to
6137 lexical variables that are reinitialized each time their enclosing block
6140 C<state> variables are only enabled when the C<feature 'state'> pragma is
6141 in effect. See L<feature>.
6148 Takes extra time to study SCALAR (C<$_> if unspecified) in anticipation of
6149 doing many pattern matches on the string before it is next modified.
6150 This may or may not save time, depending on the nature and number of
6151 patterns you are searching on, and on the distribution of character
6152 frequencies in the string to be searched--you probably want to compare
6153 run times with and without it to see which runs faster. Those loops
6154 that scan for many short constant strings (including the constant
6155 parts of more complex patterns) will benefit most. You may have only
6156 one C<study> active at a time--if you study a different scalar the first
6157 is "unstudied". (The way C<study> works is this: a linked list of every
6158 character in the string to be searched is made, so we know, for
6159 example, where all the C<'k'> characters are. From each search string,
6160 the rarest character is selected, based on some static frequency tables
6161 constructed from some C programs and English text. Only those places
6162 that contain this "rarest" character are examined.)
6164 For example, here is a loop that inserts index producing entries
6165 before any line containing a certain pattern:
6169 print ".IX foo\n" if /\bfoo\b/;
6170 print ".IX bar\n" if /\bbar\b/;
6171 print ".IX blurfl\n" if /\bblurfl\b/;
6176 In searching for C</\bfoo\b/>, only those locations in C<$_> that contain C<f>
6177 will be looked at, because C<f> is rarer than C<o>. In general, this is
6178 a big win except in pathological cases. The only question is whether
6179 it saves you more time than it took to build the linked list in the
6182 Note that if you have to look for strings that you don't know till
6183 runtime, you can build an entire loop as a string and C<eval> that to
6184 avoid recompiling all your patterns all the time. Together with
6185 undefining C<$/> to input entire files as one record, this can be very
6186 fast, often faster than specialized programs like fgrep(1). The following
6187 scans a list of files (C<@files>) for a list of words (C<@words>), and prints
6188 out the names of those files that contain a match:
6190 $search = 'while (<>) { study;';
6191 foreach $word (@words) {
6192 $search .= "++\$seen{\$ARGV} if /\\b$word\\b/;\n";
6197 eval $search; # this screams
6198 $/ = "\n"; # put back to normal input delimiter
6199 foreach $file (sort keys(%seen)) {
6203 =item sub NAME BLOCK
6206 =item sub NAME (PROTO) BLOCK
6208 =item sub NAME : ATTRS BLOCK
6210 =item sub NAME (PROTO) : ATTRS BLOCK
6212 This is subroutine definition, not a real function I<per se>.
6213 Without a BLOCK it's just a forward declaration. Without a NAME,
6214 it's an anonymous function declaration, and does actually return
6215 a value: the CODE ref of the closure you just created.
6217 See L<perlsub> and L<perlref> for details about subroutines and
6218 references, and L<attributes> and L<Attribute::Handlers> for more
6219 information about attributes.
6221 =item substr EXPR,OFFSET,LENGTH,REPLACEMENT
6222 X<substr> X<substring> X<mid> X<left> X<right>
6224 =item substr EXPR,OFFSET,LENGTH
6226 =item substr EXPR,OFFSET
6228 Extracts a substring out of EXPR and returns it. First character is at
6229 offset C<0>, or whatever you've set C<$[> to (but don't do that).
6230 If OFFSET is negative (or more precisely, less than C<$[>), starts
6231 that far from the end of the string. If LENGTH is omitted, returns
6232 everything to the end of the string. If LENGTH is negative, leaves that
6233 many characters off the end of the string.
6235 my $s = "The black cat climbed the green tree";
6236 my $color = substr $s, 4, 5; # black
6237 my $middle = substr $s, 4, -11; # black cat climbed the
6238 my $end = substr $s, 14; # climbed the green tree
6239 my $tail = substr $s, -4; # tree
6240 my $z = substr $s, -4, 2; # tr
6242 You can use the substr() function as an lvalue, in which case EXPR
6243 must itself be an lvalue. If you assign something shorter than LENGTH,
6244 the string will shrink, and if you assign something longer than LENGTH,
6245 the string will grow to accommodate it. To keep the string the same
6246 length you may need to pad or chop your value using C<sprintf>.
6248 If OFFSET and LENGTH specify a substring that is partly outside the
6249 string, only the part within the string is returned. If the substring
6250 is beyond either end of the string, substr() returns the undefined
6251 value and produces a warning. When used as an lvalue, specifying a
6252 substring that is entirely outside the string is a fatal error.
6253 Here's an example showing the behavior for boundary cases:
6256 substr($name, 4) = 'dy'; # $name is now 'freddy'
6257 my $null = substr $name, 6, 2; # returns '' (no warning)
6258 my $oops = substr $name, 7; # returns undef, with warning
6259 substr($name, 7) = 'gap'; # fatal error
6261 An alternative to using substr() as an lvalue is to specify the
6262 replacement string as the 4th argument. This allows you to replace
6263 parts of the EXPR and return what was there before in one operation,
6264 just as you can with splice().
6266 my $s = "The black cat climbed the green tree";
6267 my $z = substr $s, 14, 7, "jumped from"; # climbed
6268 # $s is now "The black cat jumped from the green tree"
6270 Note that the lvalue returned by the 3-arg version of substr() acts as
6271 a 'magic bullet'; each time it is assigned to, it remembers which part
6272 of the original string is being modified; for example:
6275 for (substr($x,1,2)) {
6276 $_ = 'a'; print $x,"\n"; # prints 1a4
6277 $_ = 'xyz'; print $x,"\n"; # prints 1xyz4
6279 $_ = 'pq'; print $x,"\n"; # prints 5pq9
6282 Prior to Perl version 5.9.1, the result of using an lvalue multiple times was
6285 =item symlink OLDFILE,NEWFILE
6286 X<symlink> X<link> X<symbolic link> X<link, symbolic>
6288 Creates a new filename symbolically linked to the old filename.
6289 Returns C<1> for success, C<0> otherwise. On systems that don't support
6290 symbolic links, produces a fatal error at run time. To check for that,
6293 $symlink_exists = eval { symlink("",""); 1 };
6295 =item syscall NUMBER, LIST
6296 X<syscall> X<system call>
6298 Calls the system call specified as the first element of the list,
6299 passing the remaining elements as arguments to the system call. If
6300 unimplemented, produces a fatal error. The arguments are interpreted
6301 as follows: if a given argument is numeric, the argument is passed as
6302 an int. If not, the pointer to the string value is passed. You are
6303 responsible to make sure a string is pre-extended long enough to
6304 receive any result that might be written into a string. You can't use a
6305 string literal (or other read-only string) as an argument to C<syscall>
6306 because Perl has to assume that any string pointer might be written
6308 integer arguments are not literals and have never been interpreted in a
6309 numeric context, you may need to add C<0> to them to force them to look
6310 like numbers. This emulates the C<syswrite> function (or vice versa):
6312 require 'syscall.ph'; # may need to run h2ph
6314 syscall(&SYS_write, fileno(STDOUT), $s, length $s);
6316 Note that Perl supports passing of up to only 14 arguments to your system call,
6317 which in practice should usually suffice.
6319 Syscall returns whatever value returned by the system call it calls.
6320 If the system call fails, C<syscall> returns C<-1> and sets C<$!> (errno).
6321 Note that some system calls can legitimately return C<-1>. The proper
6322 way to handle such calls is to assign C<$!=0;> before the call and
6323 check the value of C<$!> if syscall returns C<-1>.
6325 There's a problem with C<syscall(&SYS_pipe)>: it returns the file
6326 number of the read end of the pipe it creates. There is no way
6327 to retrieve the file number of the other end. You can avoid this
6328 problem by using C<pipe> instead.
6330 =item sysopen FILEHANDLE,FILENAME,MODE
6333 =item sysopen FILEHANDLE,FILENAME,MODE,PERMS
6335 Opens the file whose filename is given by FILENAME, and associates it
6336 with FILEHANDLE. If FILEHANDLE is an expression, its value is used as
6337 the name of the real filehandle wanted. This function calls the
6338 underlying operating system's C<open> function with the parameters
6339 FILENAME, MODE, PERMS.
6341 The possible values and flag bits of the MODE parameter are
6342 system-dependent; they are available via the standard module C<Fcntl>.
6343 See the documentation of your operating system's C<open> to see which
6344 values and flag bits are available. You may combine several flags
6345 using the C<|>-operator.
6347 Some of the most common values are C<O_RDONLY> for opening the file in
6348 read-only mode, C<O_WRONLY> for opening the file in write-only mode,
6349 and C<O_RDWR> for opening the file in read-write mode.
6350 X<O_RDONLY> X<O_RDWR> X<O_WRONLY>
6352 For historical reasons, some values work on almost every system
6353 supported by perl: zero means read-only, one means write-only, and two
6354 means read/write. We know that these values do I<not> work under
6355 OS/390 & VM/ESA Unix and on the Macintosh; you probably don't want to
6356 use them in new code.
6358 If the file named by FILENAME does not exist and the C<open> call creates
6359 it (typically because MODE includes the C<O_CREAT> flag), then the value of
6360 PERMS specifies the permissions of the newly created file. If you omit
6361 the PERMS argument to C<sysopen>, Perl uses the octal value C<0666>.
6362 These permission values need to be in octal, and are modified by your
6363 process's current C<umask>.
6366 In many systems the C<O_EXCL> flag is available for opening files in
6367 exclusive mode. This is B<not> locking: exclusiveness means here that
6368 if the file already exists, sysopen() fails. C<O_EXCL> may not work
6369 on network filesystems, and has no effect unless the C<O_CREAT> flag
6370 is set as well. Setting C<O_CREAT|O_EXCL> prevents the file from
6371 being opened if it is a symbolic link. It does not protect against
6372 symbolic links in the file's path.
6375 Sometimes you may want to truncate an already-existing file. This
6376 can be done using the C<O_TRUNC> flag. The behavior of
6377 C<O_TRUNC> with C<O_RDONLY> is undefined.
6380 You should seldom if ever use C<0644> as argument to C<sysopen>, because
6381 that takes away the user's option to have a more permissive umask.
6382 Better to omit it. See the perlfunc(1) entry on C<umask> for more
6385 Note that C<sysopen> depends on the fdopen() C library function.
6386 On many UNIX systems, fdopen() is known to fail when file descriptors
6387 exceed a certain value, typically 255. If you need more file
6388 descriptors than that, consider rebuilding Perl to use the C<sfio>
6389 library, or perhaps using the POSIX::open() function.
6391 See L<perlopentut> for a kinder, gentler explanation of opening files.
6393 =item sysread FILEHANDLE,SCALAR,LENGTH,OFFSET
6396 =item sysread FILEHANDLE,SCALAR,LENGTH
6398 Attempts to read LENGTH bytes of data into variable SCALAR from the
6399 specified FILEHANDLE, using the system call read(2). It bypasses
6400 buffered IO, so mixing this with other kinds of reads, C<print>,
6401 C<write>, C<seek>, C<tell>, or C<eof> can cause confusion because the
6402 perlio or stdio layers usually buffers data. Returns the number of
6403 bytes actually read, C<0> at end of file, or undef if there was an
6404 error (in the latter case C<$!> is also set). SCALAR will be grown or
6405 shrunk so that the last byte actually read is the last byte of the
6406 scalar after the read.
6408 An OFFSET may be specified to place the read data at some place in the
6409 string other than the beginning. A negative OFFSET specifies
6410 placement at that many characters counting backwards from the end of
6411 the string. A positive OFFSET greater than the length of SCALAR
6412 results in the string being padded to the required size with C<"\0">
6413 bytes before the result of the read is appended.
6415 There is no syseof() function, which is ok, since eof() doesn't work
6416 very well on device files (like ttys) anyway. Use sysread() and check
6417 for a return value for 0 to decide whether you're done.
6419 Note that if the filehandle has been marked as C<:utf8> Unicode
6420 characters are read instead of bytes (the LENGTH, OFFSET, and the
6421 return value of sysread() are in Unicode characters).
6422 The C<:encoding(...)> layer implicitly introduces the C<:utf8> layer.
6423 See L</binmode>, L</open>, and the C<open> pragma, L<open>.
6425 =item sysseek FILEHANDLE,POSITION,WHENCE
6428 Sets FILEHANDLE's system position in bytes using the system call
6429 lseek(2). FILEHANDLE may be an expression whose value gives the name
6430 of the filehandle. The values for WHENCE are C<0> to set the new
6431 position to POSITION, C<1> to set the it to the current position plus
6432 POSITION, and C<2> to set it to EOF plus POSITION (typically
6435 Note the I<in bytes>: even if the filehandle has been set to operate
6436 on characters (for example by using the C<:encoding(utf8)> I/O layer),
6437 tell() will return byte offsets, not character offsets (because
6438 implementing that would render sysseek() very slow).
6440 sysseek() bypasses normal buffered IO, so mixing this with reads (other
6441 than C<sysread>, for example C<< <> >> or read()) C<print>, C<write>,
6442 C<seek>, C<tell>, or C<eof> may cause confusion.
6444 For WHENCE, you may also use the constants C<SEEK_SET>, C<SEEK_CUR>,
6445 and C<SEEK_END> (start of the file, current position, end of the file)
6446 from the Fcntl module. Use of the constants is also more portable
6447 than relying on 0, 1, and 2. For example to define a "systell" function:
6449 use Fcntl 'SEEK_CUR';
6450 sub systell { sysseek($_[0], 0, SEEK_CUR) }
6452 Returns the new position, or the undefined value on failure. A position
6453 of zero is returned as the string C<"0 but true">; thus C<sysseek> returns
6454 true on success and false on failure, yet you can still easily determine
6460 =item system PROGRAM LIST
6462 Does exactly the same thing as C<exec LIST>, except that a fork is
6463 done first, and the parent process waits for the child process to
6464 complete. Note that argument processing varies depending on the
6465 number of arguments. If there is more than one argument in LIST,
6466 or if LIST is an array with more than one value, starts the program
6467 given by the first element of the list with arguments given by the
6468 rest of the list. If there is only one scalar argument, the argument
6469 is checked for shell metacharacters, and if there are any, the
6470 entire argument is passed to the system's command shell for parsing
6471 (this is C</bin/sh -c> on Unix platforms, but varies on other
6472 platforms). If there are no shell metacharacters in the argument,
6473 it is split into words and passed directly to C<execvp>, which is
6476 Beginning with v5.6.0, Perl will attempt to flush all files opened for
6477 output before any operation that may do a fork, but this may not be
6478 supported on some platforms (see L<perlport>). To be safe, you may need
6479 to set C<$|> ($AUTOFLUSH in English) or call the C<autoflush()> method
6480 of C<IO::Handle> on any open handles.
6482 The return value is the exit status of the program as returned by the
6483 C<wait> call. To get the actual exit value, shift right by eight (see
6484 below). See also L</exec>. This is I<not> what you want to use to capture
6485 the output from a command, for that you should use merely backticks or
6486 C<qx//>, as described in L<perlop/"`STRING`">. Return value of -1
6487 indicates a failure to start the program or an error of the wait(2) system
6488 call (inspect $! for the reason).
6490 If you'd like to make C<system> (and many other bits of Perl) die on error,
6491 have a look at the L<autodie> pragma.
6493 Like C<exec>, C<system> allows you to lie to a program about its name if
6494 you use the C<system PROGRAM LIST> syntax. Again, see L</exec>.
6496 Since C<SIGINT> and C<SIGQUIT> are ignored during the execution of
6497 C<system>, if you expect your program to terminate on receipt of these
6498 signals you will need to arrange to do so yourself based on the return
6501 @args = ("command", "arg1", "arg2");
6503 or die "system @args failed: $?"
6505 If you'd like to manually inspect C<system>'s failure, you can check all
6506 possible failure modes by inspecting C<$?> like this:
6509 print "failed to execute: $!\n";
6512 printf "child died with signal %d, %s coredump\n",
6513 ($? & 127), ($? & 128) ? 'with' : 'without';
6516 printf "child exited with value %d\n", $? >> 8;
6519 Alternatively you might inspect the value of C<${^CHILD_ERROR_NATIVE}>
6520 with the W*() calls of the POSIX extension.
6522 When the arguments get executed via the system shell, results
6523 and return codes will be subject to its quirks and capabilities.
6524 See L<perlop/"`STRING`"> and L</exec> for details.
6526 =item syswrite FILEHANDLE,SCALAR,LENGTH,OFFSET
6529 =item syswrite FILEHANDLE,SCALAR,LENGTH
6531 =item syswrite FILEHANDLE,SCALAR
6533 Attempts to write LENGTH bytes of data from variable SCALAR to the
6534 specified FILEHANDLE, using the system call write(2). If LENGTH is
6535 not specified, writes whole SCALAR. It bypasses buffered IO, so
6536 mixing this with reads (other than C<sysread())>, C<print>, C<write>,
6537 C<seek>, C<tell>, or C<eof> may cause confusion because the perlio and
6538 stdio layers usually buffers data. Returns the number of bytes
6539 actually written, or C<undef> if there was an error (in this case the
6540 errno variable C<$!> is also set). If the LENGTH is greater than the
6541 available data in the SCALAR after the OFFSET, only as much data as is
6542 available will be written.
6544 An OFFSET may be specified to write the data from some part of the
6545 string other than the beginning. A negative OFFSET specifies writing
6546 that many characters counting backwards from the end of the string.
6547 In the case the SCALAR is empty you can use OFFSET but only zero offset.
6549 Note that if the filehandle has been marked as C<:utf8>, Unicode
6550 characters are written instead of bytes (the LENGTH, OFFSET, and the
6551 return value of syswrite() are in UTF-8 encoded Unicode characters).
6552 The C<:encoding(...)> layer implicitly introduces the C<:utf8> layer.
6553 See L</binmode>, L</open>, and the C<open> pragma, L<open>.
6555 =item tell FILEHANDLE
6560 Returns the current position I<in bytes> for FILEHANDLE, or -1 on
6561 error. FILEHANDLE may be an expression whose value gives the name of
6562 the actual filehandle. If FILEHANDLE is omitted, assumes the file
6565 Note the I<in bytes>: even if the filehandle has been set to
6566 operate on characters (for example by using the C<:encoding(utf8)> open
6567 layer), tell() will return byte offsets, not character offsets (because
6568 that would render seek() and tell() rather slow).
6570 The return value of tell() for the standard streams like the STDIN
6571 depends on the operating system: it may return -1 or something else.
6572 tell() on pipes, fifos, and sockets usually returns -1.
6574 There is no C<systell> function. Use C<sysseek(FH, 0, 1)> for that.
6576 Do not use tell() (or other buffered I/O operations) on a file handle
6577 that has been manipulated by sysread(), syswrite() or sysseek().
6578 Those functions ignore the buffering, while tell() does not.
6580 =item telldir DIRHANDLE
6583 Returns the current position of the C<readdir> routines on DIRHANDLE.
6584 Value may be given to C<seekdir> to access a particular location in a
6585 directory. C<telldir> has the same caveats about possible directory
6586 compaction as the corresponding system library routine.
6588 =item tie VARIABLE,CLASSNAME,LIST
6591 This function binds a variable to a package class that will provide the
6592 implementation for the variable. VARIABLE is the name of the variable
6593 to be enchanted. CLASSNAME is the name of a class implementing objects
6594 of correct type. Any additional arguments are passed to the C<new>
6595 method of the class (meaning C<TIESCALAR>, C<TIEHANDLE>, C<TIEARRAY>,
6596 or C<TIEHASH>). Typically these are arguments such as might be passed
6597 to the C<dbm_open()> function of C. The object returned by the C<new>
6598 method is also returned by the C<tie> function, which would be useful
6599 if you want to access other methods in CLASSNAME.
6601 Note that functions such as C<keys> and C<values> may return huge lists
6602 when used on large objects, like DBM files. You may prefer to use the
6603 C<each> function to iterate over such. Example:
6605 # print out history file offsets
6607 tie(%HIST, 'NDBM_File', '/usr/lib/news/history', 1, 0);
6608 while (($key,$val) = each %HIST) {
6609 print $key, ' = ', unpack('L',$val), "\n";
6613 A class implementing a hash should have the following methods:
6615 TIEHASH classname, LIST
6617 STORE this, key, value
6622 NEXTKEY this, lastkey
6627 A class implementing an ordinary array should have the following methods:
6629 TIEARRAY classname, LIST
6631 STORE this, key, value
6633 STORESIZE this, count
6639 SPLICE this, offset, length, LIST
6644 A class implementing a file handle should have the following methods:
6646 TIEHANDLE classname, LIST
6647 READ this, scalar, length, offset
6650 WRITE this, scalar, length, offset
6652 PRINTF this, format, LIST
6656 SEEK this, position, whence
6658 OPEN this, mode, LIST
6663 A class implementing a scalar should have the following methods:
6665 TIESCALAR classname, LIST
6671 Not all methods indicated above need be implemented. See L<perltie>,
6672 L<Tie::Hash>, L<Tie::Array>, L<Tie::Scalar>, and L<Tie::Handle>.
6674 Unlike C<dbmopen>, the C<tie> function will not use or require a module
6675 for you--you need to do that explicitly yourself. See L<DB_File>
6676 or the F<Config> module for interesting C<tie> implementations.
6678 For further details see L<perltie>, L<"tied VARIABLE">.
6683 Returns a reference to the object underlying VARIABLE (the same value
6684 that was originally returned by the C<tie> call that bound the variable
6685 to a package.) Returns the undefined value if VARIABLE isn't tied to a
6691 Returns the number of non-leap seconds since whatever time the system
6692 considers to be the epoch, suitable for feeding to C<gmtime> and
6693 C<localtime>. On most systems the epoch is 00:00:00 UTC, January 1, 1970;
6694 a prominent exception being Mac OS Classic which uses 00:00:00, January 1,
6695 1904 in the current local time zone for its epoch.
6697 For measuring time in better granularity than one second,
6698 you may use either the L<Time::HiRes> module (from CPAN, and starting from
6699 Perl 5.8 part of the standard distribution), or if you have
6700 gettimeofday(2), you may be able to use the C<syscall> interface of Perl.
6701 See L<perlfaq8> for details.
6703 For date and time processing look at the many related modules on CPAN.
6704 For a comprehensive date and time representation look at the
6710 Returns a four-element list giving the user and system times, in
6711 seconds, for this process and the children of this process.
6713 ($user,$system,$cuser,$csystem) = times;
6715 In scalar context, C<times> returns C<$user>.
6717 Note that times for children are included only after they terminate.
6721 The transliteration operator. Same as C<y///>. See
6722 L<perlop/"Quote and Quote-like Operators">.
6724 =item truncate FILEHANDLE,LENGTH
6727 =item truncate EXPR,LENGTH
6729 Truncates the file opened on FILEHANDLE, or named by EXPR, to the
6730 specified length. Produces a fatal error if truncate isn't implemented
6731 on your system. Returns true if successful, the undefined value
6734 The behavior is undefined if LENGTH is greater than the length of the
6737 The position in the file of FILEHANDLE is left unchanged. You may want to
6738 call L<seek> before writing to the file.
6741 X<uc> X<uppercase> X<toupper>
6745 Returns an uppercased version of EXPR. This is the internal function
6746 implementing the C<\U> escape in double-quoted strings. Respects
6747 current LC_CTYPE locale if C<use locale> in force. See L<perllocale>
6748 and L<perlunicode> for more details about locale and Unicode support.
6749 It does not attempt to do titlecase mapping on initial letters. See
6750 C<ucfirst> for that.
6752 If EXPR is omitted, uses C<$_>.
6755 X<ucfirst> X<uppercase>
6759 Returns the value of EXPR with the first character in uppercase
6760 (titlecase in Unicode). This is the internal function implementing
6761 the C<\u> escape in double-quoted strings. Respects current LC_CTYPE
6762 locale if C<use locale> in force. See L<perllocale> and L<perlunicode>
6763 for more details about locale and Unicode support.
6765 If EXPR is omitted, uses C<$_>.
6772 Sets the umask for the process to EXPR and returns the previous value.
6773 If EXPR is omitted, merely returns the current umask.
6775 The Unix permission C<rwxr-x---> is represented as three sets of three
6776 bits, or three octal digits: C<0750> (the leading 0 indicates octal
6777 and isn't one of the digits). The C<umask> value is such a number
6778 representing disabled permissions bits. The permission (or "mode")
6779 values you pass C<mkdir> or C<sysopen> are modified by your umask, so
6780 even if you tell C<sysopen> to create a file with permissions C<0777>,
6781 if your umask is C<0022> then the file will actually be created with
6782 permissions C<0755>. If your C<umask> were C<0027> (group can't
6783 write; others can't read, write, or execute), then passing
6784 C<sysopen> C<0666> would create a file with mode C<0640> (C<0666 &~
6787 Here's some advice: supply a creation mode of C<0666> for regular
6788 files (in C<sysopen>) and one of C<0777> for directories (in
6789 C<mkdir>) and executable files. This gives users the freedom of
6790 choice: if they want protected files, they might choose process umasks
6791 of C<022>, C<027>, or even the particularly antisocial mask of C<077>.
6792 Programs should rarely if ever make policy decisions better left to
6793 the user. The exception to this is when writing files that should be
6794 kept private: mail files, web browser cookies, I<.rhosts> files, and
6797 If umask(2) is not implemented on your system and you are trying to
6798 restrict access for I<yourself> (i.e., (EXPR & 0700) > 0), produces a
6799 fatal error at run time. If umask(2) is not implemented and you are
6800 not trying to restrict access for yourself, returns C<undef>.
6802 Remember that a umask is a number, usually given in octal; it is I<not> a
6803 string of octal digits. See also L</oct>, if all you have is a string.
6806 X<undef> X<undefine>
6810 Undefines the value of EXPR, which must be an lvalue. Use only on a
6811 scalar value, an array (using C<@>), a hash (using C<%>), a subroutine
6812 (using C<&>), or a typeglob (using C<*>). (Saying C<undef $hash{$key}>
6813 will probably not do what you expect on most predefined variables or
6814 DBM list values, so don't do that; see L<delete>.) Always returns the
6815 undefined value. You can omit the EXPR, in which case nothing is
6816 undefined, but you still get an undefined value that you could, for
6817 instance, return from a subroutine, assign to a variable or pass as a
6818 parameter. Examples:
6821 undef $bar{'blurfl'}; # Compare to: delete $bar{'blurfl'};
6825 undef *xyz; # destroys $xyz, @xyz, %xyz, &xyz, etc.
6826 return (wantarray ? (undef, $errmsg) : undef) if $they_blew_it;
6827 select undef, undef, undef, 0.25;
6828 ($a, $b, undef, $c) = &foo; # Ignore third value returned
6830 Note that this is a unary operator, not a list operator.
6833 X<unlink> X<delete> X<remove> X<rm> X<del>
6837 Deletes a list of files. Returns the number of files successfully
6840 $cnt = unlink 'a', 'b', 'c';
6844 Note: C<unlink> will not attempt to delete directories unless you are superuser
6845 and the B<-U> flag is supplied to Perl. Even if these conditions are
6846 met, be warned that unlinking a directory can inflict damage on your
6847 filesystem. Finally, using C<unlink> on directories is not supported on
6848 many operating systems. Use C<rmdir> instead.
6850 If LIST is omitted, uses C<$_>.
6852 =item unpack TEMPLATE,EXPR
6855 =item unpack TEMPLATE
6857 C<unpack> does the reverse of C<pack>: it takes a string
6858 and expands it out into a list of values.
6859 (In scalar context, it returns merely the first value produced.)
6861 If EXPR is omitted, unpacks the C<$_> string.
6863 The string is broken into chunks described by the TEMPLATE. Each chunk
6864 is converted separately to a value. Typically, either the string is a result
6865 of C<pack>, or the characters of the string represent a C structure of some
6868 The TEMPLATE has the same format as in the C<pack> function.
6869 Here's a subroutine that does substring:
6872 my($what,$where,$howmuch) = @_;
6873 unpack("x$where a$howmuch", $what);
6878 sub ordinal { unpack("W",$_[0]); } # same as ord()
6880 In addition to fields allowed in pack(), you may prefix a field with
6881 a %<number> to indicate that
6882 you want a <number>-bit checksum of the items instead of the items
6883 themselves. Default is a 16-bit checksum. Checksum is calculated by
6884 summing numeric values of expanded values (for string fields the sum of
6885 C<ord($char)> is taken, for bit fields the sum of zeroes and ones).
6887 For example, the following
6888 computes the same number as the System V sum program:
6892 unpack("%32W*",<>) % 65535;
6895 The following efficiently counts the number of set bits in a bit vector:
6897 $setbits = unpack("%32b*", $selectmask);
6899 The C<p> and C<P> formats should be used with care. Since Perl
6900 has no way of checking whether the value passed to C<unpack()>
6901 corresponds to a valid memory location, passing a pointer value that's
6902 not known to be valid is likely to have disastrous consequences.
6904 If there are more pack codes or if the repeat count of a field or a group
6905 is larger than what the remainder of the input string allows, the result
6906 is not well defined: in some cases, the repeat count is decreased, or
6907 C<unpack()> will produce null strings or zeroes, or terminate with an
6908 error. If the input string is longer than one described by the TEMPLATE,
6909 the rest is ignored.
6911 See L</pack> for more examples and notes.
6913 =item untie VARIABLE
6916 Breaks the binding between a variable and a package. (See C<tie>.)
6917 Has no effect if the variable is not tied.
6919 =item unshift ARRAY,LIST
6922 Does the opposite of a C<shift>. Or the opposite of a C<push>,
6923 depending on how you look at it. Prepends list to the front of the
6924 array, and returns the new number of elements in the array.
6926 unshift(@ARGV, '-e') unless $ARGV[0] =~ /^-/;
6928 Note the LIST is prepended whole, not one element at a time, so the
6929 prepended elements stay in the same order. Use C<reverse> to do the
6932 =item use Module VERSION LIST
6933 X<use> X<module> X<import>
6935 =item use Module VERSION
6937 =item use Module LIST
6943 Imports some semantics into the current package from the named module,
6944 generally by aliasing certain subroutine or variable names into your
6945 package. It is exactly equivalent to
6947 BEGIN { require Module; Module->import( LIST ); }
6949 except that Module I<must> be a bareword.
6951 In the peculiar C<use VERSION> form, VERSION may be either a numeric
6952 argument such as 5.006, which will be compared to C<$]>, or a literal of
6953 the form v5.6.1, which will be compared to C<$^V> (aka $PERL_VERSION). A
6954 fatal error is produced if VERSION is greater than the version of the
6955 current Perl interpreter; Perl will not attempt to parse the rest of the
6956 file. Compare with L</require>, which can do a similar check at run time.
6957 Symmetrically, C<no VERSION> allows you to specify that you want a version
6958 of perl older than the specified one.
6960 Specifying VERSION as a literal of the form v5.6.1 should generally be
6961 avoided, because it leads to misleading error messages under earlier
6962 versions of Perl (that is, prior to 5.6.0) that do not support this
6963 syntax. The equivalent numeric version should be used instead.
6965 use v5.6.1; # compile time version check
6967 use 5.006_001; # ditto; preferred for backwards compatibility
6969 This is often useful if you need to check the current Perl version before
6970 C<use>ing library modules that won't work with older versions of Perl.
6971 (We try not to do this more than we have to.)
6973 Also, if the specified perl version is greater than or equal to 5.9.5,
6974 C<use VERSION> will also load the C<feature> pragma and enable all
6975 features available in the requested version. See L<feature>.
6976 Similarly, if the specified perl version is greater than or equal to
6977 5.11.0, strictures are enabled lexically as with C<use strict> (except
6978 that the F<strict.pm> file is not actually loaded).
6980 The C<BEGIN> forces the C<require> and C<import> to happen at compile time. The
6981 C<require> makes sure the module is loaded into memory if it hasn't been
6982 yet. The C<import> is not a builtin--it's just an ordinary static method
6983 call into the C<Module> package to tell the module to import the list of
6984 features back into the current package. The module can implement its
6985 C<import> method any way it likes, though most modules just choose to
6986 derive their C<import> method via inheritance from the C<Exporter> class that
6987 is defined in the C<Exporter> module. See L<Exporter>. If no C<import>
6988 method can be found then the call is skipped, even if there is an AUTOLOAD
6991 If you do not want to call the package's C<import> method (for instance,
6992 to stop your namespace from being altered), explicitly supply the empty list:
6996 That is exactly equivalent to
6998 BEGIN { require Module }
7000 If the VERSION argument is present between Module and LIST, then the
7001 C<use> will call the VERSION method in class Module with the given
7002 version as an argument. The default VERSION method, inherited from
7003 the UNIVERSAL class, croaks if the given version is larger than the
7004 value of the variable C<$Module::VERSION>.
7006 Again, there is a distinction between omitting LIST (C<import> called
7007 with no arguments) and an explicit empty LIST C<()> (C<import> not
7008 called). Note that there is no comma after VERSION!
7010 Because this is a wide-open interface, pragmas (compiler directives)
7011 are also implemented this way. Currently implemented pragmas are:
7016 use sigtrap qw(SEGV BUS);
7017 use strict qw(subs vars refs);
7018 use subs qw(afunc blurfl);
7019 use warnings qw(all);
7020 use sort qw(stable _quicksort _mergesort);
7022 Some of these pseudo-modules import semantics into the current
7023 block scope (like C<strict> or C<integer>, unlike ordinary modules,
7024 which import symbols into the current package (which are effective
7025 through the end of the file).
7027 There's a corresponding C<no> command that unimports meanings imported
7028 by C<use>, i.e., it calls C<unimport Module LIST> instead of C<import>.
7029 It behaves exactly as C<import> does with respect to VERSION, an
7030 omitted LIST, empty LIST, or no unimport method being found.
7036 See L<perlmodlib> for a list of standard modules and pragmas. See L<perlrun>
7037 for the C<-M> and C<-m> command-line options to perl that give C<use>
7038 functionality from the command-line.
7043 Changes the access and modification times on each file of a list of
7044 files. The first two elements of the list must be the NUMERICAL access
7045 and modification times, in that order. Returns the number of files
7046 successfully changed. The inode change time of each file is set
7047 to the current time. For example, this code has the same effect as the
7048 Unix touch(1) command when the files I<already exist> and belong to
7049 the user running the program:
7052 $atime = $mtime = time;
7053 utime $atime, $mtime, @ARGV;
7055 Since perl 5.7.2, if the first two elements of the list are C<undef>, then
7056 the utime(2) function in the C library will be called with a null second
7057 argument. On most systems, this will set the file's access and
7058 modification times to the current time (i.e. equivalent to the example
7059 above) and will even work on other users' files where you have write
7062 utime undef, undef, @ARGV;
7064 Under NFS this will use the time of the NFS server, not the time of
7065 the local machine. If there is a time synchronization problem, the
7066 NFS server and local machine will have different times. The Unix
7067 touch(1) command will in fact normally use this form instead of the
7068 one shown in the first example.
7070 Note that only passing one of the first two elements as C<undef> will
7071 be equivalent of passing it as 0 and will not have the same effect as
7072 described when they are both C<undef>. This case will also trigger an
7073 uninitialized warning.
7075 On systems that support futimes, you might pass file handles among the
7076 files. On systems that don't support futimes, passing file handles
7077 produces a fatal error at run time. The file handles must be passed
7078 as globs or references to be recognized. Barewords are considered
7086 Returns a list consisting of all the values of the named hash, or the values
7087 of an array. (In a scalar context, returns the number of values.)
7089 The values are returned in an apparently random order. The actual
7090 random order is subject to change in future versions of perl, but it
7091 is guaranteed to be the same order as either the C<keys> or C<each>
7092 function would produce on the same (unmodified) hash. Since Perl
7093 5.8.1 the ordering is different even between different runs of Perl
7094 for security reasons (see L<perlsec/"Algorithmic Complexity Attacks">).
7096 As a side effect, calling values() resets the HASH or ARRAY's internal
7098 see L</each>. (In particular, calling values() in void context resets
7099 the iterator with no other overhead. Apart from resetting the iterator,
7100 C<values @array> in list context is no different to plain C<@array>.
7101 We recommend that you use void context C<keys @array> for this, but reasoned
7102 that it taking C<values @array> out would require more documentation than
7106 Note that the values are not copied, which means modifying them will
7107 modify the contents of the hash:
7109 for (values %hash) { s/foo/bar/g } # modifies %hash values
7110 for (@hash{keys %hash}) { s/foo/bar/g } # same
7112 See also C<keys>, C<each>, and C<sort>.
7114 =item vec EXPR,OFFSET,BITS
7115 X<vec> X<bit> X<bit vector>
7117 Treats the string in EXPR as a bit vector made up of elements of
7118 width BITS, and returns the value of the element specified by OFFSET
7119 as an unsigned integer. BITS therefore specifies the number of bits
7120 that are reserved for each element in the bit vector. This must
7121 be a power of two from 1 to 32 (or 64, if your platform supports
7124 If BITS is 8, "elements" coincide with bytes of the input string.
7126 If BITS is 16 or more, bytes of the input string are grouped into chunks
7127 of size BITS/8, and each group is converted to a number as with
7128 pack()/unpack() with big-endian formats C<n>/C<N> (and analogously
7129 for BITS==64). See L<"pack"> for details.
7131 If bits is 4 or less, the string is broken into bytes, then the bits
7132 of each byte are broken into 8/BITS groups. Bits of a byte are
7133 numbered in a little-endian-ish way, as in C<0x01>, C<0x02>,
7134 C<0x04>, C<0x08>, C<0x10>, C<0x20>, C<0x40>, C<0x80>. For example,
7135 breaking the single input byte C<chr(0x36)> into two groups gives a list
7136 C<(0x6, 0x3)>; breaking it into 4 groups gives C<(0x2, 0x1, 0x3, 0x0)>.
7138 C<vec> may also be assigned to, in which case parentheses are needed
7139 to give the expression the correct precedence as in
7141 vec($image, $max_x * $x + $y, 8) = 3;
7143 If the selected element is outside the string, the value 0 is returned.
7144 If an element off the end of the string is written to, Perl will first
7145 extend the string with sufficiently many zero bytes. It is an error
7146 to try to write off the beginning of the string (i.e. negative OFFSET).
7148 If the string happens to be encoded as UTF-8 internally (and thus has
7149 the UTF8 flag set), this is ignored by C<vec>, and it operates on the
7150 internal byte string, not the conceptual character string, even if you
7151 only have characters with values less than 256.
7153 Strings created with C<vec> can also be manipulated with the logical
7154 operators C<|>, C<&>, C<^>, and C<~>. These operators will assume a bit
7155 vector operation is desired when both operands are strings.
7156 See L<perlop/"Bitwise String Operators">.
7158 The following code will build up an ASCII string saying C<'PerlPerlPerl'>.
7159 The comments show the string after each step. Note that this code works
7160 in the same way on big-endian or little-endian machines.
7163 vec($foo, 0, 32) = 0x5065726C; # 'Perl'
7165 # $foo eq "Perl" eq "\x50\x65\x72\x6C", 32 bits
7166 print vec($foo, 0, 8); # prints 80 == 0x50 == ord('P')
7168 vec($foo, 2, 16) = 0x5065; # 'PerlPe'
7169 vec($foo, 3, 16) = 0x726C; # 'PerlPerl'
7170 vec($foo, 8, 8) = 0x50; # 'PerlPerlP'
7171 vec($foo, 9, 8) = 0x65; # 'PerlPerlPe'
7172 vec($foo, 20, 4) = 2; # 'PerlPerlPe' . "\x02"
7173 vec($foo, 21, 4) = 7; # 'PerlPerlPer'
7175 vec($foo, 45, 2) = 3; # 'PerlPerlPer' . "\x0c"
7176 vec($foo, 93, 1) = 1; # 'PerlPerlPer' . "\x2c"
7177 vec($foo, 94, 1) = 1; # 'PerlPerlPerl'
7180 To transform a bit vector into a string or list of 0's and 1's, use these:
7182 $bits = unpack("b*", $vector);
7183 @bits = split(//, unpack("b*", $vector));
7185 If you know the exact length in bits, it can be used in place of the C<*>.
7187 Here is an example to illustrate how the bits actually fall in place:
7193 unpack("V",$_) 01234567890123456789012345678901
7194 ------------------------------------------------------------------
7199 for ($shift=0; $shift < $width; ++$shift) {
7200 for ($off=0; $off < 32/$width; ++$off) {
7201 $str = pack("B*", "0"x32);
7202 $bits = (1<<$shift);
7203 vec($str, $off, $width) = $bits;
7204 $res = unpack("b*",$str);
7205 $val = unpack("V", $str);
7212 vec($_,@#,@#) = @<< == @######### @>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
7213 $off, $width, $bits, $val, $res
7217 Regardless of the machine architecture on which it is run, the above
7218 example should print the following table:
7221 unpack("V",$_) 01234567890123456789012345678901
7222 ------------------------------------------------------------------
7223 vec($_, 0, 1) = 1 == 1 10000000000000000000000000000000
7224 vec($_, 1, 1) = 1 == 2 01000000000000000000000000000000
7225 vec($_, 2, 1) = 1 == 4 00100000000000000000000000000000
7226 vec($_, 3, 1) = 1 == 8 00010000000000000000000000000000
7227 vec($_, 4, 1) = 1 == 16 00001000000000000000000000000000
7228 vec($_, 5, 1) = 1 == 32 00000100000000000000000000000000
7229 vec($_, 6, 1) = 1 == 64 00000010000000000000000000000000
7230 vec($_, 7, 1) = 1 == 128 00000001000000000000000000000000
7231 vec($_, 8, 1) = 1 == 256 00000000100000000000000000000000
7232 vec($_, 9, 1) = 1 == 512 00000000010000000000000000000000
7233 vec($_,10, 1) = 1 == 1024 00000000001000000000000000000000
7234 vec($_,11, 1) = 1 == 2048 00000000000100000000000000000000
7235 vec($_,12, 1) = 1 == 4096 00000000000010000000000000000000
7236 vec($_,13, 1) = 1 == 8192 00000000000001000000000000000000
7237 vec($_,14, 1) = 1 == 16384 00000000000000100000000000000000
7238 vec($_,15, 1) = 1 == 32768 00000000000000010000000000000000
7239 vec($_,16, 1) = 1 == 65536 00000000000000001000000000000000
7240 vec($_,17, 1) = 1 == 131072 00000000000000000100000000000000
7241 vec($_,18, 1) = 1 == 262144 00000000000000000010000000000000
7242 vec($_,19, 1) = 1 == 524288 00000000000000000001000000000000
7243 vec($_,20, 1) = 1 == 1048576 00000000000000000000100000000000
7244 vec($_,21, 1) = 1 == 2097152 00000000000000000000010000000000
7245 vec($_,22, 1) = 1 == 4194304 00000000000000000000001000000000
7246 vec($_,23, 1) = 1 == 8388608 00000000000000000000000100000000
7247 vec($_,24, 1) = 1 == 16777216 00000000000000000000000010000000
7248 vec($_,25, 1) = 1 == 33554432 00000000000000000000000001000000
7249 vec($_,26, 1) = 1 == 67108864 00000000000000000000000000100000
7250 vec($_,27, 1) = 1 == 134217728 00000000000000000000000000010000
7251 vec($_,28, 1) = 1 == 268435456 00000000000000000000000000001000
7252 vec($_,29, 1) = 1 == 536870912 00000000000000000000000000000100
7253 vec($_,30, 1) = 1 == 1073741824 00000000000000000000000000000010
7254 vec($_,31, 1) = 1 == 2147483648 00000000000000000000000000000001
7255 vec($_, 0, 2) = 1 == 1 10000000000000000000000000000000
7256 vec($_, 1, 2) = 1 == 4 00100000000000000000000000000000
7257 vec($_, 2, 2) = 1 == 16 00001000000000000000000000000000
7258 vec($_, 3, 2) = 1 == 64 00000010000000000000000000000000
7259 vec($_, 4, 2) = 1 == 256 00000000100000000000000000000000
7260 vec($_, 5, 2) = 1 == 1024 00000000001000000000000000000000
7261 vec($_, 6, 2) = 1 == 4096 00000000000010000000000000000000
7262 vec($_, 7, 2) = 1 == 16384 00000000000000100000000000000000
7263 vec($_, 8, 2) = 1 == 65536 00000000000000001000000000000000
7264 vec($_, 9, 2) = 1 == 262144 00000000000000000010000000000000
7265 vec($_,10, 2) = 1 == 1048576 00000000000000000000100000000000
7266 vec($_,11, 2) = 1 == 4194304 00000000000000000000001000000000
7267 vec($_,12, 2) = 1 == 16777216 00000000000000000000000010000000
7268 vec($_,13, 2) = 1 == 67108864 00000000000000000000000000100000
7269 vec($_,14, 2) = 1 == 268435456 00000000000000000000000000001000
7270 vec($_,15, 2) = 1 == 1073741824 00000000000000000000000000000010
7271 vec($_, 0, 2) = 2 == 2 01000000000000000000000000000000
7272 vec($_, 1, 2) = 2 == 8 00010000000000000000000000000000
7273 vec($_, 2, 2) = 2 == 32 00000100000000000000000000000000
7274 vec($_, 3, 2) = 2 == 128 00000001000000000000000000000000
7275 vec($_, 4, 2) = 2 == 512 00000000010000000000000000000000
7276 vec($_, 5, 2) = 2 == 2048 00000000000100000000000000000000
7277 vec($_, 6, 2) = 2 == 8192 00000000000001000000000000000000
7278 vec($_, 7, 2) = 2 == 32768 00000000000000010000000000000000
7279 vec($_, 8, 2) = 2 == 131072 00000000000000000100000000000000
7280 vec($_, 9, 2) = 2 == 524288 00000000000000000001000000000000
7281 vec($_,10, 2) = 2 == 2097152 00000000000000000000010000000000
7282 vec($_,11, 2) = 2 == 8388608 00000000000000000000000100000000
7283 vec($_,12, 2) = 2 == 33554432 00000000000000000000000001000000
7284 vec($_,13, 2) = 2 == 134217728 00000000000000000000000000010000
7285 vec($_,14, 2) = 2 == 536870912 00000000000000000000000000000100
7286 vec($_,15, 2) = 2 == 2147483648 00000000000000000000000000000001
7287 vec($_, 0, 4) = 1 == 1 10000000000000000000000000000000
7288 vec($_, 1, 4) = 1 == 16 00001000000000000000000000000000
7289 vec($_, 2, 4) = 1 == 256 00000000100000000000000000000000
7290 vec($_, 3, 4) = 1 == 4096 00000000000010000000000000000000
7291 vec($_, 4, 4) = 1 == 65536 00000000000000001000000000000000
7292 vec($_, 5, 4) = 1 == 1048576 00000000000000000000100000000000
7293 vec($_, 6, 4) = 1 == 16777216 00000000000000000000000010000000
7294 vec($_, 7, 4) = 1 == 268435456 00000000000000000000000000001000
7295 vec($_, 0, 4) = 2 == 2 01000000000000000000000000000000
7296 vec($_, 1, 4) = 2 == 32 00000100000000000000000000000000
7297 vec($_, 2, 4) = 2 == 512 00000000010000000000000000000000
7298 vec($_, 3, 4) = 2 == 8192 00000000000001000000000000000000
7299 vec($_, 4, 4) = 2 == 131072 00000000000000000100000000000000
7300 vec($_, 5, 4) = 2 == 2097152 00000000000000000000010000000000
7301 vec($_, 6, 4) = 2 == 33554432 00000000000000000000000001000000
7302 vec($_, 7, 4) = 2 == 536870912 00000000000000000000000000000100
7303 vec($_, 0, 4) = 4 == 4 00100000000000000000000000000000
7304 vec($_, 1, 4) = 4 == 64 00000010000000000000000000000000
7305 vec($_, 2, 4) = 4 == 1024 00000000001000000000000000000000
7306 vec($_, 3, 4) = 4 == 16384 00000000000000100000000000000000
7307 vec($_, 4, 4) = 4 == 262144 00000000000000000010000000000000
7308 vec($_, 5, 4) = 4 == 4194304 00000000000000000000001000000000
7309 vec($_, 6, 4) = 4 == 67108864 00000000000000000000000000100000
7310 vec($_, 7, 4) = 4 == 1073741824 00000000000000000000000000000010
7311 vec($_, 0, 4) = 8 == 8 00010000000000000000000000000000
7312 vec($_, 1, 4) = 8 == 128 00000001000000000000000000000000
7313 vec($_, 2, 4) = 8 == 2048 00000000000100000000000000000000
7314 vec($_, 3, 4) = 8 == 32768 00000000000000010000000000000000
7315 vec($_, 4, 4) = 8 == 524288 00000000000000000001000000000000
7316 vec($_, 5, 4) = 8 == 8388608 00000000000000000000000100000000
7317 vec($_, 6, 4) = 8 == 134217728 00000000000000000000000000010000
7318 vec($_, 7, 4) = 8 == 2147483648 00000000000000000000000000000001
7319 vec($_, 0, 8) = 1 == 1 10000000000000000000000000000000
7320 vec($_, 1, 8) = 1 == 256 00000000100000000000000000000000
7321 vec($_, 2, 8) = 1 == 65536 00000000000000001000000000000000
7322 vec($_, 3, 8) = 1 == 16777216 00000000000000000000000010000000
7323 vec($_, 0, 8) = 2 == 2 01000000000000000000000000000000
7324 vec($_, 1, 8) = 2 == 512 00000000010000000000000000000000
7325 vec($_, 2, 8) = 2 == 131072 00000000000000000100000000000000
7326 vec($_, 3, 8) = 2 == 33554432 00000000000000000000000001000000
7327 vec($_, 0, 8) = 4 == 4 00100000000000000000000000000000
7328 vec($_, 1, 8) = 4 == 1024 00000000001000000000000000000000
7329 vec($_, 2, 8) = 4 == 262144 00000000000000000010000000000000
7330 vec($_, 3, 8) = 4 == 67108864 00000000000000000000000000100000
7331 vec($_, 0, 8) = 8 == 8 00010000000000000000000000000000
7332 vec($_, 1, 8) = 8 == 2048 00000000000100000000000000000000
7333 vec($_, 2, 8) = 8 == 524288 00000000000000000001000000000000
7334 vec($_, 3, 8) = 8 == 134217728 00000000000000000000000000010000
7335 vec($_, 0, 8) = 16 == 16 00001000000000000000000000000000
7336 vec($_, 1, 8) = 16 == 4096 00000000000010000000000000000000
7337 vec($_, 2, 8) = 16 == 1048576 00000000000000000000100000000000
7338 vec($_, 3, 8) = 16 == 268435456 00000000000000000000000000001000
7339 vec($_, 0, 8) = 32 == 32 00000100000000000000000000000000
7340 vec($_, 1, 8) = 32 == 8192 00000000000001000000000000000000
7341 vec($_, 2, 8) = 32 == 2097152 00000000000000000000010000000000
7342 vec($_, 3, 8) = 32 == 536870912 00000000000000000000000000000100
7343 vec($_, 0, 8) = 64 == 64 00000010000000000000000000000000
7344 vec($_, 1, 8) = 64 == 16384 00000000000000100000000000000000
7345 vec($_, 2, 8) = 64 == 4194304 00000000000000000000001000000000
7346 vec($_, 3, 8) = 64 == 1073741824 00000000000000000000000000000010
7347 vec($_, 0, 8) = 128 == 128 00000001000000000000000000000000
7348 vec($_, 1, 8) = 128 == 32768 00000000000000010000000000000000
7349 vec($_, 2, 8) = 128 == 8388608 00000000000000000000000100000000
7350 vec($_, 3, 8) = 128 == 2147483648 00000000000000000000000000000001
7355 Behaves like the wait(2) system call on your system: it waits for a child
7356 process to terminate and returns the pid of the deceased process, or
7357 C<-1> if there are no child processes. The status is returned in C<$?>
7358 and C<${^CHILD_ERROR_NATIVE}>.
7359 Note that a return value of C<-1> could mean that child processes are
7360 being automatically reaped, as described in L<perlipc>.
7362 =item waitpid PID,FLAGS
7365 Waits for a particular child process to terminate and returns the pid of
7366 the deceased process, or C<-1> if there is no such child process. On some
7367 systems, a value of 0 indicates that there are processes still running.
7368 The status is returned in C<$?> and C<${^CHILD_ERROR_NATIVE}>. If you say
7370 use POSIX ":sys_wait_h";
7373 $kid = waitpid(-1, WNOHANG);
7376 then you can do a non-blocking wait for all pending zombie processes.
7377 Non-blocking wait is available on machines supporting either the
7378 waitpid(2) or wait4(2) system calls. However, waiting for a particular
7379 pid with FLAGS of C<0> is implemented everywhere. (Perl emulates the
7380 system call by remembering the status values of processes that have
7381 exited but have not been harvested by the Perl script yet.)
7383 Note that on some systems, a return value of C<-1> could mean that child
7384 processes are being automatically reaped. See L<perlipc> for details,
7385 and for other examples.
7388 X<wantarray> X<context>
7390 Returns true if the context of the currently executing subroutine or
7391 C<eval> is looking for a list value. Returns false if the context is
7392 looking for a scalar. Returns the undefined value if the context is
7393 looking for no value (void context).
7395 return unless defined wantarray; # don't bother doing more
7396 my @a = complex_calculation();
7397 return wantarray ? @a : "@a";
7399 C<wantarray()>'s result is unspecified in the top level of a file,
7400 in a C<BEGIN>, C<UNITCHECK>, C<CHECK>, C<INIT> or C<END> block, or
7401 in a C<DESTROY> method.
7403 This function should have been named wantlist() instead.
7406 X<warn> X<warning> X<STDERR>
7408 Prints the value of LIST to STDERR. If the last element of LIST does
7409 not end in a newline, it appends the same file/line number text as C<die>
7412 If the output is empty and C<$@> already contains a value (typically from a
7413 previous eval) that value is used after appending C<"\t...caught">
7414 to C<$@>. This is useful for staying almost, but not entirely similar to
7417 If C<$@> is empty then the string C<"Warning: Something's wrong"> is used.
7419 No message is printed if there is a C<$SIG{__WARN__}> handler
7420 installed. It is the handler's responsibility to deal with the message
7421 as it sees fit (like, for instance, converting it into a C<die>). Most
7422 handlers must therefore make arrangements to actually display the
7423 warnings that they are not prepared to deal with, by calling C<warn>
7424 again in the handler. Note that this is quite safe and will not
7425 produce an endless loop, since C<__WARN__> hooks are not called from
7428 You will find this behavior is slightly different from that of
7429 C<$SIG{__DIE__}> handlers (which don't suppress the error text, but can
7430 instead call C<die> again to change it).
7432 Using a C<__WARN__> handler provides a powerful way to silence all
7433 warnings (even the so-called mandatory ones). An example:
7435 # wipe out *all* compile-time warnings
7436 BEGIN { $SIG{'__WARN__'} = sub { warn $_[0] if $DOWARN } }
7438 my $foo = 20; # no warning about duplicate my $foo,
7439 # but hey, you asked for it!
7440 # no compile-time or run-time warnings before here
7443 # run-time warnings enabled after here
7444 warn "\$foo is alive and $foo!"; # does show up
7446 See L<perlvar> for details on setting C<%SIG> entries, and for more
7447 examples. See the Carp module for other kinds of warnings using its
7448 carp() and cluck() functions.
7450 =item write FILEHANDLE
7457 Writes a formatted record (possibly multi-line) to the specified FILEHANDLE,
7458 using the format associated with that file. By default the format for
7459 a file is the one having the same name as the filehandle, but the
7460 format for the current output channel (see the C<select> function) may be set
7461 explicitly by assigning the name of the format to the C<$~> variable.
7463 Top of form processing is handled automatically: if there is
7464 insufficient room on the current page for the formatted record, the
7465 page is advanced by writing a form feed, a special top-of-page format
7466 is used to format the new page header, and then the record is written.
7467 By default the top-of-page format is the name of the filehandle with
7468 "_TOP" appended, but it may be dynamically set to the format of your
7469 choice by assigning the name to the C<$^> variable while the filehandle is
7470 selected. The number of lines remaining on the current page is in
7471 variable C<$->, which can be set to C<0> to force a new page.
7473 If FILEHANDLE is unspecified, output goes to the current default output
7474 channel, which starts out as STDOUT but may be changed by the
7475 C<select> operator. If the FILEHANDLE is an EXPR, then the expression
7476 is evaluated and the resulting string is used to look up the name of
7477 the FILEHANDLE at run time. For more on formats, see L<perlform>.
7479 Note that write is I<not> the opposite of C<read>. Unfortunately.
7483 The transliteration operator. Same as C<tr///>. See
7484 L<perlop/"Quote and Quote-like Operators">.